The Role of the VWF A1 and A2 Domains in Murine Models of TTP and Thrombosis.

Abstract 3060

Poster Board II-1036

von Willebrand Factor (VWF) is a large multimeric plasma glycoprotein synthesized in endothelial cells and megakaryocytes. In humans and mice, VWF dysfunction is associated both with defects in hemostasis, and with the systemic blood clotting disease thrombotic thrombocytopenic purpura (TTP). The initial adhesion of platelets to sites of vascular injury in large part involves binding of the VWF A1 domain to the platelet glycoprotein receptor GPIb alpha. This VWF A1-GPIb alpha interaction, along with deficiency of the ADAMTS13 plasma metalloprotease, is thought to be required for the pathogenesis of TTP. Deficiency of ADAMTS13 results in the failure to cleave the Y1605-M1606 sissile bond within the A2 domain of VWF. The structure of VWF is strongly influenced by its high content of cysteine residues, all of which are involved in inter-or intra-chain disulfide bonds. The location of these cysteine residues within the A domains leads to the formation of disulfide loops within the A1 (Cys 1272-1458) and A3 (Cys 1686-1872) domains, but not within the A2 domain. The lack of a disulfide loop allows the A2 domain to assume a “flexible” conformation that is thought to “open” in response to fluid shear stress, exposing the Y1605-M1606 bond to cleavage by ADAMTS13. To investigate specifically the role of the VWF A1-Gp1b alpha interaction in the context of otherwise functional VWF in vivo, we generated a chimeric murine VWF expression construct in which the murine A1 domain sequence is replaced with the corresponding sequence from human VWF (the human VWF A1 domain is known to not interact appreciably with murine GPIb alpha). Additionally, we engineered a VWF construct in which paired cysteine residues analogous to those in the A1 and A3 domains were introduced into the A2 domain sequence, with the goal being to “lock” the A2 domain closed and prevent cleavage by ADAMTS13. Hydrodynamic tail vein injection of both the VWF-hA1 and the VWF-A2 lock constructs into VWF-deficient mice resulted in plasma VWF levels up to 20-fold higher than observed in wild-type mice, dependent on the amount of plasmid injected. Importantly, the degree of VWF multimerization appeared nearly identical both to that observed in wild-type mice, and to mice injected with wild-type murine VWF, and expression persisted for approximately 30 days. Functionally, unlike WT murine VWF, expression of VWF-hA1 failed to restore thrombus formation in a ferric chloride-induced injury model, demonstrating the crucial importance of the VWF A1-GP1b apha interaction in thrombus formation. Currently we are investigating whether expression of VWF-hA1 can support disease pathogenesis in a mouse model of TTP. Similarly, we are determining whether expression of VWF-A2 lock leads to development of TTP, even in the presence of ADAMTS13. The ultimate goal of these studies is to completely “humanize” the VWF A1-GP1b alpha interaction in mice by replacing the murine GP1b alpha sequence with that from humans. These resulting animals will be used to further investigate the role of the VWF A1 domain-GPIb alpha interaction in vivo, and should prove useful for identifying compounds to effectively inhibit this interaction in humans. In addition, the expression of a VWF construct that is unable to be cleaved by ADAMTS13 should help to elucidate the role of VWF cleavage in TTP pathogenesis.