The temporal response and mechanism of action of tranexamic acid in endothelial glycocalyx degradation.

Abstract

The endothelial glycocalyx (GCX) plays an important role in vascular barrier function. Damage to the GCX occurs due to a variety of causes including hypoxia, ischemia-reperfusion, stress-related sympathoadrenal activation, and inflammation. Tranexamic acid (TXA) may prevent GCX degradation. The therapeutic window for TXA administration and the mechanism of action has been under review. Membrane-anchored proteases (sheddases) are key components in endothelial cell biology including the regulation of vascular permeability. The effect of TXA administration on stress-related GCX damage, and the role of sheddases in this process was studied in a cell-based model. Confluent human umbilical vein endothelial cells (HUVEC) were exposed to hydrogen peroxide and/or epinephrine (EPI) to stimulate postshock reperfusion. TXA was added at various times after hydrogen peroxide (H2O2) and/or EPI exposure. GCX degradation was indexed by syndecan-1 and hyaluronic acid release. Activation of endothelial sheddases was indexed by A Disintegrin and Metalloproteinase-17 and matrix metalloproteinase-9 activity in culture supernatants. Exposure of HUVEC to either/both EPI and H2O2 resulted in a cellular stress and GCX disruption demonstrated by increased levels of syndecan-1 shedding, hyaluronic acid release, tumor necrosis factor-α release. Shedding of these GCX components was associated with increased activity of both A Disintegrin and Metalloproteinase-17 and matrix metalloproteinase. Disruption of the GCX was further demonstrated via fluorescent imaging, which demonstrated disruption after exposure to either/both H2O2 and EPI. Early administration of either TXA or doxycycline resulted in preservation of the GCX. Late administration of TXA had no effect, whereas doxycycline had some residual protective effect. Tranexamic acid as a serine protease inhibitor prevented GCX degradation via inhibition of endothelial sheddase activation. This effect was not apparent when TXA was administered greater than 60 minutes after "simulated" reperfusion. Our study supports the clinical practice of early TXA administration in the severely injured patient.