von Willebrand factor and its propeptide: the influence of secretion and clearance on protein levels and the risk of venous thrombosis.

BACKGROUND AND OBJECTIVES: Elevated levels of factor (F)VIII are associated with an increased risk of thrombosis. FVIII levels are determined mainly by von Willebrand factor (VWF). We have investigated the contribution of secretion and clearance rates to the elevated VWF antigen (VWF:Ag) and to the risk of thrombosis. VWF is secreted in equimolar amounts with its propeptide, which has a shorter half-life. VWF propeptide can be used as a measure of VWF secretion and allows estimation of the VWF half-life. METHODS AND RESULTS: We have measured VWF propeptide, VWF:Ag, FVIII:Ag and FVIII activity (FVIII:C) in the Leiden Thrombophilia Study. In controls, high VWF propeptide was associated with high VWF:Ag, FVIII:Ag and FVIII:C. In contrast to mature VWF:Ag, VWF propeptide was not influenced by blood groups. Using an ELISA-based assay we have shown that VWF propeptide lacks ABO antigens. Levels were higher in men and increased with age. A long VWF half-life was also associated with high VWF:Ag, FVIII:Ag and FVIII:C. The VWF half-life was influenced by blood group (10 h in O vs. 12 h in non-O individuals), but not by sex, and only slightly by age. VWF propeptide was higher in thrombosis patients than in controls. The VWF half-life was similar in patients and controls (11.4 and 11.1 h, respectively). CONCLUSIONS: Both secretion and clearance rates are important determinants of VWF and FVIII levels. However, mainly high VWF and FVIII levels caused by increased secretion seem to be associated with thrombosis. ABO blood group influences the clearance rates of VWF rather than VWF secretion rates.

Association between thrombin activatable fibrinolysis inhibitor genotype and levels in plasma: comparison of different assays.

Thrombin activatable fibrinolysis inhibitor (TAFI) antigen levels exhibit a large interindividual variability in which genetic control seems to play a major role. However, recent reports have questioned the association between TAFI concentration and genotype, suggesting that variable antibody reactivity towards TAFI isoforms, particularly the Thr325Ile polymorphism (1040C/T), may lead to artefacts in TAFI antigen levels. In order to compare assay outcome we determined plasma TAFI levels in 92 healthy individuals, using an enzyme-linked immunosorbent assay (ELISA) (commercial antibodies), an electroimmunoassay (in-house antibodies) and a commercial chromogenic assay (Actichrome TAFI). Each individual was genotyped for the -438A/G and 1040C/T polymorphisms in the TAFI gene. TAFI levels were significantly associated with genotype in both antigen and chromogenic assays. All assays displayed significant correlations with each other. Linear regression and Bland-Altman agreement analysis in the genotype subgroups showed that neither the genotype nor the concentration affected the relationship between the Actichrome TAFI and the electroimmunoassay. In contrast, the ELISA/Actichrome TAFI and the ELISA/electroimmunoassay relationships were concentration- and genotype-dependent. Our results demonstrate that artefacts may arise when measuring TAFI antigen levels by ELISA. Nevertheless, the electroimmunoassay and the Actichrome TAFI assay support a genotype-related variation of TAFI concentration.

Alterations in the extrinsic pathway in hypertriglyceridemia do not cause a 'procoagulant state': effects of bezafibrate therapy.

Hypertriglyceridemia (HTG) is an independent risk factor for cardiovascular disease (CVD). Hemostatic variables [factor VII antigen (FVIIag), factor VII coagulant activity (FVIIc), activated factor VII (FVIIa), free and endothelial-associated (EC) tissue factor pathway inhibitor (TFPI) antigen, pre- and post-heparin total TFPI activity, EC-TFPI activity, prothrombin fragment 1 + 2 (F1 + 2), fibrinogen and D-dimer] were compared between 18 HTG patients and 20 controls to investigate whether HTG is associated with alterations in the extrinsic pathway and whether such alterations create a procoagulant state, as expressed by F1 + 2 and D-dimer levels. In addition, the effects of bezafibrate therapy (6 weeks, 400 mg/day) on these variables were studied in 18 HTG patients in a double-blind, placebo-controlled, cross-over study. FVIIag, FVIIc, free TFPI and fibrinogen were significantly higher in HTG patients (by 44, 30, 45 and 31%, respectively; all P < 0.02), while FVIIa, EC-TFPIag and activity, total TFPI activities, F1 + 2 and D-dimer levels were similar in patients and controls. Bezafibrate reduced serum TG and fibrinogen levels (by 62 and 20%, respectively; both P < 0.001), whereas the other hemostatic variables were unaffected. In conclusion, the observed alterations in the extrinsic pathway in HTG are not associated with a procoagulant state. In contrast, the presence of elevated fibrinogen levels in HTG might enhance the risk for CVD. Bezafibrate therapy improved the adverse lipid profile and decreased fibrinogen levels in HTG patients.

Thrombin activatable fibrinolysis inhibitor and the risk for deep vein thrombosis.

Thrombin activatable fibrinolysis inhibitor (TAFI, or procarboxypeptidase B) is the precursor of a recently described carboxypeptidase that potently attenuates fibrinolysis. Therefore, we hypothesized that elevated plasma TAFI levels induce a hypofibrinolytic state associated with an increased risk for venous thrombosis. To evaluate this hypothesis, we developed an electroimmunoassay for TAFI antigen and used this assay to measure TAFI levels in the Leiden Thrombophilia Study, a case-control study of venous thrombosis in 474 patients with a first deep vein thrombosis and 474 age- and sex-matched control subjects. In 474 healthy control subjects, an increase of TAFI with age was observed in women but not in men. Oral contraceptive use also increased the TAFI concentration. TAFI levels above the 90th percentile of the controls (> 122 U/dL) increased the risk for thrombosis nearly 2-fold compared with TAFI levels below the 90th percentile (odds ratio, 1.7; 95% confidence interval, 1.1-2.5). Adjustment for various possible confounders did not materially affect this estimate. These results indicate that elevated TAFI levels form a mild risk factor for venous thrombosis. Such levels were found in 9% of healthy controls and in 14% of patients with a first deep vein thrombosis. Elevated TAFI levels did not enhance the thrombotic risk associated with factor V Leiden but may interact with high factor VIII levels. (Blood. 2000;95:2855-2859)

Toward gene therapy for hemophilia A: long-term persistence of factor VIII-secreting fibroblasts after transplantation into immunodeficient mice.

Hemophilia A is caused by the lack of functional blood-clotting factor VIII. We have used retrovirus-mediated gene transfer to generate various cell lines, rodent as well as human, that secrete the human factor VIII protein. To study whether transplantation of genetically modified fibroblasts is a feasible approach for gene therapy of hemophilia A, we implanted the factor VIII-secreting cells into immune-deficient mice. Implantation of factor VIII-secreting primary human skin fibroblasts resulted in long-term persistence of the transplanted cells; cells recovered from the implants up to 2 months post-implantation still had the capacity to secrete factor VIII when regrown in tissue culture. However, we were unable to detect any human factor VIII in plasma samples of the recipient mice. The absence of human factor VIII in the recipients' plasma is shown to be due neither to (epigenetic) inactivation of the retroviral vector in vivo, nor to inability of the stationary cells to secrete factor VIII protein. However, we did note a rapid clearing of the human factor VIII: CAg from plasma upon intravenous injection of plasma-derived human factor VIII in mice (t1/2 < 60 min vs. 10 hr in humans). This phenomenon can fully explain the apparent absence of human factor VIII in the recipients' plasma.

Activated protein C accelerates clot lysis by virtue of its anticoagulant activity.

The effect of human activated protein C (APC) on t-PA dependent fibrinolysis was studied in vitro using plasma (and whole blood) clot lysis techniques. Clot lysis was monitored by measuring the release of soluble 125I-labelled fibrin degradation products from the clot over time. It was demonstrated that the stimulatory effect of APC on plasma and blood clot lysis was specific for APC and depended on the presence of its active site and Ca2+ ions. Furthermore, the effect depended on the presence of phospholipids in plasma or cells in blood. The presence of pro-urokinase, factor XIII or alpha 2-antiplasmin was not required for the expression of the profibrinolytic effect of APC. Subsequent experiments revealed that the profibrinolytic effect of APC was only observed when thrombin was formed through the coagulation pathway during the initial phase of the clot lysis experiment. It was also shown that the addition of increasing concentrations of thrombin itself could delay the t-PA dependent lysis of clots prepared from Al(OH)3 adsorbed plasma via a mechanism not yet understood. Based on these findings we propose that (a) t-PA dependent lysis of clots prepared from pooled normal plasma is delayed by thrombin generated through the coagulation system, and (b) that by its anticoagulant properties APC blocks this thrombin generation and thereby prevents the delay in clot lysis. Because in this model the profibrinolytic effect of APC is directly related to its anticoagulant properties we predicted and confirmed that other anticoagulants--like heparin--also have profibrinolytic activity. Conversely, procoagulants such as phospholipids can be antifibrinolytic.

Expression of functional factor VIII in primary human skin fibroblasts after retrovirus-mediated gene transfer.

We have developed a retroviral-vector system for the transfer and expression of a cloned blood clotting factor VIII cDNA. Since inclusion of the complete cDNA into existing vectors is precluded by its large size, we deleted most codons for the B-domain, which is also excised during in vivo maturation of factor VIII. When inserted into the retroviral vector M5-neoR (Laker, C., Stocking, C., Bergholtz, V., Hess, N., DeLamarter, J. F., and Ostertag, W. (1987) Proc. Natl. Acad. Sci. U. S. A. 84, 8458-8462), the sequence was shown to be efficiently expressed in murine fibroblast cell lines, as well as in primary human skin fibroblasts. Upon infection of murine fibroblast cell lines, clones containing only a single copy of the integrated vector-provirus secreted up to 125 milliunits of factor VIII antigen/10(6) cells/day. Equivalent amounts were found in a factor VIII activity assay, which signifies that the factor VIII protein secreted by the infected fibroblasts is fully functional. Primary human skin fibroblasts infected with the vector virus secreted up to 30 milliunits/10(6) cells/day.

A second plasma inhibitor of activated protein C: alpha 1-antitrypsin.

Inactivation of activated protein C (APC) in normal human plasma was studied in the absence and presence of heparin. In the absence of heparin APC inactivation followed pseudo-first order kinetics. In the presence of heparin the neutralization of APC was found to be biphasic. Up to 500 nM APC could be readily inactivated in normal plasma, indicating that the concentration of the APC inhibitor must be higher than previously assumed. Plasma deficient in the protein C inhibitor (PCI-I, as described by Suzuki and coworkers) and deficient in beta 2-glycoprotein I still possessed APC neutralizing capacity, presumably through the formation of complexes of APC with another plasma protein as was demonstrated by immunoblotting with anti-protein C antibodies. Together these data made us to conclude that a second inhibitor of APC (PCI-II) must be present in normal human plasma. This second inhibitor should be heparin independent, have a relatively high plasma concentration and form complexes with APC. Subsequently, we purified this PCI-II by isolating APC-PCI-II complexes from plasma deficient of vitamin K dependent proteins, PCI-I and beta 2-glycoprotein-I, to which purified human APC had been added. Purified PCI-II has a molecular weight of 50,000 daltons and aminoacid analysis revealed that PCI-II is identical with alpha 1-antitrypsin (alpha 1-AT). The second order rate constant for the reaction between purified alpha 1-AT and APC was found to be 269 M-1 min-1 in the absence of calcium and 602 M-1 min-1 in the presence of calcium.(ABSTRACT TRUNCATED AT 250 WORDS)

Immunoradiometric assays for human coagulation factor VII using polyclonal antibodies against the Ca(II)-dependent and Ca(II)-independent conformation.

Human coagulation factor VII is a trace plasma protein belonging to the vitamin K-dependent factors. Two specific and sensitive immunoradiometric assays for factor VII were developed using immunopurified rabbit antibodies against the Ca(II)-independent and Ca(II)-dependent conformation of factor VII. Both assays were insensitive to the activation state of factor VII. The distribution of factor VII antigen was studied in 40 healthy individuals and the antigen level in normal plasma was calculated to be 0.52-0.62 micrograms/ml. The two assays were used in a comparative study of factor VII procoagulant activity and factor VII antigen in patients treated with oral anticoagulants.

Evaluation of the cuticle bleeding time in canine haemophilia A.

In this paper we describe our clinical experience and results with the cuticle bleeding time test in a colony of cross-bred Labrador retrievers with severe haemophilia A. The dogs have a severe bleeding tendency with a high incidence of fatal haemorrhages in the central nervous system. Homozygous females appeared to be especially prone to this lethal complication. Factor VIII recovery and half-life determinations yielded results similar to the data from human studies. The cuticle bleeding time proved to be a good measure of the coagulation defect. The prolongation of the bleeding time could be completely abolished by administration of 10 to 15 units of canine factor VIII per kg body weight. We conclude that the cuticle bleeding time in canine haemophilia provides us with a suitable model for the in vivo study of new therapeutic materials.

Activated protein C decreases plasminogen activator-inhibitor activity in endothelial cell-conditioned medium.

Confluent cultures of endothelial cells from human umbilical cord were used to study the effect of activated human protein C (APC) on the production of plasminogen activators, plasminogen activator-inhibitor, and factor VIII-related antigen. Addition of APC to the cells in a serum-free medium did not affect the production of tissue-type plasminogen activator (t-PA) or factor VIII-related antigen; under all measured conditions, no urokinase activity was found. However, less plasminogen activator-inhibitor activity accumulated in the conditioned medium in the presence of APC. This decrease was dose dependent and could be prevented by specific anti-protein C antibodies. No decrease was observed with the zymogen protein C or with diisopropylfluorophosphate-inactivated APC. APC also decreased the t-PA inhibitor activity in endothelial cell-conditioned medium in the absence of cells, which suggests that the effect of APC is at least partly due to a direct effect of APC on the plasminogen activator-inhibitor. High concentrations of thrombin-but not of factor Xa or IXa--had a similar effect on the t-PA inhibitor activity. The effect of APC on the plasminogen activator-inhibitor provides a new mechanism by which APC may enhance fibrinolysis. The data suggest that activation of the coagulation system may lead to a secondary increase of the fibrinolytic activity by changing the balance between plasminogen activator(s) and its (their) fast-acting inhibitor.

Coagulation factors in the premature infant born after about 32 weeks of gestation.

In plasma samples from 10 premature infants born after about 32 weeks of gestation, a number of coagulation factors have been determined. For 9 infants, who were healthy, mean values are given: fibrinogen-antigen, 311 mg/dl; factor II, +/- 0.46 U/ml; factor V, 0.80 U/ml; factor VII, 0.59 U/ml; factor VIII coagulant activity, 0.93 U/ml; factor VIII-related antigen, 1.66 U/ml; procoagulant factor VIII antigen, 1.15 U/ml; factor IX coagulant activity, 0.41 U/ml; factor IX antigen, 0.42 U/ml; factor X coagulant activity, 0.52 U/ml; factor X antigen, 0.61 U/ml, and antithrombin III-antigen (AT-III), 0.43 U/ml. In 5 infants a second sample was taken a week after the first one; for most values there was no significant rise, except for factor II and AT-III. The values we found in this group of premature infants are within the range of those reported in earlier literature. They are higher than the ones we found in early fetal samples and most of them are similar to those we found in early fetal samples and most of them are similar to those we found in the cord blood of full-term newborns.

Heterogeneity of haemophilia A: a study with three different antisera.

One human and one rabbit antibody against VIII:C--in a fluid-phase, inhibitor neutralization assay (INA)--and one human antibody--in a solid-phase, immunoradiometric assay (IRMA)--were used to investigate a group of 59 patients with severe, moderate and mild haemophilia A. Patients were classified as haemophilia A+ when the VIII:C/VIIIAG ratio was less than 0.4 while the absolute VIII:C antigen (VIIICAG) value exceeded 0.25 u/ml. Three haemophiliacs (from two pedigrees) were classified as A+ in all three assay systems. 50% of the patients were classified as haemophilia A+ on the basis of at least one assay. The other 50%, including 66% of the patients with severe haemophilia, were shown to be A- by all three assay systems. Combining the data obtained with the three different antibodies four different categories could be distinguished, in addition to the classification based on severity.

A monoclonal antibody to VIII:C produced by a mouse hybridoma.

Spleen cells of a BALB/c mouse immunized with factor VIII procoagulant activity (VIII:C) (isolated by affinity chromatography) were fused with mouse myeloma cells (P3 x 63 Ag8). After the fusion 12/32 wells produced an inhibitor to VIII:C. Cells from one well (1B3) were subcloned four times in order to isolate the hybridoma that produces the anti-VIII:C antibody. Injection of hybridoma cells in pristane pretreated BALB/c mice results in anti-VIII:C titers of 5000-10,000 Bethesda U/ml. Analysis of the produced immunoglobulin demonstrated heavy chains of IgG1 (produced by the myeloma cell line) and IgG2b subclass. The 1B3 antibody neutralizes VIII:C in LMW FVIII, crysosupernatant, cryoprecipitate, and normal plasma. It was found that binding of the IgG to FVIII results in a delay in its activation and not in an inhibition of its cofactor activity. The antibody removes VIII:C from pooled normal plasma when coupled to Sepharose; when coupled to plastic tubes, it binds VIIICAG from isolated VIII:C, purified FVIII, and pooled normal plasma; it does not bind VIIIR:AG, fibrogen, or serum VIIICAG. The 1B3 antibody can be used successfully in an IRMA for VIIICAG.