Improved hemostasis in hemophilia mice by means of an engineered factor Va mutant.

BACKGROUND: Factor (F)VIIa-based bypassing not always provides sufficient hemostasis in hemophilia. OBJECTIVES: To investigate the potential of engineered activated factor V (FVa) variants as bypassing agents in hemophilia A. METHODS: Activity of FVa variants was studied in vitro using prothrombinase assays with purified components and in FV- and FVIII-deficient plasma using clotting and thrombin generation assays. In vivo bleed reduction after the tail clip was studied in hemophilia A mice. RESULTS AND CONCLUSIONS: FVa mutations included a disulfide bond connecting the A2 and A3 domains and ones that rendered FVa resistant to inactivation by activated protein C (APC). '(super) FVa,' a combination of the A2-A3 disulfide (A2-SS-A3) to stabilize FVa and of APC-cleavage site mutations (Arg506/306/679Gln), had enhanced specific activity and complete APC resistance compared with wild-type FVa, FVL eiden (Arg506Gln), or FVaL eiden (A2-SS-A3). Furthermore, (super) FVa potently increased thrombin generation in vitro in FVIII-deficient plasma. In vivo, (super) FVa reduced bleeding in FVIII-deficient mice more effectively than wild-type FVa. Low-dose (super) FVa, but not wild-type FVa, decreased early blood loss during the first 10 min by more than two-fold compared with saline and provided bleed protection for the majority of mice, similar to treatments with FVIII. During the second 10 min after tail cut, (super) FVa at high dose, but not wild-type FVa, effectively reduced bleeding. These findings suggest that (super) FVa enhances not only clot formation but also clot stabilization. Thus, (super) FVa efficiently improved hemostasis in hemophilia in vitro and in vivo and may have potential therapeutic benefits as a novel bypassing agent in hemophilia.

Activated protein C.

Protein C is a vitamin K-dependent plasma protein zymogen whose genetic mild or severe deficiencies are linked with risk for venous thrombosis or neonatal purpura fulminans, respectively. Studies over past decades showed that activated protein C (APC) inactivates factors (F) Va and VIIIa to down-regulate thrombin generation. More recent basic and preclinical research on APC has characterized the direct cytoprotective effects of APC that involve gene expression profile alterations, anti-inflammatory and anti-apoptotic activities and endothelial barrier stabilization. These actions generally require endothelial cell protein C receptor (EPCR) and protease activated receptor-1. Because of these direct cytoprotective actions, APC reduces mortality in murine endotoxemia and severe sepsis models and provides neuroprotective benefits in murine ischemic stroke models. Furthermore, APC reduces mortality in patients with severe sepsis (PROWESS clinical trial). Although much remains to be clarified about mechanisms for APC's direct effects on various cell types, it is clear that APC's molecular features that determine its antithrombotic action are partially distinct from those providing cytoprotective actions because we have engineered recombinant APC variants with selective reduction or retention of either anticoagulant or cytoprotective activities. Such APC variants can provide relatively enhanced levels of either cytoprotective or anticoagulant activities for various therapeutic applications. We speculate that APC variants with reduced anticoagulant action but normal cytoprotective actions hold the promise of reducing bleeding risk because of attenuated anticoagulant activity while reducing mortality based on direct cytoprotective effects on cells.

Disulfide bond-stabilized factor VIII has prolonged factor VIIIa activity and improved potency in whole blood clotting assays.

BACKGROUND: Genetically engineered disulfide bonds in B-domain-deleted factor (F) VIII variants (C662-C1828 FVIII and C664-C1826 FVIII) improve FVIIIa stability by blocking A2 domain dissociation because the new disulfide covalently links the A2 and A3 domains in FVIIIa. AIM: The aim of this study was to assess the hypothesis that these variants have physiologically relevant properties because of prolonged thrombin generation and improved clot formation in whole blood. METHODS: Clot-formation properties in whole blood were measured in thromboelastogram assays. The thrombin generation capabilities of the wild-type (WT) FVIII and FVIII variants were determined, and half-lives of FVIIIa variants were determined in fresh whole blood serum. RESULTS: Thromboelastogram assays were performed with fresh, severe hemophilia whole blood reconstituted with variant and WT FVIII. The two disulfide bond-stabilized FVIII variants and WT FVIII had comparable clotting times at all studied concentrations. However, when compared with WT FVIII at low concentrations, the two FVIII variants required only 10% as much FVIII to achieve comparable clot-formation rates, clot-formation times and clot firmness values. The differences between WT and FVIII variants were quite pronounced at low FVIII concentrations. Measurement of the endogenous thrombin potential in FVIII-deficient plasma supplemented with these FVIII variants confirmed that the disulfide bond-stabilized variants supported high levels of thrombin generation at lower concentrations than did WT FVIII. During the course of clot generation in whole blood, the disulfide bond-stabilized FVIIIa variants had approximately 5-fold increased half-lives relative to WT FVIIIa. CONCLUSION: C662-C1828 FVIII and C664-C1826 FVIII have physiologically relevant superior clot-forming properties in a whole blood environment, most likely due to the increased half-life of these FVIIIa variants in whole blood.

Intrinsic stability and functional properties of disulfide bond-stabilized coagulation factor VIIIa variants.

BACKGROUND: The utility of purified coagulation factor (F)VIII for treatment of hemophilia A is limited in part by its instability following activation by thrombin, which is caused by spontaneous dissociation of the A2 domain from the activated FVIII (FVIIIa) heterotrimer. To prevent this A2 domain dissociation in FVIIIa, we previously engineered a cysteine pair (C664-C1826) in recombinant FVIII that formed a disulfide bond cross-linking the A2 domain in the heavy chain to the A3 domain in the light chain. This engineered disulfide bond resulted in a more stable FVIIIa. AIMS: Here, we characterize the functional parameters of C664-C1828 FVIII and of a new disulfide bond-stabilized FVIII (C662-C1828 FVIII). METHODS: In order to assess whether these FVIII variants might be good candidates for a new therapeutic agent to treat hemophilia A, we investigated a variety of functional parameters that might affect the in vivo properties of the variants, including half-life of disulfide bond-stabilized FVIII and FVIIIa and the potency of these FVIIIa molecules in the FXase complex. RESULTS: Both disulfide bond-stabilized variants had improved affinity for von Willebrand factor (VWF). In studies of FX activation by purified FIXa and FVIIIa, C662-C1828 FVIIIa had normal activity while C664-C1826 FVIIIa had reduced activity. Both C664-C1826 FVIIIa and C662-C1828 FVIIIa were inactivated by activated protein C (APC) but the rates of inactivation were different. CONCLUSION: Overall, the specific location of the disulfide bridge between the A2 and A3 domains appears to affect functional properties of FVIIIa. In summary, introduction of engineered interdomain disulfides results in FVIIIa variants that resist spontaneous loss of activity while retaining susceptibility to APC proteolytic inactivation and maintaining VWF binding.

An engineered interdomain disulfide bond stabilizes human blood coagulation factor VIIIa.

The blood coagulation disorder, hemophilia A, is caused by deficiency of coagulation factor (F)VIII. Hemophilia A is now treated by infusions of pure FVIII, but the activity of FVIII is limited because it is unstable following activation by thrombin. This instability of activated FVIII is the result of dissociation of the A2 subunit. To obtain increased stability in FVIIIa, a disulfide bond between the A2 domain and the A3 domain, preventing A2 subunit dissociation, has been engineered. Structural analysis of the FVIII A domain homology model allowed us to identify residues 664 and 1826 as a potential disulfide bond pair. A FVIII mutant containing Cys664 and Cys1826 was produced and purified (C664-C1826 FVIII). Immunoblotting showed that a disulfide bond did form to link covalently the A2 and the A3 domains. Following activation of the recombinant C664-C1826 FVIII by thrombin, the mutant FVIIIa had increased stability and retained more than 90% of its clotting activity at a time at which wild-type FVIIIa lost more than 90% of its activity. This remarkably stable C664-C1826 FVIIIa provides a unique approach for studies of the cofactor activity of FVIIIa and also for new, improved therapy for hemophilia A.

Update on tumor cell procoagulant factors.

Tumor cells produce tissue factor, cancer procoagulant, plasminogen activators and other factors that interact with the coagulation system, the fibrinolytic system and vascular or blood cells such that they can upset the normal homeostasis and balance between activation and inhibition of the coagulation and fibrinolytic systems. These activities play a role in tumor cell growth and metastasis, vascular wall function, and hemostasis. Proteases and their inhibitors are intimately involved in all aspects of the hemostatic, cell proliferation and cellular signalling systems. This review provides a brief examination of recent observations in this complex interaction of cellular and hemostatic factors.

Three-dimensional model of coagulation factor Va bound to activated protein C.

A complete molecular model of blood coagulation factor Va (FVa) bound to anticoagulant activated protein C (APC) and to a phospholipid membrane was constructed. The three homologous A domains and the two homologous C domains of FVA were modeled based on the X-ray crystallographic structures of ceruloplasmin and C2 domain of factor V, respectively. The final arrangement of the five domains in the complete FVa model bound to a membrane incorporated extensive published experimental data. FVa binds the phospholipid membrane through its C2 domain while the A-domain trimer is located from 40 through 100 A above the membrane plane. From our model we infer a probable role for metal ions at the interface between FVa light and heavy chains, provide an explanation for the slower APC cleavage at Arg306 relative to Arg506, and predict specific interactions between positively and negatively charged exosites in APC and FVa, respectively.

The autolysis loop of activated protein C interacts with factor Va and differentiates between the Arg506 and Arg306 cleavage sites.

The anticoagulant human plasma serine protease, activated protein C (APC), inactivates blood coagulation factors Va (FVa) and VIIIa. The so-called autolysis loop of APC (residues 301-316, equivalent to chymotrypsin [CHT] residues 142-153) has been hypothesized to bind FVa. In this study, site-directed mutagenesis was used to probe the role of the charged residues in this loop in interactions between APC and FVa. Residues Arg306 (147 CHT), Glu307, Lys308, Glu309, Lys311, Arg312, and Arg314 were each individually, or in selected combinations, mutated to Ala. The purified recombinant protein C mutants were characterized using activated partial thromboplastin time (APTT) clotting assays and FVa inactivation assays. Mutants 306A, 308A, 311A, 312A, and 314A had mildly reduced anticoagulant activity. Based on FVa inactivation assays and APTT assays using purified Gln506-FVa and plasma containing Gln506-FV, it appeared that these mutants were primarily impaired for cleavage of FVa at Arg506. Studies of the quadruple APC mutant (306A, 311A, 312A, and 314A) suggested that the autolysis loop provides for up to 15-fold discrimination of the Arg506 cleavage site relative to the Arg306 cleavage site. This study shows that the loop on APC of residues 306 to 314 defines an FVa binding site and accounts for much of the difference in cleavage rates at the 2 major cleavage sites in FVa. (Blood. 2000;96:585-593)

Blood coagulation: The outstanding hydrophobic residues.

Newly determined crystal structures suggest that the membrane-binding C2 domains of blood coagulation cofactors Va and VIIIa bind anionic phospholipids through protruding solvent-exposed hydrophobic residues, aided by a crown of positively charged residues and by specific hydrogen-bonding side chains.

Structural basis for hemophilia A caused by mutations in the C domains of blood coagulation factor VIII.

Three dimensional homology models for the C1 and C2 domains of factor VIII (FVIII) were generated. Each C domain formed a beta-sandwich, and C1 was covalently connected to C2 in a head-to-head orientation. Of the >250 missense mutations that cause FVIII deficiency and hemophilia A, 34 are in the C domains. We used the FVIII C1-C2 model to infer the structural basis for the pathologic effects of these mutations. The mutated residues were divided into four categories: 15 conserved buried residues that affect normal packing of the hydrophobic side chains, 2 non-conserved buried residues that affect structure, 11 conserved exposed residues and 6 non-conserved exposed residues. The effects of all 34 missense mutations can be rationalized by predictable disruptions of FVIII structure while at most four mutations (S2069F, T2154I, R2209Q/G/L and E2181D) may affect residues directly involved in intermolecular interactions of FVIII/VIIIa with other coagulation factors or vWF.

Binding site for blood coagulation factor Xa involving residues 311-325 in factor Va.

Factor Va inactivation by activated protein C is associated with cleavages at Arg306, Arg506, and Arg679 with Arg306 cleavage causing the major activity loss. To study functional roles of the Arg306 region, overlapping 15-mer peptides representing the sequence of factor Va residues 271-345 were synthesized and screened for anticoagulant activities. The peptide containing residues 311-325 (VP311) noncompetitively inhibited prothrombin activation by factor Xa, but only in the presence of factor Va. Fluorescence studies showed that VP311 bound to fluorescence-labeled 5-dimethylaminonaphthalene-1-sulfonyl-Glu-Gly-Arg factor Xa in solution with a Kd of 70 microM. Diisopropylphosphoryl factor Xa and factor Xa but not factor VII/VIIa or prothrombin bound to immobilized VP311 with relatively high affinity. These results support the hypothesis that residues 311-325, which are positioned between the A1 and A2 domains of factor Va and likely exposed to solvent, contribute to the binding of factor Xa by factor Va. Based on this hypothesis, it is suggested that cleavage by activated protein C at Arg306 in factor Va not only severs the covalent connection between the A1 and A2 domains but also disrupts the environment and structure of residues 311-325, thereby down-regulating the binding of factor Xa to factor Va.

Homology models of the C domains of blood coagulation factors V and VIII: a proposed membrane binding mode for FV and FVIII C2 domains.

We present homology models of the C domains of coagulation factors V (FV) and VIII (FVIII). Using a threading approach, we identified the binding domain of galactose oxidase as an appropriate template for each C domain. The C1 and C2 domains of FV associate to form an elongated cylinder of 80A long and 30A diameter. The folding unit is a beta-sandwich with a long axis of 40A and a diameter of 30A. The current model allows us to propose a membrane binding mode for the C2 domains of FV and FVIII with three major characteristics: 1) solvent-exposed hydrophobic side chains from three loops at one end of the beta-sandwich are buried in the hydrophobic layer of the outer phospholipid leaflet; 2) a crown of positively charged residues is located in the polar zone of the phospholipid head groups; and 3) the long axis of the beta-sandwich of the C2 domain is perpendicular to the plane of the membrane. This proposal satisfies experimentally observed characteristics of membrane binding for the C2 domain and the light chain of FVa.

Nonenzymatic anticoagulant activity of the mutant serine protease Ser360Ala-activated protein C mediated by factor Va.

The human plasma serine protease, activated protein C (APC), primarily exerts its anticoagulant function by proteolytic inactivation of the blood coagulation cofactors Va and VIIIa. A recombinant active site Ser 360 to Ala mutation of protein C was prepared, and the mutant protein was expressed in human 293 kidney cells and purified. The activation peptide of the mutant protein C zymogen was cleaved by a snake venom activator, Protac C, but the "activated" S360A APC did not have amidolytic activity. However, it did exhibit significant anticoagulant activity both in clotting assays and in a purified protein assay system that measured prothrombinase activity. The S360A APC was compared to plasma-derived and wild-type recombinant APC. The anticoagulant activity of the mutant, but not native APC, was resistant to diisopropyl fluorophosphate, whereas all APCs were inhibited by monoclonal antibodies against APC. In contrast to native APC, S360A APC was not inactivated by serine protease inhibitors in plasma and did not bind to the highly reactive mutant protease inhibitor M358R alpha 1 antitrypsin. Since plasma serpins provide the major mechanism for inactivating APC in vivo, this suggests that S360A APC would have a long half-life in vivo, with potential therapeutic advantages. S360A APC rapidly inhibited factor Va in a nonenzymatic manner since it apparently did not proteolyze factor Va. These data suggest that native APC may exhibit rapid nonenzymatic anticoagulant activity followed by enzymatic irreversible proteolysis of factor Va. The results of clotting assays and prothrombinase assays showed that S360A APC could not inhibit the variant Gln 506-FVa compared with normal Arg 506-FVa, suggesting that the active site of S360A APC binds to FVa at or near Arg 506.

Affinity coelectrophoresis for dissecting protein-RNA domain-domain interactions in a tRNA synthetase system.

The class I methionyl tRNA synthetase has a conserved N-terminal nucleotide binding fold which contains the active site, and a largely non-conserved C-terminal anticodon binding domain. At the C-terminal end of the anticodon binding domain is a peptide which curls back into the N-terminal nucleotide binding fold near the active site. We showed that a mutation in this peptide disrupts aminoacylation and binding of a 7 base pair microhelix substrate based on the acceptor stem of tRNA(fMet). The novel technique of affinity coelectrophoresis was applied to this system for the first time to determine dissociation constants of wild-type and mutant MetRS for small RNA substrates. A description and evaluation of this technique for measuring weak protein-nucleic acid interactions is presented here, in the context of the methionyl tRNA synthetase system.

Evidence that specificity of microhelix charging by a class I tRNA synthetase occurs in the transition state of catalysis.

Determinants for the identities of tRNAs are located in the acceptor stem and, commonly, in the anticodon as well. Although the anticodon is an important determinant for the identity of methionine tRNA, RNA microhelices whose sequences are based on the acceptor stem alone can be aminoacylated by the class I methionyl-tRNA synthetase. We show here that specific nucleotide substitutions in a microhelix significantly reduced its rate of aminoacylation. In contrast, affinity coelectrophoresis analysis showed that microhelix binding to the enzyme was not significantly affected by the same substitutions. These and additional experiments and considerations imply that specific determinants for microhelix aminoacylation are needed for orientation of the acceptor stem in the transition state of catalysis rather than for enhanced binding interactions. The effect of linking together acceptor stem interactions with those in the anticodon, as occurs in the whole tRNA molecule, was also evaluated. This analysis showed that linkage results in some of the favorable acceptor stem and anticodon interactions being used to offset the free energy cost of straining the structure of the enzyme-tRNA complex.

Isolated RNA binding domain of a class I tRNA synthetase.

Class I tRNA synthetases typically have two major domains consisting of a class-defining N-terminal nucleotide binding fold, which contains the active site, and an idiosyncratic C-terminal domain, which in many instances provides for interactions with the tRNA anticodon. Whether the C-terminal domain can function in specific RNA binding when disconnected from the catalytic core is not known. We fused the anticodon binding domain of Escherichia coli methionyl tRNA synthetase to maltose binding protein. This fusion protein and the released, isolated domain are stable and have native-like structural characteristics, as shown by circular dichroism and thermal denaturation studies. Both the fusion protein and the isolated domain bind specifically to a small RNA hairpin oligonucleotide that recapitulates the anticodon stem-loop of tRNAfMet. Neither protein binds to an RNA hairpin with a point mutation in the anticodon trinucleotide. The binding specificity and affinity of these proteins duplicate those of the interaction between methionyl tRNA synthetase and the anticodon stem-loop oligonucleotide. Thus, the anticodon binding domain is functionally independent of the class-defining catalytic core and can be joined to another protein with little change in RNA binding characteristics.

C-terminal peptide appendix in a class I tRNA synthetase needed for acceptor-helix contacts and microhelix aminoacylation.

The 10 class I tRNA synthetases have an N-terminal nucleotide-binding fold which contains the catalytic center. Insertions into the nucleotide-binding fold provide contacts for acceptor-helix interactions, which stabilize the amino acid acceptor end of the tRNA substrate in the active site. A separate and largely nonconserved C-terminal domain provides contacts with distal parts of the tRNA, including the anticodon. For Escherichia coli methionyl tRNA synthetase, whose structure is known, the C-terminal domain is predominantly alpha-helical and forms a loop which interacts with the anticodon trinucleotide located about 76 A from the amino acid attachment site. Fused to the end of this helical domain is a peptide which curls back into the N-terminal nucleotide-binding fold and region of the active site. We show here that mutations in this peptide appendix disrupt aminoacylation and binding of a 7 base pair microhelix substrate based on the acceptor stem of tRNA(fMet), without affecting interactions with ATP or methionine or with the tRNA(fMet) anticodon. The impairment of acceptor-helix interactions by mutation of the C-terminal peptide can offset favorable anticodon interactions and severely reduce aminoacylation of tRNA(fMet). Thus, in addition to, or as an alternative to, acceptor-helix-binding insertions into the N-terminal nucleotide-binding fold, C-terminal peptide epitopes in some class I enzymes may provide a mechanism for facilitating RNA microhelix interactions with the catalytic site.