Pregnancy And Childhood Epigenetics (PACE)
The first PACE project was a meta-analysis of maternal smoking in relation to methylation in newborns and children using the Illumina 450K platform. Cohort-specific results were generated by each group and meta-analyzed by Bonnie Joubert, Ph.D., who began this work as a research fellow in the NIEHS Epidemiology Branch. She now serves as a program director in the NIEHS Extramural Research and Training Division’s Population Health Branch.
We identified more than 6,000 differentially methylated CpG sites in newborns in relation to maternal smoking during pregnancy. Out of this total, nearly 3,000 CpGs, which corresponded to 2,017 genes, were not previously associated with smoking and methylation in either newborns or adults. Some of these genes have been implicated in genetic studies of orofacial clefts, asthma, and lung and colorectal cancer. Notably, many of these signals persist into childhood. The following cohorts participated in this first PACE meta-analysis (Joubert et al., 2016):
- The Avon Longitudinal Study of Parents and Children (ALSPAC)
- The Center for Health Assessment of Mothers and Children of Salinas (CHAMACOS)
- The Children’s Health Study (CHS)
- The GECKO Drenthe Cohort
- The Generation R Study
- Isle of Wight Birth Cohort Study
- The Mechanisms of the Development of Allergy (MeDALL) Consortium, which includes four component cohorts:
- The Norwegian Mother and Child Cohort Study (MoBa)
- The Norway Facial Clefts Study (NFCS)
- The Newborn Epigenetics Study (NEST)
- Project Viva
- Genes-environments and Admixture in Latino Americans (GALA II)
- The Study to Explore Early Development (SEED)
After the smoking project began, a number of additional cohorts joined PACE and are listed in a recent publication.
The next PACE publication examined traffic related air pollution, using NO2 exposure estimated at the residential address, in relation to newborn methylation. Novel associations were identified in three genes related to mitochondrial function. One of these methylation signals could still be seen at age eight years. More recently, PACE investigators reported methylation differences at birth and in childhood associated with prenatal exposure to particulate matter.
A subsequent PACE publication examined maternal body mass index at the start of pregnancy in relation to DNA methylation in 9,340 newborns from 19 cohorts. After adjustment for cell type there were 86 sites (CpGs) significantly differentially methylated in newborns. In adolescents (four additional cohorts, 1,817 children), at 72 of these 86 sites there was evidence for persistence of these signals related to maternal body mass index at the start of pregnancy.
PACE has also examined maternal alcohol intake in relation to newborn methylation in 3,075 mother-child pairs, including 1,147 mothers who consumed alcohol both before and throughout pregnancy and 1,928 mothers who consumed alcohol before pregnancy and during the first trimester but not during the second and/or third trimester. The amount of alcohol intake during pregnancy was typically low. No CpGs were significantly differentially methylated in newborns in relation to exposure.
To follow-up our PACE smoking findings, we collaborated with the Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE) consortium of adult cohorts to compare the methylation signals in newborns from having a mother who smoked during pregnancy with those seen in adults from their own smoking. We identified many signals shared between both exposures, but also a substantial number of genes differentially methylated only in relation to the newborn exposure.
PACE projects also examine methylation in offspring in relation to a range of childhood health outcomes. A PACE meta-analysis identified variation in DNA methylation in blood related to childhood asthma, prospectively in newborns and cross-sectionally in children; these signals are potential biomarkers of asthma development and its biologic effects that may shed light on disease mechanisms. A recent PACE publication reported widespread differential methylation in newborns related to their birthweight; some signals persisted into adolescence, but not adulthood. We have also published on methylation in relation to gestational hypertension and preeclampsia as well as gestational diabetes gestational diabetes. A project examining gestational age in relation to newborn methylation identified several thousand signals. Efforts are underway examining other child health outcomes including blood pressure, body mass index, and ADHD. PACE scientists are also working together to examine methodological issues involved in the analysis of the methylation data.
So far, PACE analyses have been self-funded by each cohort with consortium investigators donating their time. London, with funding from the NIEHS intramural program, has furnished the administrative support for monthly conference calls and 3 in-person meetings that have been held in Research Triangle Park, North Carolina and in Rotterdam, the Netherlands.