A comparison of the mechanical and structural properties of fibrin fibers with other protein fibers.

In the past few years a great deal of progress has been made in studying the mechanical and structural properties of biological protein fibers. Here, we compare and review the stiffness (Young's modulus, E) and breaking strain (also called rupture strain or extensibility, epsilon(max)) of numerous biological protein fibers in light of the recently reported mechanical properties of fibrin fibers. Emphasis is also placed on the structural features and molecular mechanisms that endow biological protein fibers with their respective mechanical properties. Generally, stiff biological protein fibers have a Young's modulus on the order of a few Gigapascal and are not very extensible (epsilon(max) < 20%). They also display a very regular arrangement of their monomeric units. Soft biological protein fibers have a Young's modulus on the order of a few Megapascal and are very extensible (epsilon(max) > 100%). These soft, extensible fibers employ a variety of molecular mechanisms, such as extending amorphous regions or unfolding protein domains, to accommodate large strains. We conclude our review by proposing a novel model of how fibrin fibers might achieve their extremely large extensibility, despite the regular arrangement of the monomeric fibrin units within a fiber. We propose that fibrin fibers accommodate large strains by two major mechanisms: (1) an alpha-helix to beta-strand conversion of the coiled coils; (2) a partial unfolding of the globular C-terminal domain of the gamma-chain.

B:b interactions are essential for polymerization of variant fibrinogens with impaired holes 'a'.

BACKGROUND: Fibrin polymerization is mediated by interactions between knobs 'A' and 'B' exposed by thrombin cleavage, and holes 'a' and 'b' always present in fibrinogen. The role of A:a interactions is well established, but the roles of knob:hole interactions A:b, B:b or B:a remain ambiguous. OBJECTIVES: To determine whether A:b or B:b interactions have a role in thrombin-catalyzed polymerization, we examined a series of fibrinogen variants with substitutions altering holes 'a': gamma364Ala, gamma364His or gamma364Val. METHODS: We examined thrombin- and reptilase-catalyzed fibrinopeptide release by high-performance liquid chromatography, fibrin clot formation by turbidity, fibrin clot structure by scanning electron microscopy (SEM) and factor (F) XIIIa-catalyzed crosslinking by sodium dodecylsulfate polyacrylamide gel electrophoresis. RESULTS: Thrombin-catalyzed fibrinopeptide A release was normal, but fibrinopeptide B release was delayed for all variants. The variant fibrinogens all showed markedly impaired thrombin-catalyzed polymerization; polymerization of gamma364Val and gamma364His were more delayed than gamma364Ala. There was absolutely no polymerization of any variant with reptilase, which exposed only knobs 'A'. SEM showed that the variant clots formed after 24 h had uniform, ordered fibers that were thicker than normal. Polymerization of the variant fibrinogens was inhibited dose-dependently by the addition of either Gly-Pro-Arg-Pro (GPRP) or Gly-His-Arg-Pro (GHRP), peptides that specifically block holes 'a' and 'b', respectively. FXIIIa-catalyzed crosslinking between gamma-chains was markedly delayed for all the variants. CONCLUSION: These results demonstrate that B:b interactions are critical for polymerization of variant fibrinogens with impaired holes 'a'. Based on these data, we propose a model wherein B:b interactions participate in protofibril formation.

Role of 'B-b' knob-hole interactions in fibrin binding to adsorbed fibrinogen.

BACKGROUND: The formation of a fibrin clot is supported by multiple interactions, including those between polymerization knobs 'A' and 'B' exposed by thrombin cleavage and polymerization holes 'a' and 'b' present in fibrinogen and fibrin. Although structural studies have defined the 'A-a' and 'B-b' interactions in part, it has not been possible to measure the affinities of individual knob-hole interactions in the absence of the other interactions occurring in fibrin. OBJECTIVES: We designed experiments to determine the affinities of knob-hole interactions, either 'A-a' alone or 'A-a' and 'B-b' together. METHODS: We used surface plasmon resonance to measure binding between adsorbed fibrinogen and soluble fibrin fragments containing 'A' knobs, desA-NDSK, or both 'A' and 'B' knobs, desAB-NDSK. RESULTS: The desA- and desAB-NDSK fragments bound to fibrinogen with statistically similar K(d)'s of 5.8 +/- 1.1 microm and 3.7 +/- 0.7 microm (P = 0.14), respectively. This binding was specific, as we saw no significant binding of NDSK, which has no exposed knobs. Moreover, the synthetic 'A' knob peptide GPRP and synthetic 'B' knob peptides GHRP and AHRPY, inhibited the binding of desA- and/or desAB-NDSK. CONCLUSIONS: The peptide inhibition findings show both 'A-a' and 'B-b' interactions participate in desAB-NDSK binding to fibrinogen, indicating 'B-b' interactions can occur simultaneously with 'A-a'. Furthermore, 'A-a' interactions are much stronger than 'B-b' because the affinity of desA-NDSK was not markedly different from desAB-NDSK.

Fibrin fibers have extraordinary extensibility and elasticity.

Blood clots perform an essential mechanical task, yet the mechanical behavior of fibrin fibers, which form the structural framework of a clot, is largely unknown. By using combined atomic force-fluorescence microscopy, we determined the elastic limit and extensibility of individual fibers. Fibrin fibers can be strained 180% (2.8-fold extension) without sustaining permanent lengthening, and they can be strained up to 525% (average 330%) before rupturing. This is the largest extensibility observed for protein fibers. The data imply that fibrin monomers must be able to undergo sizeable, reversible structural changes and that deformations in clots can be accommodated by individual fiber stretching.

Visualization and mechanical manipulations of individual fibrin fibers suggest that fiber cross section has fractal dimension 1.3.

We report protocols and techniques to image and mechanically manipulate individual fibrin fibers, which are key structural components of blood clots. Using atomic force microscopy-based lateral force manipulations we determined the rupture force, FR, f fibrin fibers as a function of their diameter, D, in ambient conditions. As expected, the rupture force increases with increasing diameter; however, somewhat unexpectedly, it increases as FR approximately D1.30+/-0.06. Moreover, using a combined atomic force microscopy-fluorescence microscopy instrument, we determined the light intensity, I, of single fibers, that were formed with fluorescently labeled fibrinogen, as a function of their diameter, D. Similar to the force data, we found that the light intensity, and thus the number of molecules per cross section, increases as I approximately D1.25+/-0.11. Based on these findings we propose that fibrin fibers are fractals for which the number of molecules per cross section increases as about D1.3. This implies that the molecule density varies as rhoD approximately D -0.7, i.e., thinner fibers are denser than thicker fibers. Such a model would be consistent with the observation that fibrin fibers consist of 70-80% water and only 20-30% protein, which also suggests that fibrin fibers are very porous.

Fibrin stimulates platelets to increase factor VIIIa binding site expression.

Factor (F)VIII functions as an enzymatic cofactor on the membranes of stimulated platelets. However, thrombin stimulates platelets to express only a small number of binding sites for FVIII. We wished to determine whether molecules that are likely to be present in a developing thrombus stimulate platelets to up-regulate FVIII binding site expression. Flow cytometry was utilized to measure binding of fluorescein-labeled FVIIIa to activated platelets and a FXase assay was utilized to measure platelet-dependent function. Various agonists as well as normal and mutant fibrinogens and fibrin were evaluated as co-stimuli. Thrombin-stimulated platelets expressed 214 +/- 67 binding sites for thrombin-activated FVIII (FVIIIa) and none of the established soluble agonists enhanced binding site exposure. However, the presence of 5 micro g mL(-1) fibrin increased the number of FVIIIa binding sites/platelet three- to eight-fold (1470 +/- 130, range 600-1800) with a parallel increase in platelet-based FXase assay. Binding site up-regulation was not stimulated by fibrinogen and was blocked by inhibitors of GPIIbIIIa. Mutant fibrin lacking the gamma-chain C-terminal four residues was ineffective while fibrin with altered RGD sequences did stimulate expression of FVIIIa binding sites indicating that co-stimulation is mediated by the fibrin gamma-chain termini. Fibrin-enhanced expression of FVIIIa binding sites was not supported by D364H fibrin, which does not aggregate normally, and was blocked by the GPRP peptide, which inhibits fibrin polymerization. Polymerized fibrin can function as a platelet co-stimulus, up-regulating expression of binding sites for FVIIIa.

Tall cell papillary thyroid carcinoma metastatic to femur: evidence for thyroid hormone synthesis within the femur.

A 67-year old man with a 3-month history of left hip pain had a history of Graves disease, treated with 131I 20 years before admission, and papillary thyroid carcinoma, treated with cervical lymphadenopathy 9 years before admission. Removal of a 3.5- x 5-cm mass from the left femur revealed it to be a tall cell variant of papillary thyroid carcinoma. Removal of this mass resulted in his thyrotropin level increasing from 2 (presurgery) to 23 mIU/mL, whereas his thyroxine level simultaneously decreased from 5.79 (presurgery) to 2.29 microg/dL 12 days after surgery despite continuation of levothyroxine of 0.137 mg/day. On histological examination, the tall cell variant in the femur was producing abundant thyroglobulin. This first case of a metastatic papillary thyroid carcinoma in bone producing thyroid hormone to the extent that the patient became hypothyroid after removal of this metastasis illustrates that metastatic thyroid lesion(s) may produce significant amounts of thyroid hormone.

A novel transgenic mouse model of hyperfibrinogenemia.

Hyperfibrinogenemia is a risk predictor in several diseases, including cardiovascular disease. Nevertheless, it remains unknown whether elevated fibrinogen has an etiologic role in or is a reflection of disease pathogenesis, or both. To examine this question, we generated a mouse model of hyperfibrinogenemia. We isolated the mouse fibrinogen locus, containing the three fibrinogen genes, in a single P1 clone. This approximately 100 kb clone was injected into C57Bl/6J zygotes. Three transgenic lines were identified, two with elevated fibrinogen, 1.4- and 1.7-fold relative to normal. We characterized the line with the higher level. Northern blots of total RNA showed transgene expression was liver specific, and the message levels were 2- to 3-fold enhanced. Fibrinogen in transgenic mice was normal in both immunologic and clotting assays. Our data indicate that over-expression of all three fibrinogen genes is necessary to achieve hyperfibrinogenemia. We saw no increase in mortality or morbidity, no gross abnormalities in the organs, and no histologic differences in lung, liver, spleen or kidney, in transgenic mice relative to normal littermates. We conclude that elevated fibrinogen did not cause disease in mice. We anticipate that breeding these mice to other mouse models of disease will demonstrate whether hyperfibrinogenemia has a role in the initiation or progression of symptomatic disease.

The impaired polymerization of fibrinogen Longmont (Bbeta166Arg-->Cys) is not improved by removal of disulfide-linked dimers from a mixture of dimers and cysteine-linked monomers.

This study identified a new substitution in the Bbeta chain of an abnormal fibrinogen, denoted Longmont, where the residue Arg166 was changed to Cys. The variant was discovered in a young woman with an episode of severe hemorrhage at childbirth and a subsequent mild bleeding disorder. The neo-Cys residues were always found to be disulfide-bridged to either an isolated Cys amino acid or to the corresponding Cys residue of another abnormal fibrinogen molecule, forming dimers. Removing the dimeric molecules using gel filtration did not correct the fibrin polymerization defect. Fibrinogen Longmont had normal fibrinopeptide A and B release and a functional polymerization site "a." Thus, the sites "A" and "a" can interact to form protofibrils, as evidenced by dynamic light-scattering measurements. These protofibrils, however, were unable to associate in the normal manner of lateral aggregation, leading to abnormal clot formation, as shown by an impaired increase in turbidity. Therefore, it is concluded that the substitution of Arg166-->Cys-Cys alters fibrinogen Longmont polymerization by disrupting interactions that are critical for normal lateral association of protofibrils. (Blood. 2001;98:661-666)

Insight from studies with recombinant fibrinogens.

Using a two-step cloning strategy, we have synthesized more than 20 variant human fibrinogens for biochemical studies. In preliminary experiments we showed that normal fibrinogen produced in CHO cells serves as an accurate model for plasma fibrinogen. We focus here on those variants whose characterization has provided insight into the mechanism of thrombin-catalyzed polymerization. Analysis of N-terminal variants showed that thrombin specificity dictates the ordered release of fibrinopeptides. Nevertheless, analysis of C-terminal variants indicated that fibrinopeptide B (FpB) release is dependent on polymerization. Changes in the a polymerization site and the high-affinity calcium-binding site were associated with a complete loss of polymerization. These experiments showed that alterations in the calcium-binding site influenced function of the a site; in contrast, alterations in the a site did not alter calcium binding. Analysis of variants in the N-terminus of the B beta chain provided the first direct evidence that this region impacts predominantly on lateral aggregation, as has long been presumed. These experiments also suggested that lateral aggregation facilitated by this region proceeds without the release of FpB. From these studies we learned that individual sites within fibrinogen do not function in isolation. We conclude that thrombin-catalyzed polymerization is mediated by a continuum of concerted interactions.

Synthesis of a mouse model of the dysfibrinogen Vlissingen/Frankfurt IV.

The dysfibrinogen Vlissingen/Frankfurt IV is characterized as a deletion of Asn319 and Asp320 from the C-terminus of the gamma-chain of fibrinogen. This dysfibrinogen, which was identified in several family members that are all heterozygous for the in-frame 6-bp deletion, is associated with both venous and arterial thrombosis. Here, we describe the generation of a murine model of the V/F IV dysfibrinogen using gene targeting of mouse gamma-chain DNA. Preliminary analysis shows that the human and mouse variant fibrinogens are similar: analogous to the human V/F IV protein, the D1 fragment of the variant mouse fibrinogen is partially protected from digestion in the presence of calcium or Gly-Pro-Arg-Pro. These heterozygous mice provide the first opportunity to examine the association of thrombophilia and dysfibrinogenemia in a controlled genetic background.

Fibrinogen Longmont. A heterozygous abnormal fibrinogen with B beta Arg-166 to Cys substitution associated with defective fibrin polymerization.

B beta Arg166 to Cys substitution was identified in an abnormal fibrinogen named fibrinogen Longmont. The proband, a young woman, and her mother were heterozygous; both experienced episodes of severe hemorrhage at childbirth. The neo-Cys residues were found to be disulfide-bridged to either an isolated Cys amino acid or to the corresponding Cys residue of another abnormal fibrinogen molecule, forming dimers. Thrombin and batroxobin induced fibrin polymerization were impaired, despite normal release of fibrinopeptides A and B. Moreover, the polymerization defect was not corrected by removing the dimeric species or adding calcium. Fibrinogen Longmont had normal polymerization site a, as evidenced by normal GPRP-peptide binding. Thus, the sites A and a can interact to form protofibrils, as evidenced by dynamic light scattering measurements. These protofibrils, however, do not associate laterally in a normal manner, leading to an abnormal clot formation.

Polymerization site a function dependence on structural integrity of its nearby calcium binding site.

To explore the functional relationship between the polymerization site a and the nearby high affinity calcium binding site, we analyzed four variant fibrinogens with substitutions at these sites: gamma D364A in the a site and gamma D318A, gamma D320A, and gamma D318 + gamma D320A in the Ca2+ site. In all cases fibrinopeptide A release was normal and thrombin catalyzed polymerization was markedly impaired (unpublished observations). We examined the functional connection between the Ca2+ site and the a site by testing for plasmin protection in the presence of Ca2+ or the a site peptide ligand GPRP. SDS-PAGE analysis of the products showed that gamma D364A fibrinogen was protected from plasmin cleavage by Ca2+ but not by the GPRP peptide. In contrast, neither Ca2+ nor the GPRP peptide protected gamma D318A, gamma D320A, or gamma D318 + gamma D320A fibrinogens from complete plasmin cleavage. These results suggest that the structural integrity of the calcium binding site is required for expression of the a site. In contrast, the structural integrity of the a site has no functional consequence on Ca2+ binding to this high affinity site.

The formation of beta fibrin requires a functional a site.

We used recombinant fibrinogens in which the a site is disrupted to examine beta-fibrin formation in the absence of a functional a site. Our variants have only b sites available, and they showed no evidence of fibrin polymer formation after cleavage of FpB with venzyme. We conclude that B-b interactions are not strong enough to induce clot formation. Our studies do not rule out the involvement of b in the formation of beta-fibrin, yet they do provide evidence that a is likely to be essential in the formation of beta-fibrin.

Clot lysis of variant recombinant fibrinogens confirms that fiber diameter is a major determinant of lysis rate.

Previous studies have suggested that clots with thinner fiber diameter lyse at slower rates than clots with thicker fiber diameter. We examined lysis of fibrin clots formed from three variant fibrinogens, each with substitutions in the N-terminal region of the B beta chain. When we measured lysis as the rate of decrease in turbidity at 350 nm, we found that the rate of lysis was slower than normal for clots with thinner fibers. We noted, however, that the time to complete lysis was the same for all clots. Thus, when the data were considered as the percent of lysis with time, we found that the curves were the same as normal. We suggest that a complete and accurate characterization of clot dissolution requires comparison of normalized lysis rates.

Mutations on fibrinogen (gamma 316-322) are associated with reduction in platelet adhesion under flow conditions.

In this paper we report on studies of platelet adhesion to several fibrinogen gamma chain variants under physiological flow conditions. Reduced platelet adhesion was found to patient dysfibrinogen Vlissingen and its recombinant form (deletion of gamma 319-320). Furthermore, substitutions of the amino acids 318, 320, or both in the recombinant fibrinogen gamma chain showed a strong decrease in platelet adhesion under flow conditions in our perfusion system. Antibodies raised against peptides covering these sequences inhibited platelet adhesion completely, which suggested that the gamma 316-322 sequence could be involved in platelet adhesion in flowing blood.