Mid-pregnancy PM2.5 exposure affects sex-specific growth trajectories via ARRDC3 methylation

Prenatal particulate matter <2.5 μm (PM2.5) is associated with adverse birth growth. However, the longitudinal growth impacts have been little studied, and no mechanistic relationships have been described. We investigated the association between prenatal PM2.5 exposure and growth trajectories, and the possible role of epigenetics. We enrolled 1313 neonates with PM2.5 data measured by ordinary kriging from the COhort for Childhood Origin of Asthma and allergic diseases, followed up at 1, 3, and 5 years to evaluate growth. Differential DNA methylation and pyrosequencing of cord blood leukocytes was evaluated according to the prenatal PM2.5 levels and birth weight (BW). PM2.5 exposure during the second trimester (T2) caused the lowest BW in both sexes, further adjusted for indoor PM2.5 levels [female, aOR 1.39 (95% CI 1.05–1.83); male, aOR 1.36 (95% CI 1.04–1.79)]. Bayesian distributed lag models with indoor PM2.5 adjustments revealed a sensitive window for BW effects at 10–26 weeks gestation, but only in females. Latent class mixture models indicated that a persistently low weight-for-height percentile trajectory was more prevalent in the highest PM2.5 exposure quartile at T2 in females, compared to a persistently high trajectory (36.5% vs. 20.3%, P = 0.022). Also, in the females only, the high PM2.5 and low BW neonates showed significantly greater ARRDC3 methylation changes. ARRDC3 methylation was also higher only in females with low weight at 5 years of age. Higher fetal PM2.5 exposure during T2 may cause a decreased growth trajectory, especially in females, mediated by ARRDC3 hyper-methylation-associated energy metabolism.