Use of Snake Venom Inhibitors in Studies of the Function and Tertiary Structure of Coagulation Factors

C-type lectin-like proteins (CLPs) of snake venom have a variety of biological properties, acting for example as anticoagulants, procoagulants, and agonists/antagonists of platelet activation. The structural and functional studies of the first identified venom CLP, factors IX/X-binding protein (IX/X-bp), have contributed to our understanding of the roles of magnesium ions in the blood coagulation cascade reaction. The crystal structures of gamma-carboxyglutamic acid (Gla) domains of cpagulation factors X and IX have recently been clarified in structural studies of complexes between the Gla domain of factor X and X-bp (a venom CLP) and between the Gla domain of factor IX and IX-bp (a venom CLP).     

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.     

Lupus anticoagulants, thrombosis and the protein C system.

Although lupus anticoagulants (LAs) are immunoglobulins that inhibit procoagulant reactions in vitro, these molecules are associated with thrombosis in vivo. We and others have hypothesized that this may be due to selective targeting of the activated protein C (APC) anticoagulant pathway. Populations of antibodies that interact with protein C or protein S in ways that inhibit their activity are obvious candidates for such pathological molecules. However, it is less clear how populations that appear to bind to membrane surfaces might target the APC anticoagulant complex selectively. Studies now show that the membrane requirements of the APC anticoagulant complex are significantly different from those of the procoagulant reactions. The most dramatic difference is the requirement for the presence of phosphatidylethanolamine (PE) in the membrane for optimal APC function. The inhibitory activity of at least some LAs is enhanced by the presence of PE, but the anti-APC activity is enhanced even more, resulting in the plasma from these patients clotting faster than normal when APC is present. Structure-function studies have been undertaken to understand the PE dependence of this reaction better. Chimeric proteins in which all or part of the Gla domain of protein C has been replaced by the homologous region of prothrombin have been prepared. Unexpectedly, the PE dependence resides primarily in the C-terminal half of the Gla domain. Using liposomes of various composition, we found both the presence of the PE head group and unsaturation of the fatty acid chains are required for optimal inactivation of factor Va. It is hoped that a better understanding of the biochemistry of these reactions, combined with the use of the chimeric proteins described, will permit us to design better assays for the identification of pathologic LAs.     

Activated protein C cofactor function of protein S: a novel role for a γ-carboxyglutamic acid residue

Protein S has an important anticoagulant function by acting as a cofactor for activated protein C (APC). We recently reported that the EGF1 domain residue Asp95 is critical for APC cofactor function. In the present study, we examined whether additional interaction sites within the Gla domain of protein S might contribute to its APC cofactor function. We examined 4 residues, composing the previously reported "Face1" (N33S/P35T/E36A/Y39V) variant, as single point substitutions. Of these protein S variants, protein S E36A was found to be almost completely inactive using calibrated automated thrombography. In factor Va inactivation assays, protein S E36A had 89% reduced cofactor activity compared with wild-type protein S and was almost completely inactive in factor VIIIa inactivation; phospholipid binding was, however, normal. Glu36 lies outside the ω-loop that mediates Ca(2+)-dependent phospholipid binding. Using mass spectrometry, it was nevertheless confirmed that Glu36 is γ-carboxylated. Our finding that Gla36 is important for APC cofactor function, but not for phospholipid binding, defines a novel function (other than Ca(2+) coordination/phospholipid binding) for a Gla residue in vitamin K-dependent proteins. It also suggests that residues within the Gla and EGF1 domains of protein S act cooperatively for its APC cofactor function.     

Functional mapping of anti-factor IX inhibitors developed in patients with severe hemophilia B

Development of inhibitory antibodies is a serious complication of treatment with repeated factor IX infusions in a minority of patients with hemophilia B. Such antibodies detected in 8 patients have been characterized. Typing studies revealed that patients' immune response toward factor IX is highly heterogeneous and involves immunoglobulin G (IgG) antibodies, preferentially IgG1 and IgG4. The preservation of the sequence and the 3-dimensional orientation of the amino acids constituting one epitope are highly important for the assembly of an antibody-antigen complex. To localize the epitopes on the factor IX molecule, an original approach was designed using a set of factor X chimeras carrying regions of factor IX. Results showed that some patients' antibodies were directed against both the domain containing the gamma-carboxy glutamic acid residues (Gla domain) and the protease domain of factor IX. In contrast, no binding was observed to the epidermal growth factor-like domains or to the activation peptide. Functional characterization showed that the purified IgG from patients' serum inhibited the factor VIIIa-dependent activation of factor X. Moreover, patients' IgG directed against the Gla domain inhibited the binding of factor IX to phospholipids as well as the binding of factor VIII light chain to factor IXa. These data demonstrate that inhibitors appearing in patients with severe hemophilia B display specificity against restricted functional domains of factor IX.     

Inflammation, sepsis, and coagulation

The molecular links between inflammation and coagulation are unquestioned. Inflammation promotes coagulation by leading to intravascular tissue factor expression, eliciting the expression of leukocyte adhesion molecules on the intravascular cell surfaces, and down regulating the fibrinolytic and protein C anticoagulant pathways. Thrombin, in turn, can promote inflammatory responses. This creates a cycle that logically progresses to vascular injury as occurs in septic shock. Most complex systems are regulated by product inhibition. This inflammation-coagulation cycle seems to follow this same principle with the protein C pathway serving as the regulatory mechanism. The molecular basis by which the protein C pathway functions as an anticoagulant is relatively well established compared to the mechanisms involved in regulating inflammation. As one approach to identifying the mechanisms involved in regulating inflammation, we set out to identify novel receptors that could modulate the specificity of APC in a manner analogous to the mechanisms by which thrombomodulin modulates thrombin specificity. This approach led to the identification of an endothelial cell protein C receptor (EPCR). To understand the mechanism, we obtained a crystal structure of APC (lacking the Gla domain). The crystal structure reveals a deep groove in a location analogous to anion binding exosite 1 of thrombin, the location of interaction for thrombomodulin, platelet thrombin receptor and fibrinogen. Thrombomodulin blocks the activation of platelets and fibrinogen without blocking reactivity with chromogenic substrates or inhibitors. Similarly, in solution, EPCR blocks factor Va inactivation without modulating reactivity with protease inhibitors. Thus, these endothelial cell receptors for the protein C system share many properties in common including the ability to be modulated by inflammatory cytokines. Current studies seek to identify the substrate for the APC-EPCR complex as the next step in elucidating the mechanisms by which the protein C pathway modulates the response to injury and inflammation.     

Enhancement of vitamin-K-dependent protein function by modification of the gamma-carboxyglutamic acid domain: studies of protein C and factor VII

Vitamin-K-dependent proteins are found in both the pro- and anti-coagulation cascades, and their use in coagulation therapies is expanding rapidly. The vitamin-K-dependent, gamma-carboxyglutamic acid (Gla)-containing regions of proteins in this family are homologous and are responsible for membrane association. Site-directed mutations that enhance the membrane affinity of protein C, an anticoagulant, and of factor VII, a procoagulant, have been identified. These protein C and Factor VII mutants show enhanced activity in many assays, offering opportunities to study the role of membrane in blood clotting reactions and proteins that may have greater therapeutic value.     

Full-length cDNA cloning and protein three-dimensional structure modeling of porcine prothrombin

Prothrombin is a vitamin K-dependent serine protease and plays pivotal roles in both procoagulant and anticoagulant pathway of hemostasis. In this study, we cloned the full-length cDNA of porcine prothrombin by cDNA library screening and SMART RACE technique. The full-length cDNA is 2027 bp, with a 1869 bp Open Reading Frame (ORF) coding 623 amino acids. The deduced protein of porcine prothrombin contains signal peptide, propeptide, Gla domain, two kringle domains and trypsin domain. Porcine prothrombin shares 86.15% nucleotide similarity and 83% amino acid similarity with human prothrombin. The trypsin domain is highly conserved between the two species with 92.1% amino acid identity. Macromolecular interaction sites comparison between porcine and human prothrombin suggests that the Gla domain in porcine prothrombin contains an additional potential gamma-carboxyglutamic acid site. However, a thrombin cleavage site (Arg284-Thr285) in its light chain is lost. When thrombin heavy chain is concerned, the most important functional sites such as catalytic triad DHS, RGD site, Na+ binding site and anion-binding exosite-I and II are highly conserved. However, great differences have been observed between residues 145 and 158 of heavy chain which is associated with thrombomodulin binding. Two important limited proteolysis sites at Ala150 and Lys154 were lost in porcine sequence, which would affect epsilon-thrombin and gammaT-thrombin generation. Comparison on 3-D protein models demonstrates that these proteins are obviously different in autolysis loop (Lys145 to Gly155). Compared with that of human prothrombin, variation at critical recognition sites would likely alter its binding affinity and reaction velocity, which would contribute to coagulation disorder when porcine liver is transplanted into human body.     

Recombinant nematode anticoagulant protein c2 and other inhibitors targeting blood coagulation factor VIIa/tissue factor

Originally isolated from a haematophagous hookworm, recombinant nematode anticoagulant protein c2 (rNAPc2) is an 85-amino acid protein with potent anticoagulant properties. Unlike conventional anticoagulants that attenuate blood coagulation via inhibition of thrombin or activated factor X (FXa) at the downstream portion of the cascade, rNAPc2 is a potent inhibitor of the activated factor VII/tissue factor complex (FVIIa/TF), the key physiological initiator of blood coagulation. Its mechanism of action requires prerequisite binding to circulating FXa or zymogen factor X (FX) to form a binary complex prior to its interaction and inhibition of membrane-bound FVIIa/TF. The binding of rNAPc2 to FX results in an elimination half-life of longer than 50 h following either subcutaneous or intravenous administration. Recombinant NAPc2, like other inhibitors of FVIIa/TF including tissue factor pathway inhibitor (TFPI) and active site-blocked FVIIa (ASIS, FFR-rFVIIa or FVIIai), may have a promising role in the prevention and treatment of venous and arterial thrombosis, as well as potential efficacy in the management of disseminated intravascular coagulopathies because of their potent and selective inhibition of FVIIa/TF.     

Calcium ion binding to human and bovine factor X

Human and bovine factor X contain 11 and 12 glutamyl residues, respectively, within the first 40 amino terminal residues that are post-translationally modified to gamma-carboxyglutamyl (Gla) residues. We have measured calcium ion binding to human factor X by equilibrium dialysis. This is the first examination of calcium ion binding to human factor X. We have also re-examined the equilibrium dialysis binding of calcium ions to bovine facor X in order to compare the two species. The data was analysed using a variety of models that allow for more than one class of binding site and for co-operativity among binding sites. Calcium ion binding to human factor X fits a model that had two classes of sites: one class with a single site that had an affinity of 0.1 mM and a second class with 19 equivalent, non-interacting sites with an average affinity of 3.5 mM. There was no evidence for co-operativity in calcium ion binding. Calcium ion binding to bovine factor X was best stimulated by a model that assumed one tight site, four co-operative sites, and 18 equivalent, non-interacting sites. To examine the co-operativity seen in calcium ion binding to bovine factor X, calcium ion binding to isolated Gla region (residues 1-44) and Gla-domainless factor X was measured by equilibrium dialysis. Calcium ion binding to Gla-domainless factor X was simulated by a model that had two classes of sites: one class with a single site that had an affinity of 0.25 mM, and a second class that had 15 sites with very low affinity sites (greater than 15 mM).(ABSTRACT TRUNCATED AT 250 WORDS)     

Structure-function relationships in factor IX and factor IXa

Factor IX (FIX) consists of an N-terminal gamma-carboxyglutamic acid (Gla) domain followed by two epidermal growth factor (EGF)-like domains, and the C-terminal serine protease domain. During physiologic coagulation, one of the activators of FIX is the FVIIa/tissue factor (TF) complex. In this reaction, the Gla and EGF1 domains of FIX are thought to interact with TF. The FIXa that is generated then combines with FVIIIa on the platelet surface to activate FX in the coagulation cascade. In this assembly, the protease domain and possibly the EGF2 domain of FIXa are thought to provide the primary specificity in binding to FVIIIa. Disruption of the interaction of FIX/FX with TF and of the FIXa:FVIIIa interface may provide a pharmacologic target as an alternative strategy for the development of antithrombotic agents.     

The binding of protein S and the protein S-C4BP complex to neutrophils is apoptosis dependent.

Vitamin K-dependent protein S and complement regulator C4b-binding protein (C4BP) form a high-affinity complex in plasma. We have previously shown that both free protein S and the C4BP-protein S complex can bind to apoptotic Jurkat cells. It has been demonstrated in the past that protein S and C4BP can bind to neutrophils. We now show that it is only the apoptotic neutrophil population that binds these proteins. In addition, we also show that binding is mediated through the Gla domain on protein S, which binds negatively charged phospholipids, since a monoclonal antibody directed against this domain blocks the binding. Thus, we conclude that binding of protein S and the C4BP-protein S complex to neutrophils is not cell specific, but rather apoptosis dependent.     

Full-length cDNA cloning and protein three-dimensional structure modeling of factor VII of rhesus monkey, Macaca mulatta

FVII is a vitamin K dependent serine protease that plays a key role in extrinsic coagulation pathway. In this paper, we report the full-length cDNA sequences of rhesus monkey FVII. The full-length cDNA has 2424 bp, and predicts an open reading frame of 1416 bp corresponding to 472 amino acids. The deduced protein sequence of rhesus monkey FVII indicates the functional domains including signal peptide, Gla domain, two EGF domains, and catalytic domain. Rhesus monkey FVII is highly homologous to human FVII with amino acid identity of 91.0%. Comparison of three-dimensional protein structure shows high conservation between them. The important functional sites such as the N-terminal gamma-carboxyglutamic acids of the Gla domain, the Ca(2+) binding region of the EGF I domain, the TF binding region, the active site binding cleft, and the macromolecular substrate binding exosite of trypsin domain are all well conserved in FVII of rhesus monkey. Prothrombin time test shows rhesus monkey FVII has a similar clotting time with that of human. This study of rhesus monkey FVII might be helpful for understanding the function compatibility of human and rhesus monkey FVII, which is beneficial for the study of xenotransplantation.     

The gamma-carboxyglutamic acid domain of anticoagulant protein S is involved in activated protein C cofactor activity, independently of phospholipid binding

We expressed 2 chimeras between human protein S (PS) and human prothrombin (FII) in which the prothrombin gamma-carboxyglutamic acid (Gla) domain replaced the PS Gla domain in native PS (Gla(FII)-PS) or in PS deleted of the thrombin-sensitive region (TSR) (Gla(FII)-DeltaTSR-PS). Neither PS/FII chimera had activated protein C (APC) cofactor activity in plasma clotting assays or purified systems, but both bound efficiently to phospholipids. This pointed to a direct involvement of the PS Gla domain in APC cofactor activity through molecular interaction with APC. Using computational methods, we identified 2 opposite faces of solvent-exposed residues on the PS Gla domain (designated faces 1 and 2) as potentially involved in this interaction. Their importance was supported by functional characterization of a PS mutant in which the face 1 and face 2 PS residues were reintroduced into Gla(FII)-PS, leading to significant APC cofactor activity, likely through restored interaction with APC. Furthermore, by characterizing PS mutants in which PS face 1 and PS face 2 were individually replaced by the corresponding prothrombin faces, we found that face 1 was necessary for efficient phospholipid binding but that face 2 residues were not strictly required for phospholipid binding and were involved in the interaction with APC.     

Anticoagulant effects of a synthetic peptide containing residues Thr-2253-Gln-2270 within factor VIII C2 domain that selectively inhibits factor Xa-catalysed factor VIII activation

Factor VIII (FVIII), an essential cofactor that accelerates the generation of factor Xa (FXa) in the tenase complex, is activated by proteolytic cleavage by thrombin or FXa. A strong relationship has been reported between high levels of FVIII activity and thrombosis. We have demonstrated previously that an anti-FVIII C2 antibody (ESH8) with a Val-2248-Gly-2285 epitope inhibited FXa-catalysed FVIII activation, and that a synthetic peptide designated EP-2 (residues 2253-2270) blocked C2 domain binding to FXa. We investigated the inhibitory effect of EP-2 on FXa-catalysed FVIII activation and its anticoagulant effect in the blood coagulation system. EP-2 inhibited FXa-catalysed activation in a clotting assay in a dose-dependent manner and reduced FXa generation in a chromogenic assay using FVIII, factor X, factor IXa and phospholipid. The peptide only inhibited FVIII binding to FXa. We also tested the anticoagulant effect of EP-2 in the plasma milieu. The peptide prolonged the activated partial thromboplastin time and activated clotting time in a dose-dependent manner, but not prothrombin time. Our results indicate that EP-2 mediates the anticoagulant effect by specific inhibition of FVIII and FXa interaction in the intrinsic pathway, and that FXa-catalysed FVIII activation plays a significant role in blood clotting. The peptide may provide the basis for the development of novel anticoagulant therapy.     

Ixolaris, a novel recombinant tissue factor pathway inhibitor (TFPI) from the salivary gland of the tick, Ixodes scapularis: identification of factor X and factor Xa as scaffolds for the inhibition of factor VIIa/tissue factor complex

Saliva of the hard tick and Lyme disease vector, Ixodes scapularis, has a repertoire of compounds that counteract host defenses. Following sequencing of an I scapularis salivary gland complementary DNA (cDNA) library, a clone with sequence homology to tissue factor pathway inhibitor (TFPI) was identified. This cDNA codes for a mature protein, herein called Ixolaris, with 140 amino acids containing 10 cysteines and 2 Kunitz-like domains. Recombinant Ixolaris was expressed in insect cells and shown to inhibit factor VIIa (FVIIa)/tissue factor (TF)-induced factor X (FX) activation with an inhibitory concentration of 50% (IC(50)) in the picomolar range. In nondenaturing gel, Ixolaris interacted stoichiometrically with FX and FXa but not FVIIa. Ixolaris behaves as a fast-and-tight ligand of the exosites of FXa and gamma-carboxyglutamic acid domainless FXa (des-Gla-FXa), increasing its amidolytic activity. At high concentration, Ixolaris attenuates the amidolytic activity of FVIIa/TF; however, in the presence of DEGR-FX or DEGR-FXa (but not des-Gla-DEGR-FXa), Ixolaris becomes a tight inhibitor of FVIIa/TF as assessed by recombinant factor IX (BeneFIX) activation assays. This indicates that FX and FXa are scaffolds for Ixolaris in the inhibition of FVIIa/TF and implies that the Gla domain is necessary for FVIIa/TF/Ixolaris/FX(a) complex formation. Additionally, we show that Ixolaris blocks FXa generation by endothelial cells expressing TF. Ixolaris may be a useful tool to study the structural features of FVIIa, FX, and FXa, and an alternative anticoagulant in cardiovascular diseases.     

Location of the platelet binding site in zymogen coagulation factor IX.

The assembly of the tenase complex on the surface of the platelet is an essential step in maintaining normal hemostasis as evidenced by the serious hemorrhagic diathesis associated with either factor IX (FIX) or factor VIII deficiencies. Understanding the regions and or residues of FIX crucial for proper binding to platelets has important clinical implications. The ability of FIX to bind activated platelets in the presence of 4 mmol/l CaCl2 was examined using electrophoretic light-scattering experiments. Wild-type FIX binds to activated platelets with dissociation constant Kd = 7.9 nmol/l. Activated FIX binds to activated platelets with Kd = 2 nmol/l. Activated factor VII does not bind activated platelets at physiological concentrations. The Gla domain of FIX is important for the binding of FIX to activated platelets since a chimera with a factor VII (FVII) template and FIX Gla [FVII(FIXGla)] has Kd = 9.6 nmol/l, and a chimera with a FVII template and FIX Gla, A and the first epidermal growth factor domain (EGF1) [FVII(FIXGla,A,EGF1)] has Kd = 9.7 nmol/l, but a chimera with a FIX template and a FVII Gla [FIX(FVIIGla)] does not bind activated platelets. Altering the fifth residue of FIX from a lysine to an alanine (Lys5<--Ala) abolishes the mutant from binding to collagen but does not affect FIX binding to the activated platelet (Kd = 9.8 nmol/l). Point mutations involved with residues 4 and 5 (Gly4<--Phe and Lys5<--no residue), residue 9 (Phe9<--Ala), residue 10 (Val10<--Lys) and residues 9-11 (Phe9<--Met, Val10<--Lys, Glu11<--Lys) do not bind to activated platelets.     

Synergistic inhibition of the intrinsic factor X activation by protein S and C4b-binding protein

The complement protein C4b-binding protein plays an important role in the regulation of the protein C anticoagulant pathway. C4b-binding protein can bind to protein S, thereby inhibiting the cofactor activity of protein S for activated protein C. In this report, we describe a new role for C4b-binding protein in coagulation. We observed inhibition of the intrinsic factor X activating reaction by the complex of C4b- binding protein and protein S. At the plasma concentration of protein S, the factor X activation was inhibited for 50% and addition of C4b- binding protein led to a potentiation of the inhibition to almost 90%. Because C4b-binding protein alone had no effect on the activation of factor X, we hypothesized that binding of C4b-binding protein to protein S was a prerequisite for optimal inhibition of factor X activation. C4b-binding protein lacking the beta-chain, which is unable to bind to protein S, did not potentiate the inhibitory effect of protein S. In an earlier study, we observed that C4b-binding protein increased the binding affinity of protein S for factor VIII. Therefore, a possible interaction of C4b-binding protein with factor VIII was investigated. C4b-binding protein bound to factor VIII and to thrombin activated factor VIII in a saturable and specific way. Also, factor VIII in complex with von Willebrand factor was able to bind C4b-binding protein. The beta-chain of C4b-binding protein was not required for the interaction with factor VIII because C4b-binding protein lacking the beta-chain also bound to factor VIII. Monoclonal antibodies directed against the alpha-chain of C4b-binding protein inhibited the binding to factor VIII, whereas monoclonal antibodies directed against the beta- chain had no effect on the binding to factor VIII. This finding indicates that the binding site for factor VIII on C4b-binding protein is localized on the alpha-chains of C4b-binding protein. The potentiation by C4b-binding protein of the inhibition of the factor X activation by protein S was blocked by a monoclonal antibody directed against the alpha-chain of C4b-binding protein. This finding indicates that the potentiation of the inhibitory effect of protein S was mediated via an interaction of C4b-binding protein with factor VIII. C4b-binding protein did not bind to factor V and was not able to potentiate the inhibitory effect of protein S on prothrombinase activity.(ABSTRACT TRUNCATED AT 400 WORDS)     

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.