Gene-based FVIIa prophylaxis modulates the spontaneous bleeding phenotype of hemophilia A rats.

A sizable proportion of hemophilia inhibitor patients fails immune tolerance induction and requires bypass agents for long-term bleed management. Recombinant human-activated coagulation Factor VII (rhFVIIa) is an on-demand bypass hemostatic agent for bleeds in hemophilia inhibitor patients. Prophylactic use of rhFVIIa may enable sustained hemostatic management of inhibitor patients, but the critical relationship of rhFVIIa circulating levels and clinical outcome in that setting remains unclear. To address this in vivo, we used the rat hemophilia A (HA) model that exhibits spontaneous bleeds and allows longitudinal studies with sufficient statistical power. We simulated activated Factor VII (FVIIa) prophylaxis by adeno-associated virus (AAV) gene transfer of a rat FVIIa transgene. Compared with naive HA animals, rat FVIIa continuous expression affected the overall observed bleeds, which were resolved with on-demand administration of recombinant rat FVIIa. Specifically, although 91% of naive animals exhibited bleeds, this was reduced to 83% and 33% in animals expressing less than 708 ng/mL (<14 nM) and at least 708 ng/mL (≥14 nM) rat FVIIa, respectively. No bleeds occurred in animals expressing higher than 1250 ng/mL (>25 nM). Rat FVIIa expression of at least 708 ng/mL was also sufficient to normalize the blood loss after a tail vein injury. Continuous, AAV-mediated rat FVIIa transgene expression had no apparent adverse effects in the hemostatic system of HA rats. This work establishes for the first time a dose dependency and threshold of circulating FVIIa antigen levels for reduction or complete elimination of bleeds in a setting of FVIIa-based HA prophylaxis.

Prevention of beta strand movement into a zymogen-like position does not confer higher activity to coagulation factor VIIa.

Coagulation factor VIIa (FVIIa) belongs to the chymotrypsin family of S1 peptidases, whose members require the cleavage of at least one internal peptide bond to attain an active conformation. FVIIa also requires association with tissue factor (TF) to attain full catalytic competency. Without this, FVIIa has very low activity toward peptide and physiologic substrates. Reregistration of beta strands has been suggested to play a part in the activation of FVII, and their positioning is possibly important for the active conformation of FVIIa. To scrutinize this hypothesis, we have designed FVIIa variants which prevent beta strand movement and lock FVIIa in the alleged active conformation. The V299M mutation, alone or combined with the L280I mutation, was introduced to alter the first of two Leu-X-Val motifs in beta strand B2 and thereby prevent reregistration. Along the same line, C164V/V299C-FVIIa has a new disulfide which would keep beta strand B2 in the registration of active FVIIa. The amidolytic and proteolytic activities of V299M-, L280I/V299M-, and C164V/V299C-FVIIa were indistinguishable from or lower than those of wild-type FVIIa, and none of the mutants displayed an altered exposure of the N-terminal amino group of the protease domain. Moreover, the affinities of mutant and native FVIIa for TF increased to a similar extent upon incorporation of an active site inhibitor, and the enzymatic activities were equally stimulated by TF. In conclusion, we found no evidence that the mutants were in a more active state than native FVIIa. Thus, the proposed beta strand reregistration, if part of the regulatory mechanism governing FVIIa activity, apparently does not suffice for the transformation of FVIIa into an enzymatically active conformation. Our data raise the possibility that the structural differences between enzymatically latent (zymogen-like) and active FVIIa resemble those between trypsinogen and trypsin.

Large-scale production and properties of human plasma-derived activated Factor VII concentrate.

BACKGROUND AND OBJECTIVES: An activated Factor VII (FVIIa) concentrate, prepared from human plasma on a large scale, has to date not been available for clinical use for haemophiliacs with antibodies against FVIII and FIX. In the present study, we attempted to establish a large-scale manufacturing process to obtain plasma-derived FVIIa concentrate with high recovery and safety, and to characterize its biochemical and biological properties. MATERIALS AND METHODS: FVII was purified from human cryoprecipitate-poor plasma, by a combination of anion exchange and immunoaffinity chromatography, using Ca2+-dependent anti-FVII monoclonal antibody. To activate FVII, a FVII preparation that was nanofiltered using a Bemberg Microporous Membrane-15 nm was partially converted to FVIIa by autoactivation on an anion-exchange resin. The residual FVII in the FVII and FVIIa mixture was completely activated by further incubating the mixture in the presence of Ca2+ for 18 h at 10 degrees C, without any additional activators. For preparation of the FVIIa concentrate, after dialysis of FVIIa against 20 mm citrate, pH 6.9, containing 13 mm glycine and 240 mm NaCl, the FVIIa preparation was supplemented with 2.5% human albumin (which was first pasteurized at 60 degrees C for 10 h) and lyophilized in vials. To inactivate viruses contaminating the FVIIa concentrate, the lyophilized product was further heated at 65 degrees C for 96 h in a water bath. RESULTS: Total recovery of FVII from 15 000 l of plasma was approximately 40%, and the FVII preparation was fully converted to FVIIa with trace amounts of degraded products (FVIIabeta and FVIIagamma). The specific activity of the FVIIa was approximately 40 U/ micro g. Furthermore, virus-spiking tests demonstrated that immunoaffinity chromatography, nanofiltration and dry-heating effectively removed and inactivated the spiked viruses in the FVIIa. These results indicated that the FVIIa concentrate had both high specific activity and safety. CONCLUSIONS: We established a large-scale manufacturing process of human plasma-derived FVIIa concentrate with a high yield, making it possible to provide sufficient FVIIa concentrate for use in haemophiliacs with inhibitory antibodies.

Substitution of aspartic acid for methionine-306 in factor VIIa abolishes the allosteric linkage between the active site and the binding interface with tissue factor.

The enzyme factor VIIa (FVIIa) triggers the blood coagulation cascade upon association with tissue factor (TF). The TF-induced allosteric enhancement of FVIIa's activity contributes to the procoagulant activity of the complex, and Met-306 in the serine protease domain of FVIIa participates in this event. We have characterized FVIIa variants mutated in position 306 with respect to their ability to be stimulated by TF. The amidolytic activity of FVIIa mutants with Ser, Thr, and Asn in position 306 was stimulated 9-, 12-, and 7-fold, respectively, by soluble TF as compared to 22-fold for wild-type FVIIa. In contrast, the activity of Met306Asp-FVIIa only increased about 2-fold and that of Met306Asp/Asp309Ser-FVIIa increased about 1.5-fold. Modeling suggests that Asp in position 306 prevents the TF-induced stimulation of FVIIa by disrupting essential intermolecular hydrogen bonds. The ability of the FVIIa variants to catalyze factor X activation and the amidolytic activity were enhanced to a similar extent by soluble TF. This indicates that factor X does not promote its own activation through interactions with exosites on FVIIa made accessible upon FVIIa-TF assembly. Met306Asp-FVIIa binds soluble TF with a dissociation constant of 13 nM (about 3-fold higher than that of FVIIa), and, in sharp contrast to FVIIa, its binding kinetics are unaltered after inactivation with D-Phe-Phe-Arg chloromethyl ketone. We conclude that a single specific amino acid replacement, substitution of Asp for Met-306, virtually prevents the TF-induced allosteric changes which normally result in dramatically increased FVIIa activity and eliminates the effect of the active site inhibitor on TF affinity.

Structure of human factor VIIa and its implications for the triggering of blood coagulation.

Factor VIIa (EC 3.4.21.21) is a trypsin-like serine protease that plays a key role in the blood coagulation cascade. On injury, factor VIIa forms a complex with its allosteric regulator, tissue factor, and initiates blood clotting. Although the structure of the binary complex has already been determined [Banner, D. W., D'Arcy, A., Chène, C., Winkler, F. K., Guha, A., Konigsberg, W. H., Nemerson, Y. & Kirchhofer, D. (1996) Nature (London) 380, 41-46], the conformational effects of cofactor binding to factor VIIa are not known in detail because of a lack of structural information on free factor VIIa. Here we report the structure of gamma-carboxyglutamic acid-domainless human coagulation factor VIIa at a resolution of 2.8 A. The molecule adopts an extended conformation within the crystal similar to that previously observed for the full-length protein in complex with tissue factor. Detailed comparison of free and tissue factor-bound factor VIIa reveals several structural differences. The binding mode of the active-site inhibitor D-Phe-Phe-Arg methyl ketone differs in the two structures, suggesting a role for the cofactor in substrate recognition. More importantly, a surface-exposed alpha-helix in the protease domain (residues 307-312), which is located at the cofactor recognition site, is distorted in the free form of factor VIIa. This subtle structural difference sheds light on the mechanism of the dramatic tissue factor-induced enhancement of factor VIIa activity.

Proteolysis of blood coagulation factor VIII by the factor VIIa-tissue factor complex: generation of an inactive factor VIII cofactor.

Activation of factor VIII by thrombin occurs via limited proteolysis at R372, R740, and R1689. The resultant active factor VIIIa molecule consists of three noncovalently associated subunits: A1-a1, A2-a2, and A3-C1-C2 (50, 45, and 73 kDa respectively). Further proteolysis of factor VIIIa at R336 and R562 by activated protein C subsequently inactivates this cofactor. We now find that the factor VIIa-tissue factor complex (VIIa-TF/PL), the trigger of blood coagulation with restricted substrate specificity, can also catalyze limited proteolysis of factor VIII. Proteolysis of factor VIII was observed at 10 sites, producing 2 major fragments (47 and 45 kDa) recognized by an anti-factor VIII A2 domain antibody. Time courses indicated the slow conversion of the large fragment to 45 kDa, followed by further degradation into at least two smaller fragments. N-Terminal sequencing along with time courses of proteolysis indicated that VIIa-TF/PL cleaved factor VIII first at R740, followed by concomitant cleavage at R336 and R372. Although cleavage of the light chain at R1689 was observed, the majority remained uncleaved after 17 h. Consistent with this, only a transient 2-fold increase in factor VIII clotting activity was observed. Thus, heavy chain cleavage of factor VIII by VIIa-TF/PL produces an inactive factor VIII cofactor no longer capable of activation by thrombin. In addition, VIIa-TF/PL was found to inactivate thrombin-activated factor VIII. We hypothesize that these proteolyses may constitute an alternative pathway to regulate coagulation under certain conditions. In addition, the ability of VIIa-TF/PL to cleave factor VIII at 10 sites greatly expands the known protein substrate sequences recognized by this enzyme-cofactor complex.

[Recombinant activated factor VII: a new treatment for hemophilia]. / Le facteur VII activé recombinant: un nouveau traitement de l'hémophilie.

Recombinant factor VIIa (NovoSeven, NovoNordisk) initiates coagulation in inhibitor patients who either develop alloantibodies to clotting factors or those with acquired haemophilia. The human gene for FVII was transfected into a baby hamster kidney (BHK) cell line which secretes FVII in a single-chain form. Recombinant FVII is purified from the medium by four chromatographic steps including immunoaffinity chromatography with a monoclonal anti-FVII. During this process rFVII undergoes autoactivation to rFVIIa. NovoSeven does not contain any stabilising protein including human albumin. The SA is 50 KU/mg, the concentration is 0.6 mg/mL. Recovery is 45%, clearance is 31 mL/h/kg, the half-life was estimated between 2 and 3 hours. Combined to tissue factor rFVIIa directly activates factor X without implication of FVIIIa and FIXa. There is no systemic coagulation activation. Efficacy was assessed from data derived largely from the Compassionate Use Programme and clinical studies using rFVIIa as first line therapy. Various situations were proven to be efficiently resolved with rFVIIa including surgeries. Usual dosages are 90-120 mcg/kg administered every 2-3 hours. There were no side effects even in cases of prolonged administration. Drawbacks are the short half-life and the cost of the product.

Roles of the membrane-interactive regions of factor VIIa and tissue factor. The factor VIIa Gla domain is dispensable for binding to tissue factor but important for activation of factor X.

The roles of the putative membrane-interactive regions of factor VIIa (fVIIa) and tissue factor (TF) have been examined. Enzymatic removal of the 4-carboxyglutamic acid (Gla) domain of fVIIa had no effect on hydrolysis of a tripeptidyl chromogenic substrate in the absence or presence of TF. Additionally, Gla-domainless fVIIa (GdVIIa) was similar to native fVIIa in activating factor X in the absence of TF and phospholipid. However, GdVIIa in complex with recombinant soluble TF (sTF) was 76-fold less efficient in factor X activation than was fVIIa.sTF. The difference increased to 740-fold using TF relipidated in vesicles composed of 80% phosphatidylcholine and 20% phosphatidylserine (TF/PCPS). While Gla domain deletion produced a 10(3)-fold increase in the Kd for binding to TF/PCPS, the Kd for binding to TF/PC increased only 20-fold, and that for sTF in the absence of phospholipid increased 10-fold. Kd values for GdVIIa binding to TF/PCPS, TF/PC, or sTF were nearly identical. Thus, most of the binding energy required for formation of the fVIIa.TF complex was present even after Gla domain deletion. Both fVIIa and GdVIIa were capable of binding sTF in the presence of excess divalent metal-ion chelator, suggesting Ca(2+)-independent binding or the presence of a novel very high affinity Ca2+ binding site in fVIIa. The results demonstrate that the effect of the Gla domain on the Kd is apparent only in the presence of PS, and that interactions involving the fVIIa Gla domain and phospholipid are critical for efficient proteolysis of factor X on a membrane surface.

Isolation and characterization of proteolytic fragments of human factor VIIa which inhibit the tissue factor-enhanced amidolytic activity of factor VIIa.

The interaction of circulating factor VII/VIIa with tissue factor presented by cells in extravascular tissues represents the initial event in the extrinsic pathway of blood coagulation. To determine the tissue factor binding domains in human factor VIIa, we have subjected recombinant human factor VIIa to tryptic digestion and isolated two proteolytic fragments (molecular mass = 32 and 20 kDa) by a combination of immunoaffinity chromatography and reversed phase high performance liquid chromatography (HPLC) which strongly inhibits the tissue factor-enhanced amidolytic activity of factor VIIa and inhibits the activation of factor X by factor VIIa in the presence of tissue factor. The 32-kDa factor VIIa fragment consisted of residues 1-137/143 from the light chain of factor VIIa connected by a disulfide bond to residues 153-277 from the heavy chain. The 20-kDa factor VIIa fragment consisted of residues 1-137 of the light chain of factor VIIa in disulfide linkage with residues 248-266 of the heavy chain. The 32- and 20-kDa factor VIIa fragments inhibited the tissue factor apoprotein-enhanced factor VIIa amidolytic activity with Ki values of 35 and 65 nM, respectively. The Ki values for the inhibition of relipidated tissue factor apoprotein-enhanced factor VIIa amidolytic activity by the 32- and 20-kDa factor VIIa fragments were 70 and 610 nM, respectively. Factor X activation by factor VIIa-relipidated tissue factor was inhibited half-maximally by the 32- and 20-kDa factor VIIa fragments at 65 and 680 nM concentrations, respectively. Equilibrium binding studies indicated that the 32- and 20-kDa factor VIIa fragments interacted with cell surface tissue factor expressed on J82 cells in a specific and saturable manner with Kd values of 30 and 64 nM, respectively. In addition, a peptide consisting of residues 1-109 from the light chain of factor VIIa obtained by reduction and HPLC of the 20-kDa factor VIIa fragment retained inhibitory activity, but the selective removal of the gamma-carboxyglutamic acid domain from the 20-kDa factor VIIa fragment by cathepsin G cleavage resulted in the complete loss of inhibitory activity in this fragment. Our data strongly suggest that the epidermal growth factor-like domains covalently linked to the gamma-carboxyglutamic acid domain in factor VIIa constitute the high affinity tissue factor binding domain in this molecule.

Binding of factor VIIa to tissue factor permits rapid antithrombin III/heparin inhibition of factor VIIa.

Because free factor VIIa is inactivated only very slowly by a plasma concentration of antithrombin III (AT III) even in the presence of heparin, it has been assumed that AT III plays no significant role in regulating the initiation of tissue factor-dependent blood coagulation. However, in the present study, we present evidence that factor VIIa bound to tissue factor, unlike free factor VIIa, is readily inactivated by AT III in the presence of heparin. In a reaction mixture containing calcium ions and approximately equimolar concentrations of relipidated tissue factor (8.9 nmol/L) and factor VIIa (10 nmol/L), AT III (100 micrograms/mL) plus heparin (1 U/mL) inhibited 50% of the factor VIIa coagulant activity of the reaction mixture within 5 minutes. AT III/heparin was also shown to inhibit the catalytic activity towards factor X of factor VIIa/tissue factor complexes formed on monolayers of an ovarian carcinoma cell line (OC-2008) that constitutively expresses surface membrane tissue factor. AT III, even in the absence of exogenously added heparin, substantially inhibited the functional activity of factor VIIa/cell surface tissue factor complexes on intact monolayers. AT III alone and AT III/heparin, to a greater extent, also inhibited factor VIIa on "nonfunctional" factor VIIa/tissue factor complexes on intact monolayers, with resultant inhibition of their expression of factor VIIa/tissue factor catalytic activity toward factor X after cell lysis. The potential physiologic significance of these findings is discussed.

Deletion of the membrane anchoring region of tissue factor abolishes autoactivation of factor VII but not cofactor function. Analysis of a mutant with a selective deficiency in activity.

The activation of human blood coagulation factor VII can occur by the feedback activity of either factor VIIa (autoactivation) or factor Xa. Both of these reactions are known to be enhanced by the presence of tissue factor, an integral membrane protein and the cofactor for factor VIIa. We examine here the activation of 125I-factor VII by both factor VIIa and factor Xa employing a mutant soluble form of tissue factor which has had its transmembrane and cytoplasmic domains deleted (sTF1-219). This mutant soluble tissue factor retains cofactor activity toward factor VIIa in a single-stage clotting assay but shows a strong dependence on initial plasma levels of factor VIIa (from 1 to 10,000 ng/ml) when compared to wild-type tissue factor. We show that this dependence is due to a deficiency of sTF1-219 in ability to both promote autoactivation and enhance the factor Xa-catalyzed activation of 125I-factor VII. sTF1-219 does not, however, inhibit the tissue factor-independent activation of 125I-factor VII by factor Xa. The results strongly suggest that the phospholipid anchoring region of tissue factor is essential for autoactivation and beneficial for factor Xa-catalyzed activation of 125I-factor VII. In addition, when taken together with the dependence of clotting times on initial factor VIIa levels observed with sTF1-219, these results indicate that factor VII autoactivation may be of greater importance in the initiation of blood coagulation via tissue factor than has been previously realized.

The role of factor VII in haemostasis: infusion studies of factor VIIa in a canine model of factor VIII deficiency.

The role of factor VIIa in haemostasis has been studied using a canine model of factor VIII deficiency. Highly purified human factor VIIa was administered to dogs at a dosage of 0.5 microgram/kg. At selected times pre- and post-infusion, haemostasis was evaluated by the cuticle bleeding time. Plasma was collected for the assay of various parameters, including fibrinopeptide A (FPA) as a marker for thrombin generation in vivo. Factor VIIa infusion resulted in a 6-fold increase of factor VII clotting activity with a t1/2 of 2 h. FPA levels, which were 1.4 ng/ml before infusion, did not increase significantly in haemophilic dogs. In normal dogs, however, FPA levels rose to a mean value of 190 ng/ml 30 min post-infusion. It appeared that thrombin generation by factor VIIa infusion had occurred mainly via the intrinsic, factor VIII-dependent pathway. In factor VIII-deficient dogs, factor VIIa infusion did not correct cuticle bleeding, but an inconsistent haemostatic effect was observed 15-30 min post-infusion. Similar results were obtained in haemophilic dogs with circulating antibodies against factor VIII. The haemostatic effectivity could not be improved by increasing the factor VIIa dosage up to 40-fold. Although these data suggest that the extrinsic, factor VII-dependent factor X activation provides only a minor pathway of thrombin generation in vivo, it is possible that the suboptimal haemostatic effect noted may be promoted in bleeding situations where tissue factor availability is less limited. As such, factor VIIa may prove useful in the treatment of haemophilia A patients with acquired inhibitors to factor VIII.

Comparison of the effect of plasma-derived and recombinant human FVIIa in vitro and in a rabbit model.

Both purified plasma-derived FVIIa (pFVIIa) and recombinant FVIIa (rFVIIa) were shown to shorten the in vitro APTT in haemophilic plasma significantly. After addition of 3.8 micrograms/ml of pFVIIa or rFVIIa, corresponding to 110-140 FVII U/ml, the APTT was normalized in both haemophilia A and haemophilia B plasma. Furthermore, the prolonged APTT in plasma from a patient with acquired inhibitors against FVIII:C was normalized by addition of pFVIIa or rFVIIa in vitro. These results indicate an alternative pathway to the FVII or FVIIa-TF complex formation involving phospholipids (present in the APTT reagent). No signs of systemic activation of the coagulation system in rabbits following infusion of 500, 1000, 5000 U pFVIIa or rFVIIa/kg body weight, corresponding to about 150 micrograms/kg body weight, were seen. This corresponded to 2-3 times the clinical dose of FVIIa.