Abstract
BACKGROUND: A critical need exists for objective biomarkers and novel therapeutic targets in major depressive disorder (MDD). Although dysfunction in mitochondrial immunometabolism is implicated in MDD, the specific causal genes suitable for clinical translation remain largely unidentified. This study aimed to bridge this gap by identifying mitochondria-related genes that have a causal impact on MDD risk through their expression in specific immune cells. METHODS: We integrated multi-omics data with machine learning to pinpoint key mitochondria-related energy metabolism genes (MEMRGs) linked to immune cell infiltration, assessed via ssGSEA and CIBERSORT algorithms. Cell-type-specific two-sample Mendelian randomization (MR) was employed to evaluate causal relationships between gene expression and MDD risk. Findings were validated in a chronic unpredictable mild stress (CUMS) rat model. RESULTS: Our analysis identified five genes-HK2, NDUFS4, NEU1, SOD1, and UCP2-whose expression in distinct immune populations had significant causal effects on MDD risk. Notably, HK2, NDUFS4, and NEU1 were identified as protective factors, while UCP2 and SOD1 were risk factors in specific cell types. The clinical relevance of this panel was supported by its diagnostic performance in an independent cohort, and the upregulation of the principal risk gene, UCP2, was confirmed in the hippocampus of CUMS rats. CONCLUSION: This study provides robust genetic evidence establishing a causal link between the expression of specific mitochondrial genes in immune cells and the risk of MDD. By prioritizing UCP2, SOD1, HK2, NDUFS4, and NEU1, our findings highlight novel, immune-mediated pathways in depression and nominate promising targets for future diagnosis and therapeutic intervention.