Search

An attempt will be made in this article to classify known abnormal Plgs based only on studies of the isolated highly purified proteins

A new abnormal plasminogen, Frankfurt I, has been identified in the plasma of a 42 year-old male patient who had recurring thromboses, thrombophlebitis and pulmonary embolism since his age of 29. Reduced functional and also slightly reduced antigen plasminogen concentrations were found in both the proposituts and his mother. Plasmin generation rates carried out by Streptokinase and Urokinase were also abnormal. The plasmin generated was very unstable in the absence of stabilizing ligands and/or substrates. Crossed immunoelectrophoresis of the purified Frankfurt I revealed a peak with normal size and shape, but displaced with respect to normal Glu-plasminogen toward the anode. Isoelectric focusing followed by zymography on an agarose-fibrin plate proved this observation but did not indicate a separation of the normal from the abnormal plasminogen molecular species, also, fewer bands were found in the abnormal plasminogen isozyme pattern. Kinetic studies of Frankfurt I Glu-plasminogen and plasmin showed that most of the functional abnormality is related to absence of active sites in half of the molecules.

A comparison was made of the esterase and activator activities of the various activated forms of human plasminogen and their streptokinase complexes with Nalpha-Cbz-L-lysine-p-nitrophenyl ester as the substrate. The steady state kinetic properties of Glu- and Lys-plasmins, and Glu- and Lys-plasminogen-streptokinase complexes were identical, while the Lys-plasmin-streptokinase complex showed a 2-fold increase in Km with the same kcat and a 3-fold increase in Ki for the competitive inhibitor leupeptin. Lys-plasminogen (zymogen with an active site) was prepared which incorporated 0.7 mol of [3H]idisopropyl phosphorofluoridate and 0.43 mol of p-nitrophenyl-p'-guanidinobenzoate/mol of protein. The Km for Lys-plasminogen was 3-fold higher than that of Lys-plasmin, and its maximum velocity 10-fold lower. The steady state kinetic parameters of a plasmin-derived light (B) chain (CmCys)3, and a derived equimolar light (B) chain-streptokinase complex (CmCys)3, isolated from human plasmin and equimolar plasmin-streptokinase, or plasminogen-streptokinase, complexes, respectively, were determined. When the light (B) chain-streptokinase complex is isolated from its parent complexes, there is a complete retention of the original parent's esterase activities, with respect to Km and kcat, and interaction with the competitive inhibitors benzamidine and leupeptin. The plasmin-derived light (B) chain does not retain its parent esterase activities. This chain has very similar kinetic properties to Lys-plasminogen except that streptokinase, in an equal molar amount, does not impart full esterase activity to the light (B) chain whereas the zymogen can be completely activated by streptokinase. The kcat of the plasmin-derived light (B) chain, and its streptokinase complex can be enhanced by 50 and 30%, respectively, in the presence of 10(-4) M leupeptin, a competitive inhibitor of plasmin, attesting to the increased structural flexibility within the active site of this enzyme species. Urokinase hydrolyzes Nalpha-Cbz-L-lysine p-nitrophenyl ester efficiently with a kcat/Km of one-third that of plasmin. The human plasminogen activator activities of various activated forms of human plasminogen and their equimolar streptokinase complexes were compared in a kinetic assay. The Lys-plasmin-streptokinase complex, and streptokinase were the least active of the activator species and were approximately equal in their activator activities. Glu- and Lys-plasminogen-streptokinase complexes had approximately 1.5 times the activity of streptokinase, whereas the equimolar light (B) chain-streptokinase complexes had approximately 2- to 3-times the activator activity of streptokinase. Since the esterase activity remained unchanged, this indicates a greater degree of specificity in the active site of the equimolar light (B) chain-streptokinase activator complex. Urokinase proved to be a poor activator species...

Comparative physical and chemical data are described for the human NH2-terminal Glu-plasminogen and Lys-plasminogen forms in order to determine the exact relationship between these two types of the zymogen. The molecular weights of Glu-plasminogen and Lys-plasminogen were similar and were determined to be 83, 800 plus or minus 4, 500 and 82, 400 plus or minus 3, 300, respectively, by sedimentation equilibrium methods. The molecular weights were identical in dodecyl sulfate solutions, approximately 83, 000, by sedimentation equilibrium methods. The sedimentation coefficients, s-020, w of Glu-plasminogen and Lys-plasminogen were determined to be 5.0 S, and 4.4 S, respectively. These two plasminogen forms had different partial specific volumes, and calculations of the frictional coefficients from sedimentation coefficients and molecular weights indicated conformation differences. Glu-plasminogen appeared to be larger in size than Lys-plasminogen in acrylamide gel-dodecyl sulfate electrophoresis. The amino acid compositions of Glu-plasminogen and Lys-plasminogen, and their major isolated isoelectric forms, were found to be similar, but several amino acid residues (glutamic acid, alanine, isoleucine, phenylalanine, and lysine) were found to be significantly higher in the Glu-plasminogen forms. The derived plasmins from both the Glu- and Lys-plasminogens with an nh2-terminal Lys- heavy (A) chain were found to have identical molecular weights of 76, 500 plus or minus 2, 500, and sedimentation coefficients, s-020, w of 4.3 S.

SummaryA new abnormal plasminogen, Frankfurt I, has been identified in the plasma of a 42 year-old male patient who has recurrent deep vein thrombosis. Clinical laboratory data showed normal hemostasis test results. Since plasma plasmin generation rates gave low values, the fibrinolytic system was analyzed for a possible fibrinolytic system defect. Functional and antigen plasminogen concentrations both in the plasma and with the isolated, purified plasminogen showed that only 49% of the antigen concentration had potential functional active sites. Also, a reduced antigen concentration was found in both the propositus, and his mother (46% active sites).Sodium dodecyl sulfate polyacrylamide gel electrophoresis of the purified Frankfurt I plasminogen showed a normal native Glu-plasminogen band. Crossed-immunoelectrophoresis revealed a peak with normal size and shape, but displaced with respect to normal Glu-plasminogen toward the anode, i. e., was, as a whole, more negatively charged. Isoelectric focusing followed by zymog-raphy on a agarose-fibrin plate proved this observation, but did not indicate a separation of the normal from the abnormal plasminogen molecular species, also, fewer bands were found in the abnormal plasminogen isozyme pattern.Kinetic studies of Frankfurt I Glu-plasminogen and plasmin led to the conclusion that most of the functional abnormality is related to absence of active sites in half of the molecules. The plasmin generated was very unstable in the absence of stabilizing ligands and/or substrates. After reduction, the plasmin was completely converted to the typical two plasmin chains, A and B.

SummarySpecific assay methods for the human plasmin-derived light (B) chain · streptokinase (B·SK) complex, in terms of both streptokinase (SK) and urokinase (UK) International Units, are described. The kinetic properties of various SK activator complexes with plasminogen, Val 442-plasmin, and the plasmin-derived light (B) chain were compared to SK in terms of their catalytic efficiencies and Lineweaver-Burk plots. Similar kinetic data, and Lineweaver-Burk plots, are described for both highly purified high-molecular weight UK and low-molecular weight UK, including different clinical UK preparations. The B·SK complex has the highest catalytic efficiency of all the activator species studied. The Lineweaver-Burk plots of each of the various activator species are “fingerprints” of the enzymatic character of the activator. The B-SK complex is more like UK than SK, as an activator, in activating non-human plasminogen species. The biological halflife of the B·SK complex, in a dog model, was determined to be about 4 hr which is longer than the biological half-life(s) of SK in the same animal model, namely 0.6 hr (47%) and 2.8 hr (53%). This new second-generation activator complex may prove to be a useful thrombolytic agent in the treatment of thromboembolic diseases.

Affinity chromatography forms, 1 and 2, were each isolated from human Glu- and Lys-plasminogens by gradient elution from a L-lysine-substituted Sepharose column with a linear gradient of epsilon-aminocaproic acid. Although each of the two zymogen forms contains two affinity chromatography forms, the relative concentrattions of these forms in each of the zymogen preparations depended upon the plasma sample or enriched plasma fraction used for the preparation of the zymogen. Specific analytical acrylamide gel electrophoretic systems were used for the characterization of the zymogen and enzyme forms, and their component affinity chromatography forms, 1 and 2. The four zymogen affinity chromatography forms, Glu-1-plasminogen, Glu-2-plasminogen, Lys-1-plasminogen, and Lys-2-plasmingoen, show distinct stepwise differences in their molecular size and charge. The Glu-1-form is the largest in molecular size and the most acidic, and the Lys-2-form is the smallest in molecular size and the most basic. The proteolytically altered Lys-1- and Lys-2- forms appear to be specifically df the zymogen affinity chromatography forms showed a different distribution of isoelectric forms. The major isoelectric forms isolated from Glu-plasminogen with pI values of 6.2, 6.3, 6.4, and 6.6, and the major isoelectric forms isolated from Lys-plasminogen with pI values of 6.7, 7.2, 7.5, 7.8, and 8.1, (Summaria, L., Arzadon, L., Bernabe, P., Robbins, K. C., and Barlow, G. H. (1973) J. Biol. Chem. 248, 2984-2991) were shown to be mixtures of the Glu-1- and Glu-2- forms, or the Lys-1- and Lys-2- forms, respectively. Although the sialic acid contents of the Glu- and Lys- forms appear to be similar, the isolated affinity chromatography forms show distinct differences. The sialic acid contents of the Glu-1- and Lys-1- forms are identical, and are substantially higher than the sialic acid contents of the Glu-2- and Lys-2- forms which are also identical to each other. It is possible that the charge difference between the zymogen-1- and -2- forms may be related to the differences in their sialic acid content. Each of the four zymogen affinity chromatography forms, when activated by urokinase in the presence of the plasmin inhibitor, Trasylol, was converted to an apparently unique and different enzyme form. The four enzyme forms show distinct stepwise differences in molecular size; Glu-1-plasmin is the largest in size whereas Lys-2-plasmin is the smallest in size. Each plasmin-derived carboxymethyl heavy(A) chain was found to be different in molecular size, but the two carboxymethyl light(B) chains found in each of the four enzyme forms appeared to be identical and of the same molecular sizes. The four heavy(A) chains show a stepwise difference in molecular size; the Glu-1-heavy(A) chain is the largest in size whereas the Lys-2-heavy(A) chain is the smallest in size...

The availability of a cell line derived from an adenocarcinoma of the liver has made it possible to study the plasminogen activator(s) (PA) biosynthesized in culture by liver cells. Conditioned cultured media purified on fibrin-celite, benzamidine-Sepharose and immunoabsorbent anti-urokinase columns have shown the presence of multiple plasminogen activators when separated on sodium dodecyl sulfate polyacrylamide electrophoresis (SDS-PAGE). These PAs differ in molecular weight but all are urokinase-like based on their reaction with goat anti-urokinase and rabbit antitissue activator. Subcellular fractionation of the cultured cells shows the presence of activator in both the cytoplasmic and membrane fractions, but the higher molecular weight forms appear primarily in the cytoplasm.

SummaryA new abnormal, variant, plasminogen Chicago III has been isolated from a patient with recurring deep vein thrombosis. Studies on the plasma fibrinolytic system of four family members showed no inheritance pattern. Kinetics of activation parameters of Chicago III plasminogen with different activators showed lowered catalytic rate constants from 159fold with urokinase, to 3fold with light (B) chain ‧ streptokinase complex, to 1.3fold with streptokinase. The Michaelis constants of activation of Chicago III plasminogen were 16fold higher with streptokinase, 6fold higher with light (B) chain ‧ streptokinase complex, and similar with urokinase. Each of the three activators exhibited different interaction characteristics with this variant zymogen, as they did with variant Chicago I and Chicago II plasminogens (previously reported). However, the latter variants were characterized by having normal catalytic rate constants of activation, with higher than normal apparent Michaelis constants of activation. Chicago III plasminogen, as well as Chicago I and II plasminogens, has a homogeneous population of molecules; the interpretation of the kinetic data was possible only in terms of a single population of molecules. The plasmins derived from Chicago I plasminogen, and also Chicago II and Chicago III plasminogens, have 100% active sites with normal amidase parameters. Basically, a single population of normal isoelectric forms were found in the Chicago III plasma, with three minor forms, about 11% of the total.The kinetic parameters have permitted us to classify the four known plasminogen variants into three different classes. The Class A homozygote (Tochigi) has both an active center defect and a charge mutation difference with normal cleavage of the Arg560-Val peptide bond; the Class B homozygote (Chicago I and Chicago II) has a Km of activation defect with impaired activator binding and normal cleavage of the Arg560-Val peptide bond; and, the Class C homozygote (Chicago III) has both a Km of activation defect with impaired activator binding and a kcat of activation defect, and impaired Arg560-Val peptide bond cleavage.

In an attempt to elucidate the mechanism of fibrinolytic enhancement by orally administered urokinase, studies on the intestinal transport of urokinase were carried out, using 125I-labeled human high mol wt urokinase, administered intraduodenally in the experimental dog model with a saphenous vein thrombus. Using the plasma sample obtained from blood 45 minutes after intraduodenal administration of the urokinase, protein fractions were isolated by a sequential two-step affinity chromatography method, first with [N alpha-(epsilon-aminocaproyl)-DL- homoarginine hexylester]-Sepharose followed by a specific anti-human low mol wt urokinase rabbit IgG-Sepharose (adsorbed-eluted and unadsorbed). Each of the isolated protein fractions was further purified by gel filtration on Sephacryl S-300. The proteins isolated by the two-step affinity chromatography method were transported human urokinase with radioactivity in the adsorbed-eluted fraction, and newly synthesized and/or released dog plasminogen activators, probably urokinase-type and tissue-activator type, without radioactivity. In an antibody quenching assay, dog urokinase and the immuno-affinity unadsorbed fraction were not neutralized, but the immuno-affinity adsorbed-eluted fraction was completely neutralized by the specific anti-urokinase IgG antibody. Proteins isolated from control plasma (after administration of saline) by the two-step affinity chromatography method in the unadsorbed fraction had negligible amounts of activator activity. In these studies, we were able to show that synthesis of plasminogen activators was stimulated, with the activators being released, from either the liver or the vascular endothelium. Also we showed that urokinase is transported across the intestinal tract in the dog model.

Highly purified human (Lys-forms), cat, dog, and rabbit plasminogens showed significantly different sensitivities to activation by highly purified streptokinase. The most sensitive plasminogen is the human zymogen, followed by the cat, dog, and rabbit zymogens, respectively. These human, cat, dog, and rabbit zymogens reacted with streptokinase to form homogeneous equimolar plasminogen-streptokinase complexes. The electrophoretic mobilities of the human and cat complexes on cellulose acetate were different from their respective zymogens, but the electrophoretic mobilities of the dog and rabbit complexes were similar to their respective zymogens. The human, cat, and dog complexes showed different mobilities from their respective zymogens in acrylamide gel electrophoretic systems and appeared to show multiple electrophoretic forms. With these species, all of the multiple isoelectric forms of the zymogens reacted with streptokinase. Analyses of the human (Glu- and Lys-forms), cat, dog, and rabbit equimolar plasminogen-streptokinase complexes, incubated for various time intervals, in an acrylamide gel-dodecyl sulfate-urea electrophoretic system, showed the gradual conversion, or transformation, of the plasminogen-streptokinase complexes into plasmin-streptokinase complexes. These transformations occurred at different rates in each of the mammalian plasminogen-streptokinase complexes. This acrylamide gel system also permitted a comparison between the plasmin-derived carboxymethyl heavy (A) and carboxymethyl light (B) chains produced from these mammalian zymogens by two different methods of activation, namely with equal molar ratios of streptokinase to zymogen and with low molar ratios of urokinase to zymogen (1:1000). The carboxymethyl heavy (A* and A') chains produced by streptokinase activation of the human (Glu-forms), cat, and rabbit plasminogens in the complex appeared to have higher molecular weights than the major carboxymethyl heavy (A and A1) chains derived from each of these zymogens by urokinase activation; the human (Lys-forms) and dog carboxymethyl heavy (A and A') chains appeared to be similar in both activation systems. The carboxymethyl heavy (A* and A') chains of the human (Glu- and Lys-forms), cat, dog, and rabbit plasmins found in the complexes were similar but not identical with each other in molecular weight. The plasmin-derived carboxymethyl light (B') chains produced by streptokinase activation of the human (Glu- and Lys-forms), cat, dog, and rabbit plasminogens in the complex appeared to have molecular weights similar to the carboxymethyl light (B) chains derived from each of these plasminogens by urokinase activation. But, the plasmin-derived carboxymethyl light (B') chains of the four species obtained from the equimolar complexes appeared to have somewhat different molecular weights. Streptokinase fragmentation within the complexes occurs within a few seconds after the complexes are formed, and intact streptokinase disappears after several minutes of incubation. Four major streptokinase fragments (SK1, SK2, SK3, and SK4), having molecular weights between 47,600 and 25,700, are produced in each of the equimolar human (Glu- and Lys-forms), cat, dog, and rabbit complexes. They appear to be the same fragments in each of the mammalian plasminogen-streptokinase and plasmin-streptokinase complexes; each of the complexes contains varying amounts of streptokinase fragments, SK1, SK2, SK3, and SK4. But, the rate of degradation of streptokinase into these major fragments differed with each species. Plasmin and streptokinase moieties prepared by dissociation of an equimolar plasminogen (Lys-forms)-streptokinase complex at pH 3.0 (Summaria, L., Robbins, K. C., and Barlow, G. H. (1971) J. Biol. Chem. 246, 2136–2142) were also analyzed in the acrylamide gel-dodecyl sulfateurea electrophoretic system. The isolated plasmin moiety gave carboxymethyl heavy and carboxymethyl light chains, both of which were of lower molecular weight than the plasmin-derived carboxymethyl heavy (A) and carboxymethyl light (B) chains produced by urokinase activation. The isolated streptokinase moiety contains approximately equal amounts of streptokinase fragments SK2 and SK4, and a third smaller fragment.

The equimolar human light (B) chain·streptokinase activator complex was shown to be an effective fibrinolytic agent in the dissolution of human cross-linked plasma fibrin clots, in vitro; its activator activity was similar to that of the equimolar human plasmin·streptokinase complex. 125 I-radiolabelled preparations of the human plasmin-derived light(B) and heavy(A) chains, and Lys-plasminogen were shown to be adsorbed to cross-linked plasma fibrin clots invitro to the same extent. Clot lysis by the light(B) chain·streptokinase complex, the plasmin·streptokinase complex, streptokinase and urokinase was enhanced by a factor of at least 10-fold when the fibrin clots were pretreated with either the light(B) chain or Lys-plasminogen, but was not enhanced when the fibrin clots were pretreated with the heavy(A) chain. The activator activities of the light(B) chain·streptokinase complex, the plasmin·streptokinase complex and streptokinase were similar on these pretreated cross-linked fibrin clots. Limited proteolysis of fibrinogen by the light(B) chain and by plasmin showed the same specificity giving Fragments X, Y, D and E, but the rate of hydrolysis of fibrinogen by the light(B) chain was much slower.

Summary A highly allergenic protein-carbohydrate preparation has been isolated from a low ragweed pollen extract (Ambrosia elatoir) by precipitation with ammonium sulfate followed by chromatography on DEAE-cellulose, gel-filtration on Sephadex G-75, chromatography on DEAE-Sephadex and gel-filtration on Sephadex G-200. The preparation is skin reactive at 10-9 to 10-10 µg, on intradermal injection, in the treated rag-weed-sensitive patient. Immunoelectrophoretic and gel-diffusion methods show multiple antigens.

The isolated equimolar human plasmin-streptokinase activator complex was found to be an excellent fibrinolytic agent in an experimental dog model. It will dissolve artificially produced human clots placed in the femoral vein of the animals. There appears to be a critical dose level for fibrinolytic activity which may be approximately 0.1 mg per kg body weight. At the critical dose level, the activator complex appeared to be more effective than equivalent individual doses of either streptokinase or human plasmin.

With the use of photoelectric scanning optics on the analytical ultracentrifuge molecular weights on human plasminogen and plasmin have been determined at levels as low as 50 µg per ml and as low as 10 µg per ml for the sedimentation coefficient. The results give a molecular weight of 81,000 for human plasminogen and 75,400 for human plasmin. The molecular weight difference has been verified with differential sedimentation equilibrium techniques. The s20, w0 value of human plasminogen and plasmin were determined to be 4.2 S and 3.9 S, respectively.

Bovine plasminogen activator preparations were isolated at 0° and 25° from a mixture of streptokinase and human plasminogen. The degree of incorporation and inhibition with radioactive 32P-labeled diisopropyl phosphofluoridate (DFP) demonstrated a definite difference between these two preparations. Incorporation of approximately 1 mole of DFP-32P per mole of bovine plasminogen activator prepared at 25° and per mole of human plasmin indicated that bovine plasminogen activator and human plasmin each contains a single "active site" which may be identical. Incorporation of 0.50 mole of DFP-32P per mole of bovine plasminogen activator prepared at 0° and its retention of partial activity indicated that this activator preparation was partly a human plasminogen-streptokinase complex. Acid treatment of this preparation and inhibition studies with e-aminocaproic acid also indicated that this preparation contains a plasminogen-streptokinase complex. Complex formation also occurred when streptokinase was allowed to react with DFP-32P-treated human plasminogen and plasmin. Starch gel electrophoretic analyses indicated that all the complexes formed under these conditions possessed the same mobility. Dissociation of the 25° activator complex inhibited with DFP-32P after its isolation revealed that the radioactivity was located in the plasmin moiety of the activator complex.

NH2-terminal sequences were determined for human plasminogen and the S-carboxymethyl heavy (A) and light (B) chain derivatives of plasmin with a protein sequenator. The NH2-terminal 5 residues of plasminogen and the isolated S-carboxymethyl heavy (A) chain derivative isolated from plasma Fraction III2,3 were determined to be Lys-Val-Tyr-Leu-Leu-. A second identical repeating des-lysyl sequence was found. These sequences were identical for all Fraction III2,3 plasminogen preparations and isolated isoelectric forms, pH values 6.7 and 7.8. Plasma Fraction III contains, in addition to the plasma Fraction III2,3 isoelectric forms, other predominant isoelectric forms; the NH2-terminal 5 residues of Fraction III isolated isoelectric forms, pH values 6.2, 6.4, and 6.6, were determined to be Glu-Pro-Leu-Asp-Asp-. The NH2-terminal 20 residues of the isolated S-carboxymethyl light (B) chain derivative of plasmin isolated from plasma Fraction III2,3 were determined to be Val-Val-Gly-Gly-Gln-Val-Ala-His-Pro-His-Ser-Trp-Pro-Trp-Gln-Val-Val-Leu-Leu-Arg-. This light (B) chain sequence shows extensive homology to the NH2-terminal sequence of other serine proteases.

The NH2-terminal 12-residue sequences of human and rabbit plasminogens were determined to be Glu-Pro-Leu-Asp-Asp-Tyr-Val-Asn-X-Gln-Gly-Ala-. The cat plasminogen 12-residue sequence differed only in the 1st residue, in which aspartic acid replaced glutamic acid. The bovine plasminogen NH2-terminal 7-residue sequence was determined to be Asp-Leu-Leu-Asp-Asp-Tyr-Val-. The NH2-terminal residue of dog plasminogen is blocked since the protein could not be sequenced. The NH2-terminal 5-residue sequences of the isolated S-carboxymethyl heavy (A) chain derivatives of cat, dog, rabbit, and bovine plasmins (urokinase-activated) were determined and found to be homologous with each other and with the human S-carboxymethyl heavy (A) chain 5-residue sequence. These NH2-terminal sequences were different from those found in the five mammalian zymogens studied, indicating a specific cleavage at the NH2-terminal end of each zymogen during activation at apparently the same position in the peptide chain. These sequences are also homologous with the human Lys-plasminogen NH2-terminal sequence. The NH2-terminal 21-residue sequences of the isolated S-carboxymethyl light (B) chain derivatives of cat, dog, rabbit, and bovine plasmins (urokinase-activated) were determined and found to be homologous with each other and with a previously reported human S-carboxymethyl light (B) chain 20-residue sequence (Robbins, K. C., Bernabe, P., Arzadon, L., and Summaria, L. (1972) J. Biol. Chem. 247, 6757–6762). The human NH2-terminal 21-residue sequence was determined to be Val-Val-Gly-Gly-Cys-Val-Ala-His-Pro-His-Ser-Trp-Pro-Trp-Gln-Val-Val-Leu-Leu-Arg-Arg-. The 5th residue previously reported as glutamine is incorrect and is cysteine in all the sequences. This cysteine is part of the hinge disulfide bond connecting the heavy (A) and light (B) chains of human plasmin. The evolutionary changes in this portion of the molecule appear to be in the pairs of basic residues at positions 8 and 10, and 20 and 21, and in the pairs of neutral residues at positions 16 and 17. The replacement of serine at the 17th residue of bovine light (B) chain is a rather significant evolutionary change which could be related to the activator binding site in bovine plasminogen for streptokinase. The mechanism of activation of plasminogen to plasmin by urokinase involves two specific peptide bond cleavages instead of one previously reported by us for the activation of human Lys-plasminogen (Robbins, K. C., Summaria, L., Hsieh, B., and Shah, R. J. (1967) J. Biol. Chem. 242, 2333–2342). The specific cleavage of the sensitive arginyl-valine peptide bond in the COOH-terminal portion of the molecule to give the two-chain plasmin molecule and the specific cleavage of an X-arginyl, or X-lysyl, peptide bond in the NH2-terminal portion of the molecule are the two principal events which occur during activation of the zymogen. The latter event apparently results in the release of a peptide.

Human plasminogen was prepared from both plasma Fractions III and III2,3 by affinity chromatography on lysine-substituted Sepharose columns. Plasminogen prepared by this method, from either Fraction III or III2,3, appeared to be homogeneous when analyzed by slab acrylamide gel electrophoresis in 6 m urea-0.1% sodium dodecyl sulfate. The specific activities of the preparations were 20 to 24 casein units per mg of protein. Acrylamide gel electrophoretic analyses in 0.3 m e-aminocaproic acid at pH 8.4 showed that the distribution of the multiple molecular forms of plasminogen prepared from Fraction III differed depending upon whether the extracts were either frozen, or unfrozen, prior to affinity chromatography or processed more quickly during affinity chromatography. The major electrophoretic components in Fraction III plasminogen prepared from extracts which were unfrozen had greater electronegative mobilities than the major components in Fraction III2,3 plasminogen. The electrophoretic forms seen in Fraction III plasminogen prepared from extracts which had been frozen and thawed, prior to affinity chromatography appeared to be identical with those in Fraction III2,3 plasminogen. Electrophoretic analyses in acrylamide gels containing 0.4% casein showed that all of the electrophoretic forms of plasminogen prepared from Fractions III and III2,3 could be activated with urokinase as determined by the zones of casein digestion seen after incubation and staining of the gels. Electrophoretic analyses of Fractions III and III2,3 plasminogens in acrylamide gel-isoelectric focusing systems showed similar patterns to those obtained in the acrylamide gel-0.3 m e-aminocaproic acid system. At pH 3.1, in acrylamide gel electrophoresis, Fraction III plasminogen showed two groups of components, whereas Fraction III2,3 plasminogen showed a single component similar to the minor, faster moving component found in Fraction III plasminogen. Analyses and isolation of the individual isoelectric forms from Fractions III and III2,3 plasminogens were carried out with isoelectric focusing methods. Sixty-six per cent of the plasminogen isolated from Fraction III had isoelectric points between pH 6.2 and 7.2 (six forms). The specific activity of these forms was found to be 20 to 21 casein units per mg of protein. Each of these forms contained a major, and a minor electrophoretic component in slab acrylamide gel electrophoresis in 0.3 m e-aminocaproic acid. Eighty-three per cent of the plasminogen isolated from Fraction III2,3 had isoelectric points between pH 7.2 and 8.1 (four forms). The specific activity of these forms was found to be 22 to 24 casein units per mg of protein. Each of these forms contained two major isoelectric components ("doublets"). The NH2-terminal amino acid of the isoelectric forms with pI values of 6.2, 6.4, and 6.6 was determined to be glutamic acid. The NH2-terminal amino acid of the isoelectric forms with pI values of 6.7, 7.2, 7.8, and 8.3 isolated from Fraction III2,3, and of the isoelectric forms with pI values of 7.2 and 7.8 isolated from Fraction III was determined to be lysine (and valine). Sedimentation-equilibrium analyses on isolated isoelectric forms showed that the NH2-terminal lysine forms with pI values of 7.2, 7.5, and 7.8, and the NH2-terminal glutamic acid forms with pI values of 6.2 and 6.4 all had the same molecular weights, approximately 80,000. The sedimentation coefficients, s°20,??w, of the Fraction III plasminogen forms were different from the Fraction III2,3 plasminogen forms. Each plasminogen isoelectric form, from pI values 6.2 to 8.5, can be activated with urokinase. The resulting plasmins which were inhibited with l-1-chloro-3-tosylamido-7-amino-2-heptanone, reduced with mercaptoethanol and alkylated with iodoacetic acid, gave Cm-heavy (A) and Cm-light (B) chain derivatives that appeared to be identical with each other when analyzed in acrylamide gel electrophoresis in urea-sodium dodecyl sulfate systems. Cm-heavy (A) and Cm-light (B) chain derivatives isolated from urokinase-activated plasminogen III A appeared to be identical with the same derivatives isolated from urokinase-activated plasminogens 1112,3 A and III2,3 S.

The S-carboxymethyl heavy chain derivative of human plasmin was prepared in a highly purified form by precipitating out the contaminating S-carboxymethyl light chain derivative in 1 m trichloracetic acid-8 m urea. The S-carboxymethyl heavy chain derivative was shown to be homogeneous by ultracentrifugal, carboxyl-terminal amino acid, and cellulose acetate electrophoretic analyses. Its weight average molecular weight was determined to be 48,800 ± 3,000 and its sedimentation coefficient was determined to be 2.8 S. Arginine was verified to be its carboxyl-terminal amino acid. The heavy chain derivative consists of approximately 410 amino acid residues.

Specific rabbit precipitating antibodies have been preparted to human plasminogen and plasmin. The preoenzyme-enzyme pair were found to be indentical in gel diffusion. In immunoelectrophoresis, the proenzyme-enzyme pair differ in their electrophoretic mobilities. Antibody γ-globulin preparations to either human plasminogen or plasmin were found to neutralize the proteolytic activity of plasmin with maximum inhibitions of 80 percent. There were some differences in the ability of antibody γ-globulin to neutralize plasminogen activated in situ, with either streptokinase or urikinase, as compared to plasmin. The antibodies also partially inhibit the activation of proenzyme to enzyme. Trypsin inhibitors which inhibit plasmin, appear to be different, in certain respects, from specific antibodies.

Summary and ConclusionsA simple method is described for preparation of highly purified intrinsic factor PI from porcine stomach pyloric mucosa. In pernicious anemia patients in relapse, 35 mg of intrinsic factor PI when given with 15 μg vit. B12 is defined as one U.S.P. oral unit. Dialysis of intrinsic factor PI gives a material with a potency of about 15 mg/U.S.P. unit. In the urinary excretion test, one-half U.S.P. unit of intrinsic factor PI gives significant and measurable urinary excretion of vit. B12 Co60 when given orally with 1 μg vit. B12 Co60. Approximately 1.5 U.S.P. units of intrinsic factor PI will give maximum urinary excretion of vit. B12 Co60. Intrinsic factor PI can be further purified on a DEAE-cellulose ion exchanger to give a material with an estimated potency of 9 mg/U.S.P. unit.

Summary The serological characteristics of a toxin producing strain of H. pertussis, # 29, are compared with a typical phase I strain, # 18-323. A water extract of strain # 29 organisms, heated at 56 C for 45 minutes, contains a protective antigen against a typical phase I strain, # 18-323, the challenge organism in the mouse protection test. The protective antigen can be partially separated from the heated, non-toxic, water extract by fractionation with methanol-water mixtures under controlled conditions of pH, ionic strength and temperature.

Bovine plasminogen activator was isolated from a mixture of streptokinase and human plasminogen. The method of preparation was based on the difference in solubility between the activator and its precursors. The activator can be prepared from either crude or highly purified streptokinase and highly purified human plasminogen. The activator complex gave a single component in starch gel electrophoresis, in either lysine or e-aminocaproic acid buffers, and in immunochemical studies with a specific rabbit antibody to human plasminogen. However, the mobility of the activator was different from human plasmin, human plasminogen, or streptokinase both in starch gel electrophoresis and in immunoelectrophoresis. The activator can be dissociated into two components, apparently corresponding to streptokinase and human plasmin, by electrophoresis in starch gels. Gel diffusion analyses with specific antibody revealed that the activator was immunologically identical with both human plasmin and human plasminogen. In sedimentation velocity experiments, the activator showed a single component in e-aminocaproic acid buffers with a sedimentation coefficient of approximately 5 S. In lysine buffers, activator showed some heterogeneity, but the sedimentation coefficient of the major component was found to be approximately 11 S. The apparent molecular weight of activator by sedimentation equilibrium methods, in e-aminocaproic acid buffers, was found to be approximately 150,000. In lysine buffers the apparent molecular weight in dilute solutions was found to be approximately 138,000, whereas in concentrated solutions the molecular weight was found to be approximately 308,000. These data suggest a possible monomer-dimer relationship which is concentration dependent. The dimer form appeared to be present when activator was dissolved in lysine buffers, in concentrated solutions only, whereas the monomer form was always found when activator was dissolved in e-aminocaproic acid buffers.

A method for the preparation of highly purified bovine liver arginase (proenzyme) is described. This procedure gives an enzyme preparation which on activation by Mn++ will liberate up to 3300 μmoles urea/min./ mg. protein at 37 °, in the presence of optimum substrate, at pH 9.5. Electrophoretic analysis in Veronal buffer, ionic strength 0.1 at pH 8.6, showed a heterogeneous mixture with a major component of 60% having a mobility of approximately −2.8. The activity-pH curve indicates two pH optima. Crystalline catalase can be obtained as a by-product from arginase fraction PII.

An isolated equimolar human plasmin-streptokinase complex (native complex) was dissociated either in 8 m urea or at pH 3.0, both before and after treatment with diisopropyl phosphorofluoridate (DIP complex). Dissociation of the native complex either in 8 m urea or at pH 3.0, resulted in an immediate loss of both the bovine plasminogen and human plasminogen activator activities with complete restoration of the original plasmin proteolytic activity. Urea or acid dissociation of the inactive DIP complex did not regenerate either activator or proteolytic activities. The weight average molecular weight of native complex was redetermined with the use of photoelectric scanning optics in the analytical untracentrifuge and found to be 123,900 ± 1,800. The sedimentation coefficient, s020,w, of native complex was redetermined to be 5.0 S. On immunoelectrophoretic analysis, native complex gave a single antigen-antibody band with specific rabbit antibodies to both plasminogen and streptokinase. The DIP complex dissociated with either urea or acid still reacted with both antibodies. However, urea dissociation of native complex caused the mixture to lose only its ability to react with the streptokinase antibody. Gel filtration on Sephadex G-150 in 0.001 m HCl-0.15 m NaCl, pH 3.0, and on Sephadex G-200 in 6 m urea-0.02 m Tris buffer, pH 8.0, was used to separate the dissociated plasmin and streptokinase moieties from the urea- and acid-dissociated complexes. The plasmin moiety isolated from the DIP complex dissociated at pH 3.0 was found to be homogeneous by electrophoretic and ultracentrifugal analysis with a weight average molecular weight of 84,000 ± 4,100 and a sedimentation coefficient, s20,w, of 4.2 S. Carboxyl-terminal amino acid analysis showed that this moiety contained 0.99 residue of asparagine, 0.47 residue of arginine, and 0.53 residue of lysine per mole of protein. It reacted with specific antibodies to both plasminogen and streptokinase. The weight average molecular weight of this moiety in 6 m guanidine-0.05 m phosphate buffer, pH 7.0, was determined to be 67,250. It was calculated that two species of molecules could be present in solution, in equimolar amounts, with molecular weights of 75,000 (plasmin) and 9,000 (streptokinase fragment). The plasmin moiety isolated from native complex dissociated in 8 m urea was found to be homogeneous by electrophoretic and ultracentrifugal analysis with a weight average molecular weight of 65,000 ± 4,000 and a sedimentation coefficient, s20,w, of 3.6 S. Carboxyl-terminal amino acid analysis indicated that considerable cleavage of the moiety had occurred at arginine- and lysine-peptide bonds. The streptokinase moieties isolated from both native and DIP complexes dissociated either in 8 m urea or at pH 3.0 were recovered primarily as fragments. In the ultracentrifuge, they appeared as mixtures of components with sedimentation coefficients between 2 and 0.8 S. This fragmentation occurs very shortly after formation of the complex without loss of any of the fragments from the complex.

Summary Heat-coagulated beef fibrinogen, Ca-fibrin, T-fibrin and heat-coagulated Ca-fibrin produce precipitins to fibrinogen. Heat coagulation of fibrinogen and fibrin changes their solubilities in 0.03 per cent hydrochloric acid and 0.5 per cent sodium carbonate. From the results obtained with the immunologic method, it would appear that the antigenic nucleus remains unaffected by clotting or by heat denaturation of fibrinogen regardless of its changes in solubility. The fibrinogen and fibrin molecules do not contain either albumin or pseudoglobulin.

The S-carboxymethyl β chain derivative from both moderately reduced urokinase-activated plasmin and extensively reduced streptokinase-activated plasmin was separated in a highly purified form by gel filtration through Sephadex G-200. Electrophoresis in urea-starch gel systems at pH 3.2 showed that the faster moving β electrophoretic component was the light chain derivative. The sedimentation coefficient, s20,w, of this derivative was found to be 1.4 S. The apparent molecular weight was found by sedimentation equilibrium analysis with absorption optics to be 25,700. The apparent molecular weight of the S-carboxymethyl heavy chain derivative was found to be approximately 57,200. Carboxyl-terminal amino acid analysis with carboxypeptidases A and B showed that asparagine was the carboxyl-terminal amino acid of the light chain and indicated that arginine must be the carboxyl-terminal amino acid of the heavy chain. Incorporation of 1.18 ± 0.08 moles of DFP-32P per mole of human plasmin indicated a single proteolytic "active site." Starch gel electrophoresis and Sephadex gel filtration experiments both showed that the proteolytic active site of human plasmin is located on the light chain. The amino acid compositions of the S-carboxymethyl light chain derivatives of both urokinase-activated plasmin and streptokinase-activated plasmin were found to be almost identical, confirming the specificity of the activation of human plasminogen to plasmin by both urokinase and streptokinase.

Summary and Conclusions1. Some of the nutritional requirements for the in vitro cultivation of hog kidney epithelial cells were determined. The data indicate that cultures grown in a “simple” medium containing Synthetic Mixture #199 (92.5%) and hog serum (7.5%) were consistently more proliferative than cultures cultivated in more complex media containing embryo extract, serum ultra-filtrate and serum of homologous and heterologous origin. 2. None of the media tested adequately supported growth of either splenic or testicular fibroblasts. 3. The influence of various Porcine Serum Fractions (II, IV, V and IV + V) on growth of hog kidney epithelial cells was also investigated. When Fraction IV -f- V was used, excellent growth was achieved. When Fraction II, IV or V was used independently, growth inferior to that of whole serum resulted.

Human plasminogen and plasmin were found to be monomers by physical measurements. The cleavage of a single disulfide bond in both the proenzyme and enzyme, at pH 9, by reduction with 2-mercaptoethanol in 8 m urea, resulted in complete loss of proteolytic activity. At pH 3, activity losses paralleled the number of disulfide bonds cleaved with complete loss in activity only after total cleavage of all the disulfide bonds (21 to 22). The cleavage of the single interchain disulfide bond of plasmin at pH 9 resulted in the formation of two major chains (α and β) which were first demonstrated by starch gel electrophoresis in 8 m urea-2-mercaptoethanol systems at pH 3. S-Sulfo and S-carboxymethyl derivatives of plasmin were prepared; they were also found to contain two major chains on electrophoresis in starch gels. Sedimentation velocity measurements of both proenzyme and enzyme and their S-carboxymethyl derivatives, in urea-2-mercaptoethanol systems, confirmed the presence of plasmin chains. The amino-terminal amino acid residue of plasminogen and one plasmin chain was found to be lysine, and the amino-terminal amino acid residue of the second plasmin chain was found to be valine. The carboxyl-terminal amino acid sequence of plasminogen and one plasmin chain was found to be Leu-Asn-COOH, and the carboxyl-terminal amino acid sequence of the second plasmin chain was found to be Lys-Arg-COOH. The activation of human plasminogen monomer to plasmin monomer by urokinase is probably due to the cleavage of a single arginyl-valine bond by the activator, which results in a two-chain molecule held together by a single disulfide bond. The spontaneous proteolytic activity of human plasminogen preparations is probably due to contaminating plasmin. It is suggested that the activation of human plasminogen by other activators, such as streptokinase, trypsin, and pig heart activator, may involve a similar hydrolytic cleavage of an arginyl-valine bond.