Single-cell transcriptome profiling reveals dynamic cell populations and immune infiltration in cerebral cavernous malformation

Abstract

Introduction: The cellular subpopulations and signaling pathways in the pathological tissues of cerebral cavernous malformation (CCM) remain incompletely understood. To gain a deeper understanding of the pathogenesis of CCM, we aimed to comprehensively map the cellular subpopulations and signaling pathway alterations in the pathological tissues of sporadic CCM patients.

Methods: Lesional brain vascular tissues from CCM patients and normal brain vascular tissues from controls were collected. Multiplex fluorescent immunohistochemistry and single-cell RNA sequencing were performed on the lesional tissues. Differential gene expression, pathway enrichment analysis, and cell-cell communication analysis were conducted to investigate disease-related changes.

Results: We identified 8 major cell types in the lesion tissues of CCM patients. We observed an increased proportion of monocytes, neutrophils, and NK cells in the lesion tissues of CCM patients. Twenty-eight significantly differentially expressed genes were identified, and pathways such as NK cell-mediated cytotoxicity showed alterations. Cell-cell communication analysis revealed an increase in both the types and strength of communication between cells in the CCM lesion tissues.

Conclusion: This study provides the single-cell transcriptomic analysis of CCM lesions, revealing increased monocytes, neutrophils, and NK cells, along with dysregulated gene expression and signaling pathways. Enhanced intercellular communication, particularly via VEGF and ADGRE5 pathways, highlights potential therapeutic targets for CCM.

Keywords: cell populations; cerebral cavernous malformation; immune infiltration; multiplex fluorescent immunohistochemistry; single-cell RNA sequencing.

Figure 1

Multiplex fluorescent immunohistochemistry result. CD31 labels endothelial cells, α-SMA labels vascular smooth muscle cells, DAPI labels DNA, Claudin5 and ZO-1 label the blood-brain barrier, and Vimentin labels fibroblasts. Scale bar=500 and 50 μm.

Figure 2

Cell types, distribution, significant differential genes, GO and KEGG enrichment results of Endothelial cells. (A) UMAP plot annotated by cell type shows 8 cell types, including NK cells, Monocytes, Neutrophils, Fibroblasts, Endothelial cells, HSC-G-CSF (hematopoietic stem cells mobilized to peripheral blood by granulocyte colony-stimulating factor), tissue stem cells, and B cells. (B) cell type proportion stacked bar plot. CCM sample shows more Monocytes, Neutrophils, NK cells and less Fibroblasts, Endothelial cells. (C) Log2FC scatter plot for differential genes in each cell type, with red representing upregulated genes, blue representing downregulated genes, and gray indicating no significant difference. (D) Bubble plot of the 28 most significant differential genes, with genes on the x-axis and cell type sample group on the y-axis. (E) GO enrichment analysis of differential genes in Endothelial cells. The most significant signaling pathways for the BP, CC, and MF terms are regulation of multicellular organismal processes, cell surface and signaling receptor binding. (F) Enrichment plot of the top 10 most significant upregulated and downregulated pathways in Endothelial cells. The most significant upregulated and downregulated signaling pathway are Natural Killer cell-mediated cytotoxicity and EGFR tyrosine kinase inhibitor resistance.

Figure 3

GO and KEGG enrichment results of Monocytes. (A) GO enrichment analysis results for differential genes in Monocytes. The most significant signaling pathways for the BP, CC, and MF terms are system development, extracellular matrix, and extracellular matrix structural constituent. (B) Top 10 upregulated pathways in Monocytes. The most significant upregulated signaling pathway is Antigen processing and presentation. (C) Top 10 downregulated pathways in Monocytes. The most significant downregulated signaling pathway is Drug metabolism - cytochrome P450.

Figure 4

Cell communication analysis results. (A) Network plot comparing cell communication types between CCM and control samples, with red indicating more types in CCM and blue indicating fewer types. (B) Network plot comparing communication intensity between CCM and control samples, with red indicating stronger communication in CCM and blue indicating weaker communication. (C) Heatmap comparing the number of cell communication types in CCM vs. control. (D) Heatmap comparing communication strength in CCM vs. control. (E) Comparison of the number of cell communication types. CCM samples had 660 cell communication types while control had 449. (F) Comparison of the total communication strength. CCM samples had an interaction strength of 21.583 while control had 15.073.

Figure 5

Significant communication pathways between CCM and control samples. Left plot shows proportion of communication pathway strength. Right plot shows communication pathway strength values. Orange indicates significantly activated pathways in CCM relative to controls while blue indicates significantly suppressed pathways. Significantly activated pathways include VEGF, ADGRE5, EPHA, MIF, CLEC, CXCL, MK, VISFATIN, CD6, ALCAM, FN1, ICAM, IL1, CADM, CDH, PARs, BAFF, OCLN, SEMA4, PTN, SELPLG, VCAM, CALCR, SPP1, OSM, GALECTIN, and MHC-I. Significantly suppressed pathways include CD22, TNF, COMPLEMENT, GRN, EGF, GAS, CD23, THY1, SEMA7, NEGR, PROS, CSF, PDGF, CD86, MPZ, SEMA6, CD45, FGF, PTPRM, CD34, TENASCIN, IL16, ESAM, CDH5, EPHB.