Persistent DNA Methylation Changes across the First Year of Life and Prenatal NO2 Exposure in a Canadian Prospective Birth Study.

BACKGROUND: Evidence suggests that prenatal air pollution exposure alters DNA methylation (DNAm), which could go on to affect long-term health. It remains unclear whether DNAm alterations present at birth persist through early life. Identifying persistent DNAm changes would provide greater insight into the molecular mechanisms contributing to the association of prenatal air pollution exposure with atopic diseases. OBJECTIVES: This study investigated DNAm differences associated with prenatal nitrogen dioxide (${\mathrm{NO}}_2$) exposure (a surrogate measure of traffic-related air pollution) at birth and 1 y of age and examined their role in atopic disease. We focused on regions showing persistent DNAm differences from birth to 1 y of age and regions uniquely associated with postnatal ${\mathrm{NO}}_2$ exposure. METHODS: Microarrays measured DNAm at birth and at 1 y of age for an atopy-enriched subset of Canadian Health Infant Longitudinal Development (CHILD) study participants. Individual and regional DNAm differences associated with prenatal ${\mathrm{NO}}_2$ ($n=128$) were identified, and their persistence at age 1 y were investigated using linear mixed effects models ($n=124$). Postnatal-specific DNAm differences ($n=125$) were isolated, and their association with ${\mathrm{NO}}_2$ in the first year of life was examined. Causal mediation investigated whether DNAm differences mediated associations between ${\mathrm{NO}}_2$ and age 1 y atopy or wheeze. Analyses were repeated using biological sex-stratified data. RESULTS: At birth ($n=128$), 18 regions of DNAm were associated with ${\mathrm{NO}}_2$, with several annotated to HOX genes. Some of these regions were specifically identified in males ($n=73$), but not females ($n=55$). The effect of prenatal ${\mathrm{NO}}_2$ across CpGs within altered regions persisted at 1 y of age. No significant mediation effects were identified. Sex-stratified analyses identified postnatal-specific DNAm alterations. DISCUSSION: Regional cord blood DNAm differences associated with prenatal ${\mathrm{NO}}_2$ persisted through at least the first year of life in CHILD participants. Some differences may represent sex-specific alterations, but replication in larger cohorts is needed. The early postnatal period remained a sensitive window to DNAm perturbations. https://doi.org/10.1289/EHP13034.