Deciphering the Intercellular Communication Between Immune Cells and Altered Vascular Smooth Muscle Cell Phenotypes in Aortic Aneurysm From Single-Cell Transcriptome Data

Abstract

Background: Vascular smooth muscle cell (VSMC) phenotype switching has been preliminarily found in aortic aneurysms. However, two major questions were raised: (1) What factors drive phenotypic switching of VSMCs in aortic aneurysms? (2) What role does VSMC phenotype transformation play in aortic aneurysms? We speculated that the interaction between infiltrated immune cells and VSMCs played a pivotal role in aortic aneurysm expansion.

Materials and methods: We obtained single-cell transcriptome data GSE155468 that incorporate eight aortic aneurysm samples and three normal aorta samples. A standard single-cell analysis procedure was performed by Seurat (v3.1.2) for identifying the general cell components. Subsequently, VSMCs were extracted separately and re-clustered for identifying switched VSMC phenotypes. VSMC phenotype annotation was relied on the definitions of specific VSMC phenotypes in published articles. Vital VSMC phenotypes were validated by immunofluorescence. Next, identified immune cells and annotated vital VSMC phenotypes were extracted for analyzing the intercellular communication. R package CellChat (v1.1.3) was used for investigating the communication strength, signaling pathways, and communication patterns between various VSMC phenotypes and immune cells.

Result: A total of 42,611 cells were identified as CD4 + T cells, CD8 + T cells, VSMC, monocytes, macrophages, fibroblasts, endothelial cells, and B cells. VSMCs were further classified into contractile VSMCs, secreting VSMCs, macrophage-like VSMCs, mesenchymal-like VSMCs, adipocyte-like VSMCs, and T-cell-like VSMCs. Intercellular communication analysis was performed between immune cells (macrophages, B cells, CD4 + T cells, CD8 + T cells) and immune related VSMCs (macrophage-like VSMCs, mesenchymal-like VSMCs, T-cell-like VSMCs, contractile VSMCs). Among selected cell populations, 27 significant signaling pathways with 61 ligand-receptor pairs were identified. Macrophages and macrophage-like VSMCs both assume the roles of a signaling sender and receiver, showing the highest communication capability. T cells acted more as senders, while B cells acted as receivers in the communication network. T-cell-like VSMCs and contractile VSMCs were used as senders, while mesenchymal-like VSMCs played a poor role in the communication network. Signaling macrophage migration inhibitory factor (MIF), galectin, and C-X-C motif chemokine ligand (CXCL) showed high information flow of intercellular communication, while signaling complement and chemerin were completely turned on in aortic aneurysms. MIF and galectin promoted VSMC switch into macrophage-like phenotypes, CXCL, and galectin promoted VSMCs transform into T-cell-like phenotypes. MIF, galectin, CXCL, complement, and chemerin all mediated the migration and recruitment of immune cells into aortic aneurysms.

Conclusion: The sophisticated intercellular communication network existed between immune cells and immune-related VSMCs and changed as the aortic aneurysm progressed. Signaling MIF, galectin, CXCL, chemerin, and complement made a significant contribution to aortic aneurysm progression through activating immune cells and promoting immune cell migration, which could serve as the potential target for the treatment of aortic aneurysms.

Keywords: VSMC phenotype switching; aortic aneurysm; immune cells; intercellular communication; single-cell transcriptome.

FIGURE 1

Clustering all cells filtrated from GSE155468. The t-distributed stochastic neighbor embedding (t-SNE) plot of the aligned gene expression of cells from the normal aorta and aortic aneurysms, respectively, showing the distribution of nine cell identities (A). The expression level and expression percentage of marker genes are illustrated on the dot plot (B). The change of proportion of identified cell types was shown in a donut chart (C). The proportion of adipocyte-like VSMCs, mesenchymal-like VSMCs, macrophage-like VSMCs, and T-cell-like VSMCs significantly increased in aortic aneurysms (inner donut), and the proportion of contractile VSMCs dramatically decreased in the normal aorta (outer donut).

FIGURE 2

Reanalysis of VSMC subtypes. t-SNE of identified 16 cell clusters (A), which were then annotated as six VSMC phenotypes (B). The expression level and expression percentage of specific marker genes were illustrated on the dot plot (C). The proportion of monocyte, B cells, T cells, and macrophages was increased in aortic aneurysms (inner donut), while the proportion of smooth muscle cells, fibroblasts, and HSC was higher in the normal aorta (outer donut) (D).

FIGURE 3

Validation of VSMC subtypes through immunofluorescence staining. Macrophage-like VSMC was labeled by αSMA and CD68 (A). T-cell-like VSMC was labeled by αSMA and CD3D (B); mesenchymal-like VSMC was labeled by αSMA and CD34 (C).

FIGURE 4

Global cell–cell communication between immune cells and selected VSMC phenotypes. The overall intercellular interaction number (A) and weight (B) were visualized in the network. The line width represents the intensity. Subsequently, the communication intensity from each immune cell type to all VSMC phenotypes or from each VSMC phenotype to all immune cell types was visualized, respectively (C). The incoming and outgoing communication capacity of each cell type (D).

FIGURE 5

MIF signaling pathway-mediated intercellular communication intensity is shown in the hierarchy plot (A) and heatmap (B). Network center score showed the role each cell type played, including sender, receiver, mediator, and influencer (C). Contribution of each ligand–receptor pair to MIF signaling (D). Expression level of every ligand and receptor gene in each cell type (E).

FIGURE 6

Alterations of intercellular communication patterns between the aortic aneurysm and normal aorta. The number of interactions increased (A), but the mean interaction strength decreased in aortic aneurysms (B). Relative information flow of the significant signaling pathway in the normal aorta and aortic aneurysm (C). Sources and targets of signaling complement (D) and chemerin (E). Alteration of incoming/outgoing communication ability of each cell type between the aortic aneurysm and normal aorta (F).

FIGURE 7

Validation of the expression level of significant ligand–receptor pairs of signaling MIF, galectin, CXCL, chemerin, and complement by qPCR. *P < 0.05, **P < 0.001, ***P < 0.0001.