Perfluorooctane sulfonate (PFOS) in follicular fluid and human granulosa cell dysfunction: a physiologically based toxicokinetic model translation of low-level, long-term in vitro exposure data

STUDY QUESTION Are bioactive human-equivalent doses (HEDs) of perfluorooctane sulfonate (PFOS), derived from long-term low-level in vitro exposure of human granulosa cells comparable to HEDs inferred from follicular fluid PFOS concentrations in women undergoing ART and in occupationally exposed women? SUMMARY ANSWER The bioactive HEDs overlapped with and, in some cases, were lower than the median HEDs inferred from follicular fluid PFOS concentrations. WHAT IS KNOWN ALREADY PFOS exposure is a growing public health concern, with evidence suggesting adverse female reproductive effects. However, the relevance of current human exposure levels to granulosa cell function remains unclear. STUDY DESIGN, SIZE, DURATION Four independent vials of human granulosa cells (HGrC1 cells) were thawed and expanded into separate flasks (biological replicates). Cells were allocated to four experimental groups and exposed to PFOS (0.01, 0.1, or 1 µM) or vehicle control (0.05% DMSO) for up to 12 weeks, with re-dosing at each passage. Different apical endpoints, along with transcriptomic changes, were evaluated at designated time points. Clinical relevance of PFOS risk to human granulosa cells was assessed by integrating experimental data with physiologically based toxicokinetic (PBTK) modeling. PARTICIPANTS/MATERIALS, SETTING, METHODS Viability of HGrC1 cells was assessed using the Alamar Blue assay. Estradiol and progesterone secretion were quantified by enzyme-linked immunosorbent assay. Flow cytometry was used to determine the proportions of live, apoptotic and necrotic cells, as well as cell cycle distribution. Global mRNA expression was assessed by DNA nanoball sequencing technology (DNBSEQ), whereas pathway-level molecular functions were derived using bioinformatic tools. Benchmark concentrations (BMCs) were calculated from key endpoints with concentration-dependent responses and used to estimate HEDs via PBTK modeling. These HEDs were compared with HEDs inferred from follicular fluid PFOS levels reported in the literature to derive bioactivity exposure ratios (BERs) and assess relevance to human exposure. MAIN RESULTS AND THE ROLE OF CHANCE In HGrC1 cells, long-term PFOS exposure altered steroidogenesis, apoptosis/necrosis, cell cycle distribution (P < 0.05), and gene expression (at least 2-fold change, Q-value ≤ 0.05). Median transcriptomic HEDs were 18.1 (95% CI: 1.1–35.1) and 17.5 ng/kg bw/day (95% CI: 8–27.1) for 6- and 12-week exposures, respectively, with corresponding 5th percentile HEDs of 3.7 ng/kg bw/day (95% CI: 0.4–9.3) and 1.4 ng/kg bw/day (95% CI: 0.5–3.5). Pathway-level HEDs ranged from 2.8 to 24.1 ng/kg bw/day, with eicosanoid synthesis showing the greatest sensitivity. HEDs for apical endpoints ranged from 0.4 to 203 ng/kg bw/day, with the sub-G1 cell cycle phase being most sensitive. HEDs derived from the 5th percentile transcriptomic data, eicosanoid metabolism, and the sub-G1 phase yielded BERs below 1, indicating that PFOS levels measured in follicular fluid of ART patients may be sufficient to induce these biological effects. For occupational exposure, BERs derived from all endpoints were below 1. A subset of nine granulosa-cell genes, including CYP1B1 and TIPARP (aryl hydrocarbon receptor signaling), showed HEDs that were below the follicular-fluid-inferred HED, highlighting potential high-priority targets and candidate biomarkers. LARGE SCALE DATA Raw and processed RNA-sequencing data are deposited in NCBI Gene Expression Omnibus (GEO) under accession number GSE315651. LIMITATIONS, REASONS FOR CAUTION The estimated exposure values were based on predictions from a PBTK model rather than empirical human exposure data. Also, differences in protein concentrations in vitro and in vivo may affect free PFOS levels and bioactivity estimates. We addressed this with additional adjustments for PFOS-albumin binding. Finally, follicular fluid PFOS concentrations in occupational settings were approximated from serum concentrations using blood-to-follicular fluid transfer efficiency (BFTE) values. WIDER IMPLICATIONS OF THE FINDINGS Our findings suggest that PFOS concentrations in follicular fluid from women undergoing ART and those who have been occupationally exposed may be sufficient to perturb granulosa cell mRNA expression and key pathways, including eicosanoid, interleukin, and GPCR signaling. The identified genes may serve as candidate biomarkers linking PFOS exposure to clinical outcomes in ART settings. Overall, this study provides a framework for interpreting PFOS reproductive toxicity and refining health-protective exposure thresholds. STUDY FUNDING/COMPETING INTEREST(S) This research was supported by the Ministry of Science, Technological Development and Innovation of the Republic of Serbia (Faculty of Sciences, Novi Sad: Grants No. 451-03-137/2025-03/200125 & 451-03-136/2025-03/200125), the Institute of Physics, Belgrade, National Institute of the Republic of Serbia, and the Science Fund of the Republic of Serbia, Grant No. 7010, ‘Integration of Biological Responses and PBTK Modeling in Chemical Toxicity Assessment: А Case Study of Perfluorooctanoic Acid (PFOA)—ToxIN’. The authors declare no conflicts of interest.