Obesity Affects IVF Outcomes and Mevalonic Acid Rescues the Side Effects of Statins on Follicle Growth.

Fei Mao, Yun Qian, and Ruizhi Feng conceived and designed the study. Fei Mao conducted all experiments and analyzed data with the assistance of Zhihan Guo, Tianqing Chang, Guangping Yang and Heqi Dai. Fei Mao and Jihong Yang supported the collection of clinical epidemiological information. Jihong Yang contributed to the clinical sample collection; Fei Mao wrote the manuscript. Yun Qian, Ruizhi Feng, Jihong Yang, and Hua Wu revised and polished the manuscript. All authors have read and agreed to the published version of the manuscript.

Informed consent was obtained from all participating subjects. The Ethics Committee of the Second Affiliated Hospital of Nanjing Medical University reviewed and approved this study. The study was conducted in accordance with the guidelines of the Declaration of Helsinki, and approved by Nanjing Medical University Animal Care and Use Committee (Nanjing, China).

The patients were grouped according to their BMI: normal (18.5 ≤ BMI < 28, n  = 204) and obesity (≥ 28 kg/m 2 , n  = 128) groups. The differences in clinical epidemiological data between the two groups were compared. As displayed in Table  1 , there were no significant differences in patients' age and basal serum anti‐Müllerian hormone (AMH) level. The number of antral follicle count (AFC) on the day of oocyte retrieval, gonadotrophin (Gn) dose, and Gn duration in obese patients was significantly higher than those in normal patients ( p  < 0.001). Meanwhile, the number of retrieved oocytes, mature oocytes, fertilized oocytes, cleavage embryos, and high‐quality embryos showed significant opposite differences between the two groups. Clinical epidemiological information for participants. Note: All values are presented as the mean ± SEM. Abbreviations: AFC, antral follicle count; AMH, anti‐Müllerian hormone; Gn, gonadotrophin. p  < 0.05. p  < 0.001. To clarify the direct relationship between BMI and IVF outcomes, the correlation was evaluated by Spearman correlation analysis. The results demonstrate that BMI was positively correlated with AFC ( R  = 0.431, p  < 0.001), Gn dose ( R  = 0.371, p  < 0.001), Gn duration ( R  = 0.390, p  < 0.001), the difference of which was statistically significant (Figure  1A ). Negative correlations were found between BMI and the number of retrieved oocytes ( R  = −0.315, p  < 0.001), mature oocytes ( R  = −0.258, p  < 0.001), fertilized oocytes ( R  = −0.250, p  < 0.001), cleavage embryos ( R  = −0.171, p  = 0.002) and high‐quality embryos ( R  = −0.162, p  = 0.003) (Figure  1A ). (A) Correlations between BMI and clinical parameters. (B) Representative images of follicular growth on Days 0, 2, 4, and 6 in each group (magnification, 40×; scale bar, 100 μm). There were at least 40 follicles per group. (C) In vitro follicle growth curve. (D) Growth diameters of surviving follicles (the diameter on Day 6 minus that on Day 0). (E) Survival rate of follicles in different groups. Results presented as the mean ± SEM. Different letters in superscript (a, b, c) indicate significant differences between groups ( p  < 0.05). To further investigate the effects of lipid metabolites on follicular development, we added 1 and 10 μM SV to an in vitro culture medium of mouse follicles. During the culture process, photographs were taken on Days D0, D2, D4, and D6 to observe growth, as shown in Figure  1B,C . On D0, the day of follicle collection, the sizes of the follicles in the different groups were no significant difference. By D4, the diameters of the follicles in the BL and DMSO control groups were significantly larger than those in the model groups (1 and 10 μM SV). On D6, the follicle surrounding the theca layer was complete in the BL and DMSO groups, with a clear boundary and translucent follicular cavity structures, and the cumulus‐oocyte complex faintly visible in the center. The GCs surrounding the follicles atrophied in the model groups and the central oocytes were deformed. The follicle growth diameters in the BL, DMSO, 1 μM SV, and 10 μM SV groups were 303.20 ± 15.94 μm, 302.90 ± 16.63 μm, 80.34 ± 15.84 μm, and 16.06 ± 5.67 μm, respectively (Figure  1C ). The growth diameters of the SV model groups were markedly less than those in the control groups (Figure  1D ). Furthermore, compared with the control groups, SV supplementation led to a significant and dose‐dependent reduction in follicle survival. The survival rate of the 10 μM SV group (62.07%) was the lowest at Day 4. On Day 6 of in vitro growth, the follicle survival rate was significantly lower in the 1 μM SV and 10 μM SV groups (43.59% and 5.31%, respectively) than in the BL and DMSO groups (92.84% and 93.97%, respectively) (Figure  1E ). To further explore the effects of lipid metabolites on periovulatory follicles, we used SV to construct an in vitro oocyte maturation model. It was found that the addition of 1, 10, or 100 μM SV to the IVM media did not alter the maturation, fertilization, and two‐cell rates of oocytes in any of the groups (Figure  S1 ). Wide‐spectrum targeted metabolomics was employed to analyze FF collected from paired small ( n  = 26) and large ( n  = 26) follicles from 26 patients [ 34 ]. The P‐value of MVA, the key metabolite screened from the data, is 2.74 × 10 −8 , the VIP value is 2.903, and the fold change is 0.393 [ 34 ]. The relative content of MVA was significantly greater in the FF of large follicles compared with small follicles (193 572 ± 14 014 vs. 76 031 ± 6256, respectively) ( p  < 0.001, Figure  2A ). Receiver operating characteristic (ROC) curves were used to determine the value of MVA in predicting follicle size. The relative content of MVA had a high predictive value for follicular development, with an area under the curve (AUC) of 0.9978 ( p  < 0.001). A cut‐off value of 112 535 for the relative content of MVA represented a sensitivity of 92.31% and specificity of 96.15% for the prediction of follicle size (Figure  2B ). (A) Relative MVA content in FF at different developmental stages. (B) Follicle size: the area under the ROC curve for MVA was 0.9778 (*** p  < 0.001); C Representative images of follicle growth in vitro on Days 0, 2, 4, and 6 in different groups (magnification, 40×; scale bar, 100 μm). There were at least 40 follicles per group. (D) Growth curves of follicle growth at different MVA concentrations in the rescue experiment. (E) Concentrations of estradiol in the follicular culture medium on Days 2,4 and 6. (F) Follicle diameters on Day 6 in different groups. (G) Survival rates of follicles in different groups. (H) Antrum formation rates of follicles on Day 6 in different experiments. Results are presented as the mean ± SEM. Different superscript letters (a, b, c) indicate significant differences between groups ( p  < 0.05). To clarify the relationships between FF MVA, oocyte mass, and pregnancy outcomes, the relevant clinical epidemiological information of the 26 patients was collected for analysis. In the clinic, the size of the follicle was monitored before oocyte retrieval. Larger follicles were more likely to be associated with oocytes with development potential [ 35 ]. Therefore, the associations between MVA levels in the FF from large follicles and clinical features of the patients were examined. This showed significant positive correlations between MVA levels and the basal estrogen level ( r  = 0.407, p  = 0.039) and the oocyte maturation rate ( r  = 0.527, p  = 0.006) (Table  2 ). Spearman correlation coefficients between FF MVA contents in large follicles and clinical demographic characteristics. Note: oocyte maturation rate = number of oocytes at MII stage/total number of oocytes retrieved; fertilization rate = number of fertilized embryos/total number of oocytes retrieved; Cleavage rate = number of cleavage embryos/number of 2PN; High‐quality embryo rate = number of high‐quality embryos (Grades I and II embryos)/number of 2PN. Abbreviations: AFC, antral follicle count; AMH, anti‐Müllerian hormone; BMI, body mass index; E2, estradiol; FF, follicular fluid; FSH, follicle‐stimulating hormone; GnRH, gonadotrophin‐releasing hormone; HCG, human chorionic gonadotrophin; LH, luteinizing hormone; MVA, mevalonic acid; P, progestogen. p  < 0.05. p  < 0.01. We used an in vitro mouse follicle culture system added 1 μM SV for subsequent MVA rescue experiments. The addition of 0.5, 1, and 2 mM MVA could partially rescue the growth of follicles in model groups, and improvements in the morphology of the follicles in relation to MVA concentration. The growth morphology of follicles cultured in the medium supplemented with 2 mM MVA alone was similar to that of the BL group (Figure  2C ). There were no significant differences in follicle diameters between the groups on the day of follicle acquisition (Figure  2D and Table  3 ). It was observed that follicles grew markedly faster in the BL and 2 mM MVA groups relative to the other groups, with the 2 mM MVA group showing the same growth trend as the BL group. The rescue groups supplemented with 0.5, 1, and 2 mM MVA all showed improvement in follicle growth after treatment with SV (Figure  2D ). Results of experiments on preantral follicles in culture. 1 μM SV +2 mM MVA Note: Different letters in superscript (a, b, c) indicate significant differences between groups (p < 0.05). Measurement of estradiol levels indicated that blocking the MVA pathway with 1 μM SV reduced estradiol production by follicular GCs significantly on Day 6 and, while 2 mM MVA could partially rescue this effect, the change was non‐significant (Figure  2E ). Follicle diameters, survival rates, and antrum formation rates were assessed for each group on Day 6. It was found that MVA supplementation partially improved the follicle diameter values on Day 6, although no significant differences were seen in terms of MVA concentration (Figure  2F ). Compared with the model group, the addition of 0.5, 1, and 2 mM MVA partially and dose‐dependently rescued survival and antrum formation rates on Day 6, with 2 mM MVA having the best rescue effect (Figure  2G,H ). As shown in Table  3 , the rate of antrum formation in the 1 μM SV + 2 mM MVA group was 41.30%, showing greater significance than that in the 1 μM SV group (0.00%), and less than in the BL (65.91%) and 2 mM MVA groups (60.87%). When the surviving oocytes were subjected to IVM, the proportion of MII oocytes in relation to the overall number of surviving follicles was determined, showing that mature oocyte formation in the BL and 2 mM MVA groups (18.18% and 26.09%, respectively) was significantly higher than those in the 1 μM SV and 1 μM SV + 2 mM MVA groups (0.00% and 6.52%, respectively). Maturation rates did not differ significantly between the BL and 2 mM MVA groups. The rescue group with 1 μM SV + 2 mM MVA showed partial improvement in the maturation rate relative to the model group, although this was non‐significant (Table  3 ). To further confirm the impact of downstream metabolites on granulosa cells growth and development, we utilized a simplified in vitro KGN cell culture model. Flow cytometry was used to assess apoptosis in KGN cells under each treatment condition. Results showed that 1 μM SV can significantly increase the apoptosis of KGN cells compared to BL group, while both Chol and GGOH can partially rescue the apoptosis elevation (Figure  3A,B ). CCK‐8 assays revealed significantly suppressed proliferation of KGN cells in the 1 μM SV treatment group compared to other groups at both 48 h and 72 h time points (Figure  3C ). (A) The apoptosis of KGN cells in each group was detected using Annexin V/PI staining. (B) Histogram of apoptosis of KGN cells in each group. (C) Cell proliferation was measured at the indicated time points using Cell Counting Kit‐8. Results are presented as the mean ± SEM. Different superscript letters (a, b, c) indicate significant differences between groups ( p  < 0.05). RNA‐seq was performed to compare mRNA expression levels in the GCs of follicles between the three experimental groups, namely, the BL (B), 1 μM SV (S), and 1 μM SV + 2 mM MVA (SM) groups. Hierarchical clustering (Figure  4A ) showed separation between the GCs treated with 1 μM SV and the other two groups. GCs treated with 1 μM SV +2 mM MVA were clustered together with those in the BL group. (A) Hierarchical cluster analysis of DEGs identified by transcriptomic analysis of GCs. Blue, white, and red representing gradients gene expression levels, with blue representing low and red representing high. (B) Heatmap showing DEGs between the SM and S groups. Color coding as in (A). Horizontal bars represent genes with columns representing samples. (C) GO annotations of DEGs in the SM versus S groups. The horizontal coordinate represents DEG annotations, and the left and right vertical coordinates indicate the proportions of up‐and down‐regulated DEGs. (D) The q‐value distribution map of enriched KEGG pathways. Colors indicate q ‐values. (E) KEGG enrichment of DEGs, showing the top 10 pathways in the SM versus S groups. The enrichment factor indicates the number of DEGs relative to the overall number of genes. Dot sizes represent gene numbers in the pathways, with dot colors indicating q ‐values; F RT‐qPCR verification of the expression of eight DEGs. Results are presented as the mean ± SEM.* p  < 0.05, ** p  < 0.01, *** p  < 0.001. DEGs between the groups were identified with the threshold criterion of p value  1, focusing mainly on differences between the S and SM groups. A total of 361 DEGs were identified between the S and SM groups (Figure  4B ), of which 49 genes were upregulated and 312 down‐regulated in the SM group compared with the S group. Details of the upregulated and downregulated genes in the groups are provided in Table  S2 . The specific biological annotations of DEGs in the SM versus S group comparison were investigated by GO analysis. There are three categories of GO enrichment, namely, cellular component (CC), biological process (BP), and molecular function (MF), and 15, 23, and 10 DEGs were enriched in these categories, respectively. Processes that were altered in the SM group relative to S were found to be “cellular process”, “biological regulation” and “response to stimulus” in the BP category, while MF and CC annotations were most associated with “cell part” and “binding” (Figure  4C ). KEGG analysis was then performed to examine differences between groups. The representative 20 pathways in the q ‐value (adjusted P) distribution map of KEGG enrichment pathways are shown in Figure  4D . Among them, only two pathways (steroid biosynthesis and terpenoid backbone biosynthesis) were significantly ( q  < 0.05) changed between the SM and B groups. When the SM group was compared with the S group, enrichment was observed in all the top 20 pathways except for steroid biosynthesis, terpenoid backbone biosynthesis, and ovarian steroidogenesis. TNF, JAK–STAT and PI3K‐Akt signaling were significantly enrichment between SM and S groups, but not between SM and B groups. The top 10 enrichment pathways in the SM versus S group were ECM‐receptor interactions, Protein digestion and absorption, hematopoietic cell lineages, AGE‐RAGE signaling in diabetic complications, TNF signaling, gastric cancer, JAK–STAT signaling, cytokine‐cytokine receptor interactions, PI3K‐Akt signaling, and Pathways in cancer (Figure  4E ). Eight genes were chosen for verification of the RNA‐seq data by real‐time PCR on independent samples. Differential expression of five genes identified as linked to the KEGG pathway of steroid metabolism were Cyp51, Sqle, Tm7sf2, Ebp, and Nadhl. The remaining genes were involved in cytokine‐cytokine receptor interaction and terpenoid backbone biosynthesis. The above verification results showed a similar expression pattern to that observed in the RNA sequencing (Figure  4F ).

In this study, we further confirmed through clinical data analysis that obesity reduces in vitro fertilization pregnancy success, with significant decreases in the number of retrieved oocytes, mature oocytes, and cleavage embryos. To explore possible causes, we started from lipid metabolism and found that the lipid‐lowering drug statins inhibited preantral follicle growth. By screening for differential metabolites at different stages of follicle development, MVA is a key metabolite that can rescue the effect of the potential reproductive toxicity of simvastatin. Both the downstream metabolites cholesterol and GGOH partially attenuated SV‐induced apoptosis elevation and proliferation suppression in KGN cells. RNA sequencing results suggested that these effects may have been mediated through the TNF, PI3K‐Akt, and JAK–STAT pathways. The findings shed light upon clinical practice for women of childbearing age who were treated with statins and might contribute to better health management in obesity populations. Consistent with previous studies [ 36 , 37 , 38 ], obesity impairs ovarian responsiveness to gonadotropin stimulation; these patients used higher doses and more days of Gn in the clinic. The obese group showed more antral follicles on the day of oocyte retrieval, which may be due to increased duration and amount of Gn. However, in order to achieve the success of IVF, the changes in the ovulation stimulation protocol have not succeeded in increasing the number of retrieved oocytes, mature oocytes, fertilized oocytes, cleavage embryos, and high‐quality embryos. Similarly, the effectiveness of short‐term weight loss remains to be improved, unrelated to clinical outcomes [ 39 ]. Therefore, it is urgent to find a way to improve the success rate of IVF in obese patients. The lipid metabolomics may provide a novel method. Statins, such as SV, are the most frequently used cholesterol‐lowering drugs both in Europe and the US [ 40 ], inhibiting the rate‐limiting enzyme of the lipid metabolic pathway [ 41 ]. Statins are used to control cholesterol levels in patients with hyperlipidemia and can also reduce proliferation in some tumor cells [ 42 ]. Surveys show that the average age of women giving birth for the first time has increased markedly [ 43 ], and the increase in age is often accompanied by an increase in the incidence of obesity, cancer, and other diseases, which means the number of reproductively active women using statins is also increasing [ 40 ]. However, some researchers believe that statins should be used with caution in women of childbearing age, as pregnancy is contraindicated with SV [ 40 ]. Statins inhibit blastocyst formation by inhibiting geranylgeranylation [ 25 ]. In vivo experiments conducted in rats showed that SV reduced serum levels of FSH and progesterone [ 40 ]. Statins can also lower the synthesis of cholesterol and enhance the apoptosis of both human and rat GCs in the periovulatory stage in vitro [ 24 ]. However, the influence of statins on the development of preantral follicles is unknown. We found SV strongly and dose‐dependently inhibited follicle growth in mice, together with adversely affecting follicle survival, reducing growth, and preventing the formation of normal follicular antra by an in vitro follicle culture system. To further explore the effects of SV on oocytes, we conducted in vitro studies on the final stages of follicular development. During IVM, the COC was cultured for 16–18 h after adding SV. The results showed that SV did not significantly affect the maturation, fertilization, and development of oocytes. The possible reasons are as follows: (1) The SV concentration was insufficient; (2) The duration of SV action is short, and metabolites remaining in the medium were sufficient for the use of cumulus GCs and oocytes. By screening for differential metabolites at different stages of follicle development, MVA was found to be a key metabolite. Our clinical data showed a positive association between MVA levels and basal patient estrogen levels. The mevalonate pathway is important in many organisms. In humans, it consists of the reactions starting with acetyl‐coenzyme A to generate MVA and leads to the formation of farnesyl pyrophosphate, which then serves as the substrate for several biologically important agents, including Chol, isoprenylated proteins, coenzyme Q, and dolichol [ 17 , 44 , 45 ]. The pathway is associated with signal transduction [ 46 ], the modulation of proliferation, energy homeostasis [ 42 ], and reduction in inflammation [ 47 ]. It also has been found that there is an association between the pathway and female reproduction [ 25 , 48 ]. A recent study showed that age‐related MVA pathway dysfunction led to defective oocyte meiosis and aneuploidy [ 49 ]. Our research shows that MVA can reverse statin‐induced follicle dysplasia. Measurement of estrogen in the follicle culture medium in vitro showed that SV decreased the level of estrogen significantly, and MVA can partially rescue it, which is in agreement with our clinical research results. Therefore, whether statins can be used in these women needs to be considered, and whether they should be combined with an appropriate dose of MVA after use may be a protocol worth investigating. On the other hand, to further explore the influence of SV on follicular development, our KGN cell culture model showed that SV had an inhibitory effect on the proliferation of KGN and a promoting effect on the apoptosis of KGN. Both Chol and GGOH, the downstream metabolites of MVA, can partially counteract this effect. Our research results are consistent with those of a previous study in granulosa cells during the periovulation period in humans and rats [ 50 ]. Chol is indispensable as a substrate for steroid hormone synthesis in the mammalian ovarian follicle [ 51 ]. The production of steroid hormones is related to the proliferation of granulosa cells and the induction of ovulation [ 52 ]. Isoprenoid and geranylgeranyl groups are essential for post‐translational lipid modification of many proteins [ 53 ] and play critical roles in the activation of small GTPases. The activated form of GGOH, geranylgeranyl pyrophosphate, serves as the key donor for the isoprenylation modification of Rho GTPases (e.g., RhoA). The small GTPase family regulates diverse signaling pathways, including contraction of intercellular junctions, migration, proliferation, and differentiation [ 54 ], and is also closely associated with apoptosis [ 55 , 56 ]. The impact of SV on follicular development may rely not only on the cholesterol pathway but also be closely linked to the synthesis of geranylgeranyl groups. Preantral follicles are not only responsive to gonadotropin, but are also regulated by diverse paracrine and autocrine factors secreted by oocytes and GCs [ 57 ]. Therefore, we investigated transcriptomic changes in GCs in the presence of SV and MVA. According to the results of the KEGG pathway analysis, steroid and terpenoid backbone biosynthesis were not rescued by MVA, consistent with the results of reduced hormone synthesis. However, TNF, PI3K‐Akt, and JAK–STAT signaling were restored to normal levels after MVA rescue. It has been found that mammalian ovary function is dependent on TNF signaling, with both TNF‐α and its receptor TNFR2 expressed by cumulus cells and oocytes in human [ 58 ]. While in rats, TNFR is found in GCs, oocytes, and interstitial cells, and TNF‐α in oocytes [ 59 ]. Inclusion of TNF‐α in follicle culture media adversely affects their survival and enhances apoptosis in bovine ovarian cells [ 60 ]. Similar effects were observed after activation of the TNF pathway by glucocorticoids [ 61 ]. The PI3K‐Akt pathway is central to the metabolism and growth of cells [ 62 ] and has also been linked to ovarian function [ 63 , 64 ]. The pathway regulates activation of primordial follicles through FOXO3 [ 63 ]. This regulation through PI3K signaling has also been linked to DNA damage and repair [ 65 ]. The PI3K pathway, together with the MAPK pathway, mediates the actions of progesterone and FSH on follicle development [ 66 ]. Blockage of the mevalonate pathway increases apoptosis through inhibition of Akt in various tumor cells [ 67 ]. JAK–STAT signaling has also been demonstrated in the human granular cell line COV434 [ 68 ]. These results indicate that SV and MVA can alter the GC transcriptome, possibly mediated by the above pathways. We intend to investigate the pathways through which MVA influences follicular growth in the future.

All authors gave their consent for publication.

The authors declare no conflicts of interest.

The Ethics Committee of the Second Affiliated Hospital of Nanjing Medical University reviewed and approved the population studies. This study was registered in the China clinical trial registration center under the registration number ChiCTR1900022319. All animal protocols in the study were approved by the Ethics Committee of Laboratory Animals and the Animal Care and Use Committee of Nanjing Medical University. A total of 332 patients who underwent ovarian stimulation for IVF between October 2022 and March 2024 at the Reproductive Medicine Center of the Second Affiliated Hospital of Nanjing Medical University, China, were recruited into this study. We included women ≤ 35 years of age who were infertile due to tubal factors or unexplained infertility and their partners could provide normal sperm for IVF. The exclusion criteria are any factors that may affect reproductive outcomes, such as a history of endometriosis, premature ovarian failure, polycystic ovary syndrome, ovarian surgery, diabetes, or other endocrine and immune‐related diseases. The ovarian stimulation protocol was as our previous study [ 28 ]. According to the Guidelines for Prevention and Control of Overweight and Obesity in Chinese Adults, all patients were classified into two groups according to their BMI: normal, 18.5 ≤ BMI < 28; obese, BMI ≥ 28 [ 29 ]. Each subject provided written informed consent. BDF1 female mice were used in this study [ 30 ]. Eight‐week‐old male DBA/2 and female C57BL/6J mice were mated in a ratio of 1:2 under conditions of 25 ± 1 and a 12‐h/12‐h light/dark cycle. Sixteen to seventeen‐day‐old F1 female mice were used for follicle culture experiments. The preantral follicles were collected as previously reported [ 30 ]. Bilateral ovaries were harvested from the peritoneal cavities of F1 mice after cervical dislocation and were placed in preheated MEM‐α medium (12561056, Gibco, USA). Using a dissecting scope (Olympus, Tokyo, Japan), the entire ovaries were separated from the uterus, bursa, and excess adipose tissue. The preantral follicles were mechanically separated using a 1‐mL insulin needle to avoid disruption of the membrane cell layer, trying to remove as much stroma as possible without puncturing the follicle. Using a homemade mouth straw, preantral follicles with diameters of 120–150 μm were transferred to a clean dish of MEM‐α containing 5% FBS + 1% penicillin–streptomycin, and excess impurities were removed by gently and repeatedly blowing and inhaling. This procedure is completed within 30 min on average. On the day before the experiment, 96‐well flat‐bottomed tissue culture plates were coated with 3 mg/mL Matrigel (354234, Corning, USA) to evenly cover the entire bottom of the wells [ 31 , 32 ]. The preantral follicles were inoculated in the wells with preheated MEM‐α containing 5% FBS + 1% penicillin–streptomycin + 1% ITS (I3146, Sigma, USA) + 100mIU/mL follicle‐stimulating factor (urofollitropin for injection, Livzon, China), with one preantral follicle/well, and cultured for 6–8 days at 37°C with 5% CO 2 . Follicle diameters (measured in two vertical meridians, averaged) were measured under an IX2 inverted microscope (Olympus, Tokyo, Japan), and a semi‐quantitative medium exchange was performed every other day. On Days 6–8 of culture, the cumulus‐oocyte complex (COC) located in the center of the antral follicle was removed using an oral pipette. The COCs were incubated in in vitro maturation (IVM) culture droplets (α‐MEM containing 5% FBS, 3 ng/mL EGF, 50mIU/mL rFSH, 0.25 mmol/L sodium pyruvate, 0.5% penicillin, and 0.5% streptomycin) (Aibei Biotechnology, Nanjing, China) for 24 h. The expanded COCs were digested with hyaluronidase (H4272, Sigma, USA) preheated at 37°C for 1 min, and the oocytes were repeatedly washed with a homemade mouth straw until no GCs surrounded the oocytes. The oocytes were then examined and counted under a microscope. The statin SV, an inhibitor of a key enzyme in the lipid metabolism pathway, was added to the follicular culture medium. The follicles were randomly allocated to four groups, namely, the BL, DMSO, 1 μM SV, and 10 μM SV groups, with at least 40 follicles per group. After 6–8 days of culture, the growth diameter, survival, and follicular cavity formation were evaluated. Based on the 1 μM SV group, culture medium supplemented with 0.5, 1, and 2 mM MVA was used to verify the rescue effects of MVA on follicular development and oocyte maturation. Three‐week‐old F1 female mice were given intraperitoneal injections of 5 IU of pregnant mare serum gonadotrophin (PMSG, Ningbo Second Hormone Factory, Ningbo, China) 24 h in advance [ 33 ]. The mice were sacrificed after 46–48 h of superovulation and both ovaries were removed. Large antral follicles were punctured and COCs with intact cumulus cell layers were collected in IVM culture drops with 1, 10, or 100 μM SV. The COCs were placed in a Petri dish with 245 μL of human tubal fluid (HTF, Aibei Biotechnology) and incubated for 16–18 h. Spermatozoa were harvested from the epididymides of mature F1 male mice and capacitation induced by pre‐incubation in HTF for 1 h. Then, for fertilization, 5 μL of the capacitated spermatozoa were introduced. Fertilized oocytes were collected after 6–7 h and transferred to KSOM medium (Easycheck, Nanjing, China) for the evaluation of fertilization and oocyte maturation. The two‐cell rate was observed 24 h after fertilization. Culture medium was collected while semi‐quantitative liquid exchange was performed on Days 2, 4, and 6 of culture. Samples were collected from 20 wells each time and stored at −80°C. Estrogen levels were measured by radioimmunoassay by the Beijing North Institute of Biotechnology. On the fourth day of follicle culture, the central oocytes were removed with an oral pipette and GCs were collected. After three washes in PBS, the total RNA was extracted using the RNA‐Easy Isolation Reagent (R701‐01, Vazyme, China), following the provided directions. The RNA was dissolved in RNase‐free water and transported on dry ice. Transcriptome sequencing was conducted by Annoroad Gene Tech, Beijing Co. Ltd. RNA purity was measured using a Kaiao K5500 spectrophotometer (China), and integrity and concentration with an Agilent Bioanalyzer 2100 (USA). Three sets of samples, each with two biological replicates, were included (BL group, 1 μM SV group, and 1 μM SV + 2 mM MVA group). Construction of cDNA libraries was undertaken with a NEBNext Ultra RNA Library Prep Kit for Illumina (#E7530L, NEB, USA) according to the provided directions. The libraries were checked for size and purity with the Agilent 2100, followed by sequencing of suitable libraries on an Illumina platform and the generation of 150‐bp paired‐end reads. The clean data were mapped to the Ensembl mouse genome (GRCm38.99) using HISAT2 (version 2.1.0). The read counts for individual genes in the samples were determined by HTSeq v0.6.0, and the fragments per kilobase of exon per million mapped reads (FPKM) value was determined to evaluate gene expression levels. Differentially expressed genes (DEGs) were identified using DESeq2 using the criteria of q ‐value < 0.05 (corrected p value) and |log2 (Fold Change)|≥ 1. GO and KEGG enrichment analyses of the DEGs were performed using the hypergeometric test with q ‐values calculated from the p values. GO terms and KEGG terms with q  < 0.05 were considered significantly enriched. The RNA was reverse‐transcribed into cDNA using HiScript III All‐in‐one RT SuperMix Perfect for qPCR (R333, Vazyme, China) and the cDNA was amplified using Taq Pro Universal SYBR qPCR Master Mix (Q712, Vazyme) on a QuantStudio 7 Real‐Time PCR System (Applied Biosystems, USA). Relative mRNA expression was calculated by the 2 −ΔΔCt method with gapdh as the internal control. The experiment was repeated in triplicate. The sequences of the primers used are provided in Table  S1 . The human granulosa cell tumor‐derived cell line, KGN, was generously provided by Prof. Ruizhi Feng's research group. KGN cells were maintained in DMEM/F12 Medium (BC‐M‐002, SenBeiJia, China) supplemented with 10% fetal bovine serum (A5256701, Gibco, USA) and 1% penicillin–streptomycin (BL505A, Biosharp, China). Cells were cultured at 37°C in a humidified atmosphere containing 5% CO 2 and passaged at a ratio of 1:2–1:3 every 2 days. To detect apoptosis, we cultured KGN cells in 6‐well plates and grouped them into BL group, 1 μM SV group, 1 μM SV + 100 μM Chol group, and 1 μM SV + 10 μM GGOH group respectively. They were harvested by adding trypsin without EDTA. Cell apoptosis was detected using the Annexin V‐FITC/PI Apoptosis Detection Kit (A211‐02, Vazyme, China) according to the manufacturer's protocol. Briefly, cells were washed twice with PBS and resuspended in 100 μL of 1× binding buffer. Each sample was then stained with Annexin V‐FITC and propidium iodide (PI), followed by incubation at room temperature in the dark for 10 min. Subsequently, 400 μL of 1× binding buffer was added to the cell suspension. Flow cytometric analysis was performed using a flow cytometer 1 h after staining. Data were analyzed using FlowJo software (v10). To detect proliferation, we cultured KGN cells in 96‐well plates and grouped them as above. The CCK‐8 assay (A311‐01, Vazyme, China) was used according to the manufacturer's protocol. The absorbance of the solution was measured at 450 nm. Data were analyzed using SPSS 12.0 and GraphPad Prism 6.0. Means and standard error of the mean (SEM) were plotted. Statistical differences were analyzed using Student's t ‐test and ANOVA. Differences were considered statistically significant at p  < 0.05.

Figure S1. Effects of simvastatin on oocyte maturation rate A, fertilization rate B and 2‐cell rate C. Table S1. Primer sequences for real‐time PCR. Table S2. Details of the upregulated and downregulated genes between SM and S group.