In vivo behavior of factor ix and certain gla domain mutants evaluated in coagulation factor k-deficient mice

To investigate the role of certain y-carboxyglutamic acid (gla) domain residues that mediate factor IX endothelium/collagen IV binding, we compared the clearance and distribution of molecules containing mutations at residue K5 (to A or R, K5A, K5R) to that of wild-type recombinant and plasma purified human factor IX. Human proteins were injected into hemophilia B mice that have no factor IX. We determined the plasma and liver concentrations of factor IX with enzyme-linked immunoassays. In addition, we used immunohistochemical techniques to examine the location of factor IX in crosssections of the mouse aorta. All proteins showed a time-dependent disappearance from blood that fits a two exponential equation, indicating at least two pathways are involved in clearance. Even when we took the first sample at 30 seconds, the highest factor IX concentration was always found at the first time point. Thus, the distribution phase is fast relative to sampling times. The factor IX's that we previously showed bind to endothelium/collagen IV (K5R, normal plasma, and wild-type) were removed from the blood faster than those that we showed did not bind (K5A). Less than 1 % of factor IX was detected associated with the cellular fraction of blood. In mice treated with the K5R or normal factor IX, about 80% of the injected factor IX appeared in the liver after 2 minutes, while 20% remained in the circulation. In contrast, in mice injected with K5A only about 60% was found in liver and 30% in blood. When we blocked circulation to the liver by tying off the hepatic artery and portal vein, 70% of factor IX was still in the plasma after 2 minutes. Immunohistochemical studies of arteries from mice injected with the various factor IX's indicated that only those factor IX's shown to bind endothelium/ collagen IV in vitro, appear bound to the endothelial surface of the arteries. Based on our results we conclude the following: 1) binding to endothelium/collagen IV plays a role in controlling factor IX concentrations in vivo, 2) liver clearance of factor IX is not primarily mediated by the endothelium/collagen IV, and 3) mutants that do not bind endothelium/ collagen IV, and thus, remain in circulation longer, such as K5A, may be candidates for gene therapy and replacement therapy in patients with hemophilia B.

Duke Scholars

Author Maureane Hoffman Pathology