Protein S is a single chain (Mr 75.000) plasma protein. It is a cofactor to activated protein C (APC) in the regulation of coagulation factors Va and Villa. It has high affinity for negatively charged phospolipids and it forms a 1:1 complex with APC on phospholipid surfaces, platelets and on endothelial cells. Patients with heterozygous protein S deficiency have a high incidence of thrombosis. Protein S is cleaved by thrombin, which leads to a loss of calcium binding sites and of APC cofactor activity. Protein S has two to three high affinity (KD 20uM) calcium binding sites - unrelated to the Gla-region - that are unaffected by the thrombin cleavage. In human plasma protein S (25 mg/liter) circulates in two forms; free (approx. 40%) and in a 1:1 noncovalent complex (KD 1× 10-7M) with the complement protein C4b-binding protein (C4BP). C4BP (Mr 570.000) is composed of seven identical 70 kDa subunits that are linked by disulfide bonds. When visualized by electron microscopy, C4BP has a spiderlike structure with the single protein S binding site located close to the central core and one C4b-binding site on each of the seven tentacles. When bound to C4BP, protein S looses its APC cofactor activity, whereas the function-of C4BP is not directly affected by the protein S binding. Chymotrypsin cleaves each of the seven C4BP subunits close to the central core which results in the liberation of multiple 48 kDa “tentacte” fragments and the formation of a 160 kDa central core fragment. We have successfully isolated a 160 kDa central core fragment with essentially intact protein S binding ability.The primary structure of both bovine and human protein S has been determined and found to contain 635 and 634 amino acids, respectively, with 82 % homology to each other. Four different regions were distinguished; the N-terminal Gla-domain (position 1-45) was followed by a region which has two thrombin-sensitive bonds positioned within a disulfide loop. Position 76 to 244 was occupied by four repeats homologous to the epidermal growth factor (EGF) precursor. In the first EGF-domain a modified aspartic acid was identified at position 95, B-hydroxaspartic acid (Hya), and in corresponding positions in the three following EGF-domains (positions 136,178 and 217) we found B-hydroxyasparagine (Hyn). Hyn has not previously been identified in proteins. The C-terminal half of protein S (from position 245) shows no homology to the serine proteases but instead to human Sexual Hormon Binding Globulin (SHBG)(see separate abstract). To study the structure-function relationship we made eighteen monoclonal antibodies to human protein S. The effects of the monoclonals on the C4BP-protein S interaction and on the APC cofactor activity were analysed. Eight of the antibodies were calciumdependent, four of these were against the Gla-domain, two against the thrombin sensitive portion and two against the region bearing the high affinity calcium binding sites. Three of the monoclonals were dependent on the presence of chelating agents, EDTA or EGTA, and were probably directed against the high affinity calcium binding region. Three other monoclonals inhibited the protein S-C4BP interaction. At present, efforts are made to localize the epitopes to gain information about functionally important regions of protein S.
Characterization of calcium-binding sites in development-specific protein S of Myxococcus xanthus using site-specific mutagenesis.
