Protein-protein interactions will be probed in the thrombomodulin- thrombin complex. The binding of thrombomodulin to thrombin causes an inhibition of the procoagulant activity of thrombin (cleavage of fibrinogen) and a selective enhancement of the anticoagulant activity of thrombin (activation of protein C). Previous studies have shown that a region of thrombomodulin containing three domains that resemble epidermal growth factor (EGF domains) are necessary and sufficient to bind to thrombin and to alter thrombin's catalytic activity. Our experiments will determine the molecular details of the interaction of the thrombomodulin EGF domains with thrombin, and of the concommitant activation of protein C. The fourth, fifth, and sixth EGF domains from thrombomodulin that have been implicated in the binding and activation of thrombin cleavage of protein C will be expressed in E coli, and large amounts of the purified domains will be obtained. We will study the function of each of these domains and pairs of domains by assaying their ability to either inhibit the thrombomodulin-thrombin complexation, or to activate thrombin to catalyze the cleavage of protein C. Thrombomodulin has been shown to contain a beta-hydroxylated aspartic acid or asparagine residue, but the location of this modified amino acid is unknown. Based on sequence similarity, it is likely that the modified residue is hydroxylated asparagine in the sixth EGF domain of thrombomodulin. We will purify bovine thrombomodulin, isolate the fifth and sixth EGF domains following limited proteolysis, and determine the location of the modified amino acid. If we find a hyroxylated asparagine, we will probe its function by synthesizing the domain containing beta-hydroxyasparagine and comparing its activity to that of the unhydroxylated domain. Our studies will concentrate on the possibility that the hydroxyl group mediates calcium-binding. The solution structures of the functionally important domains will be determined by two-dimensional NMR methods. We will then carry out measurements of transferred nuclear Overhauser effects to protons on thrombin to determine the molecular details of the interaction of the thrombomodulin EGF domains with thrombin. We will also try to co- crystallize the domains with thrombin to gain structural information about the protein-protein interaction. Information from sequence similarity and analogy to the interaction of epidermal growth factor with its receptor suggests the two beta-turn regions are involved in the interaction. This hypothesis will be probed using site-directed mutagenesis.
