Immune Phenotypes of Oligodendroglial-Lineage Cells in MDD and in Response to Chronic Stress-Induced Microglial Inflammation.

BACKGROUND: Major Depressive Disorder (MDD) is a complex and heterogenous psychiatric disorder affecting more than 300 million people worldwide. Though the mechanisms underlying MDD are not fully understood, a subset of depressive patients can manifest chronic neuroinflammatory responses. In addition, brain-imaging studies in MDD patients have provided evidence of white matter reductions and such changes are hypothesized to lead to disruption of Oligodendroglial-Lineage cells (OLN) homeostatic mechanisms, resulting in destabilization of emotional/cognitive circuitry. The present goal of the study is to investigate whether these histopathological alterations are linked with chronic inflammation through single-nucleus transcriptomics of MDD patients and in a depression mouse model.

METHODS: Bioinformatics: Raw counts were downloaded from GEO (GSE144136), and expression objects created using Seurat. OLN were subsetted, clustered (using UMAP), and annotated based on their expression of early and late OLN markers. Pseudotime trajectory analysis was conducted using Moncle3. In vivo: The Repeated Social Defeat Stress (RSDS) paradigm (10 days) was used to induce depressive-like behavior in 8-12-week-old male CX3CR1-GFP+ and CSPG4-EGFP+ mice. Behavioral tasks (BH) were performed to stratify the defeated mice to susceptible (S; depressive-like) and resilient (R; non-depressive) to stress groups. The study focuses on the MDD-affected prefrontal cortex (mPFC) area (Fig).

RESULTS: Separate clustering of MDD and Control samples yielded four similar clusters labeled based on their gene expression as: OPCs, Committed OPCs, Immature Oligodendrocytes (OLs), and Mature OLs. One MDD-specific cluster was also obtained and labeled as Immune OLs (ImOL), given its expression of immune genes, such as: ARGHAP24, ADAM28, LPAR6, C3, CD74, and P2RY12. Further analysis of the ImOL cluster revealedexpression of markers associated with later phases of OLN progeny. Finally, a predictive model of pseudotime trajectory revealed alterations in the progression of OLN progeny in MDD samples. We then sought to identify this novel subset of ImOL in the RSDS model of depression. MHCII, C3 and P2RY12 were examined in conjunction with OLN markers (CSPG4, O4, CNP), with significant changes observed in S groups. Interestingly, microglial phagocytosis of myelin elements (CX3CR1-GFP, MHCII, CNP), was significantly increased in S groups post RSDS, rendering the ImOL populations as potential recruiters of innate immunity.

CONCLUSIONS: Together, our computational data in MDD patients and depression mouse model, show that OLN can swiftly respond to chronic stress, and adopt an immune-like phenotype, potentially contributing to the white matter disturbances observed in MDD.