Increased cellular senescence in doxorubicin-induced murine ovarian injury: effect of senolytics

Abstract

Ovarian injury caused by chemotherapy can lead to early menopause, infertility, and even premature senility in female cancer patients, impairing the quality of life and overall health of the cancer survivors seriously. However, there is still a lack of effective protection strategies against such injury. Cellular senescence can be induced by chemotherapeutic agents in multiple organs and may corrode the structure and function of normal tissues. We hypothesized that the widely used first-line chemotherapy drug, doxorubicin, could increase senescent cell burden in normal ovarian tissue during the therapeutic process and that elimination of senescent cells with senolytics would ameliorate doxorubicin-induced ovarian injury. Here, we demonstrated an accumulation of cellular senescence in doxorubicin-treated ovaries through detecting p16 and p21 expression levels and senescence-associated β-galactosidase (SA-β-gal) activity as well as senescence-associated secretory phenotype (SASP) factors. Short-term intervention with the classic senolytic combination dasatinib and quercetin (DQ) or fisetin significantly reduced the load of senescent cells in ovaries after doxorubicin treatment. However, neither DQ nor fisetin alleviated doxorubicin-related ovarian dysfunction. Further experiments showed that ovarian apoptosis and fibrosis following doxorubicin exposure could not be improved by senolytics. Collectively, our study shows that senolytic treatment can eliminate accumulated senescent cells, but cannot reverse the massive follicle loss and ovarian stromal fibrosis caused by doxorubicin, suggesting that cellular senescence may not be one of the key mechanisms in doxorubicin-induced ovarian injury.

Keywords: Apoptosis; Cellular senescence; Doxorubicin; Fibrosis; Ovarian injury; Senolytics.

Fig. 1

Ovarian senescence is increased in doxorubicin-treated mice. a Schematic diagram for animal experiment design. b–d Body weight, ovary weight, and ovary index of mice after saline or doxorubicin injection (n = 16, unpaired Student’s t-test). e The proportion of regular or irregular estrous cycles after saline or doxorubicin injection (n = 16, Chi-square test). f The serum levels of AMH, FSH, and E2 at diestrus (n = 6, unpaired Student’s t-test). g The proportion of follicles at different stages (PMF: primordial follicles, GF: growing follicles, ATF: atretic follicles). h Representative H&E staining images of murine ovaries. Yellow arrows: primordial follicles, red arrows: primary follicles, blue arrows: secondary follicles, black arrows: zona pellucida remnants. ANF: antral follicles. i Follicle counting results based on H&E staining of serial ovarian sections (n ≥ 5, unpaired Student’s t-test). j Representative images of SA-β-gal staining and IHC detection of p16 and p21 in the murine ovaries. k IHC scores of relative expression of p16 and p21 based on IOD (n = 3, unpaired Student’s t-test). IOD: integrated optical density. l, m Protein expression of p16 and p21 in ovarian tissues detected by western blot. n Relative mRNA expression of genes related to cellular senescence and SASP (n = 3, unpaired Student’s t-test). Data are presented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 2

Senolytics cannot alleviate doxorubicin-induced ovarian injury. a Brief schematic diagram for animal experiment design. b–d Body weight, ovary weight, and ovary index of mice in each group (n ≥ 16, one-way ANOVA). e The proportion of regular or irregular estrous cycles in each group (n ≥ 18, Chi-square test). f The serum levels of AMH, FSH, and E2 at diestrus (n = 6, one-way ANOVA). g Representative H&E staining images of murine ovaries from each group. Yellow arrows: primordial follicles, red arrows: primary follicles, blue arrows: secondary follicles, black arrows: zona pellucida remnants. ANF: antral follicles, ATF: atretic follicles. h Follicle counting results according to H&E staining of serial ovarian sections (n ≥ 5, one-way ANOVA). i The proportion of follicles at different stages (PMF: primordial follicles, GF: growing follicles). j The proportion of successful pregnant mice in the mating test (n = 12, Chi-square test). k Average litter size of total mated mice (n = 12, one-way ANOVA). l Average litter size of post-delivery mice (n = 3~11, one-way ANOVA). m General state of pups in different groups. Data are presented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 3

Senolytics can reduce ovarian senescence in doxorubicin-treated mice. a Representative SA-β-gal staining images of murine ovarian sections. b, c Representative images of p16 and p21 expression in murine ovaries by IHC. d, e IHC scores of relative expression of p16 and p21 based on IOD (n = 3, one-way ANOVA). f, g Protein expression of p16 and p21 in ovarian tissues shown by western blot. h Relative mRNA expression levels of genes related to cellular senescence and SASP in ovaries (n = 3, one-way ANOVA). Data are presented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 4

Senolytics have no protective effect against ovarian apoptosis after doxorubicin treatment. a, c Expression of DNA damage marker γH2AX in murine ovaries detected by IHC (n = 3, one-way ANOVA). b Representative TUNEL staining images of ovarian sections. Green fluorescent dots indicate TUNEL-positive apoptotic cells. d–f Relative mRNA expression of apoptosis-related genes in ovaries (n = 3, one-way ANOVA). Data are presented as mean ± SD. *p < 0.05, **p < 0.01

Fig. 5

Senolytics cannot prevent ovarian fibrosis after doxorubicin treatment. a Immunofluorescence images of ovarian sections stained with α-SMA antibody. Red fluorescence indicates α-SMA-positive cells. b Representative images of Sirius red staining in murine ovaries. c–d The percentage of α-SMA positive cells and Sirius red positive area in ovarian sections (n = 3, one-way ANOVA). e Relative mRNA expression of fibrosis-related genes in ovaries (n = 3, one-way ANOVA). Data are presented as mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001