Multi-omics Analysis of Energy Metabolism Pathways Across Major Psychiatric Disorders

Mol Neurobiol. 2025 Jun 11. doi: 10.1007/s12035-025-05133-8. Online ahead of print.

ABSTRACT

The brain requires a large amount of energy, primarily obtained through glucose metabolism, which appears to be disrupted in various neuropsychiatric disorders. The etiology of neuropsychiatric disorders is complex and involves genetic factors that are slowly being identified. To investigate whether glucose metabolism-related genes are associated with major psychiatric disorders, we conducted secondary analyses using genome-wide association study (GWAS) summary statistics for major depressive disorder (MDD), bipolar disorder (BD), and schizophrenia (SCZ). Based on predefined glucose metabolism pathway genes, we conducted a multi-layer integrative analysis using gene-level approaches including multi-marker analysis of genomic annotation (MAGMA), transcriptome-wide association studies (TWAS) with joint-tissue imputation (JTI), and summary-based Mendelian randomization (SMR). We further explored gene expression patterns across tissues, druggability, and applied gene network analysis to evaluate the interactions. MAGMA identified 1 significant gene for MDD, 19 for BD, and 32 for SCZ, with gene RBKS shared across all three disorders. TWAS-JTI detected 17 and 35 transcriptome-wide significant genes for BD and SCZ, respectively, while SMR prioritized 2 and 9 putatively causal genes. For MDD, no convergent evidence emerged from TWAS-JTI or SMR analyses. Integrated analysis highlighted NDUFS2, NDUFS7, and NDUFC2 in the oxidative phosphorylation pathway as potential therapeutic targets. Finally, gene network analysis highlighted enrichment in mitochondrial respiratory chain complex I biogenesis, NADH dehydrogenase complex assembly, and ATP synthesis. Our results reinforce the role of energy metabolic disturbance in psychiatric disorders, particularly in BD and SCZ. These findings open avenues for targeted therapeutic interventions, warranting further validation across populations.

PMID:40498271 | DOI:10.1007/s12035-025-05133-8