Fibrinogen: evolution of the structure-function concept. Keynote address at fibrinogen 2000 congress.

Coagulation of blood is such an evident phenomenon that its observation can be traced back to earliest historical times. The great philosophers and physicians of antiquity discussed and provided interesting explanations. However, it was not until the end of the seventeenth century that the structural component of the blood clot was described by Malpighi as a white fibrous substance. In the middle of the nineteenth century this was identified as a constituent of pathological thrombi and given the name fibrin. At about that time its precursor in blood, fibrinogen, was isolated in a highly purified form by Hammarsten who suggested that, preceding fibrin formation, activation of fibrinogen by thrombin occurred by limited proteolysis. The activation mechanism was eventually clarified in the 1950s. It was shown to proceed in two discrete steps, by removal of low molecular weight activation peptides. Ferry postulated, based on physicochemical observations, that the activated molecules aligned in a half-staggered fashion to form polymers. The rapid post-war development of biochemical technology permitted evaluation of the primary structure of fibrinogen. With that followed identification of molecular domains in the activated firbinogen molecules that participate in polymer formation, crosslinking of polymeric structures, and domains for cellular attachment. Crystallization of fragments and, recently, of the entire molecule has confirmed and extended this knowledge. Lately, it has also been possible to obtain detailed information on the architecture of the fiber network in the fibrin gel. The gel structure is primarily determined by the initial rate of fibrinogen activation, but without infringement of this primary rule, several factors in blood may modulate the structure. Fibrinogen and fibrin play important roles in normal hemostasis, wound healing, and pathological processes, such as thrombosis and atherosclerosis.

Characterization of a fibrin glue-GDNF slow-release preparation.

One novel method to deliver trophic factor locally in the CNS is to mix it into fibrin glue. In the present studies, [125I]-labeled GDNF-containing fibrin glue balls were used to determine binding and spread of the trophic factor. First, the binding of different concentrations of [125I]-labeled GDNF in fibrin glue was determined in vitro. Within the six concentrations used (from 200 nM to 0.004 nM, 0 M as control), there was a strong linear correlation between the [125I]-GDNF concentration and the recovered radioactivity (r = 0.992). The mean bound radioactivity in 16 samples with 4 nM [125I]-GDNF was 71262 +/- 2710 CPM, and accounted for 89.8% of the mean initial count of free [125I]-GDNF (79369 +/- 3499 CPM). Second, [125I]-GDNF-containing glue balls were implanted into the anterior chamber of adult rats. The implanted fibrin glue balls decreased in size with time, but could still be identified on the irises 2 wk after implantation. Radioactivity was concentrated at the implantation sites in the early stages with a distribution in the surrounding iris tissue, which became separated into focal radioactive spots at the third week. Counts of radioactivity were significantly higher in the [125I]-GDNF glue ball-implanted irises than controls until 14 days after implantation. A study of the [125I] decay over time using least-squares linear regression demonstrated first-order kinetics (r = -0.98, p <0.02) with k = 0.0091 and T 1/2 = 76 h. Finally, [125I]-GDNF-containing glue balls were implanted in the spinal cord of adult rats. Radioactivity was concentrated at the implantation sites in the early stages and was later distributed more widely in the surrounding thoracic cord. The [125I]-GDNF-containing glue degraded over time and became a porous meshwork with decreasing radioactivity at the later time points. Radioactivity in the spinal cords subjected to implantation of [125I]-GDNF-containing glue balls was higher than in controls for 14 days. Study of the [125I] decay by time with least-squares linear regression demonstrated first-order kinetics (r = -0.97, p = 0.001) with T 1/2 = 75.6 h. We conclude that the trophic factor GDNF becomes bound in the fibrin glue matrix from which it is gradually released. Our results suggest that fibrin glue is an effective substrate for keeping a trophic factor localized in situ for a finite period, protected from the circulation, surrounding aqueous humor or CSF.

Application of thrombin based fibrin glue and non-thrombin based batroxobin glue on intact human blood vessels: evidence for transmural thrombin activity.

An alternative method of uniting small diameter vessels to obtain tissue union while limiting the thrombogenic effect of suture placement at a vessel anastomosis involves the use of a thrombin based fibrin glue as a surgical sealant. This investigation addresses whether the in vitro application of a thrombin based glue (TG), or batroxobin glue (BG), a non-thrombin based glue made with the snake venom enzyme batroxobin, alters intravascular platelet deposition (PD) or cleaves blood fibrinogen, as measured by fibrinopeptide A (FPA) production, when the respective glue is applied to the external surface of an intact human placental artery or an artery with an anastomosis. When TG was applied to the adventitial surface of an intact vessel or an anastomosis (n = 7) of control and experimental vessels, there was a significant increase in intraluminal platelet deposition, an effect not realized with BG (n = 12, intact vessel TG p = 0.01, BG p = 0.66, anastomosis TG p <0.01, BG p <0.01). Both TG and BG significantly increased FPA levels when human whole blood was perfused through both intact vessels or vessels containing an anastomosis when compared to control vessels (intact vessel TG and BG p <0.01, anastomosis TG and BG p <0.01). Labelled thrombin studies document the rapid passage of thrombin through an intact vessel wall or vessels with an anastomosis when TG was applied to the adventitial surface of the vessel. The data suggest that TG and BG are drug delivery systems for their respective enzymes that either pass through or transfer a message across not only a surgically created anastomosis, but also an intact vessel wall.

On the occurrence of thrombin-like enzymes in mosquitoes.

A thrombin-like enzyme has been identified in mosquitoes and partially purified. The enzyme preparation displays on SDS gel electrophoresis a major protein band of about 22 kDa and one minor of about 28 kDa. The enzyme preparation cleaves a synthetic substrate (S-2238) for thrombin with Km 113 mumol/L (as compared to 3 mumol/L for thrombin). The enzyme clots fibrinogen and plasma. During activation of fibrinogen, fibrinopeptide A (but scarcely fibrinopeptide B) is released. The enzyme is not inhibited by hirudin and activates (if at all) factor XIII differently from thrombin. Predominantly gamma-dimers are formed in the cross linking process. As compared to thrombin a larger extent of activation is required to induce gelation (clotting) by the mosquito enzyme. At a given clotting time the enzyme produces tighter gel structures than thrombin. In its action the enzyme resembles the snake venom enzyme, batroxobin.

Fibrinogenemia Tampere--a dysfibrinogenemia with defective gelation and thromboembolic disease.

Fibrinogen Tampere was found in a woman with severe thromboembolic disease. The thrombin induced clotting time of her plasma and purified fibrinogen was slightly prolonged. The activation of fibrinogen Tampere appeared to be normal but subsequent gelation was defective. We studied fibrin gels formed at different ionic strengths and at different fibrinogen and calcium concentrations by liquid permeation, turbidity, and 3D laser microscopy. Crosslinking was studied by SDS-gel electrophoresis. The gels formed from fibrinogen Tampere were at ionic strength above 0.2 much tighter and had lower fiber mass-length ratios than normal gels as judged by permeability and turbidity data. At ionic strength 0.15 and at different calcium concentrations analysis by permeability showed the same results for fibrinogen Tampere as for normal gels. Analysis by turbidity at ionic strength 0.15 suggested swelling of the fibers at low calcium concentrations. 3D microscopy revealed perturbed clot architecture under all conditions. In fibrin gels from fibrinogen Tampere, the gamma-chain crosslinking was normal but the crosslinking of alpha-chains was delayed at ionic strength 0.2 and also at lower ionic strengths on lowering the calcium concentration. The abnormal gelation may be due to a mutation in the fibrinogen molecule. Tendency to form tight fibrin gels and/or insufficient crosslinked fibrin matrix may be pathogenetic in this thrombotic disease.

Flow and antibody binding properties of hydrated fibrins prepared from plasma, platelet rich plasma and whole blood.

Previous studies, using cross-linked fibrin prepared from purified fibrinogen, showed low binding of a fibrin-specific monoclonal antibody designated T2G1 (Procyk et al., Blood 77:1469-75, 1991). In this study we investigated the binding of T2G1 and one other antibody to clots prepared from platelet poor plasma (PPP), platelet rich plasma (PRP) and whole blood. In contrast to our previous study, we used unlabelled antibodies and quantitated the level bound by ELISA, measuring antibody concentration in the non-adsorbed fraction. Antibody T2G1 bound 1.35 +/- 0.10 pmol/pmol fibrin (n = 11) to whole blood columns, 1.64 +/- 0.18 (n = 10) to PRP columns and 1.58 +/- 0.13 (n = 8) to PPP columns. The binding of T2G1 to columns made from purified fibrinogen was 0.78 +/- 0.05 pmol/pmol fibrin (n = 15). An antibody to a conformation-dependent epitope on Fragment D (Fd4-7B3) bound in comparable amounts to the different fibrins. Flow data show that whole blood columns, and also, but to a lesser extent those made with plasma, had a higher flow rate, permeability and fiber mass-length ratio than columns prepared from fibrinogen indicating a more coarse fibrin network. These data show that the presence of other proteins and blood cells, similar to what might occur in vivo, not only lead to an increase in the permeability of gels but also allow for better exposure of some epitopes.

Fibrin in human plasma: gel architectures governed by rate and nature of fibrinogen activation.

The porosity, fiber dimension and architecture of fibrin gels formed in recalcified plasma on addition of thrombin are, within a certain range of thrombin concentrations, determined by the initial rate of fibrinogen activation. Furthermore, the initial network formed in this range creates the scaffold into which subsequently activated fibrinogen molecules are deposited. Change in thrombin concentration that occurs during gelation, as a result of indigenous thrombin generation in plasma, does not qualitatively alter this scaffold. The formation of the networks obeys a more complex rule when low amounts of thrombin are added or with recalcified plasma without added thrombin. These networks are tighter than would be expected from the initial rate of fibrinogen activation. In this case an extremely porous network is probably formed initially, followed by formation of a secondary, superimposed network of a less porous architectural quality. The latter structure appears to be governed by the rate of indigenous generation in plasma of thrombin-like enzymes in combination with the particular type of fibrinmonomers being produced. In addition our findings establish the rules for proper determination of gel structures in clinical plasma samples. The sequelae of a variety of clot structures that may be formed in vivo are discussed.

The effect of plasma antithrombin concentration on thrombin generation and fibrin gel structure.

Congenital deficiency of antithrombin (AT) is associated with thrombotic events and AT consumption occurs in some severe disorders and after treatment with heparin. The aim of this study was to investigate whether variations in the level of plasma AT modify thrombin generation and the fibrin formation process after the intrinsic coagulation mechanism is triggered. Normal plasma was depleted of AT by immunoadsorption on CNBr-Sepharose coupled with the anti-AT-IgG fraction of antiserum. The AT-depleted plasma was reconstituted with AT (between 0.3 and 1.5 AT units per ml). Thrombin generation was measured as the development of thrombin-antithrombin complexes (TAT). The lag phase preceding fibrin formation depended on the concentration of AT. The short lag phase was seen in completely AT-depleted plasma and the long in plasma with 1.5 AT units per ml. TAT generation, determined in parallel consecutive samples, showed that the rate at which thrombin was generated was inverse to the AT concentration in plasma. The network structure of hydrated fibrin gels in the clotted plasma was studied by measuring the wavelength dependence of gel turbidity. The mass/length ratio value, -i.e. the thickness of fiber strands and porosity of the gel increased with increasing AT concentrations. It is concluded that plasma AT regulates the rate of prothrombin-thrombin conversion, the clotting time and the consequently network structure of the fibrin gel.

Fibrin gel network characteristics and coronary heart disease: relations to plasma fibrinogen concentration, acute phase protein, serum lipoproteins and coronary atherosclerosis.

The native fibrin gel structure formed in vitro from plasma samples was examined by liquid permeation of the hydrated fibrin gel networks in 18 men who had suffered a myocardial infarction before the age of 45 years and in 20 control subjects. Patients with an elevated plasma fibrinogen concentration had a considerably lower fibrin gel porosity (permeability coefficient, Ks) compared with patients with a normal plasma fibrinogen level and with controls. The calculated fiber mass-length ratio of the fibrin gel networks was decreased in both patient groups. Gel porosity differed markedly between individuals at a given plasma fibrinogen concentration. Fairly strong inverse correlations were found between plasma orosomucoid level on the one hand and Ks (r = -0.617, p less than 0.01) or fiber mass-length ratio (r = -0.499, p less than 0.05) on the other. The low density lipoprotein (LDL) cholesterol concentration also correlated inversely with Ks (r = -0.471, p less than 0.05) and fiber mass-length ratio (r = -0.522, p less than 0.05). Significant inverse relations, which were independent of plasma fibrinogen and lipoprotein concentrations, were detected between Ks (r = -0.519, p less than 0.05) and calculated fiber mass-length ratio (r = -0.723, p less than 0.001) and number and severity of coronary artery stenoses determined by angiography. A proneness to formation of tight, rigid and space-filling fibrin network structures with small pores thus appears to be associated with premature coronary artery disease.

Nonclottable fibrin obtained from partially reduced fibrinogen: characterization and tissue plasminogen activator stimulation.

Out of 29 disulfide bonds in human fibrinogen, 7 were cleaved during limited reduction under nondenaturing conditions in calcium-free buffer: 2 A alpha 442Cys-A alpha 472Cys and 2 gamma 326Cys-gamma 339Cys intrachain disulfide bonds in the carboxy-terminal ends of the A alpha- and gamma-chains and the symmetrical disulfide bonds at gamma 8Cys, gamma 9Cys, and A alpha 28Cys. We studied the loss of thrombin clottability that followed limited reduction and the increase in the susceptibility of the fibrinogen A alpha 19-A alpha 20 bond to hydrolysis by thrombin. Using differential scanning calorimetry, we show that the extent of unfolding and denaturation of specific domains following limited reduction is small. Heat absorption peaks corresponding to the melting of the major regions of compact structure give high calorimetric enthalpies, as in untreated nonreduced fibrinogen, indicating that substantial regions of native structure are still present in partially reduced fibrinogen. Thrombin releases fibrinopeptide A at an identical rate as in nonreduced fibrinogen while fibrinopeptide B release is slower. Sedimentation velocity studies show that thrombin treatment leads to complex formation; however, gelation does not occur. Amino-terminal analysis indicates that the second thrombin cleavage in the A alpha-chain at A alpha 19-A alpha 20 takes place only after fibrinopeptide A release. Thus, the loss of clottability appears to result from perturbation of carboxy-terminal polymerization sites, probably a consequence of gamma 326Cys-gamma 339Cys intrachain disulfide bond cleavage. The thrombin-treated partially reduced fibrinogen remains soluble in buffered saline and fully expresses at least one epitope, B beta 15-21, unique to fibrin. Furthermore, this nonclottable form accelerates the tissue plasminogen activator dependent conversion of plasminogen to plasmin.