Effects of 6-months of SSRI use on DNA-methylation and gene expression in blood.

BACKGROUND

Selective serotonin reuptake inhibitors (SSRIs) are a recommended first line medication for the treatment of major depressive disorder, due to higher tolerability and lower risk of adverse effects than other antidepressants. The mechanisms by which SSRIs reduce depressive symptoms are not well understood, but are hypothesised to include direct effects on serotonin signalling and synaptic remodelling, and indirect effects on inflammation. Indirect or off-target effects may be detectable in blood and can be investigated using methylome- and transcriptome-wide approaches.

METHODS

The Staged Treatment in Early Psychosis (STEP) clinical trial included a 6-month long randomised, placebo-controlled trial of the SSRI fluoxetine in a cohort of young people at ultra-high risk for psychosis. A methylome-wide association study (MWAS; NTotal (before/after/both) = 104 (52/52/44), NSSRI (before/after/both) = 45 (21/24/18)) and differential expression analysis were performed on longitudinal blood samples collected at the start and end of the 6 months to identify changes in DNA-methylation and gene expression associated with medium-term SSRI exposure.

RESULTS

Four methylation CpGs (cg26253898, cg09719563, cg22216017, cg26017656) were significantly associated with SSRI exposure (FDR < 0.1, 2 CpGs at FDR < 0.05) and annotated to genes involved in glucose metabolism, synaptic remodelling and inflammation (GCG, COL23A1, PEG10, SGCE, MFGE8). Gene-set enrichment analyses of genes annotated to the top 100 CpGs identified significant tissue-specific enrichments in artery, adipose and spleen tissues, and in the 'postsynaptic density' GO term. No genes were differentially expressed, including genes annotated to the significant methylation CpGs.

CONCLUSIONS

Medium-term SSRI use during the STEP trial was associated with changes in DNA-methylation that may partially explain the potential antidepressant mechanisms and adverse effects of SSRIs, however replication in other cohorts is necessary to establish if these changes are generalisable to SSRI use more broadly.