Integrative analysis identifies DCBLD2 and immune-related biomarkers for major depressive disorder: evidence from human peripheral blood, post-mortem brain, and rat models.

Major depressive disorder (MDD) is a prevalent mental illness, and inflammatory processes are considered a pivotal component of the pathogenesis of MDD. This study aims to identify novel biomarkers associated with the development of MDD and to elucidate the underlying immunological mechanisms. Mendelian randomization (MR) studies confirmed that two inflammatory cytokines are related to MDD. Subsequently, MDD-related transcriptomic datasets were retrieved from the Gene Expression Omnibus (GEO) database, and potential mechanisms were revealed by screening for differentially expressed genes (DEGs) and enrichment analyses. To identify key genes associated with MDD, we employed various machine learning methods, including univariate logistic regression, least absolute shrinkage and selection operator (LASSO) regression, and random forest algorithms, to construct a risk prediction nomogram and ROC curve to evaluate the diagnostic efficacy of candidate genes for MDD. Additionally, we analyzed the immune infiltration in MDD and healthy controls, investigated the expression of key genes in publicly available human post-mortem prefrontal cortex (PFC) transcriptomic datasets (bulk and single-cell) and validated candidate biomarkers in peripheral blood from an independent human cohort and in the PFC tissues of a chronic unpredictable stress (CUMS) rat model. Subsequently, a six-gene diagnostic signature (DCBLD2, FZD5, GP1BA, MMP8, RNF144B, and SOCS1) was constructed using logistic regression, demonstrating good discriminative ability (AUC = 0.83). Immune infiltration revealed that the infiltration levels of seven types of immune cells were significantly different between the MDD and control groups, and were correlated with DCBLD2. Bulk RNA sequencing found that DCBLD2 was significantly overexpressed in the postmortem PFC of MDD. Single-cell RNA sequencing analysis of the MDD post-mortem PFC further localized DCBLD2 overexpression specifically within microglia and endothelial cells. Experimental validation indicates that six genes (DCBLD2, FZD5, GP1BA, MMP8, RNF144B, and SOCS1) exhibited significantly elevated expression in the peripheral blood of MDD patients. In the prefrontal cortex of rats subjected to the chronic unpredictable stress (CUMS) model, significant upregulation was only confirmed for the key gene DCBLD2. Immune dysregulation is closely related to MDD, and DCBLD2, FZD5, GP1BA, MMP8, RNF144B, and SOCS1 may represent candidate biomarkers and potential therapeutic targets for MDD, with DCBLD2 being a particularly promising candidate.