Effect of Lipoprotein(a) and LDL on Plasminogen Binding to Extracellular Matrix and on Matrix-dependent Plasminogen Activation by Tissue Plasminogen Activator

Lp(a) is an LDL-like lipoprotein plus an additional apolipoprotein apo(a). Based on the structural homology of apo(a) with plasminogen, it is hypothesized that Lp(a) interferes with fibrinolysis. Extracellular matrix (ECM) produced by human umbilical vein endothelial cells was used to study the effect of Lp(a) and LDL on plasminogen binding and activation. Both lipoproteins were isolated from the same plasma in a single step. Plasminogen bound to ECM via its lysine binding sites. Lp(a) as well as LDL were capable of competing with plasminogen binding. The degree of inhibition was dependent on the lipoprotein donor as well as the ECM donor. When Lp(a) and LDL obtained from one donor were compared, Lp(a) was always a much more potent competitor. The effect of both lipoproteins on plasminogen binding was reflected in their effect on plasminogen activation. It is speculated that Lp(a) interacts with ECM via its LDL-like lipoprotein moiety as well as via its apo(a) moiety.

Role of the A+ Helix in Heparin Binding to Protein C Inhibitor

Interactions between proteins and heparin(-like) structures involve electrostatic forces and structural features. Based on charge distributions in the linear sequence of protein C inhibitor (PCI), two positively charged regions of PCI were proposed as possible candidates for this interaction. The first region, the A+ helix, is located at the N-terminus (residues 1-11), whereas the second region, the H helix, is positioned between residues 264 and 280 of PCI. Competition experiments with synthetic peptides based on the sequence of these regions demonstrated that the H helix has the highest affinity for heparin. In contrast to previous observations we found that the A+ helix peptide competed for the interaction of PCI with heparin, but its affinity was much lower than that of the H helix peptide.

Recombinant PCI was also used to investigate the role of the A+ helix in heparin binding. Full-length (wild-type) rPCI as well as an A+ helix deletion mutant of PCI (rPCI-Δ2-l 1) were expressed in baby hamster kidney cells and both had normal inhibition activity with activated protein C and thrombin. The interaction of the recombinant PCIs with heparin was investigated and compared to plasma PCI. The A+ helix deletion mutant showed a decreased affinity for heparin in inhibition reactions with activated protein C and thrombin, but had similar association constants compared to wild-type rPCI. The synthetic A+ helix peptide competed with rPCI-Δ2-11 for binding to heparin. This indicated that the interaction between PCI and heparin is fairly non-specific and that the interaction is primarily based on electrostatic interactions.

In summary, our data suggest that the H helix of PCI is the main heparin binding region of PCI, but the A+ helix increases the overall affinity for the PCI-heparin interaction by contributing a second positively charged region to the surface of PCI.

In Vitro Studies of Antiphospholipid Antibodies and Its Cofactor, β2-Glycoprotein I, Show Negligible Effects on Endothelial Cell Mediated Protein C Activation

The effect of sera and purified IgG isolated from plasma of 46 patients with systemic lupus erythematosus (SLE) and 9 healthy donors on the endothelial cell (EC) mediated protein C activation was investigated. Out of the 46 SLE sera used, 19 were antiphospholipid antibodies (aPL) positive. From 12 patients IgG was isolated, of which 6 contained aPL. EC were first incubated with IgG (7 mg/ml) or serum (1: 1 diluted) for 1 h and then tested for their ability to promote protein C activation by thrombin, with the cells either in a monolayer or in a suspension. The normal range (mean of control values ± 2 SD) of protein C activation was 80-120%. In contrast to others, we could not detect an inhibition of protein C activation by any of the patient IgG's or sera.

The recently described cofactor for binding of antiphospholipid antibodies to phospholipids, β2-glycoprotein I, was purified and added to the purified IgG's. A combination of these two components did not inhibit the EC mediated protein C activation by thrombin.

This study suggests that the inhibition of the protein C activation, mediated by EC, is not a general mechanism by which aPL related thrombosis can be explained.

Construction and Characterization of Thrombin-resistant Variants of Recombinant Human Protein S

Protein S is a vitamin K-dependent plasma protein that functions as a cofactor of activated protein C (APC) in the inactivation of coagulation factors Va and Villa.

Protein S, migrates as a doublet on reduced SDS polyacrylamide gel electrophoresis. This heterogeneity in molecular weight has been explained by limited proteolysis of protein S. Human protein S contains at Arg-49, Arg-60 and Arg-70 three potential cleavage sites. Whether cleavage occurs at all three sites is not known. To study the role of these arginine residues in human protein S, we have replaced them by leucine or isoleucine. All seven possible variants were constructed: three variants with single mutations (R49L, R60L, R70I), three variants with double mutations (R49L/R60L, R60L/R70I, R49L/R70I) and one variant with a triple mutation (R49L/R60L/R70I). On reduced SDS polyacrylamide gels the single and double variants migrate as a doublet just like the wild type protein S. The triple variant migrates as a single band at a molecular weight corresponding to the upper band of the doublet. The upper band of the single and double variants but not of the triple variant could be converted into the lower band by thrombin treatment.

All variants showed cofactor activity to APC in a clotting assay. After thrombin treatment, this cofactor activity was abolished for the single (R49L, R60L, R70I) and double variants (R49L/R60L, R60L/R70I, R49L/R70I), while the triple variant (R49L/R60L/R70I) tested at several concentrations, retained its cofactor activity completely, suggesting resistance to thrombin. This shows that thrombin can cleave at all three arginine sites and that cleavage at each of these sites results in the loss of APC cofactor activity. Finally, all variants bind to C4b-binding protein with an affinity similar as the wild type recombinant molecule.

The Identification and Significance of a Thr→Pro Polymorphism in Kringle IV Type 8 of Apolipoprotein(a)

Elevated plasma levels of lipoprotein(a) [Lp(a)] represent a significant independent risk factor for the development of atherosclerosis. Interindividual levels of apo(a) vary over 1000-fold and are mainly due to inheritance that is linked to the locus of the apolipoprotein(a) [apo(a)] gene. The apo(a) gene encodes multiple repeats of a sequence exhibiting up to 85% DNA sequence homology with plasminogen kringle IV (K.IV), a lysine binding domain. In our search for sequence polymorphisms in the K.IV coding domain, we identified a polymorphism predicting a Thr→Pro substitution located at amino acid position 12 of kringle IV type 8 of apo(a). The functional and clinical significance of this polymorphism was analysed in a case-control study and by comparing the in vitro lysine binding characteristics of the two Lp(a) subtypes.

The case-control study (involving 153 subjects having symptomatic atherosclerosis and 153 age and gender matched normolipidemic controls) revealed an overall allele frequency for the Thr12-→Pro substitution in kringle IV type 8 of 14% and a negative association between presence of the Pro12-subtype and symptomatic atherosclerosis (p <0.03). The in vitro lysine binding studies, using Lp(a) isolated from subjects homozygous for either Thr12 or Pro12 in K.IV type 8, revealed comparable lysine-Sepharose binding fractions for the two subtypes. The binding affinity (Kd) for immobilised plasmin degraded des- AA-fibrin (DesafibTM-X) was also comparable for the two subtypes, however a decreased maximal attainable binding (Bmax) for immobilised desafibTM-X was observed for the Pro12-subtype Lp(a).

Surface Independent Factor XI Activation by Thrombin in the Presence of High Molecular Weight Kininogen

A deficiency of one of the proteins of the contact system of blood coagulation does not result in a bleeding disorder. For this reason activation of blood coagulation via this system is believed to be an in vitro artefact. However, patients deficient in factor XI do suffer from variable bleeding abnormalities. Recently, an alternative pathway for factor XI activation has been described. Factor XI was found to be activated by thrombin in the presence of dextran sulfate as a surface. However, high molecular weight kininogen (HK), to which factor XI is bound in plasma, and fibrinogen were shown to block this activation suggesting it to be an in vitro phenomenon. We investigated the thrombin-mediated factor XI activation using an amplified detection system consisting of factors IX, VIII and X, which was shown to be very sensitive for factor XIa activity. This assay is approximately 4 to 5 orders of magnitude more sensitive than the normal factor XIa activity assay using a chromogenic substrate. With this assay we found that factor XI activation by thrombin could take place in the absence of dextran sulfate. The initial activation rate was approximately 0.3 pM/min (using 25 nM factor XI and 10 nM thrombin). The presence of dextran sulfate enhanced this rate about 8500-fold. A very rapid and complete factor X activation was observed in the presence of dextran sulfate. Although only minute amounts of factor XIa were formed in the absence of dextran sulfate, significant activation of factor X was detected in the amplification assay within a few minutes. HK inhibited the activation of factor XI by thrombin strongly in the presence, yet only slightly in the absence of dextran sulfate (26 and 1.2 times, respectively). Despite the strong inhibition of HK on the activation of factor XI by thrombin in the presence of dextran sulfate, HK had only a minor effect on the factor Xa generation.

We conclude that activation of factor XI by thrombin can take place regardless of the presence of a surface or HK. This activation might therefore be physiologically relevant. The inhibitory effect of HK on the thrombin-mediated factor XI activation is largely dextran sulfate dependent. Due to the amplification in the intrinsic system, trace amounts of factor XIa might generate physiological sufficient amounts of factor Xa for an adequate haemostatic response.

Identification of Different Forms of Human C4b-Binding Protein Lacking β-Chain and Protein S Binding Ability

Human C4b-binding protein (C4BP) is a multimeric regulatory component of the complement system that circulates in plasma either as a free protein or in a noncovalent complex with the vitamin K-dependent protein S. The major form of C4BP is composed of seven identical α-chains (70 kDa) and one β-chain (45 kDa). C4BP was purified from human plasma after barium citrate adsorption using anti-C4BP monoclonal antibody affinity chromatography. C4BP-high and low Mr forms were both obtained from the barium citrate precipitate and the supernatant. C4BP-high and low forms from the barium citrate precipitate were separated by sodium dodecylsulfate polyacrylamide slab gel electrophoresis and extracted with Triton X-100. Both forms contained the β-chain as was demonstrated on sodium dodecylsulfate polyacrylamide slab gel electrophoresis under reduced conditions after silverstaining and with Western-blotting using monoclonal antibodies specific for the β-chain. The C4BP-high and low forms demonstrated similar protein S binding affinity (KA: 3.18 × 108 and 3.21 × 108 M-1, respectively) in a C4BP-protein S binding assay and a protein S ligand blot using a peroxidase-conjugated monoclonal anti-protein S antibody. The barium citrate supernatant contained two forms of C4BP-high and one form of C4BP-low. One form of C4BP-high did contain the β-chain and was capable of protein S binding (KA: 4.35 × 108 M-1). The two other forms of C4BP lacked the β-chain and were unable to bind protein S.

Purified Human Plasma Kallikrein Does Not Stimulate but Primes Neutrophils for Superoxide Production

In patients with septicemia and septic shock the contact phase of blood coagulation is activated. It has been suggested that polymorphonuclear leukocytes (PMN) are directly activated by purified plasma kallikrein. This has been recently questioned because granulocytic elastase release induced by recalcification of normal and prekallikrein-deficient plasma was similar. We studied the interaction of different preparations of purified human plasma kallikrein with PMN. Cytosolic calcium shifts were measured with the quin2 method, PMN aggregation was assayed in an aggregometer, and superoxide production was quantitated as superoxide dismutase inhibitable cytochrome c reduction in a continuous assay. No increase of cytosolic free calcium was found during at least 5 min after adding 10 μg/ml plasma kallikrein to PMN. Similarly, highly purified plasma kallikrein from two different sources did not induce PMN aggregation at all, nor did it stimulate superoxide production. However, sequential exposure of PMN to plasma kallikrein and formylpeptide increased the superoxide production compared to stimulation with formylpeptide alone. This phenomenon which is called priming was observed at plasma kallikrein concentrations ≥7 μg/ml. The active site of the molecule was required for the priming, because plasma prekallikrein, active site-inactivated plasma kallikrein, and soybean trypsin inhibitor treated kallikrein did not prime PMN. This indicates that the contact activation system may play a role in host defence against bacterial infection.

Elimination of Intravenously Administered Radiolabelled Antithrombin III and Heparin in Humans

Antithrombin III was purified from normal plasma by DEAE- Sephadex chromatography and heparin affinity chromatography; the protein was subsequently radiolabelled with 125I. 125I-anti- thrombin III alone and 125I-antithrombin III in the presence of high affinity 35S-heparin fractions were injected into normal humans. 125I-radiolabel and protein bound 35S-radioactivity were followed separately. In semilogarithmic plots 125I-antithrombin III disappeared according to a double exponential curve with a half-life in the second phase of 56.8 hr in the absence of heparin and of 33.7 hr in the presence of heparin. Protein bound 35S- radioactivity disappeared much faster than the 125I-radiolabel. These data support the concept that heparin disappears as free heparin from the equilibrium heparin – antithrombin III ⇄ heparin + antithrombin III. Immuno-reactive antithrombin III decreased from 100% to 85-90% immediately after injection of 125I-antithrombin III in the presence of heparin and returned to normal values within 30 min. This suggests that antithrombin III is transiently sequestered, possibly in trimolecular complexes consisting of antithrombin III, heparin and either lipases or other vascular bound proteins.

Effect of Heparin on the Activation of Factor XI by Fibrin-bound Thrombin

Fibrin-bound thrombin is protected from inactivation by antithrombin III, while its coagulant potential is retained. In the presence of heparin, ternary complexes between thrombin, fibrin and heparin are formed. In these complexes the coagulant activity of thrombin is retained, whereas the anticoagulant activity of fibrin-bound heparin is neutralized. The limited effectiveness of heparin in the prevention of both venous thrombosis and coronary reocclusion is probably related to the protective effect of fibrin on the inactivation of thrombin by antithrombin III. Recently, it has been shown that factor XI can be activated by thrombin, resulting in the generation of additional thrombin via the intrinsic pathway. This additional thrombin is capable of stabilizing the clot by protecting it from fibrinolysis. We studied the effect of heparin on the activation of factor XI by fibrin-bound thrombin. First, we used fibrin monomers coupled to Sepharose to which thrombin and unfractionated heparin (UFH) were bound. Factor XI activation by thrombin was the same in the presence of fibrin-Sepharose or control-Sepha-rose. The addition of heparin (0.1 U/ml) resulted in a 91 and 15-fold enhancement in the presence of control-Sepharose and fibrin-Sepharose, respectively. Next, we added complexes of heparin, thrombin and fibrin monomer to factor XII and XI double-deficient plasma in the presence or absence of a reconstituting amount of factor XI. In the presence of factor XI, additional fibrin formation was observed indicating that factor XI activation by thrombin in complex with fibrin and heparin can take place in plasma. We then studied the effect of other heparin-like anticoagulants on the thrombin-mediated factor XI activation. UFH enhanced thrombin-mediated factor XI activation 68-fold, LMWH (low molecular weight heparin, Fragmin) 12-fold, danaparoid (Orgaran) 3-fold, while the pentasaccharide ORG 31540 did not result in an enhancement. Binding studies of these anticoagulants to fibrin-Sepharose showed that LMWH bound with approximately the same affinity as UFH, while danaparoid and the pentasaccharide did not bind to fibrin.

We conclude that fibrin-bound thrombin is capable of factor XI activation. Furthermore, heparin bound in a complex with fibrin can act as a cofactor for this activation. This factor XI activation capacity may play a role in the limited effectiveness of heparin. Provided that thrombin-mediated factor XI activation plays an important role in vivo, danaparoid and especially the pentasaccharide may be better anticoagulants than UFH and LMWH.

The Interaction of Protein S with the Phospholipid Surface Is Essential for the Activated Protein C-independent Activity of Protein S

Protein S is a vitamin-K dependent glycoprotein involved in the regulation of the anticoagulant activity of activated protein C (APC). Recent data showed a direct anticoagulant role of protein S independent of APC, as demonstrated by the inhibition of prothrombinase and tenase activity both in plasma and in purified systems. This anticoagulant effect of protein S can be explained either by a direct interaction of protein S with one of the components of the complexes and/or by the interference with the binding of these components to phospholipid surfaces.

During our investigation we noted that protein S preparations purified in different ways and derived from different sources, expressed discrepant APC cofactor and direct anticoagulant activity. In order to elucidate these differences and to study the mechanism of the APC-inde-pendent activity of protein S, we compared the protein S preparations in phospholipid-binding properties and anticoagulant activity. The dissociation constant for the binding of protein S to immobilized phospholipids ranged from 7 to 74 nM for the different protein S preparations. APC-independent inhibition of both prothrombinase and tenase activity performed on phospholipid vesicles and in plasma showed a strong correlation with the affinity for phospholipids. The APC-independent activity could be abolished by monoclonal antibodies that were either calcium-dependent and/or directed against epitopes in the Gla-region of protein S, suggesting that the protein S-phospholipid interaction is crucial for the APC-independent anticoagulant function of protein S. Protein S preparations with a low APC-independent activity expressed a high APC-cofactor activity suggesting that the affinity of protein S for phospholipids is of less importance in the expression of APC-cofactor activity of protein S.

We conclude that high affinity interactions of protein S with the membrane surface are essential for the direct anticoagulant activity of protein S and we suggest that inhibition of the prothrombinase and the tenase complex by protein S is a consequence of the occupation of the phospholipid surface by protein S molecules.

Changes in Haemostatic Factors and Activation Products after Exercise in Healthy Subjects with Different Ages

We studied exercise-induced changes in coagulation and fibrinolytic factors and activation products in different age categories. Thirty-eight sedentary males, divided in three age categories (cats I-III; 20-30, 35-45 and 50-60 y) were subjected to a standardized exercise test.

Pre-exercise levels (cats I-III resp) of FVII:c (105 ± 5, 121 ±6 and 123 ± 7% NP), fibrinogen (2.35 ± 0.12, 2.55 ± 0.10 and 2.66 ± 0.09 mg/ml), prothrombin activation fragment F1+2 (0.80 ± 0.10,0.80 ± 0.11 and 1.22 ±0.16 nM), t-PA (5,2 ± 0.6, 9.2 ± 1.0, 8.6 ± 1.2 ng/ml) and PAI-1 (42.8 ± 7.5, 67.6 ± 7.6, 62.2 ± 10.9 ng/ml) showed differences that seemed related to age. Regression analysis revealed associations with anthropometry (FVII:c, fibrinogen, F1+2, t-PA, PAI-1) rather than with age.

Exercise-induced changes in coagulation (increase in von Wille-brand factor and FVIII:c and a shortening of APTT) and fibrinolytic potential (increase in t-PA and u-PA) were of comparable magnitude for the three age categories. Hardly any change in F1+2 (6%) was observed, while thrombin-antithrombin complexes (93%), plasmin-antiplasmin complexes (79%) and D-dimer (77%) almost doubled during maximal exercise.

We conclude that anthropometric differences play a more significant role than age on constitutive levels of haemostatic factors in participants up to 60 years of age. The magnitude of exercise-induced changes is comparable in the age categories under study, and simply superimposed on constitutive (pre-exercise) levels. Clear evidence for prothrombin activation is lacking, but plasmin formation is enhanced during exercise.

Fibrinolytic Activity in Blood Is Distributed over a Cellular and the Plasma Fraction which Can Be Modulated Separately

Blood fibrinolytic activity is mediated by plasma and cellular components. We have studied blood fibrinolytic activity in different species and investigated the distribution pattern in rats after modulation with PAF, dexamethasone, or retinoic acid. Whole blood and plasma activity were measured in an assay system using human or endogenous fibrin as substrate. When human fibrin was used as substrate marked species differences in distribution of fibrinolytic activity were observed. In rat and murine blood most fibrinolytic activity was associated with the plasma fraction (70% and 50% respectively) while in human and canine blood the plasma fraction contained only 30% of the blood fibrinolytic activity. When endogenous fibrin was used as substrate the distribution pattern of fibrinolytic activity in rat blood changed dramatically. Less than 25% of the blood fibrinolytic activity was now present in the plasma fraction.

The fibrinolytic system was further investigated in rats using specific inhibitors of proteolytic activity. Blood fibrinolytic activity could be inhibited for 33% by antibodies raised against t-PA and 60% inhibition was obtained in the presence of amiloride. No significant effect of elas-tinal (an inhibitor of elastase) could be detected. Plasma fibrinolytic activity was not affected by these inhibitors. The fibrinolytic activity in plasma could be enhanced about 100-fold after i. v. PAF administration (10 μg/kg). This extra fibrinolytic activity could be fully blocked by antibodies raised against t-PA. Oral administration of dexamethasone or retinoic acid affected blood fibrinolytic activity by modulating selectively the activity mediated by the cellular fraction. Dexamethasone treatment (1 mg/kg) resulted in a 59% decrease of this fibrinolytic activity. Retinoic acid treatment (3 mg/kg) resulted in a 69% increase of the fibrinolytic activity in this phase. Both compounds affected selectively the t-PA activity.

We conclude that 1) the plasma as well as the cellular fraction contribute to the blood fibrinolytic activity; 2) the plasma derived fibrinolytic activity as such is a minor part of the total whole blood fibrinolytic activity (the mouse is an exception); 3) the rat blood fibrinolytic activity is evoked by u-PA (two third) and t-PA (one third) mediated processes; 4) the blood fibrinolytic activity can be modulated, in contrast to the plasma activity as such, by retinoic acid or dexamethasone; 5) the plasma fibrinolytic activity is subject to modulation by compounds which evoke a direct PA increase.

Studies of the Interaction between Human Protein S and Human C4b-Binding Protein Using Deletion Variants of Recombinant Human Protein S

Human protein S interacts noncovalently with human C4b-binding protein (C4BP). We have studied this interaction using deletion variants of recombinant human protein S. Two deletion variants were constructed by restriction enzyme digestion and in vitro site-specific mutagenesis of the human protein S cDNA. The variants were stably expressed in Cl27 cells. Recombinant proteins were purified using Fast Flow Q anion-exchange chromatography. The activated protein C (APC) cofactor activity, C4BP binding properties and reactivity to different monoclonal antibodies against human protein S were examined. The first variant (E variant), which has a deletion of the third epidermal growth factor (EGF)-like domain (deletion of exon VII, corresponding to amino acid residues ASP-160 to Asp-202) expresses normal APC cofactor • activity in a plasma system. This activity was inhibited by the addition of purified C4BP. The second variant (L variant), which has a deletion of the C-terminal loop of the sex hormone binding globulin (SHBG)- like domain (deletion of exon XV, corresponding to amino acid residues Asp-583 to Ser-635) also expresses normal APC cofactor activity in plasma. This activity could only be partially inhibited by the addition of purified C4BP.

Binding of the recombinant proteins to C4BP was studied in a system using purified proteins. The E variant binds to C4BP with the same affinity similar as recombinant wild type protein S (apparent Kd ∼ 10−10 M). The L variant, however, shows a markedly reduced affinity for binding to C4BP (apparent Kd ∼ 10−7 M).

Three different Ca2+-independent monoclonal antibodies (S5, S12, S17) against protein S, which do not interfere with the APC cofactor activity and C4BP binding of protein S, were used to screen the deletion variants for possible conformational changes. Two of these showed the same affinity for the E and L variant as for wild type recombinant protein S. The third, S12, which recognizes an epitope in the vicinity of ser-460, reacts normally with the E variant but has a strongly reduced affinity for the L variant, although the presence of the epitope could be clearly demonstrated by immunoblotting under denaturing conditions.

This suggests that the deletion of the C-terminal loop induces a conformational change in protein S which affects the epitope for S12. Therefore although our results indicate that the C-terminal loop of the SHBG-like domain of human protein S is involved in the interaction with C4BP, we cannot exclude the possibility that the deletion of the C-terminal loop induces a conformational change that results in a loss of binding affinity for C4BP elsewhere in the protein S molecule.

Retinoic Acid Enhances Fibrinolytic Activity In-Vivo by Enhancing Tissue Type Plasminogen Activator (t-PA) Activity and Inhibits Venous Thrombosis

We studied the profibrinolytic effect and the anti-thrombotic potential of retinoic acid in-vivo. Male Wistar rats were treated with retinoic acid either acutely or twice daily for a period of 5 consecutive days in a dose range of 0.01 to 3.0 mg/kg. Fibrinolytic activity was measured ex-vivo using the diluted blood clot lysis test and net t-PA activity in tissue extracts was measured in a spectrophotometric rate assay. Clot lysis was dose dependently increased by retinoic acid up to about 170% (relative to vehicle treated rats) at a dose of 3 mg/kg. No tachyphylaxis could be detected. Ex-vivo measured fibrinolytic activity after single administration of 1 mg/kg of retinoic acid peaked at 3 h after ingestion. Even after 18 h a significantly higher clot lysis rate was measured. Lysis of blood clots from vehicle and retinoic acid treated rats could be completely blocked by addition of tranexamic acid or antibodies against rat t-PA before clot formation. t-PA activity in plasma was slightly increased after retinoic acid treatment; no effects were measured on plasma PAI-1, u-PA, plasminogen, and α2-antiplasmin levels. t-PA activity in lung and kidney was marginally enhanced by retinoic acid but in heart and aortic tissue extracts t-PA activity was increased by about 50%. We confirmed this potential anti-thrombotic property in an in-vivo venous thrombosis model. A reduced clot size was observed after retinoic acid administration (35% reduction at a dose of 1 mg/kg). We could not detect any effect on ex-vivo measured platelet aggregation, bleeding time, prothrombin time, thrombin time, partial thromboplastin time or plasma fibrinogen level, indicating that there were no effects on platelet aggregation or blood coagulation. We conclude that retinoic acid enhanced fibrinolytic activity in-vivo by selectively enhancing t-PA activity and that retinoic acid is a potential anti-thrombotic agent in-vivo.

Lysine-Binding Heterogeneity of Lp(a): Consequences for Fibrin Binding and Inhibition of Plasminogen Activation

Lipoprotein(a) [Lp(a)] is recognized as an independent risk factor for atherosclerosis. Lp(a) consists of a LDL-like moiety with an additional glycoprotein, apo(a), linked to apolipoprotein B-100. Apo(a) has a high homology with plasminogen (Pg). In vivo, Pg is activated on a fibrin surface by tissue Pg activator (tPA). We prepared Lp(a) from plasma by sequential ultracentrifugation followed by lysine-sepharose affinity chromatography. We found that a changing (donor dependent) fraction of the Lp(a) did not bind to lysine-sepharose. This fraction, designated Lp(a)lys–, was further purified using gel filtration. Bound Lp(a) [Lp(a)lys+] was eluted with 0.2 M EACA. Apo(a) isoforms in both fractions were identical. In contrast Lp(a)lys+ inhibited Pg activation by tPA in vitro (IC50% 20 mg/1), whereas Lp(a)lys– did not. In addition Lp(a)lys– did not bind to CNBr-digested fibrinogen whereas Lp(a)lys+ did (K d, app = 0.2 nM). Therefore we conclude that a changing donor dependent fraction of human plasma Lp(a) does not inhibit Pg activation in vitro and does not bind to CNBr-digested fibrinogen.

Lupus Anticoagulant Activity Is Frequently Dependent on the Presence of β2-Glycoprotein I

Antiphospholipid antibodies (aPL) are defined by anticardioli-pin antibody (aCL) ELISA and prolongation of phospholipid dependent coagulation assays (lupus anticoagulant; LAC). For the binding of aCL to cardiolipin a cofactor, β2-glycoprotein I (β2-GPI), is necessary. We have investigated whether the same cofactor is essential for LAC activity. Plasma from 6 LAC positive patients and 3 controls was depleted from β2-GPI by means of affinity chromatography. From the 6 LAC positive plasmas, 4 became LAC negative (tested with dRWT) when β2-GPI was depleted and became positive again when purified β2-GPI (200 μg/ml) was added. A dose response curve showed that addition of 50 μg/ml β2-GPI to β2-GPI deficient patient plasma, led to a positive dRWT. Depletion of, and addition of β2-GPI to plasma from controls had no effect on the dRWT. Measurement of β2-GPI plasma levels in 19 LAC positive patients, 40 LAC negative patients and 15 controls showed no difference in β2-GPI levels.

These results show that a combination of aPL and β2-GPI is essential not only for binding to cardiolipin, but also for LAC activity and imply that low β2-GPI levels (<50 μg/ml) can lead to false negative LAC tests. These observations may lead to new insights in the pathophysiological complications associated with aPL.

Expression and Characterization of Recombinant Human Protein S in Heterologous Cells - Studies of the Interaction of Amino Acid Residues Leu-608 to Glu-612 with Human C4b-Binding Protein

Mouse C127 epithelioid cells were genetically engineered to produce biologically active ³-carboxylated human protein S. A full length human protein S cDNA was cloned into a bovine papilloma virus (BPV) based shuttle vector under the transcriptional control of the Moloney murine sarcoma virus enhancer and the mouse metallothionein promoter. Stable expression was obtained in transfected C127 cells. Expression of ³-carboxylated protein S was dependent on the presence of vitamin K in the culture medium. Protein sequence analysis showed that recombinant and plasma protein S have the same amino terminal sequence. Analysis of specific post-translationally modified amino acids shows that recombinant protein S is fully ³-carboxylated and fully p-hydroxylated. Immunoblotting analysis using polyclonal and monoclonal antibodies shows that recombinant protein S has a slightly higher molecular weight than plasma protein S. After N-Glycanase treatment, identical molecular weights are observed for recombinant and plasma protein S, indicating that the difference is caused by differences in the N-linked carbohydrate side chains. Recombinant protein S also demonstrates normal cofactor activity for activated protein C in a clotting assay. Binding studies with the complement component, C4b-binding protein (C4BP), shows that recombinant protein S binds to C4BP with the same apparent affinity as plasma protein S. Two variant molecules are also tested for their binding to C4BP. The first variant has a replacement of amino acid residue leu-608 by val and was designated B variant. The second variant has three alterations, at positions 609, 611 and 612 where the acidic amino acid residues asp, asp and glu were replaced by asn, asn and gin, respectively and this variant was designated C variant. The binding of these variants to C4BP was the same as wild type recombinant protein S. This suggests that amino acid residues leu-608, asp-609, asp-611 and glu-612 are not essential for binding of the intact full length protein to C4BP.

Plasma protein S contains zinc essential for efficient activated protein C-independent anticoagulant activity and binding to factor Xa, but not for efficient binding to tissue factor pathway inhibitor

Protein S (PS) is a cofactor for activated protein C (APC), which inactivates coagulation factors (F) Va and VIIIa. Deficiency of protein C or PS is associated with risk of thrombosis. We found that PS also has APC-independent anticoagulant activity (PS-direct) and directly inhibits thrombin generated by FXa/FVa (prothrombinase complex). Here we report that PS contains Zn(2+) that is required for PS-direct and that is lost during certain purification procedures. Immunoaffinity-purified PS contained 1.4 +/- 0.6 Zn(2+)/mol, whereas MonoQ-purified and commercial PS contained 0.15 +/- 0.15 Zn(2+)/mol. This may explain the controversy regarding the validity of PS-direct. Zn(2+) content correlated positively with PS-direct in prothrombinase assays and clotting assays, but APC-cofactor activity of PS was independent of Zn(2+) content. PS-direct and Zn(2+) were restored to inactive PS under mildly denaturing conditions. Conversely, o-phenanthroline reversibly impaired the PS-direct of active PS. Zn(2+)-containing PS bound FXa more efficiently (K(d)(app)=9.3 nM) than Zn(2+)-deficient PS (K(d)(app)=110 nM). PS bound TFPI efficiently, independently of Zn(2+) content (K(d)(app)=21 nM). Antibodies that block PS-direct preferentially recognized Zn(2+)-containing PS, suggesting conformation differences at or near the interface of 2 laminin G-like domains near the PS C terminus. Thus, Zn(2+) is required for PS-direct and efficient FXa binding and may play a role in stabilizing PS conformation.

Misfolded proteins activate factor XII in humans, leading to kallikrein formation without initiating coagulation

When blood is exposed to negatively charged surface materials such as glass, an enzymatic cascade known as the contact system becomes activated. This cascade is initiated by autoactivation of Factor XII and leads to both coagulation (via Factor XI) and an inflammatory response (via the kallikrein-kinin system). However, while Factor XII is important for coagulation in vitro, it is not important for physiological hemostasis, so the physiological role of the contact system remains elusive. Using patient blood samples and isolated proteins, we identified a novel class of Factor XII activators. Factor XII was activated by misfolded protein aggregates that formed by denaturation or by surface adsorption, which specifically led to the activation of the kallikrein-kinin system without inducing coagulation. Consistent with this, we found that Factor XII, but not Factor XI, was activated and kallikrein was formed in blood from patients with systemic amyloidosis, a disease marked by the accumulation and deposition of misfolded plasma proteins. These results show that the kallikrein-kinin system can be activated by Factor XII, in a process separate from the coagulation cascade, and point to a protective role for Factor XII following activation by misfolded protein aggregates.

Identification of fibronectin type I domains as amyloid-binding modules on tissue-type plasminogen activator and three homologs

The serine protease tissue-type plasminogen activator (tPA), a key enzyme in hemostasis, is activated by protein aggregates with amyloid-like properties. tPA is implicated in various pathologies, including amyloidoses. A major task is to further elucidate the mechanisms of amyloid pathology. We here show that the fibronectin type I domain of tPA mediates the interaction with amyloid protein aggregates. We found that in contrast to full-length tPA, a deletion-mutant of tPA, lacking the first three N-terminal domains (including the fibronectin type I domain), fails to activate in response to amyloid protein aggregates. Using recombinantly produced domains of tPA in direct binding assays, we subsequently mapped the amyloid-binding region to the fibronectin type I domain. This domain co-localized with congophilic plaques in brain sections from patients with Alzheimer's disease. Fibronectin type I domains from homologous proteases factor XII, hepatocyte growth factor activator and from the extracellular matrix protein fibronectin also bound to aggregated amyloidogenic peptides. Finally, we demonstrated that the isolated fibronectin type I domain inhibits amyloid-induced aggregation of blood platelets. The identification of the fibronectin type I domain as an amyloid-binding module provides new insights into the (patho-) physiological role of tPA and the homologous proteins which may offer new targets for intervention in amyloid pathology.

Regulation of fibrinolysis by thrombin activatable fibrinolysis inhibitor, an unstable carboxypeptidase B that unites the pathways of coagulation and fibrinolysis

The coagulation and fibrinolytic systems safeguard the patency of the vasculature and surrounding tissue. Cross regulation of coagulation and fibrinolysis plays an important role in preserving a balanced hemostatic process. Identification of Thrombin Activatable Fibrinolysis Inhibitor (TAFI) as an inhibitor of fibrinolysis and one of the main intermediates between coagulation and fibrinolysis, greatly improved our understanding of cross regulation of coagulation and fibrinolysis. As TAFI is an enzyme that is activated by thrombin generated by the coagulation system, its activation is sensitive to the dynamics of the coagulation system. Defects in coagulation, such as in thrombosis or hemophilia, resonate in TAFI-mediated regulation of fibrinolysis and imply that clinical symptoms of coagulation defects are amplified by unbalanced fibrinolysis. Thrombomodulin promotes the generation of both antithrombotic activated protein C (APC) and prothrombotic (antifibrinolytic) activated TAFI, illustrating the paradoxical effects of thrombomodulin on the regulation of coagulation and fibrinolysis. This review will discuss the role of TAFI in the regulation of fibrinolysis and detail its regulation of activation and its potential therapeutic applications in thrombotic disease and bleeding disorders.

Factor XI enhances fibrin generation and inhibits fibrinolysis in a coagulation model initiated by surface-coated tissue factor

In-vitro studies have shown that thrombin-mediated factor XI activation enhances thrombin and fibrin formation, rendering the clot more thrombogenic and protecting it from lysis by activation of thrombin activatable fibrinolysis inhibitor. These effects of factor XI are only observed when coagulation is initiated by a low concentration of soluble tissue factor. At high concentrations of soluble tissue factor no effects of factor XI are seen on coagulation and fibrinolysis. In vivo, tissue factor is present in large amounts in the vascular wall. This makes it difficult to extrapolate these in-vitro findings on factor XI to the in-vivo situation. To address the question of whether factor XI could play a role in coagulation initiated on a tissue factor-containing surface we devised a static in-vitro coagulation model in which clotting is initiated in recalcified citrated plasma by tissue factor coated on the bottom of microtiter plates. The effect of factor XI was studied with an antibody that blocked the activation of factor IX by activated factor XI. The tissue factor coating strategy produced clotting times similar to those obtained with cultured tissue factor-expressing vessel wall cells (smooth muscle cells, fibroblasts and activated endothelial cells) grown to confluence in the same wells. A factor XI-dependent effect on clot formation and clot lysis was observed depending on the plasma volume used. In clots formed from small amounts of plasma (100 microl) no effect of factor XI was detected. In larger clots (200-300 microl) factor XI not only increased prothrombin activation and the fibrin formation rate but also inhibited fibrinolysis. Effects of factor XI were observed at short clotting times (3-4 min) similar to the clotting times found on cultured tissue factor-expressing vessel wall cells. This is in contrast with earlier studies using soluble tissue factor, in which effects of factor XI were only observed at much longer clotting times using low soluble tissue factor concentrations. We conclude that factor XI not only enhances coagulation initiated by surface bound tissue factor but also protects the clot against lysis once it is formed. On the basis of these results, we propose a coagulation model in which initial clot formation in the proximity of the tissue factor surface is not factor XI dependent. Clot formation becomes dependent on factor XI in the propagation phase when the clot is increasing in size. These findings support a role for factor XI in the propagation of clot growth after tissue factor-dependent initiation.

Thrombin activatable fibrinolysis inhibitor (TAFI)--how does thrombin regulate fibrinolysis?

The thrombin-catalysed conversion of plasma fibrinogen into fibrin and the development of an insoluble fibrin clot are the final steps of the coagulation cascade during haemostasis. A delicate balance between coagulation and fibrinolysis determines the stability of the fibrin clot. Thrombin plays a central role in this process, it not only forms the clot but it is also involved in stabilizing the clot by activating thrombin activatable fibrinolysis inhibitor (TAFI). Activated TAFI protects the fibrin clot against lysis. Here we will discuss the mechanisms for regulation of fibrinolysis by thrombin. The role of the coagulation system for the generation of thrombin and for the activation of TAFI implies that defects in thrombin generation will directly affect the protection of clots against lysis. Thus, defects in activation of TAFI might contribute to the severity of bleeding disorders. Vice versa an increased activation of TAFI due to an increased rate of thrombin generation might lead to thrombotic disorders. Specific inhibitors of activated TAFI or inhibitors that interfere with the generation of thrombin might provide novel therapeutic strategies for thrombolytic therapy. Besides having a role in the regulation of fibrinolysis, TAFI may also have an important function in the regulation of inflammation, wound healing and blood pressure.

Role of isoleucine residues 182 and 183 in thrombin-activatable fibrinolysis inhibitor

Thrombin-activatable fibrinolysis inhibitor (TAFI) is a procarboxypeptidase that, once activated, can attenuate fibrinolysis. The active form, TAFIa, is a labile enzyme, with a half-life of a few minutes at 37 degrees C. Understanding the molecular mechanisms of TAFIa inactivation will allow the development of compounds that modulate TAFIa activity. Based on their three-dimensional model of TAFI, Barbosa Pereira et al. [J Mol Biol (2002), vol. 321, pp. 537-547] suggested that Ile182 and Ile183 were involved in the instability of TAFIa. However, these carboxypeptidases are, unlike TAFIa, stable proteases. Therefore, we constructed, expressed and characterized a TAFI mutant in which Ile182 and Ile183 were changed into the residues found in pancreas carboxypeptidase B at corresponding positions, Arg and Glu. The active form of the mutant, TAFIa-I182R-I183E, had a similar half-life as wild-type TAFIa, showing that Ile182 and Ile183 were not involved in the regulation of TAFIa stability. Remarkably, however, TAFI-I182R-I183E was activated at a lower rate by thrombin-thrombomodulin (mutant: 45 +/- 2 U L(-1) s(-1) and wild type: 103 +/- 3 U L(-1) s(-1)), thrombin (mutant: 1 +/-0.1 U L(-1) s(-1) and wild type 3 +/- 0.2 U L(-1) s(-1)) and plasmin (mutant: 0.8 +/- 0.04 U L(-1) s(-1) and wild type: 5.0 +/-0.2 U L(-1) s(-1)) compared with wild-type TAFI. Accordingly, it had a sixfold reduced antifibrinolytic potential. In conclusion, analysis of TAFI-I182R-I183E showed that I182 and I183 are not involved in TAFIa inactivation by conformational instability but that these residues may be involved in the activation of TAFI and stabilization of the fibrin clot.

Decreased plasma sensitivity to activated protein C by oral contraceptives is associated with decreases in plasma glucosylceramide

Oral contraceptive (OC) use increases venous thrombosis (VTE) risk and causes activated protein C (APC) resistance. Plasma glucosylceramide (GlcCer) deficiency is associated with VTE and GlcCer functions as an APC anticoagulant cofactor. Because estradiol decreases GlcCer in cultured cells, we hypothesized OC use would decrease plasma GlcCer and contribute to APC resistance. In a pilot study, seven female adults alternatively took second and third generation OCs and plasma samples were analyzed for GlcCer using high performance liquid chromatography and for APC sensitivity using modified prothrombin time assays. Second and third generation OC usage decreased the APC sensitivity ratio by 8.1% +/- 4.7% (P = 0.004) and 11.7% +/- 8.2% (P = 0.013) and plasma GlcCer levels by 10.1% +/- 6.8% (P = 0.008) and 11.0% +/- 5.1% (P = 0.002), respectively. The plasma GlcCer level correlated with the sensitivity of plasma to APC (P = 0.017, r = 0.51, n = 21 plasma samples). Thus, both second and third generation OC usage decreased plasma GlcCer which could cause a reduction in the plasma sensitivity to APC/protein S, thereby potentially increasing VTE risk.

Thrombin activatable fibrinolysis inhibitor (TAFI) at the interface between coagulation and fibrinolysis

The thrombin-catalysed conversion of plasma fibrinogen into fibrin and the development of an insoluble fibrin clot are the final steps of the coagulation cascade during haemostasis. A delicate balance between coagulation and fibrinolysis determines the stability of the fibrin clot. Thrombin Activatable Fibrinolysis Inhibitor (TAFI) plays an important role in this process. TAFI is activated by thrombin and protects the fibrin clot against lysis. The role of TAFI in bleeding and thrombotic disorders is discussed as well as its novel emerging role in wound healing and inflammation.

Tumor growth and metastasis are not affected in thrombin-activatable fibrinolysis inhibitor-deficient mice

Many studies have indicated that the plasminogen activation system may have a prominent role in cancer. Activation of the zymogen plasminogen into the serine protease plasmin by plasminogen activator is mediated by carboxyterminal basic amino acids in fibrin, including lysines and arginines. Thrombin-activatable fibrinolysis inhibitor (TAFI) is a circulating carboxypeptidase B-type proenzyme that, after activation, removes carboxyterminal lysine or arginine residues in fibrin, resulting in decreased plasminogen activation and attenuated fibrinolysis. To determine directly whether TAFI is involved in primary tumor growth and metastasis formation, we examined the effects of TAFI deficiency on subcutaneous growth and experimentally or spontaneously induced pulmonary metastasis formation of different tumor cell types in mice. In all tumor models TAFI deficiency did not affect the formation and growth of primary and metastasized tumors.

Generation and Characterization of a Highly Stable Form of Activated Thrombin-activable Fibrinolysis Inhibitor

Activated thrombin-activable fibrinolysis inhibitor (TAFIa) is a carboxypeptidase B that can down-regulate fibrinolysis. TAFIa is a labile enzyme that can be inactivated by conformational instability or proteolysis. TAFI is approximately 40% identical to pancreatic carboxypeptidase B (CPB). In contrast to TAFIa, pancreatic CPB is a stable protease. We hypothesized that regions or residues that are not conserved in TAFIa compared with pancreatic CPB play a role in the conformational instability of TAFIa and that replacement of these non-conserved residues with residues of pancreatic CPB would lead to a TAFIa molecule with an increased stability. Therefore, we have expressed, purified, and characterized two TAFI-CPB chimeras: TAFI-CPB-(293-333) and TAFI-CPB-(293-401). TAFI-CPB-(293-333) could be activated by thrombin-thrombomodulin, but not as efficiently as wild-type TAFI. After activation, this mutant was unstable and was hardly able to prolong clot lysis of TAFI-deficient plasma. Binding of TAFI-CPB-(293-333) to both plasminogen and fibrinogen was normal compared with wild-type TAFI. TAFI-CPB-(293-401) could be activated by thrombin-thrombomodulin, although at a lower rate compared with wild-type TAFI. The activated mutant displayed a markedly prolonged half-life of 1.5 h. Plasmin could both activate and inactivate this chimera. Interestingly, this chimera did not bind to plasminogen or fibrinogen. TAFI-CPB-(293-401) could prolong the clot lysis time in TAFI-deficient plasma, although not as efficiently as wild-type TAFI. In conclusion, by replacing a region in TAFI with the corresponding region in pancreatic CPB, we were able to generate a TAFIa form with a highly stable activity.

Characteristics of piraltin, a polyphenol concentrate, produced by freeze-drying of red wine

Moderate consumption of red wine is associated with a decreased risk for coronary heart disease. Apart from alcohol, an additive role for wine polyphenols has been suggested. However, the real contribution of these compounds can only be studied when available without the alcohol component. The objective of the study was to prepare a wine polyphenol concentrate not containing alcohol and to compare the quantitative and qualitative properties of this concentrate with those of the original wine from which the concentrate is made. This polyphenol concentrate, called piraltin, was made out of red wine by a freeze-drying technique. Both piraltin and the original red wine were analyzed quantitatively for the main polyphenols present: gallic acid, catechin, epicatechin and quercetin. The qualitative comparison comprised the inhibitory effect of the two products on LDL oxidation in vitro. In the process of freeze-drying recovery of the four determined flavonoids of red wine is fairly constant (average 68 +/- 7%). In a copper induced LDL oxidation assay both red wine and piraltin prolonged lag-times over 300% compared to controls without a significant difference between the two products. The freeze-dried polyphenol concentrate piraltin contains about 70% of the total polyphenol content of the original wine. This preparation technique does not cause a loss of antioxidative properties of the phenols. Piraltin creates the possibility to study the effects of wine polyphenols separately without the influence of alcohol both in vitro and in vivo.

New insights into factors affecting clot stability: a role for thrombin activatable fibrinolysis inhibitor (TAFI; plasma procarboxypeptidase B, plasma procarboxypeptidase U, procarboxypeptidase R)

The thrombin-catalyzed conversion of plasma fibrinogen into fibrin and the development of an insoluble fibrin clot are the final steps in the coagulation cascade during hemostasis. The delicate balance between clot formation and fibrinolysis, which determines clot stability, is controlled by a complex interplay between fibrin and other molecular and cellular components of the hemostatic system, including thrombin activatable fibrinolysis inhibitor (TAFI). TAFI is activated by thrombin and has an important role in the stability of the fibrin clot, which is reviewed here. In particular, the role of TAFI in fibrinolysis and those characteristics of the protein that affect clot stability are described. In addition, the importance of TAFI in the coagulation process and how changes in its availability may contribute to bleeding or thrombotic disorders are discussed.

Effect of second- and third-generation oral contraceptives on the protein C system in the absence or presence of the factor VLeiden mutation: a randomized trial

A plausible mechanism to explain thrombotic risk differences associated with the use of second- and third-generation oral contraceptives (OCs), particularly in carriers of factor V(Leiden), is still lacking. In a double-blind trial, 51 women without and 35 women with factor V(Leiden) were randomized to either a second- (30 microg ethinylestradiol/150 microg levonorgestrel) or third- (30 microg ethinylestradiol/150 microg desogestrel) generation OC. After 2 cycles of use and a wash-out of 2 cycles, the participants continued with the corresponding progestagen-only preparation. Hemostatic variables that probe the activity of the anticoagulant protein C system were determined. Compared with levonorgestrel, desogestrel-containing OCs significantly decreased protein S and increased activated protein C (APC) resistance in both groups. OCs with desogestrel had the most pronounced effects in carriers of factor V(Leiden). Progestagen-only preparations caused changes of anticoagulant parameters opposite to those of combined OCs, which in a number of cases were more pronounced with levonorgestrel. Our data show that progestagens in combined OCs counteract the thrombotic effect of the estrogen component. The higher thrombotic risk associated with third-generation OCs compared with second-generation OCs may be explained by the fact that desogestrel appeared less antithrombotic than levonorgestrel, especially in women with factor V(Leiden).

Impaired healing of cutaneous wounds and colonic anastomoses in mice lacking thrombin-activatable fibrinolysis inhibitor

Plasmin and other components of the plasminogen activation system play an important role in tissue repair by regulating extracellular matrix remodeling, including fibrin degradation. Thrombin-activatable fibrinolysis inhibitor (TAFI) is a procarboxypeptidase that, after activation, can attenuate plasmin-mediated fibrin degradation by removing the C-terminal lysine residues from fibrin, which play a role in the binding and activation of plasminogen. To test the hypothesis that TAFI is an important determinant in the control of tissue repair, we investigated the effect of TAFI deficiency on the healing of cutaneous wounds and colonic anastomoses. Histological examination revealed inappropriate organization of skin wound closure in the TAFI knockout mice, including an altered pattern of epithelial migration. The time required to completely heal the cutaneous wounds was slightly delayed in TAFI-deficient mice. Healing of colonic anastomoses was also impaired, as reflected by decreased strength of the tissue at the site of the suture, and by bleeding complications in 3 of 14 animals. Together, these abnormalities resulted in increased mortality in TAFI-deficient mice after colonic anastomoses. Although our study shows that tissue repair, including re-epithelialization and scar formation, occurs in TAFI-deficient mice, TAFI appears to be important for appropriate organization of the healing process.

Thrombin-activatable fibrinolysis inhibitor (TAFI, plasma procarboxypeptidase B, procarboxypeptidase R, procarboxypeptidase U)

Recently, a new inhibitor of fibrinolysis was described, which downregulated fibrinolysis after it was activated by thrombin, and was therefore named TAFI (thrombin-activatable fibrinolysis inhibitor; EC 3.4.17.20). TAFI turned out to be identical to the previously described proteins, procarboxypeptidase U, procarboxypeptidase R, and plasma procarboxypeptidase B. Activated TAFI (TAFIa) downregulates fibrinolysis by the removal of carboxy-terminal lysines from fibrin. These carboxy-terminal lysines are exposed upon limited proteolysis of fibrin by plasmin and act as ligands for the lysine-binding sites of plasminogen and tissue-type plasminogen activator (t-PA). Elimination of these lysines by TAFIa abrogates the fibrin cofactor function of t-PA-mediated plasminogen activation, resulting in a decreased rate of plasmin generation and thus downregulation of fibrinolysis. In this review, the characteristics of TAFI are summarized, with an emphasis on the pathways leading to activation of TAFI and the role of TAFIa in the inhibition of fibrinolysis. However, it cannot be ruled out that TAFI has other, as yet undefined, functions in biology.

Identification of thrombin activatable fibrinolysis inhibitor (TAFI) in human platelets

Thrombin activatable fibrinolysis inhibitor (TAFI) is a carboxypeptidase B-like proenzyme that after activation down-regulates fibrinolysis. Platelets are known to contain antifibrinolytic factors that are secreted during platelet activation. Therefore, the presence of TAFI in platelets was analyzed. TAFI was identified in platelets in a concentration of about 50 ng/1 x 109 platelets and was secreted on platelet activation. Thrombin-mediated activation of platelet-derived TAFI resembled that of plasma-derived TAFI with respect to stimulation by thrombomodulin and spontaneous loss of activity at 37 degrees C. The different glycosylation of platelet-derived TAFI compared with plasma-derived TAFI suggests that platelet-derived TAFI is synthesized in the megakaryocyte. This suggestion was substantiated by the detection of mRNA in the megakaryocytic cell lines DAMI and CHRF, representing the intermediate and late stages of megakaryocyte development. These results establish the presence of TAFI in platelets and suggest a role for platelet-derived TAFI in the protection of the clot against fibrinolysis.

Amyloid endostatin induces endothelial cell detachment by stimulation of the plasminogen activation system

Endostatin is a fragment of collagen XVIII that acts as an inhibitor of tumor angiogenesis and tumor growth. Anti-tumor effects have been described using both soluble and insoluble recombinant endostatin. However, differences in endostatin structure are likely to cause differences in bioactivity. In the present study, we have investigated the cellular effects of insoluble endostatin. We previously found that insoluble endostatin shows all the hallmarks of amyloid aggregates and potently stimulates tissue plasminogen activator-mediated formation of the serine protease plasmin. We here show that amyloid endostatin induces plasminogen activation by endothelial cells, resulting in vitronectin degradation and plasmin-dependent endothelial cell detachment. Endostatin-mediated stimulation of plasminogen activation, vitronectin degradation, and endothelial cell detachment is inhibited by carboxypeptidase B, indicating an essential role for carboxyl-terminal lysines. Our results suggest that amyloid endostatin may inhibit angiogenesis and tumor growth by stimulating the fibrinolytic system.

No acute effect of red wine on the coagulation pathway in healthy men

Binge drinking is associated with an increased risk of cardiovascular events. Those events often happen within hours after alcohol is consumed. Apart from arrhythmias and changes in blood pressure, these events may be caused by an acute (i.e., occurring within a 24-h period) shift of the hemostatic balance in a thrombogenic direction. Alcohol can influence platelet aggregation and inhibit fibrinolysis, but little is known about its direct effect on coagulation. In the current study, parameters of coagulation, reflecting either stimulation or inhibition, were measured 5 and 15 h after the consumption of four (62.5 g of alcohol) and eight (125 g of alcohol) glasses of red wine. Both doses had no direct effect on activated cephalin time, thrombin-antithrombin complexes, factors VII and VIII, and von Willebrand factor. In contrast with the observed effects on thrombocytes and fibrinolysis, the consumption of large amounts of wine does not influence the coagulation pathway.

Changes of Hemostatic Variables during Oral Contraceptive Use

The use of oral contraceptives (OCs) has been known for many years to affect significantly almost all hemostatic parameters, but the challenge to relate these changes in a meaningful way to OC-induced increased venous thrombotic risk has not been met. New insights indicate that at least part of the answer can be found in the net effect of OC use on the efficacy with which the protein C pathway down-regulates thrombin formation. During OC use the (blood) plasma of a woman becomes resistant to the anticoagulant action of activated protein C (APC). The extent of this so-called acquired APC resistance as determined in a thrombin generation-based assay correlates remarkably well with the risk increases observed in clinical studies. Recent evidence indicates that the prothrombotic effect of the estrogen component ethinylestradiol in combined OC is counteracted by the progestagen component present in these preparations and that third-generation progestagens such as desogestrel or gestodene are less efficient with respect to this than the second-generation progestagen levonorgestrel.

Regular alcohol intake and fibrinolysis

BACKGROUND

Light to moderate alcohol consumption is associated with a reduced risk of coronary heart disease. Stimulation of fibrinolysis has been suggested as one of the mechanisms involved. The present study analyses the effect of regular alcohol consumption on various parameters of fibrinolysis. The question whether the alcohol-induced plasma increase of plasminogen activator inhibitor (PAI-1) may originate from thrombocytes was also addressed.

METHODS

Six healthy male volunteers consumed three glasses of red wine daily during two periods of a week, with a week of abstinence from alcohol in between. PAI-1 antigen and activity levels, t-PA antigen and activity levels and plasmin antiplasmin (PAP) complexes were measured on days 1, 3, 8, 15, 17 and 22 of the experiment period. On the first day, PAI-1 antigen and activity before and after alcohol consumption was also measured in platelet-rich plasma (prp).

RESULTS

Although some slight shifts in the various parameters could be noticed during the drinking periods, all favouring impairment rather than stimulation, no significant effect of regular moderate alcohol use could be observed on fibrinolysis. Alcohol did not trigger a release of PAI-1 from platelets.

CONCLUSIONS

Regular moderate alcohol consumption has no significant effect on fibrinolysis. The alcohol-induced increase of plasma PAI-1 does not originate from thrombocytes. The cardioprotective effect of moderate alcohol consumption cannot be explained by a beneficial influence on fibrinolysis.

Protein C inhibitor (plasminogen activator inhibitor-3) and the risk of venous thrombosis

Protein C inhibitor (PCI), also known as plasminogen activator inhibitor-3, is a serine proteinase inhibitor that can inhibit enzymes in blood coagulation, fibrinolysis and fertility. The role of PCI in regulating the blood coagulation mechanism is not known, as it can inhibit both procoagulant (thrombin, factor Xa, factor XIa) and anticoagulant (activated protein C, thrombin-thrombomodulin, urokinase) enzymes. To determine the relevance of this inhibitor in thrombosis, PCI levels were assessed in the Leiden Thrombophilia Study, a case-control study of venous thrombosis in 473 patients with a first deep-vein thrombosis and 474 age- and sex-matched control subjects. PCI levels above the 95th percentile of the controls (136.1%) increased the risk 1.6-fold compared with PCI levels below the 95th percentile (95% confidence interval 0.9-2.8). There was a gradual increase in risk of thrombosis with further increasing levels of PCI. Adjustment for a number of possible confounders led to a reduction of the risk estimates associated with PCI. However, it is unclear whether adjustment for such factors in the risk models is justified. These results indicate that high levels of PCI may constitute a mild risk factor for venous thrombosis.

Effect of second- and third-generation oral contraceptives on fibrinolysis in the absence or presence of the factor V Leiden mutation

Third-generation oral contraceptives (OC) have been associated with an increased risk of venous thrombosis compared with second-generation OC. To find an explanation for this increased risk, the effect of a second- and third-generation OC and of the progestagens used in these pills on several fibrinolytic parameters was studied in the absence or presence of the factor V Leiden mutation. In a single-center, double-blind trial, 51 women without and 35 women with the factor V Leiden mutation were randomized to either a second-generation (30 microg ethinylestradiol/150 microg levonorgestrel) or a third-generation (30 microg ethinylestradiol/150 microg desogestrel) oral contraceptive. After two menstrual cycles of use and a wash-out period of two cycles, the participants received the corresponding progestagen-only preparation containing 150 microg levonorgestrel or 150 microg desogestrel. D-Dimers, thrombin-activatable fibrinolysis inhibitor (TAFI) and the clot lysis time in the absence (LYSmin) or the presence (LYSplus) of a blocking anti-factor XI antibody were determined in plasmas of the participating women, and the mean difference in changes between the OC were calculated. Both combined OC induced increased plasma levels of D-dimers and TAFI, and induced a prolongation of LYSplus, whereas LYSmin hardly changed. Virtually no changes in fibrinolytic parameters were observed for the progestagen-only preparations. No differential effects between levonorgestrel- and desogestrel-containing OC were found in women without factor V Leiden. Women with the mutation on levonorgestrel-containing OC showed an increased LYSplus compared with desogestrel containing OC (3.9; 95% confidence interval, 0.1-7.7). When using progestagen-only preparations, no differential effect on the fibrinolytic parameters were found, except for non-carriers on levonorgestrel who showed a reduced LYSmin compared with non-carriers on desogestrel (-4.0; 95% confidence interval, -7.8 to -0.2). In conclusion, the effect of oral contraceptives on fibrinolytic parameters is largely independent of the type of progestagen. The increased fibrinolytic activity during OC use appears to be induced by the estrogen component and may be counteracted by increased TAFI activation. This may result in an enhanced downregulation of fibrinolysis.

Plasmin-mediated activation and inactivation of thrombin-activatable fibrinolysis inhibitor

Activated thrombin-activatable fibrinolysis inhibitor (TAFIa) attenuates the fibrin cofactor function of tissue-type plasminogen activator-mediated plasmin formation and subsequently fibrin degradation. In the present study, we focused on the role of plasmin in the regulation of TAFIa activity. Upon incubation with plasmin, TAFIa activity was generated, which was unstable at 37 degrees C. Analysis of the cleavage pattern showed that TAFI was cleaved at Arg(92), releasing the activation peptide from the 35.8-kDa catalytic domain. The presence of the 35.8-kDa fragment paralleled the time course of generation and loss of TAFIa activity. This suggested that, in the presence of plasmin, TAFIa is probably inactivated by proteolysis rather than by conformational instability. TAFI was also cleaved at Arg(302), Lys(327), and Arg(330), resulting in a approximately 44.3-kDa fragment and several smaller fragments. The 44.3-kDa fragment is no longer activatable since it lacks part of the catalytic center. We concluded that plasmin can cleave at several sites in TAFI and that this contributes to the regulation of TAFI and TAFIa.

Interaction of human plasma kallikrein and its light chain with C.hivin.1 inhibitor

The light chain of human plasma kallikrein contains the enzymatic active site. The inactivation of kallikrein and of its isolated light chain by C1 inhibitor was investigated to assess the functional contributions of the heavy-chain region of kallikrein and of high molecular weight kininogen to this reaction. The second-order rate constants for the inactivation of kallikrein or its light chain were respectively 2.7 X 10(6) and 4.0 X 10(6) M -1 min -1. High molecular weight kininogen did not influence the rate of kallikrein inactivation. The nature of the complexes formed between kallikrein or its light chain and C1 inhibitor was studied by using sodium dodecyl sulfate (SDS) gradient polyacrylamide slab gel electrophoresis. Kallikrein as well as its light chain combined with C1 inhibitor to form stable stoichiometric complexes that were not dissociated by SDS and that exhibited apparent molecular weights (Mr's) of 185 000 and 135 000, respectively, on nonreduced SDS gels. Reduction of the kallikrein-C1 inhibitor complex gave a band at Mr 135 000 that comigrated with the complex seen for the light chain-C1 inhibitor complex. During the inactivation of both kallikrein and its light chain, a Mr 94 000 fragment of C1 inhibitor was formed which was unable to inactivate or bind kallikrein or its light chain. Kallikrein inactivated by diisopropyl phosphofluoridate did not form SDS-stable complexes with C1 inhibitor. These results demonstrate that the functional binding site for C1 inhibitor is localized in the light chain of kallikrein.(ABSTRACT TRUNCATED AT 250 WORDS)