Preliminary assessment of pre‐morbid DNA methylation in individuals at high genetic risk of mood disorders

OBJECTIVES: Accumulating evidence implicates altered DNA methylation in psychiatric disorders, including bipolar disorder (BD) and major depressive disorder (MDD). It is not clear, however, whether these changes are causative or result from illness progression or treatment. To disentangle these possibilities we profiled genome‐wide DNA methylation in well, unrelated individuals at high familial risk of mood disorder. DNA methylation was compared between individuals who subsequently developed BD or MDD [ill later (IL)] and those who remained well [well later (WL)]. METHODS: DNA methylation profiles were obtained from whole‐blood samples from 22 IL and 23 WL individuals using the Infinium HumanMethylation450 BeadChip. Differential methylation was assessed on a single‐locus and regional basis. Pathway analysis was performed to assess enrichment for particular biological processes amongst nominally significantly differentially methylated loci. RESULTS: Although no locus withstood correction for multiple testing, uncorrected P‐values provided suggestive evidence for altered methylation at sites within genes previously implicated in neuropsychiatric conditions, such as Transcription Factor 4 (TCF4) and Interleukin 1 Receptor Accessory Protein‐Like 1 ([IL1RAPL1]; P≤3.11×10(−5)). Pathway analysis revealed significant enrichment for several neurologically relevant pathways and functions, including Nervous System Development and Function and Behavior; these findings withstood multiple testing correction (q≤0.05). Analysis of differentially methylated regions identified several within the major histocompatibility complex (P≤.000 479), a region previously implicated in schizophrenia and BD. CONCLUSIONS: Our data provide provisional evidence for the involvement of altered whole‐blood DNA methylation in neurologically relevant genes in the aetiology of mood disorders. These findings are convergent with the findings of genome‐wide association studies.