Single-cell RNA sequencing of cerebrospinal fluid immune cells in relapsing–remitting multiple sclerosis: insights into cellular composition and immune dynamics
摘要
Relapsing–remitting multiple sclerosis (RRMS) is a chronic autoimmune disease characterized by episodic inflammation and demyelination within the central nervous system. Recent advances in single-cell RNA sequencing (scRNA-seq) have enabled high-resolution profiling of cerebrospinal fluid (CSF) immune cells, revealing distinct transcriptional and functional states. This study specifically targets the CSF compartment to delineate immune cell heterogeneity, transcriptional regulation, and cell–cell interactions underlying RRMS pathology.
MethodsCSF samples were collected from RRMS patients and matched healthy controls. Single-cell suspensions were prepared, and scRNA-seq libraries were constructed using the GEXSCOPE™ kit. Data processing involved rigorous quality control, normalization, dimensionality reduction, and clustering using established algorithms. Complementary flow cytometry was used for validation. Differential gene expression, pseudotime trajectories, and cell–cell interaction analyses (via CellPhoneDB) were performed to characterize immune dynamics.
ResultsA total of 23,247 CSF cells were initially captured (12,200 HC, 11,047 RRMS). After quality control filtering, 9528 high-quality cells remained for downstream analyses. The analysis revealed six major immune cell types, with T and natural killer (NK) cells comprising the majority. RRMS samples exhibited a significant expansion of B cells and plasma cells, accompanied by a notable reduction in plasmacytoid dendritic cells and certain monocyte subtypes compared to controls. Differential expression analysis demonstrated upregulation of genes linked to antigen processing, cytokine signaling, and autophagy, while key transcription factors (e.g., decreased JUNB and altered NFIL3) indicated a reprogramming of T cell and monocyte functional states. Pseudotime trajectory analysis placed naïve T cells at early developmental stages, transitioning toward activated phenotypes enriched in inflammatory mediators. Cell–cell interaction analysis uncovered enhanced ligand–receptor communication, notably via the CXCL12–CXCR4 axis, suggesting that aberrant cross talk among T cells, B cells, and mononuclear phagocytes may amplify central nervous system inflammation. These results delineate an inflammation-associated immune network observed in RRMS CSF that underpins localized autoimmunity and highlights potential targets for therapeutic intervention.
ConclusionThe findings provide novel insights into the cellular composition and dynamic transcriptional shifts of CSF immune cells in RRMS. The expanded B cell compartment and altered T cell activation profiles underscore the importance of localized CNS inflammation. Enhanced intercellular communication pathways further implicate a coordinated immune response driving disease activity. These findings provide preliminary, exploratory insights into RRMS-associated CSF immune dynamics and are intended to guide future mechanistic and validation studies in larger cohorts.