Deconvolution-Based Transcriptomic Analysis in the Hippocampus Reveals Cell Type–Specific Risk Genes and Pathways Associated With Depression and Suicide

Abstract

Background

Major depressive disorder (MDD) is a prevalent mental health condition with the highest associated suicide risk among major psychiatric disorders. Understanding the molecular and cellular mechanisms is crucial for assessing the risk of MDD and associated suicide.

Methods

In this study, transcriptome-based deconvolution was applied to human postmortem hippocampal samples from nonpsychiatric control subjects (n = 29), subjects with MDD who died by means other than suicide (D−S; n = 15), and subjects with MDD who died by suicide (D+S; n = 29). A reference gene expression profile for deconvolution was established using single-nucleus RNA sequencing (snRNA-Seq) analysis. The proportions of various cell types were assessed using the MuSiC2 package, and the cell type–specific gene expression was estimated from the bulk transcriptome using bMIND. Differential expression analysis using the deconvoluted expression profile was conducted to study cell type–specific gene regulation patterns in the hippocampi of patients with MDD who died by nonsuicidal or suicidal means.

Results

The snRNA-Seq analysis identified 11 major cell types in the hippocampus, which were consolidated into 5 primary categories: pyramidal, GABAergic (gamma-aminobutyric acidergic), microglia, macroglia, and endothelial cells. Variability in the cell-type proportions was noted among the sample groups, and the gene expression deconvolution showed distinct patterns among cell types and sample groups. Differential expression analysis at the cell-type level identified more differentially expressed genes than the bulk transcriptome, with variations across comparisons and cell types. Notably, pyramidal neurons displayed significant contrasts between D−S and D+S subjects; the former group exhibited enrichment in cytoskeleton-related pathways and molecular functions, while the latter demonstrated a prevalence of immune-related terms.

Conclusions

The distinct cell type–specific transcriptomic patterns, gene networks, and pathways reveal critical vulnerabilities associated with suicidality in individuals with MDD. These findings underscore the potential for targeted interventions aimed at these specific molecular pathways to mitigate suicide risk in individuals with depression.