A Phenome-Wide Association and Epigenetic Mendelian Randomization Study of Air Pollution and Health Outcomes

Abstract The health effects of ambient air pollution extend beyond well-established outcomes, yet their broader phenotypic impact and underlying biological mechanisms remain incompletely understood. In particular, the role of DNA methylation (DNAm) in these associations has not been systematically investigated. We conducted a phenome-wide association study (PheWAS) in the UK Biobank to examine associations between major air pollutants, including particulate matter with aerodynamic diameters of ≤2.5 μm (PM2.5) and ≤10 μm (PM10), nitrogen oxides (NOx) and nitrogen dioxide (NO2), and a wide range of clinical outcomes. Next, to explore potential biological mechanisms, we applied epigenetic Mendelian randomization (MR) to assess whether DNAm at pollutant-associated CpG sites is associated with disease risk, using data from the UK Biobank and the FinnGen study. Colocalization analysis was used to prioritize shared genetic signals. In the PheWAS analysis, exposure to PM2.5, PM10, NOx, and NO2 was associated with 90 clinical outcomes (81 positive associations and 9 inverse associations) across diverse systems, including circulatory, metabolic, digestive, genitourinary, neurological, respiratory, musculoskeletal, hematologic, and neoplastic conditions. Epigenetic MR analysis identified that higher genetically determined DNAm levels at 16 cytosine-phosphate-guanine (CpG) sites related to NO2 were associated with risks of metabolic, genitourinary, and neoplastic conditions. Altered methylation at cg06240200 (odds ratio [OR], 1.39; 95% confidence interval [CI], 1.22–1.59), cg09367967 (OR, 1.21; 95% CI, 1.11–1.33) and cg20959189 (OR, 1.43; 95% CI, 1.30–1.57), and cg01543603 (OR, 0.51; 95% CI, 0.42–0.61) and cg25670076 (OR, 0.67; 95% CI, 0.55–0.82) showed positive associations with type 2 diabetes and prostate hyperplasia and negative associations with nonepithelial skin cancer, respectively. Addtionly, these signals were prioritized with high colocalization evidence. The regulation of their mapped genes (CCDC12, ZBTB38, ANKRD11, BACH2) could contribute to the observed biological associations between methylation and disease risk. Our findings suggest multisystem health burdens associated with ambient air pollution. Biological pathways implicated in DNAm alterations at key genetic loci related to NO2 may act as potential mechanisms that could modify specific disease risks. These findings may therefore yield insights into how reducing air pollution could provide broad population health benefits and into the potential pathogenic role of these candidate epigenetic loci.