Effect on platelet FXIII and partial characterization of Lonomin V, a proteolytic enzyme from Lonomia achelous caterpillars

Contact with Lonomia achelous caterpillars venom induces a severe bleeding syndrome in humans. A constant finding in all reported cases is a marked decrease of blood coagulation factor XIII (FXIII), which has been attributed to the presence of a proteolytic enzyme, isolated and named Lonomin V, in the hemolymph and hair secretion. In this study, the effect of Lonomin V on transglutaminase activity from human plasma, rabbit plasma, and platelet FXIII was analyzed. The decrease of activity was more pronounced in platelet (A2) when compared with rabbit plasma (AB) and human plasma FXIII (A2B2). This finding might be explained by the differences in FXIII molecular structure. In addition, platelet FXIII molecule was degraded by Lonomin to several fragments of low molecular mass. Lonomin V was stable over a wide range of pH (6-8.5) and temperatures of -70 degrees C, -20 degrees C and between 4 to 24 degrees C, with a progressive decrease at 37 degrees C and total inactivation at 60 degrees C after 2 hours incubation. Diisopropyl fluoro-phosphate, phenylmethylsulfonyl fluoride, tosyl-l-lysine chloromethyl ketone, and iodoacetamide abolished the effect of Lonomin V on FXIII; in contrast dithiothreitol and EDTA-Na enhance the activity. We concluded that Lonomin V is a serine proteinase with a free Cys essential for the enzymatic activity. Due to its proteolytic activity on FXIII, with concomitant impairment of fibrin cross-linking, Lonomin V might be useful in association with thrombolytic drugs for preventing rethrombosis.     

Alterations of hemostasis associated with cardiopulmonary bypass.

Cardiopulmonary bypass (CPB) has become common surgery, but remains associated with poorly understood derangements in hemostasis. These alterations may lead to significant hemorrhage and subsequent increased morbidity and mortality. To better define alterations of hemostasis associated with CPB, coagulation proteins, fibrinolysis, and platelets were evaluated in thirty consecutive patients undergoing this procedure. Results of these studies have revealed activation of the fibrinolytic system with the presence of circulating plasmin, elevated fibrin(ogen) degradation products, and depletion of plasminogen and fibrinogen. All patients were void of soluble fibrin monomers of disseminated intravascular coagulation (DIC) and demonstrated only minimal to moderate thrombocytopenia. All patients undergoing CPB developed a severe functional platelet defect as assessed by measuring platelet retention. At one hour post-CPB, this defect only partially corrects. Hyperheparinemia, heparin rebound, and protamine excess were not noted in any patient with doses of heparin and protamine used in this series. Thus, these studies have revealed a primary hyperfibrino(geno)lytic state to occur in the majority of patients undergoing CPB; in addition, a severe defect in platelet function occurs in all patients subjected to this procedure. These two alterations of hemostasis occurring during CPB account for most instances of non-technical hemorrhage associated with this procedure. A quick search for these defects should lead to rapid diagnosis, efficacious therapy, and a subsequent reduction in morbidity and mortality due to CPB hemorrhage.     

Fibronectin and coagulation factor XIII increases blood platelet adhesion to fibrin

Fibronectin is covalently linked to fibrin during clot formation by coagulation factor XIII and has been suggested as a possible mediator of platelet adhesion to collagen. Fixed human platelets were found to adhere to fibrin. This adhesion was significantly increased when fibronectin was incorporated into the fibrin network, and was supported by factor XIII.     

A hamster antibody to the mouse fibrinogen gamma chain inhibits platelet-fibrinogen interactions and FXIIIa-mediated fibrin cross-linking, and facilitates thrombolysis.

Murine models employing genetically altered mice have the potential to provide important new information about the hemostatic system, but before such data can be extrapolated to humans it is necessary to define the similarities and differences between murine and human hemostasis. After establishing the similarities of murine fibrinogen to human fibrinogen in its pattern of proteolysis in response to plasmin and its cross-linking by factor XIIIa, we studied a new hamster monoclonal antibody (mAb) 7E9 that reacts with the gamma chain of mouse fibrinogen. This antibody inhibits platelet adhesion to fibrinogen, platelet-mediated clot retraction, platelet aggregation, and FXIIIa-mediated cross-linking of fibrin; it also facilitates tissue plasminogen activator (tPA)-mediated lysis of fibrin formed either in the absence or presence of platelets. These data provide evidence that the C-terminus of mouse fibrinogen gamma chain, like that of human fibrinogen, is involved in fibrinogen binding to platelets and FXIIIa-mediated cross-linking of fibrin. Our data raise the possibility that a therapeutic agent that targets the C-terminus of the gamma chain in human fibrinogen might have broad antithrombotic and profibrinolytic effects.     

Strategies to reduce hemostatic activation during cardiopulmonary bypass

INTRODUCTION: We evaluated whether a modified protocol for cardiopulmonary bypass (CPB) could reduced the systemic hemostatic activation associated with this procedure. MATERIALS AND METHODS: The in vivo rates of thrombin, fibrin, plasmin and D-dimer generation were determined in each subject during CPB using measured levels of hemostatic factors combined with a computer model of the cardiovascular and hemostatic systems. A standard CPB group using uncoated circuits, standard heparin levels and direct shed blood reinfusion (n=9) was compared to a modified CPB group using heparin-coated circuits, shed blood collection, washing and reinfusion post-operatively, lower heparin levels and epsilon-amino-caproic acid (n=10). RESULTS AND CONCLUSIONS: Standard CPB increased average thrombin generation 9-fold, decreased fibrin generation 2-fold, increased plasmin generation 11-fold and increased fibrin degradation and D-dimer generation 19-fold. During CPB in the modified group thrombin generation was not increased beyond surgical levels, lower heparin concentrations allowed each thrombin to make more fibrin prior to inhibition, while fibrin degradation was suppressed by epsilon-amino-caproic acid. At baseline for every 100 fibrins formed only 1-2 were degraded to D-dimer. During standard CPB for every 100 fibrins generated on average 34 fibrins were degraded with some subjects showing a net fibrin loss. In contrast, in the modified CPB group for every 100 fibrins formed only 4 fibrins were degraded (p<0.0001 vs. standard group). Kinetic modeling of hemostasis in individual patients showed that a modified CPB protocol could reduce excessive thrombin generation during CPB and suppress fibrin degradation moving hemostatic regulation back towards normal.     

Factor XIII, clot structure, thrombosis

Blood coagulation factor XIII (FXIII) is a tetrameric protein consisting of two catalytic A (FXIII-A) and two carrier/inhibitory B (FXIII-B) subunits. It is a zymogen, which becomes transformed into an active transglutaminase (FXIIIa) in the final phase of coagulation cascade by thrombin and Ca^2 +. FXIII is essential for hemostasis, its deficiency results in severe bleeding diathesis. FXIIIa mechanically stabilizes fibrin by cross-linking its α-, and γ-chains. It also protects newly formed fibrin from fibrinolysis, primarily by cross-linking α₂-plasmin inhibitor to fibrin. Beside the above prothrombotic effects, it is involved in limiting thrombus growth by down-regulating platelet adhesion to fibrin. Elevated FXIII level seems to be a gender-specific risk factor of both coronary artery disease and peripheral arterial disease, it represents an increased risk only in females. The association of FXIII level with the risk of ischemic stroke and venous thromboembolism was investigated only in a few studies from which no clear conclusion could be drawn. Among the FXIII subunit polymorphisms, concerning their effect on the risk of thrombotic diseases, only FXIII-A p.Val34Leu was investigated extensively. Meta-analyses of reported data suggest that this polymorphism provides a moderate protection against coronary artery disease and venous thromboembolism, but not against ischemic stroke. Gene-gene and gene-environmental interactions might modify its effect. Further studies are required to explore the effect of other FXIII subunit polymorphism on the risk of thrombotic diseases.     

In vivo effects of a low molecular weight heparin fragment on platelet aggregation and platelet dependent hemostasis in dogs

Plasma defibrinogenated dogs were used to study the influence of conventional heparin and a low molecular weight heparin fragment (Fragmin, mean MW 5,000 d) on platelet dependent hemostasis. The heparins were given intravenously in gravimetrically equal doses. The bleeding from standardized skin flap wounds and platelet aggregation (ADP and thrombin) was studied. In comparison, higher doses of the fragment than of heparin were required to increase the bleeding. ADP-induced aggregation in defibrinogenated platelet rich plasma (after addition of normal dog plasma) was potentiated by both heparins. After injection of heparin or the fragment, ADP induced platelet aggregation without prior addition of normal plasma to the test-tube. In conclusion the heparin fragment affected bleeding to a less extent than conventional heparin. One explanation might be a weaker inhibition of thrombin-induced platelet aggregation.     

Tranexamic acid inhibits fibrinolysis, shortens the bleeding time and improves platelet function in patients with chronic renal failure.

BACKGROUND: A defect in platelet function is the main determinant of the prolonged bleeding time in chronic renal failure (CRF). We previously reported a significant correlation between platelet abnormalities and elevated plasma markers of plasmin and thrombin generation. Our aim was to explore the effect of inhibiting both plasmin action with tranexamic acid (TA) and thrombin production with low molecular weight heparin (LMWH), on the bleeding time (BT) and platelet function in patients with CRF. METHODS: 37 patients with CRF (mean creatinine 8.6 +/- 4.4 mg/dl) under conservative treatment, with prolonged BT, entered this study and received TA during 6 days, with (n = 24) and without LMWH (n = 13). BT, platelet aggregation/secretion, platelet granule contents, von Willebrand factor and parameters of coagulation and fibrinolysis were recorded before and at the end of treatment. RESULTS: The BT was shortened in 26/37 (67%) patients. This effect was associated with significant improvement of platelet aggregation and secretion, with decrease to a normal range of fibrin/fibrinogen degradation products, mild increase in plasmin-antiplasmin complexes and pronounced reduction of circulating plasminogen. No differences were seen among patients with or without LMWH. No serious side effects or complications were observed. INTERPRETATION: These findings indicate that the activation of fibrinolysis plays a significant role in the defect of primary hemostasis in patients with CRF. Inhibition of plasmin activity with TA shortens the BT and improves platelet function in the majority of patients with severe disease.     

Comparison of platelet fibronectin, ADP-induced platelet aggregation and serum total nitric oxide (NOx) levels in angiographically determined coronary artery disease

INTRODUCTION: Coronary thrombosis is an important determinant of prognosis in patients with acute coronary syndromes. Fibronectin is also found in platelets within the alpha secretory granules and secreted following platelet stimulation by a variety of agonist. Available data suggest that expression of platelet fibronectin on the cell surface may indicate a role in platelet aggregation and adhesion to fibrin thrombi and connective tissue. Clear evidence has emerged that a concerted action of nitric oxide (NO) generated by either endothelial or platelet NO synthases regulates platelet activation, causing inhibition of adhesion and aggregation. The aim of the present study was determining and correlating the serum total NO (NOx), platelet fibronectin and ADP-induced platelet aggregation levels in coronary artery disease (CAD) patient subgroups. MATERIALS AND METHODS: A total of 43 coronary artery disease patients were included in this study. Peripheral blood samples from patients with coronary artery disease were obtained from the Department of Cardiology. Platelet aggregation tests with adenosine diphosphate (ADP) were analyzed by using aggregometer. Platelet fibronectin concentrations were determined by enzyme-linked immunosorbent assay (ELISA). Serum total nitric oxide (NOx) levels were determined by colorimetric method. RESULTS: In patients with double-vessel disease, platelet fibronectin levels were found to be significantly higher than no-vessel disease (p<0.001), single-vessel disease (p<0.01) and triple-vessel disease (p<0.001). In addition, in patients with single-vessel disease platelet fibronectin levels significantly higher than no-vessel disease (p<0.05). We could not find any significant differences in ADP-induced platelet aggregation and serum NOx values between CAD patient subgroups. There was a positive correlation between platelet fibronectin levels and severity of disease (r=0.315, p<0.05).     

Thrombospondin binds to amino-terminal fragments of plasma fibronectin

Thrombospondin is a 420-kD trimeric glycoprotein that can bind type V collagen, heparin, fibrinogen and certain cells and may be one of the lectins responsible for platelet aggregation. Thrombospondin binds another glycoprotein, fibronectin, that is also released during platelet aggregation and also binds similar ligands. This work shows that the amino-terminal 29-kD segment of fibronectin binds thrombospondin, the interaction occurs within minutes, and one 29-kD molecule binds per thrombospondin subunit. The interaction was not inhibited by fibrinogen, type V collagen, or heparin. Two subfragments of the 29-kD fragment, an amino-terminal 20-kD and a carboxyl-terminal 8-kD subfragment, the latter containing a single disulfide-rigidified type I loop of the five homologous loops in the 29-kD fragment, reacted with thrombospondin while the reduced counterparts did not.     

Aprotinin counteracts heparin-induced inhibition of platelet contractile force

BACKGROUND: Aprotinin interferes with heparin binding to platelets and decreases blood loss during cardiopulmonary bypass (CPB). Heparin abolishes platelet force during CPB, and the extent of platelet force recovery after protamine administration appears to correlate with blood loss. This study assessed the effect of aprotinin on heparin suppression of platelet force. METHODS: Platelet force was measured using the Hemodyne Hemostasis Analyzer. Clots were formed from platelet-rich plasma (PRP) by the addition of batroxobin and 10 mM CaCl(2). Clotting conditions included pH 7.4, ionic strength 0.15 M, fibrinogen level 1 mg/ml and 75,000 platelets/microl. RESULTS: After 1200 s of clotting, force was reduced from 7110+/-1190 to 450+/-450 dyn by 0.2 U/ml of heparin. Platelet force in aprotinin [20 microg/ml (140 KIU/ml)] containing PRP was not suppressed by heparin addition (7480+/-2410 dyn). Aprotinin [40 microg/ml (280 KIU/ml)] addition to previously heparinized plasma counteracted heparin force suppression. Aprotinin (40 microg/ml) increased platelet force from 5630 to 11,138+/-562 in PRP devoid of heparin. Aprotinin did not affect thrombin activity, fibrin structure, platelet aggregation or secretion. CONCLUSIONS: Aprotinin counteracts heparin suppression of platelet force and enhances platelet force in the absence of heparin. Aprotinin-heparin-platelet interactions may help explain aprotinin's ability to reduce blood loss during CPB.     

Bleeding in uremic patients after carbenicillin.

Hemorrhagic diathesis was observed in patients with renal insufficiency after carbenicillin at serum levels greater than 300 mug/ml. Normal coagulation factors (F. I, II, V, VII, VIII, X), normal PTT, normal platelet counts, negative ethanol gelation test (fibrin monomers) were found as well as a prolongation of thromboplastin time (Quick), thrombin time, reptilase time and thrombin coagulase time. Platelet function was disturbed. In addition, the plasmatic system was involved: inhibition of fibrinogen-fibrin conversion (Belitser assay) and enhanced antithrombin III activity; in vivo the latter was ascribed to a heparin-like activity. In vitro, abnormal III was seen: however an enhanced antithrombin III activity in vitro was not found with carbenicillin and various penicillin derivatives. This study demonstrates that carbenicillin, in addition to its known effect on platelet function, also disturbs the plasmatic coagulation system. This additional effect of carbenicillin is clinically important since protamin chloride effectively blocks bleeding without interfering with antibacterial activity. Both penicillin and penicillin derivatives have been shown to interfere with hemostasis and to cause clinically manifest hemorrhagic diathesis (Fleming and Fish 1947, Lurie et al. 1970a, b, McClure et al. 1970, Yudis et al. 1972, Demos 1971, Waisbren et al. 1971). Carbenicillin interferes with ADP-, collagen- or thrombin-induced platelet aggregation and with the release reaction both in vivo (McClure et al. 1970, Cazenae et al. 1973) and in vitro (McClure et al. 1970, Cazenave et al. 1973). In addition Lurie and colleagues (1970b) concluded that an inhibition of the conversion of fibrinogen to fibrin is involved although no experimental details were given. Later Brown and colleagues (1974) concluded that carbenicillin at usual dose levels "only affects the platelet component of hemostasis and has little effect on fibrin formation or other phases of coagulation in patients with normal renal function".     

Prevention of platelet adhesion and aggregation by a glutardialdehyde-stabilized heparin surface

A stable heparinized surface was prepared by sequential treatment of polyethylene with water solutions of hexadecylamine hydrochloride, heparin and glutardialdehyde. In order to explain the "non-thrombogenic" properties of this surface, it was evaluated with regard to prevention of platelet adhesion and aggregation. Human heparinized blood (2 and 10 IU/ml) with 51Cr-labelled autologous platelets was rotated for 60 minutes in untreated and heparin-treated circular tubings. The surface area/blood volume ratio was varied and an air-blood interface was present. In untreated tubings, platelet adhesion and aggregation increased in proportion to the size of the surface area/blood volume ratio, irrespective of the heparin concentrations of the blood. In the heparin-treated tubings, there was no measurable platelet adhesion to the surface and no platelet aggregation in the blood. The difference between the heparinized and the untreated surfaces with regard to platelet adhesion was discernible even after 10 minutes storage of stagnant blood. It is concluded that platelet adhesion and aggregation induced by exposure of blood to a foreign surface in an in vitro experimental model can be prevented by a stable heparin coating of the surface.     

Inaccessibility to ligands of the amino-terminal region of plasma fibronectin

Fragments of plasma fibronectin can display properties not expressed by fibronectin. This work shows that the relative affinities of smaller fragments for gelatin or heparin increases with decreasing fragment size. When heparin or gelatin were added to fragments or to fibronectin, the affinities for the alternate ligand increased. The UV and CD spectra of a mixture of amino-terminal 29-kD and 50-kD fragments differed from that of the precursor, the intact 72-kD fragment, suggesting differences in secondary structure. Therefore, biologic activities of the 29-kD and 50-kD segments in fibronectin may be suppressed by structural constraints but may be expressed more fully by interaction with ligands.     

Free N-terminal fibronectin 30-kD-domain mediates binding of soluble fibrin to gelfiltered unstimulated thrombocytes

Gelfiltered unstimulated human platelets neither bound 125-I-fibrinogen nor 125-I-fibrin. Fibrin-binding was, however, stimulated by N-terminal fibronectin 30 kD-and 70 kD-fragments while fibronectin was ineffective. The 30 kD-fragment also stimulated some platelet preparations to bind fibrinogen which, however, was suppressed by minute amounts of the thrombin inhibitor PPACK. PPACK hardly influenced fibrin-binding. Fragment-promoted fibrinogen-binding was also inhibited by a monoclonal antibody recognizing the membrane glycoprotein IIb/IIIa complex known to act as fibrinogen receptor. This antibody failed to influence fragment-stimulated fibrin-binding giving evidence that fibrinogen and fibrin were retained by different receptors. In contrast to 125-I-fibrin its plasmin-derived and 125-I-labelled fragment X was not recognized by the platelets in presence of the fibronectin 30 kD-fragment. Fragment-stimulated binding of 125-I-fibrin showed a lag phase and was completely inhibited by 0.25 mM putrescine as well as by 1 mM EDTA or 0.1 mM N-ethylmaleinimide. Evidently, a cell-attached transamidase was involved in fibrin-binding possibly by forming a ternary complex with fibrin and the fibronectin fragment.     

Treatment of uremic anemia with recombinant erythropoietin also reduces the defects in platelet adhesion and aggregation caused by uremic plasma.

In the present study, uremic patients on chronic maintenance hemodialysis were treated with recombinant erythropoietin. Before and after 20 weeks of treatment, platelet adhesion and aggregation were studied with perfusions over a sprayed collagen surface and over matrix of cultured endothelial cells with high tissue factor activity. The influence of the erythropoietin induced raise in hematocrit on platelet transport and adhesion was excluded by performing the perfusions at a standard red blood cell concentration. The present study clearly demonstrates that erythropoietin treatment improves platelet adhesion and aggregation in addition to and independent of its effect on the hematocrit. Studies with control platelets resuspended in plasma of untreated patients showed that a uremic plasma factor reduced adhesion and thrombin- and collagen-dependent aggregation. Patient platelets resuspended in control plasma showed no defects. After erythropoietin treatment, the plasma-induced inhibition of adhesion and aggregation had almost completely disappeared from patient plasma. The beneficial effect of the erythropoietin treatment on uremic hemostasis is therefore twofold. The increase of the red blood cell mass improves transport of platelets, and thus adhesion to the vessel wall. The intrinsic defect due to the presence of an inhibitory toxin in uremic plasma is, in large part, corrected. Improved neutralization of uremic toxins by red blood cells or less production of toxins by better oxygenated tissue might play a role in the observed phenomena.     

Comparison between commercial heparin, low molecular weight heparin and pentosan polysulfate on hemostasis and platelets in vivo

The effect of commercial heparin (MW 15000; 60 U/kg), low molecular weight heparin (MW 4500, 60 U/kg), sodium pentosan polysulfate (PZ-68) in doses of 0.5, 1, 2 and 5 mg/kg and 1 mg/kg combined with dextran 70 (0.5 g/kg) on microvascular hemostasis and platelet activity in vivo was investigated in the microcirculation of rabbit mesentery and ear chamber, respectively. Low molecular weight heparin did not alter the hemostasis: nonetheless, commercial heparin and PZ-68 interfered with the hemostatic mechanism by decreasing the stability of the hemostatic plug. This abnormality was more apparent in the venular than in the arteriolar system. Both heparins exhibited a trend towards an increase in platelet reactivity and the reverse was observed following the administration of PZ-68.     

A thiocholine ester of cinnamic acid inhibits crosslinking of fibrin without specific binding to donor lysines.

In the presence of activated factor XIII, 2-diethylbenzyl-aminoethylthiol-14C-transcinnamate bromide completely inhibited the crosslinking of fibrin. However, all three fibrin chains bound the cinnamic acid, and, in the alpha- and gamma-chains, the binding of label was not restricted to the crosslinking donor sites as might be expected. Furthermore, even in the absence of activated factor XIII, fibrinogen and fibrin incorporated cinnamic acid. Thus, as well as reacting with the functional thiol group of factor XIII, the thiocholine ester of cinnamic acid is incorporated non-specifically throughout the fibrinogen and fibrin subunit chains. The thiocholine ester used differs in this respect from dansyl cadaverine which is incorporated enzymatically and exclusively to the (acceptor) sites involved in crosslinking. Thiocholine ester of cinnamic acid cannot be used as a label for localization of specific crosslinking donor sites.     

Platelet factor XIII increases the fibrinolytic resistance of platelet-rich clots by accelerating the crosslinking of alpha 2-antiplasmin to fibrin.

Platelet clots resist fibrinolysis by plasminogen activators. We hypothesized that platelet factor XIII may enhance the fibrinolytic resistance of platelet-rich clots by catalyzing the crosslinking of alpha 2-antiplasmin (alpha 2AP) to fibrin. Analysis of plasma clot structure by polyacrylamide gel electrophoresis and immunoblotting revealed accelerated alpha 2AP-fibrin crosslinking in platelet-rich compared with platelet-depleted plasma clots. A similar study of clots formed with purified fibrinogen (depleted of factor XIII activity), isolated platelets, and specific factor XIII inhibitors indicated that this accelerated crosslinking was due to the catalytic activity of platelet factor XIII. Moreover, when washed platelets were aggregated by thrombin, there was evidence of platelet factor XIII-mediated crosslinking between platelet alpha 2AP and platelet fibrin(ogen). Specific inhibition (by a monoclonal antibody) of the alpha 2AP associated with washed platelet aggregates accelerated the fibrinolysis of the platelet aggregate. Thus in platelet-rich plasma clots, and in thrombin-induced platelet aggregates, platelet factor XIII actively formed alpha 2AP-fibrin crosslinks, which appeared to enhance the resistance of platelet-rich clots to fibrinolysis.     

Inhibition of platelet adhesion to fibrin(ogen) in flowing whole blood by Arg-Gly-Asp and fibrinogen gamma-chain carboxy terminal peptides.

We have employed synthetic peptides with sequences corresponding to the integrin receptor-recognition regions of fibrinogen as inhibitors of platelet aggregation and adhesion to fibrinogen- and fibrin-coated surfaces in flowing whole blood, using a rectangular perfusion chamber at wall shear rates of 300 s-1 and 1,300 s-1. D-RGDW caused substantial inhibition of platelet aggregation and adhesion to fibrinogen and fibrin at both shear rates, although it was least effective at blocking platelet adhesion to fibrin at 300 s-1. RGDS was a weaker inhibitor, and produced a biphasic dose-response curve; SDRG was inactive. HHLGGAKQAGDV partially inhibited platelet aggregation and adhesion to fibrin(ogen) at both shear rates. These results support the identification of an RGD-specific receptor, most likely the platelet integrin glycoprotein IIb:IIIa, as the primary receptor responsible for platelet:fibrin(ogen) adhesive interactions under flow conditions, and indicate that platelet adhesion to surface bound fibrin(ogen) is stabilized by multivalent receptor-ligand contacts.