Senolytic treatment fails to improve ovarian reserve or fertility in female mice

Abstract

Senescent cell number increases with age in different tissues, leading to greater senescent cell load, proinflammatory stress, and tissue dysfunction. In the current study, we tested the efficacy of senolytic drugs to reduce ovarian senescence and improve fertility in reproductive age female mice. In the first experiment, 1-month-old C57BL/6 female mice were treated every other week with D + Q (n = 24) or placebo (n = 24). At 3 and 6 months of age, female mice were mated with untreated males to evaluate pregnancy rate and litter size. In the second experiment, 6-month-old C57BL/6 female mice were treated monthly with D + Q (n = 30), fisetin (n = 30), or placebo (n = 30). Females were treated once a month until 11 months of age, then they were mated with untreated males for 30 days to evaluate pregnancy rate and litter size. In the first experiment, D + Q treatment did not affect pregnancy rate (P = 0.68), litter size (P = 0.58), or ovarian reserve (P > 0.05). Lipofuscin staining was lower in females treated with D + Q (P = 0.04), but expression of senescence genes in ovaries was similar. In the second experiment, D + Q or fisetin treatment also did not affect pregnancy rate (P = 0.37), litter size (P = 0.20), or ovarian reserve (P > 0.05). Lipofuscin staining (P = 0.008) and macrophage infiltration (P = 0.002) was lower in fisetin treated females. Overall, treatment with D + Q or fisetin did not affect ovarian reserve or fertility but did decrease some senescence markers in the ovary.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11357-024-01089-0.

Introduction

Cellular senescence is a state of cell cycle arrest where cells develop a pro-inflammatory senescence-associated secretory phenotype (SASP), which attract macrophages for clearance [1]. Cell cycle arrest in senescent cells is associated with increased expression of cyclin-dependent kinase inhibitors such as p16 and p21 [2]. Senescent cell burden increases with age in different tissues and the associated inflammation may be the cause of several age-related diseases [3] such as cancer, diabetes, atherosclerosis, and neurodegenerative diseases [4]. Despite the great number of recent studies investigating senescence and aging, literature is scarce regarding the presence and effects of senescent cells in reproductive organs.

The reproductive lifespan of a female is linked to the initial size of the ovarian reserve of primordial follicles and its depletion rate [5]. In women, as follicle numbers decrease with aging, fertility also declines, progressing through irregularities in the menstrual cycle until complete cessation of menstruation, known as menopause [6]. The decreasing number of follicles with advancing age is also accompanied by a decrease in oocyte quality [7], resulting in decreased embryo quality and higher frequency of abortions with age [8]. In C57Bl/6 mice, the ovarian reserve is severely reduced from 3 to 12 months of age [9], with the sharpest drop observed between 6 and 9 months of age. Moreover, C57Bl/6 female mice have significant fertility reduction at 10 months of age [10]. These data indicate that, although a clear menopausal transition is not observed in mice, there is a severe reduction in the ovarian reserve followed by reduced fertility, similar to what is observed in humans [7]. We previously showed that there is an increase in ovarian senescent cell markers from 3 to 12 months of age in parallel with the decreasing ovarian reserve [9] and that this burden is higher in infertile obese mice [11].

Senolytic drugs can selectively remove senescent cells [12]. Treatment of mice with dasatinib and quercetin (D + Q), as well as fisetin, promotes selective reduction of senescent cells in vivo, extending lifespan [13, 14]. Treatment with D + Q reduces SASP, and, consequently, systemic inflammation [13]. Recent studies using D + Q have also noted improvement in symptoms of age-related diseases in humans, followed by a reduction in senescent cells in the lungs and adipose tissue [15, 16]. However, there is still controversy regarding the effects of senolytics in young mice. Specifically, treatment of young females during the reproductive decline window could give us more insight about the effects of senolytics to prevent reproductive aging. Most studies have been focusing in older mice, long past the fertility window. However, recent studies suggested that D + Q can have negative effects in young female mice [17]. Additionally, the effects of senolytic drugs depend on the tissue studied and are not yet fully understood. Therefore, it is important to understand if senolytics can affect ovarian aging and improve fertility in young female mice of reproductive age. We showed before, in young leptin deficient obese mice, that treatment with D + Q decreased markers of senescence in the ovary [11]. Therefore, the aim of this study was to evaluate the effect of senolytics (D + Q and fisetin) on ovarian senescence and improve ovarian reserve and fertility in reproductive age female mice.

Materials and methods

Animals and treatment

The experiments were approved by the ethics committee on animal experimentation from UFPel under the protocol 58357-2018 and 18974-2020. Mice were kept under controlled temperature, light, and humidity (22 ± 2 °C, 12-h light cycles, and 40–60% humidity). Mice received water and a standard diet ad libitum throughout the duration of the experiment. Body weight was assessed twice a month during the experiment. Additionally, mice were observed daily for clinical signs of prostration.

The first experiment was performed with 1-month-old female C57BL/6 mice. Females were divided into two groups: D + Q group (n = 24), receiving dasatinib (5 mg/kg) and quercetin (50 mg/kg), and control group (n = 24) receiving a vehicle solution for three consecutive days every 2 weeks via oral gavage [13]. D + Q were dissolved in PHOSAL (60%, v/v), PEG 400 (30%, v/v), and ethanol (10%, v/v). The treatment persisted until females were 6 months of age. At 3 and 6 months of age, female mice were mated for a period of 7 days with 3-month-old untreated males to verify pregnancy rate and litter size. At 6 months of age, females (n = 8/group) were anesthetized and euthanized. One ovary was collected and stored at −80 °C and the other in 4% paraformaldehyde solution.

The second experiment was performed with 6-month-old female C57BL/6 mice. Mice were divided into three groups and administered: D + Q (n = 30; D = 5 mg/kg, Q = 50 mg/kg), fisetin (n = 30, 100 mg/kg) [14] diluted as described in the experiment 1, or vehicle (n = 30) via oral gavage. Mice received the treatment for three consecutive days, once a month. Treatments continued until females were 10-month-old, then females (n = 21/group) were mated for a period of 30 days with control 3-month-old male mice to verify pregnancy rate and litter size. The remaining female mice (n = 9/group) were anesthetized and euthanized, and one ovary and the liver were collected and stored at −80 °C. The other ovary was stored in 4% paraformaldehyde solution.

Histological analysis

Fixed ovaries were dehydrated in an alcohol gradient, diaphanized with xylol and embedded in paraplast (Sigma-Aldrich, Saint-Louis, MO, USA). Ovaries were then serially sectioned at 5 μm thickness using a semi-automatic microtome (Leica RM2245, Leica Biosystems Newcastle Ltd, Newcastle Upon Tyne, UK). The sections of different areas of the ovary were selected and placed on standard histological slides. After drying at 56 °C for 24 h, the slides were stained with hematoxylin–eosin and mounted with coverslips and synthetic resin (Sigma, St. Louis, MO, USA). Sections were evaluated using a digital camera coupled to a microscope (Nikon Eclipse E200, Nikon Corporation, Japan) and the software TCapture (Tucsen Photomics Co.) was used to take photos of the sections and, later the number of follicles count per section was performed. The quantified follicles were those that had a clearly visible oocyte nucleus. Follicles were classified as primordial when surrounded by a single layer of squamous granulosa cells, as primary when surrounded by a single layer of cuboid granulosa cells, as a transition follicle when surrounded by both squamous and cuboid cells, as secondary follicle when surrounded by more than one layer of granulosa cuboid cells, with no visible antrum and when the follicle had a clearly defined antral space and a layer of cumulus granulosa cells around the oocyte, it was classified as a tertiary follicle [18]. The number of follicles was calculated as an average/section/mouse [9].

Sudan Black staining

Lipofuscin staining was performed using Sudan Black dye on a subset of the histological slides from each mouse according to a previously protocol [9]. Lipofuscin is a heterogeneous pigmented by-product due to failure of intracellular catabolism, conventionally found in the lysosomes or cytosol of post-mitotic aged cells [19]. Briefly, the histological slides were dewaxed with xylol, washed in an alcohol gradient until reaching 70% alcohol. After diluting the Sudan black in 70% alcohol and using syringe with a disc filter was used to drop Sudan Black on a clean slide, after which the slide with the tissue section positioned downwards was placed over the slide with the dye (Sudan Black solution) for approximately 2 min. The slide was then washed with 50% alcohol and distilled water. A glycerol droplet and coverslip were placed over the section and observed under the light microscope. Images obtained at 10× magnification were evaluated for five mice per group. For each mouse 3 sections were used to quantify the lipofuscin positive area. The number of pixels in the images was calculated using the ImageJ software (National Institutes of Health—USA) and presented as percentage relative to the total area of the section evaluated.

Picrosirius red staining

Collagen fiber staining was performed using Picrosirius red dye in histological slides of ovary tissue from each mouse, according to previous protocol [20]. Briefly, the histological slides were dewaxed with xylol, washed in an alcohol gradient until reaching 50% alcohol and rehydrated in water. After the slides were stained using a Picrosirius solution 0.1% (Direct red 80, Sigma-Aldrich/365548 and picric acid) for 1 min and washed with acidified water (glacial acetic acid). A resin droplet and coverslip were placed over the section and observed under the light microscope. For collagen analysis, five mice per group were evaluated. For each mouse, 3 sections were observed at 100× magnification using 552 nm laser to acquire fluorescence in a confocal laser scanning microscope (Leica TCS SP8). The software ImageJ (National Institutes of Health—USA) was used to quantify the collagen positive area and presented as a fluorescence count per section evaluated.

Immunofluorescence

For immunofluorescence analysis, the ovarian samples were deparaffinized with xylene and rehydrated with gradient alcohol chain as previously described [21]. The anti-CD68 monoclonal antibody (ab955, Abcam, Cambridge, UK) was diluted in 1.5% BSA solution, and used to indicate macrophage presence [22] and used at 1:400 dilution. Anti-Lamin B1 (ab133741, Abcam) was also diluted in 1.5% BSA solution and used at 1:1000 dilution [23]. The blockage of the endogenous peroxidase activity was achieved with hydrogen peroxide blocking solution, while antigen recovery was performed in humid heat, during 3 min at boiling point in citrate solution (pH 6.0). Non-specific background staining was reduced by covering the tissue sections with 10% BSA and 7% goat serum. Slides were incubated overnight with the primary antibody in a humid chamber at 4 °C, for 1 h with secondary antibody Alexa Fluor® 488 (ab150077, Abcam) and for 3 min with DAPI (Invitrogen, Carlsbad, USA) for nuclei stanning. The slides were mounted with a drop of mounting medium (Fluoroshield, Sigma-Aldrich, St Louis, USA) under coverslips. For analysis of CD68 and LB1, five mice per group were evaluated. For each mouse image of one section was acquired at 100× magnification using a confocal laser scanning microscope (Leica TCS SP8). Sequential scanning was employed using the 488 nm laser to visualize Alexa 488 and the 405 nm-UV lasers to visualize DAPI staining. Fluorescence intensity was calculated using the software Image J® to calculate the proportion of positive staining in each section of ovary.

Gene expression analysis

Ovarian and liver RNA extraction was performed using a TRIzol protocol. The final RNA was dissolved in 20 μL RNAse-free water, and its concentration and quality were estimated by spectrophotometry (Epoch™ Microplate Spectrofotometer, BioTek, Winooski, VT, USA). Complementary DNA (cDNA) was synthesized from 1 ug of total RNA using random hexamer nucleotide (RH) primers (iScript, Biorad). After that, samples were used for real-time PCR analysis to detect the expression of target genes (Suppl. Table 1). Real-time PCR was performed using SYBR Green qPCR Master Mix (ThermoFisher, Vilnius, Lithuania) in a 20 μL reaction in a Quant studio 7 system (Applied Biosystems). Each reaction was performed in duplicate, using 2 μL of cDNA (20 ng), 10 μL of SYBR Green, 0.4 μL of each primer (10 μM) and 7.2 μL of ultrapure water. For each assay, 40 cycles (95 °C for 15 s and 60 °C for 1 min) were performed and a dissociation curve was included at the end of reaction to detect the amplification specificity of a single PCR product. The level of expression was calculated using the delta Ct method, the internal control β2m as a reference.

Statistical analysis

Statistical analyses were performed using the GraphPad Prism 6.0 software. Repeated measures ANOVA was used for evaluating body weight. For experiment 1, a t-test was used, and for experiment 2, one-way ANOVA test was used. Chi-square was used to evaluate pregnancy rate in experiment 1 and a Kaplan-Meyer survival analysis for experiment 2. P values lower than 0.05 were considered significant. Data is presented as mean ± SEM.

Results

Experiment 1—senolytics treatment from 1 to 6 months of age

Body weight

There was no difference in body weight in female mice treated from 1 to 6 months with D + Q (Suppl. Figure 1).

Ovarian reserve and fertility

There was no difference between groups for primordial (P = 0.30; Fig. 1a), transition (P = 0.25; Fig. 1b), primary (P = 0.76; Fig. 1c), secondary (P = 0.21; Fig. 1d), tertiary (P = 0.68; Fig. 1e), and total follicles (P = 0.74; Fig. 1f). There was also no effect of D + Q in the pregnancy rate at 3 and 6 months of age during the 7 days of mating (P = 1.00 and P = 0.68, respectively; Fig. 2a) or litter size (P = 0.58; Fig. 2b). We did observe an age-related decline non-significant on pregnancy rate and litter size with senolytic treatment.

Fig. 1.

Follicular count per section of six-month-old females in the control or treated (D + Q) groups. a Primordial, b transition, c primary, d secondary, e tertiary, and f total follicles. P values < 0.05 were considered statistically different

Fig. 2.

Pregnancy rate and litter size of 6-month-old females in the control or treated (D + Q) groups. Mice were mated for seven days with control 3-month-old male mice. P values < 0.05 were considered statistically different

Ovarian senescence

D + Q treated females had a lower percentage of lipofuscin staining at six months of age (P = 0.04; Fig. 3a). However, there was no difference in the relative ovarian expression of genes related to cellular senescence, SASP and inflammation between groups (P > 0.05; Fig. 4).

Fig. 3.

Percentage of lipofuscin stain of 6-month-old females in the control or treated (D + Q) groups. a Ovarian section from a control female. b Ovarian section from a D + Q treated female. P values < 0.05 were considered statistically different and represented by one asterisk

Fig. 4.

Relative expression of genes related with senescence, inflammation from ovaries of 6-month-old females in treated (D + Q) relative to control females. The black line represents the control group average for each gene and all the bars represented the D + Q group for each gene. P values < 0.05 were considered statistically different and represented by asterisk

Experiment 2—senolytic treatment from 6 to 10 months of age

Body weight

There was no difference in body weight in female mice treated from 6 to 10 months of age with D + Q and fisetin (Suppl. Figure 2).

Ovarian reserve and fertility

Similar to experiment 1, there was no difference between groups for primordial (P = 0.51; Fig. 5a), transition (P = 0.40; Fig. 5b), primary (P = 0.68; Fig. 5c), secondary (P = 0.61; Fig. 5d), tertiary (P = 0.91; Fig. 5e), and total follicles (P = 0.61; Fig. 5f).

Fig. 5.

Follicular count per section of 10-month-old control, D + Q-, and fisetin-treated females. a Primordial, b transition, c primary, d secondary, e tertiary, and f total follicles. P values < 0.05 were considered statistically different

Additionally, there was no difference in the pregnancy rate at 11 months of age during 30 days of mating (P = 0.37; Fig. 6a). Litter size was also similar between groups (P = 0.20; Fig. 6b).

Fig. 6.

Pregnancy rate (a) and litter size (b) of 10-month-old control, treated with D + Q or fisetin females. Mice were mated for 30 days with control 3-month-old male mice. P values < 0.05 were considered statistically different

Ovarian and liver senescence

Fisetin-treated females had lower lipofuscin staining compared to control females (P = 0.008; Fig. 7). Fisetin-treated females had lower macrophage infiltration than control females (P = 0.002; Fig. 7). However, there was no difference in lamin B1 staining between groups (P = 0.25; Fig. 7). Collagen staining was also not different between groups (P = 0.91; Fig. 7).

Fig. 7.

Staining of lipofuscin (a–d), collagen (e–h), macrophages (i–l), and Lamin B1 (m–p) in 10-month-old control females, treated with D + Q or fisetin. The first column of images are ovarian section from control females, the second treated with fisetin and the third treated with D + Q. The blue stain in images N to P is DAPI. P values < 0.05 were considered statistically different. **P < 0.01

Some changes were observed for senescence markers gene expression. The relative expression of Cdkn1 (P = 0.004; Fig. 8a), Serpin1 (P = 0.0002; Fig. 8a), Col1a (P = 0.03; Fig. 8a), and Efnb2 (P = 0.004; Fig. 8a) were lower in the fisetin compared to the control group. However, the relative expression of Il1b (P < 0.0001; Fig. 8a) and Col4a (P = 0.0005; Fig. 8a) were higher in the fisetin compared to the control group. The D + Q group had lower expression of Serpin1 (P = 0.0002; Fig. 8a) and Ccxl5 (P = 0.009; Fig. 8a) compared to the control group. Expression of Ccl2/Mcp1 (P = 0.03; Fig. 8a), Il1b (P < 0.0001; Fig. 8a), and Col4a (P = 0.0005; Fig. 8a) were higher in fisetin than in the D + Q group.

Fig. 8.

Relative expression of senescence and inflammation associated genes from ovaries of fisetin and dasatinib and quercetin (D + Q)-treated 10-month-old females relative to control females. The black line represents control average expression. P values < 0.05 were considered statistically different and represented by asterisk. **P < 0.001. ***P < 0.0001. ^###P < 0.0001 and difference between the fisetin and D + Q group

In the present study, we used the liver expression of senescence genes as control extra-ovarian tissue to observe the effects of senolytics. In the liver, there was no difference in the relative expression of genes related to cellular senescence, SASP and inflammation between groups for most of the genes. Expression of Col1a (P = 0.03; Suppl. Figure 3), Bcl2 (P = 0.04; Suppl. Figure 3), and Efnb2 (P = 0.03; Suppl. Figure 3) were lower in the fisetin group compared to the control group.

Discussion

In the present study, we observed that D + Q and fisetin treatment in females during reproductive age did not improve the ovarian reserve and fertility. Expression of some senescence markers decreased in the ovary after treatment, but changes were not consistent across all measured markers and suggest a weak effect at the ovarian level. The ovarian reserve is severely reduced from 3 to 12 months of age in female mice [9] and significant fertility reduction is observed at 10 months of age [10]. We choose our timeline for intervention based on this reproductive window, aiming to preserve and prevent damage to the ovarian reserve and improve fertility. However, previous studies only showed beneficial effects of senolytics using much older mice, around 19–21 months old [13, 14]. Our previous study indicated a reduction in markers of ovarian senescence in D + Q treated mice using genetically obese mice [11]. Others have shown increased senescent cell load and reduced follicle numbers and fertility in young female mice after doxorubicin treatment [24]. Mice received doxorubicin the day after the first senolytic treatment, which lasted for 3 weeks. Although D + Q and fisetin reduced senescent cell load, it did not revert follicular loss and ovarian stromal fibrosis caused by doxorubicin [24]. In a similar study, young females received the chemotherapeutical drug cisplatin 1 week after the start of a 4-week-long senolytic treatment. In this scenario, senolytic treatment prevented follicle decline, reduced ovarian DNA damage and increased the number of pups per litter [25]. This suggests that senolytics may be more effective in preventing the damage caused by chemotherapy, but less effective to reverse the damage already caused by senescence in the ovary.

Our data suggests that the use of senolytics in normal young female mice of reproductive age does not benefit the ovarian reserve and fertility. Therefore, senolytics may only be efficient in young mice when senescence was induced by stressors such as chemotherapy [25], severe obesity [11], aging progeroid models [14], or neuropathies such as Alzheimer [26]. This may be related to relatively small accumulation of senescent cells in ovaries of young female mice or possible physiological role of these cells in reproductive functions. In addition, studies have suggested that senolytics promote the elimination of senescent cells or decrease cell senescence [14], but there is no evidence that senolytics can prevent cells from entering a senescent state. Therefore, in order to protect the ovarian reserve is important to also reduce the inflammation generated by the senescent cells through SASP secretion. Increased inflammation generated by proinflammatory mediators increase the activation of primordial follicles [27]. However, this is something we did not observe in the current study with the treatment protocol used, suggesting a different treatment protocol is needed for reproductive longevity. Additionally, the ovulatory process itself is preceded by inflammatory reactions that occur in mature follicles [28]. Therefore, inflammation and tissue remodeling is present in the ovaries at all stages of aging, making it difficult to identify the physiological and pathological roles of senescent cells in this context.

A recent study treated young (4–13 months of age) female and male mice with D + Q and fisetin (the same scheme used in our experiment 2) and observed negligible and even harmful effects of senolytics [17]. Fisetin, however, improved several metabolic parameters in young male mice, while D + Q had minimal effects [17]. SASP markers were elevated in males compared to females and may explain the sexual dimorphic effects [17]. In young females, D + Q had negative effects, increasing body weight gain, fat accumulation, and expression of senescence markers [17]. These findings are in line with our current observations, indicating that D + Q has no effects in young female health and can even be harmful. This indicates that, although removal of senescent cells is good in old age or disease, disrupting physiological role of senescent cells at young age might be harmful. Further confirming this sexual dimorphic effect, our previous study showed a positive effect of D + Q in young male mice in testosterone levels, sperm concentration, and sperm morphology [29]. These findings suggest that age and sex are key determinants in the efficacy of senolytics. In the present study, we observed that fisetin decreased lipofuscin accumulation, macrophage infiltration and decreased ovarian expression of Cdkn1, Serpin1, and Col1a. D + Q reduced lipofuscin accumulation in the first study using younger females only. A previous study in young females at the same age window also showed that fisetin treatment had beneficial effects in some senescence marker while D + Q did not [18]. These subtle positive effects in ovarian senescence were not sufficient to improve pregnancy rate and litter size in our study. Expression of some ovarian senescence markers such as Mcp1, Il1b, and Col4a were increased after fisetin treatment, indicating a possible negative effect, which can decrease therapeutical effects. The increased Col4a expression can indicate fibrosis resulting from continuous removal of senescent cells from a young age [30]. Interestingly, expression of Efnb2 and Bcl2 in liver was decreased after fisetin treatment. Efnb2 is an apoptosis inhibitor [31] and Bcl2 is an antiapoptotic protein [31]. Fisetin is known to target the anti-apoptotic pathways in senescent cells such as Bcl-2 and Bcl-xl [32]. Moreover, the reduced expression of these genes indicates that fisetin was reaching therapeutical potential, activating genes that promote senescent cell removal.

Based on this context, we can observe that fisetin decreased some senescence markers in the ovary; however, it did not result in improved ovarian reserve or fertility. There are few studies evaluating the effect of senolytics for longer periods in young mice. Therefore, it is important to evaluate the effect of longevity interventions in young females to uncover treatments that can prevent or delay the decline in female reproductive potential. Successful treatments for reproductive aging must be safe to use long term in young mice. The present study suggests that senolytics cannot improve fertility or ovarian reserve in females during reproductive age, although some senescence markers were improved. More studies are needed to better understand the role of senescent cells in ovaries, focusing on the prevention of senescent cell accumulation and SASP secretion to prevent declines in ovarian reserve and fertility. In addition, lack of effect of senolytics in our study may be related to the physiological role of these cells in reproductive functions, as inflammation and tissue remodeling are key points in the ovulatory process in females. For future studies, it is necessary to define how to improve the use of senotherapy in reproductive aging and define best frequency, dosage, drug class and starting age of therapies aiming at ovarian senescence. In addition, a better understanding of ovarian senescence biomarkers itself is needed in order to predict treatment efficacy.

Conclusion

The treatment with D + Q or fisetin in females during reproductive age did not affect the ovarian reserve and fertility. Fisetin treatment decreased macrophage accumulation and some senescence markers in the ovary, while D + Q effects were negligible in females during reproductive age.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contribution

DNG performed experiments, data interpretation, analysis, and wrote the manuscript. JDH, BMZ, JVVI, JP, SA, TL, and ICS assisted in experiments, data analysis, and manuscript revision. JBM, MMM, CCB, and MBS assisted in experimental design, data interpretation, and manuscript revision. AS performed data analysis, interpretation, and manuscript preparation.

Funding

Research reported in this publication was supported by CAPES, CNPq, and FAPERGS to A. S., the National Institute on Aging of the National Institutes of Health under Award Number R56AG074499 to M.M.M.

Data availability

Data will be available from the corresponding author upon request.

Declarations

Competing interests

The authors declare no competing interests.