Abstract
Abstract 372
The dependence of procoagulant activity on phosphatidylserine (PS) has been recognized for at least four decades but the location of physiologically relevant membranes with PS exposure remains uncertain. PS is exposed on apoptotic cells and cell microparticles but in vitro and in vivo studies have failed to demonstrate a clear relationship of microparticles or apoptotic cells to fibrin deposition. Exposure of endothelial cells to stimulants or toxins leads to retraction of cell margins, mounding of the central cell, and extension of filopodia. We have also found that cell stress also leads to limited, focal PS exposure. Furthermore, we found that binding sites for lactadherin, a PS-binding protein that shares homology with factor VIII and factor V, are concentrated on convex surfaces such as filopodia. In this study we ask whether the limited, focal PS exposure on stressed human umbilical vein endothelial cells is sufficient to support prothrombinase complex assembly and whether the prothrombinase complex assembly is restricted to the convex membrane features that bind lactadherin.
We allowed Human Umbilical Vein Endothelial Cells (HUVEC) to grow to confluent monolayers prior to exposure to TNF-α, 10 ng/ml, for 5–24 hours. PS exposure was detected by simultaneous staining using 10 nM lactadherin–Alexa 488 and annexin V–Cy 3.18, both exhibiting high affinity for PS. Stressed cells withdrew from their prior borders, leaving residual fibrils connected to original attachment points. In addition, they extended filopodia that were up to several cell diameters in length. Confocal microscopy demonstrated focal staining of filopodia, fibrils and cell margins with lactadherin and patches near the nucleus with annexin A5. We asked whether the selective binding might be determined by the membrane topology. To mimic the curvature of a cell membrane we prepared nano-fabricated silica substrates with ridge radii of 10 nm. The AFM topographic and fluorescent images of synthetic membrane bilayers supported by the substrates showed that, over a PS content of 4–15%, lactadherin preferentially binds to the convex nano-ridges with a ridge: valley staining ratio >80:1, while annexin V selectively binds the concave areas of the nano-trenches with a ridge. Combined fluorescence/AFM imaging of TNF-α treated HUVEC's, demonstrated that the new thin filaments staining with lactadherin had radii of curvature of approx. 12 nm, similar to the ridges of our synthetic bilayers. We asked whether factor Va and factor Xa share preference for convex surfaces, analogous to lactadherin. Supported membranes of 4% PS had preferential ridge staining by factor Va-fluorescein-maleimide with a ridge/valley ratio > 10/1. Co-staining with factor Va and factor Xa-EGRck-biotin (complexed to Alexa 647-steptavidin) indicated that factor Va enhanced binding of factor Xa to ridges, thus the prothrombinase complex has highly preferential binding to convex ridges. TNF-α-treated endothelial cells bound factor Va, like lactadherin, selectively on filopodia and fibrils near the retracted edges of endothelial cells. Factor Xa also localized to these features in the presence of factor Va, indicating prothrombinase complex assembly. Stressed endothelial cells exhibited at least 8-fold higher support for thrombin production and prothrombinase activity. Prothrombinase activity was efficiently inhibited by lactadherin, demonstrating that the lactadherin-binding sites were the functional sites for prothrombinase activity. Together, these data indicate that stressed endothelial cells can support the prothrombinase complex and that prothrombinase activity is compartmentalized near the periphery of the cell and in the intracellular area through binding sites on highly convex membrane features with exposed PS. We have hypothesized that this compartment of procoagulant activity is relatively protected from anti-coagulant proteins that are localized elsewhere on the stimulated/stressed endothelial cell.
Disclosures:
No relevant conflicts of interest to declare.
Hsp90–Sgt1 and Skp1 target human Mis12 complexes to ensure efficient formation of kinetochore–microtubule binding sites