Revisiting endothelial tropism of SARS-CoV-2 using a cell-specific hACE2 mouse model

Severe COVID-19 is frequently associated with vascular complications, raising ongoing debate about whether SARS-CoV-2 can directly infect endothelial cells and thereby contribute to disease pathogenesis. Although endothelial cells express angiotensin-converting enzyme 2 (ACE2), the in vivo relevance of endothelial-restricted viral tropism remains unclear. To directly assess the consequences of endothelial-restricted SARS-CoV-2 tropism in vivo, we generated a transgenic mouse model expressing human ACE2 under control of the endothelial-specific Cdh5 promoter (Cdh5-hACE2). Despite confirmed pulmonary endothelial expression and protein presence of hACE2, SARS-CoV-2 infection of Cdh5-hACE2 mice did not induce clinical illness, detectable viral replication, immune cell influx in the lung, or histopathological abnormalities in the lung or brain. These findings indicate that endothelial-restricted SARS-CoV-2 tropism alone is insufficient to drive productive infection and clinical disease in vivo, suggesting that endothelial involvement in COVID-19 likely arises in the context of broader cellular infection or systemic host responses rather than from primary endothelial infection.

Although SARS-CoV-2 primarily infects the upper and lower airways, COVID-19 was quickly recognized as a multi-organ disease, in which vascular complications are a recurring feature. This has raised the possibility that direct infection of endothelial cells contributes to disease pathogenesis. However, whether vascular injury arises from productive endothelial infection or instead represents a secondary consequence of systemic inflammation remains unresolved. To directly disentangle these possibilities and define the in vivo consequences of endothelial-restricted viral tropism, we generated a transgenic mouse model expressing human ACE2 under the control of the endothelial-specific Cdh5 promoter (Cdh5-hACE2).