Integrative genomic characterization of the FADS1-2-3 mood disorder risk locus: implications for the role of polyunsaturated fatty acids.

BACKGROUND

Polyunsaturated fatty acids (PUFA) have been implicated in mood disorder pathophysiology, with Mendelian randomisation studies suggesting a causal role. However, much of this evidence centres on the highly pleiotropic FADS1-2-3 gene cluster, raising concerns about horizontal pleiotropy and uncertainty regarding the true effector gene(s). To assess the validity of these concerns, we performed integrative genetic characterization of the FADS1-2-3 mood disorder risk locus.

METHODS

We analysed summary statistics from the largest (up to n = 4 million) genome-wide association studies (GWASs) of bipolar disorder (BD) and major depressive disorder (MDD) and performed pairwise genetic colocalization. Lead and proxy variants were annotated across global populations, and cis-eQTL effects were assessed using bulk tissue and single-cell RNA-seq datasets.

RESULTS

BD and MDD GWAS signals at the FADS1-2-3 locus shared a common causal variant (PPH4 = 0.956) with directionally consistent risk alleles. Colocalization analyses prioritised FADS1 over FADS2 and FADS3 as the most likely effector gene, with robust colocalization across multiple brain tissues. Beyond PUFA metabolism, TMEM258 and MYRF emerged as credible alternative effector genes, implicating N-linked glycosylation and axonal myelination as alternative mechanisms through which this locus could influence mood disorder risk.

CONCLUSIONS

These findings illustrate the mechanistic complexity of the FADS1-2-3 mood disorder risk locus. While the evidence prioritising FADS1 as a likely effector gene is consistent with a potential aetiological role for PUFA in mood disorders, our findings also highlight genetic co-regulation of other local genes as a potential source of horizontal pleiotropy. Future studies using orthogonal approaches are necessary to fully address this open question.