Metabolic profiling in major depressive disorder with high psychological resilience: changes in amino acid and carbohydrate metabolism

TL;DR

The study finds that major depressive disorder, not psychological resilience, mainly affects metabolic profiles, with specific amino acid and carbohydrate changes in resilient patients.

Contribution

Identifies novel metabolic signatures in high-resilience MDD patients and suggests potential biomarkers for depression.

Findings

MDD status, not resilience, primarily drives serum metabolic changes.

High-resilience MDD patients show altered amino acid and pentose-glucuronate metabolism.

A diagnostic model achieved 81.1% accuracy in distinguishing MDD patients.

Abstract

Psychological resilience varies among major depressive disorder (MDD) patients, with some exhibiting high resilience. This challenges the notion of resilience as purely protective and suggests biological heterogeneity. Both resilience and MDD have been linked to metabolic alterations, but their independent and interactive effects remain unclear. This study aims to investigate how resilience and MDD jointly affect metabolic profiles, with a focus on identifying key metabolic and pathway alterations in high-resilience MDD patients compared to healthy controls, and exploring their potential for diagnostic biomarkers.

Targeted serum metabolomics using UPLC-MS/MS was conducted in MDD patients and healthy controls. Resilience was assessed via the Ego Resiliency Scale (ERS). Interaction effect analysis examined the main and interactive influences of resilience and MDD. Key metabolites in high-resilience MDD were identified by OPLS-DA and pathway enrichment. A logistic regression model with cross-validation assessed diagnostic accuracy in training and test sets.

A total of 271 participants were enrolled, and about one-third of MDD patients exhibited high resilience. The MDD×resilience interaction was not significant, whereas MDD showed a significant main effect on metabolite levels. Five key metabolites were identified in high-resilience individuals, with arginine, methionine, and kynurenine downregulated and threonic and erythronic acids upregulated in the high-resilience MDD group. The diagnostic model achieved an area under the curve (AUC) of 0.811 in the test set.

MDD status—rather than psychologically resilience—was the primary driver of serum metabolic variation. In high-resilience individuals, alterations converged on amino acid pathways (driven by lower arginine, methionine, kynurenine) and pentose-glucuronate axis (with higher threonic and erythronic acids), the latter is potentially linked to redox imbalance. These features may provide novel insights into depression-related metabolic dysregulation.

Not applicable.

The online version contains supplementary material available at 10.1186/s12888-026-07798-4.

Figures