Human recombinant tissue factor pathway inhibitor induces apoptosis in cultured human endothelial cells.

Tissue factor pathway inhibitor (TFPI) is mainly synthesized in vascular endothelial cells and exhibits a strong and specific inhibitory activity against tissue factor-mediated blood coagulation. In the present study, we demonstrate that human recombinant TFPI (h-rTFPI) inhibits the growth of cultured human umbilical vein endothelial cells (HUVECs) by inducing apoptosis. In a growth-rate assay of HUVECs, the growth of the cultured HUVECs is completely abolished by the addition of 1 microM h-rTFPI to the culture medium containing fetal bovine serum (FBS), basic fibroblast growth factor, and epidermal growth factor. In addition, h-rTFPI and h-rTFPI-C which lacks the carboxyl-terminal basic region prevent the survival of growth-arrested HUVECs which are starved in a medium containing 2%, FBS alone, suggesting that h-rTFPI directly induces the death of these HUVECs. This hypothesis is supported by the finding that h-rTFPI does not inhibit the synthesis of DNA in HUVECs during proliferation, as shown by a 5-bromo-2'-deoxyuridine (BrdU) incorporation assay. Furthermore, Giemsa staining and a gel electrophoretic analysis of DNA fragmentation show that the HUVEC death mediated by h-rTFPI has the typical characteristics of apoptosis. However, the apoptosis in HUVECs is considerably inhibited in the presence of 1 microg/ml of the protein synthesis inhibitor, cycloheximide. Therefore, the process of apoptosis triggered by h-rTFPI is, at least in part, actively conducted by the cells.

Human recombinant tissue-factor pathway inhibitor prevents the proliferation of cultured human neonatal aortic smooth muscle cells.

Tissue-factor pathway inhibitor (TFPI) inhibits the procoagulant activity of the tissue-factor/factor VIIa complex. It was recently reported that TFPI prevented restenosis following tissue injury in a rabbit atherosclerotic model. In order to clarify the mechanism behind this successful prevention of restenosis, we investigated the direct effect of human recombinant TFPI (h-rTFPI) on the proliferation of cultured human neonatal aortic smooth muscle cells (hSMC). We found that h-rTFPI exhibits inhibitory activity toward hSMC proliferation, while h-rTFPI-C which lacks the carboxyl (C)-terminal region does not. Furthermore, we found that h-rTFPI binds to hSMCs with K(d) = 526 nM but that this binding is inhibited by the addition of the synthetic C-terminal peptide, Lys254-Met276, to h-rTFPI. Thus, the interaction of h-rTFPI with hSMCs mediated via the C-terminal region is responsible for the anti-proliferative action of h-rTFPI. On the basis of these results, we presume that the anti-proliferative effect of h-rTFPI in addition to its anticoagulant function plays a significant role in preventing restenosis following tissue injury.

Antithrombotic effect of human recombinant tissue factor pathway inhibitor on endotoxin-induced intravascular coagulation in rats: concerted effect with antithrombin.

In our current study, we examined the antithrombotic effect of Chinese hamster ovary cell-derived human recombinant tissue factor pathway inhibitor (h-rTFPI) by intravenous injection of h-rTFPI with or without antithrombin into endotoxin-treated rats. An injection of h-rTFPI at a high dose (4 mg/kg of h-rTFPI or three doses of 1 mg/kg) significantly prevented the decrease of fibrinogen and factor VIII and the increase of fibrin/fibrinogen degradation products and glutamic-pyruvic transaminase in rats, while a single injection of 1 mg/kg of h-rTFPI or 250 U/kg of antithrombin did not significantly prevent intravascular coagulation. However, a simultaneous injection of 1 mg/kg of h-rTFPI and 250 U/kg of antithrombin did significantly prevent intravascular coagulation. From the studies on the clearance rate and immunohistochemical staining of injected h-rTFPI into normal rats, we found that most of the intravenously-injected h-rTFPI was localized on the central vein and sinusoids in the liver and catabolized via the proteoglycan-mediated pathway with a half-life of 48 min. These results suggest that h-rTFPI and antithrombin prevented endotoxin-induced intravascular coagulation in concert by binding to the vascular walls of the liver and by inhibiting fibrin formation on Kupffer cells in hepatic sinusoids.

Effect of protein C and activated protein C on coagulation and fibrinolysis in normal human subjects.

Although protein C (PC) and activated protein C (APC) have been postulated to be useful for treating patients with thrombosis, their critical effect remains to be studied in human subjects. To examine whether purified PC or APC are useful for treating patients with thrombosis without showing any adverse effect, we studied effects on coagulation and fibrinolysis in normal human subjects. When highly purified human PC was administered intravenously to healthy subjects, plasma levels of immunoreactive PC decreased with a half-life of 10.9 h. Intravenously administered APC decreased with a half-life of 23 min as measured by prolongation of activated partial thromboplastin time (APTT). However, 1.7 h was obtained for the plasma half-life of APC when it was measured immunologically. These findings suggested that a significant fraction of the administered APC was rapidly inhibited by plasma inhibitor. Upon administration of APC, APTT was prolonged and plasma levels of clotting factor VIII (F-VIII) decreased transiently as measured by clotting assay. However, when determined by a chromogenic assay method in which 120-fold diluted plasma samples were used, plasma levels of F-VIII remained unchanged. Plasma levels of F-V did not decrease after APC administration. These findings suggested that prolongation of APTT and apparent decrease in plasma F-VIII clotting activity might be due to the in vitro-effect of APC present in plasma samples used. Diurnal fluctuation of plasminogen activator inhibitor in normal subject was not affected by administration of APC. Thus, PC or APC seems to function selectively at the site of thrombin-formation without lowering plasma levels of coagulation factors.

Chemical cross-linking of activated coagulation factor VII with soluble tissue factor: calcium ions are not essential for full amidolytic activity of the factor VIIa-tissue factor complex after complex formation.

In the previous study involving a yeast expression system, a high molecular mass extracellular domain of human tissue factor (denoted as sTF alpha) with a high content of mannose residues was produced in abundance and 37 kDa sTF beta was obtained in a low yield [Shigematsu et al. (1992) J. Biol. Chem. 267, 21329-21337]. To obtain sTF beta in a high yield, we constructed four kinds of mutant sTF with partial or total replacement of the N-potential glycosylation Asn residues with Ala, and expressed them in yeast. We found that the yield of the beta form of the Asn137-to-Ala mutant (designated as sTF beta NNA) was threefold higher (3 mg/liter) than that of the wild type, suggesting that the replacement of one of the three potential N-glycosylation Asn residues with Ala could be a good way to minimize the addition of mannose repeats. Since it has been reported that calcium ions are required for the effective hydrolysis of peptidyl substrates by the factor VIIa-sTF complex, it is believed to be essential for the expression of full protease activity. Here, we report the enzymatic characterization of a factor VIIa-sTF beta NNA complex cross-linked with a homobifunctional reagent, bis(sulfosuccinimidyl) suberate. The factor VIIa-sTF beta NNA complex cross-linked in the presence of 5 mM calcium ions or 50 mM EDTA was purified. The cross-linked complex did not show factor X activation in the presence of phospholipids. However, it showed essentially the same activity toward peptidyl substrates as before cross-linking, even in the presence of EDTA.(ABSTRACT TRUNCATED AT 250 WORDS)

Monoclonal antibody (VII-M31) to bovine factor VII: a specific epitope in the gamma-carboxyglutamic acid domain.

A murine monoclonal antibody (designated VII-M31) directed against bovine factor VII was prepared and characterized. Antibody VII-M31 inhibited the activations of both factors IX and X catalyzed by factor VIIa in the presence of tissue factor, phospholipids, and Ca2+. It possessed a strong affinity for factor VII in the presence of 5 mM Ca2+ (Kd = 1.12 x 10(-10)M). The immunoblotting test of other bovine proteins with the antibody, such as prothrombin, factor X, factor IX, protein C, protein S, and protein Z, in addition to human factor VII, revealed that it recognizes only a Ca2(+)-dependent epitope in bovine factor VII. Furthermore, this antibody VII-M31 covalently coupled with Affi-Gel allowed a simple and rapid purification of bovine factor VII. To localize the antigenic site in factor VII, various segments including a gamma-carboxyglutamic acid (Gla)-domainless protein, a Gla-domain peptide and the fragments isolated from the lysyl endopeptidase digest, were prepared. Among them, the isolated Gla-domain peptide and Gla-domainless factor VII were no longer recognized by antibody VII-M31, indicating that the sequence around the cleavage site by a-chymotrypsin is required for the interaction between the antibody and factor VII. In accordance with this result, the antibody bound specifically to a Gla-containing peptide corresponding to the NH2-terminal 23-50 residues of factor VII, which contains the chymotryptic cleavage site. These results suggest that the specific epitope of this antibody is localized in the carboxy-terminal 28 residues of the Gla-domain constituting the amino-terminal portion of bovine factor VII.

Amino acid sequence and inhibitory activity of rhesus monkey tissue factor pathway inhibitor (TFPI): comparison with human TFPI.

Rhesus monkey cDNA for tissue factor pathway inhibitor (TFPI) was cloned by means of the reverse transcriptase-polymerase chain reaction, using liver mRNA, and its nucleotide sequence was determined by sequencing five independent clones. Monkey TFPI was found to have a signal peptide of 28 amino acid residues and to be a mature protein of 276 amino acid residues, in which three and seventeen amino acid residue substitutions compared to human TFPI were found, respectively. All the cysteine residues, three putative carbohydrate-linked asparagine residues, and the P1 amino acid residues of each of the three Kunitz inhibitor domains were conserved in the two species. Recombinant monkey TFPI (rTFPI) was isolated from the culture medium of transformed Chinese hamster ovary cells. Amino acid sequence analysis and immunoblotting analysis, using polyclonal and monoclonal antibodies, showed that the carboxyl-terminal basic part of Rhesus monkey rTFPI had been truncated. The inhibitory activity of monkey rTFPI was compared with that of human rTFPI without the carboxyl-terminal basic part. The prothrombin time of human plasma was slightly more prolonged by the addition of monkey rTFPI than by that of human rTFPI. However, no significant differences were found between the potencies of human and monkey rTFPI as to the inhibition of factor Xa and tissue factor-factor VIIa complex.

Flow cytometric analysis of surface membrane proteins on activated platelets and platelet-derived microparticles from healthy and thrombasthenic individuals.

We used flow cytometry to investigate surface membrane protein expression by platelets and platelet-derived microparticles from normal individuals and a patient with Glanzmann's thrombasthenia. Microparticles were detected by both forward scatter and side scatter using FACScan. The binding of coagulation factors on microparticles was investigated by using monoclonal anti-Factor IX (IXa) and anti-Factor X (Xa) antibodies. Furthermore, the procoagulant activity of microparticles was measured with a chromogenic substrate (S-2222) using a microtiter enzyme-linked immunosorbent assay. Both types of platelets showed similar release of microparticles. Microparticles released from platelets after activation with the calcium ionophore A23187 did not bind factors IXa and Xa, but when purified factors Va and Xa were added to the incubation buffer, factor Xa binding increased markedly in both normal and thrombasthenic platelets. Both normal and thrombasthenic platelets showed a similar time-dependent release of microparticles when activated with A23187. However, the binding of an antibody to granule membrane protein-140 also increased time-dependently in normal microparticles, but was little increased in thrombasthenic microparticles. These findings suggest that glycoprotein IIb/IIIa does not participate in the expression of prothrombinase activity on the surface of activated platelets and microparticles, whereas this glycoprotein appears to have an important role in the movement of granule membrane protein-140 from platelets to microparticles.

Measurement of Lp(a) with a two-step monoclonal competitive sandwich ELISA method.

OBJECTIVE: To evaluate the results of Lipoprotein (a)[Lp(a)] measurements by a competitive two-step monoclonal enzyme-linked immuno sorbent assay method comparing them with those by a conventional ELISA. METHODS: Serum having various isoforms of Lp(a) and purified Lp(a) were assayed using the method described here and commercially available kits. The reference range was determined with the use of 324 normal subjects by means of calculation from Lp(a) results of logarithmic transformation. RESULTS: Our method takes advantage of a competitive reaction between fixed antibody and free antibody to Lp(a), having the detection range up to 1000 mg/L with the lowest detection limit of 2 mg/L. The anti-Lp(a) monoclonal antibody employed in the assay system reacts uniformly with all phenotypes of Lp(a) but showing very low cross-reactivity for plasminogen and LDL. Within-run and between-run precisions were excellent, giving CVs of 2.9 and 4.0% with mean values of 145 and 635 mg/L, respectively. In comparison of the results by our method with those by a polyclonal method (Biopool) or a monoclonal antibody method (Terumo), they correlated well; Y (our method) = 0.99 x (polyclonal method, Biopool) - 1.9, r = 0.994 (n = 60), and Y = 0.94 X(monoclonal method, Terumo) -9.8, r = 0.97 (n = 60), respectively. The reference range was 105.9 +/- 25.4 mg/L, the difference between the sexes was not significant. CONCLUSION: Our method has proven highly accurate and specific. It is applicable with auto analyzer because it does not require such a pre-dilution step as is necessary for Lp(a) determination by conventional ELISA assay. Accordingly, we can conclude that our test method is workable for both clinical laboratories and mass screening.

The antithrombotic effect of human activated protein C on He-Ne laser-induced thrombosis in rat mesenteric microvessels.

The effect of activated protein C (APC) on He-Ne laser-induced thrombosis in rat mesenteric microvessels was studied. APC inhibited platelet-rich thrombus formation in a dose-dependent manner more significantly in venules but also in arterioles. Inhibition of thrombus formation was maximal immediately after intravenous administration and persisted for 40 minutes in venules and 20 minutes in arterioles. The maximum effective blood concentration of APC was 1.04 micrograms/ml. A small but statistically significant prolongation of the activated partial thromboplastin time (APTT) was observed at this blood concentration, demonstrating the anticoagulant effect of APC. Our findings show that APC has a significant inhibitory effect on platelet thrombus formation and suggest that APC is a potent natural antithrombotic agent.

Characteristic effects of activated human protein C on tissue thromboplastin-induced disseminated intravascular coagulation in rabbits.

Protein C (PC) is the zymogen of an anticoagulant serine protease and is converted to its active form (activated protein C: APC) by thrombin in the presence of thrombomodulin. APC plays an important role in regulating thrombosis and fibrinolysis by inhibiting not only blood coagulation factors Va and VIIIa but also type-1 plasminogen activator inhibitor (PAI-1). In the present study we examined the effects of human APC on tissue thromboplastin-induced disseminated intravascular coagulation (DIC) in rabbits and compared them with those of heparin. Both APC (300-3000 U/kg) and heparin (100-300 IU/kg) inhibited the decreases in platelet count and fibrinogen level equally. APC improved the prolonged bleeding time, but heparin aggravated bleeding with potent prolongation of activated partial thromboplastin time (APTT). Furthermore, in APC-treated animals, fibrin deposition in glomeruli was less than in heparin-treated animals. This result that APC accelerated local fibrinolysis by neutralizing PAI-1. From our findings, we concluded that APC can improve both coagulation and fibrinolysis in a DIC model and should be useful for the clinical remedy of DIC without having an adverse side effect like a bleeding tendency.

Induction of acquired factor IX inhibitors in cynomolgus monkey (Macaca fascicularis): a new primate model of hemophilia B.

Inherited hemophilia dog and other transient hemophilic animal models have been used for evaluation of hemostatic agents for use in treatment of hemophilia. We established the first nonhuman primate hemophilic model by immunizing cynomolgus monkeys with human FIX (hFIX) in adjuvants. FIX activities of all three hFIX-immunized monkeys decreased transiently to less than 10% in accordance with prolongation of activated partial thromboplastin time (APTT). Forty micrograms of human factor VIIa (hFVIIa) per kilogram body weight (that was reported to be clinically effective) was administered to the monkey with the highest inhibitor titer to evaluate its usefulness as a hemophilia inhibitor model. Results of thromboelastography (TEG) after the injection demonstrated that the hemostatic effect of FVIIa in this model would be similar to that in hemophiliacs with inhibitors. The antibodies purified from the monkey's plasma by hFIX-immobilized gel were composed of two types: Ca(2+)-dependent and -independent antibodies, with features of IgG(1) and IgG(4). Both types of antibodies reacted to cynomolgus FIX, and only Ca(2+)-dependent antibodies also expressed inhibitory activity against cynomolgus FIX. Immunoblotting analyses of Ca(2+)-dependent antibodies using hFIX and its derivatives suggested that they recognized the Ca(2+)-dependent conformation related to the gamma-carboxyglutamic acid (Gla) domain. Comparison of FIX cDNA from human, cynomolgus monkey, and other species, and the results of immunization of various animals (goats, beagle dogs, rabbits, and rats) with hFIX in adjuvants strongly suggested that the development of acquired FIX inhibitors in the monkeys might be due to high cross-reactivity of the antibodies to molecular mimic antigens, hFIX, and cynomolgus FIX.

A kinetic analysis of the interaction of human recombinant tissue factor pathway inhibitor with factor Xa utilizing and immunoassay and the effect of antithrombin III/heparin on the complex formation.

We have recently shown that a complex formation of tissue factor pathway inhibitor (TFPI) and factor Xa (Xa) promotes a clearance of proteoglycans-associated TFPI. In the current studies, the interaction between human recombinant TFPI (h-rTFPI) and Xa were kinetically analyzed by utilizing both a protease inhibitor, p-(amidophenyl) methanesulfonyl fluoride hydrochloride, and a specific enzyme-linked immunosorbent assay for the complex of h-rTFPI with Xa. We further investigated the effect of antithrombin III on the complex formation between h-rTFPI and Xa. We found that the h-rTFPI/Xa complex formed in a time-dependent manner: the second-order rate constant (K1) for the complex formation was calculated to be 0.86x10(6) M(-1)s(-1). The addition of antithrombin III to the h-rTFPI solution modestly reduced the rate of the complex formation between h-rTFPI and Xa. Heparin strikingly enhanced antithrombin III's inhibition of Xa and resulted in complete abrogation of the complex formation between h-rTFPI and Xa in the absence or presence of acidic phospholipids. Furthermore, antithrombin III induced dissociation of the preformed h-rTFPI/Xa complex in the presence of heparin. These results suggest that in the presence of heparin, antithrombin III interferes with the catabolism of TFPI mediated via Xa.

Large-scale preparation of human thrombin: polyethylene glycol potentiates the factor Xa-mediated activation of prothrombin.

We investigated the ability of polyethylene glycol 4000 to accelerate thrombin generation in a mixture of prothrombin and factor X at concentrations of 1-30%. In the presence of 5 mM of CaCl2, polyethylene glycol 4000 promoted prothrombin activation at concentrations above 1%. The peak of activation was seen at levels of 14 and 20% of polyethylene glycol 4000. The effect of the polyethylene glycol was remarkably dependent on its molecular weight; molecular weights greater than 2000 were required for accelerating thrombin generation. Under optimal conditions, polyethylene glycol 4000, in the presence of CaCl2, promoted conversion of all of the prothrombin into thrombin and its derivatives. We conclude that polyethylene glycol 4000, at concentrations ranging from 14 to 20%, effectively accelerates thrombin generation in the presence of 5 mM of CaCl2. This new method for preparing thrombin is based on the use of polyethylene glycol 4000 and CaCl2 and is applicable to the manufacture of thrombin.

The clearance of proteoglycan-associated human recombinant tissue factor pathway inhibitor (h-rTFPI) in rabbits: a complex formation of h-rTFPI with factor Xa promotes a clearance rate of h-rTFPI.

The very rapid clearance of human recombinant tissue factor pathway inhibitor (h-rTFPI) may result from its binding to vascular proteogly can and LDL receptor-related protein (LRP). To investigate the effect of factor Xa on the clearance of h-rTFPI, we developed a specific ELISA for h-rTFPI/factor Xa complex, and compared the pharmacokinetic parameters of h-rTFPI/factor Xa complex and the clearance rate of the cellular proteogly can-associated h-rTFPI/factor Xa complex with those of h-rTFPI by itself in rabbits. We found that the h-rTFPI/factor Xa complex disappeared from circulation at a rapid rate of clearance, having pharmacokinetic parameters similar to those of non-complexed h-rTFPI. After the rapid disappearance of the h-rTFPI complex from plasma, an intravenous injection of heparin resulted in a release of h-rTFPI/factor Xa complex into plasma. However, the recovery of heparin-releasable h-rTFPI/factor Xa decreased significantly in a time-dependent manner. Therefore, we examined the half-life of proteogly can-associated h-rTFPI/factor Xa and determined it to be 51 min, which was significantly shorter than that of h-rTFPI by itself (107 min). These results suggest that a complex formation of h-rTFPI with factor Xa promotes a clearance of proteogly can-associated h-rTFPI existing in the liver and kidney.

Synthesis of monolayer platinum nanosheets.

Platinum (Pt) nanosheets were developed by exfoliating layered platinum oxide. Moreover, we succeeded in synthesizing monolayer Pt nanosheets for the first time by adjusting the conditions for reduction. Monolayer Pt nanosheets were highly active in oxygen reduction reaction.

Comparison of hemorrhagic effect of heparin and human activated protein C with use of Thrombostat 4000.

The importance of bleeding as a complication of anticoagulant therapy is clearly recognized. We previously reported that amelioration of hemorrhage associated with disseminated intravascular coagulation by the human activated protein C (APC) was greater than that by heparin. In this study, we compared the bleeding complication of intravenously administered APC and heparin in rabbits, and also estimated primary hemostasis. When both anticoagulants were intravenously infused, the bleeding time from a punctured ear vein was prolonged dose-dependently. However, at doses which prolonged the activated partial thromboplastin time nearly equally, the prolongation of bleeding was greater in heparin-administered rabbits. Blood withdrawn from heparin-administered animals showed increases in in vitro bleeding parameters which correlated with the in vivo bleeding time. However, only small changes were observed in the blood withdrawn from APC-administered animals. Both drugs induced either no change or only a slight decrease in the platelet count, hematocrit and fibrinogen content. These observations suggest that APC may be a more useful anticoagulant than heparin since it causes less bleeding tendency.

[Role of protein C in endotoxin-induced release of plasminogen activator inhibitor from endothelial cell].

To elucidate the role of protein C (PC) in the release of plasminogen activator inhibitor (PAI) from endothelial cells, the effect of PC and activated protein C (APC) on plasma levels of PAI in endotoxin (ET)-treated rats was examined. When activated by snake venom, human PC significantly prolonged the activated partial thromboplastin time (APTT) of both human and rat plasma samples. Addition of APC also prolonged the APTT of both human and rat plasma samples. PAI activity in plasma from septicemic patients and ET-treated rats was neutralized by APC. A small dose of ET (0.1 microgram/kg) gradually increased plasma PAI activity, which became maximum 3h after ET-treatment. APC administered prior to ET-treatment, PC decreased PAI activity, however, no such inhibition was seen when administered after ET-treatment. A significant negative correlation between PC concentrations and PAI activities was observed in plasma from septicemic patients. These findings indicated that activation of PC on endothelial surface plays a regulatory role in releasing PAI and that endotoxin might inhibit the surface activation of PC.