The aims of this project are to conduct genomewide DNA methylation analyses of peripheral blood leukocytes (PBL) in two groups of children; those born late preterm (32-36 weeks gestation) and those born at term (greater than 37 weeks) with low birthweight who are participants of the Avon Longitudinal Study of Parents and Children (ALSPAC). The associations of genomewide and gene specific methylation patterns at birth (cord blood) with lung function outcomes at 8 years and 15-17 years will be investigated initially. For differentially methylated regions (DMRs) identified in this discovery phase, methylation analyses will be repeated at 7 years and 15-17 years in the same subjects to examine longitudinal changes. Additionally, associations of prenatal exposure to tobacco smoke and maternal nutrition with methylation of cord blood DNA will be examined.

The proposed study will take advantage of the large scale epigenomic resource, ARIES (Accessible Resource for Integrated Epigenomic Studies) established by a grant from the BBSRC (PIs: Davey Smith & Relton). This will provide comparable genomewide methylation data of 1000 mother-child pairs in ALSPAC, from which term-born, normal birthweight controls will be selected (as well as using data from any eligible preterm and low birth weight children included in this random sample). Data generated by this project will be integrated with the ARIES project and will thereby create a unique and valuable epigenomic resource for studies of other outcomes of interest, such as neurocognitive development17, in preterm and low birthweight infants.

Specific hypotheses to be tested in this project are:

1. Differences in genome-wide and gene-specific DNA methylation are observed when comparing cord PBL from preterm infants and term infants with evidence of intrauterine growth restriction compared with term, appropriately grown infants and these DMRs are associated with abnormal lung function in later childhood.

2. DMRs identified at birth in association with abnormalities of long-term lung function are modified by postnatal exposures including growth rate, diet and environmental tobacco smoke, and these modifications are associated with different lung function trajectories.

3. Maternal smoking and nutrition during pregnancy; total energy and protein intake and estimated micronutrient intake of substances involved in 1-carbon metabolism, including synthetic folate supplements, are associated with DNA methylation of genes linked with abnormal long term lung function.

Additionally, the availability of genomewide SNP data in the ALSPAC population will enable exploration of whether methylation patterns that are associated with particular exposures and disease outcomes are likely to be causal. A two-step epigenetic Mendelian randomization approach will be used, based on Mendelian randomisation18 and genetical epigenomics19. The establishment of causal inference for associations will be critical to the translation of findings from this study to future clinical interventions. However, even in the absence of evidence of causality, epigenetic marks may serve as useful biomarkers of exposure in both aetiological and intervention studies.