Trimester-Specific Prenatal Air Pollution Effects on the Newborn Epigenome

Abstract

Trimester-Specific Prenatal Air Pollution Effects on the Newborn EpigenomeAbstract Number:2227 Carrie Breton*, Yu-Chen Su, Jie Ren, Kim Siegmund, Frank Gilliland Carrie Breton* University of Southern California, United States, E-mail Address: [email protected] Search for more papers by this author , Yu-Chen Su University of Southern California, United States, E-mail Address: [email protected] Search for more papers by this author , Jie Ren University of Southern California, United States, E-mail Address: [email protected] Search for more papers by this author , Kim Siegmund University of Southern California, United States, E-mail Address: [email protected] Search for more papers by this author , and Frank Gilliland University of Southern California, United States, E-mail Address: [email protected] Search for more papers by this author AbstractA limited body of evidence suggests that air pollution is associated with altered DNA methylation patterns in peripheral blood of adults. We aimed to investigate whether prenatal air pollution exposure might also influence DNA methylation patterns in newborn blood, and if these associations were trimester specific. Trimester-specific exposures to EC, PM2.5, PM10, N02, and O3 were calculated using data from the Children’s Health Study (CHS) air quality monitoring data as well as the US EPA air Quality System for 580 CHS subjects based on reported birth address. DNA methylation in LINE1 and ALuYb8 repetitive elements was measured in newborn dried bloodspots using Pyrosequencing and gene-specific DNA methylation loci were measured using the Illumina HumanMethylation450 array (n=280). HM450 data were subject to a custom pre- processing workflow to reduce background noise, normalize data, and filter unwanted probes. Generalized linear regression models were used to analyze associations between air pollutants and DNA methylation levels adjusting for covariates, in which a quasi-binomial link was used for the HM450 array data. FDR was used to adjust for multiple testing. First trimester PM2.5, PM10, NO2, but not EC were consistently and significantly associated with hypomethylation of LINE1 but not AluYb8. Exposure in other trimesters was not associated with repetitive element methylation. We identified 1 locus marginally associated with second trimester PM10 exposure and 6 CpG loci associated with third trimester exposure to PM10 and none with first trimester exposure after FDR adjustment. In conclusion, first trimester exposure to air pollutants is associated with decreased LINE1 methylation in newborn bloodspots whereas PM10 exposure in later trimesters was associated with gene-specific methylation. The health effects of these effects are currently unknown but warrant further investigation.