Distinct effects of sevoflurane and propofol on vascular permeability in vitro and in a mouse model.

General anesthetics may substantially influence endothelium function, potentially affecting outcomes of surgical patients, but their effects are unclear. Here, we studied a commonly used inhaled anesthetic, sevoflurane, and an intravenous anesthetic, propofol, on vascular endothelial permeability using multiple in vitro assays and a mouse model. Human umbilical vein endothelial cells (HUVECs) and mouse pulmonary endothelial cells (MPECs) were used for in vitro models to test the effect of anesthetics on endothelial permeability. The effect of anesthetics on pulmonary vascular leakage was analyzed using AngioSense® 750 fluorescent tracer and rhodamine-labeled 3-kD dextran in a mouse model. Downstream targets were identified using RNA sequencing and confirmed by qRT-PCR and western blot. Sevoflurane at clinically relevant concentrations disrupted the endothelial monolayer formed by HUVECs and MPECs in transwell permeability models. Sevoflurane, but not propofol, induced 1.8-fold increase of AngioSense® dye accumulation in mouse lung than control, indicating pulmonary vascular leakage in sevoflurane group. RNA sequencing analysis, qRT-PCR, and western blot analysis revealed that sevoflurane induced the expression and activation of hypoxic-induced factor 1α (HIF-1α) in vitro and in vivo. The activation of HIF-1α led to the increased expression of its downstream vascular endothelial growth factor (VEGF). The knockdown of HIF-1α restored the change of endothelial permeability and abolished the increase of VEGF induced by sevoflurane in MPECs. Our results demonstrate that sevoflurane increased endothelial and pulmonary vascular permeability via HIF-1α and VEGF. Propofol had no significant effect on the permeability of endothelium.