BACKGROUND: We conducted an epigenome-wide association study (EWAS) of DNA methylation in placenta in relation to maternal tobacco smoking during pregnancy and examined whether smoking-induced changes lead to low birthweight. METHODS: DNA methylation in placenta was measured using the Illumina HumanMethylation450 BeadChip in 179 participants from the INfancia y Medio Ambiente (INMA) birth cohort. Methylation levels across 431 311 CpGs were tested for differential methylation between smokers and non-smokers in pregnancy. We took forward three top-ranking loci for further validation and replication by bisulfite pyrosequencing using data of 248 additional participants of the INMA cohort. We examined the association of methylation at smoking-associated loci with birthweight by applying a mediation analysis and a two-sample Mendelian randomization approach. RESULTS: Fifty CpGs were differentially methylated in placenta between smokers and non-smokers during pregnancy [false discovery rate (FDR) < 0.05]. We validated and replicated differential methylation at three top-ranking loci: cg27402634 located between LINC00086 and LEKR1, a gene previously related to birthweight in genome-wide association studies; cg20340720 (WBP1L); and cg25585967 and cg12294026 (TRIO). Dose-response relationships with maternal urine cotinine concentration during pregnancy were confirmed. Differential methylation at cg27402634 explained up to 36% of the lower birthweight in the offspring of smokers (Sobel P-value < 0.05). A two-sample Mendelian randomization analysis provided evidence that decreases in methylation levels at cg27402634 lead to decreases in birthweight. CONCLUSIONS: We identified novel loci differentially methylated in placenta in relation to maternal smoking during pregnancy. Adverse effects of maternal smoking on birthweight of the offspring may be mediated by alterations in the placental methylome.

This study was supported by grants from the Instituto de Salud Carlos III and Spanish Ministry of Health (Red INMA G03/176; CB06/02/0041; FIS 97/0588; 00/0021–2, PI061756; PS0901958; FIS-FEDER 03/1615, 04/1509, 04/1112, 04/1931, 05/1079, 05/1052, 06/1213, 07/0314, 09/02647, 11/0178, 11/01007, 11/02591, 11/02038, 13/1944, 13/2032, 14/0891 and 14/1687; Miguel Servet-FEDER CP15/00025; FIS-PI041436, FIS- PI081151, FIS-PI06/0867, FIS-PS09/00090, FIS-FEDER PI11/0610 and FIS-FEDER PI11/010007), FISS-PI042018, FISS-PI0902311, Fundació La Marató de TV3 (Project No. 090430), EC Contract No. QLK4-CT-2000–00263, Universidad de Oviedo, Conselleria de Sanitat Generalitat Valenciana, Generalitat de Catalunya-CIRIT 1999SGR 00241, the Department of Health of the Basque Government (2005111093 and 2009111069), the Provincial Government of Gipuzkoa (DFG06/004 and DFG08/001) and Fundación Roger Torné. EM is supported by a Miguel Servet Research Grant (CP14/00046) from the Instituto de Salud Carlos III, Spanish Ministry of Economy and Competitiveness, Madrid (Spain). We also acknowledge support of the Spanish Ministry of Economy and Competitiveness, Centro de Excelencia Severo Ochoa 2013–2017, SEV-2012–0208