The ovary holds a peculiar distinction in human biology: it is arguably the fastest-aging organ in the body. Women are born with a fixed reserve of ovarian follicles, and the depletion of that reserve dictates both fertility and the timing of menopause, which typically arrives around age 50. Because the ovary’s decline precedes overt aging elsewhere by decades, and because menopause is linked to elevated risks of cardiovascular disease, osteoporosis, neurodegeneration, and ovarian cancer, the organ’s influence on women’s health span is outsized. Yet despite growing longevity that leaves many women spending a third or more of their lives postmenopausal, remarkably little is known about what the ovary is actually doing during those years. Classical histology described it as a reduced-volume, fibrotic organ with cortical thinning, loss of cortico-medullary distinction, epithelial inclusion cysts, and abundant corpora albicantia, the scar-like bodies left behind by spent follicles. Molecular studies, meanwhile, have focused overwhelmingly on premenopausal tissue or the early menopausal transition, leaving the proteome of the truly aged, normal ovary essentially unmapped.

To fill that gap, the team, working through the Northwestern University Reproductive Tissue Library, collected non-pathological ovarian tissue from women aged 50 to 75 undergoing surgery for benign gynecologic indications, with every sample confirmed as histologically normal by a board-certified gynecological pathologist. The cohort was stratified into three groups: 50 to 59 years (11 participants), 60 to 69 years (10 participants), and 70 or older (7 participants). The researchers then applied a deep quantitative proteomics workflow known as data-independent acquisition mass spectrometry with parallel accumulation-serial fragmentation, or diaPASEF, a technique in which the mass spectrometer systematically fragments and quantifies peptides across every sample in an unbiased fashion. The analysis yielded 5812 quantified protein groups, each supported by at least two unique peptides, providing an unusually comprehensive molecular census of the postmenopausal ovary.

The pairwise comparisons between age cohorts told a story of progressive transformation. When ovaries from women aged 60 to 69 were compared with those from women aged 50 to 59, 34 protein groups increased and 6 decreased, with strict statistical thresholds of greater than 1.5-fold change and q-value below 0.01. The rising proteins clustered around complement and humoral immunity, including Complement Factor H-related proteins 1 and 5, Complement Factor B, Complement Component C9, and Clusterin, alongside matrix and scavenger proteins such as the metalloproteinase ATS16, the crosslinking enzyme transglutaminase-2, and vitronectin. By the oldest comparison, the shift was far larger: 84 protein groups decreased and 33 increased in ovaries from women aged 70 or older relative to the youngest cohort. Structural and cytoskeletal proteins, including Myosin-11, Filamin-A and C, Dystrophin, and multiple keratins, were depleted, while damage-associated secretory proteins such as Alpha-2-macroglobulin and EMILIN-3 rose. A core set of seventeen proteins was significantly altered in both comparisons, pointing to overlapping biology across the postmenopausal decades, from complement activation to lipid handling and sulfated glycosaminoglycan metabolism.

Multivariate modeling sharpened the picture. Partial least squares-discriminant analysis, a supervised technique that finds the protein axes best separating predefined groups, showed only modest separation between the 50 to 59 and 60 to 69 cohorts, with a cross-validated accuracy of 0.52, but a much clearer split between the 50 to 59 and 70-or-older groups, reaching an accuracy of 0.70 and an area under the ROC curve of roughly 0.73. The proteins driving that separation were telling. Ovaries from the younger postmenopausal women were enriched for proteostasis and chromatin machinery, including the folding chaperone FKBP4, Prefoldin subunit 6, the non-histone chromosomal protein HMGN1, and Histone deacetylase 3, consistent with a proteome still biased toward gene regulation and cellular maintenance. Ovaries from the oldest women instead shifted toward lipid metabolism and innate immunity, with rising Apolipoprotein A-II, Phospholipid transfer protein, 11-beta-hydroxysteroid dehydrogenase type 2, the immune effector PARP9, Complement Factor H-related protein 4, and the antiviral receptor RIG-I. Pathway enrichment mirrored the split: RNA processing and mRNA splicing dominated the younger signature, while the oldest ovaries were enriched for plasma lipoprotein assembly, extracellular matrix organization and degradation, and neutrophil degranulation.

Among the multivariate hits, one secreted signaling molecule stood out. WNT4, a ligand with established roles in ovarian development and stromal organization, increased progressively with age across all three cohorts, and its abundance was positively associated with continuous age in the proteomic dataset, with an R-squared of 0.42 and a p-value of 0.0002. Immunohistochemistry on tissue sections from 10 participants confirmed the finding, showing stronger WNT4 staining in the stroma and blood vessels of older ovaries. Its rise suggests that the aging postmenopausal ovary retains not merely remodeling features but active signaling capacity, potentially broadcasting molecular instructions to neighboring tissues long after reproduction has ended.

To identify proteins that changed continuously rather than between discrete age groups, the researchers performed covariate-adjusted regression, modeling age as a continuous variable while controlling for body mass index and self-reported race and ethnicity. Linear models outperformed quadratic alternatives across nearly all proteins by Akaike and Bayesian information criteria, indicating that most proteomic changes accumulated steadily rather than accelerating late. Two proteins reached genome-wide-style significance: Fatty acid-binding protein 5, which declined by roughly 29 percent per decade, and Heparan-sulfate 6-O-sulfotransferase 2, which climbed by roughly 58 percent per decade. Gene set enrichment analysis on the age coefficients revealed significant downregulation of keratinization and Hedgehog signaling, and significant upregulation of inflammatory response and matrisome-related pathways. Matrisome subcategory analysis showed collagens declining while secreted factors rose, and targeted regression within the matrisome highlighted fibromodulin, a small leucine-rich proteoglycan that organizes collagen fibrils and restrains TGF-beta signaling. Fibromodulin fell with age, with an R-squared of 0.46 in the proteomic data, and immunohistochemistry on 8 participants confirmed reduced staining in both cortical and medullary regions of older ovaries, correlating with age at an R-squared of 0.62. Its loss, previously linked to disordered matrix assembly and fibrotic scarring in other tissues, suggests a shift from organized matrix toward a crosslinked, remodeling-prone stroma, a pattern accentuated by the concurrent rise of transglutaminase-2, which promotes extracellular crosslinking during wound repair and fibrosis.

The inflammatory and matrix signatures raised a provocative question: does healthy ovarian aging converge with cellular senescence, the stress-induced state in which cells secrete inflammatory and matrix-degrading factors collectively known as the senescence-associated secretory phenotype? Comparisons with independent ovarian senescence datasets, including proteomic and transcriptomic profiles of doxorubicin-treated postmenopausal ovarian explants and spatial transcriptomic data from p16-positive cortical and medullary regions, revealed partial but consistent overlap. Serum amyloid P-component, vitronectin, transglutaminase-2, and nicotinamide phosphoribosyltransferase rose concordantly in both native aging and senescence models, while the matrix proteoglycan Versican fell. The authors caution that acute DNA damage-induced senescence does not fully recapitulate chronological aging, but the convergence across datasets supports the existence of shared senescence-adjacent remodeling programs in the postmenopausal ovary, marked by extracellular matrix turnover, metabolic stress, and innate immune signaling.

Perhaps the most striking finding emerged when the team compared their age-associated proteins with a literature-curated panel of 35 ovarian cancer-associated proteins, of which 21 were quantified in the dataset. Many of these proteins, including transglutaminase-2, Clusterin, the CA125 antigen MUC16, the HE4 antigen WFDC2, and Matrix metalloproteinase-2, exhibited their highest mean abundance in the 60 to 69 cohort, the decade surrounding the median age at ovarian cancer diagnosis of roughly 63 years. A participant-level score calculated as the mean log2 fold abundance across the 21 detected proteins was highest in the 60 to 69 group, at 0.19 versus -0.19 for the youngest cohort, and a planned permutation test comparing that group with all remaining participants yielded a p-value of 0.049. Because every tissue examined was histologically normal, the enrichment cannot be read as evidence of malignancy. Instead, it suggests that healthy ovarian tissue in this age range shares molecular features with ovarian cancer-associated remodeling, a permissive niche concept that adds nuance to debates over prophylactic ovary removal. Risk-reducing salpingo-oophorectomy clearly lowers ovarian cancer incidence in high-risk BRCA1 and BRCA2 carriers, but early or indiscriminate removal carries long-term risks of cardiovascular disease, osteoporosis, and cognitive decline, especially near the natural age of menopause.

To distill these convergent findings into a usable framework, the researchers integrated four complementary analytical approaches, pairwise differential abundance, multivariate variable importance in projection, continuous-age regression, and matrisome-focused enrichment, into a 26-protein consensus signature they named the Buck Postmenopausal Ovary Molecular Signature, or BuckPOMS. Proteins declining with age included lipid metabolism and transport factors such as Microsomal triglyceride transfer protein and Fatty acid-binding protein 5, matrix organizers such as Fibromodulin, and cellular differentiation markers, while rising proteins included extracellular remodelers and secreted signaling molecules such as ATS16, WNT4, Heparan-sulfate 6-O-sulfotransferase 2, Norrin, and the acute-phase protein Serum amyloid P-component. When the 26 proteins were standardized, directionally weighted, and combined into a participant-level score, the BuckPOMS score rose strongly with chronological age, with a Pearson correlation of 0.837 and a Spearman correlation of 0.875, and differed significantly across age cohorts with an effect size explaining 58 percent of the variance. The signature captures a coordinated transition from a proteome enriched in structural maintenance, RNA processing, and proteostasis toward one dominated by innate immunity, extracellular matrix remodeling, and altered intercellular signaling. The authors note limitations, including the absence of data on age at natural menopause, years since menopause, and menopausal hormone therapy history, all of which may shape postmenopausal remodeling. Still, the message is unambiguous: the postmenopausal ovary is not a static fibrotic remnant but a biologically dynamic organ, one that may continue to participate in local and even inter-organ signaling during aging, and whose ongoing molecular remodeling could hold keys to understanding chronic disease risk in the decades after reproduction ends.

Subject of Research: Cell Biology

Article Title: The Human Ovary Exhibits Dynamic Molecular Remodeling in the Decades Post‐Menopause

Article References: Watson, M. A., Soygur, B., King, C. D., Devrukhkar, P. R., Melov, S., Shanes, E. D., Pavone, M. E. G., Duncan, F. E., & Schilling, B. (2026). The Human Ovary Exhibits Dynamic Molecular Remodeling in the Decades Post‐Menopause. Aging Cell, 25(10), Article e70714. https://doi.org/10.1111/acel.70714

Image Credits: AI Generated

DOI: 10.1111/acel.70714

Keywords: aging, menopause, proteomics, ovary, ovarian cancer, senescence, extracellular matrix, inflammaging, Human, Exhibits, Dynamic, Molecular