Introduction
Premature ovarian insufficiency (POI), diagnosed mainly by elevated FSH (> 40 IU/L) and amenorrhea before 40 years old, affects 1 in 100 females[
1]. Due to very poor ovarian reserve, it is difficult for females with POI to become pregnant naturally or with the aid of assisted reproductive technology. In addition to infertility, POI also increases other health risks, including osteoporosis, cardiovascular disease, and earlier mortality [
2‐
4]. Several causes of POI have been reported, including genetic alterations, ovarian surgery, radio- or chemotherapy, environmental factors, viral infections, and metabolic and autoimmune diseases [
5]. Of them, chemical agents such as alkylating agents (AAs) that are commonly used for the treatment of cancer cause great damage to the ovaries, thereby significantly increasing the risk of POI [
6]. However, the exact molecules underlying AA-induced POI remain largely obscure.
p16, a well-known cell cycle inhibitor encoded by the CDKN2A gene locus (which also encodes p19), plays an important role in arresting the cell cycle and maintaining the state of cell senescence [
7,
8]. p16 has been widely used as a biomarker reflecting cell senescence [
9]. Targeting p16 has been demonstrated to be an effective way to lengthen the healthy lifespan and prevent multiple diseases in mice, including emphysema, tubulointerstitial injury and osteoporosis [
10‐
13]. Recent studies have indicated a role for p16 in promoting the development of POI. It has been observed that p16 is significantly upregulated in the ovaries of several mouse models of POI, including a natural ovarian ageing mouse model and cyclophosphamide-, d-galactose- and consecutive superovulation-induced POI mouse models [
14‐
18]. The agents that were reported to exert a protective effect against ovarian insufficiency, including metformin, moxibustion and curcumin, could significantly downregulate the expression of p16 in the ovaries [
14,
15,
17]. In addition, a study conducted by Xiong et al. indicated that p16 was the ultimate effector that mediated cyclophosphamide-induced ovarian failure in mice [
16]. These studies suggest that the upregulation of p16 may be a universal mechanism involved in promoting POI, including AA-induced POI.
However, the abovementioned studies indicated rather than demonstrated a role of p16 in promoting POI. As yet, no in vivo data from p16-deficient (KO) mice are available to demonstrate a critical role of p16 in POI. Without this evidence from KO mice, it remains unknown whether p16 could be a target for preserving ovarian function and fertility in females treated with chemical agents.
To address this issue, we employed p16-deficient female mice in the present study. p16 KO and wild-type (WT) female mice were treated with busulfan (BUL) in combination with cyclophosphamide (CTX) to establish a POI mouse model as previously described [
19]. The reproductive parameters, including oestrous cycles, hormone levels, fertility, and follicle counts, were measured. In addition, cell proliferation and apoptosis and ovarian stromal vessels and fibrosis were compared among WT and KO mice treated with/without BUL + CTX.
Materials and Methods
Animals and treatment
All mice were housed in an SPF laboratory characterized by a 12 h light/dark cycle, 23 ± 1 °C and 40–60% humidity, with free access to food and water. p16 heterozygous (p16
+/−) mice on FVB N2 background were provided by Dr. Dengshun Miao from Nanjing Medical University. Male and female p16
+/− were mated to generate WT and p16 KO mice for further experiments. Eight-week-old WT and p16 KO mice received a single dose of BUL (Sigma, China, 20 mg/kg) and CTX (Selleck, China, 120 mg/kg) by intraperitoneal injection as previously described [
19]. One month after administering AAs, the oestrous cycle for each mouse was monitored for 34 consecutive days. Three months later, except for 6 mice in each group for the fertility test, the remaining mice were sacrificed by cervical dislocation. Serum and ovaries were collected for further analysis. Twenty-four hours before sacrifice, 3 mice in each group were injected with BrdU solution (Sigma, China) at a dose of 50 mg/kg. All animal procedures and experiments were approved by the ethics committee of Changzhou Maternal and Healthy Care Hospital.
Fertility test
One female mouse and one WT fertile male mouse was mated in one cage for one month. Fertile male mice were confirmed by mating experiments. After one month or female pregnancy, the male mouse was removed from the cage. Female mice without pregnancy were observed for another month. The pregnant mice and litter size were recorded.
Oestrous cycle detection
A vaginal smear was performed at approximately 9:00 am. Briefly, 10 µL of saline was pushed into the vagina and drawn with a micropipette 3 times. Then, saline containing exfoliated vaginal cells was smeared on a slide. After air drying, the smear was fixed with 75% ethanol. HE staining was performed to evaluate the cell types. The oestrous cycle was read and recorded for each mouse.
Hormone level assay
Serum levels of AMH, FSH, LH and E2 were determined by ELISA kits. ELISA kits including anti-FSH (RJ-17,024), anti-E2 (RJ-17,014) and anti-LH (RJ17209) were used. All ELISA kits were purchased from Shanghai Renjie Biological Company. All procedures were performed strictly according to the instructions provided by the manufacturer.
Follicle counts
Paraffin-embedded ovaries were serially sectioned. The first section containing ovarian tissue was collected. Every six sections, another section was collected until 20 sections were collected. HE staining was performed to evaluate the morphology of follicles. The criteria for the classification of follicles are as follows: primordial follicle, the oocyte was enclosed by a layer of squamous granulosa cells; primary follicle, the oocyte was enclosed by cuboidal granulosa cells; secondary follicle, the oocyte was enclosed by 2 or more layers of granulosa cells; and antral follicle, an antrum cavity was present. We avoided recording the same follicle more than once.
Immunohistochemistry
After dewaxing and hydration, ovarian sections were immersed in citrate solution for antigen retrieval by a high-pressure method. After naturally cooling down the citrate solution, the slides were removed, and ovarian sections were incubated in 3% H2O2 for half an hour at room temperature (RT). Then, ovarian sections were incubated in 10% donkey serum for 1 h at RT. After this, ovarian sections were incubated with primary antibodies, including anti-p16 (YM0494, Immunoway, China), anti-cleaved caspase 3 (GB11532, Servicebio, China), anti-αSMA (67735-1, Proteintech, China), anti-CD31 (#77699, Cell Signaling Technology, China) and anti-BrdU antibodies (B2531, Millipore, China), at 4 °C overnight. Then, ovarian sections were incubated with HRP-labelled goat anti-mouse and anti-rabbit antibodies (A0208 and A0216, Beyotime Biotechnology, China) for 1 h at RT. Positive staining was visualized by DAB staining. For BrdU staining, additional steps were needed. After dewaxing and hydration, the sections were incubated with diluted hydrochloric acid (1:5; v:v) for 30 min at RT, followed by direct incubation with a boric acid solution for 20 min at RT. The remaining steps were the same as those of conventional immunohistochemistry.
Picrosirius red staining (PRS)
After dewaxing and hydration, ovarian sections were stained with picrosirius red solution (R21890, Saint-bio, China) at RT for 1 h. Then, ovarian sections were dehydrated using graded ethanol and became transparent by xylene. The slides were mounted, and photos were taken.
Statistical analysis
All values are presented as the mean ± SD. One-way ANOVA followed by post hoc Tukey’s honestly significant difference test was used to compare data among groups. SPSS software (ver. 18.0; SPSS, Inc., Chicago, IL, USA) was used for the statistical analyses.
Discussion
This is the first study to explore the role of the p16 gene in the development of AA-induced POI by using p16-deficient mice. We found that genetic ablation of the p16 gene had no impact on the ovarian reserve, function or fertility of mice. More importantly, we demonstrated that genetic ablation of the p16 gene did not attenuate ovarian damage or help preserve the fertility of mice challenged by AAs. This study demonstrated for the first time a dispensable role of p16 in POI caused by AAs. Our preliminary findings suggest that targeting p16 alone may not work for preserving the ovarian reserve and fertility of females with POI caused by AAs.
Primordial follicles that initiate folliculogenesis are the foundation of female fertility [
24]. A previous study showed that physiologically, p16 was strongly expressed in primordial follicles but weakly expressed in growing follicles, indicating that the downregulation of p16 may initiate the growth of primordial follicles [
25]. However, in the present study, we found that p16-deficient mice did not show an apparently reduced number of primordial follicles compared to WT mice. Therefore, p16 may not play a key role in attenuating the growth of primordial follicles under physiological conditions. However, it is possible that p16 may help prevent the overactivation of primordial follicles under stress. It has been reported that AAs induce depletion of primordial follicles by overactivation of primordial germ cells [
26]. The upregulation of p16 has been observed in the ovaries of mice with POI induced by AAs [
27]. Here, we found that p16 deficiency did not lead to a decreased number of primordial follicles in response to AAs compared to WT mice. Therefore, we concluded that the p16 gene is dispensable for initiating the growth of primordial follicles under either physiological conditions or stress conditions caused by AAs.
Due to the high proliferation activity of granulosa cells, growing follicles are more sensitive to the toxicity of AAs, which can induce cell death by direct intercrossing of DNA double strands. In the present study, a single dose of AAs was given, and the direct killing effect of AAs on growing follicles in WT or p16-deficient mice should be similar at the initial stage. The time point for follicle counts in the present study was 3 months after administering AAs. Undoubtedly, the growing follicles within ovaries treated with AAs at that time point originated from the surviving primordial follicles. The normal function of GCs is crucial for the growth of follicles [
28]. Several studies have indicated that the insufficient proliferation of granulosa cells may be associated with the occurrence of POI, indicating that increasing the proliferation of granulosa cells may help to prevent the occurrence of POI [
29‐
32]. It has been observed that p16 is significantly upregulated in granulosa cells in mice with POI [
17,
33]. In vitro data indicated that p16 plays a role in decreasing the proliferation of cultured granulosa cells [
16,
34]. Therefore, we expected a beneficial role of p16 deficiency in preventing POI by promoting the proliferation of granulosa cells in mice treated with AAs. Unexpectedly, we found that the number of proliferating granulosa cells within growing follicles (normal appearance and nonatretic follicles) was similar between WT and p16-deficient mice treated with or without AAs. In addition, it seems that these growing follicles in WT and KO mice treated with AAs were relatively healthy without obvious apoptosis of granulosa cells. Although “normally developed follicles” existed, the fertility test showed that some WT and KO female mice treated with BUL + CTX were infertile, indicating a poor quality of oocytes. Consistently, p16 deficiency did not increase the number of growing follicles, including antral follicles, or reduce the number of atretic follicles in mice treated with AAs. More importantly, p16 deficiency did not increase the litter size of female mice treated with AAs. These results indicated that p16 is dispensable for the growth of follicles in the ovaries of mice treated with AAs.
As a well-known cell cycle inhibitor and a widely accepted marker for cell senescence, the expression of p16 parallels the progression of ageing and ageing-related disease in multiple organs [
35‐
37]. A large number of studies have demonstrated that targeting p16 could effectively attenuate the progression of ageing and ageing-related disorders [
38‐
40]. However, the expression of p16 is not always detrimental. Issac et al. reported that the expression of p16 was significantly induced in the lung tissue of a mouse model of chronic obstructive pulmonary disease (COPD); however, genetic ablation of p16 could not prevent cellular senescence or alleviate the symptoms of COPD, indicating that p16 alone is dispensable for the development of COPD induced by chronic smoking [
41]. The protective role of p16 has also been demonstrated by several studies. Studies conducted by Lv et al. have demonstrated that genetic ablation of p16 promotes liver fibrosis in mice induced by CCL
4 or a methionine- and choline-deficient diet [
42,
43]. Another study showed that deletion of p16 shortened the lifespan and accelerated the disorders of multiple organs of mice bearing homozygous mutations of Pot1b, which plays a critical role in stabilizing the structure of telomeres [
44]. A recent study showed that removal of p16
− high senescent cells in aged mice is detrimental to the lifespan [
45]. Therefore, the detrimental/protective role of p16 may depend on the type of disease and disease context. The present study showed for the first time a dispensable role of p16 in folliculogenesis following treatment with AAs. Other cell cycle inhibitors were also upregulated in the ovaries of mice with POI, such as p21 [
46,
47]. Recent studies have highlighted a critical role of p21 in the progression of some diseases. Genetic ablation of p21 has been reported to improve the symptoms of COPD induced by chronic smoking [
48]. Similarly, genetic ablation of p21 also prevents liver fibrosis induced by CCL
4 [
49]. In addition, ablation of p21, not p16, could lengthen the lifespan and improve the organ disorder of mice bearing homozygous mutation of Pot1b [
44]. Therefore, it is possible that p21 may play a critical role in mediating the occurrence of POI caused by AAs. This will be further investigated in the future by employing p21-deficient mice.
In addition to defective folliculogenesis, ovarian stromal abnormalities, including ovarian stromal fibrosis and damaged vessels, are also common following treatment with chemical drugs [
21]. It has been reported that p16 may play a role in promoting CCl4-induced liver fibrosis. Therefore, we speculated that p16 may play a role in regulating fibrosis in the ovary following treatment with AAs. However, in the present study, no significant difference in ovarian fibrosis was observed among groups. Our data indicated that treatment with AAs did not apparently induce the formation of ovarian fibrosis. Patients with cancer showing obvious ovarian fibrosis receive constant treatment with chemical drugs. In contrast, in the present study, the mice received only a single dose of AAs. It is possible that a single dose of AAs may not effectively induce fibrosis. Furthermore, at the age of approximately 18 months (the end of reproductive life), ovarian fibrosis was obvious in naturally aged mice [
22]. However, in the present study, the female mice treated with AAs were approximately 6 months old, and most of them were still fertile. The time point set in the present study may not be appropriate for exploring ovarian fibrosis. Consistent with previous studies [
23,
50], a reduced ovarian vascularized area of mice treated with AAs was also observed in the present study. However, p16 deficiency had no impact on the ovarian vascularized area caused by AAs. Collectively, we demonstrated that p16 deficiency had no impact on ovarian stromal abnormalities caused by AAs.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.