Investigating the Role of the Vascular Endothelium in Insulin Delivery

Introduction Insulin is produced in the pancreas and circulates in the blood before crossing the vascular endothelium to gain access to target tissues. The delivery of insulin from the blood to tissues occurs in two stages: first, blood carrying insulin must perfuse the capillary beds. In arteries, insulin stimulates the endothelial cells to produce nitric oxide, which induces vasodilation of the surrounding vascular smooth muscle to increase blood flow. This in turn recruits capillaries that irrigate peripheral tissues. Second, insulin must cross the endothelial monolayer in order to reach the surrounding tissue (e.g. smooth or striated muscle). Classic studies demonstrated that insulin delivery is rate‐limiting to its metabolic action; notably, capillary recruitment and perfusion are compromised in insulin resistant states, thereby decreasing insulin access to the tissues. However, the contribution of insulin transfer across the endothelial layer to insulin action is unclear. In vivo studies cannot define this contribution due to confounding from vessel dilation or capillary recruitment, which contribute to perfusion. On the other hand, in vitro data from cultured endothelial cells suggest that insulin crosses endothelial monolayers by a saturable transport process (e.g. transcytosis). However, it is uncertain whether cells in culture adequately model normal physiology, as cultured endothelial cells are known to rapidly undergo phenotypic drift. Most of the in vitro work on insulin transcytosis has been performed using endothelial cells derived from large vessels such as the aorta. As mentioned, smooth muscle cells respond to insulin stimulation but how insulin crosses the aortic endothelium is unclear. Rationale Because of inherent limitations of both in vivo and in vitro studies, whether the endothelium constitutes a barrier to insulin and restricts its delivery is unknown. Objective To evaluate the contribution of the endothelial barrier to insulin delivery independent of confounding by hemodynamic forces, through an ex vivo perfusion assay. Methods Murine aortas were isolated and segments perfused ex vivo with insulin alone or in the presence of inducers of endothelial barrier dysfunction (histamine or platelet activating factor). Insulin action in vascular smooth muscle was measured by immunoblotting for Akt phosphorylation; blots were normalized to total Akt and to smooth‐muscle actin. The effectiveness of histamine/PAF to induce endothelial leak was established by perfusion with 70kDa dextran. Results The ex vivo aorta responded to insulin in a time‐ and dose‐dependent manner; insulin (10nM) perfusion elicited significant increases in Akt phosphorylation after 5 min (1.4 fold) that were maintained at 30 min (1.3 fold). Perfusion with histamine and PAF induced rapid endothelial leak, demonstrated by accumulation of dextran in the vascular intima; however there was no change in the kinetics of insulin action. Aortic perfusion with 100 nM insulin showed similar results. Conclusions Our results suggest that the endothelium is not rate‐limiting to insulin delivery in large vessels, contrary to existing literature. We are currently using similar approaches to study the microvascular endothelium. Support or Funding Information This work was supported by Canadian Institutes of Health Research Grant MOP‐130493 to A. Klip and W. Lee. V. Tokarz was supported by a graduate studentship from the Banting and Best Diabetes Centre and a Queen Elizabeth II Graduate Studentship in Science and Technology from the University of Toronto.