Natural Variation in Transplacental Transfer Efficiency Reveals Distinct Transcriptional Network Architectures of PFAS Effects on Birth Weight and Gestational Age

Distinguishing effects of environmental exposures mediated through maternal pathophysiology from those operating through direct fetal mechanisms remains a central challenge in environmental epidemiology, with implications for understanding effect heterogeneity on related, but distinct, health and clinical outcomes. Per- and polyfluoroalkyl substances (PFAS) are a broad class of ubiquitous environmental contaminants with suspected, but inconsistently observed, associations with adverse perinatal outcomes such as gestational age and birth weight. Variation in transplacental transfer efficiency (TPTE) of PFAS presents an opportunity to decompose maternal and fetal effects through effective dose–response gradients: Using isoform-resolved placental transcriptomics from n = 124 term deliveries with eight PFAS measured in maternal and cord blood, validated in a patient-derived placental explant experiment (n = 18), we find that PFAS with relatively higher transplacental transfer efficiency (TPTE) are associated with increasingly numerous and tightly coordinated transcriptional mediators for both birth weight and gestational age. Importantly, we find that these effects are mediated through coexpression network hubs rather than the “top hits” that are commonly targeted in differential expression studies. Moreover, the TPTE gradient revealed distinct network architectures for the two related perinatal outcomes: for birth weight, but not gestational age, higher TPTE was associated with increasing centrality and maternal-fetal compartmentalization of transcription signals, providing molecular support consistent with distinct architectures underlying PFAS effects on fetal growth restriction versus more diffuse, maternal effects governing parturition timing. Consistent mediators across outcomes (including TM7SF3, NBEAL1, and ANXA5) implicate ER stress, cholesterol homeostasis, and trophoblast membrane integrity as convergent therapeutic targets in PFAS-associated perinatal outcomes. These patterns are detectable only at the isoform level. This framework identifies experimentally tractable intervention candidates and offers a novel generalizable approach for resolving mechanistic heterogeneity from nonexperimental data in environmental epidemiology.