Cellular senescence of granulosa cells in the pathogenesis of polycystic ovary syndrome

Abstract

Polycystic ovary syndrome (PCOS) is one of the most common endocrine disorders in women of reproductive age, but its pathology has not been fully characterized and the optimal treatment strategy remains unclear. Cellular senescence is a permanent state of cell-cycle arrest that can be induced by multiple stresses. Senescent cells contribute to the pathogenesis of various diseases, owing to an alteration in secretory profile, termed 'senescence-associated secretory phenotype' (SASP), including with respect to pro-inflammatory cytokines. Senolytics, a class of drugs that selectively eliminate senescent cells, are now being used clinically, and a combination of dasatinib and quercetin (DQ) has been extensively used as a senolytic. We aimed to investigate whether cellular senescence is involved in the pathology of PCOS and whether DQ treatment has beneficial effects in patients with PCOS. We obtained ovaries from patients with or without PCOS, and established a mouse model of PCOS by injecting dehydroepiandrosterone. The expression of the senescence markers p16INK4a, p21, p53, γH2AX, and senescence-associated β-galactosidase and the SASP-related factor interleukin-6 was significantly higher in the ovaries of patients with PCOS and PCOS mice than in controls. To evaluate the effects of hyperandrogenism and DQ on cellular senescence in vitro, we stimulated cultured human granulosa cells (GCs) with testosterone and treated them with DQ. The expression of markers of senescence and a SASP-related factor was increased by testosterone, and DQ reduced this increase. DQ reduced the expression of markers of senescence and a SASP-related factor in the ovaries of PCOS mice and improved their morphology. These results indicate that cellular senescence occurs in PCOS. Hyperandrogenism causes cellular senescence in GCs in PCOS, and senolytic treatment reduces the accumulation of senescent GCs and improves ovarian morphology under hyperandrogenism. Thus, DQ might represent a novel therapy for PCOS.

Keywords: cellular senescence; dasatinib; p16INK4a; p21; polycystic ovary syndrome; quercetin; senescence-associated secretory phenotype; senescence-associated β-galactosidase; senolytic; γH2AX.

Figure 1.

p16^INK4a, p21, and p53 are upregulated in the granulosa cells (GCs) of patients with polycystic ovary syndrome (PCOS). (A) The expression of p16^INK4a, p21, and p53 mRNAs in GCs from control patients (n = 12) and those with PCOS (n = 12) was measured by quantitative real-time PCR and normalized to that of GAPDH. (B–D) Immunohistochemical analysis was performed on GCs in the antral follicles of ovaries from control patients (n = 4) and those with PCOS (n = 4). Cross-sections of ovaries, stained with (B) anti-p16^INK4a, (C) anti-p21, and (D) anti-p53 antibodies, and counterstained with hematoxylin. (E) Corresponding negative control. For the panels (c), two or three antral follicles were randomly selected from each ovary (total eight antral follicles per group) for quantitative analysis of immunohistochemical staining. Representative images and data are shown. Scale bars: 50.0 µm. Values are mean ± SEM, and the mean differences are analyzed using Student’s t-test. *P < 0.05 and **P < 0.01. NC, negative control.

Figure 2.

p16^INK4a, p21, p53, and γH2AX are upregulated in the ovaries of polycystic ovary syndrome (PCOS) mice. (A) The mRNA expression of p16^INK4a, p21, and p53 in whole ovaries from control mice (n = 4) and those from PCOS mice (n = 4) was measured by quantitative real-time PCR and normalized to that of GAPDH expression. (B–E) Immunohistochemical analysis was performed on the GCs in antral follicles of ovaries from control mice (n = 5) and those from PCOS mice (n = 5). Cross-sections of ovaries were stained with (B) anti-p16^INK4a, (C) anti-p21, (D) anti-p53, and (E) anti-γH2AX antibodies, and counterstained with hematoxylin. (F) Corresponding negative control. Scale bars: 100 µm. (c) One or two antral follicles were selected from each ovary (total 8 antral follicles per group) for quantitative analysis of immunohistochemical staining. Values are mean ± SEM, and the mean differences are analyzed using Student’s t-test. *P < 0.05 and **P < 0.01. GCs, granulosa cells; NC, negative control. We repeated experiments three times using different mice, and representative data are shown.

Figure 3.

Effects of testosterone on the expression of markers of senescence in cultured human granulosa cells (GCs). Human GCs were treated with testosterone (15 µg/ml). (A–D) Protein expression levels of p16^INK4a, p21, p53, and γH2AX in cultured human GCs, measured using western blotting (n = 10). β-actin was used as a loading control. (b) Quantitative analysis of the western blots and (a) representative images are shown. All full-size, uncropped images of western blots used for analysis are shown in Supplementary Fig. S3. Results are expressed relative to the mean control value. Values are mean ± SEM, the mean differences are analyzed using Student’s t-test. *P < 0.05 and **P < 0.01. (E) SA-β-gal staining of human GCs. Scale bars: 200 µm. (A–E) The experiments were performed 10 times using 10 different women’s GCs, and representative images are shown.

Figure 4.

A combination of dasatinib and quercetin (DQ) reduces the testosterone-induced senescence of, and secretion of a SASP-related factor by, human granulosa cells (GCs). Human GCs were treated with testosterone (15 µg/ml), dasatinib (1 nM), and quercetin (20 µM), concurrently for 24 h. The protein expression of (A) p16^INK4a and (B) p21 in cultured human GCs was analyzed using western blotting (n = 5). β-actin was used as a loading control. (b) Quantitative analysis of the western blots using five different women’s GCs are expressed relative to the mean control value, and (a) representative images are shown. All full-size, uncropped images of western blots used for analysis are shown in Supplementary Fig. S4. (C) Concentration of IL-6 in the culture medium, measured by ELISA using six different women’s GCs. Values are mean ± SEM, and the mean differences were analyzed using one-way ANOVA followed by Tukey–Kramer honest significant difference test. *P < 0.05 and **P < 0.01.

Figure 5.

A combination of dasatinib and quercetin (DQ) treatment reduces the senescence of, and the secretion of a senescence-associated secretory phenotype (SASP)-related factor by, granulosa cells in the antral follicles of polycystic ovary syndrome (PCOS) mice. The mice were divided into three groups, which were s.c. injected with dehydroepiandrosterone (DHEA 60 mg/kg per day) (PCOS, n = 6), s.c. injected with DHEA and orally administered vehicle (10% polyethylene glycol 400) (PCOS + vehicle, n = 6), or s.c. injected with DHEA and orally administered DQ (D: dasatinib 5 mg/kg per day, Q: quercetin 50 mg/kg per day) (PCOS + DQ, n = 6). Subcutaneous injections were performed daily for 20 days, and oral administration was performed on Days 1, 6, 11, and 16. Ovaries were collected on Day 21. Cross-sections of ovaries were stained with (A) anti-p16^INK4a, (B) anti-p21, (C) anti-p53, and (D) anti-IL-6 antibodies and counterstained with hematoxylin. (d) One or two antral follicles were randomly selected from each ovary for quantitative analysis of the immunohistochemical staining (total eight follicles per group). Values are mean ± SEM, and the mean differences were analyzed using one-way ANOVA followed by Tukey–Kramer honest significant difference test. **P < 0.01. (E) Corresponding negative control. Scale bars: 100 µm. n.s., not significant; GCs, granulosa cells; NC, negative control. We repeated experiments three times using different mice, and representative images and data are shown.

Figure 6.

A combination of dasatinib and quercetin (DQ) treatment restores the ovarian morphology of polycystic ovary syndrome (PCOS) mice. To evaluate the ovarian morphology, the mice were divided into four groups; control (n = 4), PCOS (n = 4), PCOS + vehicle (n = 4), and PCOS + DQ (n = 4). Control mice were s.c. injected with sesame oil, PCOS mice were s.c. injected with dehydroepiandrosterone (DHEA 60 mg/1 kg per day), PCOS + vehicle mice were s.c. injected with DHEA and orally administered vehicle (10% polyethylene glycol 400), and PCOS + DQ mice were s.c. injected with DHEA and orally administered DQ (D: dasatinib 5 mg/kg per day, Q: quercetin 50 mg/kg per day). Subcutaneous injection was performed daily for 20 days and oral administration was performed on Days 1, 6, 11, and 16. Ovaries were collected on Day 21. Cross-sections of ovaries were stained with hematoxylin and eosin. (A) ★ indicate atretic follicles. Scale bars: 500 µm. (B) The number of atretic follicles per ovary (four ovaries per group). Values are mean ± SEM, and the mean differences were analyzed using one-way ANOVA followed by Tukey–Kramer honest significant difference test. n.s., not significant; *P < 0.05. We repeated experiments three times using different mice, and representative images and data are shown.