Associations between serum amino acids and genetic liability to depression in the LURIC cohort
摘要
Blood concentrations of amino acids have been associated with depression, although findings have been inconsistent. We investigated associations between serum amino acid concentrations and genetic liability to depression, assessed by two genetic depression risk scores, in the Ludwigshafen Risk and Cardiovascular Health (LURIC) study.
MethodsThe LURIC study enrolled 3,316 participants referred for coronary angiography; serum amino acid measurements were available for 2,143 individuals and were quantified by ion-exchange chromatography. Two weighted genetic depression risk scores, based on 101 and 220 genome-wide significant variants, were calculated. Associations between genetic depression risk scores and amino acid concentrations were assessed using multivariable linear regression models adjusted for age, sex, body mass index, renal function (eGFR), diabetes, hypertension, and coronary artery disease. Clinical depression data were not available; thus, analyses reflect genetic liability rather than manifest depression.
ResultsIn multivariable-adjusted analyses, higher genetic depression risk (per SD increase) was associated with lower concentrations of α-aminoadipic acid (GDRS₁₀₁: β = −0.097, 95% CI − 0.161 to − 0.033, p = 0.003; GDRS₂₂₀: β = −0.076, 95% CI − 0.141 to − 0.011, p = 0.021), which remained significant after false discovery rate (FDR) correction. α-aminobutyric acid showed inverse associations that did not remain significant after FDR correction. No statistically significant associations for other amino acids remained after multivariable adjustment and correction for multiple testing. Restricted cubic spline analyses suggested predominantly linear relationships, with evidence of non-linearity observed for selected amino acids, particularly homoarginine with respect to GDRS₂₂₀ (p for non-linearity = 0.005).
ConclusionGenetic liability to depression was associated with differences in serum amino acid concentrations, particularly lower α-aminoadipic acid concentrations in this cardiovascular cohort, whereas associations with other amino acids were not robust after correction for multiple testing. These findings are associative and hypothesis-generating and should be confirmed in cohorts with detailed depression phenotyping and longitudinal follow-up.