Diversity of arterial cell and phenotypic heterogeneity induced by high-fat and high-cholesterol diet

Objective: Disturbed blood flow characterized by low and oscillatory shear stress and complex microenvironment is one of the causes of the atherosclerosis progression. During the development, endothelial cells generally develop into a dysfunction state, so do other cells. To further analyze the heterogeneity of different cell types, single-cell RNA sequencing was used to provide detailed transcriptomic information.
Approach and results: Transcriptional analysis of single cells from aortic arch of 16-week Apoe-/- mice under normal and atherosclerotic condition was carried out using single-cell RNA sequencing. By unsupervised clustering, 10 types of cells were identified: endothelial cells, immune cells, fibroblasts, erythroids, smooth muscle cells, cycling basal cells, clara cells, mesothelial cells, adipocytes and neurons. Among these cells, endothelial cells and immune cells exhibited great heterogeneity especially between normal and atherosclerotic group. Gene expression profile of later screened cells identified ten subpopulations of endothelial cells of different functions. Distribution of these cells in the control group and HFD group varies. Immune cells consist of macrophages, foam cells, B cells, T cells, neutrophils and NK cells mainly. During the data processing, the cell sorting is done by recognition with specific markers and rules.
Conclusions: Our data provide a comprehensive description of all kinds of cells in aortic arch and the heterogeneity brought by western diet. These findings open a new sight into the transcriptional development of atherosclerosis. Single cell suspension of the two aortas from each group was pooled together as one sample. Single cell RNA sequencing, library construction and quality control were executed using the illumina-HiSeq3000 platform by Genergy Bio-technology (Shanghai) Co., Ltd. Dying cells with mitochondrial RNA above 30% and cells lacking information with UMI<200 were eliminated during the quality control and 8155 dissociated cells, with 5416 cells from control group and 2739 from HFD one, were filtered out then. The median reads per cell were 3955 for control group and 3421 for HFD group, transcripts detected per cell in the two groups were 1364 and 1069 respectively.