Mevalonate metabolites boost aged oocyte quality through prenylation of small GTPases

Human ovaries are one of the earliest organs to show aging-associated dysfunctions, with a remarkable functional decline after the age of 35 (refs. ^1,2). Ovarian aging mainly manifests as declines in oocyte quality and quantity, resulting in female subfertility and infertility^3,4. Cortical F-actin affects oocyte contractions, cytoplasmic organization and developmental potential⁵. Abnormal cortical F-actin assembly contributes to aneuploidy and declined oocyte quality^5,6,7,8,9. It has been reported that cortical F-actin formation requires the polarized activation of the small guanosine triphosphatases (GTPases) and the Arp2/3 complex^10,11,12. Dysfunction of small GTPases leads to failure in the formation of cortical F-actin and the loss of cortical actomyosin polarity^13,14. Most of the small GTPases could be prenylated, a lipidic modification that tethers proteins to cell membranes, enabling proper protein localization and function^15,16. The mevalonate (MVA) pathway is crucial for multiple cellular processes through protein prenylation¹⁶. Recently, we reported that abnormal metabolism of the MVA pathway in granulosa cells (GCs) contributes to oocyte meiotic defects and aneuploidy¹⁷. However, whether MVA metabolites regulate oocyte cortical F-actin assembly via prenylation of small GTPases during ovarian aging remains unknown.

Recent studies have reported several therapeutic strategies for improving the quality of aged oocytes, including the use of antioxidants, growth hormones, melatonin, nicotinamide mononucleotide and the polyamine metabolite spermidine^{18,19,20,21,22}. The effectiveness of these substances is yet to be tested in the clinic, and more effective drugs for the treatment of ovarian aging are worthy to be explored. MVA is a pivotal water-soluble upstream metabolite for cholesterol synthesis and the synthesis of more than 50,000 terpenoids, and is required for cell cycle progression and cell proliferation^16,23,24. Previous studies have shown that MVA is important for the coordination of regulatory T cell proliferation by enhancing transforming growth factor-β signaling and the mediation of training immunity via IGF1-R and mTOR activation^25,26. MVA supplementation has also been shown to counteract the adverse developmental effects of statins on blastocyst formation^27,28.

Here we found that Arp2/3-dependent assembly of cortical F-actin was decreased in aged oocytes, accompanied by reduced MVA pathway metabolite (MVA, farnesyl pyrophosphate (FPP) and geranylgeranyl pyrophosphate (GGPP)) levels. MVA supplementation improved the quality of aged oocytes in cumulus–oocyte complexes (COCs) by promoting cortical F-actin assembly, ameliorating meiotic errors and promoting embryonic development both in vitro and in vivo. Mechanistically, MVA activated the FPP metabolic pathway in GCs and transported FPP promoted cortical F-actin assembly by increasing CDC42 and RAC1 prenylation and Arp2/3 complex formation in aged oocytes. In addition, 8-isopentenyl flavone (8-IPF), a natural chemical drug, also ameliorated the decline in ovarian reserve and significantly increased oocyte quality in aged mice by enhancing the prenylation of CDC42 and RAC1 to assemble the oocyte cortical F-actin. These data reveal a potential mechanism for ovarian aging mediated by metabolism coupling between GCs and oocytes and provide more effective and suitable interventions for clinical use to improve female fertility with advanced maternal age.

Decreased cortical F-actin distribution and MVA pathway metabolites in aged oocytes

Cortical F-actin is crucial for the correct segregation of chromosomes, asymmetric divisions and subsequent embryonic developmental potential^5,29. To explore the changes of cortical F-actin during ovarian aging, cortical F-actin distribution of oocytes at metaphase I (MI) from the COCs of young and aged mice was examined. The fluorescence intensity of cortical F-actin of aged oocytes was much weaker than that of young oocytes (0.2 ± 0.03 versus 1.0 ± 0.09, P < 0.0001) (Fig. 1a,b). The assembly of cortical F-actin requires the activity of the Arp2/3 complex, and it was reported that the protein and messenger RNA levels of the Arp2/3 complex were decreased in aged metaphase II (MII) oocytes^5,30,31. We further detected Arp3 in aged MI oocytes and found that Arp3 expression was markedly decreased in aged MI oocytes compared with that of young oocytes (0.2 ± 0.05 versus 1.0 ± 0.08, P < 0.0001) (Fig. 1c,d). Western blotting results further confirmed that the protein level of Arp3 was decreased in aged MI oocytes (0.7 ± 0.09 versus 1.0 ± 0.02, P = 0.02) (Extended Data Fig. 1a,b). Our recent study indicates that downregulation of the MVA pathway in aged ovaries is involved in the decrease of quality in aged oocytes¹⁷. Interestingly, atorvastatin (an MVA pathway blocker) treatment of young COCs considerably decreased the oocyte cortical F-actin (0.4 ± 0.03 versus 1.0 ± 0.06, P < 0.0001) and Arp3 expression (0.7 ± 0.06 versus 1.0 ± 0.1, P = 0.05) at MI (Fig. 1e–h). Western blotting results further confirmed the decreased Arp3 in MI oocytes from atorvastatin-treated COCs (0.8 ± 0.06 versus 1.0 ± 0.03, P = 0.04) (Extended Data Fig. 1c,d).

a, Fluorescence imaging showing F-actin expression at the MI oocyte cortex from the young and old groups. Young group: COCs from young mice (6 weeks old) cultured in MEMα maturation medium; old group: COCs from aged mice (10 months old) cultured in MEMα maturation medium. Scale bar, 25 µm. b, F-actin fluorescence intensity in the young (n = 17 oocytes) and old (n = 25 oocytes) groups. c, Fluorescence imaging showing Arp3 expression at the MI oocyte cortex from the young and old groups. Scale bar, 25 µm. d, Arp3 fluorescence intensity in the young (n = 20 oocytes) and old (n = 20 oocytes) groups. e, Fluorescence imaging showing F-actin expression at the MI oocyte cortex from the young and ATO groups. Young group: COCs from young mice (6 weeks old) cultured in MEMα maturation medium; ATO group: COCs from young mice (6 weeks old) cultured in MEMα maturation medium with 40 µM atorvastatin. Scale bar, 25 µm. f, F-actin fluorescence intensity in the young (n = 26 oocytes) and ATO (n = 24 oocytes) groups. g, Fluorescence imaging showing Arp3 expression at the MI oocyte cortex from the young and ATO groups. Scale bar, 25 µm. h, Arp3 fluorescence intensity in the young (n = 20 oocytes) and ATO (n = 20 oocytes) groups. i, Schematic illustration showing the collection of MI oocytes from 6-week-old mice and 10-month-old mice for targeted metabolomics. YMIO, MI oocyte from young mice (6 weeks old); OMIO, MI oocyte from aged mice (10 months old). j, Heatmaps showing the abundance of MVA, FPP and GGPP in oocytes by LC–MS/MS analysis. The scaled value bar indicates the relative concentration. k–m, Box plot showing the levels of MVA (k), FPP (l) and GGPP (m) per MI oocyte from young and aged mice. The data are shown as the mean ± s.e.m. of seven independent experiments. An unpaired two-tailed Student’s t-test was used for statistical analysis. Panel i created with BioRender.com.

Source data

Full size image

Our previous study showed that the effect of the MVA pathway in GCs on oocytes was mainly mediated by the non-sterol isoprenoid¹⁷. To further comprehensively study the change of MVA pathway metabolites in aged oocytes, we established a targeted metabolomics approach to detect MVA, FPP and GGPP using low cell inputs. We collected 10 oocytes at the MI stage from young (6 weeks old) and aged (10 months old) mice with 7 biological replicates for the MVA pathway metabolite test (Fig. 1i). Remarkably, aged oocytes contained significantly less MVA (911.9 ± 22.3 ng ml⁻¹ versus 1,297.0 ± 68.2 ng ml⁻¹, P = 0.0002), FPP (2.9 ± 0.2 ng ml⁻¹ versus 7.1 ± 0.7 ng ml⁻¹, P = 0.0001) and GGPP (7.0 ± 0.5 ng ml⁻¹ versus 16.0 ± 2.2 ng ml⁻¹, P = 0.002) compared with young oocytes (Fig. 1j–m and Supplementary Table 1). These results indicate that the decreased MVA pathway metabolites seem to be involved in abnormal assembly of the cortical F-actin in aged oocytes during ovarian aging.

MVA promotes cortical F-actin assembly and aged oocyte quality via the surrounding GCs

As the levels of MVA, FPP and GGPP were considerably reduced in aged oocytes and MVA is the precursor for terpenoids, including FPP and GGPP^23,24, we next investigated the effects of MVA supplementation on aged oocyte quality. We isolated aged COCs from 10-month-old female mice and cultured them in in vitro maturation (IVM) medium supplemented with 50 μM MVA (Fig. 2a). We first compared the cortical F-actin between the two groups treated with or without MVA and found that MVA supplementation markedly promoted cortical F-actin assembly of MI oocytes from aged COCs (6.3 ± 0.4 versus 1.0 ± 0.2, P < 0.0001) (Fig. 2b,c). In addition, the decreased Arp3 expression in the aged MI oocyte cortex was rescued after MVA supplementation (1.5 ± 0.08 versus 1.0 ± 0.04, P < 0.0001) (Fig. 2d,e). Western blotting analysis also confirmed the increased Arp3 level in MI oocytes from aged COCs after MVA supplementation (1.4 ± 0.008 versus 1.0 ± 0.07, P = 0.006) (Extended Data Fig. 2a,b). As actin is responsible for various essential functions, including spindle migration, accurate chromosome segregation and polar body extrusion (PBE) during oocyte meiosis³⁰, we next compared meiotic progression in oocytes after MVA treatment. The PBE rate was significantly higher in the MVA-treated group than in the control group (95.1% ± 3.3% versus 81.2% ± 4.0%, P = 0.02), although the germinal vesicle breakdown (GVBD) rates of oocytes collected from aged COCs showed no difference between groups (Fig. 2f and Extended Data Fig. 2c,d). We further evaluated meiotic defects and developmental potential of MII oocytes collected from the two groups. The incidence of meiotic defects (18.2% ± 5.9% versus 38.2% ± 4.1%, P = 0.03) and aneuploidy (16.7% ± 1.9% versus 38.9% ± 2.0%, P = 0.001) were markedly reduced after MVA supplementation (Fig. 2g,h and Extended Data Fig. 2e,f). Moreover, the MII oocytes derived from MVA-treated aged COCs were more competent to develop into 2-cell embryos (79.7% ± 4.2% versus 54.1% ± 5.1%, P = 0.005) and blastocysts (28.2% ± 3.6% versus 13.1% ± 2.4%, P = 0.008) than those derived from the control group (Fig. 2i–k). To further clarify whether MVA acted on oocytes or GCs, we also treated denuded oocytes (DOs) with 50 μM MVA. Notably, the incidence of GVBD, PBE and meiotic defects were not improved in DOs after MVA supplementation (Extended Data Fig. 2g–k), indicating that the effects of MVA on aged oocytes were executed via surrounding GCs.

a, Schematic illustration of the experimental protocol used to analyze the effect of MVA supplementation on the mouse oocyte meiotic process. b, Fluorescence imaging showing F-actin expression at the oocyte cortex in the old and MVA groups. Old group: COCs from aged mice (10 months old) cultured in MEMα maturation medium; MVA group: COCs from aged mice (10 months old) cultured in MEMα maturation medium with 50 µM MVA. Scale bar, 25 µm. c, F-actin fluorescence intensity in the old (n = 25 oocytes) and MVA (n = 28 oocytes) groups. d, Fluorescence imaging showing Arp3 expression at the oocyte cortex in the old and MVA groups. Scale bar, 25 µm. e, Arp3 fluorescence intensity in the old (n = 13 oocytes) and MVA (n = 13 oocytes) groups. f, Rate of PBE in the old (n = 76 oocytes) and MVA (n = 83 oocytes) groups. The data are shown as the mean ± s.e.m. of six independent experiments. g, Chromosome spread analysis showing representative images of aneuploid and euploid MII oocytes from the old and MVA groups. h, Rate of aneuploidy in the old (n = 33 oocytes) and MVA (n = 37 oocytes) groups. The data are shown as the mean ± s.e.m. of three independent experiments. i, Images of 2-cell embryos and blastocysts from the old and MVA groups. The arrowheads denote blastocysts. Scale bar, 100 µm. j,k, Rates of 2-cell embryos (j) and blastocysts (k) in the old (n = 78 oocytes) and MVA (n = 84 oocytes) groups. The data are shown as the mean ± s.e.m. of five independent experiments. An unpaired two-tailed Student’s t-test was used for statistical analysis. Panel a created with BioRender.com.

Source data

Full size image

To further evaluate the effects of MVA on aged COCs in vivo, 9-month-old female mice were intraperitoneally injected with 5 mg kg⁻¹ MVA for 30 days (Fig. 3a). The ovarian index of aged mice was considerably increased (13.8% ± 0.7% versus 11.2% ± 0.8%, P = 0.03) without affecting body weight after MVA supplementation (Fig. 3b and Extended Data Fig. 3a,b). MVA supplementation also ameliorated age-related depletion in ovarian reserve and increased the number of follicles for all developmental stages, especially the antral stage (Extended Data Fig. 3b–d). MVA-treated aged mice showed an increase in estradiol (E2) (384.5 ± 33.8 pg ml⁻¹ versus 211.3 ± 16.3 pg ml⁻¹, P = 0.002) and anti-Mullerian hormone (AMH) (975.4 ± 37.7 pg ml⁻¹ versus 620.9 ± 47.8 pg ml⁻¹, P = 0.0004) levels (Extended Data Fig. 3e,f). Moreover, the number of ovulated oocytes doubled after MVA injection (9.8 ± 1.2 versus 4.5 ± 1.1, P = 0.003) (Fig. 3c). MVA treatment significantly increased the cortical F-actin assembly (2.8 ± 0.2 versus 1.0 ± 0.1, P < 0.0001) and Arp3 expression (1.7 ± 0.3 versus 1.0 ± 0.2, P = 0.05) of MI oocytes collected from aged mice (Fig. 3d–g). As shown in Extended Data Fig. 3g,h, the percentages of abnormal spindles and misaligned chromosomes were considerably lower in oocytes from MVA-treated mice than those from control mice (26.3% ± 2.3% versus 46.3% ± 6.5%, P = 0.04). Chromosome spread experiments further confirmed a markedly lower frequency of aneuploidy in oocytes after MVA injection (25.1% ± 4.3% versus 50.5% ± 2.9%, P = 0.001) (Fig. 3h,i). Moreover, more 2-cell embryos (63.8% ± 2.4% versus 43.7% ± 3.8%, P = 0.02) and blastocysts (53.8% ± 2.3% versus 39.3% ± 4.1%, P = 0.04) could be obtained from oocytes from MVA-treated mice (Fig. 3j–l). Taken together, these results suggest that MVA ameliorates cortical F-actin assembly, meiotic defects and aneuploidy in aged oocytes via GCs both in vitro and in vivo.

a, Schematic illustration of in vivo injection of MVA into 9-month-old mice. NS, normal saline. b, Ovarian index of mice in the old and MVA-injected groups. Old group: 9-month-old female mice intraperitoneally injected with normal saline every day for 30 days. MVA group: 9-month-old female mice intraperitoneally injected with 5 mg kg⁻¹ MVA every day for 30 days. The data are shown as the mean ± s.e.m. of ten independent experiments. c, The number of MII oocytes in the old (n = 54 oocytes) and MVA-injected (n = 118 oocytes) groups. The data are shown as the mean ± s.e.m. of 12 independent experiments. d, Fluorescence imaging showing F-actin expression at the oocyte cortex in the old and MVA-injected groups. Scale bar, 25 µm. e, F-actin fluorescence intensity in the old (n = 15 oocytes) and MVA-injected (n = 21 oocytes) groups. f, Fluorescence imaging showing Arp3 expression at the oocyte cortex in the old and MVA-injected groups. Scale bar, 25 µm. g, Arp3 fluorescence intensity in the old (n = 9 oocytes) and MVA-injected (n = 8 oocytes) groups. h, Chromosome spread showing representative images of aneuploid and euploid MII oocytes from the old and MVA-injected groups. i, The aneuploidy rate was measured in the old (n = 59 oocytes) and MVA-injected (n = 50 oocytes) groups. The data are shown as the mean ± s.e.m. of five independent experiments. j, Images of two-cell embryos and blastocysts from the old and MVA-injected groups. The arrowheads indicate blastocysts. Scale bar, 200 µm. k,l, Two-cell-embryo (k) and blastocyst (l) rates were measured in the old (n = 51 oocytes) and MVA-injected (n = 63 oocytes) groups. The data are shown as the mean ± s.e.m. of three independent experiments. An unpaired two-tailed Student’s t-test was used for statistical analysis. Panel a created with BioRender.com.

Source data

Full size image

MVA activates the FPP metabolic pathway in aged GCs

To further investigate how MVA supplementation improved the quality of aged oocytes, we performed RNA sequencing analysis of oocytes and neighboring GCs from aged COCs treated with or without MVA (Fig. 4a). Principal component analysis (PCA) and comparison between the two groups revealed a highly dynamic change in gene expression after MVA supplementation. A total of 2,116 downregulated and 1,287 upregulated differentially expressed genes (DEGs) were identified in oocytes from aged COCs after MVA supplementation (Extended Data Fig. 4a,b). Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis of the DEGs revealed that genes associated with the cell cycle and oocyte meiosis were upregulated, whereas genes associated with ribosomes and thermogenesis were downregulated in oocytes from MVA-treated aged COCs (Fig. 4b and Extended Data Fig. 4c). Many meiosis-associated genes such as Cdc20, Bub3, Ccnb1 and Mos have important roles during oocyte meiosis, and the downregulation of these genes results in meiotic defects and aneuploidy^32,33,34. As shown in Fig. 4c, the expressions of Ccna2, Cdc20, Bub3, Ccnb1, Ppp2r1a, Mos, Mad2l1, Pk1, Map2K1, Cdc42ep2 and Cdc42se2 were markedly increased in oocytes from MVA-treated aged COCs.

a, Schematic showing the collection of oocytes and GCs from aged mice (10 months old) for RNA sequencing (RNA-seq). CTL, control. b, KEGG analysis of the upregulated DEGs in MO groups compared with the OO groups. OO group: oocytes from aged COCs cultured in MEMα maturation medium. MO group: oocytes from aged COCs cultured in MEMα maturation medium with 50 µM MVA. Statistical significance was determined by two-sided Fisher’s precision probability test. c, Violin plot showing the expression levels of meiosis-associated genes in OO groups and MO groups. d, PCA plot of GCs in the OGC groups and MGC groups based on gene expression patterns separated by PC1 and PC2. OGC group: GCs from aged COCs cultured in MEMα maturation medium. MGC group: GCs from aged COCs cultured in MEMα maturation medium with 50 µM MVA. e, Violin plot showing the expression levels of MVK, PMVK, MVD, IDI1 and FDPS in OGC groups and MGC groups based on the RNA-seq results. f, mRNA levels of the MVK, FDPS, GGPPS and SQLE genes in aged GCs from the OGC groups and MGC groups. The data are shown as the mean ± s.e.m. of ten independent experiments. g, MVK and FDPS protein expression in KGN cells (human granulosa-like tumor cell line) treated with 50 µM MVA. h,i, Box plot showing the levels of MVA (h) and FPP (i) in primary mouse GCs treated with or without 50 µM MVA. The data are shown as the mean ± s.e.m. of four independent experiments. j, Prenylation levels in aged oocytes treated with 50 µM MVA. The membrane was incubated with rabbit polyclonal anti-farnesyl. The molecular weight of small GTPases is around 26 kDa (arrow). Data are presented as the mean ± s.e.m. An unpaired two-tailed Student’s t-test was used for statistical analysis. Panel a created with BioRender.com.

Source data

Full size image

To investigate how MVA improved cortical F-actin assembly and meiosis in aged oocytes via GCs, we further analyzed the transcriptome of GCs. PCA and volcano plotting revealed differences in the transcriptome profile of the MVA-treated GCs and control GCs, with 9,317 downregulated and 1,279 upregulated DEGs (Fig. 4d and Extended Data Fig. 4d). KEGG analysis revealed that genes involved in ribosome, oxidative phosphorylation and proteasome were upregulated, while genes associated with neuroactive ligand‒receptor interaction and calcium signaling pathway were downregulated after MVA supplementation (Extended Data Fig. 4e,f). Because MVA is an early rate metabolite in the cholesterol synthesis pathway¹⁶, we next analyzed the expression of metabolic genes related to the MVA pathway (Extended Data Fig. 5a). The expression of genes involved in the metabolism of MVA to FPP, including MVK, PMVK, MVD, IDI1 and FDPS, was increased 1.8- to 4.0-fold in GCs after MVA supplementation (Fig. 4e and Extended Data Fig. 5a). However, genes encoding metabolic enzymes at biosynthetic steps from acetyl-CoA to MVA and FPP to cholesterol did not show obvious changes in expression (Extended Data Fig. 5a–c). Quantitative real-time reverse transcription PCR (qRT-PCR) and western blotting analysis also confirmed that both the mRNA and protein levels of MVK and FDPS were significantly elevated after MVA supplementation (Fig. 4f,g and Extended Data Fig. 5d). Moreover, the MVA (896.8 ± 7.6 ng ml⁻¹ versus 506.6 ± 42.5 ng ml⁻¹, P = 0.001) and FPP (0.17 ± 0.005 ng ml⁻¹ versus 0.14 ± 0.007 ng ml⁻¹, P = 0.03) metabolite levels were considerably increased after MVA supplementation (Fig. 4h,i and Supplementary Table 2). Furthermore, protein prenylation levels were reduced in oocytes from both atorvastatin-treated young COCs and aged COCs compared with those from untreated young COCs (Extended Data Fig. 5e,f). Conversely, MVA supplementation restored the protein prenylation levels in both KGN cells (human granulosa-like tumor cell line) and oocytes from aged COCs (Fig. 4j and Extended Data Fig. 5g). Taken together, our results suggest that MVA supplementation activates the FPP synthesis pathway and increases the protein prenylation levels in aged COCs.

MVA promotes cortical F-actin assembly via prenylation by FPP from GCs

To clarify whether the effects of MVA on oocyte cortical F-actin assembly and meiosis was mediated by prenylation via FPP synthesis from GCs, aged COCs were isolated from 10-month-old female mice and subsequently cultured in IVM medium supplemented with farnesol (FOH), which can be catabolized to FPP in cells³⁵. FOH supplementation promoted cortical F-actin assembly (0.6 ± 0.009 versus 0.2 ± 0.03, P < 0.0001) in MI oocytes, increased the rate of PBE (89.5% ± 2.5% versus 70.2% ± 1.9%, P = 0.004) and reduced the incidence of meiotic defects (23.9% ± 2.5% versus 43.3% ± 1.7%, P = 0.003) in oocytes from aged COCs, similar to MVA treatment (Fig. 5a–d and Extended Data Fig. 6a,b). To further explore the relationships between prenylation and oocyte quality in aged COCs, FTI-277, an inhibitor of protein prenylation, was added to IVM medium supplemented with MVA. Indeed, the addition of FTI-277 reduced cortical F-actin assembly (0.5 ± 0.03 versus 1.0 ± 0.07, P < 0.0001), decreased the PBE rate (70.0% ± 1.8% versus 86.8% ± 3.4%, P = 0.01) and increased the incidence of meiotic defects (46.0% ± 0.2% versus 20.5% ± 1.5%, P < 0.0001) in oocytes from aged COCs in the presence of MVA (Fig. 5a–d and Extended Data Fig. 6a,b). Moreover, supplementation with geranylgeraniol (GGOH), another activator of protein prenylation, rescued cortical F-actin assembly, PBE rates and meiotic defect rates in oocytes from aged mice to a similar extent as MVA or FOH treatment, while cholesterol failed to rescue these defects (Extended Data Fig. 6c–i).

a, Fluorescence imaging showing F-actin expression at the oocyte cortex in the old, MVA, MVA + FTI and FOH groups. Old group: COCs from aged mice (10 months old) cultured in MEMα maturation medium; MVA group: COCs from aged mice (10 months old) cultured in MEMα maturation medium supplemented with 50 µM MVA; MVA + FTI group: COCs from aged mice (10 months old) cultured in MEMα maturation medium supplemented with 50 µM MVA and 10 µM FTI-277; FOH group: COCs from aged mice (10 months old) cultured in MEMα maturation medium supplemented with 10 µM FOH. Scale bar, 25 µm. b, F-actin fluorescence intensity in the old (n = 25 oocytes), MVA (n = 28 oocytes), MVA + FTI (n = 22 oocytes) and FOH (n = 43 oocytes) groups. c, Images of oocytes isolated from aged COCs of 10-month-old mice after 14 h of maturation in the old, MVA, MVA + FTI and FOH groups. Scale bar, 100 µm. d, Rate of PBE in the old (n = 44 oocytes), MVA (n = 56 oocytes), MVA + FTI (n = 43 oocytes) and FOH (n = 38 oocytes) groups. The data are shown as the mean ± s.e.m. of three independent experiments. e, Synthesis of the prenylation reporter alk-FOH. PPTS, pyridinium 4-toluenesulfonate. f, Western blotting analysis showing prenylated proteins in KGN cells treated with 0, 20, 50 and 100 µM alk-FOH. g, Western blotting analysis of prenylated protein expression around 26 kDa in the CTL, alk-FOH and alk-FOH + FOH groups. CTL group: KGN cells; alk-FOH group: KGN cells treated with 50 µM alk-FOH; alk-FOH + FOH group: KGN cells treated with 50 µM alk-FOH and 50 µM FOH. The data are shown as the mean ± s.e.m. of three independent experiments. h, Schematic of mouse DO and COC labeling. DOs and COCs were incubated with 50 µM alk-FOH for 14 h. i, Fluorescence imaging showing the alk-FOH signals (FITC labeled) in the gap junction from COCs after incubation with 50 µM alk-FOH for 14 h. The green particles indicated by the white arrowheads represent transported FPP between GCs and oocyte. The yellow fibers indicated by the white arrowheads denote the gap junctions between GCs and oocyte. Three experiments were repeated independently with similar results. Scale bar, 5 µm. Data are presented as the mean ± s.e.m. An unpaired two-tailed Student’s t-test was used for statistical analysis. Panels e and h created with BioRender.com.

Source data

Full size image

To investigate whether FPP can be transferred from GCs to oocytes, an alkynyl-FOH chemical reporter based on FOH (alk-FOH) was used (Fig. 5e). To study the incorporation of alk-FOH, KGN cells were first incubated with different concentrations of alk-FOH for 24 h, then the total cellular protein was ligated to azide biotin via Cu-catalyzed azide–alkyne cyclo addition (CuAAC) (Extended Data Fig. 7a). Western blotting analysis revealed that the alk-FOH was incorporated at 20 μM and increased considerably with 100 μM (Fig. 5f). Moreover, alk-FOH incorporation was sensitive to competition by the natural metabolite FOH, suggesting that addition of the alkyne group did not compromise the activity of FOH in cells (Fig. 5g and Extended Data Fig. 7b). Next, we isolated young DOs and COCs from 3-week-old female mice, cultured them with alk-FOH for 14 h and visualized them using fluorescence imaging (Fig. 5h). No signals were detected in DOs, indicating that oocytes could not directly incorporate alk-FOH (Extended Data Fig. 7c). Interestingly, obvious fluorescence signals were detected in both GCs and oocytes from young COCs (Extended Data Fig. 7d). Furthermore, alk-FOH signals were also observed in the transzonal projections connecting GCs and the oocytes at the MI stage (Fig. 5i), whereas no detectable signals were present in the controls (Extended Data Fig. 7e). Collectively, these results indicate that MVA activates the synthesis of FPP in GCs and that the transfer of FPP from GCs to oocytes ameliorates aged oocyte cortical F-actin assembly and quality through prenylation.

Prenylation increases CDC42 and RAC1 cortical localization and F-actin assembly

Prenylation involves the covalent addition of either farnesyl or geranylgeranyl isoprenoids to conserved cysteine (Cys) residues at the carboxyl-terminal CaaX or C(X)C motifs (Fig. 6a). Immunofluorescence analysis revealed that the KGN cells also incorporate alk-FOH (Extended Data Fig. 8a). To identify potential prenylated proteins involved in cortical F-actin assembly and oocyte meiosis, we performed large-scale profiling of prenylated proteins in KGN cells using alk-FOH. After incubation with azide biotin, affinity enrichment and elution, liquid chromatography–tandem mass spectrometry (LC‒MS/MS) was used to identify prenylated proteins (Extended Data Fig. 7a). Western blotting analysis revealed that a diverse range of proteins, especially a cluster of approximately 26-kDa proteins, were labeled by alk-FOH (Fig. 6b). A total of 140 putative prenylated proteins were identified by alk-FOH supplementation when compared with the control group with two independent experiments (Supplementary Table 3); 39 of these proteins had a carboxyl-terminal CaaX or Rab motif (Extended Data Fig. 8b). In addition, 52% (73/140) of the detected prenylated proteins were membrane-associated proteins (Extended Data Fig. 8c). Prenylation can facilitate the anchoring of proteins to cell membranes, mediating protein‒protein interactions and signal transduction²⁴. A Venn diagram showed that 24 proteins labeled by alk-FOH, which included CDC42 and RAC1, had a carboxyl-terminal CaaX or Rab motif and localized to the cell membrane (Extended Data Fig. 8d,e). KEGG analysis revealed that 12 detected prenylated proteins were involved in the regulation of actin cytoskeleton, also including CDC42 and RAC1 (Fig. 6c).

a, Schematic showing prenylated proteins with a CaaX motif. n = 2, farnesylated proteins; n = 3, geranylgeranylated proteins. b, Western blotting analysis of proteins after pull-down experiments in KGN cells. Three experiments were repeated independently with similar results. c, KEGG analysis of the 140 prenylated proteins. P_adjust, P value adjusted. d, Western blotting analysis validated the labeling of the prenylated proteins CDC42 and RAC1 in KGN cells. PD, pull-down; TL, total. e, CDC42 and RAC1 protein expression in the membrane fractions of CTL, MVA, MVA + FTI and FOH groups. CTL group: KGN cells cultured in DMEM/F12 medium; MVA group: KGN cells cultured in DMEM/F12 medium supplemented with 50 µM MVA; MVA + FTI group: KGN cells cultured in DMEM/F12 medium supplemented with 50 µM MVA and 10 µM FTI-277; FOH group: KGN cells cultured in DMEM/F12 medium supplemented with 10 µM FOH. f, Western blotting analysis of CDC42 and RAC1 expression in the membrane fractions of CTL, MVA, MVA + FTI and FOH groups. The data are shown as the mean ± s.e.m. of three independent experiments. g,h, Endogenous CDC42–N-WASP, CDC42–Arp2 and CDC42–Arp3 interactions were detected by immunoprecipitation analysis in KGN cells with or without FOH treatment (g), and relative protein expression was determined (h). The data are shown as the mean ± s.e.m. of three independent experiments. i,j, Endogenous RAC1–WAVE2, RAC1–Arp2 and RAC1–Arp3 interactions were detected by immunoprecipitation analysis in KGN cells with or without FOH treatment (i), and relative protein expression was determined (j). The data are shown as the mean ± s.e.m. of three independent experiments. k, Fluorescence imaging showing Arp3 expression at the oocyte cortex in the old, MVA, MVA + FTI and FOH groups. Old group: COCs from aged mice (10 months old) cultured in MEMα maturation medium; MVA group: COCs from aged mice (10 months old) cultured in MEMα maturation medium supplemented with 50 µM MVA; MVA + FTI group: COCs from aged mice (10 months old) cultured in MEMα maturation medium supplemented with 50 µM MVA and 10 µM FTI-277; FOH group: COCs from aged mice (10 months old) cultured in MEMα maturation medium supplemented with 10 µM FOH. Scale bar, 25 µm. l, Arp3 fluorescence intensity in the old (n = 15 oocytes), MVA (n = 15 oocytes), MVA + FTI (n = 15 oocytes) and FOH groups (n = 15 oocytes). Data are presented as the mean ± s.e.m. An unpaired two-tailed Student’s t-test was used for statistical analysis. Panel a created with BioRender.com.

Source data

Full size image

CDC42 and RAC1 are ~26-kDa, small GTP-binding proteins of the Rho subfamily and have critical roles in the establishment of oocyte polarity and subsequent meiosis^36,37. To further validate whether CDC42 and RAC1 could be prenylated, we conducted immunoblot analysis using affinity enrichment of the labeled proteins. Both of these proteins could be labeled by alk-FOH, and the labeling was sensitive to competition with natural FOH, indicating that CDC42 and RAC1 were indeed prenylated (Fig. 6d). Next, we examined whether MVA affected CDC42 and RAC1 membrane partitioning and cellular localization via prenylation. As shown in Fig. 6e,f, MVA supplementation considerably increased the localization of CDC42 and RAC1 to the cell cortex, similar to the effect observed following treatment with FOH. As expected, supplementation with exogenous FTI-277 markedly reduced the MVA-induced anchoring of CDC42 and RAC1 to cell membranes, suggesting that MVA induces CDC42 and RAC1 prenylation and promotes their localization to the cell cortex.

CDC42 and RAC1 are enriched in the oocyte cortex after chromosome migration to the periphery and regulate actin dynamics during oocyte meiosis^36,37. To investigate the relationship between CDC42 and RAC1 prenylation and cortical F-actin assembly, we examined CDC42 and RAC1 localization in oocytes at MI by immunofluorescence. CDC42 and RAC1 failed to localize to the cortex in oocytes from aged COCs, and MVA supplementation induced CDC42 and RAC1 relocation to the cortex in oocytes from aged COCs, showing a distribution pattern similar to that of F-actin, while the ameliorative effect of MVA was disrupted by FTI-277 (Extended Data Fig. 9a–d). Next, we perturbed the CDC42/RAC1 signaling by overexpressing a CDC42/RAC1 dominant-negative mutant (CDC42-C188Y/RAC1-C189Y; which is the conserved position to the cysteine^188/189 at the CVLL/CLLL motif to tyrosine respectively; previously shown to inhibit CDC42/RAC1 prenylation)^38,39. Expressing constitutively active CDC42 and RAC1 mutant failed to locate on the oocyte cortical F-actin, suggesting that CDC42 and RAC1 prenylation is essential for cortical F-actin assembly (Extended Data Fig. 9e,f). The Arp2/3 complex can be activated by the CDC42–N-WASP or RAC1–WAVE2 signaling pathways to promote the cortical F-actin assembly in oocytes^12,40. Reciprocal coimmunoprecipitation analysis further confirmed the endogenous interaction between CDC42 with N-WASP and the Arp2/3 complex, as well as the interaction between endogenous RAC1, WAVE2 and the Arp2/3 complex. In addition, FOH supplementation significantly enhanced the formation of the CDC42–N-WASP–Arp2/3 and RAC1–WAVE2–Arp2/3 complexes (Fig. 6g–j). Moreover, FOH supplementation significantly increased the Arp3 expression (2.9 ± 0.4 versus 1.0 ± 0.2, P = 0.0004) at the cortex of MI oocytes from aged COCs, while the addition of exogenous FTI-277 reduced the cortical Arp3 expression (1.1 ± 0.2 versus 2.6 ± 0.3, P = 0.0003) (Fig. 6k,l). GGOH supplementation also increased the cortical Arp3 expression while cholesterol had no ameliorative effect (Extended Data Fig. 9g,h). Together, these results suggest that MVA or FOH promotes the formation of the CDC42–N-WASP–Arp2/3 and RAC1–WAVE2–Arp2/3 complexes for cortical F-actin assembly in oocytes from aged COCs. We found that prenylation of CDC42 and RAC1 by FPP facilitates their localization to the membrane and contributes to cortical F-actin assembly in oocytes.

The isopentenyl compound 8-IPF facilitates cortical F-actin assembly in aged oocytes

Although MVA can effectively ameliorate ovarian reserve and oocyte quality in aged mice, currently, there are no drugs containing MVA in the clinic. To improve age-related infertility, approved and preclinical drugs were screened for their efficacy in increasing the expression of MVK and FDPS and protein prenylation. Among them, we identified 8-IPF, a traditional natural chemical used in China with an isopentenyl side chain⁴¹. Treatment with 50 µg l⁻¹ 8-IPF considerably increased the prenylation level and the expression of MVK and FDPS in KGN cells (Fig. 7a and Extended Data Fig. 10a,b). In addition, 8-IPF supplementation also increased CDC42 and RAC1 cortical localization, similar to the results observed following MVA treatment in KGN cells (Fig. 7b,c).

a, Prenylation levels in the CTL and 8-IPF groups. The molecular weight of small GTPases is around 26 kDa (arrow). 8-IPF group: KGN cells treated with 50 µg l⁻¹ 8-IPF. b, CDC42 and RAC1 protein expression in the membrane fractions of the CTL and 8-IPF groups. c, Western blotting analysis of CDC42 and RAC1 expression in the membrane fractions of the CTL and 8-IPF groups. The data are shown as the mean ± s.e.m. of three independent experiments. d, Schematic of 8-IPF supplementation in vivo. e, Follicle counts of old and 8-IPF-treated mouse ovaries. Old group: 9.5-month-old female mice intragastrically gavaged with normal saline for 14 days. 8-IPF group: 9.5-month-old female mice intragastrically gavaged with 5 mg kg⁻¹ d⁻¹ 8-IPF for 14 days. The data are shown as the mean ± s.e.m. of seven independent experiments. f, Oocyte immunofluorescence showing the expression of F-actin at the oocyte cortex from the old and 8-IPF groups. Scale bar, 25 µm. g, Oocyte immunofluorescence intensity of F-actin in the old (n = 23 oocytes) and 8-IPF (n = 15 oocytes) groups. h, Fluorescence imaging showing Arp3 expression at the oocyte cortex in the old and 8-IPF groups. Scale bar, 25 µm. i, Arp3 fluorescence intensity in the old (n = 13 oocytes) and 8-IPF (n = 12 oocytes) groups. j, Images showing aneuploidy in MII oocytes from old groups and euploid MII oocytes from 8-IPF groups. k, Histogram showing the incidence of aneuploidy in MII oocytes from the old (n = 35 oocytes) and 8-IPF (n = 29 oocytes) groups. The data are shown as the mean ± s.e.m. of three independent experiments. l, Fertility of mice treated with normal saline or 8-IPF. m, Pregnancy rates in the old (n = 7) and 8-IPF-treated (n = 7) mice. n, Litter sizes in old (n = 4) and 8-IPF-treated (n = 6) mice. Data are presented as the mean ± s.e.m. An unpaired two-tailed Student’s t-test was used for statistical analysis. Panel d created with BioRender.com.

Source data

Full size image

We next examined the therapeutic effects of 8-IPF on ovarian aging in vivo using aged mice. Briefly, 5 mg kg⁻¹ 8-IPF was intragastrically administered to 9.5-month-old female mice for 14 days via gavage (Fig. 7d). The ovarian index (20.5% ± 1.1% versus 14.1% ± 0.9%, P = 0.0001) and the numbers of follicles considerably increased for all developmental stages after 8-IPF supplementation. Improvement effects were especially evident in growing follicles (secondary follicles (235.7 ± 28.4 versus 87.1 ± 21.7, P = 0.001) and antral follicles (76.4 ± 6.5 versus 22.9 ± 5.2, P < 0.0001)) (Fig. 7e and Extended Data Fig. 10c–e). Aged mice treated with 8-IPF showed an increase in E2 (393.5 ± 46.3 pg ml⁻¹ versus 267.5 ± 14.8 pg ml⁻¹, P = 0.02) and AMH (1,205 ± 58.7 pg ml⁻¹ versus 914.1 ± 36.3 pg ml⁻¹, P = 0.001) levels (Extended Data Fig. 10f,g). Moreover, the number of ovulated oocytes increased significantly after 8-IPF treatment (10.4 ± 1.3 versus 3.7 ± 0.3, P = 0.0002) (Extended Data Fig. 10h). More importantly, cortical F-actin staining of MI oocytes from aged COCs isolated from 8-IPF-treated mice was stronger than those from control oocytes (1.3 ± 0.1 versus 1.0 ± 0.1, P = 0.04) (Fig. 7f,g). In addition, the decreased Arp3 expression in the aged MI oocyte cortex was rescued after 8-IPF treatment (2.6 ± 0.4 versus 1.0 ± 0.1, P = 0.001) (Fig. 7h,i). Western blotting results further confirmed the increased Arp3 in MI oocytes from 8-IPF-treated aged COCs (1.3 ± 0.04 versus 1.0 ± 0.05, P = 0.006) (Extended Data Fig. 10i,j). When compared with the control, 8-IPF treatment markedly reduced the incidence of meiotic defects (32.4% ± 3.1% versus 54.6% ± 2.4%, P = 0.005) and aneuploidy (28.3% ± 5.1% versus 53.9% ± 2.1%, P = 0.01) (Fig. 7j,k and Extended Data Fig. 10k,l). Moreover, more 2-cell embryos (63.7% ± 4.7% versus 46.6% ± 1.7%, P = 0.03) and blastocysts (52.2% ± 4.7% versus 24.3% ± 7.3%, P = 0.03) could be obtained after 8-IPF treatment (Extended Data Fig. 10m–o). Excitingly, after natural mating with 8-week-old male mice, the pregnancy rate (85.7% versus 57.1%) and litter size (6.3 ± 0.3 versus 2.0 ± 0.6, P = 0.0001) were significantly higher in aged female mice subjected to 8-IPF treatment than in control aged mice (Fig. 7l–n). Overall, 8-IPF ameliorates ovarian reserve depletion and oocyte meiotic defects, improving fertility outcomes in aged female mice by increasing CDC42 and RAC1 prenylation and cortical F-actin assembly in aged oocytes.

The identification of effective methods to improve the quality of aged oocytes is a grand challenge in reproductive medicine. We have revealed the correlations between the MVA pathway and oocyte aging in our previous work¹⁷. In this study, we further show that MVA is the critical metabolite in the MVA pathway that regulates the oocyte meiotic process and MVA supplementation could promote cortical F-actin assembly and reduce meiotic defects in oocytes from COCs of aged mice. Furthermore, we showed that the decline in MVA pathway metabolites transferred from GCs to oocytes downregulates small GTPase prenylation, resulting in impaired F-actin cortex formation and ultimately contributing to age-related meiotic defects. Notably, we identified 8-IPF, an orally bioavailable compound derived from the traditional Chinese medicine Epimedium brevicornu Maxim, as a potent prenylation-activating agent. Strikingly, 8-IPF effectively restores aged oocyte quality, providing a clinically therapeutic strategy to ameliorate age-related infertility.

Oocytes and surrounding GCs form follicles, which constitute the fundamental reproductive units in the ovary. In addition, GCs establish a metabolic community with oocytes and transmit metabolites such as pyruvate, amino acids and cholesterol for oocyte development and maturation^42,43,44. Our recent study reported that GCs exhibit a robust increase in MVA pathway gene expression during oocyte meiotic resumption, whereas oocytes show low expression of these regulatory genes¹⁷. This finding indicates that active MVA pathway metabolic synthesis occurs in GCs, in which MVA pathway metabolites are subsequently supplied to oocytes for meiosis. Our study further confirmed the metabolic coupling between GCs and oocytes using alk-FOH chemical reporters. Fluorescent labeling of FOH metabolism showed that active uptake of FOH occurred in GCs and then FOH metabolites were transferred to oocytes through gap junctions in COCs, affecting oocyte meiosis, whereas denuded oocytes were unable to directly incorporate FOH. In addition, our previous study also showed the downregulation of MVA pathway gene expression in aged GCs¹⁷, which means that synthesis of the MVA pathway metabolites is reduced in GCs from aged COCs. Therefore, the decrease in MVA, FPP and GGPP in aged oocytes may be due to the reduced synthesis of MVA pathway metabolites in GCs from aged COCs.

Mammalian oocyte meiosis represents a distinctive form of asymmetric cell division, in which a mature egg and a diminutive polar body are produced. The cortical F-actin assembly in the oocyte is essential for oocyte polarization and the completion of chromosome segregation²⁹. The Arp2/3 complex comprises actin nucleators that promote actin formation and can be activated through the CDC42–N-WASP or RAC1–WAVE2 signaling pathways⁴⁰. CDC42 and RAC1 are members of the Rho GTPase family, a small GTPase subfamily of proteins that oscillate between the GTP-bound (active) and GDP-bound (inactive) states⁴⁵. CDC42 and RAC1 are enriched at the cortex after spindle migration, and cortical F-actin assembly is necessary for the extrusion of the first polar body during oocyte meiosis^{13,46,47,48,49,50}. We revealed that CDC42 and RAC1 have CaaX motifs at the carboxyl terminus and can be prenylated by FPP in oocytes. Prenylation facilitates the membrane localization of CDC42 and RAC1, promoting the assembly of the CDC42–N-WASP–Arp2/3 and RAC1–WAVE2–Arp2/3 complexes and subsequent assembly of cortical F-actin at the MI stage. Decreased CDC42 and RAC1 membrane localization and abnormal cortical F-actin assembly led to meiotic defects and aneuploidy in oocytes from aged COCs. More importantly, supplementation with MVA or FOH rescued oocyte quality from aged COCs by increasing prenylation and cortical F-actin assembly.

For several decades, many studies have focused on developing therapeutic strategies to improve the quality of aged oocytes. However, there are still no effective clinical methods for reducing meiotic defects and aneuploidy in aged oocytes. Recent studies have shown that the MVA pathway participates in the aging of multiple organs, such as the testis, muscle and ovary^17,51,52. Our study further shows that MVA supplementation restores aberrant F-actin distribution, reduces meiotic defects and lowers aneuploidy in oocytes derived from aged COCs. Therefore, we are conducting a single-center clinical study on supplementation of maturation media for aged COCs using MVA, a water-soluble chemical, to improve aged oocyte euploidy and quality (Clinical Trial: NCT05788822). Interestingly, we found that 8-IPF, a chemical from the traditional Chinese medicine Epimedium brevicornu Maxim, could activate the FPP synthesis pathway in GCs and promote the localization of CDC42 and RAC1 to the oocyte membrane for cortical F-actin assembly. Indeed, 8-IPF supplementation improved the quality of oocytes from aged COCs and increased female mouse fertility. Furthermore, 8-IPF can be orally administered, does not cause direct damage to the skin or mucous membranes, and has a relatively short half-life⁴¹. Therefore, with further clinical study, 8-IPF may be suitable for clinical use via oral administration to improve female fertility during reproductive aging.

In conclusion, our study revealed that decreased prenylation of CDC42 and RAC1 in oocytes affects cortical F-actin assembly during oocyte meiosis, consequently contributing to meiotic defects and aneuploidy in oocytes from aged COCs. MVA and 8-IPF supplementation is instrumental in facilitating the synthesis of FPP from neighboring GCs, enhancing the cortical localization of CDC42 and RAC1 to activate the Arp2/3 complex for normal cortical F-actin assembly in oocytes from aged COCs. Thus, our study provides insights into the mechanisms of age-related oocyte cortex abnormality and aneuploidy, and clinical directions for improving the reproductive outcomes or extending the female reproductive lifespan.

Animals and drug administration

Female C57BL/6 mice, 6 weeks old and 8 months old, and female Institute of Cancer Research (ICR) mice, 3 weeks old, were purchased from SPF Biotechnology. The mice were raised until experiments were conducted at the Animal Laboratory Center of Nanjing Drum Tower Hospital or at the Animal and Plant Center of the National Institute of Biological Sciences, Beijing under a 12-h light–dark cycle at a constant temperature (20–23 °C) and humidity (30–70%) and had free access to food and water according to institutional guidelines. Female C57BL/6 mice, 9 months old, were intraperitoneally injected with 5 mg kg⁻¹ MVA (Sigma, M4667) every day at 1800 hours for 30 consecutive days. Female C57BL/6 mice, 9.5 months old, were intragastrically gavaged with 5 mg kg⁻¹ 8-IPF (Yuanye, B21576) every day at 1800 hours for 14 consecutive days. Approval for the procedures involving mouse protocols and experiments was obtained from the Experimental Animal and Welfare Ethics Committee of Nanjing Drum Tower Hospital (2023AE01053) and from the Animal Ethics Committee of the National Institute of Biological Sciences, Beijing.

COC and DO IVM in MEMα maturation medium

Female mice were intraperitoneally injected with 10 IU pregnant mare serum gonadotropin (Sansheng Pharmaceuticals) and killed 48 h later. COCs and DOs were isolated from the ovarian antral follicles using a disposable syringe with a 20-gauge needle and subsequently cultured in MEMα (Gibco, 32561037) maturation medium covered with liquid paraffin oil in an incubator at 37 °C with 5% CO₂. The MEMα maturation medium contained 10% fetal bovine serum (Gibco, 10270106), 10 ng ml⁻¹ epidermal growth factor (Gibco, 53003-018) and 1.5 IU ml⁻¹ human chorionic gonadotropin (Sansheng Pharmaceutical). After culturing in IVM medium for 4 h and 14 h, respectively, the GVBD and PBE rates were analyzed. Subsequently, 40 µM atorvastatin (MedChemExpress, HYB0589), 50 µM MVA, 10 µM FOH (Sigma, F203), 10 µM FTI-277 (Aladdin, F331609), 10 µM GGOH (Sigma, G3278) and 100 µM cholesterol (MedChemExpress, HY-N0322A) were administered by addition to the MEMα maturation medium in this study.

Oocyte immunofluorescence

Oocytes were fixed in 4% paraformaldehyde (PFA; Sigma, 158127) for 30 min before permeabilizing in 0.5% Triton X-100 (Sigma, T9284) for 20 min. Then, blocking was performed using 1% bovine serum albumin (Sigma, 10711454001) for 1 h. Oocytes were incubated with mouse monoclonal anti-α-tubulin-FITC (Sigma, F2168; 1:200), 594-phalloidin (US Everbright, YP0052L; 1:200), rabbit polyclonal anti-Connexin 43 (GJA1) (Proteintech, 26980-1-AP; 1:1,000), mouse monoclonal anti-Arp3 (Santa Cruz Biotechnology sc-48344; 1:100), mouse monoclonal anti-CDC42 antibodies (Santa Cruz Biotechnology, sc-8401; 1:200) and mouse monoclonal anti-RAC1 antibodies (Santa Cruz Biotechnology, sc-514583; 1:200) at 4 °C overnight. The oocytes were then treated with a secondary antibody for 1 h at room temperature. After three 5-min washes in phosphate-buffered saline with 0.05% Tween 20 (PBST), the oocytes were incubated with Hoechst 33342 (Thermo Fisher Scientific, H3570) for 10 min at room temperature. Then, the oocytes were mounted on glass slides and observed under a fluorescence microscope (Leica, DM3000). The maximum cross-section of oocytes was chosen to capture and analyze under the exact same microscope settings during the same analysis session. For spindle and chromosome staining experiments, oocytes with chromosome misalignment at metaphase, anaphase lagging and chromatin bridges, clumped chromosomes or spindle defects were identified as meiotic defective oocytes. And the meiotic defect rate was determined based on the meiotic defective oocytes/oocytes × 100%. For the statistical analysis of the meiotic defect rate, at least 10 oocytes were tested in each experiment and at least three independent repeated experiments were conducted. ImageJ was used to quantify the signals of F-actin, Arp3, CDC42 and RAC1 in the oocyte cortex.

Chromosome spread

MII oocytes were first treated with Tyrode’s buffer (Sigma, T2397) for 3 min at 37 °C to remove the zona pellucida. Then, the cells were cultured in M2 medium (Sigma, M7167) for 10 min and fixed in 1% PFA with 0.15% Triton X-100 (pH 9.2) on a glass slide. After drying, the samples were incubated with a human anti-centromere antibody (Antibodies Incorporated, 15-234; 1:100) at 4 °C overnight. After three 5-min washes in PBST, the samples were incubated with Hoechst 33342 for 10 min at room temperature. Finally, the number of spread chromosomes was counted under a DM3000 LED microscope (Leica).

In vitro fertilization and embryo culture

The sperm from the epididymides of 12-week-old C57BL/6 male mice were collected and capacitated for 1 h in human tubal fluid medium (Merck Millipore, MR-070). COCs after IVM for 14 h or obtained from oviductal ampullae of C57BL/6 female mice were added to be fertilized by the addition of capacitated sperm for 6 h in a 37 °C incubator with 5% CO₂. The fertilized oocytes were subsequently transferred to KSOM (Merck Millipore, MR-106-D) for subsequent culture. The two-cell embryo and blastocyst formation rates were subsequently calculated.

Metabolite extraction and targeted metabolomics analysis

Metabolites were extracted using a method described previously with some modifications⁵³. Oocytes collected from mice and primary GCs supplemented with or without MVA were used to detect MVA pathway metabolites. In brief, cells were washed with PBS after isolation before quenching. To quench cells and stop metabolism, cells were collected in tubes and quick-frozen in liquid nitrogen immediately after washing. Then they were incubated in an extraction buffer containing 80% MS-grade methanol and 2 μg ml⁻¹ 4-CL-phenylalanine (Sigma, C6506) as an internal standard, which was precooled in a −80 °C freezer. Metabolites were extracted by three rounds of bead-beating. After two rounds of centrifugation at 15,000 g for 10 min at 4 °C, the supernatant fraction containing soluble metabolites was collected and dried using a CentriVap Concentrator system (Labconco FreeZone 6).

For running samples, a dried metabolite extract sample was resuspended in 50% MS-grade acetonitrile for injection. Levels of MVA pathway metabolites such as MVA, FPP and GGPP were quantitatively analyzed in negative ion mode using multiple reaction monitoring acquisition with a triple quadrupole mass spectrometer (Triple Quad 6500+, AB SCIEX) coupled to a high-performance liquid chromatograph. A mix standard containing MVA, FPP (Sigma, F6892) and GGPP (Sigma, G6025) with series dilution for 20 ng ml⁻¹, 40 ng ml⁻¹, 200 ng ml⁻¹, 500 ng ml⁻¹, 2 μg ml⁻¹ and 5 μg ml⁻¹ was used to make standard curves. Metabolites were separated chromatographically on a UPLC HSS T3 column (ACQUITY, 1.8 μm, 150 × 2.1 mm, Waters). The flow rate was set to 0.25 ml min⁻¹ using the following method: buffer A: 10 mM ammonium carbonate, buffer B: 100% acetonitrile. T = 0 min, 10% B; T = 1 min, 10% B; T = 4 min, 65% B; T = 6 min, 65% B; T = 6.5 min, 95% B; T = 8.5 min, 95% B; T = 9 min, 10% B; T = 12 min, 10% B stop. The retention time for each multiple reaction monitoring peak was compared with an appropriate standard. The area under each peak was then quantitated by using SCIEX OS software and re-inspected for accuracy.

RNA-seq library construction and analysis

Oocytes and GCs isolated from COCs were lysed, and cDNA was synthesized using a Discover-sc WTA Kit V2 (Vazyme Biotech, N712). Sequencing libraries were prepared with the TruePrep DNA Library Prep Kit V2 for Illumina (Vazyme Biotech, TD503) and sequenced on the Illumina HiSeq X platform (Nanjing Genemap). High-quality reads were aligned to the Mus musculus UCSC mm9 reference genome, and gene expression levels were quantified as fragments per kilobase of exon model per million mapped fragments (FPKM) values. Highly variable genes (coefficient of variation > 1) were selected for PCA, and the resulting PCA plot was generated using the ggplot2 package in R studio. The DESeq2 package was used to identify DEGs. KEGG enrichment analysis was performed with KEGG Mapper.

qRT-PCR

All primers were mixed in nuclease-free water to a final concentration of 0.1 μM as a primer assay pool. For each mixture, 5 μl of solution was prepared as follows: 2.5 μl of reaction mixture, 0.5 μl of primer mixture, 0.1 μl of RT-Taq mixture and 1.9 μl of nuclease-free water (Vazyme Biotech, P621). GCs isolated from COCs were added to the mixture and placed in a −80 °C freezer for 2 min. The mixture was centrifuged at 1,000 g for 2 min, after which PCR was performed. For qRT-PCR, 5 μl of SYBR-Green (Vazyme Biotech, Q121), 0.5 μl of primer-forward (10 μM), 0.5 μl of primer-reverse (10 μM), 2 μl of cDNA and 2 μl of ddH₂O were mixed. The following primer sequences were used: MVK (mouse): forward, 5′-AGCGTCAATTTACCCAACATCG-3′, reverse, 5′-GAGACATCACCTTGCTCAAGAAA-3′; FDPS (mouse): forward, 5′-GGGTTTGACCGTGGTACAAG-3′, reverse, 5′-AAGCCTGGAGCAGTTCTACAC-3′; GGPPS (mouse): forward, 5′-TTTTGCATACACTCGACACACT-3′, reverse, 5′-GGCCTCAATTTGTTTGTAGGCT-3′; SQLE (mouse): forward, 5′-GACCTCGTTCGTGACGGAC-3′, reverse, 5′-CTCCCCAACTATCCTGTCGG-3′; and 18S (mouse): forward, 5′-ATGGCCGTTCTTAGTTGGTG-3′, reverse, 5′-CGGACATCTAAGGGCATCAC-3′. All data were normalized to the expression of 18S using the comparative 2^−ΔΔCt method.

Cell culture and metabolic labeling

Human granulosa-like tumor cell line KGN cells and primary mouse GCs were cultured in DMEM/F12 (Gibco, C11330500BT) supplemented with 10% fetal bovine serum (Gibco) and 1% penicillin‒streptomycin (Gibco, 15140122) in a humidified 37 °C incubator with 5% CO₂. KGN cells were grown in 10-cm plates to approximately 70% confluence and treated with alk-FOH (20 μM, 50 μM or 100 μM; from 100 mM stock solution in DMSO) or DMSO only for 48 h. For antagonistic coincubation, both 50 μM alk-FOH and 50 μM FOH (Sigma, F203) were added to KGN cells for 48 h. Then, the cells were washed three times with cold phosphate-buffered saline. The cells were lysed for western blotting and pull-down assays.

Western blotting

The proteins were quantified using a BCA protein assay kit (Thermo Fisher Scientific, 23227) following the manufacturer’s instructions. Next, the equivalent proteins were separated via 10% SDS-PAGE and transferred to polyvinylidene difluoride membranes (Merck Millipore, 03010040001). The membranes were incubated with primary antibodies, rabbit polyclonal anti-farnesyl (Invitrogen, PA1-12554; 1:1,500), FDPS (Abcam, ab153805; 1:1,000), MVK (Proteintech, 12228-1-AP; 1:1,000), rabbit monoclonal anti-CDC42 (Abcam, ab187643; 1:1,000), rabbit monoclonal anti-RAC1 (Cell Signaling Technology, 4651S; 1:1,000), rabbit monoclonal anti-Arp2 (Abcam, ab128934; 1:1,000), rabbit monoclonal anti-Arp3 (Abcam, ab181164; 1:1,000), rabbit monoclonal anti-N-WASP (Abclonal, A2270; 1:1,500), mouse monoclonal anti-WAVE2 (Abclonal, A19601; 1:1,500), rabbit monoclonal anti-β-actin (Bioworld, AP0060; 1:100,000), rabbit polyclonal anti-Calnexin (Proteintech, 10427-2-AP; 1:10,000), mouse monoclonal anti-α-tubulin (Proteintech, 66031-1-Ig; 1:10,000) and mouse monoclonal anti-GAPDH (Proteintech, 60004-1-Ig; 1:50,000) at 4 °C overnight after blocking with 5% nonfat milk in PBST for 1 h at room temperature. The next day, the polyvinylidene difluoride membranes were washed three times with PBST, incubated with HRP-conjugated goat anti-rabbit IgG (Abcam, ab97051; 1:10,000) at room temperature for 1 h, washed three times with PBST and then developed using enhanced chemiluminescence reagents (GE); the mean gray value was estimated with ImageJ software (NIH, version 1.0).

Labeling of prenylated proteins with biotinylated probes via CuAAC reaction

Metabolic labeling and protein extraction were performed as described above. Protein concentrations were adjusted to 2.2 mg ml⁻¹. For each, 3,000 μl of protein lysate, click reagent mixture was prepared as follows: 2,520 μl of protein (2.2 mg ml⁻¹), 60 μl of azide biotin (Confluore, 908007-17-0) (0.1 mM, 5 mM stock solution in DMSO), 120 μl of ‌2-(4-((bis((1-(tert-butyl)-1H-1,2,3-triazol-4-yl)methyl)amino)methyl)-1H-1,2,3-triazol-1-yl)acetic acid (BTTAA)–CuSO₄ (2:1, 1 mM:0.5 mM) and 300 μl of fresh sodium ascorbate (2.5 mM, 25 mM stock solution in PBS), which were added last. Then, the reaction mixture was incubated on a shaker at room temperature for 3 h.

Streptavidin enrichment of biotinylated proteins

After the CuAAC reaction, the biotin-labeled lysates were precipitated overnight in methanol at −40 °C. The next day, the samples were centrifuged at 5,000 g for 15 min at 4 °C and subsequently washed twice with 7.5 ml of precooled methanol. The precipitated proteins were subsequently completely dissolved in 1.67 ml of buffer (1.2% SDS in PBS) after the methanol was discarded, after which the mixture was left at room temperature for 20 min of volatilization. Then, 100 μl of Pierce High Capacity Streptavidin Agarose (Thermo Fisher Scientific, 20359) was washed with PBS three times and resuspended in 8.33 ml of PBS in a 15-ml tube. The redissolved proteins were added to 15-ml tubes filled with the agarose and incubated gently with rotation at room temperature for 4 h. The nonspecific binding proteins were then washed with 0.2% SDS in PBS (10 ml, once), PBS (10 ml, three times) and ddH₂O (10 ml, three times) and centrifuged at 500 g for 1 min, after which the beads were transferred to 1.5-ml centrifuge tubes with ddH₂O. Then, 2× loading buffer (Beyotime, P0015L) was added to the precipitated beads, which were subsequently heated at 95 °C for 12 min. Next, the samples were centrifuged, after which approximately 40 µl of 2× loading buffer was transferred to a new 1.5-ml tube to obtain the protein released from the beads. The proteins were subsequently separated via 10% SDS-PAGE for western blotting or were subjected to LC‒MS/MS for bottom-up proteomic analysis.

LC‒MS/MS analysis and data processing

The enriched protein samples were reduced with 5 mM TCEP for 30 min at 37 °C, and alkylated with 15 mM IAA for 30 min in the dark at room temperature. Each sample was trypsinized by the filter-aided proteome preparation (FASP) method. The peptides were desalted using a C18 cartridge, lyophilized and reconstituted by 40 μl of 0.1% formic acid solution. Then the peptides were analyzed using an Easy-nLC 1000 (Thermo Fisher) (buffer A: 0.1% formic acid solution; buffer B: 0.1% formic acid + 80% acetonitrile solution; Thermo Fisher). After the column was equilibrated with 95% buffer A, the sample was loaded onto a trap column. Then, the samples were separated by chromatography and analyzed via mass spectrometry (MS). MS analysis was performed with a Q Exactive mass spectrometer (Thermo Fisher). Detection parameters were used as follows: detection method, positive ion; scanning range of the precursor ion, 300–1,800 m/z; resolution of the first-order mass spectrum, 70,000 at 200 m/z; automatic gain control target, 1 × 10⁶; maximum IT, 50 ms; and dynamic exclusion time, 30.0 s. Mass-to-charge ratios of peptides and peptide fragments were collected according to the following parameters: fragment maps taken after each full scan, 20; MS2 activation type, high-collision dissociation; isolation window, 2 m/z; MS2 resolution, 17,500 at 200 m/z; normalized collision energy, 27 eV; and underfill, 0.1%. MaxQuant software (version 1.6.14) was used for database searching against UniProt_Homo_sapiens_207393_20230103. The LC–MS/MS experiments mentioned above were done in two replicates, and the identified proteins in both experiments are listed in Supplementary Table 3.

CuAAC reaction for fluorescence imaging

COCs were incubated with 50 µM alk-FOH for 9 h. COCs were fixed with 4% PFA for 30 min. Then, the COCs were incubated with 50 µM azide AZDye 488, the BTTAA–CuSO₄ complex (BTTAA/CuSO₄ 6:1) and 2.5 mM sodium ascorbate in PBS at room temperature for 30 min. After the click reaction, COCs were incubated with 594-Phalloidin and Hochest 333342 at room temperature for 30 min. Fluorescence imaging was subsequently performed under an LSM 900 confocal laser scanning microscope (Zeiss).

Total, membrane and cytoplasmic protein isolation

Total proteins were isolated by lysing cells with radioimmunoprecipitation assay lysis buffer (Beyotime, P0013B) supplemented with protease inhibitors at 4 °C for 30 min. The supernatant of the lysates was collected after centrifugation at 13,600 g for 15 min.

Cell membrane and cytoplasmic proteins were isolated using a Membrane and Cytosol Protein Extraction Kit with protease inhibitors according to the manufacturer’s instructions (Beyotime, P0033). In brief, approximately 3 × 10⁷ cells were homogenized using 1 ml of membrane protein extraction reagent A supplemented with PMSF and protease inhibitors for 30 min at 4 °C. The nuclei and unbroken cells were removed by centrifuging at 700 g for 10 min. The supernatant was then collected for further centrifugation at 13,600 g for 30 min to obtain the plasma proteins. The remaining sediment was fully vortexed in 200 µl of membrane protein extraction reagent B on ice for 10 min and subsequently centrifuged at 13,600 g for 30 min to obtain the membrane proteins.

Coimmunoprecipitation assay

KGN cells were treated with 10 µM FOH for 48 h, total protein was extracted with radioimmunoprecipitation assay buffer supplemented with protease inhibitor, and total protein was quantified with a BCA protein assay kit. Next, 500 μg of total protein was incubated with 1 μg of anti-mouse IgG, anti-CDC42 (Santa Cruz Biotechnology, sc-8401) or anti-RAC1 (Santa Cruz Biotechnology, sc-514583) at 4 °C overnight. Then, 30 μl of agarose beads was added to the cell lysates, which were incubated at 4 °C for 4 h. Subsequently, the cell lysates were incubated with 2 × loading buffer at 95 °C for 10 min. Finally, western blotting was performed to analyze the protein samples.

Expression constructs, mRNA synthesis and microinjection

The mcherry coding sequences and CDC42 or RAC1 coding sequences were fused and inserted into pGEMHE plasmids to obtain pGEMHE–mCherry–CDC42 and pGEMHE–mCherry–RAC1. pGEMHE–mCherry–CDC42-C188Y(M-CDC42) and pGEMHE–mCherry–RAC1-C189Y(M-RAC1) were obtained using a site-directed mutagenesis kit (NEB, E0554) for in vitro transcription. Briefly, after linearization of the template with AscI (NEB, R0558V), capped mRNA was synthesized using a HiScribe T7 ARCA mRNA Kit (NEB, E2065S), column-cleaned with NucleoSpin RNA Clean-up XS (Macherey-Nagel, 740903) and dephosphorylated for uncapped RNA using Antarctic phosphatase (NEB, M0289). Finally, total mRNA was purified using acidic phenol–chloroform and resuspended in 11 μl of nuclease-free H₂O for mRNA analysis. mRNA concentrations were determined by NanoDrop (Thermo Fisher Scientific).

Mouse oocytes were held between two glass coverslips in a custom-made microinjection chamber and microinjected with 3.5 pl of mRNA using a single-sided CellTram4m Oil pump (Eppendorf). All mRNAs were microinjected at a needle concentration (final concentration in the microinjection needle) of 100 ng μl⁻¹. Oocytes were allowed to express the mRNAs for 3 h before release in M2 medium containing 250 μM dbcAMP (Sigma-Aldrich). After 3 h, the oocytes were released in dbcAMP-free M2 medium and subjected to confocal imaging using the LSM 900 laser scanning microscope (Zeiss).

Body weight and ovarian index

The mice in each group were euthanized after weighing. A U-shaped incision in the hypogastrium was made to explore the main organs. The bilateral ovaries connected to the bicorned uterus via the fallopian tubes were separated, and the peri-ovarian adipose tissue was removed under a stereoscope (Leica). The ovaries were subsequently weighed on an electronic analytical balance when no liquid remained. A representative picture was taken on dry sterile gauze, and the ovarian index was determined based on the ovarian wet weight (mg)/body weight (g) × 100%. By using this formula, ovarian weight can be standardized with body weight to more accurately evaluate the ovarian reserve.

Histological analysis and follicle counting

The estrous cycles of the mice were detected by daily examination of vaginal smears, and the mice were killed in the interestrous phase to collect ovaries. Ovaries were fixed with 4% paraformaldehyde in PBS overnight; dehydrated in 70%, 80%, 90%, 95% and 100% ethanol; cleared with xylene; and embedded in paraffin. The ovarian tissues were serially sectioned at 5 µm, sequentially deparaffinized in xylene, rehydrated in a descending series of graded ethanol solutions and stained with hematoxylin and eosin.

One of every five slices was used to count the follicles. The follicles at the primordial, primary, secondary and antral stages in the histological sections of the ovaries were observed and classified based on the histological morphology under a microscope at ×200 magnification. Only follicles containing a visible nucleus were counted independently by two researchers. The result was calculated as a fivefold counting value.

Serum isolation and hormonal assay in mice

After phenobarbital anesthesia of the mice during the interestrous phase, mouse blood was collected from the eye veins. Whole blood was allowed to clot at room temperature for 30 min before overnight incubation at 4 °C. Serum was isolated by centrifugation at 3,000 g for 15 min at 4 °C.

For hormonal quantification, E2 and AMH levels were determined using commercial ELISA kits (CEA461Ge/CEA830Mu, Cloud-Clone) according to the manufacturer’s protocol. Absorbance was measured at 450 nm with wavelength using a Spark multimode microplate reader, and the sample concentration was calculated.

Fertility test

After 2 weeks of gavage, to evaluate the fertility function after 8-IPF administration, female mice were mated with 10-week-old C57BL/6 male mice with proven fertility. The next day morning, female mice were checked for the presence of vaginal plugs to determine whether they were pregnant. The female mice were considered infertile if, after continuous mating for 2 weeks, no vaginal plugs were observed. The pregnancy was terminated by cesarean section at 18.5 days of gestation, and after routine anesthesia, the uterus was exposed through a U-shaped incision in the lower abdomen. The pregnancy and embryo implanted conditions of the mice were observed and photographed, after which the embryos and placenta were separated, weighed and recorded. The number of viable offspring in each litter was recorded.

Statistics and reproducibility

No statistical methods were used to determine the sample size in advance, but we used similar sample sizes for each experiment and no data were excluded from the analyses. The specific sample sizes are provided in the figure legends. Data collection and analysis were performed without blinding to the conditions of the experiments. Data distribution was assumed to be normal, but this was not formally tested; therefore, data distributions are visualized in each figure. All statistical data are presented as the mean ± standard error of the mean (s.e.m.) unless otherwise stated. Data were compared by unpaired two-tailed Student’s t-test following the Shapiro–Wilk test for normality, unless indicated otherwise in the figure legends or methods. Values of P < 0.05 were considered statistically significant. The exact P value is presented in the corresponding figures, and statistical test methods are detailed in the figure legends. All the statistical comparisons were performed with GraphPad Prism 8.0.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

1. Broekmans, F. J., Soules, M. R. & Fauser, B. C. Ovarian aging: mechanisms and clinical consequences. Endocr. Rev. 30, 465–493 (2009).

   CAS  PubMed  Google Scholar 

2. Laisk, T. et al. Demographic and evolutionary trends in ovarian function and aging. Hum. Reprod. Update 25, 34–50 (2019).

   CAS  PubMed  Google Scholar 

3. Perheentupa, A. & Huhtaniemi, I. Aging of the human ovary and testis. Mol. Cell. Endocrinol. 299, 2–13 (2009).

   CAS  PubMed  Google Scholar 

4. Gruhn, J. R. et al. Chromosome errors in human eggs shape natural fertility over reproductive life span. Science 365, 1466–1469 (2019).

   CAS  PubMed  PubMed Central  Google Scholar 

5. Nikalayevich, E. et al. Aberrant cortex contractions impact mammalian oocyte quality. Dev. Cell 59, 841–852.e7 (2024).

   CAS  PubMed  Google Scholar 

6. Chaigne, Agathe et al. A soft cortex is essential for asymmetric spindle positioning in mouse oocytes. Nat. Cell Biol. 15, 958–966 (2013).

   CAS  PubMed  Google Scholar 

7. Chaigne, A. et al. A narrow window of cortical tension guides asymmetric spindle positioning in the mouse oocyte. Nat. Commun. 6, 6027 (2015).

   CAS  PubMed  Google Scholar 

8. Shi, L. Y. et al. NLRP4E regulates actin cap formation through SRC and CDC42 during oocyte meiosis. Cell. Mol. Biol. Lett. 29, 68 (2024).

   CAS  PubMed  PubMed Central  Google Scholar 

9. Dunkley, S. & Mogessie, B. Actin limits egg aneuploidies associated with female reproductive aging. Sci. Adv. 9, eadc9161 (2023).

   PubMed  PubMed Central  Google Scholar 

10. Deng, M. et al. The Ran GTPase mediates chromatin signaling to control cortical polarity during polar body extrusion in mouse oocytes. Dev. Cell 12, 301–308 (2007).

    CAS  PubMed  Google Scholar 

11. Yi, Kexi et al. Dynamic maintenance of asymmetric meiotic spindle position through Arp2/3-complex-driven cytoplasmic streaming in mouse oocytes. Nat. Cell Biol. 13, 1252–1258 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

12. Bement, W. M., Goryachev, A. B., Miller, A. L. & von Dassow, G. Patterning of the cell cortex by Rho GTPases. Nat. Rev. Mol. Cell Biol. 25, 290–308 (2024).

    CAS  PubMed  Google Scholar 

13. Dehapiot, Benoit et al. Polarized Cdc42 activation promotes polar body protrusion and asymmetric division in mouse oocytes. Dev. Biol. 377, 202–212 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

14. Dehapiot, Benoit et al. RhoA- and Cdc42-induced antagonistic forces underlie symmetry breaking and spindle rotation in mouse oocytes. PLoS Biol. 19, e3001376 (2021).

    CAS  PubMed  PubMed Central  Google Scholar 

15. Jeong, A., Suazo, K. F., Wood, W. G., Distefano, M. D. & Li, L. Isoprenoids and protein prenylation: implications in the pathogenesis and therapeutic intervention of Alzheimer’s disease. Crit. Rev. Biochem. Mol. Biol. 53, 279–310 (2018).

    PubMed  PubMed Central  Google Scholar 

16. Mullen, PeterJ. et al. The interplay between cell signalling and the mevalonate pathway in cancer. Nat. Rev. Cancer 16, 718–731 (2016).

    CAS  PubMed  Google Scholar 

17. Liu, C. et al. Granulosa cell mevalonate pathway abnormalities contribute to oocyte meiotic defects and aneuploidy. Nat. Aging 3, 670–687 (2023).

    CAS  PubMed  Google Scholar 

18. Tesarik, J., Galán-Lázaro, M. & Mendoza-Tesarik, R. Ovarian aging: molecular mechanisms and medical management. Int. J. Mol. Sci. 22, 1371 (2021).

    CAS  PubMed  PubMed Central  Google Scholar 

19. Liu, C. et al. Growth hormone ameliorates the age-associated depletion of ovarian reserve and decline of oocyte quality via inhibiting the activation of Fos and Jun signaling. Aging 13, 6765–6781 (2021).

    CAS  PubMed  PubMed Central  Google Scholar 

20. Zhang, H. et al. Melatonin improves the quality of maternally aged oocytes by maintaining intercellular communication and antioxidant metabolite supply. Redox Biol. 49, 102215 (2022).

    CAS  PubMed  Google Scholar 

21. Miao, Y., Cui, Z., Gao, Q., Rui, R. & Xiong, B. Nicotinamide mononucleotide supplementation reverses the declining quality of maternally aged oocytes. Cell Rep. 32, 107987 (2020).

    CAS  PubMed  Google Scholar 

22. Zhang, Y. et al. Polyamine metabolite spermidine rejuvenates oocyte quality by enhancing mitophagy during female reproductive aging. Nat. Aging 3, 1372–1386 (2023).

    CAS  PubMed  Google Scholar 

23. Liang, W. F. et al. Biosensor-assisted transcriptional regulator engineering for Methylobacterium extorquens AM1 to improve mevalonate synthesis by increasing the acetyl-CoA supply. Metab. Eng. 39, 159–168 (2017).

    CAS  PubMed  Google Scholar 

24. Chakrabarti, R. & Engleman, E. G. Interrelationships between mevalonate metabolism and the mitogenic signaling pathway in T lymphocyte proliferation. J. Biol. Chem. 266, 12216–12222 (1991).

    CAS  PubMed  Google Scholar 

25. Acharya, S., Timilshina, M. & Chang, J. H. Mevalonate promotes differentiation of regulatory T cells. J. Mol. Med. 97, 927–936 (2019).

    CAS  PubMed  Google Scholar 

26. Bekkering, S. et al. Metabolic induction of trained immunity through the mevalonate pathway. Cell 172, 135–146.e9 (2018).

    CAS  PubMed  Google Scholar 

27. Alarcon, V. B. & Marikawa, Y. Statins inhibit blastocyst formation by preventing geranylgeranylation. Mol. Hum. Reprod. 22, 350–363 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

28. Marikawa, Y., Menor, M., Deng, Y. & Alarcon, V. B. Regulation of endoplasmic reticulum stress and trophectoderm lineage specification by the mevalonate pathway in the mouse preimplantation embryo. Mol. Hum. Reprod. 27, gaab015 (2021).

    PubMed  PubMed Central  Google Scholar 

29. Sardet, C., Speksnijder, J., Terasaki, M. & Chang, P. Polarity of the ascidian egg cortex before fertilization. Development 115, 221–237 (1992).

    CAS  PubMed  Google Scholar 

30. Uraji, J., Scheffler, K. & Schuh, M. Functions of actin in mouse oocytes at a glance. J. Cell Sci. 131, jcs218099 (2018).

    PubMed  Google Scholar 

31. Sun, S. C. et al. Degradation of actin nucleators affects cortical polarity of aged mouse oocytes. Fertil. Steril. 97, 984–990 (2012).

    CAS  PubMed  Google Scholar 

32. Li, M. et al. Bub3 is a spindle assembly checkpoint protein regulating chromosome segregation during mouse oocyte meiosis. PLoS ONE 4, e7701 (2009).

    PubMed  PubMed Central  Google Scholar 

33. Yang, W. L., Li, J., An, P. & Lei, A. M. CDC20 downregulation impairs spindle morphology and causes reduced first polar body emission during bovine oocyte maturation. Theriogenology 81, 535–544 (2014).

    CAS  PubMed  Google Scholar 

34. Cao, Lan-Rui et al. Positive feedback stimulation of Ccnb1 and Mos mRNA translation by MAPK cascade during mouse oocyte maturation. Front. Cell Dev. Biol. 8, 609430 (2020).

    PubMed  PubMed Central  Google Scholar 

35. Verdaguer, I. B., Crispim, M., Hernández, A. & Katzin, A. M. The biomedical importance of the missing pathway for farnesol and geranylgeraniol salvage. Molecules 27, 8691 (2022).

    CAS  PubMed  PubMed Central  Google Scholar 

36. Heasman, S. J. & Ridley, A. J. Mammalian Rho GTPases: new insights into their functions from in vivo studies. Nat. Rev. Mol. Cell Biol. 9, 690–701 (2008).

    CAS  PubMed  Google Scholar 

37. Zhang, Y. et al. The small GTPase CDC42 regulates actin dynamics during porcine oocyte maturation. J. Reprod. Dev. 63, 505–510 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

38. Zima, M., O’Brien, J. M., Ouellette, L. A., Church, W. R. & Johnson, D. I. Mutational analysis of CDC42Sc, a Saccharomyces cerevisiae gene that encodes a putative GTP-binding protein involved in the control of cell polarity. Mol. Cell. Biol. 11, 3537–3544 (1991).

    Google Scholar 

39. Boutin, J. A. et al. Chromatographic assay and peptide substrate characterization of partially purified farnesyl- and geranylgeranyltransferases from rat brain cytosol. Arch. Biochem. Biophys. 354, 83–94 (1998).

    CAS  PubMed  Google Scholar 

40. Duan, X. & Sun, S. C. Actin cytoskeleton dynamics in mammalian oocyte meiosis. Biol. Reprod. 100, 15–24 (2019).

    PubMed  Google Scholar 

41. Cheng, T. et al. Comparative pharmacokinetics study of icariin and icariside II in rats. Molecules 20, 21274–21286 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

42. Clarke, H. J. Regulation of germ cell development by intercellular signaling in the mammalian ovarian follicle. Wiley Interdiscip. Rev. Dev. Biol. https://doi.org/10.1002/wdev.294 (2018).

43. Gu, L. et al. Metabolic control of oocyte development: linking maternal nutrition and reproductive outcomes. Cell. Mol. Life Sci. 72, 251–271 (2015).

    CAS  PubMed  Google Scholar 

44. Su, Y. Q., Sugiura, K. & Eppig, J. J. Mouse oocyte control of granulosa cell development and function: paracrine regulation of cumulus cell metabolism. Semin. Reprod. Med. 27, 32–42 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

45. Van Aelst, L. & D’Souza-Schorey, C. Rho GTPases and signaling networks. Genes Dev. 11, 2295–2322 (1997).

    PubMed  Google Scholar 

46. Zhang, X. et al. Polar body emission requires a RhoA contractile ring and Cdc42-mediated membrane protrusion. Dev. Cell 15, 386–400 (2008).

    PubMed  PubMed Central  Google Scholar 

47. Wang, Z. B. et al. Specific deletion of Cdc42 does not affect meiotic spindle organization/migration and homologous chromosome segregation but disrupts polarity establishment and cytokinesis in mouse oocytes. Mol. Biol. Cell 24, 3832–3841 (2013).

    PubMed  PubMed Central  Google Scholar 

48. Kincade, J. N., Hlavacek, A., Akera, T. & Balboula, A. Z. Initial spindle positioning at the oocyte center protects against incorrect kinetochore–microtubule attachment and aneuploidy in mice. Sci. Adv. 9, eadd7397 (2023).

    CAS  PubMed  PubMed Central  Google Scholar 

49. Halet, G. & Carroll, J. Rac activity is polarized and regulates meiotic spindle stability and anchoring in mammalian oocytes. Dev. Cell 12, 309–317 (2007).

    CAS  PubMed  Google Scholar 

50. Song, S. J. et al. Inhibition of Rac1 GTPase activity affects porcine oocyte maturation and early embryo development. Sci. Rep. 6, 34415 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

51. Zhang, J. L., Lv, M., Yang, C. F., Zhu, Y. X. & Li, C. J. Mevalonate pathway and male reproductive aging. Mol. Reprod. Dev. 90, 774–781 (2023).

    CAS  PubMed  Google Scholar 

52. Bae, J. H. et al. Farnesol prevents aging-related muscle weakness in mice through enhanced farnesylation of Parkin-interacting substrate. Sci. Transl. Med. 15, eabh3489 (2023).

    CAS  PubMed  Google Scholar 

53. Ye, C., Sutter, B. M., Wang, Y., Kuang, Z. & Tu, B. P. A metabolic function for phospholipid and histone methylation. Mol. Cell 66, 180–193.e8 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

Download references

Author notes

1. These authors contributed equally: Chuanming Liu, Huidan Zhang, Jialian Mao, Sainan Zhang.

Authors and Affiliations

1. Center for Reproductive Medicine and Obstetrics and Gynecology, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical School, Nanjing University, Nanjing, China

   Chuanming Liu, Huidan Zhang, Junshun Fang, Nannan Kang, Yang Zhang, Jidong Zhou, Xin Zhen, Guijun Yan, Yali Hu, Haixiang Sun & Lijun Ding

2. Center for Molecular Reproductive Medicine, Nanjing University, Nanjing, China

   Chuanming Liu, Huidan Zhang, Nannan Kang, Yang Zhang, Jidong Zhou, Xin Zhen, Guijun Yan, Yali Hu, Haixiang Sun & Lijun Ding

3. Jiangsu Human Reproductive Function Remodeling Engineering Research Center, Nanjing, China

   Chuanming Liu, Huidan Zhang, Haixiang Sun & Lijun Ding

4. Nanjing Drum Tower Hospital, Clinical College of Nanjing University of Chinese Medicine, Nanjing, China

   Jialian Mao & Lijun Ding

5. State Key Laboratory of Reproductive Medicine and Offspring Health, Nanjing Medical University, Nanjing, China

   Sainan Zhang, Chaojun Li & Haixiang Sun

6. State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Chemistry and Biomedicine Innovation Center (ChemBIC), Nanjing University, Nanjing, China

   Xiao Tian, Changjiang Wang, Huijie Pan & Ran Xie

7. Zhejiang Provincial Key Laboratory for Cancer Molecular Cell Biology, Life Sciences Institute, Zhejiang University, Hangzhou, China

   Yibing Zhu & Cunqi Ye

8. National Institute of Biological Sciences, Beijing (NIBS), Beijing, China

   Chun So

9. Tsinghua Institute of Multidisciplinary Biomedical Research, Tsinghua University, Beijing, China

   Chun So

10. Assisted Reproductive Technology Unit, Department of Obstetrics and Gynecology, Faculty of Medicine, The Chinese University of Hong Kong, Shatin, Hong Kong

    Chun So

11. State Key Laboratory of Analytic Chemistry for Life Science, Nanjing University, Nanjing, China

    Lijun Ding

12. Clinical Center for Stem Cell Research, Nanjing Drum Tower Hospital, Nanjing University, Nanjing, China

    Lijun Ding

13. MOE Key Laboratory of Model Animal for Disease Study, Nanjing University, Nanjing, China

    Lijun Ding

Contributions

L.D., H.S. and C.S. designed the research. C. Liu, H.Z., J.M., S.Z., L.D., X.T., Y. Zhu, C.W., H.P., N.K., J.F., Y.Z., J.Z. and X.Z. performed the experiments. C. Liu, H.Z., J.M., S.Z., C.S. and L.D. analyzed the data and wrote the paper. L.D., H.S., C.S., R.X., C.Y., Y.H., C. Li and G.Y. revised the paper.

Corresponding authors

Correspondence to Chun So, Haixiang Sun or Lijun Ding.

Competing interests

The authors declare no competing interests.

Peer review information

Nature Aging thanks the anonymous reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions