PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “ANTITHROMBIN”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3Linked to original sources

Factor Xa is highly protected from antithrombin-fondaparinux and antithrombin-enoxaparin when incorporated into the prothrombinase complex.

Antithrombin and its cofactor, heparin, target both the product of prothrombin activation by prothrombinase, thrombin, as well as the enzyme responsible for the reaction, factor (F)Xa. These studies were carried out to quantify the effects of each of the prothrombinase components on the half-life of FXa in the presence of antithrombin and the low-molecular-weight heparins (enoxaparin, Aventis, Laval, Quebec, Canada) or the heparin pentasaccharide (fondaparinux, Organon Sanofi-Synthelabo, Cypress, TX, USA). Experiments were carried out using a recombinant form of prothrombin in which the active site serine has been mutated to cysteine and subsequently labeled with fluorescein. This mutant allowed calculation of the second order rate constant for inhibition of FXa by antithrombin in such a way that competition for antithrombin by thrombin is eliminated and competition for FXa by prothrombin is accounted for. Intrinsic rate constants for the inhibition of FXa by antithrombin-enoxaparin and antithrombin-fondaparinux, in the presence of the various prothrombinase components, were calculated. Addition of phospholipid had no significant effect on the second order rate constant for inhibition of FXa by antithrombin, while addition of FVa appeared to be mildly protective. Further addition of prothrombin however, caused profound protection of FXa, increasing its half-life from 1.1 to 353 s in the case of fondaparinux, and from 0.4 to 42 s in the case of enoxaparin. Similar results were reported for unfractionated heparin previously [1]. Therefore, in the presence of unfractionated heparin, fondaparinux, or enoxaparin, prothrombinase is profoundly protected from antithrombin.

Antithrombin III↗

Antithrombin-TRI (Ala382 to Thr) causing severe thromboembolic tendency undergoes the S-to-R transition and is associated with a plasma-inactive high-molecular-weight complex of aggregated antithrombin.

An antithrombin (AT) variant Ala382 to Thr (AT-TRI) was identified by mass spectrometric techniques. The variant behaved as a substrate rather than a thrombin inhibitor, but, contrary to previously described P12 AT variants, AT-TRI, expressed as a heterozygous dominant trait, caused severe thromboembolic tendency beginning in their teens in affected members of an English family. In addition, it underwent the S-to-R conformational state transition as evidenced by an increased resistance to thermal denaturation on active centre cleavage, but did not react with a monoclonal antibody, 4C9, directed against a neoepitope that is present on complexed and cleaved normal AT. Antithrombin-TRI, in plasma, was also associated with an abnormal high molecular weight (M(r)) 194,000) component composed of non-covalently-linked antithrombin molecules. This component (D194) showed low affinity for heparin and was devoid of antithrombin progressive activity. D194, isolated by ammonium sulphate precipitation and three chromatographic steps (heparin Sepharose, ion exchange and immunoaffinity), migrated as a single band of M(r) 60,000 on SDS-PAGE under both reducing and non-reducing conditions and was recognized by monospecific anti-human antithrombin antibodies, but did not immunoreact with antibodies raised against a number of proteins including albumin and thrombin. The above data and the fact that the 15 N-terminal amino acids of this M(r) 60,000 band were identical to that of normal antithrombin indicated that the inactive D194 component was composed of aggregated antithrombin molecules, possibly antithrombin trimers. In conclusion, early adulthood severe thromboembolic tendency, failure to expose the 4C9 epitope, and presence of aggregated AT molecules in the plasma are characteristic features of AT-TRI not previously described in other ALA-382 THR mutations.

Adolescent↗

Binding of heparin to human antithrombin III activates selective chemical modification at lysine 236. Lys-107, Lys-125, and Lys-136 are situated within the heparin-binding site of antithrombin III.

A new water-soluble color reagent, 4-N,N-dimethylaminoazobenzene-4'-isothiocyano-2'-sulfonic acid (S-DABITC), was used to identify lysine residues of antithrombin III which participate in the binding of heparin. Antithrombin, modified with S-DABITC in the presence and absence of low molecular weight heparin (Mr 5000) was reduced, carboxymethylated, and digested with trypsin. The digest was analyzed by high-performance liquid chromatography and monitored at 465 nm. In the absence of heparin, four major colored peptides (T1, T2, T3, and T4) were identified. When antithrombin was preincubated with heparin (2-fold by weight), followed by S-DABITC modification, the recovery of peptide T4 remained unchanged, but the recoveries of T1, T2, and T3 were reduced by 93, 86, and 98%, respectively. In addition, a new colored peptide, TA, appeared. Amino acid sequencing of peptides T1, T2, T3, and TA localized S-DABITC modification sites as Lys-136, Lys-125, Lys-107, and Lys-236, respectively. Thus, binding of heparin to human antithrombin diminished S-DABITC modification at Lys-107, Lys-125, and Lys-136, but at the same time enhanced S-DABITC modification at Lys-236. This phenomenon was further characterized by varying the molar ratio of heparin/antithrombin (from 0.04 to 20). The shielding of Lys-125 and Lys-136 was inversely proportional to the activation of Lys-236. At a heparin/antithrombin molar ratio of 1, the extent of shielding of Lys-125 and Lys-136 and the unmasking of Lys-236 were 25-33%. This shielding-unmasking effect correlated with enhanced antithrombin inhibition of thrombin. We conclude that Lys-107, Lys-125, and Lys-136 are situated within the heparin-binding site of human antithrombin and that binding of heparin to antithrombin causes a conformational change of antithrombin that leads to the exposure of Lys-236 for S-DABITC modification.

Amino Acid Sequence↗

Antithrombin conformation and the catalytic role of heparin. I. Does cleavage by thrombin induce structural changes in the heparin-binding region of antithrombin?

Heparin has been shown to exhibit lower affinity for the antithrombin-thrombin complex than for antithrombin alone (Carlstrom, A.-S., Lieden, K., and Bjork, I. (1977) Thromb. Res. 11, 785-797), suggesting that structural alterations in antithrombin may accompany its reaction with thrombin. The hydroxy-nitrobenzyl (HNB) group attached to a unique tryptophan has been used in the present study as an extrinsic probe for localization of conformational changes to the heparin-binding region within antithrombin III using immunochemical and spectral techniques. Site-specific modification of tryptophan-49 in antithrombin with the hydroxynitrobenzyl reagent blocks heparin binding to the protein and provides a chemical label in the heparin-binding region of the protein (Blackburn, M. N., Smith, R. L., Carson, J., and Sibley, C. C. (1984) J. Biol. Chem. 259, 939-941). Antibodies specific for the hydroxynitrobenzyl hapten, which bind to HNB-tryptophan-49 in antithrombin, were used to detect a change in conformation in the region of tryptophan-49 which occurs upon thrombin binding to antithrombin. This thrombin-induced structural change was also apparent from spectral perturbations which were detected with the environmentally sensitive HNB moiety. Thus, the HNB group was used as an immunochemical probe as well as a spectral reporter group to provide insight into an allosteric mechanism of control in the catalytic role of heparin. The thrombin-promoted alteration of the structure in the heparin-binding region is presumably responsible for recycling of heparin, allowing it to catalyze further reactions between antithrombin and thrombin.

Antibodies↗

Crossed immunoelectrophoresis as applied to studies on complex formation. The binding of heparin to antithrombin III and the antithrombin III--thrombin complex.

A two-dimensional immunoelectrophoretic method has been used to obtain information on the binding of heparin to purified antithrombin III and the antithrombin III--thrombin complex. The difference in mobility of the components in a gel containing heparin enables distinction between free and complexed forms of antithrombin III. The results obtained with purified preparations show that heparin is bound more strongly to antithrombin III than to the antithrombin III--thrombin complex. In plasma heparin is bound to several components, only a fraction being bound to antithrombin III. Several components containing antithrombin III are detectable in serum.

Animals↗

Net biosynthesis of antithrombin III by the isolated rat liver perfused for 12--24 hours. Compared with rat fibrinogen and alpha-2 (acute-phase) globulin, antithrombin III is not an acute phase protein.

Antithrombin III-heparin cofactor has been isolated from normal rat plasma, purified to homogeneity on acrylamide gel electrophoresis and used to prepare a monospecific antiserum in rabbits. Measurements of rat antithrombin III were made by a single radial immunodiffusion assay. Net synthesis of antithrombin III was investigated during 12- or 24-h perfusions of the isolated rat liver. In perfusions performed under basal conditions cumulative synthesis of antithrombin-III was observed to occur at a rate sufficient to replace the total circulating plasma antithrombin III in about 6 h. In perfusions performed under full supplementation conditions which greatly enhanced synthesis of fibrinogen and alpha-2 (acute-phase) globulin (known acute-phase reactant proteins) net synthesis of antithrombin III was not significantly greater than that observed in control perfusions. Although these prolonged perfusion studies conclusively demonstrate net synthesis of antithrombin III by the isolated rat liver, they afford no evidence that this protein is an acute-phase reactant.

Animals↗

Calcium enhances heparin catalysis of the antithrombin-factor Xa reaction by promoting the assembly of an intermediate heparin-antithrombin-factor Xa bridging complex. Demonstration by rapid kinetics studies.

Heparin catalyzes the inhibition of factor Xa by antithrombin mainly through an allosteric activation of the serpin inhibitor, but an alternative heparin bridging mechanism has been suggested to enhance the catalysis in the presence of physiologic calcium levels due to calcium interactions with the Gla domain exposing a heparin binding exosite in factor Xa. To provide direct evidence for this bridging mechanism, we studied the heparin-catalyzed reaction of antithrombin with factor Xa, Gla-domainless factor Xa (GDFXa), and a heparin binding exosite mutant of GDFXa in the absence and presence of calcium using rapid kinetic methods. The pseudo-first-order rate constant for factor Xa inhibition by antithrombin complexed with a long-chain approximately 70-saccharide heparin showed a saturable dependence on inhibitor concentration in the presence but not in the absence of 2.5 mM Ca(2+), indicating the formation of an intermediate heparin-serpin-proteinase encounter complex with a dissociation constant of approximately 90 nM prior to formation of the stable serpin-proteinase complex with a rate constant of approximately 20 s(-1). Similar saturation kinetics were observed for the inhibition of GDFXa by the antithrombin-heparin complex, except that Ca(2+) was not required for the effect. By contrast, no Ca(2+)-dependent saturation of the inhibition rate constant was detectable over the same range of inhibitor concentrations for reactions of either a short-chain approximately 26-saccharide high-affinity heparin-antithrombin complex with factor Xa or the long-chain heparin-antithrombin complex with the heparin binding exosite mutant, GDFXa R240A. These findings suggest that binding of full-length heparin chains to an exosite of factor Xa in the presence of Ca(2+) produces a chain-length-dependent lowering of the dissociation constant for assembly of the intermediate heparin-antithrombin-factor Xa encounter complex, resulting in a several 100-fold rate enhancement by a heparin bridging mechanism.

Antithrombins↗

Anticoagulating and non-enzymatic fibrinolytic activities of heparin-antithrombin III and antithrombin III-heparin-thrombin complexes in vitro and in vivo.

The heparin-antithrombin III and antithrombin III-heparin-thrombin complexes were prepared in vitro. The formation of complexes was controlled by crossed electrophoretic and spectrophotometric methods. All the binary and ternary complexes prepared at the weight ratio of components 1: 1: 1 and at some other weight correlations were active as non-crosslinked fibrin solvents and kept this activity in the presence of inhibitors of plasmin fibrinolysis. A considerable prolongation of thrombin time and an increasing of total and in particular non-enzymatic lytic actions of plasma were achieved in 7 and 30 min after intravenous injection of 1 ml 0.1-0.2% solution of heparin-antithrombin III complex in rats. The obtained results of experiments allow to conclude that the heparin-antithrombin III and antithrombin III-heparin-thrombin complexes possess the non-enzymatic fibrinolytic action on non-crosslinked fibrin. The degree of this activity is dependent on the quantitative correlation of heparin-antithrombin III or antithrombin III-heparin-thrombin concentrations during the reaction of complex formation.

Animals↗

Commercial antithrombin concentrate contains inactive L-forms of antithrombin.

The preparation of antithrombin concentrate for clinical use requires a viral inactivation step. In most commercial preparations this is achieved by heat pasteurisation. This process would be expected to alter the conformation of antithrombin from the active native species to an inactive latent (L-form) state (1, 2). To determine if this occurs during commercial preparation and to identify the proportion of the product in the inactive state, we examined the various antithrombin conformations within a therapeutic concentrate. The antithrombin concentrate was separated into five fractions by heparin-Sepharose chromatography. The fraction with the highest heparin affinity retained full activity, whereas the four fractions with reduced heparin affinity (approximately 40% of the total antithrombin) had lost their inhibitory function. These inactive antithrombins were intact, monomeric, thermostable and resistant to unfolding in 8 M urea. Moreover, the protein patterns on isoelectric focusing and non-denaturing-PAGE showed that there were at least two different L-forms with isoelectric points separate from the native active species. Our findings demonstrate that approximately 40% of the antithrombin preparation examined exists as inactive L-forms. The clinical significance of administering this altered material is uncertain.

Antithrombin III↗

Structural effects of a covalent linkage between antithrombin and heparin: covalent N-terminus attachment of heparin enhances the maintenance of antithrombin's activated state.

We have produced a molecule comprising of permanently-activated covalently linked antithrombin and heparin (ATH). This study was designed to elucidate the covalent linkage point(s) for heparin on antithrombin and conformational properties of the ATH molecule. ATH was produced using Schiff base/Amadori rearrangement by incubating antithrombin with unfractionated heparin for 14 d at 40 degrees C. ATH was then digested using Proteinase K, and the heparin-peptide was reacted with NaIO4/NaBH4/mild acid to degrade the heparin moiety. Sequencing of the remaining peptide was performed by Edman degradation with linkage point confirmation by LC-MS. The degree of insertion of the reactive center loop (RCL) of antithrombin into the A-sheet of ATH was examined using synthesized antithrombin RCL peptides. Binding between the peptides and ATH, and the formation of ATH in the presence of the peptides were tested. CD was used to further examine the secondary and tertiary structures of ATH. The results suggest that heparin is conjugated to the amino terminal of antithrombin in the majority of ATH molecules, proximal to the previously determined heparin binding domain of antithrombin. From the linkage data, a model is proposed for the structure of ATH. Studies using the RCL peptides and CD analysis of ATH support this model.

Heparin↗

Antithrombin III metabolism in two colitis patients with acquired antithrombin III deficiency.

125I-Antithrombin III metabolism studies were performed in 2 patients with ischemic and ulcerative colitis, respectively. Both patients had acquired antithrombin III deficiency and objectively diagnosed deep venous thrombosis. A decreased 125I-antithrombin III plasma disappearance halflife and an increased fractional catabolic rate was found in both patients. The transcapillary flux ratio was elevated in the patient with ischemic colitis. A follow-up study of the first patient during a period when no signs of an ischemic colitis were present and no medication was taken showed completely normal tracer data. The data are consistent with both gastrointestinal loss and intravascular consumption of antithrombin III. The antithrombin III deficiency could not be explained by other causes such as proteinuria, liver dysfunction, or obvious disseminated intravascular coagulation. Reduced antithrombin III plasma levels were considered to have contributed to the development of deep venous thrombosis in both patients.

Aged↗

Antithrombin BM from human plasma: an antithrombin binding moderately to heparin.

A human antithrombin was purified app. 60 fold from Cohn fraction IV, to give a single band of about 70.000 molecular weight in polyacrylamide gel electrophoresis. Compared to the similar antithrombin III, this glycoprotein binds only moderately to porcine heparin (hence its name Antithrombin BM), thus requiring higher heparin concentration for full thrombin inhibitor function, and lower ionic strength for elution from a heparin sepharose column. In these respects it resembles "heparin cofactor A", which is, however, characterized by a substantially larger molecular weight. From AT III, AT BM further differs in its absolute dependency on the presence of heparin(oids) for antithrombin activity, in its more pronounced inhibitory specificity largely restricted to thrombin, and in the absence of substantial immunological crossreaction with antibody to AT III. Based on comparative measurements of antithrombin activity in the presence of different amounts of heparin, up to 40% of the antithrombin activity present in human blood may be attributed to AT BM. The in vivo role of this new inhibitor remains to be elucidated.

Adult↗

Control of contact activation on end-point immobilized heparin: the role of antithrombin and the specific antithrombin-binding sequence.

The uptake and activation of FXII from blood plasma was studied in small-diameter polyethylene tubing, surface-modified by end-point immobilization of heparin. Two preparations of heparin were used to modify the contact-activating properties of the plastic tubing: unfractionated, functionally active heparin and low-affinity heparin, lacking the specific antithrombin-binding sequence and virtually devoid of anticoagulant activity. The uptakes of FXII on the two heparin surfaces were similar. No activated FXII could be demonstrated on the unfractionated heparin surface, whereas on the low-affinity heparin surface nearly all FXII underwent spontaneous activation. The suppression of FXII activation on the unfractionated heparin surface was investigated by using plasma depleted of antithrombin, complement C1 esterase inhibitor, or both. The removal of antithrombin resulted in extensive activation of FXII, whereas the depletion of C1 esterase inhibitor had only a minor effect. Experiments with recalcified plasma showed rapid clot formation during exposure to the low-affinity heparin surface. After depletion of antithrombin, but not complement C1 esterase inhibitor, the recalcified plasma clotted in contact with the unfractionated heparin surface as well. We conclude that antithrombin and the antithrombin-binding sequence in the surface-immobilized heparin are essential for the prevention of surface activation of FXII and triggering of the intrinsic coagulation system.

Animals↗

Association of thrombin, plasmin, thrombin-antithrombin III complex and plasmin-antithrombin III complex with isolated hepatocytes.

The interaction of thrombin, plasmin or their antithrombin III complexes with isolated mouse hepatocytes was studied. Plasmin bound to hepatocytes in a concentration-dependent manner with an apparent Kd of 6.4.10(-8) M, attaining equilibrium within 10 min, and the interaction was inhibited by 6-amino-n-hexanoic acid. Plasmin treated with diisopropylfluorophosphate (DFP) bound to the cells in similar way as the untreated form of the enzyme. Thrombin bound also to hepatocytes, in a concentration-dependent manner, with a Kd of 5.4.10(-8) M reaching a steady state after 180 min. Thrombin inactivated with DFP, however, was inhibited in its binding to these cells. These data suggest that, whereas the kringle domains of plasmin are responsible for the enzyme-cell interaction, the active center of thrombin may be involved in the binding of this enzyme to hepatocytes. Plasmin-antithrombin III and thrombin-antithrombin III complexes were also associated with hepatocytes in a time-dependent manner, reaching a plateau after 180 min, and the two complexes competed in the interaction. While the interaction of active proteinases plasmin or thrombin with hepatocytes did not result in their internalization, the antithrombin III complexes were taken up by the cells, and thrombin-antithrombin III complex was degraded. These results indicate that hepatocytes may participate in the elimination of proteinase-antithrombin III complexes from the plasma, while the association of plasmin and thrombin with hepatocytes could imply distinct biological importance.

Animals↗

Monoclonal antibodies against plasma protease inhibitors: production and characterization of 15 monoclonal antibodies against human antithrombin III. Relation between antigenic determinants and functional sites of antithrombin III.

Fifteen hybridomas secreting monoclonal antibodies against human antithrombin III, originating from two mouse strains, have been produced by the cell fusion technique. Eight monoclonal antibodies belong to the class IgG1, five to the class IgG2a, and two to the class IgG2b. All light chains belong to the kappa group. No cross-reaction of the monoclonal antibodies have been observed with a crude preparation of albumin nor with alpha 1-antitrypsin and alpha 2-antiplasmin. Five of these monoclonal antibodies exhibit a relatively high avidity for antithrombin III. Inhibition experiments showed that the 15 monoclonal antibodies define seven more or less independent antigenic regions on the antithrombin III molecule. Examination of the effects of these antibodies on the inhibitory capacity of antithrombin III toward thrombin activity, either in the presence or in the absence of heparin, showed that several monoclonal antibodies inhibit the antithrombin III activity and allowed to relate some of the antigenic determinants to functional sites on the antithrombin III molecule.

Animals↗

Antithrombin activity of fucoidan. The interaction of fucoidan with heparin cofactor II, antithrombin III, and thrombin.

Fucoidan, poly(L-fucopyranose) linked primarily alpha 1----2 with either a C3- or a C4-sulfate, is an effective anticoagulant in vitro and in vivo (Springer, G. F., Wurzel, H. A., McNeal, G. M., Jr., Ansell, N. J., and Doughty, M. F. (1957) Proc. Soc. Exp. Biol. Med. 94, 404-409). We have determined the antithrombin effects of fucoidan on the glycosaminoglycan-binding plasma proteinase inhibitors antithrombin III and heparin cofactor II. Fucoidan enhances the heparin cofactor II-thrombin reaction more than 3500-fold. The apparent second-order rate constant of thrombin inhibition by heparin cofactor II increases from 4 x 10(4) (in the absence of fucoidan) to 1.5 x 10(8) M-1 min-1 as the fucoidan concentration increases from 0.1 to 10 micrograms/ml and then decreases as fucoidan is increased above 10 micrograms/ml. The fucoidan reaction with heparin cofactor II-thrombin is kinetically equivalent to a "template model." Apparent fucoidan-heparin cofactor II and fucoidan-thrombin dissociation constants are 370 and 1 nM, respectively. The enhancement of thrombin inhibition by fucoidan, like heparin and dermatan sulfate, is eliminated by selective chemical modification of lysyl residues either of heparin cofactor II or of thrombin. The fucoidan-antithrombin III reactions with thrombin and factor Xa are accelerated maximally 285- and 35-fold at fucoidan concentrations of 30 and 500 micrograms/ml, respectively. Using human plasma and 125I-labeled thrombin in an ex vivo system, the heparin cofactor II-thrombin complex is formed preferentially over the antithrombin III-thrombin complex in the presence of 10 micrograms/ml fucoidan. Our results indicate that heparin cofactor II is activated by fucoidan in vitro and in an ex vivo plasma system and suggest that the major antithrombin activity of fucoidan in vivo is mediated by heparin cofactor II and not by antithrombin III.

Antithrombin III↗

The heparin-enhanced antithrombin III/thrombin reaction is saturable with respect to both thrombin and antithrombin III.

The heparin-enhanced antithrombin III/thrombin reaction was studied under experimental conditions where the dependence of the reaction velocity on the concentrations of thrombin and antithrombin III could be determined. The results have shown that the reaction is saturable with respect to both thrombin (KT = 3.6 x 10(-8) M) and antithrombin III (KAT = 1.0 x 10(-7) M) when the heparin concentration is low relative to the initial protein concentrations. The apparent first order rate constant for the rate-limiting step in the reaction was approximately 800 min-1. The reaction was subject to inhibition by antithrombin III/thrombin, the product of the reaction. Inhibition appeared to be noncompetitive with respect to antithrombin II with KP, the apparent heparin product dissociation constant, approximately equal to KT. When the heparin-enhanced antithrombin III/thrombin reaction was studied under conditions where the heparin concentration was high relative to the initial protein concentrations the overall reaction was second order. The initial reaction velocity, under any set of experimental conditions, could be described by the general rate equation for a random order bireactant, enzyme-catalyzed reaction, which is mathematically identical with the "template" model for the mechanism of action of heparin (Griffith, M. J. (1982) J. Biol. Chem. 257, 7360-7365).

Antithrombin III↗

Molecular bases of antithrombin deficiency: twenty-two novel mutations in the antithrombin gene.

Antithrombin (AT) is a major physiological inhibitor of hemostasis. We report 22 novel antithrombin gene (SERPINC1) mutations associated with antithrombin deficiency in 17 French and five German families. They were all present at the heterozygous state. Nine missense mutations accounted for type I deficiency, defined by equally low antithrombin activity and antigen level. Most of them (7/9) affected highly conserved serpin residues and were associated with venous thrombosis occurring at a young age (before age 32). One splice site, one nonsense mutation, three small deletions and one insertion were also identified as a cause for type I antithrombin deficiency. Seven other missense mutations were identified in type II or unclassified AT deficiency; g.5270C>T (p.T147I, T115I) and g.5281A>T (p.I151F, I119F) change residues in the heparin binding region, g.13267C>G (p.P439A, P407A) and g.13271T>C (p.F440S, F408S) affect amino acids in the pleiotropic region, g.2372G>A (p.G25D, G-8D) changes a signal peptide amino acid, g.2456G>C (p.C53S, C21S) affects one of the three disulfide bonds of the protein, and g.7585A>T (p.M347K, M315K) changes a nonconserved residue on strand 2C.

Antithrombin III↗