A partially degraded form of human plasminogen in circulating blood

Summary A method for separate measurement of the amounts of the native form and partially degraded form of plasminogen and plasmin in human circulating blood was devised based on their different mobilities on disc polyacrylamide gel electrophoresis. The relative amounts in the circulating blood as estimated by the method were as follows: 78.5% native form, 6.1% partially degraded form and 15.5% plasmin, in the resting state of healthy adults. The partially degraded form increased to about 50% after strenuous exercise load. The present study showed that relatively large amounts of the partially degraded form can be produced in the circulating blood, even though abundant plasmin inhibitors exist there. These data suggest that the partially degraded form may play an important role in thrombolysis in vivo.     

High-dose, brief-duration intravenous infusion of streptokinase in acute myocardial infarction: description of effects in the circulation

The effects in the circulating blood of a 1-hour intravenous infusion of 1.5 X 10(6) units of streptokinase (SK) were measured during the subsequent 24-hour period in 7 patients with acute myocardial infarction. At the end of the infusion, the activator activity, expressed in SK units, averaged 65 U/ml, all of the plasminogen had disappeared, only a small amount of free plasmin was still present and functionally active alpha 2 antiplasmin had been reduced to 21% of the preinfusion level. All of the native fibrinogen had been degraded and the thrombin-coagulable protein was composed entirely of fragment X species, but the circulating plasma also contained significant amounts of the more extensively degraded fragments Y, D and E. The biologic half-life of the SK-induced activator activity was 23 minutes and that of the fibrinogen degradation products was 6.3 hours. The lytic effects persisted for 4 hours before any signs of recovery from the hemostatic defect were evident; considerable recovery was present at 25 hours.     

The reaction between plasmin and C1-inhibitor results in plasmin inhibition by the serpin mechanism.

C1-inhibitor is an important inhibitor of plasma kallikrein and C1, but also has inhibitory activity against numerous other plasma proteinases such as plasmin. The relevance of plasmin inhibition by the C1-inhibitor has been debated, with some evidence showing that plasmin causes significant proteolysis of C1-inhibitor. In the present study, we show that C1-inhibitor in its native state will inhibit plasmin without being significantly degraded, in a manner typical of all serpin reactions. However, if C1-inhibitor is in a denatured polymeric state (as can easily occur during storage, or as produced by heating of the native protein), it will be extensively degraded by plasmin. In addition, we show that hydrophobic interaction chromatography is an effective method to remove trace contaminants of inactive C1-inhibitor polymers.     

Potential mechanisms for the plasmin-mediated release and activation of latent transforming growth factor-beta1 from the extracellular matrix of growth plate chondrocytes

Chondrocytes produce latent transforming growth factor-beta1 (TGF-beta1) in a small, circulating form of 100 kDa and also store latent TGF-beta1 in their matrix in a large form of 290 kDa containing the latent TGF-beta1 binding protein 1. As growth plate cartilage cells are exceptionally sensitive to TGF-beta1 and are known to produce plasminogen activator, the role of plasmin in the activation of soluble and matrix-bound latent TGF-beta1 was examined. As is true for other cell types, low-dose plasmin (0.01 U/ml) was found to release both active and latent TGF-beta1 from chondrocyte matrix in a time-dependent manner over 3 h. However, high-dose plasmin (1.0 U/ml) was found to release active TGF-beta1 more rapidly than low-dose plasmin, and this release ceased within 30 min; latent complex continued to be released over time (3 h). When high-dose plasmin was titrated against the serine protease inhibitors, aprotinin and alpha-(2-aminoethyl)benzenesulfonyl fluoride, results similar to low-dose plasmin were obtained, indicating that the effects of high-dose plasmin could be altered to mimic those of low-dose plasmin. No differences were observed on the effects of plasmin on the release of TGF-beta1 from the matrices of either growth zone or resting zone chondrocytes. We examined whether plasmin could further activate the truncated large latent TGF-beta1 complex of 230 kDa that was released into the media by plasmin. It is known that plasmin will activate the small latent complex, so this was compared with the truncated form. Plasmin completely activated the small latent complex, whereas a smaller, but significant, activation of the truncated form of latent TGF-beta1 also occurred. These studies may have relevance to normal physiological conditions, where plasminogen and/or plasmin is present in very small amounts in the cartilage and, therefore, small amounts of active TGF-beta1 would be present, and to pathological conditions such as fractures, where chondroprogenitor cells would be exposed to high concentrations of plasmin and, therefore, to short-term high concentrations of this potent chondrogenic growth factor.     

An Analysis of Mechanisms Underlying the Antifibrinolytic Properties of Radiographic Contrast Agents

Background: Radiographic contrast agents inhibit fibrinolysis, although by poorly defined pathways. The purpose of this study was to define specific mechanisms by which contrast agents inhibit clot lysis. Methods and Results: Diatrizoate (high osmolar ionic agent), ioxaglate (low osmolar ionic), and ioversol (nonionic) were studied in vitro. Diatrizoate inhibited clot lysis by 81.3±0.6% vs. control (p<0.001). Ioxaglate inhibited clot lysis by 41.7±11.9%, which was of borderline significance (p=0.07). Ioversol did not significantly inhibit clot lysis (14.9±11.5% decrease vs. control; p>0.3). Inhibition of fibrinolysis was not explained by the high osmolarities of contrast agents, by their iodine content, or by their effects on the amidolytic activities of t-PA, urokinase, or plasmin. However, plasminogen activation by t-PA, urokinase, or streptokinase was significantly inhibited by contrast agents. Diatrizoate, ioxaglate, and ioversol inhibited plasminogen binding to plasma clots by 51±4% (p<0.001), 30.1±4% (p<0.01), and 19.4±7% (p=0.07), respectively. Plasma clots formed in the presence of contrast agents were resistant to lysis by plasmin. Diatrizoate produced the most potent effect, inhibiting clot lysis by 40±5.7% (p<0.03). Contrast agents did not inhibit plasminogen binding to fibrin or plasmin-mediated fibrinolysis if they were added after clot formation. Contrast agents altered clot turbidity, an index of fibrin structure, if present during clot formation, but not if added to preformed clots. Contrast agents did not affect plasminogen activator inhibitor-1 or ₂-antiplasmin function. Conclusions: Contrast agents inhibit clot lysis by inhibiting plasminogen activation and by disrupting interactions of plasminogen and plasmin with fibrin by altering fibrin structure. Significant variation in antifibrinolytic properties exists between different contrast agents. Abbreviated Abstract. The purpose of this study was to define specific mechanisms by which contrast agents inhibit clot lysis. In both a purified clot lysis system and a plasma clot lysis system, diatrizoate, an ionic agent, produced the most potent inhibition of fibrinolysis. Contrast agents did not inhibit the active sites of plasminogen activators or plasmin, but did inhibit plasminogen activation. Binding of plasminogen to fibrin and lysis of fibrin by plasmin were inhibited by contrast agents if they were present during clot formation, but not if they were added after clot formation was complete. Contrast agents altered clot turbidity, an index of fibrin structure, if present during clot formation, but not if added to preformed clots. Contrast agents did not affect plasminogen activator inhibitor-1 or ₂-antiplasmin function. The effects of contrast agents on fibrinolytic parameters were not explained by their high osmolarities. These results suggest that contrast agents inhibit clot lysis by inhibiting plasminogen activation and by disrupting interactions of plasminogen and plasmin with fibrin by altering fibrin structure.     

The elastase-mediated pathway of fibrinolysis

Plasmin and elastase degrade fibrin and inhibit the blood coagulation system by degrading key proteins. Elastase can facilitate plasmin expression via an alternative pathway of plasminogen activation. Elastase modifies plasminogen to yield a zymogen that is a better substrate for activators than native plasminogen. Furthermore, elastase inactivates the inhibitor system of plasmin and plasminogen activators without affecting plasmin and plasminogen activators. While plasmin activity develops from a blood zymogen as a consequence of activators synthesized and secreted by endothelium and possibly other cells, elastase is secreted in an active form primarily by polymorphonuclear leukocytes. Plasmin and elastase may play mutual roles in thrombolysis, inflammation, and tumour invasion and metastasis.     

Recurring thromboembolic disease and pulmonary hypertension associated with severe hypoplasminogenemia

In a patient with pulmonary hypertension and a history of recurrent venous thrombosis, plasma concentrations of all known coagulant and inhibitor proteins were normal except for severe deficiency of plasminogen. Repeated analyses showed the circulating plasma plasminogen level to be 30% of normal by either functional or immunologic methods. We sought evidence for either increased activation of plasminogen or for dysplasminogen. There was no evidence for the former. Purified plasminogen studies disclosed a normal number of active sites and normal activation. Generated plasmin had normal catalytic activity. Isoelectric focusing disclosed normal distribution of isoforms. Affinity chromatography with lysine-sepharose showed the presence of the two variant forms; however, an increased proportion of the protein eluted in the first peak. Danazol administration induced an increase in circulating plasminogen, but the differences in affinity chromatography elution profile remained. We conclude that this patient has a deficiency of normally functioning plasminogen, probably due to decreased synthesis.     

Critical role for conversion of glu-plasminogen to Lys-plasminogen for optimal stimulation of plasminogen activation on cell surfaces

When Glu-plasminogen, the native circulating form of the zymogen, is bound to cell surfaces, its activation is markedly enhanced compared with the reaction in solution. This results in localization of the broad-spectrum proteolytic activity of plasmin on cell surfaces. The cell-associated plasmin plays a key role in fibrinolysis, cell migration, and prohormone processing. It is well established that the localization of plasminogen and plasminogen activators on cell surfaces promotes the enhanced plasminogen activation on the cell surface. The focus of this article is to review recent studies demonstrating that the conversion of Glu-plasminogen to the more readily activated Lys-plasminogen derivative is necessary for optimal stimulation of plasminogen activation on the cell surface, and that the interaction of Glu-plasminogen with cells serves to increase processing of Glu-plasminogen to Lys-plasminogen, thereby enhancing plasminogen activation on the cell surface.     

Plasmin generation and fibrin(ogen)olysis following desmopressin infusion.

Desmopressin acetate (DDAVP) is known to stimulate the release of tissue-type plasminogen activator (t-PA) from endothelial cells, but it is unclear whether the increased t-PA actually elicits the plasmin generation and fibrin(ogen)olysis in the circulating blood. We measured plasma levels of plasmin-alpha 2-plasmin inhibitor complex, fibrinogen degradation products (FgDP) and fibrin degradation products (FbDP) following desmopressin infusion in 19 patients with bleeding disorders or thrombophilia. Administration of desmopressin (0.3-0.4 microgram/kg) produced a 4.0-fold increase in plasmin-alpha 2-plasmin inhibitor complex at 30 min, whereas neither FgDP nor FbDP was elevated significantly. These findings indicate that desmopressin infusion provokes the generation of plasmin in vivo, but most of the plasmin generated is complexed to alpha 2-plasmin inhibitor and does not degradate fibrin or fibrinogen.     

Monoclonal antibodies against receptor-induced binding sites detect cell-bound plasminogen in blood

Binding of Glu-plasminogen (the native, circulating form of the zymogen) to cells results in enhancement of its activation. Cell-associated plasmin proteolytic activity is a key component of physiologic and pathologic processes requiring extracellular matrix degradation. Recently, we developed antiplasminogen mAbs that recognize receptor-induced binding sites (RIBS) in Glu-plasminogen and, therefore, preferentially react with cell-associated Glu-plasminogen in the presence of soluble Glu-plasminogen. Here we have used FACS with a representative antiplasminogen receptor-induced binding site mAb, mAb49, to examine whether plasminogen associates with peripheral blood cells in blood. Plasminogen binding to neutrophils, monocytes, B-lymphocytes, T-lymphocytes, and platelets was clearly detected. Treatment of whole blood with lipopolysaccharide or 12-0 tetradecanoylphorbol-13-acetate up-regulated plasminogen binding to neutrophils and in vivo treatment with all-trans retinoic acid decreased plasminogen binding to acute promyelocytic leukemia blasts. Our results demonstrate that mAb49 can be used to monitor cell-bound plasminogen in blood under both normal and pathologic conditions.     

Isolation and characterization of fragments of reduced and S-carbamidomethylated human growth hormone produced by plasmin digestion. I. Chemistry.

Reduced and S-carbamidomethylated human GH (RCAM-hGH) was digested with human plasmin, yielding a mixture of products. These were partially separated by chromatography on DEAE-cellulose, yielding three major fractions: Da and Db, which were equipotent with native hGH in the weight gain test; and Dc, which was about half as active as hGH. Each of these was further purified by gel filtration, yielding a number of subfractions which were characterized as follows: Da1 is a very stable noncovalent complex of residues 1--134 and 141--191 of RCAM-hGH; Da2 represents residues 20--41; Db1 is very similar to Da1, but appears to have lost one or more amide groups; Db3 represents residues 95--134; Dc2 is a heterogenous fraction containing a further deamidated version of Da1 and Db1 plus a similar complex of residues 42--134 and 141--191, apparently with some carboxyterminal heterogeneity; Dc3, like Db3, represents residues 95--134. The biological activities of these fragments are discussed in the accompanying paper. Earlier work has shown that native hGH, upon digestion with plasmin, is cleaved primarily at residues 134 and 140. It is shown here that when RCAM-hGH is digested with plasmin, in about 87% of the molecules at least one cleavage takes place in addition to those at residues 134 and 140.     

Inhibition of cell surface mediated plasminogen activation by a monoclonal antibody against alpha-Enolase

Localization of plasmin activity on leukocyte surfaces plays a critical role in fibrinolysis as well as in pathological and physiological processes in which cells must degrade the extracellular matrix in order to migrate. The binding of plasminogen to leukocytic cell lines induces a 30- to 80-fold increase in the rate of plasminogen activation by tissue-type (tPA) and urokinase-type (uPA) plasminogen activators. In the present study we have examined the role of alpha-enolase in plasminogen activation on the cell surface. We produced and characterized a monoclonal antibody (MAb) 11G1 against purified alpha-enolase, which abrogated about 90% of cell-dependent plasminogen activation by either uPA or tPA on leukocytoid cell lines of different lineages: B-lymphocytic, T-lymphocytic, granulocytic, and monocytic cells. In addition, MAb 11G1 also blocked enhancement of plasmin formation by peripheral blood neutrophils and monocytes. In contrast, MAb 11G1 did not affect plasmin generation in the presence of fibrin, indicating that this antibody did not interact with fibrinolytic components in the absence of cells. These data suggest that, although leukocytic cells display several molecules that bind plasminogen, alpha-enolase is responsible for the majority of the promotion of plasminogen activation on the surfaces of leukocytic cells.     

Primary Dysbetalipoproteinemia: Predominance of a Specific Apoprotein Species in Triglyceride-Rich Lipoproteins

Lipoproteins of very low density that are unusually rich in cholesteryl esters accumulate in blood plasma in a characteristic primary form of human hyperlipoproteinemia. These lipoproteins, which are thought to be products of the initial catabolic step in the metabolism of normal triglyceride-rich lipoproteins, have beta rather than pre-beta mobility on electrophoresis, presumably because they have lost certain protein components from their surface. In this study, we have used polyacrylamide gel electrophoresis of apoprotein components that are soluble in tetramethylurea to show that the very low-density lipoprotein fraction of blood serum from seven patients with hyperlipoproteinemia contains unusually large amounts of an arginine-rich protein. Pre-beta migrating, very low-density lipoproteins separated from serum of post-absorptive patients and chylomicrons obtained after a fat-rich meal contain normal amounts of this arginine-rich protein, but beta-migrating, very low-density lipoproteins and chylomicron-like particles separated from serum of post-absorptive patients contain more than twice as much. These apparently partially degraded lipoproteins also contain more tetramethylurea-insoluble protein and smaller amounts of the other soluble protein components than their normal counterparts.     

The receptor for urokinase plasminogen activator is present in plasma from healthy donors and elevated in patients with paroxysmal nocturnal haemoglobinuria

The urokinase plasminogen activator (uPA) is a proteolytic enzyme which converts the proenzyme plasminogen to the active serine protease plasmin. A cell surface receptor for uPA (uPAR) is attached to the cell membrane by a glycosyl-phosphatidylinositol anchor. Binding of uPA to uPAR leads to an enhanced plasmin formation and thereby an amplification of pericellular proteolysis. We have shown previously that uPAR is expressed on normal blood monocytes and granulocytes, but is deficient on affected blood monocytes and granulocytes in patients with paroxysmal nocturnal haemoglobinuria (PNH), and that uPAR is present in plasma from these patients. In this study a newly established sensitive enzyme-linked immunosorbent assay (ELISA) has been applied for quantttation of uPAR in plasma. Unexpectedly, we found that uPAR is not only present in PNH plasma but also in plasma from healthy individuals. In 39 healthy individuals the mean plasma-uPAR value ±SD was 31 ± 15 p m , median 28 (range 11-108), and the corresponding value for six PNH patients was 116±67 p m , median 90 (range 61-228). The elevated uPAR-level in PNH patients was highly significant (Mann-Whitney test; P < 0.0001), and may possibly contribute to the propensity for thrombosis in PNH by inhibition of the fibrinolytic system. Binding of pro-uPA by uPAR in plasma may interfere with the appropriate binding of pro-uPA to cell-bound uPAR and therefore inhibit cell-associated plasmin generation and fibrinolysis. It is likely that the uPAR in normal plasma reflects the overall level of activity of the uPAR-mediated cell surface proteolysis. The present ELISA may be used for studies of uPAR levels in plasma from patients with conditions in which this activity might be increased, such as cancer and inflammatory disorders. Future studies will determine if uPAR in plasma is a parameter of clinical importance in these diseases.     

A Solid-Phase Radioassay for the Quantitative Determination of Antiplasmin Activity in Human Plasma

The caseinolytic activity of one CTA (Committe on Thrombolytic Agents) unit of human plasmin is inhibited by a series of plasma dilutions containing antiplasmin. Then neutralization of the standard plasmin by increasing amounts of antiplasmin shows a steeper linear decrease of plasmin activity betwwen 1.0 and 0.5 CTA units and a much smaller further inactivation below 0.5 CTA units. It is thought that the standard plasmin is partially damaged at the antiplasmin combining site during the purification procedrue and might be responsible for the differences in plasmin-antiplasmin neutralization in the standard curve. Using the steeper slope of the plasmin neutralization curve, an average of 8.6 +/- 1.0 CTA units plasmin neutralizing activity per ml human plasma was found in 36 healthy donors. The difficulty of obtaining 'native' standard plasmin with full antiplasmin combining capacity represents the main problem of a reproducible reliable antiplasmin assay.     

HYPOFIBRINOGENEMIA DUE TO INCREASED FIBRINOLYSIS IN TWO PATIENTS WITH ACUTE PROMYELOCYTIC LEUKEMIA

Two patients with acute promyelocytic leukemia and severe bleeding associated with hypofibrinogenemia were studied. The markedly shortened whole blood clot lysis time and dilute clot lysis time suggested that the defect was an increase in fibrinolysis. Although disseminated intravascular coagulation could not be totally excluded as an alternative mechanism, excessive fibrinolysis was confirmed as the pathogenic cause by the prompt response to the administration of tranexamic acid. The low circulating plasminogen, α₂ plasmin inhibitor level and the presence of α₂ plasmin inhibitor-protease complex in both patients suggested that the increased fibrinolysis probably resulted from the liberation of plasminogen activator from the promyelocyte.     

The molecular nature of circulating growth hormone in normal and acromegalic man: evidence for a principal and minor monomeric forms

Human GH (hGH) extracted from pituitary glands consists of several molecular forms. Monomeric pituitary forms include the single chain 22,000-dalton polypeptide (22K; hGH-B), a 20,000-dalton variant with a 15-amino acid deletion (20K), 3 proteolytically cleaved 2-chain forms (hGH-C, -D, and -E), 2 deamidated forms, an acetylated form (fast hGH), and other, only partially characterized forms. It is not known which of these forms is secreted, nor what the precise nature of circulating hGH is. To answer these questions, we have extracted hGH from human plasma obtained by plasmapheresis from normal volunteers after L-dopa stimulation of hGH secretion and from acromegalic patients. We extracted and concentrated hGH by immunoadsorbent chromatography and examined its chemical nature by polyacrylamide gel electrophoresis under native and denaturing (sodium dodecyl sulfate and urea), nonreducing and reducing (dithiothreitol) conditions as well as by isoelectric focusing. In all cases, the predominant form of hGH present in plasma was 22K, which accounted for approximately 85% of all immunoreactive hGH. In addition, we found evidence for the presence of 20K as a minor form (approximately 7% of all hGH) and of one or more acidic forms (N-acetylated, deamidated, or cleaved hGH; 5-10% of all hGH). Exact identification of the acidic form(s) was not possible. However, the highly bioactive cleaved forms hGH-D and -E were judged to be undetectable (less than 5% of all hGH). We conclude that 1) several monomeric molecular forms of hGH circulate in normal and acromegalic man; 2) the pattern of circulating hGH forms reflects in part their relative prevalence in the pituitary gland; 3) proteolytically cleaved 2-chain hGH forms with enhanced bioactivity are not detectable in blood; and 4) monomeric hGH circulating in acromegaly is qualitatively indistinguishable from normal hGH.     

Quantitation of the Reticuloendothelial System Clearance of Soluble Fibrin

Phagocytosis of particulate fibrin has been previously established as a function of the reticuloendothelial system (RES). More recently, reticuloendothelial cells have been shown to bind soluble fibrin/fibrinogen (f/F) complexes in vitro. To quantitate RES clearance vs microthrombus formation, varying doses (0.1-6 mg/kg) of 125I-soluble f/F was injected into rabbits. One hour later the animals were killed, at which time 48 +/- 8% of the 125I f/F had been cleared from the blood. Homogenized organ samples were separated into insoluble pellet, soluble protein bound, and free 125I. Treatment of other samples with plasmin prior to homogenization differentiated the insoluble 125I into RES cleared (intracellular-plasmin resistant) vs microthrombi (plasmin sensitive pellet 125I). 125I-f/F was chiefly found in liver and spleen. Injection of low f/F concentrations resulted in no plasmin sensitive pellet 125I. 3 mg/kg f/F caused small, variable amounts of plasmin sensitive pellet 125I, chiefly in the kidney. With 6 mg/kg, 21-50% of the insoluble 125I in all organs was plasmin sensitive, and occasional 1-2 mm thrombi were found. The data indicate complete and rapid RES clearance of small amounts of soluble f/F from the blood, without microthrombi being formed. The RES was acutely saturated at 1.5-3.0 mg f/F/kg, which is equivalent to immediate conversion to fibrin of 1-2% of the intravascular fibrinogen pool.     

Plasmin-mediated proteolysis of casein in bovine milk.

Plasminogen was found to be present in bovine milk by crossreactivity between rabbit antiserum to plasminogen and casein prepared from milk by acid precipitation. This result was further supported by recovery of intact 125I-labeled plasminogen from rabbit milk after its intravenous injection. Freshly isolated whole bovine casein was observed to undergo slow autoproteolysis at 37 degrees C. Polyacrylamide gel electrophoresis revealed gradual disappearance of major caseins accompanied by appearance and increase in intensity of numerous electrophoretic bands. This autoproteolysis was inhibited by low concentrations of epsilon-aminocaproic acid (0.1 mM) and diisopropyl fluorophosphate (1 mM); catalytic amounts of urokinase accelerated the process. Autoproteolysis of isolated bovine beta-casein was shown by both urea and sodium dodecyl sulfate gel electrophoresis to result in formation of gamma 1- and gamma 2-caseins. Similar electrophoretic bands were formed when beta-casein was degraded by plasmin prepared from bovine blood serum. These results support the hypothesis that bovine plasmin occurs in milk and is identical to alkaline milk protease.