Background Thrombin activity increases in patients treated with coronary thrombolysis for acute myocardial infarction, but the mechanisms are not well defined. We have shown that thrombin activity increases in plasma and whole blood incubated with plasminogen activators and appears to be plasmin mediated and dependent on activity of the factor VIIIa/IXa complex.
Methods and Results In the present study, increases in thrombin activity induced by incubation of recalcified citrated plasma with 0.16 to 0.5 μmol/L plasmin at 37°C were markedly attenuated in recalcified citrated plasma deficient in factors XI or XII, prekallikrein, or high molecular weight kininogen, as well as in plasma incubated with plasmin in the presence of 3.5 μmol/L corn trypsin inhibitor, a specific factor XIIa inhibitor. Increases in thrombin activity also occurred in nonanticoagulated whole blood incubated with pharmacological concentrations of plasminogen activators and were markedly attenuated in the presence of corn trypsin inhibitor. Plasmin-mediated (0.25 μmol/L) activation of purified factor XII occurred in 0.05 mol/L Tris-HCl and 0.012 mol/L NaCl (pH 7.8) at 37°C, resulting in equimolar quantities of two fragments that corresponded to cleavage of factor XII at Arg353-Val354, the site involved in kallikrein-mediated activation of factor XII, and cleavage at Lys346-Ser347, an apparently novel site of plasmin-mediated hydrolysis of factor XII. Contact activation was also demonstrated in plasma samples from patients after treatment with fibrinolytic agents for myocardial infarction, by demonstrating cleavage of high molecular weight kininogen from its one-chain to its two-chain form by ligand blotting with 125I-prekallikrein.
Conclusions Plasmin-mediated activation of the contact system of coagulation appears to account, at least in part, for increases in procoagulant activity in patients treated with fibrinolytic agents. It may also explain hypotension, by release of bradykinin from high molecular weight kininogen, and complement activation, by activated factor XII, that has been demonstrated in these patients.
Treatment of patients for acute myocardial infarction with fibrinolytic agents reduces mortality and improves left ventricular function. However, because of recurrent thrombosis, mortality is increased in approximately 20% of patients.1 Recurrent thrombosis and failure of clot lysis in patients treated with fibrinolytic agents are associated with increases in plasma concentrations of fibrinopeptide A (FPA), a marker of thrombin activity.2 3 The persistence of thrombin activity appears to be a critical determinant of the success of coronary thrombolysis, judging from the reduction in time to reperfusion and prevention of reocclusion observed with conjunctive administration of hirudin, a specific thrombin inhibitor.4 5 Thrombin activity increases almost immediately after initiation of streptokinase (SK) or recombinant tissue-type plasminogen activator (rTPA) infusion, judging from the rapid increase in plasma concentrations of FPA documented in clinical studies.3 6 It appears likely that increases in thrombin activity are attributable, at least in part, to increased elaboration of thrombin, since plasma concentrations of prothrombin fragment 1.2 and thrombin–antithrombin III complexes, markers of thrombin elaboration, are also increased after initiation of treatment with fibrinolytic agents.7 8 Thus, the results of clinical studies suggest that pharmacological activation of plasminogen induces activation of the coagulation system and subsequent elaboration of thrombin.
We have shown that incubation of fibrinolytic agents with recalcified citrated plasma or nonanticoagulated whole blood induces marked increases in thrombin activity.9 10 Activation of the coagulation system under these conditions was attenuated by inhibition of plasmin with aprotinin. Increased procoagulant activity in recalcified citrated plasma in response to plasminogen activation was shown to be associated with increases in the activity of the factor VIIIa/IXa complex, either by direct activation or by activation of the contact system of coagulation,9 but appeared to be independent of factor VII (unpublished data). Others have shown that plasmin induces activation of the coagulation cofactor V, which would markedly increase activity of the factor Xa/Va or prothrombinase complex.11 Although previous studies have also shown that plasmin may induce activation of factors VII and XII in purified systems,12 13 14 15 pharmacological activation of plasminogen is believed to decrease the activity of these and other coagulation factors.16 17 Accordingly, we designed the present study to define the mechanisms responsible for increases in thrombin activity in response to pharmacological activation of plasminogen in plasma and blood and to determine whether the mechanisms identified in vitro could occur in patients with myocardial infarction treated with fibrinolytic agents.
Citrated plasma was purchased from The American Red Cross. Plasma was pooled from a minimum of two donors. Citrated plasma from patients deficient in specific coagulation factors was purchased from George King Biomedical. Purified human factor XII was purchased from Enzyme Research Laboratories or was a gift from Dr Joseph Shore (Henry Ford Institute). Purified human prekallikrein was a gift of Dr Joseph Shore. Corn trypsin inhibitor was purchased from Enzyme Research Laboratories or purified according to the method of Hojima et al.18 Purified human plasmin and SK were from Kabi Diagnostica. rTPA was from Genentech. Plasminogen was purified from pooled human plasma according to the method of Castellino and Powell.19 Aprotinin and α2-antiplasmin were purchased from Sigma Chemical.
Phe-Pip-Arg-paranitroanalide (S-2238), Pro-Phe-Arg-paranitroanalide (S-2302), and Val-Leu-Lys-paranitroanalide (S-2251) were purchased from Chromogenix. Pro-Phe-Arg-chloromethyl ketone (PPACK), an inhibitor of factor XIIa and kallikrein activity, was purchased from Bachem Bioscience. Dextran sulfate was purchased from Sigma Chemical, polystyrene microtiter plates (Immulon-2) were from Dynatech, and 125I was from Amersham Corporation. Lactoperoxidase reagent (Enzymobead) was purchased from Bio-Rad Laboratories, polyvinylidene difluoride membranes (PVDF) were from Millipore Corporation, and 4% to 15% Tris gradient Mini-PROTEAN II Ready Gels were from Bio-Rad Laboratories.
Assays of Thrombin Activity
Hydrolysis of Synthetic Chromogenic Substrate
Recalcified (25 mmol/L CaCl2 final concentration) pooled citrated plasma or plasma deficient in specific coagulation factors was diluted 1:8 in 50 mmol/L Tris and 175 mmol/L NaCl (pH 8.4) at 37°C in a polystyrene microtiter plate. The plasma was incubated with human plasmin (0 to 1 μmol/L) for 3 minutes, followed by the addition of 1000 KIU/mL aprotinin to inhibit plasmin activity. Thrombin activity was determined by measurement of the change in absorbance at 405 nmol/L due to hydrolysis of 250 μmol/L S-2238 for 5 minutes in a microplate reader at 37°C (THERMOmax, Molecular Devices) and expressed as a percentage of hydrolysis of S-2238 in recalcified citrated plasma not incubated with plasmin.
Fibrin Formation as Characterized by Elaboration of FPA
Thrombin activity was characterized by elaboration of fibrin I in plasma and whole blood by measuring the concentration of FPA, a peptide released by thrombin from fibrinogen, with radioimmunoassay (Byk-Sangtek Diagnostica). Plasma was incubated before assay with bentonite to remove cross-reacting fibrinogen and fibrinogen degradation products.
Plasmin-Mediated Cleavage of Factor XII
Amino Acid Sequencing
Purified human factor XII was labeled with 125I by the lactoperoxidase reaction with Enzymobeads according to the method recommended by the manufacturer (Bio-Rad Laboratories). 125I–Factor XII (30 μg/mL) was incubated with 2.4 μmol/L plasminogen and 1000 IU/mL SK or 0.5 μmol/L purified human plasmin in a final volume of 200 μL phosphate-buffered saline (PBS) at 37°C. Aliquots were removed at 1 to 30 minutes and placed into 0.5 mol/L Tris-HCl containing 2% sodium dodecyl sulfate (SDS) in the presence or absence of 2-β-mercaptoethanol and boiled for 5 minutes. Samples were subjected to 10% gradient SDS–polyacrylamide gel electrophoresis (PAGE). Gels were dried for autoradiography or subjected to electrophoretic transfer to PVDF membranes in Tris-glycine (0.025 mol/L Tris, 0.192 mol/L glycine, and 10% methanol [pH 8.3]) buffer. Protein bands were stained with Coomassie blue, isolated, and subjected to N-terminal amino acid sequencing by the Protein Chemistry Laboratory, Department of Biochemistry, Washington University.20
Factor XIIa Activity Assay
Purified human factor XII was incubated with 5 μmol/L PPACK to inhibit potential contamination with factor XIIa present in the preparation and was subjected to dialysis in 1500 vol of Tris-NaCl (50 mmol/L Tris, 12 mmol/L NaCl, pH 7.8). Complete inhibition of factor XIIa in the preparation was confirmed by lack of hydrolysis of S-2302, even after a 30-minute incubation. Factor XII (100 nmol/L) was added to Tris-NaCl in the presence or absence of 1 μg/mL dextran sulfate in a polystyrene microtiter plate (final volume 200 μL). The mixture was incubated with plasmin (0 to 2 μmol/L) at 37°C for 10 minutes. Plasmin was inhibited with 1000 KIU/mL aprotinin, and 200 μmol/L S-2302 was added. Inhibition of plasmin was confirmed by lack of hydrolysis of 200 μmol/L S-2251 after addition of the aprotinin. Change in absorbance at 405 nmol/L was measured for 5 minutes in a microplate reader at 37°C. Factor XIIa concentration was determined by comparison with the rate of hydrolysis of S-2302 produced by 0 to 100 nmol/L factor XIIa.
Plasmin-Mediated Cleavage of Prekallikrein
Purified human prekallikrein was labeled with 125I by the lactoperoxidase reaction, as described. 125I-Prekallikrein (0.4 μmol/L) was incubated with 0.5 μmol/L purified human plasmin, and aliquots were removed at 1 to 60 minutes, placed into 0.5 mol/L Tris-HCl (pH 6.8) containing 2% SDS in the presence or absence of 2-β-mercaptoethanol, and boiled for 5 minutes. Samples were subjected to 4% to 15% Tris-gradient PAGE, followed by autoradiography.
Purified human prekallikrein (100 nmol/L) was incubated with purified human plasmin (0 to 1 μmol/L) in Tris-NaCl buffer (50 mmol/L Tris and 12 mmol/L NaCl [pH 7.8]) in a polystyrene microtiter plate at 37°C for 60 minutes. α2-Antiplasmin (2 μmol/L) was added to inhibit plasmin, and inhibition of the plasmin was confirmed by lack of hydrolysis of S-2251. S-2302 (200 μmol/L) was added, and change in absorbance at 405 nmol/L due to kallikrein-mediated hydrolysis of S-2302 was measured in a microplate reader for 5 minutes at 37°C. Kallikrein concentration was determined by comparison with the rate of hydrolysis of S-2302 produced by 0 to 50 nmol/L kallikrein.
Characterization of Role of Factor XII in Plasmin-Mediated Elaboration of Thrombin Activity
Incubations in Plasma
Pooled citrated plasma, citrated factor XII–deficient plasma, or citrated factor XII–deficient plasma to which 1.25 μmol/L purified human factor XII had been added was placed into polypropylene tubes containing CaCl2 (25 mmol/L final concentration) and incubated with or without 0.5 μmol/L purified plasmin in the presence or absence of 3.5 μmol/L corn trypsin inhibitor. Aliquots (200 μL) were removed at 1, 3, 5, and 7 minutes and placed into polypropylene tubes containing an anticoagulant mixture composed of 5 mmol/L EDTA, 1000 KIU/mL aprotinin, and 20 μmol/L PPACK to inhibit plasmin, kallikrein, and thrombin activity.21 The concentration of FPA in the sample was determined by radioimmunoassay, as described.
Incubations in Nonanticoagulated Whole Blood
Blood samples were collected from healthy volunteers, after they gave informed consent, by application of a light tourniquet and venipuncture of the antecubital vein with a 19-gauge needle. The first 2 mL of blood was discarded, and blood (5 mL) was allowed to drip into polypropylene tubes containing either 5 μg/mL rTPA, 5 μg/mL rTPA, and 3.5 μmol/L corn trypsin inhibitor; 5 μg/mL rTPA and 1 IU/mL heparin sodium; or 0.15 mol/L NaCl (pH 7.4) adjusted to a volume equal to RTPA. Aliquots (1 mL) were removed at 1, 3, 5, and 7 minutes and placed into polypropylene tubes containing an anticoagulant mixture composed of 5 mmol/L EDTA, 1000 KIU/mL aprotinin, and 20 μmol/L PPACK to inhibit plasmin, kallikrein, and thrombin activity. The samples were then centrifuged, and the plasma was removed for determination of the concentration of FPA by radioimmunoassay.
Ligand Blotting of High Molecular Weight Kininogen With 125I-Labeled Prekallikrein
In Vitro Studies
Contact activation in plasma samples was characterized by ligand blotting of high molecular weight kininogen with 125I-labeled prekallikrein by use of a modified procedure of Lammle et al.22 Kallikrein-mediated cleavage of high molecular weight kininogen releases an approximately 10 000-kDa fragment containing bradykinin from the heavy chain, resulting in a shift in apparent molecular weight that is characterized on nonreduced autoradiograms. Pooled recalcified citrated plasma or high molecular weight kininogen-deficient plasma (200 μL) was incubated with 100 or 500 IU/mL SK for 30 minutes in polypropylene tubes at 37°C. A 3-μL aliquot was removed and placed into 0.5 mol/L Tris-HCl containing 2% SDS and boiled for 5 minutes. Samples were subjected to electrophoresis on 4% to 15% gradient Tris-HCl polyacrylamide gels, followed by electrophoretic transfer to PVDF membranes in Tris-glycine buffer containing 10% methanol. After electrotransfer, the membranes were soaked in a solution of 5% dry milk and 1% bovine serum albumin (BSA) for 2 hours. Prekallikrein was labeled with 125I by the lactoperoxidase reaction, as described. The membranes were incubated with 125I-prekallikrein (500 000 cpm/mL) in PBS, 1% BSA for 4 hours at room temperature, followed by extensive washing with PBS and exposure to XAR film (Kodak) for 5 to 120 hours.
Plasma Samples From Patients Treated With Fibrinolytic Agents
Contact activation was characterized in samples from patients before and after treatment with SK or rTPA for acute myocardial infarction. An aliquot of plasma from samples collected as part of a protocol approved by the Human Studies Committee at Washington University or at Ospedale G.B. Morgagni-L. Pierantoni from patients treated for acute myocardial infarction with 1.5 million units of SK or 100 mg rTPA was placed into a polypropylene tube containing an anticoagulant mixture of 5 mmol/L EDTA, 1000 KIU/mL aprotinin, and 20 μmol/L PPACK to inhibit plasmin, kallikrein, and thrombin activity. A 3-μL aliquot of plasma was subjected to electrophoresis, electrotransferred to PVDF membranes, and ligand blotted with 125I-prekallikrein as described above.
Data are expressed as mean±SEM. Paired or unpaired Student’s t test was performed where applicable. A significant difference was assumed when P<.05.
Plasmin-Mediated Increases in Thrombin Activity in Pooled Plasma
To characterize plasmin-mediated increases in thrombin activity in plasma, we incubated 0.05 to 1 μmol/L purified human plasmin (concentrations consistent with pharmacological activation of plasminogen23 ) in polystyrene microtiter plates at 37°C with recalcified (25 mmol/L) pooled citrated plasma for 3 minutes. Thrombin activity was determined by measuring the rate of hydrolysis of a synthetic substrate for thrombin (S-2238) over 5 minutes and was expressed as a percentage of the rate of hydrolysis in recalcified citrated plasma alone. As shown in Fig 1⇓ (Inset), incubation of recalcified citrated plasma with plasmin induced concentration-dependent increases in thrombin activity in pooled recalcified citrated plasma. Surprisingly, incubation of recalcified citrated plasma with 0.20 μmol/L plasmin for 3 minutes resulted in a marked increase in thrombin activity, which was 1000-fold higher than that in plasma not incubated with plasmin and equal to that induced by 17 nmol/L of purified human thrombin incubated with S-2238 in 0.05 mol/L Tris-HCl and 0.175 mol/L NaCl (pH 7.8) at 37°C. These concentrations of plasmin, except those near 1 μmol/L, were rapidly inhibited by α2-antiplasmin in the plasma, as judged by the lack of hydrolysis of S-2251 by plasmin that had been incubated with recalcified citrated pooled plasma for 1 minute at 37°C. Thus, plasmin-mediated activation of the coagulation system occurs rapidly in response to concentrations of plasmin that do not deplete α2-antiplasmin, the natural inhibitor of plasmin in plasma, suggesting that plasmin-mediated activation of the coagulation system is a favored reaction. Hydrolysis of S-2238 was markedly attenuated when incubations were performed in the presence of 0.5 μmol/L recombinant hirudin, confirming that hydrolysis of the substrate was attributable to thrombin activity (Fig 1⇓).
Effect of Specific Coagulation Factor Deficiency on Plasmin-Mediated Increases in Thrombin Activity
To determine the specific coagulation factors involved in plasmin-mediated activation of the coagulation system, we incubated 0.16 μmol/L plasmin with recalcified citrated pooled plasma or recalcified plasma from patients deficient in contact system factors XII or XI, prekallikrein, or high molecular weight kininogen. As shown in Fig 1⇑, increases in thrombin activity were markedly attenuated when plasmin was incubated with plasma deficient in factors XI or XII, prekallikrein, or high molecular weight kininogen compared with thrombin activity in recalcified citrated pooled plasma incubated with plasmin. The attenuation of increases in thrombin activity in association with factor XII deficiency, despite the presence of normal concentrations of factor XI, prekallikrein, and high molecular weight kininogen in factor XII–deficient plasma, suggests that plasmin-mediated activation of factor XII is important to plasmin-mediated activation of the coagulation system.
Incubation of recalcified citrated pooled plasma with 0.5 μmol/L plasmin for 7 minutes induced marked increases in the concentration of FPA, a marker of fibrin-I formation, consistent with plasmin-mediated activation of the coagulation system (Table⇓). Increases in FPA did not occur in plasma incubated in the presence of 0.5 μmol/L recombinant hirudin, confirming that the increases in FPA were attributable to thrombin activity rather than plasmin-mediated release of FPA from fibrinogen. Increases in FPA were markedly attenuated when recalcified citrated plasma was incubated with 0.5 μmol/L plasmin in the presence of 3.5 μmol/L corn trypsin inhibitor, a specific inhibitor of factor XIIa, or when factor XII–deficient plasma was substituted for pooled plasma. Increases in thrombin activity in recalcified citrated plasma, as measured by hydrolysis of S-2238, were also attenuated in the presence of corn trypsin inhibitor (data not shown). Repletion of factor XII–deficient plasma with 1.25 μmol/L purified human factor XII resulted in a 10-fold increase in the concentration of FPA induced by incubation with plasmin compared with factor XII–deficient plasma incubated with plasmin (Table⇓). The increase in thrombin activity was less than that observed in recalcified citrated pooled plasma incubated with plasmin; however, the increases in FPA observed in individual patient plasmas have demonstrated marked variability compared with increases in FPA in plasma pooled from several donors.
Plasmin-Mediated Activation of Purified Factor XII
To confirm previous findings indicating the potential for plasmin to activate factor XII in buffer,15 100 nmol/L purified human factor XII was incubated with 0 to 2 μmol/L plasmin in Tris-NaCl for 10 minutes at 37°C. Activation of factor XII was characterized by hydrolysis of the synthetic substrate S-2302 after inhibition of the plasmin with aprotinin. Incubation of purified factor XII with plasmin induced concentration-dependent increases in factor XIIa activity (Fig 2⇓), similar to the results of Griffin.15 To confirm that a negatively charged surface accelerates plasmin-mediated activation of factor XII, purified factor XII was incubated with 0 to 2 μmol/L plasmin in the presence of 1 μg/mL dextran sulfate for 10 minutes, and the rate of factor XIIa–mediated hydrolysis of S-2302 was determined. As expected, the presence of dextran sulfate markedly increased the activation of factor XII by plasmin (Fig 2⇓). A negatively charged surface also accelerated apparent plasmin-mediated activation of the contact system in plasma. Measurement of thrombin activity by hydrolysis of S-2238 in recalcified pooled citrated plasma containing dextran sulfate incubated with 0.16 μmol/L plasmin and in recalcified pooled citrated plasma incubated with plasmin in the absence of dextran sulfate revealed a twofold increase in activity in the former (56.3±1.6 versus 25.3±0.4 mOD/min, P<.01).
To characterize plasmin-mediated activation of factor XII, the N-terminal amino acid sequences of factor XII intermediates elaborated during incubation with plasmin were determined. Purified human factor XII was incubated with 0.5 μmol/L purified plasmin or with 1000 IU/mL SK and 2.4 μmol/L purified human plasminogen for 30 minutes. Incubations with SK-plasminogen resulted in more extensive proteolysis of factor XII, but the apparent molecular weights of the factor XII intermediates elaborated were the same as those observed after incubation with plasmin. Accordingly, factor XII intermediates resulting from the SK-plasminogen incubation were characterized further by subjecting the mixture to electrophoresis under reducing conditions and transferring the proteins to PVDF membranes. A protein band with an apparent molecular weight of 28 000 kDa, identified by staining with Coomassie blue and consistent with the molecular weight of the light chain of activated factor XII, was isolated and subjected to N-terminal amino acid sequencing to determine the initial 12 N-terminal amino acids. Equimolar quantities of two fragments were present that corresponded to cleavage of factor XII at Arg353-Val354, the site involved in kallikrein-mediated activation of factor XII, and cleavage at Lys346-Ser347, an apparently novel site of plasmin-mediated hydrolysis of factor XII.
Plasmin-Mediated Activation of Prekallikrein
To determine whether plasmin also cleaves prekallikrein to kallikrein, another potential mechanism for plasmin-mediated contact system activation in plasma, 0.4 μmol/L 125I-labeled human prekallikrein was incubated with 0.5 μmol/L human plasmin for 60 minutes. Analysis of the incubation mixture by 4% to 15% Tris gradient PAGE and autoradiography failed to demonstrate significant plasmin-mediated cleavage of prekallikrein (data not shown). In addition, incubation of 100 nmol/L purified human prekallikrein with 0 to 1 μmol/L purified human plasmin for 30 minutes did not result in elaboration of kallikrein activity, as assessed by hydrolysis of S-2302, a synthetic substrate for kallikrein.
Characterization of Fibrin Elaboration in Nonanticoagulated Whole Blood in Response to Pharmacological Plasminogen Activation
To confirm that pharmacological activation of plasminogen also induces increases in fibrin elaboration in blood, 5 μg/mL rTPA was incubated with freshly collected nonanticoagulated whole blood in polypropylene tubes, and the concentration of FPA was determined over 7 minutes. This concentration of rTPA was selected to be similar to that in patients undergoing coronary thrombolysis and to be consistent with that used in previous studies.10 Increases in FPA similar to those documented in recalcified citrated plasma were observed during a 7-minute incubation with rTPA (Fig 3⇓). The increases in thrombin activity were markedly attenuated when whole blood was incubated with rTPA in the presence of 3.5 μmol/L corn trypsin inhibitor, consistent with the critical role of factor XII in plasmin-mediated activation of the coagulation system in whole blood as demonstrated in plasma.
Contact Activation in Response to Plasminogen Activation as Characterized by Cleavage of High Molecular Weight Kininogen
To determine whether activation of the contact system occurs in patients treated with plasminogen activators for acute myocardial infarction, we characterized kallikrein-mediated conversion of high molecular weight kininogen from the one-chain to the two-chain form. Based on the results in vitro, conversion of high molecular weight kininogen in vivo is most likely attributable to factor XIIa–mediated cleavage of prekallikrein to kallikrein and subsequent kallikrein activity because plasmin does not appear to activate prekallikrein directly. The circulating forms of high molecular weight kininogen were characterized by ligand blotting of plasma with 125I-labeled–prekallikrein, which forms a complex with the light chain of high molecular weight kininogen in plasma. Two-chain high molecular weight kininogen migrates under nonreducing conditions at 97 kDa because of the kallikrein-mediated release of the nonapeptide bradykinin, followed by further cleavage of the light chain of high molecular weight kininogen by kallikrein from a 56- to a 47-kDa fragment.24 Incubation of recalcified citrated pooled plasma with 100 IU/mL SK for 15 or 30 minutes induced significant conversion of one-chain high molecular weight kininogen to the two-chain form, as demonstrated by the appearance of a faint band at 97 kDa (Fig 4A⇓, lanes 4 and 5) that comigrates with high molecular weight kininogen cleaved by kallikrein (Fig 4A⇓, lane 2). Incubation of recalcified citrated plasma with 500 IU/mL SK for 15 or 30 minutes induced complete conversion and depletion of one-chain high molecular weight kininogen to the two-chain form (Fig 4A⇓, lanes 6 and 7). Incubation of recalcified citrated plasma with 0.5 μmol/L plasmin also resulted in cleavage of high molecular weight kininogen (data not shown). Conversion of high molecular weight kininogen was more pronounced in plasma in which pharmacological activation of plasminogen was induced by SK than in plasma incubated with purified plasmin. As shown in Fig 4B⇓, conversion of one-chain high molecular weight kininogen to the two-chain form was also observed by ligand blotting with 125I-prekallikrein plasma samples from patients after administration of 1.5 million units of SK (Fig 4B⇓, lanes 4 and 6) or 100 mg rTPA (Fig 4B⇓, lane 8) for acute myocardial infarction. Almost complete conversion and depletion of one-chain high molecular weight kininogen occurred in these patients within 60 minutes of the initiation of SK or rTPA infusion.
The results of the present study indicate that pharmacological activation of plasminogen induces activation of the contact system of coagulation, which may be responsible, at least in part, for increases in thrombin activity observed in response to the administration of fibrinolytic agents. Although these results are consistent with those of previous studies that have shown that plasmin potentiates activation of factors of the contact system, the concentrations of plasmin required in previous studies in purified systems would require activation of all plasminogen present in plasma.14 15 In contrast, we have shown in plasma and blood that concentrations of plasmin consistent with those achieved by pharmacological activation of plasminogen activate the contact system of coagulation in the absence of a negatively charged surface. In fact, even relatively low concentrations of plasmin, insufficient to deplete α2-antiplasmin, induced marked increases in thrombin activity in plasma. Furthermore, we have shown that factor XIIa–dependent increases in thrombin activity occur in nonanticoagulated whole blood incubated with pharmacological concentrations of plasminogen activators. Activation of the contact system also appears to occur in patients treated with SK and rTPA and may account for the marked increases in thrombin activity observed in such patients in response to pharmacological plasminogen activation.
Plasmin has been shown to increase factor XIIa activity when incubated with factor XII and to increase the activity of factor XIIa in purified systems. Kaplan and Austen14 demonstrated plasmin-mediated cleavage of purified factor XIIa, yielding factor XIIa fragments with enhanced ability to activate prekallikrein. Griffin15 demonstrated that factor XIIa activity increased when purified factor XII and plasmin were incubated in buffer. However, as initially demonstrated by Griffin and confirmed in the present study, in the absence of a negatively charged surface, high concentrations of plasmin are necessary.
Plasmin may also potentiate activation of the contact system by limited proteolysis of high molecular weight kininogen, which increases susceptibility of high molecular weight kininogen to kallikrein-mediated cleavage and accelerates the release of bradykinin, as shown by Kleniewski et al.25 Plasmin-mediated hydrolysis of high molecular weight kininogen at sites other than those involved in kallikrein-mediated cleavage results in a two-chain molecule of similar size to that elaborated by kallikrein-mediated conversion of high molecular weight kininogen without release of bradykinin.25 Although we cannot exclude that plasmin cleavage of high molecular weight kininogen was the mechanism of elaboration of the two-chain form observed in patient samples, our results in vitro suggest that plasmin-mediated activation of factor XII and subsequent factor XII–mediated activation of prekallikrein to kallikrein, converting high molecular weight kininogen to its two-chain form, constitutes the most likely mechanism.
Pharmacological activation of plasminogen not only appears to induce activation of the contact system but also may result in depletion of the contact system coagulation proteins. In the purified system, SK-plasminogen induced marked hydrolysis of factor XII. Plasmin-mediated activation and depletion of factor XII may account for recent observations of decreased factor XII–dependent fibrinolytic activity in patients after administration of fibrinolytic agents.16 In addition, in plasma samples from patients treated with SK or rTPA for myocardial infarction, one-chain high molecular weight kininogen was almost completely converted to the two-chain form. Depletion of high molecular weight kininogen may contribute to the decrease in factor XII–dependent fibrinolytic activity as well as inhibit neutrophil activation and influence blood pressure regulation in these patients.
Activated factor XII converts the first component of complement to its active form and may account for observations of complement activation in patients undergoing coronary thrombolysis.26 27 28 Kallikrein-mediated conversion of high molecular weight kininogen and the release of bradykinin may also account for the previously unexplained phenomenon of hypotension in patients treated with fibrinolytic agents.26 29 Recent studies have shown that conjunctive anticoagulation accelerates the rapidity of recanalization and prevents recurrent thrombosis. Inhibition of the activity of factors Va, VIIa, VIIIa, and Xa and of thrombin have been shown to attenuate the increases in thrombin activity induced by administration of fibrinolytic agents. However, the responsible mechanisms have not been well defined. Our results are consistent with those of previous studies, since activation of the contact system would result in increased activity of IXa/VIIIa and Xa/Va complexes, with consequent elaboration of thrombin. However, our data suggest that conjunctive administration of thrombin inhibitors during thrombolysis, although effective in inhibiting thrombin activity, may not attenuate activation of the contact system. If this were the case, thrombin inhibition alone would not attenuate effects secondary to contact system activation, such as hypotension due to bradykinin release from high molecular weight kininogen, depression of factor XII–dependent fibrinolytic activity, and activation of the complement cascade.
Although multiple mechanisms are likely to contribute to increases in thrombin activity during coronary thrombolysis, plasmin-mediated activation of the contact system of coagulation appears to be novel and significant. Additional studies characterizing the effects of factor XII inhibition on the efficacy of fibrinolysis, prevention of hypotension during administration of fibrinolytic agents, and prevention of complement activation will need to be completed. Plasmin-mediated activation of the contact system during pharmacological thrombolysis may also be a model for the further study of the surface-independent activation of the contact system that occurs under other pathophysiological conditions of increased plasmin and coagulant activity, such as in the presence of disseminated intravascular coagulation.30
This work was supported in part by SCOR in Coronary and Vascular Heart Disease (grant HL-17646), National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Md. Dr Ewald is an American Heart Association, Missouri Affiliate fellow. We would like to acknowledge the gift of purified proteins as well as manuscript review and support by Dr Joseph Shore; the collection of plasma samples from patients treated with SK by Drs Marcello Galvani, Filippo Ottani, Donatella Ferrini, and Franco Rusticali; and the expert technical assistance of Mary Jane Eichenseer, Joan Lee, and Matthew Ohlendorf, the editorial assistance of Beth Engeszer, and the manuscript preparation by Ellen Visse.
- Received April 28, 1994.
- Accepted August 4, 1994.
- Copyright © 1995 by American Heart Association
Rapold HJ. Promotion of thrombin activity by thrombolytic therapy without simultaneous anticoagulation. Lancet. 1990;1:481-482.
Cannon CP, McCabe CH, Henry TD, Rogers WJ, Schweiger M, Gibson RS, Anderson JL, Williams DO, Braunwald E. Hirudin reduces reocclusion compared to heparin following thrombolysis in acute myocardial infarction: results of the TIMI 5 trial. J Am Coll Cardiol. 1993;21:136A.
Gulba DC, Westhoff-Bleck M, Jost S, Rafflenbeul W, Daniel WG, Hecker H, Lichtlen PR. Increased thrombin levels during thrombolytic therapy in acute myocardial infarction. Circulation. 1991;83:937-944.
Lee CD, Mann KG. The activation of human coagulation factor V by plasmin. Blood. 1989;73:185-190.
Gjonnaess H. Cold promoted activation of factor VII: V, relation to the fibrinolytic system. Thrombs Diasthes Haemorrh. 1973;29:143-153.
Kaplan AP, Austen KF. A prealbumin activator of prekallikrein. J Exp Med. 1971;133:696-712.
Griffin JH. Role of surface in surface-dependent activation of Hageman factor. Proc Natl Acad Sci U S A. 1978;75:1998-2002.
Castellino FJ, Powell RJ. Human plasminogen. Meth Enzymol. 1980;80:365-378.
Matsudaira P. Sequence from picomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes. J Biol Chem. 1987;262:10035-10038.
Tiefenbrunn AJ, Graor RA, Robison AK, Lucas FV, Hotchkiss A, Sobel BE. Pharmacodynamics of tissue-type plasminogen activator characterized by computer assisted simulation. Circulation. 1986;73:1291-1299.
Winters KJ, Santoro SA, Miletich JP, Eisenberg PR. Relative importance of thrombin compared with plasmin-mediated platelet activation in response to plasminogen activation with streptokinase. Circulation. 1991;84:1552-1560.
Kaplan AP, Silverberg M. The coagulation-kinin pathway of human plasma. Blood. 1987;70:1-15.
Munkvad S, Brandslund I, Gram J, Jespersen J. The complement-system is activated by streptokinase, but not by combined heparin recombinant tissue-type plasminogen-activator therapy in patients with myocardial-infarction—a placebo-controlled study. Cor Art Dis. 1991;2:889-896.
Frangi D, Gardinali M, Conciato L, Cafaro C, Pozzoni L, Agostoni A. Abrupt complement activation and transient neutropenia in patients with acute myocardial infarction treated with streptokinase. Circulation. 1994;89:76-80.
Munkvad S, Gram J, Brandslund I, Jespersen J. Coronary thrombolysis with streptokinase, but not rt-pa, activates complement in vivo—a placebo-controlled study. Thromb Haemost. 1991;65:696.
Gemmill J, Hogg K, Douglas J, Dunn F, Lowe G, Rae A, Hillis W. The incidence and mechanism of hypotension following thrombolytic therapy for acute myocardial infarction with streptokinase-containing agents: lack of relationship to pretreatment streptokinase resistance. Eur Heart J. 1993;14:819-825.