Exploratory Analysis of DNA Methylation and Socioeconomic Status of Children in the Greater Vancouver Area

Sarah J. Goodman^1,2, W. Thomas Boyce^3,4,5, Anat Zaidman-Zait⁴, Stephanie Lam⁴, Julia L. MacIsaac¹, Meaghan J. Jones¹, Sarah M. Mah¹, Michael S. Kobor^1,2

1. Centre for Molecular Medicine and Therapeutics, Child and Family Research Institute, Vancouver, British Columbia, Canada; 2. Medical Genetics, University of British Columbia, Vancouver, British Columbia, Canada; 3. School of Population and Public Health, University of British Columbia, Vancouver, British Columbia, Canada; 4. Human Early Learning Partnership, University of British Columbia, Vancouver, British Columbia, Canada; 5. Pediatrics, UCSF School of Medicine, San Francisco, California, USA.

Rationale: Socioeconomic status (SES) is the most potent yet enigmatic environmental determinant of human health. Childhood experiences of SES are known to alter development and well-being and cause health disparities later in life, suggesting that they are somehow ‘biologically embedded’. Gene Expression Collaborative for Kids Only (GECKO) was established to study the mechanisms of biological embedding in socioeconomically diverse children. An important part of the study is investigating how DNA methylation may interact with both the environment and DNA variants, thereby stably imprinting genetic and environmental information into the epigenome. We hypothesize that socially driven developmental trajectories in children are mediated by the interactions of environment, epigenetics and genetic variants, which ultimately result in phenotypic differences and health disparities later in life.

Methods: Using the Illumina Human Methylation 450K array platform, we have examined genome-wide DNA methylation of buccal epithelial cells (BECs) in 400 8-10 year old individuals from the GECKO cohort. For a subset of participants, whole blood and newborn dried blood spot (DBS) samples were obtained to compare BEC methylation patterns to those present in peripheral blood mononuclear cells (PBMCs) and neonatal blood from each child. Genetic variants were evaluated by running saliva DNA from each participant on the Illumina PsychArray BeadChip.

Results and implications: We interrogated epigenomic correlates of childhood experiences and exposures to identify a set of loci showing experience-dependent epigenetic modifications. A differential analysis of methylation by SES, as measured by family income, parent occupation and education, produced a list of candidate loci which may be involved in biological embedding. Replication of this analysis on the DBSs and PBMCs reinforced the tissue-specific nature of DNA methylation. Unsupervised clustering of the paired DBS and PBMC samples showed that the methylation of two unrelated individuals in a single tissue is more alike than the methylation of two tissue types from one individual. Candidate methylation loci will be interrogated for DNA variants to identify potential risk alleles which are more prone to methylation disturbances by SES. Understanding the molecular and epigenetic mechanisms by which early adversity and SES lead to disturbances in health is a fundamental first step in elucidating the origins of health disparities.