Brain-Targeted Epigenome-Wide Association Study of Childhood Internalizing and Externalizing Behaviors: Evidence from...

Background Identifying biologically meaningful epigenetic signatures of childhood internalizing and externalizing behaviors has been challenging because conventional epigenome-wide association studies are limited by small effect sizes, genetic confounding, and uncertain tissue relevance. We therefore combined a brain-targeted epigenome-wide association study (BEWAS) with a monozygotic twin difference design to improve biological specificity while isolating environmentally driven DNA methylation patterns. Methods Buccal-derived DNA methylation was analyzed in 415 children from the Arizona Twin Project, including 102 monozygotic twin pairs. DNA methylation analyses were restricted to genes with elevated brain expression or established roles in neurotransmission. Linear mixed-effects models were conducted in the full sample, followed by within-pair MZ difference analyses. Nominally associated CpG sites underwent pathway enrichment analyses, and gene-level findings were compared with previous pediatric DNA methylation studies. Results Although no individual CpG sites survived epigenome-wide correction, nominally associated loci converged on robust neurodevelopmental pathways involved in neurogenesis, neuronal differentiation, synaptic organization, and neuronal projection development. Internalizing behaviors demonstrated pathway enrichment in both the population-based and MZ analyses, whereas externalizing enrichment emerged only in the MZ analyses despite the substantially smaller sample size. Gene-level overlap with previous pediatric methylation studies was greater in the MZ analyses, with recurrent associations observed for neurodevelopmental genes including PTPRN2 , SHANK2 , CACNA1C , and ADCY8 . Overall, convergence emerged primarily at the level of genes and biological pathways rather than individual CpG sites. Conclusions These findings suggest that childhood internalizing and externalizing behaviors are characterized by distributed epigenetic regulation across neurodevelopmental systems. More broadly, they demonstrate how biologically informed epigenomic restriction combined with genetically informative study designs can improve detection and interpretation of molecular signatures relevant to developmental psychopathology, providing a foundation for future biomarker discovery and precision mental health approaches.