Abstract
The interaction between platelet glycoprotein (GP) Ibα and von Willebrand Factor (VWF) is essential to initiate platelet deposition at sites of vascular injury and sustain platelet thrombus formation when the shear rate exceeds a threshold value. With human blood, the dependence of normal platelet adhesion and aggregation on VWF-GP Ibα function becomes evident at shear rates above 1,000 s−1. In the last several years, mouse models have been increasingly used to study the mechanisms of thrombus formation in circulating blood, and mice deficient in both VWF and GP Ibα have been generated. These animals offer the opportunity to evaluate whether the pathways of platelet adhesion and aggregation mediated by VWF and GP Ibα are equally important in mouse and human blood as well as to define the threshold shear rate at which the function of these pathways may become essential in the mouse circulation. To address this issue, we used an ex vivo perfusion system using fibrillar collagen type I as the thrombogenic surface and a flow chamber in which the shear rate varied according to a predictable function from the inlet to the outlet in relation to the x,y position in the flow path. Thus, wall shear rates between 5,000 at the inlet and 0 s−1 at the outlet could be evaluated in a single experiment, allowing a precise definition of the threshold at which platelet deposition on the surface could initiate. In these studies we used wild type control animals (WT), mice deficient in VWF (VWF-KO) and mice in which most of the extracellular domain of GP Ibα was replaced by a domain of the human interleukin 4 receptor (GPIb-KO/IL-4R). In the latter case, the ligand binding function of GP Ibα was obliterated, but unlike in GP Ib-KO mice platelet morphology and count were essentially normal. Blood was obtained from the retroorbital vein plexus and contained 100 u/ml heparin as an anticoagulant. Experiments were recorded in real time for the visualization of platelet-surface contacts and confocal videomicroscopy was used for the direct measurement of platelet thrombus volume. With normal mouse blood, platelet formed large thrombi throughout the tested range of shear rates. In contrast, with VWF-KO and GPIb-KO/IL-4R blood, thrombus volume was less than 5% of normal at 5,000 s−1, approximately 50% of normal at 3,000 s−1, but entirely normal at 1,500 s−1. Essentially the same results were observed when the extracellular matrix of mouse fibroblasts, which may better represent the complex thrombogenic properties of the vascular wall, was used as a reactive substrate instead of isolated collagen type I. The different threshold shear rate at which VWF and GP Ibα function are essential for thrombus formation with human and mouse platelets may be explained by the smaller size of the latter, which consequently are subjected to a lower drag at equivalent shear rate levels. Moreover, the similar behavior of VWF-KO and GPIb-KO/IL-4R platelets suggests that, under the conditions of these studies, VWF binding is the predominant GP Ibα function required for normal platelet thrombus formation at high shear rates. The present results should allow a more critical evaluation of the findings derived from mouse models of hemostasis and thrombosis.
Platelet adhesion to cyanogen-bromide fragments of collagen alpha 1(I) under flow conditions