New Insights into Stress Metabolomics. Looking for new Diagnostic Biomarkers.

Stress is associated with the onset of various neurological disorders such as depression, PTSD, and anxiety. Despite the extensive research performed, metabolic changes triggered in response to acute psychological stress remain unclear. This study evaluates acute stress biomarkers and its adverse effects in university students through psychometric, biochemical, and metabolomic analyses, implementing Machine Learning on statistical models. In the study, forty participants were subject to relaxation and stress induction using autogenic training and a modified Trier Social Stress Test (TSST-M). Psychometric questionnaires confirmed the achievement of these states, while saliva and blood were sampled for biochemical analyses. Additionally, blood samples were applied to untargeted metabolomic approaches. The results reveal that although most biomarkers showed changes under stress state, the machine learning predictive model successfully identified salivary α-amylase and State-Trait Anxiety Inventory-state (STAI-s) as prominent stress markers. Additionally, several metabolic pathways, including steroid hormone biosynthesis, glycerophospholipid metabolism, linoleic acid metabolism, tyrosine metabolism, and aminoacyl-tRNA biosynthesis, were affected. Alterations of this sort, we conclude, allow us to gain further understanding into the adverse effects systematically associated with stress. In this way, our findings highlight the significant impact of acute mental stress on multiple metabolic pathways directly implicated in stress-related disorders.