Metabolomic and Lipidomic Signatures of Depressive Symptoms in the REasons for Geographic And Racial Differences in Stroke (REGARDS) Study

Background Depression is a highly prevalent condition with a substantial health burden and known links to metabolic dysfunction. We investigated plasma metabolites and lipids associated with depressive symptoms at baseline, including sex- and race-specific differences, and subsequently examined longitudinal associations of baseline metabolites and lipids with depressive symptoms over follow-up. Methods Baseline plasma samples were profiled for 162 metabolites and 195 lipid species using targeted mass spectrometry in the REasons for Geographic And Racial Differences in Stroke (REGARDS) stroke case-cohort. Depressive symptoms were assessed using the 4-item Center for Epidemiologic Studies Depression Scale (CES-D-4) at baseline and during follow-up. Cross-sectional associations were evaluated using weighted logistic regression, and longitudinal associations were assessed using weighted generalized estimating equations. Effect modification by sex and race was examined. Model 1 was adjusted for age, race, and sex (and time in longitudinal analyses). Model 2 was further adjusted for body mass index, smoking status, alcohol use, physical activity, perceived stress, and incident stroke. False discovery rate (FDR) correction was applied to account for multiple testing. Results Of 1,938 participants, 205 had depressive symptoms at baseline. At baseline, several metabolites (e.g., glycine, cyclic AMP) and lipid species were nominally associated with depressive symptoms, though none remained significant after FDR correction. In longitudinal analyses, stronger and more consistent associations emerged. Higher levels of anserine (OR=1.27, 95% CI: 1.16-1.39) and glucose (OR=1.36, 95% CI: 1.10-1.68) were associated with increased odds of depressive symptoms, with anserine remaining significant after FDR correction. Multiple lipid species, particularly phosphatidylethanolamines (PEs) and triglycerides (TGs), were significantly associated with depressive symptoms after FDR correction in longitudinal models. Significant sex interactions were observed for several lipid species, with stronger positive associations in females and attenuated or inverse associations in males. Conclusions In this stroke-enriched case-cohort, PEs and TGs were associated with depressive symptoms over time. Sex-specific lipid differences highlight biological heterogeneity. These results identify amino acid, phospholipid, and triacylglycerol pathways in depressive symptoms, supporting further mechanistic and biomarker investigations.