Increased resistance to plasmic degradation of fibrin with highly crosslinked alpha-polymer chains formed at high factor XIII concentrations

We have previously demonstrated that increasing factor XIII concentrations above that present in plasma (1 U/mL) results in the formation of very high molecular weight alpha fate polyacrylamide and agarose gel electrophoresis (SDS-PAGE). In this report, we have examined the effect of such crosslinking on plasmic susceptibility of fibrin prepared from purified fibrinogen and from plasma in the presence of factor XIII concentrations between 0 and 10 U/mL. The crosslinking achieved with purified fibrinogen at 1 U/mL factor XIII increased resistance to plasmic degradation by 32% as measured in a radiolabeled clot lysis system. However, further increases in plasmic resistance occurred at factor XIII concentrations of 2 and 10 U/mL, the latter decreasing the lysis rate to 45% of that which occurred in the absence of factor XIII. To achieve the same rate of clot lysis with fibrin formed using 10 U/mL rather than 1 U/mL of factor XIII, an increase in plasmin concentration of up to 4.2-fold was required. Similar results were obtained using clots prepared from plasma in the presence of factor XIII concentrations greater than 1 U/mL. Since the alpha 2-plasmin inhibitor content was the same for fibrin at 1 or 10 U/mL factor XIII, the increasing plasmic resistance could not be attributed to increased binding of the inhibitor. We conclude that fibrin prepared in the presence of factor XIII at concentrations exceeding that in plasma shows increased resistance to plasmic degradation, which is likely explained by the formation of very high molecular weight alpha polymer chains.     

Specific identification of fibrin polymers, fibrinogen degradation products, and crosslinked fibrin degradation products in plasma and serum with a new sensitive technique

A new method is described for identifying low concentrations of circulating derivatives of fibrinogen and fibrin, even when present in heterogeneous mixtures. This technique is applicable to plasma and serum and uses electrophoresis in 2% agarose in the presence of sodium dodecyl sulfate (SDS) followed by immunological identification of separated derivatives, using radiolabeled antifibrinogen antiserum and autoradiography. Unique electrophoretic patterns distinguish plasmic derivatives of crosslinked fibrin from those of fibrinogen and also identify crosslinked fibrin polymers produced by the combined action of thrombin and factor XIII on fibrinogen. The assay is sensitive to a concentration of 0.1 micrograms/mL of fibrinogen in serum or plasma. Fibrin polymers, plasmic degradation products of fibrinogen, and plasmic degradation products of crosslinked fibrin were detected in the plasma or serum of a patient with disseminated intravascular coagulation. Plasmic derivatives of both fibrinogen and crosslinked fibrin appeared in serum in the course of fibrinolytic therapy for pulmonary embolism, whereas during acute myocardial infarction a marked increase in the proportion of fibrin polymers in plasma was found in comparison with normal controls. Thus, the procedure can distinguish between the simultaneous processes of fibrin polymer formation, fibrinogenolysis, and fibrinolysis, and is sufficiently sensitive to detect relevant quantities of derivatives in pathologic conditions.     

Monoclonal antibodies to crosslinked fibrin degradation products (XL-FDP) II. EVALUATION IN A VARIETY OF CLINICAL CONDITIONS

Plasmas from patients with a wide variety of thrombotic and presumed prethrombotic conditions were examined for high molecular weight crosslinked fibrin degradation products (known as X-oligomers) using a two-site enzyme-linked immunospecific assay (ELISA). This assay employed a catcher-tag principle using two monoclonal antibodies (mabs) directed towards different epitopes on the complex X-oligomer fraction. In general, thrombotic events (pulmonary embolism, PE, myocardial infarction, MI, peripheral vascular disease, PVD, and disseminated intravascular coagulation, DIC) were accompanied by elevated levels of X-oligomers in the plasma. During pregnancy the value of X-oligomer assays was demonstrated to be a clear-cut marker for pre-eclampsia. Patients following a variety of forms of surgery present with heterogeneous plasma levels of X-oligomers and this may merely reflect the formation and lysis of the fibrin formed during and after surgery. The possible value of this ELISA procedure in monitoring thrombolytic therapy is discussed with a critical analysis of the data presented herein. While the assay of X-oligomer was demonstrated to be a valuable marker of fibrinolysis in plasma, more extensive data are required in order to assess whether such an assay is of diagnostic value in thrombosis-related conditions.     

Monoclonal antibodies to crosslinked fibrin degradation products (XL-FDP) I. CHARACTERIZATION AND PRELIMINARY EVALUATION IN PLASMA

Monoclonal antibodies (mabs) were raised against X-oligomers, the earliest soluble fragments released from crosslinked fibrin (XL-FN), by the action of plasmin. Two of the mabs (NIBn 52 and NIBn 123) were monospecific for X-oligomers in that they showed no binding to fibrinogen, the plasmic fragments of fibrinogen (D and E) and non-crosslinked fibrin (X, Y, D and E), or the terminal digestion product of XL-FN, fragment DD-E. One other mab (NIBn 178) was panspecific for X-oligomers in that it exhibited a weak affinity for fibrinogen. The mabs were used to develop a two-site immunoradiometric assay (IRMA) and an enzyme-linked immunospecific assay (ELISA) which permitted the specific measurement of X-oligomers directly in plasma, rather than in serum. This immunoassay is a true assay of fibrinolysis as distinct from fibrinogenolysis and may be a potential aid in the diagnosis and evaluation of thrombosis. In preliminary studies, the assay detected low levels of X-oligomers in normal plasma and elevated levels in patients with disseminated intravascular coagulation.     

Serum crosslinked fibrin (XDP) and fibrinogen/fibrin degradation products (FDP) in disorders associated with activation of the coagulation or fibrinolytic systems

Soluble crosslinked fibrin derivatives (XDP) in serum were determined by enzyme immunoassay utilizing monoclonal antibodies and compared with serum fibrinogen/fibrin degradation products (FDP) assayed by conventional techniques. In healthy subjects and patients with miscellaneous disorders not usually associated with activation of the haemostasis mechanism, mean XDP levels were 45 and 70 ng/ml respectively. However, elevated levels of XDP occurred in conditions commonly associated with intravascular and possibly extravascular activation of the coagulation system. Markedly raised mean XDP values (677-6900 ng/ml) occurred in treated pulmonary embolism, disseminated neoplasia, severe inflammatory disorders and complicated postoperative states, and lesser but significant elevation (mean 150-400 ng/ml) in treated venous thrombosis, uneventful postsurgical states, localized neoplasia, liver disease and symptomatic arterial disease. Levels during initial streptokinase therapy (mean 24 000 ng/ml) fell tenfold as treatment was continued. The degree of XDP elevation over normal values was significantly higher than that of FDP in conditions with a propensity for venous thrombosis (post-operative states, disseminated neoplasia and inflammatory diseases) than in liver disease, localized neoplasia or patients receiving heparin therapy for venous thromboembolism.     

Plasmic degradation of crosslinked fibrin. I. Structural analysis of the particulate clot and identification of new macromolecular-soluble complexes

Plasmic degradation of crosslinked fibrin has been studied to identify the proteolytic cleavages that convert the clot into a soluble lysate and also to identify the derivatives that are likely to circulate during clot dissolution. Initial polypeptide chain cleavages do not disrupt the solid clot matrix. With continued exposure to plasmin, high molecular weight derivatives are produced that remain attached to the clot by noncovalent forces. Further degradation then results in the liberation into solution of several large, noncovalently bound complexes. Progressive degradation of the largest, initially liberated complexes to the terminal derivatives, DD/E, DD, and E, occurs in solution after their release from the clot. As the fibrin clot is exposed to plasmin for longer intervals, progressive dissolution occurs, but the structure of the covalently bound insoluble fibrin core, the noncovalently attached derivatives, and the liberated complexes remains constant. Since much of the initially liberated protein is in complexes larger than DD/E, these derivatives probably represent the more prevalent plasmic degradation products of crosslinked fibrin in vivo.     

Reaction of plasmic degradation products of fibrinogen in the radioimmunoassay of human fibrinopeptide A.

A radioimmunoassay (RIA) technique has been devised for the measurement of human fibrinopeptide A (FPA). The system utilizes rabbit antiserum to native human FPA and a synthetic fibrinopeptide, with tyrosine substituted for phenylanine in amino acid position 8. The test detects native human FPA at a concentration of 0.1 ng/ml, but does not cross react with human fibrinopeptide B or with fibrinopeptides A from canine, porcine, or bovine fibrinogen. Fibrinogen and chemical or plasmic degradation products with 2 moled of FPA per mole react fully in this test system. This includes the large-molecular-weight intermediate fragments X and Y and the NH2-terminal disulfide knot, and indicates that this antibody recognizes and reacts with FPA in the presence of the contiguous peptide structures present in fibrinogen. Fragment E, which is derived from the NH-2-terminal portion of fibrinogen, loses most of its FPA content after its liberation from its precursor derivative and reacts to a lesser extent in the RIA than do fragments X and Y. This correlated with the recovery of FPA-positive material from ultrafilitrates of extensive but not partial plasmic digests of fibrinogen. Although FPA immunoreactivity liberated from fibrinogen does not necessarily reflect thrombin activity and/or fibrin formation, only extensive plasmic degradation yields peptide material which reacts in this RIA system. This should not be a serious limitation to the application of the RIA in the detection of venous thrombosis.     

Composition of fibrin glues significantly influences axial vascularization and degradation in isolation chamber model

In this study, different fibrin sealants with varying concentrations of the fibrin components were evaluated in terms of matrix degradation and vascularization in the arteriovenous loop (AVL) model of the rat. An AVL was placed in a Teflon isolation chamber filled with 500 μl fibrin gel. The matrix was composed of commercially available fibrin gels, namely Beriplast (Behring GmbH, Marburg, Germany) (group A), Evicel (Omrix Biopharmaceuticals S.A., Somerville, New Jersey, USA) (group B), Tisseel VH S/D (Baxter, Vienna, Austria) with a thrombin concentration of 4?IU/ml and a fibrinogen concentration of 80 mg/ml [Tisseel S F80 (Baxter), group C] and with an fibrinogen concentration of 20 mg/ml [Tisseel S F20 (Baxter), group D]. After 2 and 4 weeks, five constructs per group and time point were investigated using micro-computed tomography, and histological and morphometrical analysis techniques. The aprotinin, factor XIII and thrombin concentration did not affect the degree of clot degradation. An inverse relationship was found between fibrin matrix degradation and sprouting of blood vessels. By reducing the fibrinogen concentration in group D, a significantly decreased construct weight and an increased generation of vascularized connective tissue were detected. There was an inverse relationship between matrix degradation and vascularization detectable. Fibrinogen as the major matrix component showed a significant impact on the matrix properties. Alteration of fibrin gel properties might optimize formation of blood vessels.     

A fibrin based model for rheumatoid synovitis

Intracavitary fibrin clots may initiate pannus formation and the immunopathology of RA. Two critical steps, probably host dependent, may determine the development of RA: an altered regulation of extravascular haemostasis or an aberrant reactivity of synovial fibroblasts to the adhered fibrin clots. Current treatments for RA target events downstream of fibrin deposition, perhaps agents acting at an earlier stage should be tried.     

Plasma cross-linked fibrin degradation product (XLFbDP) assays in an in vivo model of fibrinolysis

Assumptions regarding the elaboration of plasma cross-linked fibrin degradation products (XLFbDPs) in vivo were tested using an experimental model in which particulate human fibrin was infused into rabbits and the products of lysis monitored with an immunoassay utilizing DD-3B6/22, a monoclonal antibody to human cross-linked derivatives. XLFbDPs were generated following the infusion of a suspension of cross-linked fibrin, attaining a peak between 40 and 60 min, then falling at a rate approximating a plasma half-life of 2 h. The major in vivo products of lysis of cross-linked fibrin, identified by SDS-PAGE of immunoextracted plasma, were D-dimer and high-molecular-weight moieties. Peak levels of XLFbDPs achieved correlated with the amount of fibrin administered. Since XLFbDP levels were no higher when fibrin infusion was followed by infusions of streptokinase and human plasminogen, it is concluded that endogenous mechanisms of lysis were already maximally stimulated. Infusions of non-cross-linked (NXL) fibrin or of fibrinogen led to much smaller, but measurable, rises in XLFbDP. In the latter group, XLFbDP levels rose further following fibrinolytic therapy. Treatment with epsilon aminocaproic acid (EACA) caused partial (greater than 50%) inhibition of lysis while pre-treatment with nitrogen mustard, inducing leucopenia, virtually abolished the appearance of XLFbDPs in the circulation. This implies that fibrinolytic responses are substantially dependent upon cellular functions sensitive to nitrogen mustard.     

Multimer size dependence of von Willebrand factor binding to crosslinked or noncrosslinked fibrin

von Willebrand factor (vWF) is synthesized in endothelial cells (EC) and may be either secreted constitutively or stored in Weibel-Palade bodies (WPB) for regulated release. Because fibrin stimulates rapid vWF release from EC, we examined the binding of EC synthesized vWF to fibrin. Culture medium containing constitutively secreted vWF was removed from metabolically labeled primary cultures of human umbilical vein EC, and vWF released from WPB was obtained after stimulation by A23187. vWF-deficient fibrinogen with or without factor XIII was added to releasate or media and clotted with thrombin to form crosslinked or noncrosslinked fibrin. vWF was immunopurified from releasate or media before and after clotting, and the amount and multimeric pattern of vWF bound was determined after sodium dodecyl sulfate agarose gel electrophoresis. High molecular weight multimers of vWF, whether secreted constitutively or released from WPB, bound preferentially to fibrin. Multimers of greater than 20 subunits represented 60% +/- 4% (SEM) of A23187 released vWF and 11% +/- 5% of media vWF, but binding to fibrin was similar, 96% +/- 1% and 94% +/- 2%, respectively. A progressively smaller proportion of vWF bound as multimer size decreased, and dimeric vWF binding was least, with 34% +/- 5% binding from A23187 releasate and 51% +/- 4% from media. The amount of vWF binding to crosslinked or noncrosslinked fibrin was similar, and preferential binding of high molecular weight multimers occurred with both. As measured by enzyme-linked immunosorbent assay, 45% +/- 2% of constitutively secreted vWF bound to crosslinked fibrin and 50% +/- 2% to noncrosslinked fibrin. The propolypeptide of vWF did not bind to fibrin. These findings indicate that binding of EC secreted vWF binding to fibrin depends on multimeric size but not on factor XIII crosslinking. This suggests that vWF released from EC in the presence of fibrin will bind locally, thereby facilitating platelet adhesion to the hemostatic plug or thrombus.     

Identification of covalently linked trimeric and tetrameric D domains in crosslinked fibrin.

Following proteolytic conversion of fibrinogen to fibrin, clot assembly commences with formation of double-stranded fibrils that subsequently branch extensively in forming a three-dimensional network. Plasmin digests of fibrin clots that had first been covalently crosslinked by plasma transglutaminase (factor XIIIa) contained multimeric proteolytic fragments composed of crosslinked outer (D) domains of neighboring fibrin molecules. Two of these were larger than the well-known "D dimer" fragment and corresponded to D trimers and D tetramers, respectively. Whereas D dimers originate from crosslinked D domains at bimolecular junctions within two-stranded fibrils, D trimers and D tetramers evidently arise through crosslinking of contiguous D domains at trimolecular and tetramolecular junctions or at fibril branch points, respectively. Measurement of the widths of fibrils comprising trifunctional branches in thin fiber networks revealed tetramolecular branch points, which are formed by bifurcation of two double-stranded fibrils. In addition, another type of trifunctional structure, which we term the trimolecular branch point, was composed of three double-stranded fibrils. Crosslinking of D domains to form trimers may occur at this type of junction. These findings add to our understanding of the crosslinking arrangements that stabilize fibrin clot structure and the ways that fibrin molecules polymerize to form branches in the clot matrix.     

Fibrin degradation products, fibrin monomer and soluble fibrin in disseminated intravascular coagulation.

Disseminated intravascular coagulation (DIC) is characterized by activation of hemostasis and fibrinolysis resulting in the formation of thrombin and plasmin, and the characteristic effects of these enzymes on plasma fibrinogen can be useful in diagnosis. Thrombin cleaves fibrinopeptides from fibrinogen, forming fibrin monomer that rapidly polymerizes to form a clot. Small amounts can circulate in plasma as "soluble fibrin," which may have a complex composition and include fibrinogen and a variable amount of cross-linking. Plasmic degradation of cross-linked fibrin forms a heterogeneous group of degradation products reactive in assays for D-dimer, and their levels provide a measure of the amount of fibrin formation and lysis. Caution should be exercised in comparing quantitative results using different assays because of problems with standardization and variable reactivity with different molecular forms. Marked elevations of fibrin(ogen) degradation products are a constant finding in experimental animal models of DIC. In human models of DIC resulting from endotoxin infusion, D-dimer is elevated early and high levels persist, reflecting lysis of microvascular fibrin deposits. Elevated levels of D-dimer and soluble fibrin are very sensitive for the diagnosis of DIC, and a normal level has a high negative predictive value. Serial monitoring of soluble fibrin or D-dimer assays may be of value in evaluating the response to therapy and possibly in identifying at-risk patients.     

Demonstration of in situ fibrin degradation in pathologic thrombi

Fibrin prepared from 15 pathologic thrombi was examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis to determine the extent and pattern of fibrinolysis that occurs in vivo. Two groups of patients could be distinguished on the basis of the polypeptide chain composition of fibrin in their thrombi. Those patients who presented with acute vascular obstruction, either arterial or venous, showed a minimal degree of fibrin degradation, with a dominance of intact, undegraded crosslinked gamma-gamma dimers. On the other hand, patients with long-standing symptoms associated with chronic aortic aneurysms had thrombi containing extensively degraded fibrin. Thrombi in large aortic aneurysms were dissected into concentric layers that showed different degrees of fibrinolysis. The luminal surface consisted of fresh, red thrombus and contained undegraded crosslinked fibrin similar to that found in patients with acute occlusive disease. Deeper layers of the thrombus showed gamma-gamma chain degradation throughout, indicating that this portion was undergoing active thrombolysis. The findings demonstrate that the variability in the pathophysiologic balance between coagulation and fibrinolysis is reflected in vivo by the polypeptide chain composition of crosslinked fibrin in thrombi. The results support the hypothesis of a dynamic equilibrium between clotting and lysis, but indicate that the balance between these two processes may be distinctly different in separate areas of a single clot.