Major depressive disorder (MDD) is a significant global health concern, affecting millions and leading to substantial disability. While genetic factors contribute to MDD heritability, the precise molecular mechanisms remain elusive. Brain transcriptomic differences have been observed in MDD, but the underlying etiology is only partially understood. Adding complexity, significant sex differences exist in MDD prevalence, symptomatology, and treatment response. Women experience MDD twice as often as men, with women more likely to present with comorbid anxiety and atypical depression, while men are more prone to substance use disorders and suicide. These clinical disparities suggest sex-specific molecular underpinnings. Previous human studies using bioinformatic and meta-analysis approaches have revealed largely sex-specific brain transcriptomic differences in MDD, with minimal overlap in differentially expressed genes (DEGs). However, these studies were limited by their reliance on bulk tissue analysis, which obscures cell type-specific contributions. Single-nucleus RNA sequencing (snRNA-seq) offers a powerful approach to dissect cell-type specific transcriptional changes in complex neuropsychiatric disorders. This study leveraged snRNA-seq to investigate cell-type and sex-specific transcriptomic alterations in the dorsolateral prefrontal cortex (dIPFC) of individuals with MDD. The dIPFC is a crucial brain region implicated in higher cognitive functions and mood regulation, often exhibiting abnormalities in MDD. The researchers combined newly generated data from a female cohort with previously published data from a male cohort to perform a comprehensive analysis, representing the largest snRNA-seq study of human brain in MDD to date. The researchers hypothesized that while overall patterns of MDD-associated gene expression may show similarities between sexes, the specific cell types and genes involved would differ significantly.

A substantial body of literature indicates cellular abnormalities in depression, ranging from altered morphology and distribution of cell types to proteomic and transcriptomic profile changes. Earlier studies using classical cytological techniques revealed abnormalities in neuronal and glial cell numbers, sizes, and neuropil density, particularly in the prefrontal cortex. More recent transcriptomic studies using bulk tissue samples have implicated a wide range of cell types, including excitatory and inhibitory neurons, astrocytes, oligodendrocyte precursor cells (OPCs), and microglia. However, these studies lacked the cellular resolution to identify the specific cell types most affected. Studies specifically examining sex differences in the brain transcriptome of MDD patients reported distinct gene expression patterns in males and females with limited overlap of DEGs. These findings suggested that the molecular mechanisms underlying MDD might differ substantially between sexes. These earlier studies, however, did not have the resolution to assess cell-type specific differences within males and females.

The study employed single-nucleus RNA sequencing (snRNA-seq) to analyze the dorsolateral prefrontal cortex (dIPFC) of 71 individuals (37 with MDD and 34 controls). The cohort included both males and females. The researchers combined newly generated snRNA-seq data from a female cohort with previously published data from a male cohort. A unified analysis pipeline was used to ensure comparability across datasets. The pipeline included quality control measures to filter out low-quality nuclei, normalization, and dimensionality reduction techniques (SCTransform and Harmony) to account for batch effects and technical variability. An optimized clustering algorithm (scclusteval) was used to identify distinct cell populations. A total of 41 clusters were identified and annotated based on the expression of known marker genes, covering seven major brain cell types: excitatory neurons, inhibitory neurons, oligodendrocytes, astrocytes, OPCs, endothelial cells, and microglia. Differential gene expression analysis was performed separately for males and females to identify DEGs associated with MDD. A rank-rank hypergeometric overlap (RRHO) analysis was used to compare the overall patterns of gene expression changes between the sexes, regardless of statistical significance thresholds. A meta-analysis combining the male and female datasets was performed to increase statistical power in identifying genes commonly altered in both sexes. Gene set enrichment analysis (GSEA) was used to identify pathways enriched in the DEGs, followed by protein-protein interaction analysis (STRING) and ligand-receptor interaction analysis (CellChat) to explore potential functional interactions among affected cell types. Weighted Gene Co-expression Network Analysis (WGCNA) was also conducted to identify gene modules associated with MDD within microglia and PV interneurons.

The study revealed significant sex-specific differences in the cell types most affected by MDD and the DEGs within those cell types. In males, the most prominent changes were observed in deep-layer excitatory neurons, astrocytes, and OPCs. In females, microglia and parvalbumin (PV) interneurons showed the strongest MDD associations. The RRHO analysis indicated substantial concordance in the overall patterns of MDD-associated gene expression changes between sexes, except for the oligodendrocyte lineage. The meta-analysis confirmed the prominence of microglia and ExN10_L46 (deep layer excitatory neurons) clusters as showing significant changes across both sexes. In females, the majority of microglial DEGs (69%) were confirmed as being transcribed and translated in microglia using a separate dataset. Functional analysis in females showed downregulation of inflammation-related pathways in microglia (interferon gamma, interleukin 4 and 13, interleukin 10, and TNFR2 pathways), which may seem counterintuitive but may not simply be a pro-inflammatory/anti-inflammatory dichotomy. Both PV interneuron clusters showed downregulation of heat shock factor 1 (HSF1)-related terms and enrichment of immune-related gene sets. STRING analysis revealed interactions between protein products of DEGs in microglia and PV interneurons, suggesting potential crosstalk between these cell types through cell surface molecules and the extracellular matrix. CellChat analysis indicated altered communication between microglia and PV interneurons in females with MDD, with an increase in overall communication strength but specific changes in certain signaling pathways (e.g., increased SPP1-integrin and decreased GAS6-MERTK communication). WGCNA in females further supported the findings, with modules highly correlated with MDD status showing an overlap with DEGs in microglia and PV interneurons, and enrichment for pathways identified via GSEA.

This study provides compelling evidence for both sex-specific and shared transcriptomic changes in the dIPFC in MDD. The findings highlight the importance of considering sex as a biological variable in understanding MDD pathophysiology and treatment strategies. The sex-specific involvement of different cell types suggests that therapeutic interventions might need to be tailored to each sex. The strong MDD association with microglia in females aligns with the hypothesis of reduced microglial activation and increased synaptic connectivity in females with MDD. The simultaneous downregulation of both pro-inflammatory and anti-inflammatory pathways in female microglia challenges the simplistic pro-inflammatory/anti-inflammatory paradigm and underscores the complex interplay of microglial functions in MDD. The observed dysregulation of PV interneurons, along with the evidence of impaired communication between microglia and PV interneurons, suggests complex interactions and shared pathways contributing to MDD pathogenesis, especially in females. The study provides valuable insights into potential cellular and molecular targets for the development of sex-specific therapies.

This large-scale snRNA-seq study offers novel insights into the cell-type and sex-specific transcriptomic alterations in the dIPFC of individuals with MDD. The findings highlight the importance of considering sex as a biological variable and underscore the complex interplay between different cell types in MDD pathogenesis. The identification of microglia and PV interneurons as key players in the female MDD transcriptome warrants further investigation to better understand the mechanisms driving MDD and to develop targeted therapeutic strategies. Future studies could explore the functional consequences of the observed gene expression changes, further investigate the microglia-PV interneuron crosstalk, and examine the interactions between genetic and environmental factors in shaping sex-specific MDD pathophysiology.

This study has limitations that should be acknowledged. The use of data from two separate cohorts (male and female) prevents a direct comparison of sex and disease interaction. Although a unified pre-processing pipeline was employed, methodological differences in data acquisition could still influence the results. The relatively small sample size, while comparable to other snRNA-seq studies in neuropsychiatric conditions, may limit the statistical power to detect subtle transcriptomic changes. The lack of spatial information inherent to snRNA-seq prevents definitive conclusions about cellular proximity and interactions. Further research is needed to validate the findings and explore the functional consequences of the observed gene expression changes using larger sample sizes and incorporating spatial transcriptomic techniques.