Abstract Body: Objective: Comprehensive investigation of endothelial cell (EC) dysfunction during atherosclerosis with single-cell ‘omics’ has resulted in the proposal that ECs undergo multiple alternative cell fate decisions during disease, but lack of lineage tracing and/or spatial localization complicates interpretation of these data. TWIST1, a causal gene for multiple atherosclerotic vascular diseases, is activated with disturbed flow in endothelial cells (ECs), and EC-Twist1 knockout results in reduced atherosclerosis. However, it remains unclear how Twist1 affects EC phenotype and plaque biology.

Approach and Results: We performed EC lineage tracing, in situ analysis and scRNA-Seq in ApoE-/- mice, both prior to disease and after 16 weeks of high-fat diet. We also performed these studies with two mouse models of EC Twist1 deletion. We overexpressed TWIST1 in human coronary artery endothelial cells (HCAECs) exposed to different flow conditions, followed by bulk RNAseq. Human scRNA-Seq data were used to validate key findings in the mouse model.
We found that EC phenotypic modulation during atherosclerosis is characterized by both pro-inflammatory and EndMT gene programs, occurring simultaneously along a single cell fate transition. Human scRNA-Seq data validated a similar EndMT transition during disease. We found that the commonly-used Twist1 conditional allele is hypomorphic, leading to reduced Twist1 expression in multiple cell types. Using a mouse model of EC-specific Twist1 deletion, we found reduced EC phenotypic modulation, decreased lesion size, and a more stable lesion phenotype. Integration of TWIST1 overexpression in HCAECs with the mouse scRNAseq data identified specific TWIST1-induced targets including CXCL12 and E-selectin during EC phenotypic modulation.

Conclusion: Our study revealed important aspects of EC phenotypic modulation during atherosclerosis, unifying disparate observations in the field. We identified key cellular and molecular mechanisms underlying a top risk locus for multiple vascular diseases, highlighting the promotion of inflammatory EndMT by TWIST1 as a key driver of disease risk.