Modulating IL-11-dependent matrix stiffness to delay ovarian aging

Recent studies have highlighted the crucial role of mechanical properties in the ovarian microenvironment for ovarian function. However, the mechanisms that cause ovarian matrix stiffening during aging remain incompletely understood. Here we utilized atomic force microscopy (AFM) to demonstrate that human ovarian matrix stiffness increases with aging and in pathophysiological conditions, such as chemotherapy-induced premature ovarian insufficiency (POI), polycystic ovary syndrome (PCOS) and ovarian endometriosis. By integrating proteomic analysis of human ovarian tissue with transcriptomic profiling of human ovarian fibroblasts, we identified that IL-11, which is elevated in aging ovaries of mice, rats and humans, activates fibroblasts to secrete extracellular matrix (ECM), thereby increasing ovarian matrix stiffness. Genetic deletion of Il11ra1 in mice mitigated the increase in ovarian matrix stiffness and the decline in ovarian function associated with aging, chemotherapy-induced POI and PCOS. Single-nuclei RNA sequencing (snRNA-seq) revealed that blocking Il11ra1 reduces the proportion of activated fibroblasts. Furthermore, administration of siIl11 nanoparticles to aged mice and rats enhanced fertility and reduced ovarian matrix stiffness. Together, these findings highlight the pro-inflammatory factor IL-11 in regulating ovarian matrix stiffness. We propose that anti-IL-11 therapy represents a promising translational strategy for delaying ovarian aging. Similar content being viewed by others Main According to the World Health Organization, infertility affects approximately 15% of the global population, equating to around 60 million couples worldwide, with its prevalence on the rise1. Female factors are responsible for 50–70% of infertility cases, encompassing conditions such as age-related ovarian aging, polycystic ovary syndrome (PCOS), endometriosis and premature ovarian insufficiency (POI)2,3. These physiological and pathological factors contribute to reduced follicle quantity, declined follicle quality and/or ovulation disorders. In addition to increased risk of infertility, aneuploidy and congenital disabilities in offspring, diminished ovarian function also impacts overall health and lifespan4. Despite its importance, the fundamental biological mechanisms driving ovarian dysfunction and infertility remain poorly understood. Recent studies have highlighted the crucial role of mechanical properties in the ovarian microenvironment in the regulation of ovarian function, distinct from the well-established effects of pituitary hormones, steroids and growth factors5,6. Extracellular biomechanical signals are transduced into intracellular signals, thereby modulating key cellular processes, such as proliferation, differentiation, motility and apoptosis7. The deposition and cross-linking of the extracellular matrix (ECM) directly shape the mechanical properties of the ovary8. Ovarian follicles recognize matrix stiffness through various receptors, including integrins, which activate intricate signaling pathways such as Hippo and Akt. These pathways subsequently regulate follicle growth, differentiation, maturation and ovulation9,10,11,12. High matrix stiffness around primordial follicles helps maintain their long-term quiescent state13,14, while growing follicles encounter a softer ECM, facilitating their expansion and development15,16. Ovulation requires ECM degradation and a reduction in matrix stiffness to facilitate oocyte expulsion17. A recent study has revealed that follicles cultured in a stiff environment exhibited downregulated expression of genes associated with follicular growth and upregulated expression of genes related to inflammation and ECM reorganization, compared with those in a softer environment18. These findings highlight that the changes in ovarian matrix stiffness directly impact follicular dynamics. Studies have revealed increased collagen deposition in postmenopausal ovaries19 and animal models of reproductive aging20,21,22. Notably, instrumental indentation revealed a quantitative increase in ovarian-tissue stiffness between reproductively young (6–12 weeks) and old (14–17 months) mice20. Similarly, shear-wave elastography revealed that the ovarian stromal stiffness was markedly higher in a 41-year-old individual than in a 31-year-old control individual (30 kPa versus 16 kPa)23. Intriguingly, treatment with pirfenidone and BGP-15 [(O-[3-piperidino-2-hydroxy-1-propyl]-nicotinicamidoxime)] has been shown to reduce collagen deposition and enhance the ovulation of healthy oocytes in reproductively aged and obese mice24. Ovarian aging is accompanied by a cluster of interconnected hallmarks, including genomic instability, telomere attrition, epigenetic drift, autophagic decline, cellular senescence, nutrient-sensing dysregulation, mitochondrial dysfunction, oxidative stress and chronic low-grade inflammation25. These hallmarks will also promote ovarian fibrosis and cause an increase in ovarian matrix stiffness. Oxidative stress and chronic inflammation are key drivers of ovarian aging. They promote cellular senescence and apoptosis and induce the secretion of numerous pro-inflammatory and pro-fibrotic factors, such as interleukin-6 (IL-6) and tumor necrosis factor (TNF)26. These factors activate ovarian fibroblasts, leading to excessive deposition of ECM, thereby increasing ovarian matrix stiffness. In addition, the accumulation of senescent cells disrupts ovarian tissue architecture and impairs function, further elevating matrix stiffness27. Impaired autophagy compounds also contribute to the accumulation of damaged mitochondria and proteins, exacerbating oxidative stress and cellular senescence and thereby promoting the progression of ovarian fibrosis28. Hallmarks of ovarian aging drive fibrotic remodeling and matrix stiffening through intertwined pathways, which in turn accelerates ovarian aging, creating a vicious cycle. Excessive ECM deposition has also been observed in various reproductive disorders, including PCOS29,30,31, ovarian endometriosis32,33 and POI34. Using ultrasound-based shear-wave elastography, researchers have demonstrated that people with PCOS exhibit markedly increased ovarian tissue stiffness35,36. Notably, those with concurrent metabolic syndrome (MetS) display even greater matrix stiffness than do people with PCOS without MetS37. In PCOS, ECM deposition is considered a potential mechanism underlying oligo-ovulation and anovulation, and ovarian laparoscopic drilling has been shown to restore ovulation by correcting the biomechanical environment38,39. Clinically, reducing ovarian stromal stiffness through surgical isolation of ovarian fragments has been demonstrated to activate primordial follicles and restore fertility in people with POI40,41,42,43. Despite these findings, the changes and regulatory mechanisms of human ovarian matrix stiffness during aging and in various pathological conditions, the specific ovarian cell types involved in regulating matrix stiffness and how follicles integrate these mechanical signals remain largely unexplored. Exploring the role of mechanical signals in ovarian function could uncover innovative and promising strategies to enhance fertility. Here, we investigated human ovarian matrix stiffness across different ages and pathological conditions utilizing atomic force microscopy (AFM). Our results revealed a significant increase in ovarian matrix stiffness in individuals undergoing aging, as well as in those with chemotherapy-induced POI, PCOS, and endometriosis. We conducted proteomics atlas of human ovarian aging and snRNA-seq analysis of mouse ovarian aging, identifying the critical role of IL-11-ERK signaling in regulating ovarian matrix stiffness. Furthermore, we demonstrated that blocking IL-11 signaling in both physiological ovarian aging model and various pathological ovarian dysfunction models holds significant therapeutic potential for improving ovarian function.

Ovarian matrix stiffness increases with aging and pathological ovarian dysfunction, impairing follicle development We collected healthy human ovaries in the follicular phase from volunteers who underwent hysterectomy and oophorectomy surgery due to cervical or endometrial cancer, categorizing them into reproductively young (18–28 years, n = 30), middle-aged (35–42 years, n = 37) and older (47–52 years, n = 40) groups (Fig. 1a and Supplementary Table 1). The level of anti-Müllerian hormone (AMH), a well-established clinical biomarker for evaluating ovarian reserve function, was high in the young group (3.857 ± 1.647 ng ml−1), lower in the middle-aged group (1.809 ± 0.864 ng ml−1) and nearly undetectable in the older group (<0.06 ng ml−1) (Supplementary Table 1). Furthermore, we collected ovarian tissue samples from individuals in a similar age range (30–40 years) who exhibited ovarian dysfunction attributable to a diverse array of pathological etiologies, including chemotherapy-induced POI (n = 16), PCOS (n = 10) and ovarian endometriosis (n = 20) (Fig. 1a and Supplementary Table 1). Hematoxylin and eosin (H&E) staining revealed that, during the aging process and under the influence of various pathological factors, there was a significant increase in connective tissue in the ovarian cortex, accompanied by a proportional decline in parenchymal cell density (Fig. 1b and Extended Data Fig. 1a). To further investigate, we measured matrix stiffness (defined as Young’s modulus) of the ovarian cortex using AFM (Fig. 1c). The results demonstrated that ovarian matrix stiffness progressively increased with age (Fig. 1d). Quantitative analysis of ovarian stiffness revealed significant variations across different pathological conditions in the age-matched cohort (30–40 years). Ovaries from individuals with chemotherapy-induced POI, PCOS or endometriosis showed increased matrix stiffness (Fig. 1e). There was a significant accumulation of hydroxyproline, a key component of collagen, during both ovarian aging and pathological states (Figs. 1f,g). Masson and picrosirius-red staining further revealed increased collagen deposition throughout the ovary in aged individuals and those with reproductive-associated pathologies (Figs. 1h–l and Extended Data Figs. 1b,c). Under circularly polarized light, picrosirius-red staining revealed a marked enrichment of thick, yellow-red birefringent fibers (collagen I) in both physiologically aged ovaries and in ovaries from pathological conditions. By contrast, thin, green-birefringent fibers (collagen III) remained mostly unchanged in aged and pathological ovaries compared with controls (Extended Data Fig. 1d). In addition, utilizing 68Ga-FAPI-04 PET, a clinically available modality for fibrosis detection44, we observed higher ovarian-tracer uptake in aged individuals compared with young counterparts (Extended Data Fig. 1e). Nevertheless, these observations need to be verified in a larger cohort. Collectively, these data demonstrate that physiological ovarian aging and the pathological states of POI, PCOS and endometriosis are all associated with excessive ECM deposition and increased ovarian stromal stiffness. To investigate the impact of increased ovarian matrix stiffness on follicle development, we utilized alginate solutions at varying concentrations. The 0.5% alginate concentration was established as the standard for follicle cultivation45,46,47, whereas the 2% alginate exhibited enhanced stiffness. When cultured for 12 days, follicles in 0.5% alginate reached a markedly larger final diameter than those in 2% alginate, indicating a softer matrix supports more extensive growth (Extended Data Figs. 2a,b). Notably, the secretion of estrogen and progesterone by granulosa cells in the 2% alginate condition was markedly reduced (Extended Data Figs. 2c,d). Ki67 staining revealed a substantial decrease in granulosa cell proliferation under increased matrix stiffness (Extended Data Fig. 2e). Additionally, to examine the effect of matrix stiffness on granulosa cells, we cultured mouse ovarian primary granulosa cells on collagen-coated polyacrylamide gels with elastic moduli of 3 kPa (‘soft’) and 30 kPa (‘stiff’) (Extended Data Fig. 2f). On the stiff matrix, granulosa cells exhibited enhanced spreading and formed thicker stress fibers; on the soft matrix, they maintained a more rounded morphology, resembling their physiological shape (Extended Data Fig. 2g–i). Compared with the soft matrix, granulosa-cell proliferation declined, accompanied by parallel reductions in estrogen and progesterone secretion (Extended Data Fig. 2j–n). Collectively, these findings demonstrate that increased matrix stiffness inhibits follicular development and hormone secretion. IL-11 is a key target in regulating the ovarian matrix stiffness To investigate age-related proteomic changes in ovarian tissue, we performed proteomic profiling of ovarian samples from young, middle-aged and older cohorts. Comparative analysis of differentially expressed proteins (DEPs) demonstrated substantial molecular distinctions between perimenopausal ovaries and those maintaining reproductive function in those from the young or middle-aged cohorts (Extended Data Fig. 3a and Supplementary Table 2). Gene Ontology (GO) enrichment analysis identified extracellular matrix organization and extracellular space as the predominant biological processes associated with these DEPs (Extended Data Fig. 3b). Quantitative assessment revealed an age-dependent upregulation of key collagen genes, including COL1A1, COL1A2, COL3A1 and COL4A1 (Extended Data Fig. 3c,d). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis further demonstrated significant activation of the transforming growth factor-β (TGFβ) signaling pathway, a critical regulator of ECM biosynthesis, in postmenopausal ovarian tissue (Extended Data Fig. 3e). Immunohistochemical evaluation confirmed elevated TGFβ1 expression in both age-related ovarian changes and pathological conditions (Extended Data Fig. 3f,g). Multiplex immunohistochemical analysis revealed a significant positive correlation between coexpression patterns of COL1A1, TGFβ1 and the ovarian stromal cell marker DCN in ovarian tissue (Fig. 2a and Extended Data Fig. 3h). Notably, regions exhibiting high expression of COL1A1 and TGFβ1 demonstrated corresponding increases in tissue stiffness, as measured by AFM (Fig. 2b). These findings collectively suggest that TGFβ1-mediated collagen biosynthesis could represent a key molecular mechanism underlying the increases in ovarian matrix stiffness. The TGFβ1 signaling pathway has a pivotal role in ovarian function, mediating processes such as follicle development, ovulation and hormone synthesis48,49. TGFβ1 inhibition is associated with substantial adverse effects, including disruptions of ovarian function, oocyte maturation arrest and early stalling of embryonic development50. Targeting its downstream effectors for intervention might overcome these issues and offer a more viable therapeutic approach. Next, we analyzed expression of TGFβ1 and its receptor, TGFβR1, using our previously published human ovarian single-cell transcriptome sequencing data51. The results showed that TGFβ1 and TGFβR1 are expressed in most cell types in the human ovary, including granulosa cells, oocytes, thecal cells, stromal cells, smooth muscle cells, endothelial cells, monocytes, natural killer cells and T lymphocytes (Extended Data Fig. 4a). Furthermore, analysis of single-cell data from aged human ovaries (47–49 years) revealed that TGFβR1 is highly expressed in thecal cells, stromal cells and endothelial cells (Extended Data Fig. 4b). We isolated primary human ovarian fibroblasts (pHOFs) using the differential adhesion method (Extended Data Fig. 4c) and confirmed their identity by immunostaining for fibronectin and vimentin (Extended Data Figs. 4d,e). Then, we performed RNA-seq on paired unstimulated and TGFβ1-stimulated pHOFs (Fig. 2c and Extended Data Fig. 5a). GO and KEGG analyses of the differentially expressed genes (DEGs) identified ECM secretion as a key biological process induced by TGFβ1 stimulation in pHOFs (Extended Data Fig. 5b). This finding was corroborated by western blot results (Extended Data Fig. 5c), which demonstrated consistent upregulation of ECM expression following TGFβ1 treatment. Significantly, IL-11 expression exhibited the most pronounced increase following TGFβ1 treatment (Fig. 2d), as corroborated by western blot and ELISA assays (Extended Data Figs. 5d,e). Beyond TGFβ1, treatment of pHOFs with CTGF, bFGF, PDGF or CCN2 also significantly increased IL-11 protein levels (Extended Data Fig. 5d). Remarkably, IL-11 expression in the ovary increased with age not only in mice, but also in rats and, importantly, in humans (Fig. 2e and Extended Data Fig. 5f–h). To our knowledge, this cross-species conservation of age-related IL-11 upregulation specifically in the ovary has not been previously reported, highlighting the novelty of this finding. Furthermore, IL-11 expression was significantly elevated in various ovarian pathological conditions (Fig. 2f and Extended Data Fig. 5i). IL-11 expression is positively correlated with age and negatively correlated with AMH level (Extended Data Fig. 5j). The receptor for IL-11, IL-11RA, was highly expressed in human ovary fibroblasts (Extended Data Fig. 5k). To investigate the functional role of IL-11, we incubated pHOFs with recombinant human IL-11 (rhIL-11) at a concentration of 10 ng ml−1 for 24 h. The results revealed that IL-11 induced ACTA2 expression and collagen secretion, as evidenced by high-content imaging (Fig. 2g and Extended Data Fig. 5l). Notably, the administration of a neutralizing anti-IL-11 antibody abolished TGFβ1-induced ACTA2 expression and collagen secretion. Considering that the increase in matrix stiffness is also involved in the invasion and migration of fibroblasts, we explored the role of IL-11 on the migration of pHOFs. Wound-healing percentage was remarkably increased in the TGFβ1- and IL-11-treated group, whereas it was notably suppressed in the presence of neutralizing anti-IL-11 antibodies (Extended Data Figs. 5m,n). Similarly, migration was significantly induced in the group treated with TGFβ1 and IL-11, whereas it was substantially inhibited by the presence of neutralizing anti-IL-11 antibody (Extended Data Fig. 5o). These findings collectively demonstrate that IL-11 plays a crucial role in activating human ovarian fibroblasts and promoting collagen secretion, which could contribute to increased ovarian matrix stiffness. IL-11 regulates ovarian matrix stiffness through the ERK1–ERK2 signaling pathway To explore the mechanism by which IL-11 regulates ovarian matrix stiffness, we performed RNA-seq on paired unstimulated and IL-11-stimulated pHOFs (Extended Data Fig. 6a). The GO and KEGG analysis indicated a significant enrichment of genes associated with the ECM and its receptor (Fig. 2h). Following IL-11 treatment, several collagen genes, including COL1A1, COL1A2, COL3A1, COL4A1, COL4A2, COL5A1, COL5A2, COL6A1, COL6A2, COL6A3 and COL7A1, exhibited increased expression (Extended Data Fig. 6b). These results indicate that IL-11 can promote the secretion of collagen in pHOFs. These findings were corroborated by gene set enrichment analysis (GSEA) analysis (Extended Data Fig. 6c). To further explore the mechanism, we analyzed the activated signaling pathways in response to IL-11 stimulation and found that NOTCH, PI3K–AKT, cGMP–PKG, MAPK and mTOR pathways were significantly activated in pHOFs (Extended Data Fig. 6d). To further dissect the role of these pathways, we introduced IL-11, along with inhibitors targeting the aforementioned pathways. We specifically divided the MAPK pathway into p38 MAPK, ERK1 and ERK2 (ERK1/2) and JNK sub-pathways. Reverse transcription PCR (RT-PCR) results showed that inhibition of ERK1/2 had the most significant impact on reducing COL1A1 upregulation induced by IL-11 (Fig. 2i). Additionally, phosphoproteomic analysis using dot blotting revealed that ERK1/2 phosphokine levels were upregulated upon IL-11 treatment (Figs. 2j,k and Extended Data Fig. 6e). After 24 h of IL-11 stimulation, significant activation of ERK and its downstream effectors (such as p90RSK, p70S6K and GSK-3β) was observed in pHOFs. These effects were abrogated by the ERK inhibitor SCH772984 (Extended Data Fig. 6f). Western blot and immunofluorescence assays confirmed that SCH772984 significantly mitigated the increase in COL1A1 and ACTA2 expression induced by IL-11 (Figs. 2l,m). Furthermore, phosphorylated ERK (p-ERK) and ERK levels exhibited age-associated elevation in both human and mice ovaries (Extended Data Figs. 6g,h). The above results suggest that IL-11 promotes collagen secretion in human ovarian fibroblasts by inducing the phosphorylation of ERK1/2. IL-11 induces increased ovarian matrix stiffness in mice Next, we assessed the effect of IL-11 on ovarian function and matrix stiffness in mice by administering recombinant mice IL-11 (rmIL-11) (Fig. 3a). After rmIL-11 administration, no statistically significant alterations were observed in body weight (Fig. 3b). Except for a decrease in the heart index, the other organ indices showed no significant changes (Fig. 3c and Extended Data Fig. 7a). Consistent with this observation, an analysis of serum levels of biomarkers for cardiac (CK), hepatic (ALT) and renal (LDH-L, CRE) function revealed no significant differences between the two groups (Extended Data Fig. 7b). Analysis of Young’s modulus indicated that ovarian matrix stiffness increased after rmIL-11 treatment (Fig. 3d). Hydroxyproline assays, Picrosirius red staining and immunohistochemical analysis collectively validate the augmentation of collagen deposition in ovaries in the rmIL-11 group (Fig. 3e–g). With rmIL-11 treatment, there was progressive activation of ACTA2 and p-ERK1/2 (Fig. 3h–j), suggesting that IL-11 activates the secretion of collagen by ovarian fibroblasts through the ERK1/2 signaling pathway. Additionally, we assessed ovarian function in mice and found that the proportion of estrous-cycle disorders was significantly increased in the rmIL-11 group (Fig. 3k). Statistical analysis of follicle count indicated that the number of primary follicles, antral follicles and total healthy follicles was increased (Fig. 3l). Hormone level measurements revealed a decline in AMH and estrogen (E2) levels, coupled with an elevation in follicle-stimulating hormone (FSH) levels in the IL-11-treated group (Fig. 3m). Furthermore, IL-11 treatment significantly reduced fertility in mice, as reflected by a decrease in the number of pups born per pregnant mouse in a single delivery (Fig. 3n). As such, IL-11 activates the ERK1/2 pathway to promote collagen secretion, thereby increasing matrix stiffness and causing a decline in ovarian function. Deletion of Il11ra1 reduces ovarian matrix stiffness and delays ovarian aging To further explore the relationship between IL-11 upregulation and ovarian matrix stiffness in old mice, we studied 48-week-old Il11ra1−/− mice and wild-type littermate controls (Fig. 4a). Compared with wild-type littermate controls, ovarian matrix stiffness was significantly decreased in Il11ra1−/− mice (Fig. 4b). Comprehensive analyses, including hydroxyproline assays, picrosirius red staining, immunohistochemistry and immunoblotting, consistently demonstrated decreased collagen levels in the ovaries of aged Il11ra1−/− mice (Figs. 4c–e,g,h). Furthermore, we observed diminished ACTA2 expression and reduced ERK1/2 activation in Il11ra1−/− mice (Fig. 4f–h). These findings collectively indicate that Il11ra1−/− mice exhibit reduced ovarian collagen deposition and significant improvement in matrix stiffness during aging. We subsequently evaluated ovarian function in both Il11ra1−/− mice and wild-type controls. Aged Il11ra1−/− mice demonstrated enhanced ovarian reserve, as evidenced by an increased number of secondary follicles and reduced atretic follicles (Figs. 4i,j). Serum analysis revealed higher levels of AMH and E2 in aged Il11ra1−/− mice than in wild-type littermates (Fig. 4k). Moreover, aged Il11ra1−/− mice showed increased responsiveness to gonadotropins, with a greater number of ovulated oocytes (Figs. 4l,m). Importantly, these ovulated oocytes maintained normal viability and fertilization capacity, as demonstrated by comparable rates of in vitro fertilization (IVF) and subsequent embryo development between the two groups (Fig. 4n). These results provide evidence that age-related ovarian matrix stiffness can be modulated to potentially extend reproductive lifespan. Il11ra1 deficiency decreases ovarian-activated fibroblasts in old mice Ovarian tissues from 48-week-old Il11ra1−/− mice and their wild-type littermate controls were subjected to snRNA-seq analysis (n = 3 per group). Following stringent quality control, a total of 56,363 high-quality cells were retained for downstream analysis (Fig. 5a). On the basis of their transcriptomic profiles, these cells were classified into nine major cell types: thecal cells, granulosa cells, stromal cells, luteal cells, vascular endothelial cells (ECs), epithelial cells, lymphatic ECs, T cells and macrophages (Fig. 5a, Extended Data Figs. 8a,b and Supplementary Table 3). Notably, Il11ra1−/− mice exhibited a reduced proportion of luteal cells and an increased proportion of granulosa cells, indicating a higher number of growing follicles in the ovaries of Il11ra1-deficient mice (Fig. 5b). In addition, the proportions of ovarian macrophages and T cells were significantly decreased in Il11ra1-deficient mice (Fig. 5b). To identify the primary cell type responsible for ovarian collagen synthesis, we conducted further analyses. Our data demonstrate that the majority of collagen genes, including Col1a1, Col1a2, Col3a1, Col4a1, Col4a2, Col4a3, Col4a4, Col5a1, Col5a2, Col6a3 and Col11a1, are predominantly expressed in ovarian stromal cells (Fig. 5c and Extended Data Fig. 8c). Additionally, we observed a decrease in the proportion of stromal cells in the ovaries of Il11ra1-deficient mice compared with that in wild-type controls (Fig. 5b). KEGG analysis revealed that the downregulated genes in stromal cells of the Il11ra1-deficient group were significantly enriched in pathways associated with matrix stiffness, such as ECM–receptor interaction, focal adhesion and cell adhesion molecules (Fig. 5d). These analyses suggest that blockade of the IL-11 signaling pathway results in reduced collagen production and decreased matrix stiffness in the ovary. Given that the ECM components are predominantly synthesized and maintained by stromal cells, we proceeded to investigate the subtypes of stromal cells. Through reclustering analysis, we identified seven distinct stromal cell clusters (Fig. 5e). To characterize these clusters, we employed DEGs analysis, GSEA, and module scoring, which enabled us to identify specific markers for each cluster (Supplementary Table 3). Subsequently, we evaluated the specific DEGs associated with each stromal cell subtype to validate cluster identification (Extended Data Fig. 8d). On the basis of the DEG profiles and pathway analysis, we classified the clusters as follows: (1) Mki67_stromal cells, characterized by elevated expression of ki67; (2) pericytes, marked by high expression of Rgs5, Notch3 and Abcc9; (3) Col26a1_matrix fibroblasts, distinguished by high expression of Col26a1 and Col25a1; (4) activated fibroblasts, exhibiting elevated levels of Piezo2, Kif26b and Col12a1; (5) Slc24a2_stromal cells, identified by Slc24a2 expression; (6) Col6a6_matrix fibroblasts, characterized by high expression of Col6a6; and (7) myofibroblasts, defined by high expression of Acta2 and Tagln (Fig. 5e and Extended Data Fig. 8d). Comparative analysis demonstrated a marked decrease in activated fibroblast populations alongside a concomitant increase in Col26a1_matrix fibroblast proportions in Il11ra1-deficient mice relative to wild-type controls (Fig. 5f). Activated fibroblasts, identified by their robust ECM biosynthesis capacity, displayed significant enrichment of matrix-associated pathways in their transcriptional profile, particularly in ECM-receptor interactions, focal adhesion complexes and adherens junction formation (Fig. 5g). By contrast, Col26a1_matrix fibroblasts exhibited a distinct molecular signature characterized by elevated expression of genes regulating endoplasmic reticulum-protein processing, DNA-repair mechanisms (particularly nucleotide excision repair), nucleocytoplasmic transport systems and ATP-dependent chromatin remodeling processes (Fig. 5h). This unique gene-expression pattern suggests a potential role for Col26a1_matrix fibroblasts in maintaining tissue homeostasis through coordinated damage response and repair mechanisms in the ovarian stroma. These results indicate that Il11ra1 deficiency leads to a reduction in activated fibroblasts in the ovary. Building on our previous findings that IL-11 activates ovarian fibroblasts through the ERK1/2 pathway (Fig. 2), we further investigated the expression of ERK1/2 across different stromal cell subtypes. The results demonstrated that both the ERK1 and ERK2 cascade (GO:0070371) and the regulation of ERK1 and ERK2 cascade (GO:0070372) were most prominently enriched in activated fibroblasts (Figs. 5i,j), suggesting that these cells could be regulated by the IL-11–ERK1/2 signaling pathway. Deletion of Il11ra1 reduces ovarian matrix stiffness in a chemotherapy-induced premature ovarian insufficiency mouse model Chemotherapy-induced POI can lead to increased ovarian matrix stiffness (Fig. 1) and impaired ovarian function. To establish a chemotherapy-induced POI mouse model, we utilized four clinically relevant chemotherapeutic agents: cyclophosphamide (CTX), cisplatin (CIS), doxorubicin (DOX) and paclitaxel (PTX) (Extended Data Fig. 9a). We then conducted a comprehensive evaluation of alterations in ovarian matrix stiffness. Quantitative analysis revealed that DOX treatment induced the most significant increase in ovarian matrix stiffness (Extended Data Fig. 9b), accompanied by substantial collagen deposition (Extended Data Fig. 9c). On the basis of these findings, we selected DOX for further experiments and administered it to 8-week-old wild-type and Il11ra1-deficient mice (KO mice) (Fig. 6a). The ovarian matrix stiffness, reflected by the Young’s modulus, was significantly elevated in both treatment groups compared with that of the control group (Fig. 6b). Notably, this increase was attenuated in the KO + DOX group, which showed lower stiffness levels relative to the WT + DOX group. Consistent with this finding, picrosirius red staining for collagen deposition revealed that the DOX-induced expansion of fibrotic areas was partially reversed in KO mice. Although the WT + DOX group exhibited a significant increase in picrosirius-red-positive area, the KO + DOX group showed reduced collagen deposition compared with the WT + DOX group (Figs. 6c,d). Biochemical analysis of hydroxyproline content further supported these observations. Although both DOX-treated groups displayed elevated hydroxyproline levels relative to controls, the KO + DOX group demonstrated a marked reduction in total collagen content compared with the WT + DOX group (Fig. 6e). Immunofluorescence analysis of COL1A1 and ACTA2 density revealed a similar pattern. The pronounced upregulation of COL1A1 and ACTA2 observed in the WT + DOX group was significantly blunted in the KO + DOX group (Figs. 6f,g). In WT mice, DOX treatment resulted in marked ERK activation, as evidenced by increased phosphorylation levels. By contrast, this activation was largely abrogated in the KO + DOX group (Figs. 6h,i). These results demonstrate that although DOX treatment induces significant ovarian fibrosis, it is markedly improved in the KO + DOX group. To assess ovarian function following DOX treatment, we measured follicular development, serum hormone levels and oocyte-related parameters. Follicle counts at various developmental stages (Figs. 6j,k) revealed that DOX treatment significantly reduced the number of primordial, secondary and total healthy follicles in WT mice. This depletion was markedly attenuated in the KO + DOX group, indicating improved follicular preservation. Serum AMH and E2 levels were substantially reduced in the WT + DOX group compared with those in controls, whereas FSH levels were elevated, indicating impaired ovarian function (Fig. 6l). Notably, these hormonal disturbances were partially ameliorated in the KO + DOX group, which exhibited higher AMH and E2 levels along with lower FSH levels relative to those in WT + DOX mice. DOX treatment significantly reduced the number of ovulated oocytes (Figs. 6m,n) in WT mice. These detrimental effects were substantially alleviated in the KO + DOX group. Notably, there was no difference in the fertilization capacity of oocytes between the three groups (Fig. 6o). These findings suggest that Il11ra1 deletion improves ovarian matrix stiffness and ovarian function in chemotherapy-induced POI. Deletion of Il11ra1 reduces ovarian matrix stiffness in a polycystic ovary syndrome mice model To investigate the role of IL-11 in ovarian matrix stiffness in a PCOS model, we established a dehydroepiandrosterone (DHEA)-induced PCOS mouse model using wild-type (WT) and Il11ra1-knockout (KO) mice (Fig. 7a). Ovarian matrix stiffness was markedly increased in both DHEA-treated groups compared with that in controls (Fig. 7b). Notably, this increase was significantly blunted in the KO + DHEA group. The collagen content in the ovaries of PCOS-like Il11ra1-deficient mice was lower than that in the PCOS group (Fig. 7c–f). IHC images and scoring revealed a marked decrease in ACTA2 expression in the PCOS-like Il11ra1−/− mice (Fig. 7g), which was further corroborated by western blot analysis and quantitative statistics (Fig. 7h,i). Furthermore, Il11ra1 depletion significantly reduced the relative expression of p-ERK (Figs. 7h,i). These results demonstrate that DHEA-induced PCOS-like conditions promote ovarian fibrosis, whereas Il11ra1 knockout substantially ameliorates these fibrotic changes, suggesting a key role for Il11ra1 in the pathogenesis of ovarian fibrosis in PCOS. Next, we examined the ovarian function of the PCOS model. PCOS-like Il11ra1−/− mice exhibited an increased number of antral follicles compared with PCOS controls (Figs. 7j,k). Notably, the deletion of Il11ra1 normalized the elevated serum testosterone levels in DHEA-treated mice, a key indicator of PCOS (Fig. 7l). The disrupted estrous cycle induced by DHEA was ameliorated in PCOS-like Il11ra1−/− mice (Fig. 7m). Elevated ovarian stromal hardness, leading to oligo-ovulation, is a characteristic feature of PCOS (Figs. 7n,o). The ovulation rate in DHEA-treated Il11ra1−/− mice was significantly improved, whereas the fertilization rate of oocytes remained comparable between the two groups (Fig. 7p). These findings underscore the pivotal role of IL-11 in increasing ovarian matrix stiffness in PCOS, a model of ovarian injury distinct from aging and chemotherapy-induced damage. Anti-IL-11 reverses matrix stiffness and restores ovarian function in old mice and rats We further investigated the effects of anti-IL-11 therapy on ovarian aging in aged mice and rats. Three short interfering RNA (siRNA) sequences were designed to suppress IL-11 expression in primary ovarian fibroblasts derived from mice. Among these, siIl11_1 demonstrated the highest interference efficiency (Extended Data Fig. 10a,b) and was subsequently selected as the therapeutic siRNA. Similarly, the effect of siIl11 was validated in primary rat ovarian fibroblasts (Extended Data Fig. 10c,d). The siRNA drug, formulated using liposome-based delivery systems, is approved by the US Food and Drug Administration (FDA), ensuring its safety and efficacy. We developed siIl11-loaded liposome nanoparticles (siIl11 NPs) (Extended Data Fig. 10e). These nanoparticles exhibited excellent dispersion in water, with a hydrodynamic diameter of approximately 116 nm and a polydispersity index (PDI) of 0.203 (Extended Data Fig. 10f). Owing to the electronegative nature of siRNA, the zeta potential of the siRNA-loaded liposomes was reduced to 0.426 mV at 25 °C (Extended Data Fig. 10f). The liposomes demonstrated exceptional siRNA encapsulation efficiency, exceeding 96% (Extended Data Fig. 10f). Transmission electron microscopy (TEM) imaging revealed that the nanoparticles possessed a spherical morphology (Extended Data Fig. 10g,h) and maintained uniform distribution and stability over 24 h (Extended Data Fig. 10i). Following tail vein intravenous injection of siIl11 NPs into aged mice, ovaries were collected for four weeks following treatment (Fig. 8a). Compared with the control group, ovarian matrix stiffness in siIl11-NP-treated mice decreased by 35.7% (Fig. 8b). Collagen content was also significantly reduced after siIl11 NP treatment (Fig. 8c–e). The expression of fibroblast activation markers, pERK/ERK and ACTA2, was notably downregulated in siIl11-NP-treated mice (Figs. 8f,g). Furthermore, ovarian function was restored in aged mice treated with siIl11 NPs, as evidenced by an increased number of secondary and antral follicles, as well as an improved ovulation rate (Fig. 8h–j). Notably, fertility in aged mice was significantly enhanced following siIl11 NP treatment, as reflected by an increase in the number of pups per pregnant mouse per delivery (Fig. 8k,l). These results highlight the potential of IL-11-targeted therapy in mitigating ovarian aging and restoring reproductive function. To exclude the potential systemic effects of siIl11 NPs injected through the tail vein on ovarian stromal stiffness, we performed bursal micro-injection of AAV-sh-Il11ra1 to locally knock down the IL-11 receptor in the ovary (Extended Data Fig. 10j). After 4 weeks of treatment, the AAV-sh-Il11ra1-treated mice exhibited significantly lower ovarian matrix stiffness and collagen deposition compared with controls (Extended Data Fig. 10k–m). Morphological analysis and follicle quantification revealed increased numbers of primary follicles, secondary follicles, antral follicles and total healthy follicles in ovaries of the treated mice compared with control ovaries (Extended Data Fig. 10n,o). Fertility assessment showed that the pregnant mice treated with AAV-sh-Il11ra1 gave birth to an increased number of pups in a single delivery (Extended Data Fig. 10p). Our results showed that a single local administration of AAV-sh-Il11ra1 also reduced ovarian stromal stiffness and improved ovarian function, independent of systemic effects. Next, we evaluated the anti-ovarian aging effects of siIl11 NPs in a rat model (Fig. 8m). Intravenous administration of siIl11 NPs effectively reduced ovarian matrix stiffness in aged rats (Fig. 8n) and significantly alleviated collagen deposition (Fig. 8o–q). IHC and western blot analyses demonstrated a marked reduction in the expression of pERK/ERK and ACTA2 in siIl11-NP-treated rats (Figs. 8r,s). Histological examination of the ovaries revealed that siIll NPs promoted follicle development, as evidenced by an increased number of secondary follicles, antral follicles, total healthy follicles and corpus luteum (Figs. 8t,u). Fertility assessments further indicated that siIl11 NPs significantly improved the pregnancy rate (Fig. 8v) and increased the litter size in aged rats (Fig. 8w). Together, these findings demonstrate that treatment with siIl11 NPs effectively ameliorates ovarian matrix stiffness and enhances ovarian function in rodent models of aging.

Abnormal tissue stiffness is primarily attributed to the rigidity of the ECM resulting from accumulation, contraction and cross-linking. This biomechanical alteration engenders the transduction of mechanical stimuli into intracellular biochemical signaling pathways, thereby modulating cellular biological processes and the function of organs. The ovaries, crucial for both reproduction and endocrine functions, are the subject of limited research regarding their stiffness under various pathological states. Consequently, the regulatory mechanisms governing the stiffness of the ovarian matrix remain elusive. In this study, we demonstrate that, in human physiological aging and various pathological changes (chemotherapy-induced POI, PCOS and endometriosis) of the ovary, there is an increase in ovarian matrix stiffness, which leads to the obstruction of follicular development, decreased release of oocytes and disordered hormone synthesis. The increase in ovarian matrix stiffness is critically dependent on IL-11. Notably, blocking the IL-11 pathway can reduce the ovarian matrix stiffness and improve ovarian function in aging individuals, as well as in the presence of various pathological mediators, indicating that this common cellular program can be triggered in response to ovarian injury from diverse causes. Ovaries contain a broad array of ECM proteins, mainly including collagens and elastins, that are crucial for maintaining the proper physiologic environment for ovarian function52,53. This encompasses a range of conditions, such as Ehlers–Danlos syndrome, characterized by diverse collagen gene mutations54; Marfan syndrome, resulting from fibrillin-1 gene mutations55; and various muscular dystrophies involving mutations in genes such as dystrophin, laminin and different types of collagens56. An extensive survey involving more than 1,200 individuals with Ehlers–Danlos syndrome revealed notable reproductive health issues: 41.1% of the participants cited infertility, 67.2% experienced irregular menstrual cycles and a concerning 57.2% had a history of spontaneous abortion57. Marfan syndrome can lead to underdeveloped ovaries or adverse pregnancy outcomes58,59. People with myotonic dystrophy type I have a reduced ovarian-reserve function and respond poorly to controlled ovarian stimulation60. Studies on these genetic disorders indicate a correlation between the accumulation of ECM components and the decline in ovarian function, potentially linked to augmented ovarian stromal rigidity; however, further investigation is warranted to elucidate these relationships. Studying the ovarian mechanical microenvironment not only enhances our understanding of ovarian physiology and pathology, but also provides a scientific basis for the development of new diagnostic and therapeutic methods for ovarian dysfunction. Through scRNA-seq, we discovered that the knockout of Il11ra1 suppresses the activation of fibroblasts in the ovaries, as evidenced by a reduction in the proportion of activated fibroblasts and myofibroblasts, which are two forms of fibroblast activation. Activated fibroblasts are the primary source of the ovarian ECM, and an increase in their proportion leads to ECM deposition, which results in increased matrix stiffness. Notably, Il11ra1-deficient mice also exhibited a selective reduction in ovarian macrophage and T-cell density, suggesting that inhibition of IL-11 signaling reshapes the ovarian immune landscape. IL-11 is increasingly recognized as a pro-inflammatory cytokine across multiple organ systems61. In chronic kidney disease, anti-IL-11 therapy rekindles proliferation of tubular epithelial cells, reverses fibroinflammation and restores both renal mass and function62. In the context of neuroinflammation, targeting IL-11 and IL-11R signaling curbs NLRP3 inflammasome activation in monocytes and dampens inflammatory-cell migration, positioning this pathway as a promising therapeutic target in relapsing–remitting multiple sclerosis63. These findings raise the possibility that, in the ovary, IL-11 fosters a local inflammatory milieu that indirectly amplifies ECM deposition and increases stromal stiffness. Therefore, the immunomodulatory function of IL-11 in the ovarian microenvironment warrants dedicated investigation. With age, IL-11 expression is progressively elevated across various tissues, and IL-11 inhibition has been shown to extend both the healthspan and lifespan of mammals64. In the ovaries, IL-11 is produced by both thecal and granulosa cells, with higher concentrations detected in human follicular fluid from preovulatory follicles compared with other types of follicles65. Clinical studies have shown that IL-11 levels in follicular fluid are elevated in people with PCOS66 and ovarian hyperstimulation syndrome (OHSS)67. Furthermore, in vitro studies have revealed that IL-11 treatment stimulates steroidogenesis in rat preovulatory follicles68 and bovine granulosa cells69. However, IL-11 has not been extensively studied and was not previously thought to be important for matrix stiffness and ovarian aging. We found that aging, chemotherapy, PCOS and endometriosis all lead to the secretion of IL-11 in the ovaries, causing an increase in matrix stiffness. The deletion of Il11ra1 mitigated the increase in ovarian matrix stiffness in mouse models of aging, doxorubicin-based chemotherapy and DHEA-induced PCOS, consequently improving ovarian function. Previous studies have revealed the prominent role of IL-11 in some other organ fibrosis62,70,71. These findings, combined with our result, suggest that IL-11 is a promising target, and effective inhibition of IL-11 signaling pathway could open up a new therapeutic strategy to delay ovarian aging. The finding that siIl11 restores fertility in reproductively aged animals appears, at first glance, to contradict reports that germline deletion of Il11 or Il11ra1 causes complete infertility72,73. However, we propose that this discrepancy is resolved by considering the distinction between complete developmental ablation and therapeutic partial inhibition in the adult. First, siIl11 administration in aged mice did not alter oocyte DNA integrity or induce heritable genetic changes in offspring. Second, the partial reduction is sufficient to alleviate ovarian fibrosis and preserve reproductive function, yet insufficient to trigger the developmental defects associated with total IL-11 ablation. Third, we found that heterozygous Il11ra1 knockout mice exhibit normal fertility, indicating that partial loss of Il11ra1 does not impair reproductive function. These findings underscore a critical biological principle: IL-11 has dual roles that depend on context. It is essential for development but becomes pathogenic when aberrantly upregulated during aging. In addition, our data demonstrate that deletion of Il11ra1 preserves serum estrogen levels in aged mice and in a chemotherapy-induced POI model, suggesting that anti-IL-11 treatment could delay the endocrine transition associated with ovarian aging. Declining estrogen during menopause contributes to age-related pathologies, including obesity, Alzheimer’s disease and osteoporosis. Whether this translates to delayed menopause onset or attenuation of these aging-associated diseases warrants investigation in future studies. Collectively, our findings suggest that targeted IL-11 attenuation could offer a safe strategy to delay ovarian aging, circumventing the infertility risks historically linked to its complete ablation. This study also has some limitations. Here, we developed the liposome-encapsulated siIl11 NPs to treat ovarian aging. Strikingly, our findings reveal that treatment with siIl11 NPs reduces ovarian matrix stiffness and improves ovarian function, which was observed in both mice and rats. However, the systemic administration of siIl11 NPs through tail-vein injection lacks specificity for the ovaries and can be distributed to multiple organs. Further exploration is needed on how to deliver siRNA to the ovaries, such as by combining it with targeting peptides or using a cell membrane biomimetic nanometer, to achieve efficient targeted therapy in the ovaries. In the present study, we used nine-month-old mice, a model corresponding to the period of rapid ovarian decline in middle-aged humans, to investigate whether IL-11 inhibition can prevent the progression of ovarian fibrosis. Although this intervention represents a preventive approach, future studies using older mice with established fibrosis will be necessary to determine whether targeting IL-11 can reverse existing fibrotic damage. Notably, long-term ovarian function after treatment in mice needs to be assessed, including sex-hormone measurement by more precise methods, such as LC–MS/MS, as well as evaluation of reproductive endpoints.

Human samples and ethical statement Normal ovaries were collected from participants who underwent oophorectomy for cervical cancer or endometrial carcinoma. At the time of surgery, all individuals were in the proliferative phase. None had received hormone therapy, nor had they previously undergone radiotherapy or chemotherapy. Pathologists meticulously examined the ovarian tissues to rule out any tumor metastasis. Ovarian tissue samples were obtained from donors across three distinct age cohorts: a reproductively young group (18–28 years, n = 30), a middle-aged group (35–42 years, n = 37) and an older group (47–52 years, n = 40). The ovaries for chemotherapy-induced POI (30–40 years, n = 16) with FSH > 25 IU L−1 were derived from people with cervical or endometrial cancer who had undergone preoperative chemotherapy and required oophorectomy. The ovaries in the PCOS group (30–40 years, n = 10) were derived from people with cervical cancer or endometrial cancer, and ultrasound indicates a polycystic appearance of the ovaries. The diagnosis of PCOS adheres to the most current diagnostic criteria, which include the presence of irregular menstrual cycles and the identification of at least 12 small antral follicles in one ovary. Ovarian tissue in people with endometriosis (30–40 years, n = 20) is derived from marginal ovarian tissue carried by ovarian-cyst dissection. The amount of ovarian tissue taken is guaranteed not to affect the postoperative pathological diagnosis. The ethical approval for the sample collection was granted by the local ethics committee (TJ-IRB20210319), and all participants provided informed consent. Animals The Animal Ethical Committee of Tongji Medical College, Huazhong University of Science and Technology, approved all the experiments done involving animals (TJH-202106016). Protocols were performed according to the guide for the care and use of laboratory animals of the National Institute of Health. All mice and rats were maintained in a controlled environment at 21–24 °C with humidity levels of 40–70% under a 12-h light/dark cycle. They had unlimited access to food and water in the specific pathogen-free (SPF) environment. The C57BL/6J Il11ra1−/− mice were purchased from Shanghai Model Organisms (NM-KO-190453). The chemotherapy-induced POI model was established by i.p. injection of two doses of cyclophosphamide (100 mg kg−1), cisplatin (5 mg kg−1), doxorubicin (10 mg kg−1) or paclitaxel (7.5 mg kg−1) in 8-week-old female Il11ra1+/+ (wild-type) and Il11ra1−/− mice once weekly. To generate the PCOS-like model, dehydroepiandrosterone (DHEA, MedChemExpress, HY-14650) was first dissolved in DMSO as a 50 mg ml−1 stock, then diluted to 6 mg ml−1 corn oil. Four-week-old female Il11ra1+/+ (wild-type) and Il11ra1−/− mice received daily subcutaneous injections of DHEA (60 mg kg−1) for 28 consecutive days, following the regimen described previously74. For the model of in vivo IL-11 administration, rmIL-11 was reconstituted to a final concentration of 50 μg ml−1 in saline. Eight-week-old female C57BL/6 mice were subjected to daily subcutaneous injection with either 100 μg kg−1 of rmIL-11 or an identical volume of saline for 28 days. For in vivo therapeutic studies, female C57BL/6 mice at 36 weeks of age and Sprague–Dawley rats at 40 weeks of age were each divided into two groups and treated with siCon NPs or siIl11 NPs, respectively. A total of 100 µl siRNA NPs were administered through tail-vein injection twice weekly over a continuous four-week treatment period. The mice were euthanized seven days after the treatment for analysis of ovarian function and matrix stiffness. The sample preparation and force indentation acquisition by AFM The matrix stiffness of the ovarian cortex was measured as previously reported75,76. In brief, fresh ovarian tissues were rapidly embedded and frozen in OCT cryostat embedding medium using liquid nitrogen and stored at −80°C. We collected 30-µm-thick sections from a depth of 50–80 µm beneath the ovary surface using a cryostat at −20°C. For AFM measurements, the section was transferred to a room-temperature (RT), AFM-compatible glass-bottom dish using a brush, and 1 ml of EDTA-free D-PBS containing proteinase inhibitors was added immediately. For immunostaining, instead of filling dishes with D-PBS, tissue section was blocked with 5% serum in D-PBS with proteinase inhibitors without EDTA for 2 h at RT. To measure areas with high or low collagen content, ovarian sections were incubated with a primary anti-Collagen I antibody (GB11022, Servicebio, 1:100 dilution) in a solution containing 5% serum, 0.2% Triton X-100 in PBS and proteinase inhibitors for 1 h at RT. Following washing, the sections were incubated with Alexa-Fluor-488-conjugated anti-rabbit IgG (GB25303, Servicebio). The cantilevers were cleaned with PBS and coated with 1% pluronic for 30 min to avoid adhesion to the sample, and calibration measurements were conducted before each experiment to determine the spring constant for each cantilever, as previously described76. The sections were then transferred to AFM-compatible glass-bottom dishes coated with CellTak (Corning). For fluorescence imaging, the cantilever was positioned over areas of high or low collagen density to obtain a standard force curve. For a single measurement, the cantilever was moved to the area of interest and a standard force curve was performed. This was repeated three times on the same location. Indentations were performed at a loading force of 2.5 nN and a constant speed of 5 mm s−1. Force-time curves were analyzed using the JPK Data Processing Software. The Hertz model was used to determine the elastic properties of the tissue. The Young’s modulus was calculated by fitting the force–distance curve to the Hertzian spherical model77. Ten different positions per tissue were measured and averaged. Preparation of polyacrylamide gels substrates Granulosa cells were cultured on polyacrylamide-coated glass coverslips with tunable stiffness, prepared following established protocols as previously described78. In brief, the glass coverslips were treated with 0.1 M NaOH, followed by 3-aminopropyltriethoxysilane (Sigma) and 0.5% glutaraldehyde solution (Aladdin) for 30 min. Then, the glass coverslips were dried in the air. Acrylamide and bis-acrylamide (Servicebio) were mixed in defined ratios (determined by the stiffness) in PBS, and gel polymerization was promoted by the addition of ammonium persulfate (1/1000, Servicebio) and N,N,N′,N′-tetramethylethylenediamine (1/1000, Servicebio). The gel mixture was then dropped on the slide that had been treated with dichlorovinylmethylsilane (Aladdin). Then put the treated glass coverslips on top of the gel. After removing the top coverslips, the gels were washed with PBS and 0.05% sulfo-SANPAH (Sigma) was added. The gels were then exposed to ultraviolet light (365 nm) for 15 min to facilitate cross-linker activation and incubated with 0.1 mg ml−1 collagen I (Sigma) in PBS for overnight at 37 °C. Excess collagen I was washed off, and the gels were kept in PBS at 4 °C until cell seeding. The Young’s modulus of polyacrylamide gels was measured using AFM (Bruker). Quantitative proteomics analysis Human ovaries (n = 4 biological replicates per reproductively young, middle and old group) were snap-frozen in liquid N2 and lysed in 8 M urea, 1× protease/phosphatase inhibitor (Roche) and 1 mM PMSF. Protein concentration was determined by BCA (Pierce). Protein (100 μg) disulfides were reduced by 5 mM Tris (2-carboxyethyl) phosphine (TCEP) for 10 min at room temperature. Then, free cysteine residues were alkylated by 10 mM iodoacetamide for 30 min and were further processed with sequencing-grade trypsin for 16 h at 37 °C. Pellets were resuspended in 100 mM TEAB, digested with sequencing-grade trypsin (Promega, 1:50 wt/wt) overnight at 37 °C. Peptides were desalted (C18 Spin Columns, Thermo) and labeled with TMT-10-plex reagents (Thermo) according to the manufacturer’s protocol (1 h, room temperature). Excess reagent was quenched with 5% hydroxylamine and samples pooled in equal ratios. High-pH reverse-phase fractionation (XBridge C18, 3.5 µm, Waters) yielded 12 fractions using an acetonitrile gradient (5–35%, pH 10). Fractions were dried (SpeedVac) and resuspended in 0.1% FA. Then, 1 µg per fraction was analyzed on a Q Exactive HF-X (Thermo) coupled to an Ultimate 3000 UHPLC. Database searching against human protein sequences from the UniProt database was performed. FDR ≤ 1 % (Percolator) at peptide and protein level; unique + razor peptides used for quantification. Reporter ion intensities were log2-transformed, median-normalized and compared by moderated t-test (limma, R). DEPs were defined as |fold change| ≥ 1.2 and P ≤ 0.05. The P values in Supplementary Table 2 are nominal P values without adjustment for multiple testing. For the GO and KEGG enrichment analysis, significance was determined using Benjamini–Hochberg FDR correction for multiple testing. Isolation and culture of primary human ovarian fibroblasts Under a stereomicroscope, ovarian follicles, corpora lutea, corpora albicantia and blood vessels were removed with Vannas scissors, preserving the cortico-medullary stroma. The remaining stromal tissue was minced into 0.5-mm3 fragments. A two-step enzymatic digestion was then performed. The fragments were digested at 37 °C for 30 min with 0.8 mg ml−1 collagenase IV, 0.3 mg ml−1 dispase II and 10 μg ml−1 DNase I with orbital shaking at 100 r.p.m.; every 5 min the suspension was triturated ten times. After addition of 10% FBS to stop the reaction and centrifugation at 300g for 5 min, the pellet was resuspended in 1.2 mg ml−1 collagenase I, 0.4 mg ml−1 hyaluronidase and 10 μg ml−1 DNase I and incubated for a further 20 min under identical shaking and pipetting conditions. The reaction was again terminated with 10% FBS and centrifuged at 300g for 5 min. The cell pellet was treated with red-blood-cell lysis buffer on ice for 2 min, washed once with PBS and resuspended in complete fibroblast medium (DMEM supplemented with 5% FBS, 20 nM insulin, 20 nM selenium, 1.0 µM vitamin E and 1% penicillin-streptomycin). Cells were plated at 1 × 106 per uncoated 10-cm dish and cultured at 37 °C, 5% CO2. After 1 h, unattached epithelial cells and leukocytes were washed with fresh medium and transferred to another plate. The above steps were repeated twice to fully remove unattached cells, and the remaining cells were cultured until passage. After 48 h, the obtained P0 cells are then subjected to magnetic bead sorting, using CD45 microbeads followed by EPCAM microbeads, to retain the negative cells. Purity was confirmed by immunofluorescent staining for the fibroblast markers fibronectin and vimentin. Experiments were carried out at low cell passages (≤ passage 3) and HOFs were treated with TGFβ1 (10 ng ml−1) or IL-11 (10 ng ml−1) in serum-free DMEM for 24 h. For pathway screening studies, HOFs were stimulated with IL-11 in the presence of PF-3084014 (10 µM, MedChemExpress), IMR-1 (10 µM, MedChemExpress), PI3K/AKT-IN-1 (5 µM, MedChemExpress), BYL-719 (250 nM, MedChemExpress), KT5823 (5 µM, MedChemExpress), MBP146-78 (150 nM, MedChemExpress), p38 MAPK-IN-1 (50 nM, MedChemExpress), SB 203580 (50 nM, MedChemExpress), ravoxertinib (5 nM, MedChemExpress), SCH772984 (5 nM, MedChemExpress), JNK-IN-7 (5 nM, MedChemExpress), SP600125 (100 nM, MedChemExpress), rapamycin (50 nM, MedChemExpress) or everolimus (10 nM, MedChemExpress) for 24 h. The anti-IL-11 antibody (R&D Systems, cat. no. MAB218, human) was used to neutralize IL-11 activity, and the Neutralization Dose (ND50) was 80 µg ml−1 in the presence of 10 ng ml−1 recombinant human IL-11. Follicle isolation, encapsulation, and growth Ovaries from day 12–13 prepubertal C57BL/6 mice were placed in dissection media containing L15, 1% fetal bovine serum (FBS, Every Green) and 1% penicillin–streptomycin (Servicebio). Early secondary follicles (120–130 µm) were mechanically isolated with insulin syringes. Only follicles with an intact basement membrane and healthy oocytes surrounded by two layers of granulosa cells were selected for culture. The follicles were mixed with 10 µl of 0.5% or 2% wt/vol alginate solution. The alginate drops were then placed in a calcium solution (50 mM CaCl2 and 140 mM NaCl2) to cross-link the gel for 2 min. The resulting alginate beads were then washed and transferred to a 96-well ultra-low attachment plate (Corning) with one bead per well. Each well contained 100 µl medium (a-MEM containing 10% FBS (Every Green), 0.1% insulin–transferrin–selenium (Sigma), 10 mIU ml−1 follicle-stimulating hormone (Roche) and 1% penicillin–streptomycin (Servicebio) at 37 °C and 5% CO2. Oocyte collection for in vitro fertilization Mice were subjected to an i.p. injection of 10 IU of pregnant mare’s serum gonadotropin (PMSG). This treatment was followed by an additional intraperitoneal injection of 10 IU of human chorionic gonadotropin (HCG) at a 48-h interval post-PMSG administration. Subsequently, at 12 h after the HCG injection, the mice were euthanized. Cumulus oocyte complexes (COCs) were isolated from oviducts. The COC clusters were gently washed twice in prewarmed (37 °C) M2 Medium (MR-015-D, Sigma). 12-week-old male C57BL/6 mice were euthanized to obtain sperm. The released sperm were capacitated for 1 h (37 °C and 5% CO2) in HTF medium (MR-070, Sigma). Ovulated oocytes were added and underwent fertilization for 4 h, and zygotes were cultured in KSOM medium (MR-106-D, Sigma) at 37 °C under 5% CO2. Preparation and characterization of siIl11-loaded liposomes The mouse Il11-targeted siRNA sequences were: siIl11_1: 5′-GAGUAGACUUGAUGUCCUACC-3′, siIl11_2: 5′-GGUGGUCCUUCCCUAAAGACU-3′ and siIl11_3: 5′-GGACAGCGCUGUUCUCCUAAC-3′ (Supplementary Table 4). The specific rat Il11-targeted siRNA sequences were as follows: 5′-UGUACAUGUCGGAAGUAGGAC-3′. For transfection, lipofectamine 3000 (Invitrogen) was used according to the manufacturer’s instructions. The most effective siRNA was selected by Western Blot and RT–PCR analyses. The siRNA was dissolved in citrate buffer (10 mM, pH 3) and rapidly mixed with a lipid mixture by vortexing. The lipid mixture was composed of 54 mg cholesterol and 98 mg DOTAP dissolved in 7 ml chloroform. Following the evaporation process, the 10 mM liposome mixture was acquired. The liposomes were encapsulated with siRNA at a volume-to-mass ratio of 1.2:1. The unentrapped siRNA was removed by ultrafiltration centrifugation. The entrapment efficiency was measured by RiboGreen assay. The diameter and zeta potential were measured by dynamic light scattering (Malvern Zetasizer Nano-ZS). The morphology of the liposomes was investigated by transmission electron microscopy (Tecnai G2-20). Bulk RNA sequencing HOF cultures in six-well plates were starved overnight and stimulated with TGFβ1 (10 ng ml−1) or IL-11 (10 ng ml−1) for 24 h, total RNA was extracted using the phenol–chloroform method. The mRNA with polyA structure in total RNA is enriched using oligo(dT) magnetic beads, and the RNA is fragmented into segments of approximately 300 bp in length through ion disruption. Using RNA as a template, the first strand of cDNA is synthesized with six-base random primers and reverse transcriptase, and the second strand of cDNA is synthesized using the first-strand cDNA as a template. After the library construction is completed, PCR amplification is used to enrich the library fragments. Subsequently, the library is quality-checked using the Agilent Bioanalyzer, and the total concentration and effective concentration of the library are measured. Then, on the basis of the effective concentration of the library and the required data volume, libraries containing different Index sequences were mixed in proportion. The mixed libraries were uniformly diluted to 2 nM and denatured to form single-stranded libraries. Using next-generation sequencing (NGS) technology on the Illumina sequencing platform, these libraries are subjected to paired-end (PE) sequencing. Differential expression was performed with DESeq2 by using the raw reads count from featureCounts. Gene set enrichment analysis (GSEA) was carried out using fgsea library, MSigDB Hallmark and Gene Ontology gene sets with 105 iterations. Single-nuclei suspension preparation and RNA sequencing The nuclei of ovarian cells from 48-week-old Il11ra1−/− mice and wild-type mice were isolated using a slightly revised 10x Genomics protocol (10x Genomics). In brief, cold lysis buffer containing 10 mM Tris-HCl, 10 mM sodium chloride, 3 mM MgCl2 and 0.1% Nonidet P40 (10x Genomics) was used to break up the tissues. Then, the suspension was filtered and centrifuged, and the supernatant was discarded. The pellets containing the nuclei were washed in nuclei wash and resuspension buffer containing 1x PBS, 1% bovine serum albumin and 0.2 U µl−1 RNase inhibitor (10x Genomics), then filtered and centrifuged a second time. Single-nuclei suspensions were counted using a hemocytometer/Countess II Automated Cell Counter and concentration adjusted to 400–900 cells μl−1. Single-nuclei suspensions were loaded to 10x Chromium, according to the manufacturer’s instructions of 10x Genomics Chromium Single-Cell 3’ kit (V3). The following cDNA amplification and library-construction steps were performed according to the standard protocol. Libraries were sequenced on an Illumina NovaSeq 6000 sequencing system (paired-end multiplexing run, 150 bp) by Shanghai Personal Biotechnology. Raw sequence reads in FASTQ format from six mice ovary samples were processed and aligned to the GRCm38 mice reference transcriptome using the Cellranger v7.1.0 pipeline (https://www.10xgenomics.com/) with default parameters. The resulting gene expression matrices merged together using Seurat package v5. The preprocessing followed the guidelines provided by Seurat V5 tutorial. In short, entries with fewer than 400 genes or more than 7,500 genes in total were filtered to remove empty droplets and probable doublets, respectively, and cells that have >20% mitochondrial counts were also filtered to remove low-quality cells. To account for differences in sequencing depth across samples, we normalized expression values for total unique molecular identifiers (UMIs) per cell and log-transformed the counts using Seurat Normalize Data function. Gene set score analysis UCell (V2.3.1) was used to score gene sets in single-cell analysis, which is based on the Mann–Whitney U statistic to evaluate gene feature scores, determining whether there is a significant difference in the central tendencies between two independent samples. Estrous cycle staging For 14 consecutive days, vaginal secretions were collected between 8:00 and 9:00 a.m. from each mouse. A plastic pipette tip filled with 20 μl NaCl 0.9% was inserted into the vagina. Then, the vaginal fluid was smeared on a clean glass slide and stained with H&E. The stained glass was observed under a light microscope. Three types of cells could be recognized: epithelial cells (round and nucleated), cornified cells (irregular, without a nucleus) and leukocytes (round and small). The proportion among them was used for the determination of the estrous-cycle phases. Follicle count The mouse ovaries were collected at dioestrus. Ovaries from different groups were cut into pieces of 5 µm, and four sections were mounted on a glass slide. In brief, every fourth slide with H&E staining was observed under the microscope, and only follicles with oocytes were counted. Follicles are divided into primary follicles, secondary follicles, antral follicles and atretic follicles. Two persons counted independently, and the counting results were analyzed together79. Enzyme-linked immunosorbent assay The level of IL-11 cell culture medium was quantified using a human IL-11 ELISA kit (CSB-E04596h, CUSABIO). Blood samples were collected at 9:00 a.m. during the dioestrus stage to minimize hormonal fluctuations; cycle stage was verified by vaginal smear. The levels of AMH, E2, FSH and testosterone were measured using the following kits: mouse AMH ELISA Kit (CSB-E13156m, CUSABIO), mouse E2 ELISA Kit (CSB-E07280m, CUSABIO), Mouse FSH ELISA Kit (CSB-E06871m, CUSABIO), mouse testosterone ELISA Kit (CSB-E05098m, CUSABIO). Mating trial Virgin females were checked for estrous stage by vaginal cytology; only those in pro-estrus were paired 2:1 with proven fertile males (10 weeks) overnight (18:00–08:00). The next morning, the presence of a vaginal plug was recorded as evidence of successful mating, after which males were immediately removed. Females that did not exhibit a plug were rested for at least one estrous cycle and re-paired with a male in a subsequent round. This process was repeated such that each female was mated up to three times. Pregnancy was confirmed by plug presence and subsequent weight gain, and litter size was recorded as the number of pups delivered per pregnant female. For female mice that either did not become pregnant or experienced miscarriage, the litter size was recorded as zero and included in the statistical analysis. Collagen determination The amount of total collagen in the ovary was quantified on the basis of the colorimetric detection of hydroxyproline using Hydroxyproline Content Assay Kit (BC0250, Solarbio). The colorimetric assays were performed according to the manufacturer’s protocol. Picrosirius red staining Ovarian slides were incubated with picrosirius red staining solution (MM1036, maokangbio) for 30 min at room temperature after deparaffinization and hydration. Then, the slides were rapidly washed with 0.5% dilute acetic acid twice and then dehydrated. Digital images were captured with slide scanning systems (TEKSQRAY). ImageJ was used to quantify the area of red positive PSR staining. Masson staining Ovarian slides were stained with Masson Stain Kit (G1006, Servicebio Technology Company). The staining process was conducted in strict accordance with the guidelines outlined by the manufacturer. Subsequent to staining, the slides were examined and digital images were obtained using the microscope (version 1.8.1, Olympus). ImageJ was used to quantify the area of red positive PSR staining. Immunofluorescence and immunohistochemical For immunofluorescence (IF) staining, 4-µm ovary sections or HOFs were permeabilized with 0.2% Triton X-100 and blocked with 2% BSA. Then, the sections or cells were incubated with primary antibodies overnight at 4 °C. After washing, the sections or cells were incubated with Alexa-Fluor-488-conjugated anti-rabbit IgG (GB25303, Servicebio) secondary antibodies (1:200 dilution) for 1 h at room temperature. Images were obtained using the fluorescence microscope (Olympus). The IF primary antibodies used are listed in Supplementary Table 4. For immunohistochemical (IHC) staining, ovarian slides were blocked with 2% BSA for 30 min at room temperature after deparaffinization and hydration. Then, they were incubated with primary antibodies overnight at 4 °C and anti-rabbit isotype-specific HRP secondary antibodies. Next, the ovary sections were visualized with DAB-HRP chromogenic agent (G1212, Servicebio) at room temperature for 10–30 s. Images were acquired with the slide-scanning systems (TEKSQRAY). The relative expression was evaluated using Image Pro Plus software. The IHC primary antibodies used are listed in Supplementary Table 4. Quantitative real-time polymerase chain reactions Total RNA was extracted using the phenol–chloroform method. Reverse transcription kits (R423-01, Vazyme) were used to convert the RNA into complementary DNA. Quantitative PCR was conducted using ChamQ Universal SYBR qPCR Master Mix (Q711-02, Vazyme) in a CFX96 real-time PCR system (Bio-Rad). The messenger RNA expression level of target genes was calculated using the 2−ΔΔCt. The primer sequences are listed in Supplementary Table 4. Western blot Total protein was isolated from cultured cells or ovaries by using a radioimmunoprecipitation assay, according to the routine procedure. Membranes were incubated with different primary antibodies overnight at 4 °C. The antibodies used are listed in Supplementary Table 4. Chemiluminescence detection of blotting results was performed using the Image Lab (BioRad). GAPDH or β-actin expression was measured to verify equal loading. Proteome detection of human phospho-kinase To screen the activity of biochemical pathways, 300 µg protein lysate of PBS- and IL-11-treated HOFs were profiled for the phosphokinase activity using the respective dot blotting kit (ARY003B, R&D Systems), according to the manufacturer’s protocol. Chemiluminescence detection of dot-blotting results was performed using the Image Lab (BioRad). Statistics and reproducibility No statistical methods were used to predetermine sample sizes. Sample sizes for mouse and rat experiments were chosen on the basis of animal availability and experimental feasibility. Animals were randomly allocated to experimental groups. Human samples were selected on the basis of the diagnostic criteria for each disease; only those meeting the criteria were included. The investigators were not blinded to allocation during experiments or outcome assessment. No data were excluded from the analyses. n represents the number of samples used in the experiments. All in vitro experiments used at least three biological replicates to meet the minimum requirement for parametric statistical testing. All results are analyzed using GraphPad Prism software (version 9.0). Data are presented as mean ± s.d. Sample size (n) and the statistical test used are indicated in each legend. P < 0.05 was considered statistically significant. A Spearman’s rank correlation coefficient (r) was used to calculate the correlation-associated statistical significance in GraphPad. Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Data availability The raw mass spectrometry proteomics data that support Extended Data Figure 3a,b,e and Supplementary Table 2 have been irretrievably lost due to irreversible physical damage (head crash) to the dedicated laboratory hard drive on which they were stored. Professional data-recovery attempts were unsuccessful, and no complete off-site backup exists for these specific raw files. Consequently, these raw data cannot be deposited in any public repository. However, all processed proteomic data that directly generated the table and figures, including the DEPs list (which is Supplementary Table 2), heatmap source data, GO and KEGG enrichment results, are still intact and have been provided as source data files with this manuscript. The quantitative results and statistical comparisons presented in the paper are fully supported by these processed files. All other relevant data supporting the findings of this study are available in the article or from the corresponding authors upon reasonable request. The RNA-seq data can be found in the Sequence Read Archive (https://www.ncbi.nlm.nih.gov/) under accession numbers SRR38486614, SRR38486615, SRR38486616, SRR38486617, SRR38486618 and SRR38486619. The snRNA-seq data are available under accession numbers SRR38487513, SRR38487514, SRR38487515, SRR38487516, SRR38487517 and SRR38487518.

Seifer, D. B., Feinberg, E. C. & Hsu, A. L. Ovarian aging and fertility. JAMA 332, 1750–1751 (2024). Carson, S. A. & Kallen, A. N. Diagnosis and management of infertility: a review. JAMA 326, 65–76 (2021). Nichols, A. R., Chavarro, J. E. & Oken, E. Reproductive risk factors across the female lifecourse and later metabolic health. Cell Metab. 36, 240–262 (2024). Guo, Y. et al. Ovarian microenvironment: challenges and opportunities in protecting against chemotherapy-associated ovarian damage. Hum. Reprod. Update 30, 614–647 (2024). Fiorentino, G. et al. Biomechanical forces and signals operating in the ovary during folliculogenesis and their dysregulation: implications for fertility. Hum. Reprod. Update 29, 1–23 (2023). Salustri, A., Campagnolo, L., Klinger, F. G. & Camaioni, A. Molecular organization and mechanical properties of the hyaluronan matrix surrounding the mammalian oocyte. Matrix Biol. 78-79, 11–23 (2019). Mittelheisser, V. et al. Evidence and therapeutic implications of biomechanically regulated immunosurveillance in cancer and other diseases. Nat. Nanotechnol. 19, 281–297 (2024). Ouni, E. et al. Spatiotemporal changes in mechanical matrisome components of the human ovary from prepuberty to menopause. Hum. Reprod. 35, 1391–1410 (2020). Shah, J. S. et al. Biomechanics and mechanical signaling in the ovary: a systematic review. J. Assist. Reprod. Genet. 35, 1135–1148 (2018). Hsueh, A. J. W. & Kawamura, K. Hippo signaling disruption and ovarian follicle activation in infertile patients. Fertil. Steril. 114, 458–464 (2020). Li, J. et al. Activation of dormant ovarian follicles to generate mature eggs. Proc. Natl Acad. Sci. USA 107, 10280–10284 (2010). Kawashima, I. & Kawamura, K. Regulation of follicle growth through hormonal factors and mechanical cues mediated by Hippo signaling pathway. Syst. Biol. Reprod. Med 64, 3–11 (2018). Nagamatsu, G., Shimamoto, S., Hamazaki, N., Nishimura, Y. & Hayashi, K. Mechanical stress accompanied with nuclear rotation is involved in the dormant state of mouse oocytes. Sci. Adv. 5, eaav9960 (2019). Hornick, J. E., Duncan, F. E., Shea, L. D. & Woodruff, T. K. Isolated primate primordial follicles require a rigid physical environment to survive and grow in vitro. Hum. Reprod. 27, 1801–1810 (2012). Woodruff, T. K. & Shea, L. D. A new hypothesis regarding ovarian follicle development: ovarian rigidity as a regulator of selection and health. J. Assist Reprod. Genet 28, 3–6 (2011). Jones, A. S. K. et al. Cellular atlas of the human ovary using morphologically guided spatial transcriptomics and single-cell sequencing. Sci. Adv. 10, eadm7506 (2024). Chen, X. et al. Micromechanical analysis of the hyaluronan-rich matrix surrounding the oocyte reveals a uniquely soft and elastic composition. Biophys. J. 110, 2779–2789 (2016). Pietroforte, S., Plough, M. & Amargant, F. Age-associated increased stiffness of the ovarian microenvironment impairs follicle development and oocyte quality and rapidly alters follicle gene expression. Preprint at bioRxiv https://doi.org/10.1101/2024.06.09.598134 (2024). McCloskey, C. W. et al. Metformin abrogates age-associated ovarian fibrosis. Clin. Cancer Res. 26, 632–642 (2020). Amargant, F. et al. Ovarian stiffness increases with age in the mammalian ovary and depends on collagen and hyaluronan matrices. Aging Cell 19, e13259 (2020). Briley, S. M. et al. Reproductive age-associated fibrosis in the stroma of the mammalian ovary. Reproduction 152, 245–260 (2016). Lu, H. et al. Aging hallmarks of the primate ovary revealed by spatiotemporal transcriptomics. Protein Cell 15, 364–384 (2024). Zaniker, E. J. et al. Shear wave elastography to assess stiffness of the human ovary and other reproductive tissues across the reproductive lifespan in health and diseasedagger. Biol. Reprod. 110, 1100–1114 (2024). Umehara, T. et al. Female reproductive life span is extended by targeted removal of fibrotic collagen from the mouse ovary. Sci. Adv. 8, eabn4564 (2022). Wu, C., Chen, D., Stout, M. B., Wu, M. & Wang, S. Hallmarks of ovarian aging. Trends Endocrinol. 36, 418–439 (2025). Dai, Y. et al. FSHR-targeting tetrahedral DNA frameworks ameliorate ovarian aging through oxidative stress elimination. J. Nanobiotechnology 24, 224 (2026). Tang, W. et al. Exploration of the mechanism and therapy of ovarian aging by targeting cellular senescence. Life Med. 4, lnaf004 (2025). Niu, C. et al. Spermidine suppresses oxidative stress and ferroptosis by Nrf2/HO-1/GPX4 and Akt/FHC/ACSL4 pathway to alleviate ovarian damage. Life Sci. 332, 122109 (2023). Zhang, C., Ma, J., Wang, W., Sun, Y. & Sun, K. Lysyl oxidase blockade ameliorates anovulation in polycystic ovary syndrome. Hum. Reprod. 33, 2096–2106 (2018). Alikhani, M. et al. Proteome analysis of endometrial tissue from patients with PCOS reveals proteins predicted to impact the disease. Mol. Biol. Rep. 47, 8763–8774 (2020). Papachroni, K. K. et al. Lysyl oxidase interacts with AGE signalling to modulate collagen synthesis in polycystic ovarian tissue. J. Cell Mol. Med 14, 2460–2469 (2010). Shi, L. B. et al. Transforming growth factor beta1 from endometriomas promotes fibrosis in surrounding ovarian tissues via Smad2/3 signaling. Biol. Reprod. 97, 873–882 (2017). Garcia Garcia, J. M. et al. Endometriosis: cellular and molecular mechanisms leading to fibrosis. Reprod. Sci. 30, 1453–1461 (2023). Mendez, M. et al. Biomechanical characteristics of the ovarian cortex in POI patients and functional outcomes after drug-free IVA. J. Assist Reprod. Genet 39, 1759–1767 (2022). He, Y. et al. Evaluation of ovarian stiffness and its biological mechanism using shear wave elastography in polycystic ovary syndrome. Sci. Rep. 15, 585 (2025). Sumbul, H. E. et al. Ovarian stiffness is significantly increased in polycystic ovary syndrome and related with anti-mullerian hormone: a point shear wave elastography study. Ultrasound Q 38, 83–88 (2022). Yildirim, M., Duran Kaymak, S., Cayonu Kahraman, N., Savran Ucok, B. & Engin-Ustun, Y. Assessment of ovarian stiffness in women with polycystic ovary syndrome: a tertiary center shear wave elastography study. J. Ultrasound Med. 44, 1797–1806 (2025). Abu Hashim, H. Predictors of success of laparoscopic ovarian drilling in women with polycystic ovary syndrome: an evidence-based approach. Arch. Gynecol. Obstet. 291, 11–18 (2015). Stener-Victorin, E. et al. Polycystic ovary syndrome. Nat. Rev. Dis. Prim. 10, 27 (2024). Kawamura, K. et al. Hippo signaling disruption and Akt stimulation of ovarian follicles for infertility treatment. Proc. Natl Acad. Sci. USA 110, 17474–17479 (2013). Suzuki, N. et al. Successful fertility preservation following ovarian tissue vitrification in patients with primary ovarian insufficiency. Hum. Reprod. 30, 608–615 (2015). Tanaka, Y., Hsueh, A. J. & Kawamura, K. Surgical approaches of drug-free in vitro activation and laparoscopic ovarian incision to treat patients with ovarian infertility. Fertil. Steril. 114, 1355–1357 (2020). Lunding, S. A. et al. Biopsying, fragmentation and autotransplantation of fresh ovarian cortical tissue in infertile women with diminished ovarian reserve. Hum. Reprod. 34, 1924–1936 (2019). Zhou, Y. et al. Value of [(68)Ga]Ga-FAPI-04 imaging in the diagnosis of renal fibrosis. Eur. J. Nucl. Med Mol. Imaging 48, 3493–3501 (2021). West, E. R., Xu, M., Woodruff, T. K. & Shea, L. D. Physical properties of alginate hydrogels and their effects on in vitro follicle development. Biomaterials 28, 4439–4448 (2007). Green, L. J. & Shikanov, A. In vitro culture methods of preantral follicles. Theriogenology 86, 229–238 (2016). Converse, A., Zaniker, E. J., Amargant, F. & Duncan, F. E. Recapitulating folliculogenesis and oogenesis outside the body: encapsulated in vitro follicle growthdagger. Biol. Reprod. 108, 5–22 (2023). Mattar, D., Samir, M., Laird, M. & Knight, P. G. Modulatory effects of TGF-beta1 and BMP6 on thecal angiogenesis and steroidogenesis in the bovine ovary. Reproduction 159, 397–408 (2020). Ingman, W. V. & Robertson, S. A. Defining the actions of transforming growth factor beta in reproduction. BioEssays 24, 904–914 (2002). Ingman, W. V., Robker, R. L., Woittiez, K. & Robertson, S. A. Null mutation in transforming growth factor beta1 disrupts ovarian function and causes oocyte incompetence and early embryo arrest. Endocrinology 147, 835–845 (2006). Wu, M. et al. Spatiotemporal transcriptomic changes of human ovarian aging and the regulatory role of FOXP1. Nat. Aging 4, 527–545 (2024). Rodgers, R. J., Irving-Rodgers, H. F. & Russell, D. L. Extracellular matrix of the developing ovarian follicle. Reproduction 126, 415–424 (2003). Camaioni, A. et al. The process of ovarian aging: it is not just about oocytes and granulosa cells. J. Assist Reprod. Genet. 39, 783–792 (2022). Malfait, F. et al. The 2017 international classification of the Ehlers–Danlos syndromes. Am. J. Med. Genet. C Semin. Med. Genet. 175, 8–26 (2017). Zeigler, S. M., Sloan, B. & Jones, J. A. Pathophysiology and pathogenesis of Marfan syndrome. Adv. Exp. Med. Biol. 1348, 185–206 (2021). Mercuri, E., Bonnemann, C. G. & Muntoni, F. Muscular dystrophies. Lancet 394, 2025–2038 (2019). Hurst, B. S. et al. Obstetric and gynecologic challenges in women with Ehlers–Danlos syndrome. Obstet. Gynecol. 123, 506–513 (2014). Dodion, J. & Parmentier, R. Marfan’s syndrome with mitral insufficiency, mental retardation and ovarian agenesis]. Acta Paediatrica Belgica 23, 17–26 (1969). Meijboom, L. J. et al. Obstetric complications in Marfan syndrome. Int J. Cardiol. 110, 53–59 (2006). Sahu, B. et al. Response to controlled ovarian stimulation and oocyte quality in women with myotonic dystrophy type I. J. Assist Reprod. Genet 25, 1–5 (2008). Izquierdo, J. M. Pro-inflammatory cytokine 11 plays a pivotal role in inflammaging-associated pathologies. Aging Cell 23, e14360 (2024). Widjaja, A. A. et al. Targeting endogenous kidney regeneration using anti-IL11 therapy in acute and chronic models of kidney disease. Nat. Commun. 13, 7497 (2022). Seyedsadr, M. et al. IL-11 induces NLRP3 inflammasome activation in monocytes and inflammatory cell migration to the central nervous system. Proc. Natl Acad. Sci. USA 120, e2221007120 (2023). Widjaja, A. A. et al. Inhibition of IL-11 signalling extends mammalian healthspan and lifespan. Nature 632, 157–165 (2024). Branisteanu, I. et al. Detection of immunoreactive interleukin-11 in human follicular fluid: correlations with ovarian steroid, insulin-like growth factor I levels, and follicular maturity. Fertil. Steril. 67, 1054–1058 (1997). Zhuang, Z. et al. The effect of interleukin-6 (IL-6), interleukin-11 (IL-11), signal transducer and activator of transcription 3 (STAT3), and AKT signaling on adipocyte proliferation in a rat model of polycystic ovary syndrome. Med. Sci. Monit. 25, 7218–7227 (2019). Liu, W. et al. Successful reversal of ovarian hyperstimulation syndrome in a mouse model by rapamycin, an mTOR pathway inhibitor. Mol. Hum. Reprod. 25, 445–457 (2019). Jang, Y. J. et al. Regulation of interleukin-11 expression in ovulatory follicles of the rat ovary. Reprod., Fertil. Dev. 29, 2437–2445 (2017). Wu, H. et al. Effects of IL-11/IL-11 receptor alpha on proliferation and steroidogenesis in ovarian granulosa cells of dairy cows. Cells 12, 673 (2023). Bai, X. et al. Inhaled siRNA nanoparticles targeting IL11 inhibit lung fibrosis and improve pulmonary function post-bleomycin challenge. Sci. Adv. 8, eabn7162 (2022). Schafer, S. et al. IL-11 is a crucial determinant of cardiovascular fibrosis. Nature 552, 110–115 (2017). Ng, B. et al. Similarities and differences between IL11 and IL11RA1 knockout mice for lung fibro-inflammation, fertility and craniosynostosis. Sci. Rep. 11, 14088 (2021). Deguchi, Y. et al. Generation of and characterization of anti-IL-11 antibodies using newly established Il11-deficient mice. Biochem. Biophys. Res. Commun. 505, 453–459 (2018). Liu, Y. et al. Artemisinins ameliorate polycystic ovarian syndrome by mediating LONP1–CYP11A1 interaction. Science 384, eadk5382 (2024). Zhang, J. et al. Osr2 functions as a biomechanical checkpoint to aggravate CD8+ T cell exhaustion in tumor. Cell 187, 3409–3426 (2024). Shen, Y., Schmidt, T. & Diz-Munoz, A. Protocol on tissue preparation and measurement of tumor stiffness in primary and metastatic colorectal cancer samples with an atomic force microscope. STAR Protoc. 1, 100167 (2020). Lin, D. C., Shreiber, D. I., Dimitriadis, E. K. & Horkay, F. Spherical indentation of soft matter beyond the Hertzian regime: numerical and experimental validation of hyperelastic models. Biomech. Model Mechanobiol. 8, 345–358 (2009). Hsieh, W. T. et al. Matrix dimensionality and stiffness cooperatively regulate osteogenesis of mesenchymal stromal cells. Acta Biomater. 32, 210–222 (2016). Wu, M. et al. Microenvironmentally responsive chemotherapeutic prodrugs and CHEK2 inhibitors self-assembled micelles: protecting fertility and enhancing chemotherapy. Adv. Mater. 35, e2210017 (2023).

This work was financially supported by the National Natural Science Foundation of China (nos. 82471678, 82403482, 8240062640, 22522402), National Key Research and Development Program of China (no. 2024YFC2707404), the China Postdoctoral Science Foundation (nos. 2024M752492, 2024M751020), the Natural Science Foundation of Hubei Province (no. 2025AFA075), and the Hubei Province Postdoctoral Innovation Talent Development Program (no. 2004HBBHCXA072). We thank D. Jiang from Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, for providing the ovarian 68Ga-FAPI-04 PET images. We thank Shanghai Personalbio for their support of RNA-seq library preparation. Author information Authors and Affiliations Contributions Conceptualization: M.W., S.X.W., J.D., Y.L., J.J.Z. Methodology: M.W., Q.Q.Z., J.Q.X., Y.C.G., S.M.W., P.Z.Z., Y.Q.S. Software: M.W., J.Q.X. Formal Analysis: J.Q.X, T.D. Investigation: W.C.T., D.C., L.R.X., Z.F.L., Y.D., T.W., C.Q.W., Y.T.L. Data Curation: Y.B.H., P.Z.Z., Y.T.L., Y.Q.S., M.F.W. Writing–original draft: M.W., J.Q.X. Writing–review and editing: J.D., J.J.Z., W.W.W., Y.G.R., Q.Q.Z., Y.R.F., Y.B.H. Supervision: S.X.W., Y.L. Funding Acquisition: S.X.W., J.D., M.W., J.Q.X., W.C.T., D.C. Corresponding authors Ethics declarations Competing interests The authors declare no competing interests. Peer review Peer review information Nature Aging thanks Farners Amargant, Anna Benrick, and Stuart Cook for their contribution to the peer review of this work. Peer reviewer reports are available. Additional information Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Extended data Extended Data Fig. 1 Hematoxylin and Eosin (H&E), Masson, and Picrosirius red staining of human ovarian tissue. (A) Hematoxylin and Eosin (H&E) staining of human ovary tissue from different ages and pathophysiological states. (B) Representative Masson staining in the ovary specimens from females with different ages or ovary diseases. (C and D) Picrosirius red staining of human ovarian tissue from different ages and pathological conditions, shown under bright-field (C) and polarized light (D). (E) 68Ga-FAPI-04 PET detection of ovarian fibrosis. Representative images from young and old individuals are presented, along with the max standardized uptake values (SUV). Extended Data Fig. 2 Increased ovarian stiffness impairs follicle development. (A) Representative transmitted light images of mice follicles cultured in 0.5% (soft) and 2% (stiff) alginate beads at day 7. (B) Graph showing the growing dynamics of mice follicles. Data are presented as the mean ± SD, n = 80 per group (unpaired two-tailed t-test). (C) Estradiol concentrations in culture media at the final day of culture for mice follicles. Data are presented as the mean ± SD, n = 15 per group (unpaired two-tailed t-test). (D) Progesterone concentrations in culture media at the final day of culture for mice follicles. Data are presented as the mean ± SD, n = 15 per group (unpaired two-tailed t-test). (E) Representative images of mice follicles immunofluorescence staining with Ki67. (F) Schematic diagram for collagen-coated polyacrylamide gels with elastic moduli of 3 kPa (‘soft’) and 30 kPa (‘stiff’). (G) Phase images showing typical morphology of mouse ovarian primary granulosa cells cultured on soft and stiff gels. (H) Cell surface areas calculated by digital image analysis of phase-contrast images of cells on soft and stiff gels. Data are presented as the mean ± SD, n = 40 per group (unpaired two-tailed t-test). (I) Representative immunofluorescence images of phalloidin. (J and K) The proliferation of granulosa cells cultured on soft versus stiff substrates was measured by EdU assay. Data are presented as the mean ± SD, n = 20 per group (unpaired two-tailed t-test). (L) Representative images of immunofluorescence staining for CYP19A1 in granulosa cells cultured on soft and stiff gels. (M) The levels of estradiol in granulosa cells culture supernatant. Data are presented as the mean ± SD, n = 15 per group (unpaired two-tailed t-test). (N) The levels of progesterone in granulosa cells culture supernatant. Data are presented as the mean ± SD, n = 15 per group (unpaired two-tailed t-test). Extended Data Fig. 3 TGFβ1 are involved in the regulation of ovarian matrix stiffness. (A) Proteomic sequencing of human ovarian tissues from females at different ages. (n = 4 per group). (B) GO enrichment analysis of differentially expressed proteins in human ovarian tissue proteomics. Y (young group), M (middle-aged group), O (old-aged group). Red labels highlight extracellular matrix organization and extracellular space as the predominant biological processes. (C) Immumohistochemical staining of COL1A1, COL1A2, COL3A1 and COL4A1 in human ovaries. (D) Immumohistochemical scores of relative expression of COL1A1, COL1A2, COL3A1 and COL4A1. n = 9 per group. Data are presented as the mean ± SD (one-way ANOVA). IOD, integrated optical density. (E) KEGG enrichment analysis of differentially expressed proteins in human ovarian tissue proteomics. Red labels highlight the TGFβ signaling pathway. (F) Representative images for TGFβ1 immumohistochemical in human ovary sections from reproductively young, middle and old females (Left). The scores are listed on the right. IgG served as a negative control. n = 9 per group. Data are presented as the mean ± SD (one-way ANOVA). IOD is as defined in panel D. (G) Representative images for TGFβ1 immumohistochemical in human ovary sections from normal, chemotherapy-induced POI, PCOS, and endometriosis. n = 9 per group. Data are presented as the mean ± SD (one-way ANOVA). IOD is as defined in panel D. (H) Multicolor immunohistochemical staining of COL1A1, TGFβ1, and DCN in human ovarian tissue. Extended Data Fig. 4 The expression of TGFβ1 and TGFβR1 in human ovarian cells. (A) UMAP plots showing eight cell types in human ovary, TGFβ1 expression, and TGFβ receptor-1 (TGFβR1) expression. T&S, theca and stroma cells; TCs, T lymphocytes; ECs, endothelial cell; MONO, monocytes; GCs, granulosa cells; SMCs, smooth muscle cells; OO, oocyte. (B) Corresponding UMAPs of the aged human ovaries showing cell-type identity, TGFβ1, and TGFβR1 expression. (C) The separation process of primary human ovarian fibroblasts. (D) and (E) Immunofluorescence of fibronectin and vimentin in primary human ovarian fibroblasts. Extended Data Fig. 5 IL-11 is a key target in regulating the ovarian matrix stiffness. (A) Heatmap of differentially expressed genes in primary human ovary fibroblasts (pHOF) after in vitro TGFβ1 treatment. (B) GO and KEGG analysis of differentially expressed genes between the control and TGFβ1-treated pHOF. Red labels highlight ECM secretion as the key biological process. (C) Western blots of COL1A1, COL1A2, and ACTA2 protein expression in pHOF in response to TGFβ1. (D) IL-11 expression in pHOF following stimulation with various fibrotic factors. (E) Detection of IL-11 concentration in the culture medium of pHOF following TGFβ1 treatment using ELISA. Data are presented as the mean ± SD, n = 5 per group (unpaired two-tailed t-test). (F) Human ovarian IL11 mRNA levels in young (18–28 years), middle-aged (36–39 years), and old (47–49 years) groups. Murine ovarian Il11 mRNA levels in young (3 months), middle-aged (9 months), and old (12 months) mice. Rat ovarian Il11 mRNA levels in young (3 months), middle-aged (9 months), and old (12 months) rats. Data are presented as the mean ± SD, n = 6 per group (one-way ANOVA). (G and H) Analysis of age-related IL-11 protein abundance in human, mouse, and rat ovaries. Data are presented as the mean ± SD, n = 6 per group (unpaired two-tailed t-test). (I) Immunohistochemical detection of IL-11 in human ovarian tissue across different ages and pathophysiological states. (J) Correlation of ovarian IL-11 expression with age and AMH levels. n = 40 (Pearson correlation analysis, two-sided). (K) Immunofluorescence staining for IL-11RA in pHOF. (n = 3 experiments). (L) Quantification of pHOF immunostained for COL1A1 and ACTA2 after 24 h incubation without stimulus (Control), TGFβ1 (10 ng/ml) or IL-11 (10 ng/ml) or with TGFβ1 (10 ng/ml) and an Anti-IL-11 neutralising antibody (2 μg/ml). Data are presented as the mean ± SD, n = 5 per group (one-way ANOVA). (M and N) Representative images and quantification of wound-healing assay in pHOF treated with TGFβ1 or IL-11 or with TGFβ1 and an Anti-IL-11 neutralising antibody. Data are presented as the mean ± SD, n = 5 per group (one-way ANOVA). (O) Representative images and quantification of stained pHOF migrating from the upper chamber to the lower chamber after treatment of TGFβ1 or IL-11 or with TGFβ1 and an Anti-IL-11 neutralising antibody. Data are presented as the mean ± SD, n = 5 per group (one-way ANOVA). Extended Data Fig. 6 IL-11 regulates ovarian matrix stiffness through the ERK1/2 signaling pathway. (A) Schematic diagram of RNA-seq of primary pHOF after in vitro IL-11 treatment (10 ng/ml, 24 h). Principal component analysis of the RNA-seq. (n = 3 per group). (B) Analysis of the expression of collagen-related genes between the control and IL-11 group. (n = 3 per group). (C and D) GSEA analysis of the differentially expressed genes in two groups. (E) Human phospho-kinase array of pHOF treated with IL-11 (10 ng/mL) or vehicle for 24 h. Experiments were performed in duplicate. The numbers 1–8 correspond to the same phosphorylated residues as defined in the legend of Fig. 2j,k. (F) Western blot analysis of ERK1/2 and its downstream targets in pHOF treated with IL-11, with or without the ERK1/2 inhibitor SCH772984. (G and H) Age-related changes in ERK expression in human and mice ovaries. Data are presented as the mean ± SD, n = 6 per group for human, n = 3 per group for mice (one-way ANOVA). Extended Data Fig. 7 General condition of mice following IL-11 treatment. (A) Organ index of major organs. Data are presented as the mean ± SD, n = 16 per group (unpaired two-tailed t-test). (B) Serum biomarkers for cardiac (CK), hepatic (ALT), and renal (LDH-L, CRE) function. Data are presented as the mean ± SD, n = 5 per group (unpaired two-tailed t-test). Extended Data Fig. 8 Analysis of single-cell nuclei transcriptome sequencing of mice ovaries. (A) UMAP plots showing nine ovarian cell types in 48-week-old WT and Il11ra1 KO mice. (n = 3 per group). (B) Dot plot heat map showing markers for the nine ovarian cell types. (C) UMAP plot showing Col4a2, Col4a3, Col4a4, Col5a1, Col5a2, Col6a3, and Col11a1 expressed ovary cells. (D) Dot plot heat map showing markers for the ovarian stromal cells subclusters. Extended Data Fig. 9 Increased ovarian matrix stiffness in chemotherapy-induced POI mice models. (A) Schematic illustration of the experimental workflow for establishing POI mice model using chemotherapeutic agents, including cyclophosphamide (CTX), cisplatin (CIS), doxorubicin (DOX), paclitaxel (PTX). (B) Quantification of the matrix stiffness of ovaries from five groups. Data are presented as the mean ± SD, n = 6 per group (unpaired two-tailed t-test). (C) Quantitative of the hydroxyproline of ovaries from five groups. Data are presented as the mean ± SD, n = 6 per group (unpaired two-tailed t-test). Extended Data Fig. 10 Preparation of Il11 siRNA-loaded liposomes and AAV-sh-Il11ra1 treatment. (A) RT-PCR analyses of the interference efficiency of Il11 siRNAs in primary mice ovary fibroblasts. Data are presented as the mean ± SD, n = 3 per group (unpaired two-tailed t-test). (B) Western blot analyses of the interference efficiency of Il11 siRNAs in primary mice ovary fibroblasts. (C) RT-PCR analyses of the interference efficiency of Il11 siRNA in primary rat ovary fibroblasts. Data are presented as the mean ± SD, n = 3 per group (unpaired two-tailed t-test). (D) Western blot analyses of the interference efficiency of Il11 siRNA in primary rat ovary fibroblasts. (E) Schematic diagram showing the preparation of Il11 siRNA-loaded liposomes. (F) The hydrodynamic diameter, polymer dispersity index (PDI) and zeta potential of the liposomes (blank or siRNA-loaded) were measured by dynamic light scattering (DLS). siRNA entrapment efficiency was measured by RiboGreen assay. (G) Representative transmission electron microscopy (TEM) image of siRNA-loaded liposomes. (H) Hydrodynamic diameter distribution of siRNA-loaded liposomes. (I) Colloid stability of siRNA-loaded liposomes in PBS. (J) Flowchart for the evaluation of matrix stiffness and ovarian function in 48-week-old mice after ovarian micro-injection of AAV-sh-Il11ra1. (n = 25 per group). (K) Assessment of ovarian matrix stiffness using atomic force microscopy. (L) Quantification the collagen content within ovarian tissue by hydroxyproline assay. (M) Representative micrographs of paraffin-embedded ovary stained histochemically for Picrosirius red. (N) Representative images of H&E stained ovarian sections from two groups. (O) Primordial follicle (PMF), primary follicle (PF), secondary follicle (SF), antral follicle (ANF), atretic follicle (ATF), total healthy follicles (THF) and corpus luteum (CL) quantification in ovaries. (P) Average litter size of total mated mice. Data are presented as the mean ± SD, n = 6 per group (unpaired two-tailed t-test). Supplementary information Supplementary Table 1 (download XLSX ) Information for human ovarian samples. Supplementary Table 2 (download XLSX ) Differentially expressed proteins in human ovarian tissue. Supplementary Table 3 (download XLSX ) Markers of each cell type in ovary. Supplementary Table 4 (download XLSX ) Antibody and sequences for RT–PCR, siRNA. Source data Source Data for Main Figures and Extended Data Figures (download XLS ) Statistical source data for Figs. 1–4 and 6–8 and Extended Data Figs. 2, 3, 5–7, 9 and 10. Source Data Fig. 2 (download PDF ) Unprocessed western blots for Fig. 2. Source Data Fig. 3 (download PDF ) Unprocessed western blots for Fig. 3. Source Data Fig. 4 (download PDF ) Unprocessed western blots for Fig. 4. Source Data Fig. 6 (download PDF ) Unprocessed western blots for Fig. 6. Source Data Fig. 7 (download PDF ) Unprocessed western blots for Fig. 7. Source Data Fig. 8 (download PDF ) Unprocessed western blots for Fig. 8. Source Data Extended Data Fig. 5 (download PDF ) Unprocessed western blots for Extended Data Fig. 5. Source Data Extended Data Fig. 6 (download PDF ) Unprocessed western blots for Extended Data Fig. 6. Source Data Extended Data Fig. 10 (download PDF ) Unprocessed western blots for Extended Data Fig. 10. Source Data Extended Data Fig. 3 (download XLS ) All differentially expressed proteins in human ovarian tissue proteomics for Extended Data Fig. 3A. Source Data Extended Data Fig. 3 (download XLS ) GO enrichment analysis in human ovarian tissue proteomics for Extended Data Fig. 3B. Source Data Extended Data Fig. 3 (download XLS ) KEGG enrichment analysis in human ovarian tissue proteomics for Extended Data Fig. 3E. Rights and permissions 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/. About this article Cite this article Wu, M., Zhu, Q., Xiong, J. et al. Modulating IL-11-dependent matrix stiffness to delay ovarian aging. Nat Aging (2026). https://doi.org/10.1038/s43587-026-01159-2 Received: Accepted: Published: Version of record: DOI: https://doi.org/10.1038/s43587-026-01159-2