Prothrombin In Presence Research Articles

Prothrombin expression in cancer-derived cell lines

The link between thrombotic disorders and cancer has been known for over 150 years, although the precise mechanism of this relationship has not yet been resolved. Current data show that thrombin has a significant role in cancer metabolism, invasiveness, adhesion and survival. However, data regarding the expression of the thrombin precursor prothrombin in various cancer cell lines are scarce. Therefore, it was our objective to determine whether common cancer-derived cell lines (Caco-2, MCF-7, SK-BR-3, U-87 and U-251) express prothrombin. The prothrombin RNA expression level was assessed by qPCR, and the presence of prothrombin was analyzed by Western blot analysis. Our results show that Caco-2 cells originating from colorectal adenocarcinoma express prothrombin, whereas other analyzed cell lines do not. Our results provide a background for further research into the role of (pro)thrombin in cancer etiopathology.

MASP-1 of the complement system promotes clotting via prothrombin activation

Mannan-binding lectin-associated serine protease-1 (MASP-1), a protein of the complement lectin pathway, resembles thrombin in terms of structural features and substrate specificity, and it has been shown to activate coagulation factors. Here we studied the effects of MASP-1 on clot formation in whole blood (WB) and platelet-poor plasma (PPP) by thrombelastography and further elucidated the underlying mechanism. Cleavage of prothrombin by MASP-1 was investigated by SDS-PAGE and N-terminal sequencing of cleavage products. Addition of MASP-1 or thrombin to WB and PPP shortened the clotting time and clot formation time significantly compared to recalcified-only samples. The combination of MASP-1 and thrombin had additive effects. In a purified system, MASP-1 was able to induce clotting only in presence of prothrombin. Analysis of MASP-1-digested prothrombin confirmed that MASP-1 cleaves prothrombin at three cleavage sites. In conclusion, we have shown that MASP-1 is able to induce and promote clot formation measured in a global setting using the technique of thrombelastography. We further confirmed that MASP-1-induced clotting is dependent on prothrombin. Finally, we have demonstrated that MASP-1 cleaves prothrombin and identified its cleavage sites, suggesting that MASP-1 gives rise to an alternative active form of thrombin by cleaving at the cleavage site R393.

Essential role of the p38 mitogen-activated protein kinase pathway in tissue factor gene expression mediated by the phosphatidylserine-dependent antiprothrombin antibody

The aim of this study was to investigate the effects of phosphatidylserine-dependent antiprothrombin antibody (aPS/PT) on the expression of tissue factor (TF) and the signal transduction pathway in procoagulant cells. Peripheral blood mononuclear cells (PBMCs) from a healthy donor, murine monocyte RAW264.7 cells and human umbilical vein endothelial cells (HUVECs) were treated with either IgG fractions obtained from APS patients who were positive for aPS/PT or a murine monoclonal aPS/PT antibody, 231D, in the presence of prothrombin. The levels of TF mRNA were measured using real-time PCR. TF function, as measured by procoagulant activity, was determined with a clotting assay. 231D binding on the surface of treated cells was determined by flow cytometric analysis. Screening for phosphorylation of intracellular signalling proteins was conducted using an array assay. Phosphorylation of p38 MAPK was quantitatively analysed with ELISA, and SB203580 was used as a specific inhibitor of p38 MAPK. Specific siRNA for p38 MAPK was used for the knockdown assay. The IgG fractions from APS patients and 231D induced TF mRNA overexpression and shortening of coagulation time in cells in the presence of prothrombin. The 231D moiety induced phosphorylation of p38 MAPK after binding to the cell surface of RAW264.7 cells. SB203580 or p38 siRNA significantly hampered TF overexpression. Expression of TF in procoagulant cells was induced by aPS/PT via p38MAPK phosphorylation. This phenomenon may be correlated with the thrombogenicity of APS.

Tissue prothrombinase activity in myometrium from timed-pregnant rats.

Fgl2 and thrombin potentially play roles during inflammation-induced preterm delivery. These studies sought to demonstrate functional prothrombinase enzyme activity in rat uterus, consistent with the expression of Fgl2 in this tissue. Myometrial and other tissue obtained from non-pregnant and timed-pregnant rats was homogenized in Tris-buffered saline. Prothrombinase activity was determined based on the kinetic metabolism of a chromogenic thrombin substrate. Homogenates were incubated with prothrombin followed by the addition of the thrombin substrate. Thrombin activity was determined by comparing tissue activity to a standard curve generated using 0.01 to 0.04 units of active thrombin. Western blot studies were also performed to confirm tissue prothrombinase activity in myometrial homogenates. Prothrombin was incubated with tissue homogenates; the reactions were then terminated with sodium dodecyl sulfate (SDS) loading buffer. Prothrombinase activity in myometrial tissue was observed to be 0.047 to 0.077 U thrombin/10 min/microg protein. Heat denaturation and calcium removal eliminated prothrombinase activity, whereas the addition of Factor V enhanced activity. The Western blots confirmed the presence of prothrombin, the anticipated prethrombin fragments, and thrombin. Consistent with the enzyme studies, the thrombin band formed upon incubation with myometrial homogenates from pregnant and nonpregnant rats. In contrast, the thrombin band was not apparent with removal of calcium, heat denaturation and treatment with serine protease inhibitors. In summary, these studies have confirmed functionally active prothrombinase activity in rat myometrium, supporting the hypothesis that Fgl2 is expressed in this tissue.

Anti‐prothrombin IgG from patients with anti‐phospholipid antibodies inhibits the inactivation of factor Va by activated protein C

Interference of anti-phospholipid antibodies with the protein C pathway has been suggested to play a role in the development of thrombosis in the anti-phospholipid syndrome. We studied the effect of IgG preparations containing anti-prothrombin antibodies of 17 lupus anticoagulant-positive patients and 12 controls on the inactivation of factor Va (FVa) by activated protein C (APC) in a system with purified coagulation factors. Test IgG was incubated with human prothrombin, phospholipid vesicles and CaCl(2). Protein S, FVa and APC were added and the APC-dependent loss of FVa activity was monitored over time. The residual amount of FVa remaining after 10 min was 14 +/- 4% (mean +/- SD) when 1.5 mg/ml normal IgG was present and ranged between 17% and 82% with 1.5 mg/ml patient IgG. Twelve patients IgG gave values of residual FVa >22% (i.e. 2 SD above the mean of controls), indicating that APC-mediated inactivation of FVa was significantly inhibited. The inhibition was strictly dependent on the presence of prothrombin, proportional to the concentration of IgG and strongly diminished at a 20-fold higher phospholipid concentration. Most, although not all, IgG containing anti-prothrombin antibodies inhibit the APC-catalysed FVa inactivation, which may contribute to the increased risk of thrombosis in patients with the anti-phospholipid syndrome.

Determination of prothombinase activation after adding human purified prothrombin to human clot: comparison of hirudin, an activated factor II inhibitor, with DX9065a, an activated factor X inhibitor, on clot-associated thrombin and on prothrombin activation

Clot-associated prothrombinase and thrombin activities may contribute to thrombus extension after thrombolytic and anticoagulant treatment. We studied prothrombin activation after adding human purified prothrombin to human clot. By using two different drugs with an exclusive direct anti-activated factor X activity (DX9065a) or anti-activated factor II activity (r-hirudin), we tried to determine whether clot-bound thrombin and prothrombinase could be inhibited in our experimental system when human purified prothrombin was added. Standard clots were prepared from platelet-poor human plasma after addition of calcium. We measured clot-bound thrombin or free thrombin using a direct simple chromogenic assay. In parallel, prothrombin fragment 1+2 measurement was used to monitor prothrombin activation. For this, two protocols were used. We introduced the direct inhibitors before starting the activation process (protocol A) or at the time of the activation process (protocol B). We found a direct correlation between thrombin generation and prothrombin fragment 1+2 with an increase of thrombin activity on clots and in the incubation mixtures when clots were incubated in human purified pothrombin alone. Two protocols were used: in the first, clots were pre-incubated in presence of drugs before adding prothrombin; and in the second, clots were incubated in the presence of prothrombin and drugs. Prothrombin activation was not inhibited when clots were incubated with r-hirudin and consequently thrombin generation still occurred. However, added r-hirudin blocks thrombin activity on the clots and in the incubation mixture, but does not prevent prothrombin activation, as shown by the increase of prothrombin fragment 1+2. In contrast, DX9065a did not suppress clot-bound thrombin. However, DX9065a blocks prothrombin activation whichever protocol was used. The results show that hirudin is a poor inhibitor of thrombin generation in contrast to DX9065a. On the other hand, DX9065a cannot inhibit thrombin bound to clot in contrast to hirudin.

DX-9065a inhibition of factor Xa and the prothrombinase complex: mechanism of inhibition and comparison with therapeutic heparins

Factor Xa (fXa) is the key enzyme of the prothrombinase complex that generates thrombin hence it is a good target for antithrombotic therapy. Here, the anti-fXa and anti-prothrombinase activities of DX-9065a which is an active-site directed inhibitor of fXa, and therapeutic heparins which are dependent on antithrombin (AT) for their anticoagulant function, were studied in amidolytic and proteolytic activity assays. It was found that DX-9065a is a competitive inhibitor of the Spectrozyme FXa (SpFXa) cleavage by both fXa and prothrombinase with similar K(i) values of approximately 10-20 nM. However, DX-9065a acted as a non-competitive inhibitor of prothrombin activation by prothrombinase with a Ki of approximately 26 nM. On the other hand, therapeutic heparins were effective catalysts of both fXa and prothrombinase inhibition by AT in the presence of SpFXa, but were ineffective in the presence of prothrombin. Further studies revealed that Tyr(99), a residue in the extended S2-S4 binding pocket of fXa, plays a key role in determination of specificity of the DX-9065a interaction.

Prothrombin and beta 2-glycoprotein I frequently contribute to antiphospholipid antibody interactions with phospholipids and the generation of abnormal waveform profiles in coagulation assays.

Transmittance waveforms are generated during clot formation on photo-optical coagulation analyzers. We previously showed that 61.5% of patients with antiphospholipid antibodies (APLA) exhibited a negative deflection in the pre-coagulation phase of the prothrombin time (PT slope 1). The current studies investigated the 'molecular basis' of this abnormal parameter. We found that the negative PT slope 1 is IgG-mediated and is not dependent on the presence of fibrinogen or thrombin activity. We also found that IgG from most of the patients required a specific thromboplastin and the presence of prothrombin or beta(2)-glycoprotein I beta(2) GPI) to produce an abnormal IgG wave-form assay. In addition, the abnormal IgG waveform required cofactor binding to phospholipids when beta(2) GPI was the cofactor, and annexin V could partially block this interaction. In conclusion, these results showed that the interactions of IgG with phospholipids via beta(2) GPI or prothrombin constitute the core mechanisms of the abnormal waveforms.

Roles of Platelets and Factor XI in the Initiation of Blood Coagulation by Thrombin

To account for the variable hemostatic defect in patients with factor XI (FXI) deficiency, with normal hemostasis in contact factor deficiencies, a coagulation paradigm is presented whereby trace quantities of thrombin, generated transiently by exposure of tissue factor at sites of vascular injury, activates FXI bound to the platelet surface in the presence of prothrombin or high Mr kininogen (HK). Tissue factor pathway inhibitor (TFPI) limits the flux of thrombin generated by the tissue factor pathway, and protease nexin II (PNII), released from activated platelets, inhibits solution phase FXIa and localizes FIX activation to the platelet surface where FXIa is protected from inactivation by PNII. Either prothrombin or HK binds to the Apple 1 (A1) domain of FXI, thereby exposing a platelet-binding site in the FXI A3 domain. Dimeric FXI binds to activated platelets directly through the A3 domain of one monomer. After proteolytic activation of platelet-bound FXI by thrombin (or FXIIa), a substrate binding site for FIX is exposed in the opposite monomer that promotes FIX activation on the platelet surface resulting in the local explosive generation of thrombin and the formation of hemostatic thrombi at sites of vascular injury.

Increased Risk for Venous Thrombosis in Carriers of the Prothrombin G→A^20210 Gene Variant

A mutation in the prothrombin gene (G-->A20210) has been associated with higher plasma prothrombin levels and an increased tendency for venous thrombosis. To determine whether the prothrombin A20210 allele is independently associated with the occurrence of venous thrombosis. Case-control study. Two thrombosis centers in southern Italy. 281 consecutive patients with venous thrombosis confirmed by objective tests and 850 controls. Medical history was collected on standardized questionnaires. The presence of prothrombin G-->A2020 and factor V Leiden mutations was determined by polymerase chain reaction. The presence of anticoagulant factors and prothrombin activity was determined by tests of function. In 150 controls, increased prothrombin activity (P < 0.001) was associated with the prothrombin A20210 allele. This allele was more frequent in patients than in controls (8.01% compared with 2.29%; P < 0.001) and was associated with an increased risk for thrombosis (odds ratio, 3.88 [95% CI, 2.23 to 6.74]). The increased prevalence of this allele was independent of the presence of the factor V Leiden mutation. After adjustment for sex, age, arterial thrombosis, and factor V Leiden mutation, the risk was still significantly elevated (odds ratio, 3.13 [CI, 1.89 to 5.21]). Moreover, the overall prevalence of inherited coagulation abnormalities was significantly higher in patients with thrombosis of the lower extremities than in patients with thrombosis of the upper extremities (odds ratio, 3.77 [CI, 1.10 to 12.93]). Fourteen patients carried both the prothrombin G-->A20210 and factor V Leiden mutations. The prothrombin A20210 allele is independently associated with the occurrence of venous thrombosis, particularly in patients with a history of thrombosis of the lower extremities.

Inhibition of Lupus Anticoagulant Activity by Hexagonal Phase Phosphatidylethanolamine in the Presence of Prothrombin

We have previously demonstrated that lupus anticoagulant antibodies from patients with systemic lupus erythematosus (SLE) specifically recognize hexagonal (II) phase phosphatidylethanolamine (PE), but not bilayer PE (Thromb Haemost 1989; 62: 892). In those studies, the involvement of proteins in this recognition was not evaluated. To address this issue, we have isolated IgG lupus anticoagulant antibodies from the plasma of SLE patients and evaluated the inhibition of lupus anticoagulant activity by hexagonal (II) phase PE in the presence and absence of purified plasma proteins. All six of the IgG lupus anticoagulant antibodies tested were inhibited by hexagonal (II) phase PE in the presence, but not the absence, of human prothrombin. In contrast, little or no inhibition was observed with prothrombin alone or with PE in combination with either beta2-glycoprotein I or annexin V. These data indicate that, for certain lupus anticoagulant antibodies, inhibition by hexagonal (II) phase PE is dependent on prothrombin, suggesting that these antibodies recognize a complex of PE and prothrombin.

Purification and Characterization of TAFI, a Thrombin-activable Fibrinolysis Inhibitor

Previous studies demonstrated that tissue plasminogen activator-induced fibrinolysis in vitro is retarded in the presence of prothrombin (II) activation and that the anticoagulant-activated protein C appears profibrinolytic by preventing the formation of thrombin (IIa)-like activity during fibrinolysis. To disclose the molecular connection between the generation of IIa and the inhibition of fibrinolysis, a lysis assay that is sensitive to the antifibrinolytic effect of II activation was developed and was used to purify a 60-kDa single-chain protein from human plasma. Because the lysis of a clot, produced from purified components, is retarded when this protein is present and when II activation occurs in situ, the protein was named TAFI (thrombin-activatable fibrinolysis inhibitor). TAFI is cleaved by IIa yielding 35-, 25-, and 14-kDa products. Amino-terminal sequence analyses identified TAFI as a precursor of a plasma carboxypeptidase B (CPB). Formation of the 35-kDa product correlates with both prolongation of lysis time and CPB-like activity. Prolongation of lysis time saturates at about 125 nM TAFI. Activated TAFI inhibits the activation of Glu-plasminogen but does not prolong the lysis of clots formed in the presence of Lys-plasminogen. 2-Guanidinoethylmercaptosuccinic acid, a competitive inhibitor of CPB, completely inhibits prolongation of lysis by activated TAFI in a purified system and the prolongation induced by II activation in barium-adsorbed plasma. This suggests that TAFI accounts for the antifibrinolytic effect that accompanies prothrombin activation and that activated protein C appears profibrinolytic by attenuating TAFI activation.

Prothrombin Activation by Prothrombinase in a Tubular Flow Reactor

Thrombin production by the phospholipid-bound complex of blood clotting factors Xa and Va (prothrombinase) was studied in a tubular flow reactor. The inner wall of a glass capillary was coated with a phospholipid bilayer of 25% phosphatidylserine and 75% phosphatidylcholine. Prothrombinase was assembled on this bilayer by perfusion with a mixture containing an excess of factor Va (2 nM) and a limiting amount of factor Xa (1-100 pM), either in the absence or presence of prothrombin. The rate of assembly of prothrombinase in the presence of prothrombin appeared to be limited by the transfer rate of factor Xa to the phospholipid surface. A good agreement was found between the predicted mass transfer coefficient for factor Xa and the observed pre-steady state rate of thrombin production. The eventually obtained steady state rates of thrombin production were proportional to the prothrombin concentration and independent of the surface density of prothrombinase. The observed rate of thrombin production was in excellent agreement with the predicted mass transfer rate for prothrombin. Transport-limited prothrombin conversion was observed for prothrombinase densities exceeding 1 fmol/cm2, which corresponds to 0.05% occupation of available binding sites. The kinetic parameters of the reaction were determined at low prothrombinase densities (0.02-0.04 fmol/cm2). Even in this situation the Michaelis-Menten equation had to be corrected for substrate depletion near the catalytic surface. We hereto employed an accurate approximation of the mass transfer coefficient. The kinetic parameter kcat was 60 s-1 and the intrinsic Km had a surprisingly low value of 3 nM. Both parameters were not influenced by the wall shear rate.

Inhibition of Prothrombinase by Antithrombin-Heparin at a Macroscopic Surface

The antithrombin-dependent inhibition of prothrombinase, assembled at a macroscopic surface, was studied under flow conditions utilizing a tubular flow reactor that consists of a phospholipid-coated glass capillary. Prothrombinase activity was determined from steady-state rates of thrombin production upon perfusion with prothrombin and from factor Va-associated factor Xa activity present in the flow reactor. The prothrombinase density was maintained at a low level (0.03 fmol/cm2) to assure that the rate of thrombin production reflected the amount of prothrombinase present in the capillary. Perfusion of the flow reactor with antithrombin resulted in an exponential decrease of prothrombinase activity in time. The second order rate constant (8.5 x 10(4) M-1min-1) is comparable with the rate of inactivation of free factor Xa. Inhibition was much faster when antithrombin was complexed with heparin. The second order rate constants of inhibition decreased with decreasing heparin chain length: 9.6 x 10(7), 4.5 x 10(7) and 0.39 x 10(7) M-1min-1 for unfractionated heparin, low molecular weight heparin and synthetic pentasaccharide heparin, respectively. In the presence of prothrombin (0.2 microM), however, the heparin-dependent rate of inhibition of prothrombinase was about 50-fold lower. The heparin-independent inhibition of prothrombinase by antithrombin (4 microM) in the presence of prothrombin (0.2 microM) was virtually negligible. At a 70-fold higher surface density of prothrombinase (2 fmol/cm2) prothrombinase activity was much faster inactivated. The rate of thrombin production, however, was not affected. In conclusion, at low prothrombinase densities, prothrombin efficiently protects prothrombinase from inhibition.(ABSTRACT TRUNCATED AT 250 WORDS)

Functional and Binding Studies of the Roles of Prothrombin and β2-Glycoprotein I in the Expression of Lupus Anticoagulant Activity

AbstractWe present functional and binding data relevant to the reported roles for prothrombin and β2-glycoprotein I (β2GPI) in the expression of lupus anticoagulant activity. In a purified system containing human prothrombin, Xa, Va, and a rate-limiting concentration of phosphatidylserine (PS)/phospha-tidylcholine (PC) vesicles, the preliminary incubation of vesicles with protein A separated IgG preparations from 10 lupus anticoagulant plasmas, calcium, and prothrombin enhanced the inhibitory effect of all IgG preparations upon thrombin generation. Experiments in a purified factor X activation system provided supporting data that a similar preliminary incubation with prothrombin enhanced the inhibitory effect of many of the IgG preparations upon factor X activation. However, we could not obtain unequivocal evidence that prothrombin was an obligatory cofactor for lupus anticoagulant IgG to inhibit procoagulant phospholipid function, because lupus anticoagulant IgG separated by protein A chromatography contained traces of prothrombin. The binding of many IgG preparations to immobilized PS was enhanced by prothrombin when calcium ions were present. β2GPI enhanced binding of many of the IgG preparations to immobilized PS both in the presence and absence of calcium, yet βGPI failed to enhance the ability of the IgG preparations to inhibit phospholipid function in purified prothrombin and factor X assays. Moreover, the IgG preparations prolonged the dilute Russell’s viper venom time (dRVVT) of βGPI-depleted normal plasma. Nine of 10 IgG preparations bound to prothrombin on Western blots in the absence of calcium and phospholipid, whereas no preparation bound to β2GPI. Passage of five citrated lupus anticoagulant plasmas through a prothrombin affinity column in the absence of added calcium and phospholipid removed most of the activity prolonging the dRVVT of normal plasma, and IgG in the pass-through plasma no longer bound to PS in the presence of prothrombin and calcium ions. IgG in prothrombin column eluates had strikingly enhanced specific lupus anticoagulant activity and also specific PS binding activity in the presence of prothrombin and calcium ions. Thus, lupus anticoagulant plasmas were shown to contain IgG binding to prothrombin, in the absence of calcium ions and phospholipid, which could also, in the presence of calcium ions and prothrombin, bind to PS and express lupus anticoagulant activity.

The activation of prothrombin by platelet-bound factor Xa.

After the platelet release reaction, factor Xa binds to a receptor on the platelet surface with high affinity, Ka 5.2 × 1o9 M−1. Binding of the factor Xa substrate, prothrombin to the platelet surface has not yet been possible to demonstrate. The interaction between prothrombin and platelets was therefore studied by determination of the structural requirements on the prothrombin molecule for activation by platelet-bound factor Xa. A comparison was made between the rates of activation of normal prothrombin, prethrombin 1, prethrombin 2, and acarboxyprothrombin (prothrombin lacking γ-carboxyglutamic acid residues) in the presence of factor Xa and isolated platelets. Both the vitamin-K-dependent fragment 1 of prothrombin and fragment 2 which interacts with factor V were found to be required for rapid prothrombin activation. Acarboxyprothrombin and prethrombin 1 were slowly activated to thrombin by factor Xa on the platelet surface after the platelet release reaction. In the same system prethrombin 2 was activated more slowly to thrombin. In the presence of fragment 2, which binds noncovalently to prethrombin 2, the activation rate was enhanced and equal to that of prethrombin 1 supporting the conclusion that factor V is part of the platelet receptor. The cleavage products of prothrombin by factor Xa in the presence of platelets were found to be the same as those previously described when the activation was catalyzed by the prothrombinase complex. The coefficient for proteolytic efficiency, Kcat/Km, was found to be approximately 50 times higher in the presence of prothrombin than in the presence of acarboxyprothrombin. We conclude that prothrombin interacts with a specific platelet structure consisting of phospholipid and an activated form of factor V exposed on the platelet surface as a result of the release reaction.

The Procoagulant in Human Urine: Purification, Assay and Some Biochemical and Physiological Properties

SummaryA powerful procoagulant is excreted with the human urine. The purification, assay with purified prothrombin, definition of unit, properties and animal studies are described. The material is relatively heat resistent, and it activates prothrombin in presence of Ac-globulin and platelet factor 3 or platelet substitute (brain lipid). Fresh male human urine contains 10 u/ml. One to 4 u/ml are required to normalize clotting of hemophilic plasma. Purified preparations exhibit a specific activity around 11,000 u/mg tyrosine. Intravenous injections or infusions of the procoagulant into dogs or rabbits induce for a few hours a phase of transitory hypercoagulability as indicated by shortening of recalcification time and enhancement of prothrombin consumption. This is followed in the rabbit by the appearance of a circulating “antithromboplastin”.

Prothrombin as Co-Factor of the Circulating Anticoagulant in Systemic Lupus Erythematosus?

SummaryA 35-year old man with lupus erythematosus of the skin is demonstrated, who also shows signs of systemic L. E.: sometimes slightly increased ESR, varying thrombocytopenia, slightly prolonged coagulation and prothrombin time, positive cepnalin-cholesterol flocculation test, biologically false positive syphilis reactions and signs of a slightly increased haemolysis. Electrophoretic examination of the serum proteins presents practically normal values, certainly no hypergammaglobulinaemia. Analytical examination of coagulation reveals hypoprothrombinaemia and an anticoagulant most probably directed against thrombocyte factor 3, intrinsic thromboplastin and tissue thromboplastin. The anticoagulant seems to be only active in the presence of prothrombin. Prothrombin therefore, is considered a co-factor of the anticoagulant. The inactive anticoagulant is localized in the gamma globulins. The possible cause of the hypoprothrombinaemia (immune reaction?) and the data in the literature are discussed.

Stability of Prothrombin in the Presence of Thrombin

In earlier numbers of these PROCEEDINGS, Mertz, Seegers and Smith, allege that thrombin can inactivate its precursor prothrombin. We have published evidence that a trypsin-like enzyme is not only a thromboplastic agent but also a progressive antithrombin of major importance in natural thrombin destruction. It is present, to a varying extent, in serum and in most prothrombin and thrombin preparations. There is doubt whether such a factor was ruled out in the Iowa experiments. It should be pointed out that Smith, et al., use extremely potent thrombic agents and assay them by a dilution technic which yields values up to several thousands. In our experience, strong thrombins (and prothrombins) appear unusually stable but turn out to be much less so when diluted to, say, about 5–25 Iowa units. In Table I of another Iowa communication, there is a 9% loss of activity of thrombin in the presence of prothrombin as compared with a much more startling 95% fall in the prothrombin titer during the same 4-hour period. Reagents and Methods. Our routine prothrombins, when maximally activated with CaCl2 and brain thromboplastin, usually clot a test fibrinogen in some 5-20+ seconds (temp. = 37.5°C, pH = 7.5). A typical prothrombin solution, assayed through the courtesy of Dr. H. P. Smith (Iowa) gave 47 units per cc (38 units per mg N). Persisting in our unwillingness to use any system of thrombin “units” for comparing one preparation with another, without convincing proof of control of instability factors, we restrict the assay to a single prothrombin preparation and the thrombins freshly prepared from it. Under these narrowed conditions, thrombin concentrations are measured, with a 5–10% accuracy, in a relative manner from the clotting-times obtained with a series of dilutions tested under strictly comparable conditions.