AUTHOR=Liu Mingxia , Pan Weigang , He Jing , Ling Sihai , He Yi , Yang Jian , Mao Peixian , Sun Zuoli TITLE=Unveiling chiral amino acid alterations and glycine dysregulation in late-life depression through targeted metabolomics JOURNAL=Frontiers in Psychiatry VOLUME=Volume 16 - 2025 YEAR=2025 URL=https://www.frontiersin.org/journals/psychiatry/articles/10.3389/fpsyt.2025.1558796 DOI=10.3389/fpsyt.2025.1558796 ISSN=1664-0640 ABSTRACT=BackgroundLate-life depression (LLD) is a major depressive disorder that is highly prevalent among older people, and there are currently no validated biomarkers for the diagnosis and treatment of LLD. Although dysregulated amino acid metabolism has been increasingly implicated in neuropsychiatric disorders, including LLD, most existing studies overlook the chiral nature of amino acids, potentially leading to inaccurate or incomplete findings. To address this gap, this study aimed to precisely characterize the serum chiral amino acid profiles in patients with LLD and identify potential biomarkers.MethodsUsing liquid chromatography tandem mass spectrometry combined with a chiral derivatization technique, the serum levels of 34 amino acids were analyzed in 53 LLD patients and 37 healthy controls (HCs). ResultsSignificant alterations in both D- and L-enantiomers were observed, including reduced levels of D-methionine, D-glutamic acid, D-threonine, and L-threonine, alongside elevated glycine levels in LLD compared to HCs. The combination of D-methionine and glycine demonstrated moderate discriminatory power for distinguishing LLD from HCs, with an area under the curve of 0.71. Notably, glycine levels were significantly lower in antidepressant treatment responders than in non-responders. Additionally, D- and L-glutamic acid levels were differentially associated with specific cognitive function indicators.DiscussionThese findings underscore the importance of accounting for amino acid chirality in biomarker research and highlight chiral amino acids as promising candidates for the diagnosis of LLD and the prediction of treatment response.