Interaction of clotting factor V heavy chain with prothrombin and prethrombin 1 and role of activated protein C in regulating this interaction: analysis by analytical ultracentrifugation

Changes in the affinity of the heavy subunit of blood coagulation factor Va (Vh) for prothrombin are thought to be important in regulating the rate of thrombin production. Using analytical ultracentrifugation, we have measured the affinity of bovine Vh for prothrombin and for the prethrombin 1 fragment of prothrombin at 23.3 degrees C, pH 7.65, in 50 mM tris(hydroxymethyl)aminomethane, 0.1 M NaCl, 0.1 mM benzamidine, and either 2 mM Ca2+ or 2 mM ethylenediaminetetraacetate (EDTA). Under these conditions a 1:1 complex of Vh with prothrombin is formed that is governed by a dissociation constant (Kd) of 10 microM, regardless of whether the buffer contains Ca2+ or EDTA. An identical Kd is observed when prethrombin 1 is substituted for prothrombin. This indicates that the fragment 1 portion of prothrombin, containing the gamma-carboxyglutamic acid residues, does not influence the association. Substitution of human prethrombin 1 for the bovine molecule also results in a 1:1 Vh-prethrombin 1 complex governed by a slightly weaker Kd (27 microM). Discrete proteolysis of bovine Vh by the anticoagulant activated protein C converts the Vh to a form with little or no affinity for prethrombin 1 (Kd greater than 1 mM), without detectable change in the mass of the Vh.     

Interaction of human prothrombin with tissue thromboplastin

The binding of 125I-labeled human prothrombin to native and papain-treated tissue thromboplastin in the presence of CaCl2 or EDTA was studied. The Scatchard plots for the protein binding suggest the presence at thromboplastin surface of two types of binding sites, high affinity [Kd(app) = 7.4.10(-8) M] and moderate affinity [Kd(app) = 7.9.10(-5) M]. The removal of Ca2+ did not influence the Kd (values for these) sites but markedly reduced their number. Proteolysis by papain caused a decrease in the affinity of high affinity sites without affecting the Kd values of the moderate affinity sites yet caused a proportional increase in the number of both high and moderate affinity sites in the presence of Ca2+. At low prothrombin concentrations a positive cooperativity of protein binding at high affinity sites in the presence of Ca2+ was observed.     

Staphylocoagulase-binding region in human prothrombin

A staphylocoagulase-binding region in human prothrombin was studied by utilizing several fragments prepared from prothrombin by limited proteolysis. Bovine prothrombin, prethrombin 1, prethrombin 2, and human diisopropylphosphorylated alpha-thrombin strongly inhibited formation of the complex ("staphylothrombin") between human prothrombin and staphylocoagulase, but bovine prothrombin fragment 1 and fragment 2 had no effect on the complex formation, indicating that the binding region of human prothrombin for staphylocoagulase is located in the prethrombin 2 molecule. To identify further the staphylocoagulase-binding region, human alpha-thrombin was cleaved into the NH2-terminal large fragment (Mr = 26,000) and the COOH-terminal fragment (Mr = 16,000) by porcine pancreatic elastase. Of these fragments, the COOH-terminal fragment, which includes Asn-200 to the COOH-terminal end of the alpha-thrombin molecule, partially inhibited the complex formation between staphylocoagulase and human prothrombin. In contrast, the NH2-terminal large fragment did not show any inhibitory effect on the staphylothrombin formation. These results suggest that the staphylocoagulase interacts with human prothrombin through the COOH-terminal region of alpha-thrombin B chain. Other plasma proteins, factor X, factor IX, protein C, protein S, protein Z, all of which are structurally similar to prothrombin, did not inhibit the staphylothrombin formation at all, indicating that a specific interaction site with staphylocoagulase is contained only in the prothrombin molecule.     

Proteolytic formation of either of the two prothrombin activation intermediates results in formation of a hirugen-binding site.

Hirugen, a synthetic dodecapeptide corresponding to the carboxyl-terminal amino acids 53-64 of hirudin, binds within a deep groove in thrombin that contains a cationic region referred to as the anion-binding exosite. This region is important in many of the binary interactions of thrombin with macromolecular substrates and cofactors. Fluorescein-labeled hirugen was used to probe which steps in the prothrombin activation process generate this anion-binding exosite. Two activation cleavage sites exist in bovine prothrombin. Cleavage at Arg274-Thr275 releases the activation fragments to generate the thrombin precursor, prethrombin 2. Cleavage of prothrombin within a disulfide loop at Arg323-Ile324 leads to formation of meizothrombin with no loss of peptide material but with formation of amidolytic activity. Cleavage of the same bond in prethrombin 2 generates thrombin. Hirugen, labeled at the amino terminus with fluorescein isothiocyanate, does not bind to prothrombin but does bind to thrombin (Kd = 9.6 +/- 1.2 x 10(-8) M), prethrombin 2 (Kd = 1.3 +/- 0.1 x 10(-7) M), thrombin-fragment-2 complex (Kd = 1.1 +/- 0.2 x 10(-6) M), and meizothrombin (Kd = 1.6 +/- 0.5 x 10(-8) M). Prothrombin fragment-2 and hirugen both bind independently to thrombin. A ternary complex can form with hirugen and fragment-2 and either thrombin or prethrombin 2, suggesting that fragment-2 and hirugen bind to discrete sites. Hirugen also alters the active site conformation of thrombin as detected by modulation of synthetic substrate hydrolytic activity. These studies suggest that conformational changes, rather than alleviating steric hindrance, are responsible for the formation of the hirugen-binding site during prothrombin activation. Furthermore, this conformational change can be effected by the cleavage of either of the two bonds required for activation of prothrombin.     

Membrane binding induces lipid-specific changes in the denaturation profile of bovine prothrombin. A scanning calorimetry study.

Prothrombin denaturation was examined in the presence of Na2EDTA, 5mM CaCl2, and CaCl2 plus membranes containing 1-palmitoyl-2-oleoyl-3-sn-phosphatidylcholine (POPC) in combination with either bovine brain phosphatidylserine (PS) or 1,2-dioleoyl-phosphatidylglycerol (DOPG). Heating denaturation of prothrombin produced thermograms showing two peaks, a minor one at approximately 59 degrees C previously reported to correspond to denaturation of the fragment 1 region (Ploplis, V. A., D. K. Strickland, and F. J. Castellino 1981. Biochemistry. 20:15-21), and a main one at approximately 57-58 degrees C, reportedly due to denaturation of the rest of the molecule (prethrombin 1). The main peak was insensitive to the presence of 5mM Ca2+ whereas the minor peak was shifted to higher temperature (Tm approximately 65 degrees C) by Ca2+. Sufficient concentrations of POPC/bovPS (75/25) large unilamellar vesicles to guarantee binding of 95% of prothrombin resulted in an enthalpy loss in the main endotherm and a comparable enthalpy gain in the minor endotherm accompanying an upward shift in peak temperature (Tm approximately 73 degrees C). Peak deconvolution analysis on the prothrombin denaturation profile and comparison with isolated prothrombin fragment 1 denaturation endotherms suggested that the change caused by POPC/PS vesicles reflected a shift of a portion of the enthalpy of the prethrombin 1 domain to higher temperature (Tm approximately 77 degrees C). The enthalpy associated with this high-temperature endotherm increased in proportion to the surface concentration of PS. By contrast, POPC/DOPG (50/50) membranes shifted the prethrombin 1 peak by 4 degrees C to a lower temperature and the fragment 1 peak by 5 degrees C to a higher temperature. The data lead to a hypothesis that the fragment 1 and prethrombin 1 domains of prothrombin do not denature quite independently and that binding of prothrombin to acidic-lipid membranes disrupts the interaction between these domains. It is further hypothesized that PS containing membranes exert the additional specific effect of decoupling the denaturation of two subdomains of the prethrombin 1 domain of prothrombin.     

Phospholipid-binding properties of bovine factor V and factor Va.

Factor V and factor Va binding to single bilayer phospholipid vesicles was investigated by light-scattering intensity measurements. This technique allows the measurement of free and phospholipid-bound protein concentrations from which equilibrium constants can be obtained. As controls, the Ca2+-dependent phospholipid binding of prothrombin and factor X were also studied. The average values obtained for the dissociation constants (Kd) and lipid to protein ratio at saturation, moles/mole (n), for prothrombin (Kd = 2.3 X 10(-6) M, n = 104) and factor X (Kd = 2.5 X 10(-6) M, n = 46) binding to vesicles containing 25% Folch fraction III and 75% phosphatidylcholine in the presence of 2 mM Ca2+ were in agreement with those reported in the literature. The average factor V and factor Va values for the dissociation constants and lipid to protein ratio at saturation (moles/mole) were Kd = 7.2 X 10(-8) M and n = 270 for factor V and Kd = 4.4 X 10(-7) M and n = 76 for factor Va. In contrast to prothrombin and factor X, factor V and factor Va demonstrated Ca2+-independent lipid binding. In addition, the number of factor V and factor Va molecules bound per vesicle was found to be dependent both on the phosphatidylserine content of the vesicle and the ionic strength of the buffer.     

Calcium binding to the epidermal growth factor homology region of bovine protein C

A high affinity calcium binding site that is independent of the gamma-carboxyglutamic acid-rich amino-terminal region, has been demonstrated in bovine protein C, as well as in the other vitamin K-dependent proteins (except prothrombin) involved in blood coagulation. gamma-Carboxyglutamic acid-independent calcium binding in protein C is required for its rapid activation by the thrombin-thrombomodulin complex. We have now isolated a Ca2+-binding fragment from a tryptic digest of bovine protein C. The isolated fragment contains the two domains that are homologous to the epidermal growth factor precursor from the light chain of protein C, and a small disulfide bound peptide derived from the heavy chain. The isolated fragment bound 1 mol of Ca2+/mol of protein with a dissociation constant (Kd) of approximately 1 x 10(-4) M. This is similar to the Kd previously determined for binding of a single Ca2+ ion to protein C lacking the gamma-carboxyglutamic acid region. Immunochemical evidence indicated that Ca2+ binding induced a conformational change both in protein C lacking the gamma-carboxyglutamic acid region and in the isolated fragment.     

Calcium binding to tandem repeats of EGF-like modules. Expression and characterization of the EGF-like modules of human Notch-1 implicated in receptor-ligand interactions.

The Ca(2+)-binding epidermal growth factor (cbEGF)-like module is a structural component of numerous diverse proteins and occurs almost exclusively within repeated motifs. Notch-1, a fundamental receptor for cell fate decisions, contains 36 extracellular EGF modules in tandem, of which 21 are potentially Ca(2+)-binding. We report the Ca(2+)-binding properties of EGF11-12 and EGF10-13 from human Notch-1 (hNEGF11-12 and hNEGF10-13), modules previously shown to support Ca(2+)-dependent interactions with the ligands Delta and Serrate. Ca2+ titrations in the presence of chromophoric chelators, 5,5''-Br2BAPTA and 5-NBAPTA, gave two binding constants for hNEGF11-12, Kd1 = 3.4 x 10(-5) M and Kd2 > 2.5 x 10(-4) M. The high-affinity site was found to be localized to hNEGF12. Titration of hNEGF10-13 gave three binding constants, Kd1 = 3.1 x 10(-6) M, Kd2 = 1.6 x 10(-4) M, and Kd3 > 2.5 x 10(-4) M, demonstrating that assembly of EGF modules in tandem can increase Ca2+ affinity. The highest affinity sites in hNEGF11-12 and hNEGF10-13 had 10 to 100-fold higher affinity than reported for EGF32-33 and EGF25-31, respectively, from fibrillin-1, a connective tissue protein with 43 cbEGF modules. A model of hNEGF11-12 based on fibrillin-1 EGF32-33 demonstrates electronegative potential that could contribute to the higher affinity of the Ca(2+)-binding site in hNEGF12. These data demonstrate that the Ca2+ affinity of cbEGF repeats can be highly variable among different classes of cbEGF containing proteins.     

Ca2+ dependence of the interactions between protein C, thrombin, and the elastase fragment of thrombomodulin. Analysis by ultracentrifugation.

The two-way and three-way interactions among active-site-blocked bovine thrombin, bovine protein C, and the elastase fragment of rabbit thrombomodulin (elTM) were examined by analytical ultracentrifugation at 23.3 degrees C in 100 mM NaCl, 50 mM Tris (pH 7.65), and 1 mM benzamidine, in the presence of 0 to 5 mM calcium chloride. Thrombin and elTM form a tight (Kd less than 10(-8) M) 1:1 complex in the absence of Ca2+ that weakens with the addition of Ca2+ (Kd approximately 4 microM in 5 mM Ca2+). Without Ca2+, thrombin and protein C form a 1:1 complex (Kd approximately 1 microM) and what appears to be a 1:2 thrombin-protein C complex. The Kd for the 1:1 complex weakens over 100-fold in 5 mM CaCl2. Protein C and elTM form a Ca(2+)-independent 1:1 complex (Kd approximately 80 microM). Nearly identical binding to thrombin and elTM is observed when active-site-blocked activated bovine protein C is substituted for protein C. Thrombin inhibited by diisopropyl fluorophosphate and thrombin inhibited by a tripeptide chloromethyl ketone exhibited identical behavior in binding experiments, suggesting that the accessibility of protein C to the substrate recognition cleft of these two forms of thrombin is nearly equal. Human protein C binds with lower affinity than bovine protein C. Ternary mixtures also were examined. Protein C, elTM, and thrombin form a 1:1:1 complex which dissociates with increasing [Ca2+]. In the absence of Ca2+, protein C binds to the elTM-thrombin complex with an apparent Kd approximately 1 microM.(ABSTRACT TRUNCATED AT 250 WORDS)     

Localization of the metal-induced conformational transition of bovine prothrombin

The calcium-stabilized antigenic determinants on bovine prothrombin were localized to the NH2-terminal 1-42 residues using conformation-specific antibodies. Polyclonal antibodies to the bovine prothrombin-Ca(II) complex were raised in rabbits, and purified antibody subpopulations were isolated by sequential immunoabsorption and affinity chromatography. Anti-prothrombin-Ca(II) antibodies, characterized by their absolute specificity for the prothrombin-metal complex (Tai, M. M., Furie, B. C., and Furie, B. (1980) J. Biol. Chem. 255, 2790-2795), bound to prothrombin, fragment 1, reduced and carboxymethylated fragment 1, and CNBr fragment (1-72) in solution. However, these antibodies do not bind significantly to the gamma-carboxyglutamic acid-rich fragment (1-39), CNBr fragment (73-156), or prethrombin 1. To obviate the complex analysis of possible reasons for the lack of antibody binding to small peptides in solution, conformation-specific antibodies directed against defined regions of the whole prothrombin molecule were isolated. The influence of calcium ions on the binding of these site-specific antibody subpopulations to 125I-labeled prothrombin fragment 1 was evaluated. Anti-(1-39)N, anti-(1-42)N, anti-(1-72)N, and anti-(reduced and carboxymethylated fragment 1)N showed enhanced binding to prothrombin fragment 1 in the presence of Ca(II), indicating the presence of calcium-stabilized antigenic determinants within each of these regions on fragment 1. In contrast, calcium ions had no effect on the interaction of anti-des-(1-42)prothrombin, anti-prethrombin 1, anti-(43-72)N, and anti-(73-156)N antibodies with prothrombin fragment 1. These results indicate that the metal-induced conformational transition, monitored immunochemically, is localized to the NH2-terminal, gamma-carboxyglutamic acid-rich region of prothrombin between residues 1-42.     

Analysis of human alpha-thrombin by hydrophobic interaction high-performance liquid chromatography

The major active form of human thrombin, alpha-thrombin, was analyzed by hydrophobic interaction high-performance liquid chromatography (HIC-HPLC). The chromatographic system included a polymeric phenyl column and elution was performed by a gradient, 2-0M sodium chloride (5-20 min). Total analysis time was 30 min per injection. By this method, a good resolution between alpha-thrombin and the proteolytically modified thrombin forms, beta- and gamma-thrombin, was obtained. In addition, the thrombin preforms, prothrombin, prethrombin 1, and prethrombin 2, were also resolved from alpha-thrombin in the system. The results from the HIC method were compared to those obtained from non-reducing sodium dodecyl sulfate-polyacrylamide gel electrophoresis. By this high-resolution chromatographic method, the rapid analysis of purified alpha-thrombin is possible.     

Fibrinogen-sepharose interaction with prothrombin, prethrombin 1, prethrombin 2 and thrombin

Binding of prothrombin, prethrombin 1, prethrombin 2 and thrombin to fibrinogen-Sepharose was studied. Thrombin and prethrombin 2 bound to fibrinogen-Sepharose, while prethrombin 1 and prothrombin did not. Bound thrombin and prethrombin 2 were recovered from the column by eluting with 0.1 M NaCl/0.05 M Tris-HCl buffer (pH 7.4). The affinity of thrombin and prethrombin 2 to fibrinogen-Sepharose depended on ionic strength and reached a maximum at 50 mm concentration. Prethrombin 2 interacts with fibrinogen as well as thrombin; and prothrombin fragment 1.2 is not important in the formation of this complex. Thus, prethrombin 2, which is a precursor of thrombin without measurable enzymatic activity and which lacks the single cleavage at Arg-322-Ile-323 present in thrombin, has the same or very similar structural conformation as thrombin and has the same macromolecular substrate recognition site. These results confirm the earlier results that active center is not necessary in fibrinogen-thrombin interaction.     

Weakly bound calcium ions involved in the thermostability of aqualysin I, a heat-stable subtilisin-type protease of Thermus aquaticus YT-1.

Aqualysin I is a heat-stable protease; in the presence of 1 mM Ca(2+), the enzyme is stable at 80 degrees C and shows the highest activity at the same temperature. After gel filtration to remove free Ca(2+) from the purified enzyme sample, the enzyme (holo-aqualysin I) still bound Ca(2+) (1 mol/mol of the enzyme), but was no longer stable at 80 degrees C. On treatment of the holo-enzyme with EDTA, bound Ca(2+) decreased to about 0.3 mol/mol of the enzyme. The thermostability of holo-aqualysin I was dependent on the concentration of added Ca(2+), and 1 mM added Ca(2+) stabilized the enzyme completely, suggesting that aqualysin I has at least two Ca(2+) binding sites, i.e. stronger and weaker binding ones. Titration calorimetry showed single binding of Ca(2+) to the holo-enzyme with an association constant of 3.1 x 10(3) M(-1), and DeltaH and TDeltaS were calculated to be 2.3 and 6.9 kcal/mol, respectively, at 13 degrees C. La(3+), Sr(2+), Nd(3+), and Tb(3+) stabilized the holo-enzyme at 80 degrees C, as Ca(2+) did. These results suggest that the weaker binding site exhibits structural flexibility to bind several metal cations different in size and valency, and that the metal binding to the weaker binding site is essential for the thermostability of aqualysin I.     

Phosphorylation of brush border myosin I by protein kinase C is regulated by Ca(2+)-stimulated binding of myosin I to phosphatidylserine concerted with calmodulin dissociation.

Brush border myosin I from chicken intestine is phosphorylated in vitro by chicken intestinal epithelial cell protein kinase C. Phosphorylation on serine and threonine to a maximum of 0.93 mol of P/mol of myosin I occurs within an approximately 20 kDa region at the end of the COOH-terminal tail of the 119-kDa heavy chain. The effects of Ca2+ on myosin I phosphorylation by protein kinase C are complex, with up to 4-fold stimulation occurring at 0.5-3 microM Ca2+, and up to 80% inhibition occurring at 3-320 microM Ca2+. Phosphorylation required that brush border myosin I be in its phosphatidylserine vesicle-bound state. Previously unknown Ca2+ stimulation of brush border myosin I binding to phosphatidylserine vesicles was found to coincide with Ca2+ stimulation of phosphorylation. A myosin I proteolytic fragment lacking approximately 20 kDa of its tail retained Ca(2+)-stimulated binding, but showed reduced Ca(2+)-independent binding. Ca(2+)-dependent phosphatidylserine binding is apparently due to the concomitant phosphatidylserine-promoted, Ca(2+)-induced dissociation of up to three of the four calmodulin light chains from myosin I. Four highly basic putative calmodulin-binding sites in the Ca(2+)-dependent phosphatidylserine binding region of the heavy chain were identified based on the similarity in their sequence to the calmodulin- and phosphatidylserine-binding site of neuromodulin. Calmodulin dissociation is now shown to occur in the low micromolar Ca2+ concentration range and may regulate the association of brush border myosin I with membranes and its phosphorylation by protein kinase C.     

Calcium-dependent and calcium-independent interactions of prothrombin fragment 1 with phosphatidylglycerol/phosphatidylcholine unilamellar vesicles

We have measured the phase behavior of mixed dipentadecanoylphosphatidylglycerol (DC15PG)/dimyristoylphosphatidylcholine (DMPC) small unilamellar vesicles (SUV) in the presence of saturating (greater than 98% occupancy of binding sites) concentrations of bovine prothrombin fragment 1 and 5 mM Ca2+. Binding of fragment 1 in the presence of Ca2+ was verified by an increase in 90 degrees light scattering. Only in the cases of DC15PG/DMPC SUV below their phase transition and of pure DMPC SUV were such light scattering measurements not reversible upon addition of ethylenediaminetetraacetic acid to complex Ca2+. Phase-behavior changes of DC15PG/DMPC SUV as monitored by diphenylhexatriene fluorescence anisotropy occurred in concert with the binding of fragment 1. The major effects of peptide binding on SUV phase behavior were to raise the phase-transition temperature by 2-15 degrees C, depending on vesicle composition, and, in general, to make the phase diagram for these small vesicles closely resemble that of large vesicles. No evidence was obtained for the existence of lateral membrane domains with distinct compositions induced by the binding of prothrombin fragment 1 plus Ca2+. Surprisingly, fragment 1 without Ca2+ also altered the phase behavior of DC15PG/DMPC SUV. Most striking was the effect of fragment 1 (with or without Ca2+) on DMPC SUV phase behavior. Freeze-fracture electron microscopy demonstrated that pure DMPC vesicles were induced to fuse in the presence of fragment 1, while vesicles containing DC15PG remained intact. The rate of DMPC SUV fusion (followed by 90 degrees light scattering) increased with increasing fragment 1 concentration but was not saturable up to 40 microM fragment 1, suggesting a weak, nonspecific interaction between fragment 1 and the neutral phospholipid vesicle.(ABSTRACT TRUNCATED AT 250 WORDS)     

The alpha 1-adrenergic receptor in human erythrocyte membranes mediates interaction in vitro of epinephrine and thyroid hormone at the membrane Ca(2+)-ATPase.

Membrane Ca(2+)-ATPase activity was stimulated in vitro separately by T4 (10(-10) M) and by epinephrine (10(-6) M). In the presence of a fixed concentration of T4, additions of 10(-8) and 10(-6) M epinephrine reduced the T4 effect on the enzyme. beta-Adrenergic blockade with propranolol (10(-6) M) prevented stimulation by epinephrine of Ca(2+)-ATPase activity, but did not prevent the suppressive action of epinephrine on T4-stimulable Ca(2+)-ATPase. In contrast, alpha 1-adrenergic blockade with unlabelled prazosin restored the effect of T4 on Ca(2+)-ATPase activity in the presence of epinephrine. Like propranolol, prazosin prevented enhancement of enzyme activity by epinephrine in the absence of thyroid hormone. Neither prazosin nor propranolol had any effect on the stimulation by T4 of red cell Ca(2+)-ATPase in the absence of epinephrine. Analysis of radiolabelled prazosin binding to human red cell membranes revealed the presence of a single class of high-affinity binding sites (Kd, 1.2 x 10(-8) M; Bmax, 847 fmol/mg membrane protein). Thus, the human erythrocyte membrane contains alpha 1-adrenergic receptor sites that are capable of regulating Ca(2+)-ATPase activity.     

Calmodulin-binding proteins of calcium-independent type in rat brain synaptosomal membranes: their localization and properties.

Some properties of calmodulin(CaM)-binding proteins (CaMBPs) of the Ca(2+)-independent type were investigated in the synaptosomal membrane (SM) from rat brain using the [125I]CaM gel overlay method. When SM was prepared in the presence of Ca2+, Ca(2+)-independent CaM binding was decreased, whereas the Ca(2+)-dependent type was not altered. All Ca(2+)-independent-type CaMBPs were membrane-bound and scarcely present in the soluble fractions. When SM was heat-denatured, the 24/22.5-kDa CaMBPs could no longer be detected by [125]CaM binding and a new component with higher molecular mass (greater than 200 kDa) was shown to bind CaM in a Ca(2+)-independent manner. A possible effect of cAMP- and Ca2+/CaM-dependent phosphorylation on CaM binding was also examined.     

Fourier transform infrared spectroscopic study of Ca2+ and membrane-induced secondary structural changes in bovine prothrombin and prothrombin fragment 1.

Fourier transform infrared (FTIR) spectroscopy was used to monitor secondary structural changes associated with binding of bovine prothrombin and prothrombin fragment 1 to acidic lipid membranes. Prothrombin and prothrombin fragment 1 were examined under four different conditions: in the presence of (a) Na2EDTA, (b) 5 mM CaCl2, and in the presence of CaCl2 plus membranes containing 1-palmitoyl-2-oleoyl-3-sn-phosphatidylcholine (POPC) in combination with either (c) bovine brain phosphatidyl-serine (bovPS) or (d) 1,2-dioleoyl-phosphatidylglycerol (DOPG). The widely reported Ca(2+)-induced conformational change in bovine prothrombin fragment 1 was properly detected by our procedures, although Ca(2+)-induced changes in whole prothrombin spectra were too small to be reliably interpreted. Binding of prothrombin in the presence of Ca2+ to procoagulant POPC/bovPS small unilamellar vesicles produced an increase in ordered secondary structures (2% and 3% increases in alpha-helix and beta-sheet, respectively) and a decrease of random structure (5%) as revealed by spectral analysis on both the original and Fourier-self-deconvolved data and by difference spectroscopy with the undeconvolved spectra. Binding to POPC/DOPG membranes, which are less active as procoagulant membranes, produced no detectable changes in secondary structure. In addition, no change in prothrombin fragment 1 secondary structure was detectable upon binding to either POPC/bovPS or POPC/DOPG membranes. This indicates that a membrane-induced conformational change occurs in prothrombin in the nonmembrane-binding portion of the molecule, part of which is activated to form thrombin, rather than in the membrane-binding fragment 1 region. The possible significance of this conformational change is discussed in terms of differences between the procoagulant activities of different acidic lipid membranes.     

Metal and phospholipid binding properties of partially carboxylated human prothrombin variants

To study the specific role of gamma-carboxyglutamic acid (Gla) residues in prothrombin, we have isolated a series of partially carboxylated prothrombin variants from a patient with a hereditary defect in vitamin K-dependent carboxylation (Goldsmith, G. H., Pence, R. E., Ratnoff, O. D., Adelstein, D. A., and Furie, B. (1982) J. Clin. Invest. 69, 1253-1260). The three variant prothrombins, purified by DEAE-Sephacel, immunoaffinity chromatography, and preparative gel electrophoresis, were indistinguishable from prothrombin in molecular weight, amino acid composition, and NH2-terminal amino acid sequence, with the exception of Gla residues. Variant prothrombin 1, with 8 Gla residues, had 66% of the coagulant activity of prothrombin, one high affinity metal-binding site (Kd = 15 nM), and three lower affinity sites (Kd = 2.7 microM); prothrombin contained two high affinity (36 nM) and four lower affinity sites (Kd = 1 microM). Ca(II) induced a 23% decrease in the intrinsic fluorescence of variant prothrombin 1 fragment 1, compared to a 35% decrease in that of prothrombin fragment 1. The phospholipid binding activity of variant prothrombin 1 was 44% that of prothrombin. Variant prothrombin 2 and variant prothrombin 3, with 4 and 6 Gla residues, respectively, had about 5% of prothrombin coagulant activity and a single high affinity and two lower affinity metal-binding sites and exhibited no phospholipid binding activity. Variant prothrombin 3 fragment 1 and variant prothrombin 2 fragment 1 demonstrated 18 and 13% of Ca(II)-induced fluorescence quenching, respectively. Abnormal prothrombin, with 1 Gla residue, had 8% of prothrombin coagulant activity, a single lower affinity (1 microM) metal-binding site, and 13% Ca(II)-induced fluorescence quenching of the fragment 1 species and did not bind to phospholipid. These results indicate that Gla residues define the metal binding properties of prothrombin. Most, if not all, of the Gla residues are required for complete prothrombin function, and the prothrombin coagulant activity correlates to the phospholipid binding activity of the prothrombin species.     

Anion-binding exosite of human alpha-thrombin and fibrin(ogen) recognition

Activation of prothrombin to alpha-thrombin generates not only the catalytic site and associated regions but also an independent site (an exosite) which binds anionic substances, such as Amberlite CG-50 resin [cross-linked poly(methylacrylic acid)]. Like human alpha-thrombin with high fibrinogen clotting activity (peak elution at I = 0.40 +/- 0.01 M, pH 7.4, approximately 23 degrees C), catalytically inactivated forms (e.g., i-Pr2P-alpha- and D-Phe-Pro-Arg-CH2-alpha-thrombins) were eluted with only slightly lower salt concentrations (I = 0.36-0.39 M), while gamma-thrombin with very low clotting activity was eluted with much lower concentrations (I = 0.29 M) and the hirudin complex of alpha-thrombin was not retained by the resin. In a similar manner, hirudin complexes of alpha-, i-Pr2P-alpha-, and gamma-thrombin were not retained by nonpolymerized fibrin-agarose resin. Moreover, the ionic strengths for the elution from the CG-50 resin of seven thrombin forms were directly correlated with those from the fibrin resin (y = 0.15 + 0.96x, r = 0.95). In other experiments, the 17 through 27 synthetic peptide of the human fibrinogen A alpha chain was not an inhibitor of alpha-thrombin, while the NH2-terminal disulfide knot (NDSK) fragment was a simple competitive inhibitor of alpha-thrombin with a Ki approximately 3 microM (0.15 M NaCl, pH 7.3, approximately 23 degrees C). These data suggest that alpha-thrombin recognizes fibrin(ogen) by a negatively charged surface, noncontiguous with the A alpha cleavage site but found within the NDSK fragment. Such interaction involving an anion-binding exosite may explain the exceptional specificity of alpha-thrombin for the A alpha cleavage in fibrinogen and alpha-thrombin incorporation into fibrin clots.