Low-to-moderate arsenic exposure and DNA methylation: A review of life-course evidence and mechanistic insights

Abstract

Arsenic exposure affects an estimated 94 to 220 million people worldwide, yet most research has centered on high-dose endemic settings, leaving the epigenetic effects of environmentally relevant exposures undercharacterized. This narrative review synthesizes evidence from 68 studies on DNA methylation alterations associated with low-to-moderate arsenic exposure, integrating human epidemiological, animal, and in vitro data. Gene-specific findings were stage-patterned across the life course, whereas global and repetitive-element methylation findings were heterogeneous. Prenatal exposure perturbs mitochondrial biogenesis, growth and endocrine regulation, vascular development and neurodevelopment. Childhood evidence rests on two studies, implicating mitochondrial, growth-regulatory and mismatch repair genes. In adults, oxidative stress response, cancer and cardiovascular pathways, and metabolic dysfunction are most often affected. Experimental models corroborate several of these processes and show that methylation alterations can persist after exposure ceases. Although CpG-site replication across cohorts remains limited, pathway convergence across independent populations supports biological plausibility. Notably, methylation alterations are detectable at or below the WHO guideline of 10 μg/L, particularly during sensitive developmental windows, although whether these changes lead to disease has not been established. Beyond the nuclear genome, mitochondria are a plausible target for arsenic-induced epigenetic disruption through their roles in one-carbon metabolism and redox regulation. This review critically evaluates mitochondrial DNA methylation findings against five methodological quality criteria and proposes a framework for future investigations, addressing a field where biological rationale has outpaced technical validation. Research priorities include longitudinal designs that link methylation biomarkers to clinical outcomes and intervention trials that assess reversibility to inform regulation.