Regulation of Follicular Atresia by WIP1-Mediated Apoptosis and Autophagy

Female endocrine homeostasis and reproductive success depend on the number and quality of 14 follicles. The follicle is the basic functional unit within mammalian ovaries. Excessive follicular 15 atresia is responsible for the accelerated ovarian aging process. Therefore, exploring the molecular 16 mechanism of follicle development and atresia is essential for protecting ovarian function. In this study, we interrogate the striking correlation between follicular atresia and wild-type p53-induced 18 phosphatase 1 (WIP1) expression in mouse ovaries to understand how WIP1 phosphatase activity 19 regulates follicle development. WIP1 is mainly expressed in granulosa cells of healthy growing 20 follicles, and atretic follicles exhibit significantly weaker WIP1 expression compared with the 21 healthy ones. Our in vivo study indicates that inhibition of WIP1 phosphatase activity causes endocrine disorder, fertility decline and decreased ovarian reserve by triggering excessive follicular atresia through promoting autophagy and apoptosis. By in vitro follicle culture, we determine that 2 inhibiting the WIP1 activity impairs the follicle development, causing more follicular atresia and 3 decreased oocyte quality. Besides, downregulating WIP1 expression in granulosa cells in vitro also 4 promotes apoptosis and autophagy via WIP1-p53 and WIP1-mTOR signal pathway. Our findings 5 from the in vitro and in vivo experiments revealed that appropriate Wip1 expression is required for 6 follicle development. Downregulation of WIP1 expression accelerates follicle atresia via WIP1-p53 7 and WIP1-mTOR signal pathway related apoptosis and autophagy. It is speculated that moderate 8 up-regulation of WIP1 expression may help delaying the decline of ovarian reserve.


Introduction
The age-related decline in ovarian function is the major challenge in women's reproductive health. 12 The decline in endocrine function or reproductivity is closely related to the quality and quantity of 13 follicles. The decline of the ovarian follicular reserve is nonlinear and seems to accelerate with age 14 [1,2] . The mechanisms of the accelerated depletion of follicles in later reproductive ages remain to 15 be elucidated. Exploring the mechanism may help developing targeted treatment strategies to slow 16 down the depletion of ovarian follicles and delay ovarian aging. Moreover, menopause related 17 disease such as cardiovascular disease [3], osteoporosis [4], neurodegenerative diseases [5], etc. 18 could also be improved. 19 The healthy follicular development and oocyte maturation are the prerequisite for endocrine 20 function and obtaining high-quality oocytes. Follicular atresia is also a survival of the fittest 21 physiological phenomenon that occurs throughout the ovarian life. Abnormalities in follicular 22 development and atresia will inevitably lead to abnormal ovarian function [6,7]. Both external and 23 3 internal factors induce cell DNA damage. The DNA double-strand breaks can alter genetic integrity 1 and cause deleterious damage to cells. Previous studies have demonstrated that decreased oocyte 2 quality in older individuals is closely related to increased DNA damage [8]. DNA damage in 3 granulosa cells also accumulates with age [9]. If the DNA damage cannot be repaired, cells undergo 4 programmed cell death or senescence to avoid severe mutagenic consequences [10]. 5 The size of the follicle pool and the depletion rate of primordial follicles determine the ovarian 6 functional lifespan [11]. Apoptosis is regarded as the main mechanism of the follicular atresia, 7 whereas the follicle destiny is determined by the balance between proapoptotic and prosurvival 8 molecules [7,[11][12][13] Recently, a few studies indicate that autophagy also participates in the 9 regulation of follicle development [14,15]. However, the molecular mechanisms that regulate 10 follicular development and the ovarian aging still need to be explored. 11 Wip1, coded by Ppm1d, is a Ser/Thr protein phosphatase of PP2C family. A number of 12 proteins, regulating the DNA damage response and cellular checkpoint pathways, could be 13 dephosphorylated by Wip1 [16]. Previous studies suggested that Wip1 was highly expressed in 14 various tumor tissues [17][18][19][20][21][22]. In recent years, more and more researches pay attention to the 15 regulatory role of WIP1 in aging [23][24][25][26] [29]. While, the regulatory role of WIP1 in ovarian function remains to be further clarified. This 7 study was aimed to explore the specific role and mechanism of WIP1 phosphatase in regulating 8 ovarian follicular development in adult mice. Our objective was to investigate the possible role of 9 WIP1 and its related signal pathways in ovaries to provide guidance for finding strategies to delay 10 ovarian aging.

12
Postnatal day (PND) 14 and PND21 C57BL/6j female mice, purchased from the Center for Disease 13 and Prevention of Hubei Province (Wuhan, China), were sacrificed for primary cell or organ 14 culture. Thirty-six female mice (C57BL/6j, 6 weeks old, purchased from Beijing Huafukang Bio-15 Technology Co., Ltd.) were fed freely for one week to acclimate to the environment. All animals 16 were fed with standard chow and water ad libitum under appropriate temperature and humidity. All  The mice were divided into three groups (with 12 female C57BL/6J mice in each group):

19
Mating test was carried out after two weeks' estrous cycle monitoring as showed in Fig.1A. The  Serum AMH level by ELISA 22 Blood samples were collected after two weeks' GSK2830371 treatment. The enzyme-linked 23 6 immune sorbent assay (ELISA) Kit (Bioss) was used to detect the serum AMH levels according to 1 the manufacturer's instructions.

2
Follicle counting 3 The ovary sections harvest after two weeks' GSK2830371 treatment were stained with hematoxylin 4 and eosin for follicle counting as previously described [35]. Briefly, every four slide was observed 5 under the microscope and only follicles with oocytes were counted. Two persons counted 6 independently and the counting results were analyzed together.  The methods of follicle isolation and culture were described previously [36,37]. Briefly, the PND14 7 mouse ovaries were removed and the follicles were mechanically dissected with syringe needles.

12
The follicle culture medium was α-minimal essential medium (Life technologies corporation, were cultured at 37°C in 95% air-5% CO2 for 6 days.   Cell proliferation detection by EdU 22 The primary granulosa cell proliferation was analyzed using the Cell-Light EdU Apollo643 In Vitro  All data are expressed as Means (SD) unless noted otherwise. Statistical comparisons were 10 performed using unpaired Student's t tests, one-way ANOVA or Chi-square test. Statistical analysis 11 was performed by SPSS (version 13.0). The significance level was set at p < 0.05.

WIP1 expression in Follicles of Different Stages in Mouse Ovaries.
14 The IHC analysis in 6-week-old mouse ovaries showed that WIP1 is mainly expressed in oocytes 15 and granulosa cells, and the stromal cells exhibited weaker WIP1 expression. WIP1 was highly 16 expressed in the granulosa cells of healthy follicles, while WIP1 expression was significantly 17 downregulated in the granulosa cells of atretic follicles (Fig.1A). The relative WIP1 expression 18 levels in follicles of healthy growing follicles (secondary follicles and antral follicles) and atretic 19 follicles were showed in Figure1B, the atretic follicles showed significant weaker WIP1 expression 20 compared with the healthy growing follicles.

21
Inhibition of WIP1 Activity Led to Endocrine Disorder and Declined Reproductivity. 22 To confirm the role of WIP1 in ovarian function and its mechanism, the in vivo WIP1 inhibitor 23 (GSK2830371) intervention experiment was carried out as showed in Fig.2A. The body weight of 1 mice in the three groups showed no significant difference (Fig.2B). In the GSK-15 group, 2 significantly fewer mice (20%) showed regular estrous cycles compared with Veh group (p < 0.05) 3 (Fig.2C). We observed that fewer mice became pregnant in the GSK groups than the Veh group (p < 4 0.05) (Fig.2D). The average litter size of total mated mice in GSK groups was lower than that of the 5 Veh group with no significant difference (p > 0.05) (Fig.2E). 6 Inhibition of WIP1 Activity Led to Declined Ovarian Reserve in Mice. 7 The ovary index (ratio of ovary weight to body weight) also showed no significant difference 8 among the three groups (Fig.2F). The serum AMH level decreased significantly in the GSK-15 9 group compared with Veh group, while the GSK-7.5group showed no significant difference 10 (Fig.2G). The serial sections of ovaries were stained with H&E for evaluation of ovarian reserve 11 and follicular development (Fig.2H). The GSK2830371 treated mice exhibit fewer healthy follicles 12 and more atretic follicles. The number of secondary follicles and antral follicles in the GSK-7.5 13 group was significantly lower than that in the Veh group (*P < 0.05). In the GSK-15 group, the 14 number of primordial follicles was significantly lower than that in Veh group (*P < 0.05), while the 15 number of atretic follicles was significantly higher than that in Veh group (*P < 0.05) (Fig.2I). The 16 total healthy follicles in both intervention groups decreased significantly compared with the Veh 17 group (*P < 0.05). According to the follicle counting results, we calculated the proportion of 18 primordial, growing (containing primary, secondary and antral follicles) and atretic follicles 19 (Fig.2I). The primordial follicle proportion decreased and the atretic follicles increased significantly 20 in the GSK-15 group compared with the Veh group (*P < 0.05), and the percentage of growing 21 follicles in the total number of follicles increased slightly (Fig.2J). According to the statistics of 22 follicle counting, we found that inhibiting the activity of WIP1 phosphatase in mice could 23 11 accelerate the depletion of the primordial follicle pool and the occurrence of follicular atresia.

The WIP1 related Mitochondrial Apoptosis and Autophagy Promoted the Follicle Atresia.
2 Current studies have revealed that granulosa cell apoptosis is widely considered as the underlying 3 mechanism of follicular atresia [7,12]. WIP1 has also been proved to be involved in regulating the 4 cell apoptosis [39,40]. The TUNEL results demonstrated that the more apoptotic granulosa cells 5 appeared in GSK group ovaries than that in the Veh group (Fig.3A-B). The key proteins involved in 6 the DNA damage and the apoptosis process were detected by Western blot, and the results showed a 7 significant increase in γ-H2AX, p-ATM (S1981), p-p53 (S15), BAX and cleaved-Caspase3 8 expression in ovaries of GSK groups compared with the Veh group ( Fig.3C-D), while the anti-9 apoptosis protein Bcl-2 decreased significantly in the GSK groups.

10
Autophagy, another important form of cell death, has also been proved to be involved in 11 follicular atresia [14]. Autophagy is also highly regulated by a range of protein phosphorylation and 12 dephosphorylation [41,42]. WIP1, as a phosphatase, has also been reported be involved in 13 regulating the autophagy process in atherosclerosis [43,44]. mTOR signal pathway is a key 14 regulator of autophagy [45]. To explore the autophagy in the mouse ovaries, we detected LC3B by 15 IHC, and the results indicated that the LC3B expression, mainly in follicle granulosa cells, was 16 significant stronger in the follicles of GSK groups compared with the Veh group (Fig.4A-B), 17 indicating the autophagy activation in the follicles is one of the causes of follicular atresia. While 18 the SQSTM1 expression in follicle granulosa cells was relatively weaker. The SQSTM1 was highly 19 expressed in ovarian stroma (Fig.4C). The total SQSTM1 expression in GSK groups was 20 significantly higher than that in Veh group (Fig.4D), which indicated that the autophagy in the 21 ovarian stroma was inhibited by the GSK treatment. Western blot results showed the activation of 22 mTOR signal pathway in GSK groups and a slight increase of SQSTM1 and LC3B II in the GSK-23 12 7.5 group (Fig.4E-F). Collectively, the levels of autophagy were not consistent in different parts of 1 ovaries.  To explore the effect of WIP1 on preantral follicle growth, the follicle morphology was observed 7 and recorded daily (Fig.5A). The analysis of follicular diameter suggested that GSK2830371 was detected by Ki67 staining, and follicles in the GSK group showed a reduction in Ki67 12 fluorescence intensity (Fig.5C). The TUNEL assay was used to compare the apoptosis in follicles 13 after 4 days culture. The relative quantitative of TUNEL-positive GCs (staining with green) of the 14 GSK group follicles significantly increased compared with the control group (Fig.5D).

15
After 6 days' culture, the oocyte nuclear maturity was assessed to reflect the oocyte quality 16 (Fig.5E). As shown in Fig.5F, the proportion of degenerated oocytes significantly increased in the 17 GSK group, and the oocytes in GVBD and MII phase declined compared with the control group (P 18 < 0.05). The results indicate inhibition of phosphatase activity can limit follicle growth and oocyte 19 maturation, and the decreased proliferation and increased apoptosis granulosa cells induced by 20 WIP1 inhibitor attribute to it.

21
Downregulation of WIP1 Promoted the Mouse Granulosa Cell Autophagy and Apoptosis. 22 The follicle development is closely related to the granulosa cells viability. To verify the relationship 23 13 between WIP1 expression and granulosa cell proliferation and apoptosis, primary mouse granulosa 1 cells were isolated and cultured for 24 h and then transfected with Si-ppm1d-RNA. The 2 morphology of granulosa cells was observed and recorded (Fig.6A). After transfection 48 h, EdU 3 immunofluorescent staining results showed that cell proliferation decreased in the Si-ppm1d group 4 compared with the Si-Nc group (Fig.6B). Cell apoptosis increased significantly in the Si-ppm1d 5 group compared with the Si-Nc group, as evidenced by Fluorescence activated Cell Sorting (FACS) 6 apoptosis detection (Fig.6C) and the TUNEL staining (Fig.6D). Western blot results indicate that

10
This present study demonstrates that the WIP1 expression of atretic follicles is significantly lower 11 than that of the healthy follicles. Our in vivo and in vitro experiments suggest that WIP1 related 12 abnormal apoptosis and autophagy in granulosa cells lead to follicular atresia, also impairing the 13 oocyte quality. WIP1 plays an essential role in regulating the ovarian reserve and ovarian function.  Previous studies have demonstrated apoptosis is the main mechanism mediating the follicular 19 atresia [7,12]. Recently studies indicated that autophagy is also closely linked to follicular 20 development [14]. Our data extend the previously identified role of apoptosis and autophagy in  Our previous study has already demonstrated that the WIP1 expression in mouse ovaries 13 declined with age [29]. The WIP1 phosphatase activity decline is closely associated with impaired 14 follicular development, as well as the ovarian aging. The age-dependent WIP1 downregulation 15 accelerated the atresia of growing follicles, indirectly accelerating the depletion of the primal 16 follicle pool. Similar to our study, Wip1 has also been proved to play a key role in several 17 physiological processes such as neurogenesis and organismal aging [27,28]. Another study also 18 shows that Wip1 is highly expressed in haematopoietic stem cell (HSC) but decreases with age, and

Availability of data and materials 13
The data that support the findings of this study are included in this article. The original data underlying this 14 article will be shared on reasonable request to the corresponding author.