Gender - and age-related differences in plasma metabolic fingerprinting of healthy populations

Abstract

Abstract Background/Aims: The objective of this study was to investigate potential differences in the plasma metabolic fingerprinting of healthy individuals based on sex and age. Methods: Blood specimens were collected from patients at the First Affiliated Hospital of Nanchang University for health check-ups. The specimens were randomly divided into discovery and validation groups, maintaining a 1:1 ratio. Small molecule metabolites in each sample were detected using liquid chromatography-mass spectrometry (LC-MS) and gas chromatography-mass spectrometry (GC-MS). Results: Plasma metabolites exhibited significant variations between sexes and age groups in our population. A comparison between men and women revealed 13 differential metabolites, with acetylcholine having the highest differential expression (AUC = 0.909). A diagnostic model (model 1) combining PC 36:4, acetylcholine, L-threonine, and arginine showed an AUC value of 0.973, a sensitivity of 91%, and a specificity of 94%. Model 2, which combined creatinine, L-methionine, acetylcholine, and L-tyrosine, had an AUC value of 0.955, a sensitivity of 94%, and a specificity of 88%. When comparing age groups, the OPLSDA model exhibited moderate stability and higher predictive efficacy in the 18-44 and 45-67 age groups. The differential metabolites were predominantly lipid molecules, with the highest AUC found in Dehydroisoandrosterone-3-sulfate. Additionally, we observed that LysoPC 20:1, LysoPC 20:0, PI 38:5, PC 40:3, 3-Hydroxydodecanoic acid 12:0(3-OH), Choline, N-Formyl-L-methionine, and Homocystine positively correlated with age, while Dehydroisoandrosterone-3-sulfate and Piperine were negatively correlated with age. Conclusion: These differential metabolites provide insights for the prevention and treatment of age- and gender-related diseases.