Proposed structural models of human factor Va and prothrombinase.

BACKGROUND: The prothrombinase complex consists of factor Xa, FVa, calcium ions, and phospholipid membrane. The prothrombinase complex plays a key role in the blood coagulation process. OBJECTIVE: To derive solvent-equilibrated models of human FVa and the prothrombinase complex. METHODS: Several modeling techniques have been employed, including homology modeling, protein-protein docking, and molecular dynamics simulation methods, to build the structural models. RESULTS AND CONCLUSIONS: We found, upon simulation, a possibly significant shift towards planarity of the five FVa domains. To estimate a prothrombinase structure, we docked an FXa model to the equilibrated FVa model using experimental data as docking filters. We found that simulation of the docked complex led to some changes in the protein-protein contacts, but not buried surface area, as compared to the initial docking model. Possible locations of prothrombin binding to prothrombinase are indicated.

Three-dimensional structural studies on fragments of fibrinogen and fibrin.

Fibrinogen is a 340 kDa glycoprotein found in the blood plasma of all vertebrates. It is transformed into a fibrin clot by the action of thrombin. Recent X-ray structures of core fragments of both fibrinogen and fibrin have revealed many details about this polymerization event. These include structures of a 30 kDa recombinant gammaC domain, an 86 kDa fragment D from human fibrinogen and a cross-linked double-D fragment from fibrin.

Crystallization of fragment D from human fibrinogen.

Fragment D from human fibrinogen has been crystallized. The fragment, which is composed of three disulfide-linked chains (alpha' beta' gamma' = 88,000), was generated with either plasmin or mild trypsin digestion. The crystals diffracted out to 3.5 A; the space group is P2(1), unit cell dimensions a = 108 A, b = 48 A, c = 167 A, beta = 106 degrees. Fragment D was also co-crystallized with the ligand GPRP-amide, in which case the space group is consistent with P212121, unit cell dimensions a = 476 A, b = 82 A, c = 432 A.

Crystallization of the A alpha subunit of protein phosphatase 2A.

The A alpha subunit of human protein phosphatase 2A forms crystals in space group P2(1) with cell dimensions a = 104.0, b = 174.9, c = 168.2 A, and beta angle = 90.2 degrees. At cryogenic temperatures, the crystals diffracted to a resolution limit of approximately 3.0 A. Based on the unit cell dimensions and a calculated molecular mass of 65,277 Da, the Matthews coefficient suggests eight molecules per asymmetric unit. Two native data sets were collected to a nominal resolution of 3.0 A and merged to provide a set that is 93% complete, with Rsym of 9.9%.

Predicting the pharmacology of thrombin inhibitors.

Thrombotic disorders can lead to uncontrolled thrombin generation and clot formation within the circulatory system leading to vascular thrombosis. Direct inhibitors of thrombin have been developed and tested in clinical trials for the treatment of a variety of these thrombotic disorders. The bleeding complications observed during these trials have raised questions about their clinical use. The development of a computer-based model of coagulation using the kinetic rates of individual reactions and concentrations of the constituents involved in each reaction within blood has made it possible to study coagulation pathologies in silico. We present an extension of our initial model of coagulation to include several specific thrombin inhibitors. Using this model we have studied the effect of a variety of inhibitors on thrombin generation and compared these results with the clinically observed data. The data suggest that numerical models will be useful in predicting the effectiveness of inhibitors of coagulation.

A three-dimensional consideration of variant human fibrinogens.

Recently reported X-ray structures for large core fragments derived from human fibrinogen and fibrin make it possible to correlate structural and functional anomalies of known genetic variants. Here we examine a variety of amino acid replacements previously reported for hereditary dysfibrinogenemias, most of which are associated with impaired fibrin polymerization. For many of these we have modeled in the mutant amino acid and considered the structural consequences. We have also examined the cases of a small deletion and a large insertion, as well as the impact of substitutions in the GPRPam ligand that was co-crystallized with fragment double-D.

Localization and quantification of alpha-1-proteinase inhibitor in the human cornea.

alpha-1-Proteinase inhibitor, formerly called alpha-1-antitrypsin, was detected in human corneal epithelium, stroma, Descemet's membrane and endothelium using an indirect immunolocalization technique. The average alpha-1-proteinase inhibitor levels detected by an immunodot blot assay in the epithelium, stroma and Descemet's membrane-endothelium extracts per total human cornea were 29.5 micrograms, 54.3 micrograms and 3.5 micrograms, respectively. Immunolocalization on Western blots of SDS polyacrylamide electrophoresis gels revealed that all three layers contained a molecule with a molecular weight equal to the native alpha-1-proteinase inhibitor. Additionally, in the epithelial and stromal extracts minor bands at 75 kD and 110 kD were noted which are possibly due to complexes with proteases. The 110 kD band alternatively may be a dimer of the inhibitor. The epithelial extract contained bands at 40 kd and less than 30 kD indicating the presence of proteolysis products. alpha-1-Proteinase inhibitor probably plays a major role in the protection of the cornea from proteases released by inflammatory cells.

Evolution of vertebrate fibrin formation and the process of its dissolution.

The thrombin-catalysed conversion of fibrinogen into a fibrin gel is common to all extant vertebrates. Because fibrin formation is both temporary and risky, an effective scheme for fibrinolysis evolved concomitantly. In this regard, the fibrinogen molecule is well adapted both for network polymerization and for subsequent dismantling. The question is, has it always been so? It has long been known that the three non-identical chains that compose vertebrate fibrinogen are descended from a common ancestor, and the original molecule must have been either a homotrimer or a dimer thereof. Three-dimensional studies on core fragments of fibrinogen are revealing new insights about both fibrin formation and its destruction. These studies are also showing exactly what structural changes have accompanied changes in function for the various domains. Chief among these is the reversal of direction for the alpha chain after replacement of its C-terminal domain.

Human fibrinogen: anticipating a 3-dimensional structure.

The principal component of blood clots is a protein meshwork called fibrin. The precursor protein, fibrinogen, occurs in a soluble form in the blood plasma where it is activated by thrombin when and if the need arises. More than a century after first being purified, fibrinogen has yet to have its detailed 3-dimensional structure revealed. The situation is changing rapidly, however, and crystallographic studies in progress in several laboratories on a variety of fragments and complexes may soon reveal not only its structure but also the subtleties of how this large glycoprotein is transformed into a fibrin clot.

Crystal structures of fragment D from human fibrinogen and its crosslinked counterpart from fibrin.

In blood coagulation, units of the protein fibrinogen pack together to form a fibrin clot, but a crystal structure for fibrinogen is needed to understand how this is achieved. The structure of a core fragment (fragment D) from human fibrinogen has now been determined to 2.9 A resolution. The 86K three-chained structure consists of a coiled-coil region and two homologous globular entitles oriented at approximately 130 degrees to each other. Additionally, the covalently bound dimer of fragment D, known as 'double-D', was isolated from human fibrin, crystallized in the presence of a Gly-Pro-Arg-Pro-amide peptide ligand, which simulates the donor polymerization site, and its structure solved by molecular replacement with the model of fragment D.

Neurons of the rostral fastigial nucleus are responsive to cardiovascular and respiratory challenges.

The rostral fastigial nucleus (rFN) of the cerebellum has been implicated in the neural control of the cardiovascular and respiratory systems. Electrical stimulation and electrolytic lesions of this region produce changes in both cardiovascular and respiratory function. It has been suggested that some of these changes may result from effects on fibers of passage rather than on cell bodies of origin within the rFN. In the present study, extracellular recordings demonstrated a high percentage of units within rFN, as well as in adjacent areas, which responded to induction of acute increases or decreases in arterial blood pressure. Furthermore, units were identified in rFN which responded to respiratory stimuli as well as to changes in blood pressure. Out of the population tested, no units responding to respiratory stimuli were found in areas adjacent to rFN. In addition, a high percentage of neurons tested for response to passive movement also showed changes in firing rate to either cardiovascular or respiratory challenges, or both. Several units were identified (mostly in rFN), whose basal firing pattern was respiratory-related. This suggests the presence of cell bodies of origin within the rFN whose function is related to cardiorespiratory activity.

Stimulating fastigial nucleus alters central mechanisms regulating phrenic activity.

In paralyzed, mechanically ventilated, anesthetized cats, 30-sec trains of electrical stimulation in the rostral fastigial nucleus caused either respiratory excitation or inhibition (apnea) of respiration. The response elicited by stimulation was dependent on the frequency of stimulation: the duration of apnea decreased and respiratory excitation occurred more frequently at lower frequencies of stimulation. Excitation was characterized by increases of f and the rate of rise and peak (TNA) integrated activity of the phrenic nerve. TI and TE decreased. TNA and f remained elevated for at least three minutes after excitatory respiratory responses. Also, respiration was frequently elevated after an inhibitory response. Brief stimulation (100-200 ms) administered during expiration or inspiration altered either TE or TI and TNA, respectively. In addition, brief stimulation elicited short-latency inhibition or excitation of phrenic nerve activity. These effects were often unassociated with other phase changes. We conclude that activation of neurons or axons within the rostral fastigial nucleus can modulate activity of the phrenic nerve by altering the activity of at least three separate central mechanisms.

Conformational changes in fragments D and double-D from human fibrin(ogen) upon binding the peptide ligand Gly-His-Arg-Pro-amide.

The structure of fragment double-D from human fibrin has been solved in the presence and absence of the peptide ligands that simulate the two knobs exposed by the removal of fibrinopeptides A and B, respectively. All told, six crystal structures have been determined, three of which are reported here for the first time: namely, fragments D and double-D with the peptide GHRPam alone and double-D in the absence of any peptide ligand. Comparison of the structures has revealed a series of conformational changes that are brought about by the various knob-hole interactions. Of greatest interest is a moveable "flap" of two negatively charged amino acids (Glubeta397 and Aspbeta398) whose side chains are pinned back to the coiled coil with a calcium atom bridge until GHRPam occupies the beta-chain pocket. Additionally, in the absence of the peptide ligand GPRPam, GHRPam binds to the gamma-chain pocket, a new calcium-binding site being formed concomitantly.

Crystal structure of fragment double-D from human fibrin with two different bound ligands.

Factor XIII-cross-linked fragment D (double-D) from human fibrin was crystallized in the presence of two different peptide ligands and the X-ray structure determined at 2.3 A. The peptide Gly-Pro-Arg-Pro-amide, which is an analogue of the knob exposed by the thrombin-catalyzed removal of fibrinopeptide A, was found to reside in the gamma-chain holes, and the peptide Gly-His-Arg-Pro-amide, which corresponds to the beta-chain knob, was found in the homologous beta-chain holes. The structure shows for the first time that the beta-chain knob does indeed bind to a homologous hole on the beta-chain. The gamma- and beta-chain holes are structurally very similar, and it is remarkable they are able to distinguish between these two peptides that differ by a single amino acid. Additionally, we have found that the beta-chain domain, like its gamma-chain counterpart, binds calcium.

Crystal structure of a recombinant alphaEC domain from human fibrinogen-420.

The crystal structure of a recombinant alphaEC domain from human fibrinogen-420 has been determined at a resolution of 2.1 A. The protein, which corresponds to the carboxyl domain of the alphaE chain, was expressed in and purified from Pichia pastoris cells. Felicitously, during crystallization an amino-terminal segment was removed, apparently by a contaminating protease, allowing the 201-residue remaining parent body to crystallize. An x-ray structure was determined by molecular replacement. The electron density was clearly defined, partly as a result of averaging made possible by there being eight molecules in the asymmetric unit related by noncrystallographic symmetry (P1 space group). Virtually all of an asparagine-linked sugar cluster is present. Comparison with structures of the beta- and gamma-chain carboxyl domains of human fibrinogen revealed that the binding cleft is essentially neutral and should not bind Gly-Pro-Arg or Gly-His-Arg peptides of the sort bound by those other domains. Nonetheless, the cleft is clearly evident, and the possibility of binding a carbohydrate ligand like sialic acid has been considered.