Enhancement of human protein C function by site-directed mutagenesis of the gamma-carboxyglutamic acid domain

C-terminal residues of activated protein C light chain contribute to its anticoagulant and cytoprotective activities

BACKGROUND

Activated protein C (APC) is an important homeostatic blood coagulation protease that conveys anticoagulant and cytoprotective activities. Proteolytic inactivation of factors Va and VIIIa facilitated by cofactor protein S is responsible for APC's anticoagulant effects, whereas cytoprotective effects of APC involve primarily the endothelial protein C receptor (EPCR), protease activated receptor (PAR)1 and PAR3.

OBJECTIVE

To date, several binding exosites in the protease domain of APC have been identified that contribute to APC's interaction with its substrates but potential contributions of the C-terminus of the light chain have not been studied in detail.

METHODS

Site-directed Ala-scanning mutagenesis of six positively charged residues within G142-L155 was used to characterize their contributions to APC's anticoagulant and cytoprotective activities.

RESULTS AND CONCLUSIONS

K151 was involved in protein S dependent-anticoagulant activity of APC with some contribution of K150. 3D structural analysis supported that these two residues were exposed in an extended protein S binding site on one face of APC. Both K150 and K151 were important for PAR1 and PAR3 cleavage by APC, suggesting that this region may also mediate interactions with PARs. Accordingly, APC's cytoprotective activity as determined by endothelial barrier protection was impaired by Ala substitutions of these residues. Thus, both K150 and K151 are involved in APC's anticoagulant and cytoprotective activities. The differential contribution of K150 relative to K151 for protein S-dependent anticoagulant activity and PAR cleavage highlights that binding exosites for protein S binding and for PAR cleavage in the C-terminal region of APC's light chain overlap.

The roles of factor Va and protein S in formation of the activated protein C/protein S/factor Va inactivation complex

BACKGROUND

Activated protein C (APC)-mediated inactivation of factor (F)Va is greatly enhanced by protein S. For inactivation to occur, a trimolecular complex among FVa, APC, and protein S must form on the phospholipid membrane. However, direct demonstration of complex formation has proven elusive.

OBJECTIVES

To elucidate the nature of the phospholipid-dependent interactions among APC, protein S, and FVa.

METHODS

We evaluated binding of active site blocked APC to phospholipid-coated magnetic beads in the presence and absence of protein S and/or FVa. The importance of protein S and FV residues were evaluated functionally.

RESULTS

Activated protein C alone bound weakly to phospholipids. Protein S mildly enhanced APC binding to phospholipid surfaces, whereas FVa did not. However, FVa together with protein S enhanced APC binding (>14-fold), demonstrating formation of an APC/protein S/FVa complex. C4b binding protein-bound protein S failed to enhance APC binding, agreeing with its reduced APC cofactor function. Protein S variants (E36A and D95A) with reduced APC cofactor function exhibited essentially normal augmentation of APC binding to phospholipids, but diminished APC/protein S/FVa complex formation, suggesting involvement in interactions dependent upon FVa. Similarly, FVaNara (W1920R), an APC-resistant FV variant, also did not efficiently incorporate into the trimolecular complex as efficiently as wild-type FVa. FVa inactivation assays suggested that the mutation impairs its affinity for phospholipid membranes and with protein S within the complex.

CONCLUSIONS

FVa plays a central role in the formation of its inactivation complex. Furthermore, membrane proximal interactions among FVa, APC, and protein S are essential for its cofactor function.

Activated protein C light chain provides an extended binding surface for its anticoagulant cofactor, protein S

Protein S anticoagulant cofactor sensitivity and PAR1 cleavage activity were assayed for 9 recombinant APC mutants.Residues L38, K43, I73, F95, and W115 on one face of the APC light chain define an extended surface containing the protein S binding site.

Structural and functional studies of γ-carboxyglutamic acid domains of factor VIIa and activated Protein C: role of magnesium at physiological calcium

Crystal structures of factor (F) VIIa/soluble tissue factor (TF), obtained under high Mg(2+) (50mM Mg(2+)/5mM Ca(2+)), have three of seven Ca(2+) sites in the γ-carboxyglutamic acid (Gla) domain replaced by Mg(2+) at positions 1, 4, and 7. We now report structures under low Mg(2+) (2.5mM Mg(2+)/5mM Ca(2+)) as well as under high Ca(2+) (5mM Mg(2+)/45 mM Ca(2+)). Under low Mg(2+), four Ca(2+) and three Mg(2+) occupy the same positions as in high-Mg(2+) structures. Conversely, under low Mg(2+), reexamination of the structure of Gla domain of activated Protein C (APC) complexed with soluble endothelial Protein C receptor (sEPCR) has position 4 occupied by Ca(2+) and positions 1 and 7 by Mg(2+). Nonetheless, in direct binding experiments, Mg(2+) replaced three Ca(2+) sites in the unliganded Protein C or APC. Further, the high-Ca(2+) condition was necessary to replace Mg4 in the FVIIa/soluble TF structure. In biological studies, Mg(2+) enhanced phospholipid binding to FVIIa and APC at physiological Ca(2+). Additionally, Mg(2+) potentiated phospholipid-dependent activations of FIX and FX by FVIIa/TF and inactivation of activated factor V by APC. Since APC and FVIIa bind to sEPCR involving similar interactions, we conclude that under the low-Mg(2+) condition, sEPCR binding to APC-Gla (or FVIIa-Gla) replaces Mg4 by Ca4 with an attendant conformational change in the Gla domain ω-loop. Moreover, since phospholipid and sEPCR bind to FVIIa or APC via the ω-loop, we predict that phospholipid binding also induces the functional Ca4 conformation in this loop. Cumulatively, the data illustrate that Mg(2+) and Ca(2+) act in concert to promote coagulation and anticoagulation.

Inhibition of thrombin formation by active site mutated (S360A) activated protein C

Activated protein C (APC) down-regulates thrombin formation through proteolytic inactivation of factor Va (FVa) by cleavage at Arg(506) and Arg(306) and of factor VIIIa (FVIIIa) by cleavage at Arg(336) and Arg(562). To study substrate recognition by APC, active site-mutated APC (APC(S360A)) was used, which lacks proteolytic activity but exhibits anticoagulant activity. Experiments in model systems and in plasma show that APC(S360A), and not its zymogen protein C(S360A), expresses anticoagulant activities by competing with activated coagulation factors X and IX for binding to FVa and FVIIIa, respectively. APC(S360A) bound to FVa with a K(D) of 0.11 +/- 0.05 nm and competed with active site-labeled Oregon Green activated coagulation factor X for binding to FVa. The binding of APC(S360A) to FVa was not affected by protein S but was inhibited by prothrombin. APC(S360A) binding to FVa was critically dependent upon the presence of Arg(506) and not Arg(306) and additionally required an active site accessible to substrates. Inhibition of FVIIIa activity by APC(S360A) was >100-fold less efficient than inhibition of FVa. Our results show that despite exosite interactions near the Arg(506) cleavage site, binding of APC(S360A) to FVa is almost completely dependent on Arg(506) interacting with APC(S360A) to form a nonproductive Michaelis complex. Because docking of APC to FVa and FVIIIa constitutes the first step in the inactivation of the cofactors, we hypothesize that the observed anticoagulant activity may be important for in vivo regulation of thrombin formation.

Mouse recombinant protein C variants with enhanced membrane affinity and hyper-anticoagulant activity in mouse plasma

Mouse anticoagulant protein C (461 residues) shares 69% sequence identity with its human ortholog. Interspecies experiments suggest that there is an incompatibility between mouse and human protein C, such that human protein C does not function efficiently in mouse plasma, nor does mouse protein C function efficiently in human plasma. Previously, we described a series of human activated protein C (APC) Gla domain mutants (e.g. QGNSEDY-APC), with enhanced membrane affinity that also served as superior anticoagulants. To characterize these Gla mutants further in mouse models of diseases, the analogous mutations were now made in mouse protein C. In total, seven mutants (mutated at one or more of positions P(10)S(12)D(23)Q(32)N(33)) and wild-type protein C were expressed and purified to homogeneity. In a surface plasmon resonance-based membrane-binding assay, several high affinity protein C mutants were identified. In Ca(2+) titration experiments, the high affinity variants had a significantly reduced (four-fold) Ca(2+) requirement for half-maximum binding. In a tissue factor-initiated thrombin generation assay using mouse plasma, all mouse APC variants, including wild-type, could completely inhibit thrombin generation; however, one of the variants denoted mutant III (P10Q/S12N/D23S/Q32E/N33D) was found to be a 30- to 50-fold better anticoagulant compared to the wild-type protein. This mouse APC variant will be attractive to use in mouse models aiming to elucidate the in vivo effects of APC variants with enhanced anticoagulant activity.

Functional properties and active-site topographies of factor X Gla- and prothrombin Gla-domain chimeras of activated protein C

Substitution of the Gla-domain of activated protein C (APC) with the Gla-domain of prothrombin (APC-PTGla) improves the anticoagulant activity of APC independent of protein S. Previous FRET studies showed that this substitution alters the active-site topography of this mutant, rendering it identical to the active site of the APC-protein S complex. In this study, we characterized the functional properties and the active-site topography of another APC chimera containing the Gla-domain of factor X (APC-FXGla). We discovered that the anticoagulant activity of this mutant was similarly improved independent of protein S. The average distance of the closest approach (L) between the donor dye fluorescein attached to the active site of APC derivatives and the acceptor dye octadecylrhodamine incorporated into PC/PS vesicles was determined to be 99 A for APC and 84-86 A for both APC-PTGla and APC-FXGla. Protein S minimally influenced the L values of the APC chimeras, however, it lowered this value to 87 A for wild-type APC. Further studies revealed that neither chimera elicits a protective signaling response in the TNF-alpha-activated endothelial cells. These results suggest that unique structural features within the Gla-domain of APC enable the protease to interact with endothelial protein C receptor in the antiinflammatory pathway, while the same features also cause an inherently lower specific activity for APC in the anticoagulant pathway. This adaptation has made APC a cofactor-dependent protease, requiring the cofactor function of protein S for its optimal anticoagulant function, which appears to involve the alteration of the active-site topography of APC above the membrane surface.

Antithrombotic and anticoagulant effects of wild type and Gla-domain mutated human activated protein C in rats

The antithrombotic and anticoagulant effects of recombinant wild type (WT) and mutated human activated protein C (hAPC) were investigated using a rat model of arterial thrombosis. Recent in vitro studies using human plasma have shown enhanced anticoagulant effects of hAPC by mutagenesis of either loop 148 in the serine protease domain or of the Gla domain. The Gla-domain mutant QGNSEDY-hAPC (= H10Q/S11G/S12N/D23S/Q32E/N33D/H44Y) was found to be particularly active as an anticoagulant. We now combined the two mutations to create the variant QGNSEDY-hAPC:B148 and investigated the in vivo effects of this variant as well as of QGNSEDY-hAPC and WT hAPC using a rat model of arterial thrombosis. In vitro clotting experiments using rat plasma demonstrated WT hAPC to be inefficient, whereas both mutant hAPC variants yielded distinct dose dependent anticoagulant effects. In the arterial injury model, a segment of the left common carotid artery was opened longitudinally. An endarterectomy was performed and the arteriotomy was closed, whereafter the vessel was reperfused and the patency rate determined after 31 min. Three treatment groups each containing 10 rats and a control group of 20 animals were in a blind random fashion given intravenous bolus injections of 0.8 mg/kg WT or mutant hAPC or vehicle only. The ex vivo clotting times of plasma drawn 3 min after the injections, as compared to baseline clotting times, were approximately doubled by QGNSEDY-hAPC and tripled by QGNSEDY-hAPC:B148 infusions, while WT APC had little effect. Compared to the control group, none of the hAPC preparations had significant antithrombotic effect or increased arteriotomy bleeding.

Computational study of coagulation factor VIIa's affinity for phospholipid membranes

The interaction between the gamma-carboxyglutamic acid-rich domain of coagulation factor VIIa (FVIIa), a vitamin-K-dependent enzyme, and phospholipid membranes plays a major role in initiation of blood coagulation. However, despite a high sequence and structural similarity to the Gla domain of other vitamin-K-dependent enzymes with a high membrane affinity, its affinity for negatively charged phospholipids is poor. A few amino acid differences are responsible for this observation. Based on the X-ray structure of lysophosphatidylserine (lysoPS) bound to the Gla domain of bovine prothrombin (Prth), models of the Gla domain of wildtype FVIIa and mutated FVIIa Gla domains in complex with lysoPS were built. Molecular dynamics (MD) and steered molecular dynamics (SMD) simulations on the complexes were applied to investigate the significant difference in the binding affinity. The MD simulation approach provides a structural and dynamic support to the role of P10Q and K32E mutations in the improvement of the membrane contact. Hence, rotation of the Gly11 main chain generated during the MD simulation results in a hydrogen bond with Q10 side chain as well as the appearance of a hydrogen bond between E32 and Q10 forcing the loop harbouring Arg9 and Arg15 to shrink and thereby enhances the accessibility of the phospholipids to the calcium ions. Furthermore, the application of the SMD simulation method to dissociate C6-lysoPS from a series of Gla domain models exhibits a ranking of the rupture force that can be useful in the interpretation of the PS interaction with Gla domains. Finally, adiabatic mapping of Gla6 residue in FVIIa with or without insertion of Tyr4 confirms the critical role of the insertion on the conformation of the side chain Gla6 in FVIIa and the corresponding Gla7 in Prth.

Enhanced Rate of Cleavage at Arg-306 and Arg-506 in Coagulation Factor Va by Gla Domain-mutated Human-activated Protein C

A Gla domain-mutated protein C variant, QGNSEDY, modified at positions 10-12, 23, 32-33, and 44, having enhanced affinity for negatively charged phospholipid and increased anticoagulant potential, was used to elucidate the importance of the interaction between the Gla domain and the phospholipid for the ability of activated protein C (APC) to inactivate factor Va (FVa). FVa degradation by wild type (WT)-APC and QGNSEDY-APC yielded similar fragments on Western blotting; QGNSEDY-APC was, however, considerably more efficient. The kinetic parameters for individual APC-mediated cleavages in FVa, i.e. at Arg-306 and Arg-506, were investigated at high and low phospholipid concentrations in the presence and absence of protein S. FVa variants 306Q679Q and 506Q679Q, which can only be cleaved at Arg-506 and Arg-306, respectively, were used. In the absence of protein S, QGNSEDY-APC was 17.8- and 4-fold more efficient than WT-APC in cleaving at Arg-306 and Arg-506, respectively, at high phospholipid. Similar values were obtained at low phospholipid. In the presence of protein S, QGNSEDYAPC was 6.8- and 3.2-fold more active than WT-APC in cleaving at Arg-306 and Arg-506, respectively, at high phospholipid. At low phospholipid, the corresponding values were 14- and 6.5-fold. In conclusion, the modification of the Gla domain in QGNSEDY-APC yielded increased rates of cleavage at both sites in FVa, the increase being particularly pronounced for the Arg-306 site in the absence of protein S. The results obtained with QGNSEDY-APC provide insights into the importance of the APC-phospholipid interaction for the APC-mediated cleavages at Arg-306 and Arg-506 in FVa.

L-alpha-glycerylphosphorylcholine inhibits the transfer function of phosphatidylinositol transfer protein alpha

Phosphatidylinositol transfer protein alpha (PITP-alpha) is a bifunctional phospholipid transfer protein that is highly selective for phosphatidylinositol (PtdIns) and phosphatidylcholine (PtdCho). Polar lipid metabolites, including L-alpha-glycerylphosphorylcholine (GroPCho), increasingly have been linked to changes in cellular function and to disease. In this study, polar lipid metabolites of PtdIns and PtdCho were tested for their ability to influence PITP-alpha activity. GroPCho inhibited the ability of PITP-alpha to transfer PtdIns or PtdCho between liposomes. The IC(50) of both processes was dependent on membrane composition. D-myo-inositol 1-phosphate and glycerylphosphorylinositol modestly enhanced PITP-alpha-mediated phospholipid transfer. Choline, phosphorylcholine (PCho), CDP-choline, glyceryl-3-phosphate, myo-inositol and D-myo-inositol 1,4,5-trisphosphate had little effect. Membrane surface charge was a strong determinant of the GroPCho inhibition with the inhibition being greatest for highly anionic membranes. GroPCho was shown to enhance the binding of PITP-alpha to anionic vesicles. In membranes of low surface charge, phosphatidylethanolamine (PtdEtn) was a determinant enabling the GroPCho inhibition. Anionic charge and PtdEtn content appeared to increase the strength of PITP-alpha-membrane interactions. The GroPCho-enhanced PITP-alpha-membrane binding was sufficient to cause inhibition, but not sufficient to account for the extent of inhibition observed. Processes associated with strengthened PITP-alpha-membrane binding in the presence of GroPCho appeared to impair the phospholipid insertion/extraction process.

Gla domain-mutated human protein C exhibiting enhanced anticoagulant activity and increased phospholipid binding

Protein C is a member of the vitamin K- dependent protein family. Proteins in this family have similar gamma-carboxyglutamic acid (Gla)-rich domains, but their affinities for negatively charged phospholipid membranes vary more than 1000-fold. We have shown that it is possible to enhance anticoagulant activity and membrane affinity of protein C by selective mutagenesis of the Gla domain. In this study, 3 new mutants, Q10G11N12 (QGN), S23E32D33Y44 (SEDY), and Q10G11N12S23E32D33Y44 (QGNSEDY), were created. In plasma-based coagulation assays, the activated form of QGNSEDY (QGNSEDY-APC) demonstrated approximately 20-fold higher anticoagulant activity than wild-type activated protein C (WT APC), while QGN-APC and SEDY-APC did not. Both normal activated factor V (FVa) and FVa Leiden (Arg506Gln) were degraded much more efficiently by QGNSEDY-APC than by WT APC in the presence as well as in the absence of protein S. Binding of protein C variants to negatively charged phospholipid membranes was investigated using light scattering and the BIAcore technique. QGNSEDY demonstrated 3- to 7-fold enhanced binding as compared with WT protein C, suggesting the membrane affinity to be influenced by several residues located at different parts of the Gla domain. The anticoagulant activity as well as phospholipid binding ability was only enhanced when multiple regions of the Gla domain were modified. The results provide insights into the molecular mechanisms that are involved in determining the binding affinity of the interaction between Gla domains and phospholipid membranes. The unique properties of QGNSEDY-APC suggest this APC variant possibly to have greater therapeutic potential than WT APC.

Mutagenesis of the gamma-carboxyglutamic acid domain of human factor VII to generate maximum enhancement of the membrane contact site

Site-directed mutagenesis of the 40 N-terminal residues (gamma-carboxyglutamic acid domain) of blood clotting factor VII was carried out to identify sites that improve membrane affinity. Improvements and degree of change included P10Q (2-fold), K32E (13-fold), and insertion of Tyr at position 4 (2-fold). Two other beneficial changes, D33F (2-fold) and A34E (1.5-fold), may exert their impact via influence of K32E. The modification D33E (5.2-fold) also resulted in substantial improvement. The combined mutant with highest affinity, (Y4)P10Q/K32E/D33F/A34E, showed 150-296-fold enhancement over wild-type factor VIIa, depending on the assay used. Undercarboxylation of Glu residues at positions 33 and 34 may result in an underestimate of the true contributions of gamma-carboxyglutamic acid at these positions. Except for the Tyr(4) mutant, all other beneficial mutations were located on the same surface of the protein, suggesting a possible membrane contact region. An initial screening assay was developed that provided faithful evaluation of mutants in crude mixtures. Overall, the results suggest features of membrane binding by vitamin K-dependent proteins and provide reagents that may prove useful for research and therapy.

Protein S Gla-domain mutations causing impaired Ca(2+)-induced phospholipid binding and severe functional protein S deficiency

We have identified 2 PROS1 missense mutations in the exon that encodes the vitamin K-dependent Gla domain of protein S (Gly11Asp and Thr37Met) in kindred with phenotypic protein S deficiency and thrombosis. In studies using recombinant proteins, substitution of Gly11Asp did not affect production of protein S but resulted in 15.2-fold reduced protein S activity in a factor Va inactivation assay. Substitution of Thr37Met reduced expression by 33.2% (P <.001) and activity by 3.6-fold. The Gly11Asp variant had 5.4-fold reduced affinity for anionic phospholipid vesicles (P <.0001) and decreased affinity for an antibody specific for the Ca(2+)-dependent conformation of the protein S Gla domain (HPS21). Examination of a molecular model suggested that this could be due to repositioning of Gla29. In contrast, the Thr37Met variant had only a modest 1.5-fold (P <.001), reduced affinities for phospholipid and HPS21. This mutation seems to disrupt the aromatic stack region. The proposita was a compound heterozygote with free protein S antigen levels just below the lower limit of the normal range, and this is now attributed to the partial expression defect of the Thr37Met mutation. The activity levels were strongly reduced to 15% of normal, probably reflecting the functional deficit of both protein S variants. Her son (who was heterozygous only for Thr37Met) had borderline levels of protein S antigen and activity, reflecting the partial secretion and functional defect associated with this mutation. This first characterization of natural protein S Gla-domain variants highlights the importance of the high affinity protein S-phospholipid interaction for its anticoagulant role.

Modeling zymogen protein C

A solution structure for the complete zymogen form of human coagulation protein C is modeled. The initial core structure is based on the x-ray crystallographic structure of the gamma-carboxyglutamic acid (Gla)-domainless activated form. The Gla domain (residues 1-48) is modeled from the x-ray crystal coordinates of the factor VII(a)/tissue factor complex and oriented with the epidermal growth factor-1 domain to yield an initial orientation consistent with the x-ray crystal structure of porcine factor IX(a). The missing C-terminal residues in the light chain (residues 147-157) and the activation peptide residues 158-169 were introduced using homology modeling so that the activation peptide residues directly interact with the residues in the calcium binding loop. Molecular dynamics simulations (Amber-particle-mesh-Ewald) are used to obtain the complete calcium-complexed solution structure. The individual domain structures of protein C in solution are largely unaffected by solvation, whereas the Gla-epidermal growth factor-1 orientation evolves to a form different from both factors VII(a) and IX(a). The solution structure of the zymogen protein C is compared with the crystal structures of the existing zymogen serine proteases: chymotrypsinogen, proproteinase, and prethrombin-2. Calculated electrostatic potential surfaces support the involvement of the serine protease calcium ion binding loop in providing a suitable electrostatic environment around the scissile bond for II(a)/thrombomodulin interaction.

Vitamin K-dependent proteins

Vitamin K is required for the synthesis of gamma-carboxyglutamate (Gla) during postribosomal protein modification. Substrates include blood clotting proteins, bone proteins, cell signaling, and receptor proteins. In addition, Gla is a component of short toxin peptides from the marine snail Conus. Studies of structure-function relationships are the most advanced for the blood coagulation proteins. Reviews of vitamin K action and blood coagulation are presented. Special focus is on the structure-function role of Gla in blood coagulation and the impact of this amino acid on enzyme reaction kinetics. This amino acid forms calcium and membrane binding sites for these proteins. Two proposed mechanisms of protein-membrane attachment are reviewed. One involves membrane attachment by protein insertion into the hydrocarbon region of the membrane, while another considers attachment by specific interactions with phospholipid head groups. Membrane attachment generates the potential for several forms of nonclassical enzyme kinetic behaviors, all of which have been observed in vitro. For example, the reaction may be limited by properties of the enzyme active site, a condition that allows use of classic steady-state enzyme kinetic parameters. However, the reaction may be limited by substrate binding to the membrane, by substrate flux through solution, and/or by solvent flow rates across the membrane surface. These states provide special mechanisms that are not anticipated by classical steady-state kinetic derivations. They may be used to regulate coagulation in vivo. Overall, vitamin K research spans the spectrum of biological research and experience. Exciting new ideas and findings continue to emanate from vitamin K-related research.