Von Willebrand disease (VWD) is the most common inherited bleeding disorder, affecting 1 to 3% of the U.S. population. It is caused by defects in expression and processing of Von Willebrand factor (VWF), a blood glycoprotein required for normal hemostasis, that mediates the adhesion of platelets to sites of vascular damage by binding to specific platelet glycoproteins and to constituents of exposed connective tissue. Since only 1 to 5% of normal levels of VWF are sufficient to ameliorate the bleeding phenotype, VWF represents an excellent target for therapy using gene transfer. However, since the normal sites of VWF production, endothelial cells, megakaryocytes and platelets, are difficult to target with traditional gene transfer methodologies, we are investigating the strategy of using gene transfer to the liver. However, while the liver is a more easily transduced site for VWF production, it is not known if the liver is capable of the complex post-translational modifications of the VWF protein. The study employed a murine knockout model of VWD created by disrupting the naturally occurring VWF gene (VWF-KO mice) which exhibit defects in hemostasis with a highly prolonged bleeding time and spontaneous bleeding events closely mimicking severe human VWD. Tail wounds in untreated VWF-KO mice will bleed to death unless cauterized, whereas similar wounds in WT mice stop bleeding in 2 to 3 min. To investigate the impact of VWF gene transfer on this phenotype, a plasmid containing the complete murine VWF cDNA directed by a cytomegalovirus immediate/early promoter/enhancer was delivered to the VWF-KO mice via hydrodynamic tail vein injection. An identical plasmid containing the cDNA for green fluorescent protein (GFP) or saline were used as controls. Immunohistochemical analysis of VWF in liver from the VWF cDNA administered animals indicated that VWF was expressed in hepatocytes, and possibly the liver endothelium. Bleeding time after wounding was determined 24, 48 or 72 hr after transfer of 10 to 250 g of plasmid, Bleeding time was normalized by 48 hr after delivery of 50 g or more of VWF cDNA, but not by GFP cDNA. Western analysis of plasma from animals receiving VWF cDNA, but not GFP cDNA, revealed high-molecular weight multimers similar to the pattern observed in wild type mice. Circulating Factor VIII in untreated VWF-KO mice was 10% of the values seen in WT mice, owing to the protective carrier function of VWF for the factor. In contrast, VWF-KO animals receiving 100 to 250 g of the VWF plasmid, but not GFP or saline, had factor VIII levels 2- to 3-fold above normal WT values, indicating restoration of carrier function for VWF in the treated mice. These data suggest that it may be possible to correct VWD using gene therapy delivered to tissues other than the naturally occurring sites of VWF manufacture, and that the liver is capable of expressing, processing and secreting VWF in a physiologically active form.