Ovaries age faster than most organs, and chronological age only roughly captures this process. Epigenetic clocks built on DNA methylation allow researchers to estimate the biological age of a tissue. Earlier work by our group showed that environmental chemicals accumulate in follicular fluid and are linked to lower oocyte counts. But the mechanism behind this effect remained unclear. This study sits at the intersection of two fields: epigenetics of reproductive tissues and exposome science, where individual pollutants have been studied but not their combined effect on the ovarian epigenome.
The study enrolled 76 patients at the Emory reproductive center in Atlanta who were undergoing ovarian stimulation for IVF or oocyte cryopreservation. Follicular fluid was collected during retrieval, the cell pellet was isolated, and DNA was extracted. Methylation was assessed on Illumina EPIC v1 and v2 arrays, then the data were harmonized and normalized. Chemical composition had been determined earlier using untargeted high-resolution mass spectrometry (gas and liquid chromatography). For each sample, the researchers calculated the GrimAge, Horvath PanTissue, and Granulosa Cell epigenetic clocks. Aging acceleration was computed as the residuals from regressing epigenetic age on chronological age. Associations with outcomes were assessed with quasi-Poisson and linear models adjusted for age and surrogate variables reflecting technical and biological sources of variation.
The mean patient age was 36 years, and the mean number of retrieved oocytes was 16. The GrimAge epigenetic clock correlated with chronological age (r=0.31), and the Granulosa Cell clock correlated more strongly (r=0.51), while PanTissue showed no correlation at all (r=0.02) and was excluded. GrimAge aging acceleration was significantly associated with fewer oocytes: one standard deviation of acceleration corresponded to 16% fewer oocytes (95% CI: 5.8–25%, p=0.004). Reductions were also seen in antral follicle count, AMH level, and the ovarian sensitivity index. All four chemical mixture indices (based on GC, LC-C18, LC-HILIC, and targeted compounds) were positively associated with accelerated epigenetic aging (p<0.05). For example, a one-standard-deviation increase in the LC-C18 mixture produced a 0.31 SD gain in acceleration (95% CI: 0.095–0.53, p=0.0063). Among individual chemicals, benzo[a]pyrene showed a significant association (0.25 SD, p=0.035). Pathway analysis revealed enrichment of genes related to olfactory transduction and glycosphingolipid biosynthesis, as well as follicle-specific developmental pathways: primordial follicle growth initiation, cumulus cell differentiation, and FSH signaling.
This work is the first to combine the untargeted chemical exposome and the epigenome in human follicular fluid, showing that pollutant mixtures act on the epigenetic mechanisms of ovarian aging. Unlike neighboring studies that examined either individual chemicals or epigenetic clocks in blood, this one used a comprehensive approach on the target tissue. That makes it possible to see the cumulative effect of mixtures, which slips through the cracks when single substances are analyzed. The results support the hypothesis that accelerated epigenetic aging may be the missing link between exposure and reduced ovarian reserve.
The field could move toward creating reproductive-specific epigenetic clocks, trained on ovarian cells and aimed at predicting fertility rather than simply chronological age. Larger cohorts with a longitudinal design will be needed to separate cause and effect from mere association. Expanding the multi-omics approach also looks promising: adding transcriptomics, proteomics, and metabolomics of follicular fluid to build a complete picture of the response to chemical load.
The findings could influence how reproductive risks from chemical exposure are assessed, the development of biomarkers for ovarian reserve, and toxicological testing of mixtures rather than individual substances.
The immediate steps include validating the associations found in independent cohorts, identifying untargeted chemical features with high confidence, and developing epigenetic clocks specific to granulosa cells and aimed at IVF outcomes.
The work touches on a fundamental problem in reproductive biology: why the same chronological age yields different ovarian reserve and different responses to stimulation. Epigenetic clocks specific to ovarian tissue could become a tool for predicting an individual's trajectory of follicular aging. What remains an open question is which components of chemical mixtures contribute most to epigenetic changes, and whether those changes are reversible when exposure is reduced.
Follicular fluid, usually discarded after oocyte retrieval, turns out to be a valuable archive: it lets you read both the body's chemical history and the ovary's epigenetic age at the same time. The benzo[a]pyrene found in this fluid forms during incomplete combustion of organic matter and enters the body through fried food, tobacco smoke, and exhaust fumes.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.