Factor Xa and VIIa inhibition by tissue factor pathway inhibitor is prevented by a monoclonal antibody to its Kunitz-1 domain.

Essentials Activated FVII (FVIIa) and FX (FXa) are inhibited by tissue factor pathway inhibitor (TFPI). A monoclonal antibody, mAb2F22, was raised against the N-terminal fragment of TFPI (1-79). mAb2F22 bound exclusively to the K1 domain of TFPI (KD ∼1 nm) and not to the K2 domain. mAb2F22 interfered with inhibition of both FVIIa and FXa activities and restored clot formation. SUMMARY: Background Initiation of coagulation is induced by binding of activated factor VII (FVIIa) to tissue factor (TF) and activation of factor X (FX) in a process regulated by tissue factor pathway inhibitor (TFPI). TFPI contains three Kunitz-type protease inhibitor domains (K1-K3), of which K1 and K2 block the active sites of FVIIa and FXa, respectively. Objective To produce a monoclonal antibody (mAb) directed towards K1, to characterize the binding epitope, and to study its effect on TFPI inhibition. Methods A monoclonal antibody, mAb2F22, was raised against the N-terminal TFPI(1-79) fragment. Binding data were obtained by surface plasmon resonance analysis. The Fab-fragment of mAb2F22, Fab2F22, was expressed and the structure of its complex with TFPI(1-79) determined by X-ray crystallography. Effects of mAb2F22 on TFPI inhibition were measured in buffer- and plasma-based systems. Results mAb2F22 bound exclusively to K1 of TFPI (KD ~1 nm) and not to K2. The crystal structure of Fab2F22/TFPI (1-79) mapped an epitope on K1 including seven residues upstream of the domain. TFPI inhibition of TF/FVIIa amidolytic activity was neutralized by mAb2F22, although the binding epitope on K1 did not include the P1 residue. Binding of mAb2F22 to K1 blocked TFPI inhibition of the FXa amidolytic activity and normalized hemostasis in hemophilia human A-like plasma and whole blood. Conclusion mAb2F22 blocked TFPI inhibition of both FVIIa and FXa activities and mapped a FXa exosite for binding to K1. It reversed TFPI feedback inhibition of TF/FVIIa-induced coagulation and restored clot formation in FVIII-neutralized human plasma and blood.

Concizumab, an anti-tissue factor pathway inhibitor antibody, induces increased thrombin generation in plasma from haemophilia patients and healthy subjects measured by the thrombin generation assay.

AIMS: Concizumab, a humanized monoclonal antibody against tissue factor pathway inhibitor (TFPI), is being developed as a subcutaneously (s.c.) administered treatment for haemophilia. It demonstrated a concentration-dependent procoagulant effect in functional TFPI assays; however, global haemostatic assays, such as the thrombin generation assay (TGA), offer a more complete picture of coagulation. We investigated how concizumab affects thrombin generation following ex vivo spiking in plasma from haemophilia patients using the TGA, and if the assay can detect the effect of multiple s.c. concizumab doses in healthy subjects. METHODS: For the ex vivo spiking study, platelet-poor plasma (PPP) from 18 patients with severe haemophilia was spiked with 0.001-500 nm concizumab. For the multiple-dosing study, four healthy males received concizumab 250 µg kg-1 s.c. every other day for eight doses; blood was collected before and after dosing and processed into PPP. In both studies, thrombin generation was measured using a Calibrated Automated Thrombogram® system with 1 pm tissue factor. RESULTS: In spiked samples from haemophilia patients, peak thrombin and endogenous thrombin potential (ETP) increased concentration dependently, reaching near-normal levels at concizumab concentrations >10 nm. Repeated s.c. doses of concizumab in healthy subjects increased both peak thrombin and ETP; these effects were sustained throughout the dosing interval. CONCLUSIONS: Thrombin generation assay demonstrated increased thrombin generation with concizumab after ex vivo spiking of haemophilia plasma and multiple s.c. doses in healthy subjects, supporting both the utility of the TGA in evaluating concizumab treatment and the potential of s.c. concizumab as a novel haemophilia therapy.

Thrombin generation assay using factor XIa to measure factors VIII and IX and their glycoPEGylated derivatives is robust and sensitive.

BACKGROUND: Conventional coagulation factor assays are associated with certain limitations, as they do not always reflect the clinical heterogeneity of bleeding in hemophilic patients or correctly reflect the individual patient response to treatment with bypassing agents or novel factor concentrates. The thrombin generation assay (TGA) is currently being assessed as a possible method for characterizing bleeding phenotypes in individuals with hemophilia. OBJECTIVES: This study assessed the robustness and sensitivity of the TGA for measuring the activity of recombinant factor VIII (rFVIII), recombinant factor IX (rFIX) and their glycoPEGylated derivatives, N8-GP and N9-GP, in vitro. METHODS: Factor-deficient plasma was spiked with 0.13-130 IU dL(-1) rFVIII or N8-GP (hemophilia A [HA] plasma), or rFIX or N9-GP (hemophilia B [HB] plasma). A calibrated automated thrombogram triggered with tissue factor (TF) or activated FXI (FXIa) was used to measure thrombin generation over time. Endogenous thrombin potential, peak thrombin, velocity index, lag time and time to peak thrombin were analyzed. RESULTS: FXIa-triggered assays were not affected by glycoPEGylation and were sufficiently sensitive to differentiate between spiked samples mimicking severe and moderate HB and HA; TF-triggered assays were not sufficiently sensitive for this distinction in HA. Both FXIa-triggered and TF-triggered assays had an acceptable level of variability (≤ 20%), although TF-triggered assays were associated with greater variability. CONCLUSIONS: FXIa-triggered TGA reactions produced more robust and sensitive results than TF-triggered TGA reactions, and have the potential for use in monitoring patients treated with glycoPEGylated or non-PEGylated coagulation factor concentrates. These promising results merit confirmation with clinical samples to correlate in vitro and in vivo data.

Functional characteristics of N8, a new recombinant FVIII.

The aim of this study was to evaluate the in vitro function of the new recombinant factor VIII (FVIII) compound, N8. The specific activity of N8 as measured in a FVIII:C one-stage clot assay was 9300±400 IU mg(-1) based on the analysis of seven individual batches. The ratio between the FVIII:C activity measured in clot and chromogenic assays was 1.00 (95% confidence interval 0.97-1.03). N8 bound to von Willebrand factor with Kd values of 0.2 nm when measured by ELISA and by surface plasmon resonance. FVIIIa cofactor activity was determined from the kinetic parameters of factor IXa-catalysed factor X (FX) activation. The rate of activation of N8 by thrombin as well as Km and kcat for FX activation was in the same range as those observed for Advate®. The rate of activated protein C (APC)-catalysed inactivation was similar for activated N8 and Advate®. N8 improved thrombin generation in a dose-dependent manner and induced similar rates of thrombin generation as Advate® and the plasma-derived FVIII product Haemate®. Using thromboelastography (TEG®), N8 was shown to improve the clot formation and clot stability in whole blood from haemophilia A patients. Comparable potency and efficacy of N8 and Advate® was found based on TEG® parameters. Finally, similar binding profiles to immobilized lipoprotein receptor-related protein (LRP) of N8 and Advate® were observed. The study demonstrated that N8 is fully functional in a variety of assays measuring FVIII activity. No functional differences were found between N8 and comparator compounds.

Characterization and crystallization of an active N-terminally truncated form of the Escherichia coli glycogen branching enzyme.

The prokaryotic glycogen branching enzymes (GBE) can be divided into two groups on the basis of their primary structures: the first group of enzymes, which includes GBE from Escherichia coli, is characterized by a long N-terminal extension that is absent in the enzymes of the second group. The extension consists of approximately 100 amino-acid residues with unknown function. In order to characterize the function of this region, the 728 amino-acid residue, full-length E. coli GBE, and a truncated form (nGBE) missing the first 107 amino-acid residues were overexpressed in E. coli. Both enzymes were purified to homogeneity by a simple purification procedure involving ammonium sulphate precipitation, ion-exchange chromatography, and a second ammonium sulphate precipitation. Purified full-length enzyme was poorly soluble and formed aggregates, which were inactive, at concentrations above 1 mg.mL-1. In contrast, the truncated form could be concentrated to 6 mg.mL-1 without any visible signs of aggregation or loss of activity on concentration. The ability to overexpress nGBE in a highly soluble form has allowed us to produce diffracting crystals of a branching enzyme for the first time. A comparison of the specific activities of purified GBE and nGBE in assays where amylose was used as substrate demonstrated that nGBE retained approximately half of the branching activity of full-length GBE and is therefore a suitable model for the study of the enzymes' catalytic mechanism.

Tricistronic operon expression of the genes gcaD (tms), which encodes N-acetylglucosamine 1-phosphate uridyltransferase, prs, which encodes phosphoribosyl diphosphate synthetase, and ctc in vegetative cells of Bacillus subtilis.

The gcaD, prs, and ctc genes were shown to be organized as a tricistronic operon. The transcription of the prs gene, measured as phosphoribosyl diphosphate synthetase activity, and of the ctc gene, measured as beta-galactosidase activity specified by a ctc-lacZ protein fusion, were dependent on the promoter in front of the gcaD gene. Analysis of cDNA molecules prepared with gcaD-prs-ctc-specified mRNA as the template revealed an RNA transcript that encompassed all three cistrons.

Chemienzymatic synthesis of uridine nucleotides labeled with [15N] and [13C].

UTP, labeled with 15N and 13C (at all carbon atoms of the ribose moiety), was obtained enzymatically from [15N]uracil and [13C6]glucose. Eleven enzymes and suitable substrates reconstituted a metabolic pathway in which glucose was first transformed to 5-phosphoribosyl-1-pyrophosphate. The latter compound plus uracil yielded UMP in a second step by the reaction catalyzed by uracil phosphoribosyltransferase. UMP was subsequently phosphorylated to the corresponding di- and triphosphate. ATP, required for five phosphorylation reactions, was regenerated from creatine phosphate, whereas NADP+ necessary for the oxidation of glucose 6-phosphate and 6-phosphogluconate was recycled by glutamate dehydrogenase and excess of ammonia and alpha-oxoglutarate. Despite the number and complexity of the enzymatic steps, the synthesis of [15N, 13C]UTP is straightforward with an overall yield exceeding 60%. This method, extended and diversified to the synthesis of all natural ribonucleotides, is a more economical alternative for obtaining nucleic acids for structural analysis by heteronuclear NMR spectroscopy.

Inactivation of Escherichia coli phosphoribosylpyrophosphate synthetase by the 2',3'-dialdehyde derivative of ATP. Identification of active site lysines.

The enzyme 5-phosphoribosyl-alpha-1-pyrophosphate (PRPP) synthetase from Escherichia coli was irreversibly inactivated on exposure to the affinity analog 2',3'-dialdehyde ATP (oATP). The reaction displayed complex saturation kinetics with respect to oATP with an apparent KD of approximately 0.8 mM. Reaction with radioactive oATP demonstrated that complete inactivation of the enzyme corresponded to reaction at two or more sites with limiting stoichiometries of approximately 0.7 and 1.3 mol of oATP incorporated/mol of PRPP synthetase subunit. oATP served as a substrate in the presence of ribose-5-phosphate, and the enzyme could be protected against inactivation by ADP or ATP. Isolation of radioactive peptides from the enzyme modified with radioactive oATP, followed by automated Edman sequencing allowed identification of Lys181, Lys193, and Lys230 as probable sites of reaction with the analog. Cysteine 229 may also be labeled by oATP. Of these four residues, Lys193 is completely conserved within the family of PRPP synthetases, and Lys181 is found at a position in the sequence where the cognate amino acid (Asp181) in human isozyme I PRPP synthetase has been previously implicated in the regulation of enzymatic activity. These results imply a functional role for at least two of the identified amino acid residues.