Sex-dependent relationships between PFAS and placental transcriptomics identified by weighted gene co-expression analysis

Abstract

Background: Per- and polyfluoroalkyl substances (PFAS) are environmental toxicants associated with adverse neonatal outcomes. The exact mechanisms by which PFAS impairs neonatal health are undefined, but the placenta is a likely target.

Objective: We applied a systems biology approach to identify placental RNA co-expression modules (gene sets) associated with PFAS exposure and birth weight.

Methods: Placental tissue samples (n = 147) from the GLOWING study underwent RNA-sequencing, and PFAS concentrations were quantified using liquid chromatography-tandem mass spectrometry. We constructed a weighted gene co-expression network using Spearman correlations across 15,028 transcripts, identifying 20 gene modules. Linear regression models were used to examine associations between PFAS and module eigengenes, adjusting for potential confounders. Effect modification by fetal sex was also tested.

Results: One module showed a negative association with perfluorononanoic acid (PFNA; β = -0.012, q = 0.009). This association was sex-specific, with the sexes exhibiting varied PFAS associations but similar directional effects. Genes within the PFNA-associated module were involved in histone modification (q ≤ 0.05) and were enriched for targets of the Vitamin D Receptor (VDR), a transcription factor previously linked to PFAS.