The Fibrinolytic Activity of Streptokinase Activated Human Plasmin

THE FIBRINOLYTIC ACTIVITY OF STREPTOKINASE ACTIVATED HUMAN PLASMIN

Sol Sherry

J Clin Invest. 1954;33(7):1054-1063. https://doi.org/10.1172/JCI102973.

Research Article

Find the latest version: https://jci.me/102973/pdf THE FIBRINOLYTIC ACTIVITY OF STREPTOKINASE ACTIVATED HUMAN PLASMINI By SOL SHERRY (From the May Institute for Medical Research of the Jewish Hospital Association, and De- partment of Medicine, University of Cincinnati, College of Medicine, Cincinnati, Ohio) (Submitted for publication February 5, 1954; accepted March 22, 1954) The successful introduction of enzymes for the contaminating plasminogen is rapidly activated by dissolution of undesirable collections of sanguine- streptokinase in the presence of trace amounts of ous, fibrinous, and purulent exudations in local- human plasminogen or plasmin. The activation of ized areas of disease in man (1), has stimulated the bovine plasminogen, under these circumstances, studies on the experimental in vivo dissolution of results in abnormally high values for the fibrino- intravascular blood clots by enzymatic means lytic activity of human plasmin preparations. In (24). Activation of plasminogen, the proteolytic general, the values obtained with the present fibri- enzyme precursor present in plasma, to plasmtin, nolytic assay for streptokinase activated plasmin by streptokinase 2 has received considerable at- are approximately five hundred fold greater than tention in these studies (2, 3, 5) not only because the actual fibrinolytic activity of these human of plasminogen's natural occurrence, but because plasmin preparations. of the special fibrinolytic propensities attributed The significance of these findings in relation to to SK activated human plasmin (6, 7). the diversity of data concerning the streptokinase Preparations of the human proenzyme prepared activated plasma proteolytic system will be dis- by the Kline procedure (8) and activated by strep- cussed. tokinase, consistently have fibrinolytic activities of METHODS approximately 200,000 units per mg. nitrogen.8 The striking fibrinolytic activity of streptokinase Enzyme preparations: activated human plasmin preparations is very large Human plasminogen. Prepared by the method of Kline (8). This procedure consistently yielded preparations of when compared to their known and measurable high activity. proteolytic activity. It seems unusual for an en- Human plasmin. Three preparations of this enzyme zyme which extensively attacks a great variety of were studied; protein substrates (9-i 1), and has the same spec- a) Human plasminogen activated by streptokinase, trum of activity as trypsin on synthetic substrates and referred to as streptokinase activated plasmin (12), to have such a unique fibrinolytic activity. or SK plasmin. b) A preparation of human plasmin derived from It is the purpose of this report to present evi- human plasminogen which spontaneously activated dence that, (a) the fibrinolytic activity of human while human plasminogen was being reprecipitated plasmin preparations is comparable to that of other three times during a three day period, according known proteolytic enzymes, and (b) the extra- to the Kline procedure for plasminogen precipita- ordinary fibrinolytic activity of streptokinase action. c) A spontaneously activated human plasmin prepara- tivated human plasmin, when measured with the tion (Cohn's plasma Fraction IIL1), obtained conventional bovine clot, is due to the activation through the courtesy of Dr. D. M. Surgenor, Har- of a contaminating bovine plasminogen in the fibri- vard University Laboratory of Physical Chemistry. nogen preparation used in the assay system. This Bovine plasminogen. Bovine plasma Fraction III2 4 or fresh plasma were used as sources of bovine plasminogen. 1 This study was supported by a grant from the Lederle Botine pksmin (or bovine fibrinolysin). Obtained Laboratories Division, American Cyanamid Company. through the courtesy of Dr. E. Loomis, Parke, Davis and 2 Aso referred to as SK. Co. This is a chloroform activated enzyme derived from 8 An activity of this amount is represented by the bovine plasma (13). ability of 0.005 gamma of plasmin N to digest a standard 4 Obtained through the courtesy of Dr. A. H. Holland, fibrin clot in 30 minutes. Jr. of Armour Laboratories. 1054 FIBRINOLYTIC ACTIVITY OF SK ACTIVATED HUMAN PLASMIN toss

Other animal plasminogens. Fresh plasma was used as lowed by the addition of fibrinogen and thrombin.7 a source of dog plasminogen. Prothrombin free dried When streptokinase was used as an activating agent for rabbit plasma (Difco) served as a rabbit plasminogen plasminogen, 1,000 units was incorporated into each of source. the assay tubes prior to the addition of thrombin. A Trypsin and Chymotrypsin (crystalline) (Worthington fibrisolytic unit is defined as the smallest amount of en- Labs.). zyme which produces clot lysis in 30 minutes under the Streptokinase. A commercial preparation of Varidase conditions of the test (14). (Lederle) (Lot No. 7-1089-287A) containing only small All reagents in this assay were made up in 0.025 M amounts of contaminating inhibitor. This preparation phosphate buffer in 0.85 per cent sodium chloride at pH contained a number of other impurities including several 7.5.8 non-proteolytic streptococcal enzymes. All references to Proteolytic assay. According to the casein assay of unitage are in terms of Christensen units (14). Remmert and Cohen (10). One casein unit is equiva- lent to the liberation of 450 gamma of acid soluble tyrosine Substrates: under the conditions of the test. Bovine fibrinogen. Fraction I (Armour), an alcohol TAMe and LEe esterase activity. According to the fraction of bovine plasma, was used in these studies. A method previously described (12). The results are ex- number of different lots of this reagent were employed. pressed as micromoles of acid liberated in 30 minutes at Clots made with this preparation plus bovine thrombin5 370 C. under the conditions of the test. in the presence of various amounts of SK, were stable for 24 hours. RESULTS Plasminogen-deficient human fibrinogen. A sample of human fibrinogen obtained for study through the courtesy I. The "unusual" fibrinolytic activity of SK acti- of Cutter Laboratories was further purified by three am- vated human plasmin as compared to other monium sulfate precipitations according to Laki's pro- plasmin preparations and cedure (15). Although this preparation contained traces trypsin of human plasminogen, the contamination was not signifi- In Table I is shown a comparison of crystalline cant enough to influence the fibrinolytic assay. Clots trypsin and several plasmin preparations, against produced with this reagent plus human 6 or bovine thrombin, in the presence of SK, were stable for several three types of substrates which are attacked by hours. Furthermore, this human fibrinogen preparation both enzymes, i.e., casein, synthetic basic amino was found to be relatively free of both plasmin inhibitor acid esters, and fibrin. The conventional assay and antistreptokinase, and was suitable for use as a of fibrinolytic activity was carried out with a bovine reagent in the fibrinolytic assay. fibrinogen-bovine thrombin clot. Three types of Casein. A preparation (Hammarsten quality) ob- plasmin preparations were tained from Nutritional Biochemicals. assayed: a) an SK ac- TAMe. (p-toluenesulfonyl I-arginine methyl ester). tivated preparation, b) two spontaneously activated Synthesized in our laboratories by a modification of the preparations, and c) a chloroform activated bovine method of Bergman, Fruton, and Pollok (12, 16). plasmin. LEe (1-lysine ethyl ester). Synthesized according to From the data in Table I it will be noted that conventional methods (17). per milligram of nitrogen, trypsin, when com- Assays: pared to the most purified human plasmin prepara- Fibrinolytic assay for streptokinase activated plasmin, tions presently available, had 30 times as much spontaneously activated plasmin, and other proteolytic proteolytic activity against casein, 140 fold the enzymes. The conventional assay utilizing a clot made TAMe esterase with bovine reagents, according to the method of Christen- activity, and 10 times the LEe sen (14), was employed. When a clot made with human esterase activity. However, the streptokinase ac- fibrinogen and thrombin was used, the procedure was tivated human plasmin had a fibrinolytic activity similar to the Christensen method except for the substi- 12 fold greater than trypsin, resulting in a fibrino- tution of human reagents in similar concentrations to the bovine reagents. In the fibrinolytic assay, the enzyme 7 At the present state of purity of the reagents the ac- solution to be assayed is serially diluted with buffer, fol- tion of trypsin and chymotrypsin appears to be greater on fibrinogen than fibrin, whereas the reverse is true with 5 The commercial bovine thrombin preparations of Up- SK plasmin. john and Parke, Davis and Co. were utilized. They con- 8 The presence of phosphate in the clotting mixture tain approximately 25 N.I.H. clotting units per mg. markedly accelerates plasmin fibrinolysis. Optimal ac- material. celeration was noted with 0.025 M phosphate concentra- 6 A preparation obtained from Cutter Laboratories tion. The effect of phosphate ion on fibrinolysis is be- containing 20 N.I.H. clotting units per mg. material. ing investigated. 1056 SOL SHERRY

TABLE I Comparison of the proteolytic, amino acid esterase, and fibrinolytic activity (bovine reagents) of several plasmin preparations, and trypsin

Fibrinolytic Esterase activity assay Fibrinolytic Casein PM hyd./30 min./mg. N units per units per units per mg. N casein Enzyme mg. N TAMe* LEet (Bovine clot) unit Trypsin 2,015 107,000 5,000 14,900 7.4 Plasmin, human (Streptokinase activ. human plasminogen) 63 748 551 176,000 2,790 Plasmin, human (Spont. activated) 58 658 450 438 7.6 Plasmin, human Fr. 1113 (Spont. activated) 4.3 - 32 7.4 Plasmin, bovine 1.0 25 10 8 8.0 * p-toluenesulfonyl 1-arginine methyl ester. t 1-lysine ethyl ester. lytic/proteolytic ratio 400 times greater than The striking fibrinolytic activity of the SK ac- trypsin. This implies that per unit of proteolytic tivated human plasmin, as measured by the bovine activity against casein, the streptokinase activated clot assay, does not appear to be an accurate enzyme had a fibrinolytic activity 400 times that measure of the actual fibrinolytic activity of the of trypsin. These differences in fibrinolytic ac- SK plasmin preparation, and can be shown to be tivity, per proteolytic unit, entirely disappeared due to two phenomena: 1) streptokinase, in the when streptokinase was not used as the activating presence of trace amounts of human plasminogen agent. The spontaneously activated human en- or plasmin, is an excellent activator for bovine and zymes, and the chloroform activated bovine plasmin other animal plasminogens, and 2) bovine fibrino- had the same fibrinolytic activity per unit of pro- gen in the amounts used in the fibrinolytic assay is teolytic activity as trypsin. seriously contaminated with bovine plasminogen. II. Lack of "unusual" fibrinolytic activity of SK III. Activation of animal plasminogens by strep- activated human plasmin when measured with tokinase in the presence of human plasmino- a human clot gen or plasmin The striking fibrinolytic activity of streptokinase Figure 1 illustrates that streptokinase, in the activated human plasmin also disappeared when presence of trace amounts of human plasminogen, the fibrinolytic activity was assayed against a human fibrinogen-human thrombin clot rather than TABLE II the usual bovine clot. Comparison of the proteolytic and fibrinolyti activity (human In Table II the fibrinolytic activity of crystalline reagents) of several plasmin preparations, trypsin, crystalline chymotrypsin, and several trypsin, and chymotrypsin plasmin preparations as determined by assay with Fibrinolytic assay Fibrinolytic a clot made of human reagents, is compared to casein (human clot) uits Per units per ("us cse their proteolytic activity. Enzyme mg. N g. N) unit The data in Table II reveal that the strepto- Plasmin, human* kinase activated human plasmin, which had assayed (SK activated at 176,000 fibrinolytic units/mg. N. with the bo- human plasminogen) 63 480 7.6 Plasmin, human vine reagents, had only 480 fibrinolytic units/mg. (Fraction IIIs) N. when assayed against a standard clot com- (Spont. activated) 4.3 30 7.0 posed of human reagents. Furthermore, the Plasmin, bovine 1.0 7.5 7.5 fibrinolytic activity per unit of proteolytic activity Trypsin 2,015 12,800 6.3 for the streptokinase activated plasmin was now Chymotrypsin 1,177 2,870 2.4 closely comparable to that observed with the other * This preparation contained 176,000 fibrinolytic units plasmin preparations, trypsin and chymotrypsin. per mg. N when assayed with bovine reagents. FIBRINOLYTIC ACTIVITY OF SK ACTIVATED HUMAN PLASMIN -1057 A. B.. 2J1- X- SK 0-SK+ Humon Plosminogen (0.3jN) 1.81- 0-Humon Plasminogen c (0.3gN) 0% 0 .

.4a.2

4 9

.- 0 .6

.a5 x x .-W V * _ E-P _a __ w _ b~~ w c w 0.48 096 1.44 .3 .6 .9 mg.N Bovine Fraction EL ml. Dog Plasma FIG. 1. ACTIVATION OF BOVINE AND DoG PLASMINOGENS BY SK IN THE PRESENCE OF HUMAN PLASMINOGEN Proteolytic activity formed measured by casein assay (ordinate). When SK was used, 1,000 units was added to test system. A. BOVINE SYSTEM Abscissa plots concentration of bovine fraction III , used as source of bovine plasminogen in test mixture. B. DoG SYSTEM Amounts of dog plasma, used as source of dog plasminogen, in test mixture is shown along abscissa.

is an excellent activator for both bovine and dog jection of small amounts of human plasminogen plasminogen, yet is without significant effect in the mixed with SK (3). absence of the human factor. The addition of bovine, dog or rabbit plasma, For these experiments, bovine Fraction III, or of bovine Fraction III2 to a mixture of strepto- was used as a source of bovine plasminogen (Fig- kinase and trace amounts of human plasmin, ure 1A) while dog plasma was used as source of markedly accelerated the fibrinolysis of a stand- dog plasminogen (Figure IB). Plasmin activity ard human or bovine fibrinogen-thrombin clot. by was measured casein The human The activation of these animal plasminogens by proteolysis. streptokinase was observed to occur either in the plasminogen employed was the highly purified presence of human plasminogen or human plasmin. preparation referred to in Table I as having 63 However, the presence of human plasminogen, or casein units per mg. N following SK activation. of human plasmin, alone, was without effect as an Note that only 0.3 gamma N of this human plas- activator. The possible mechanism of this acti- minogen preparation was necessary to show this vation will be subsequently referred to in the dis- effect. cussion. It is probable that the human factor re- Similar observations on the activation of animal quired in this activation is plasmin, or perhaps plasminogens have been made on: a) bovine plasma plasminogen, rather than some contaminant of the as the source of bovine plasminogen; b) dried rab- plasminogen preparations, since a) purification of bit plasma as the source of rabbit plasminogen; and human plasminogen or plasmin, of up to 300 fold, c) in vivo in the dog following the intravenous in- by a number of methods (8, 10, 17, 18) has resulted 1058 SOL SHERRY in a simultaneous purification of this activator but not in the presence of plasminogen or strepto- function (12), b) in the presence of streptokinase, kinase alone. These results indicate the activa- only minute traces of some of our purified plas- tion of a proteolytic enzyme in the bovine fibrino- minogen preparations (as little as 0.005 gamma N) gen preparation by the mixture of SK plus human are necessary to show this effect, and c) inactiva- plasminogen. tion of human plasminogen or plasmin by a va- Figure 2B illustrates the effect of bovine fibrino- riety of methods, results in a simultaneous loss of gen, due to its bovine plasminogen contaminant, the activator function. on fibrinolysis of a human clot. For this experiment, various amounts of bovine fibrinogen in 0.5 IV. Evidence that the bovine fibrinogen used in ml. buffer were added to 0.5 ml. of 0.5 per cent hu- the fibrinolytic assay is contaminated with man fibrinogen. One thousand units of strepto- bovine plasminogen kinase in 0.1 ml. were then added and the mixture The data presented in Figure 2 illustrate that clotted by the addition of 1 unit of human throm- the bovine fibrinogen preparation usually em- bin in 0.1 ml. buffer. The lysis time of the clots ployed in the fibrinolytic assay is seriously con- were then measured. The data in Figure 2B. re- taminated with bovine plasminogen. veal that significant fibrinolysis appeared with In the experiment described in Figure 2A., the the incorporation of small amounts of bovine digestion of casein, by the bovine fibrinogen used fibrinogen in the test system, and was progres- in the fibrinolytic assay, was measured in the pres- sively accelerated with increasing amounts of bo- ence of streptokinase with and without trace vine fibrinogen. With the test system employed, amounts of human plasminogen (0.3 gamma N). a concentration of 0.25 per cent bovine fibrinogen, It will be seen that significant proteolysis was ob- the amount used in the routine fibrinolytic assay, served in the presence of the mixture of strepto- produced clot lysis in 12 minutes. kinase plus trace amounts of human plasminogen, These observations have been confirmed on all

A. S. X-SK 90- O-SKIHuman Plosminogen (O.3UN) *-Human Plosminogen & 75. C i o 60

0 Ez 45- x0

E 30 ._

-zi 15-

.04 .08 J2 .i6 .20 A. % Bovine Fibrinogen % Bovine Fibrinogen FIG. 2. A. ACTIVATION OF A PRoTOLYTIC ENZYME IN BovINz F=NoGEN REAGENT BY STREPTOXINASE IN THE PRESENCE OF HUMAN PLASMINOGEN Proteolytic advity found was measured by casein assay (ordinate). Abscissa plots concentration of bovine fibrinogen in test mixture. When SK used, 1,000 units was added to test system. B. AC TION OF FxINOLtYsIS OF A HUMAN CLOT BY THE ADDITION OF BOVINE FIBRINOGEN TO THE CLOTTNG MIXrURx The ordinate plots the lysis time of the human clot in the presence of SK, and the abscissa the amount of bovine fibrinogen added to the clotting mixture. See text for details. FIBRINOLYTIC ACTIVITY OF SK ACTIVATED HUMAN PLASMIN 1059 of the lots of bovine fibrinogen tested. Although plasmin specimen assayed at 350,000 units per mg. each lot differed somewhat quantitatively in the N. After purification of this fibrinogen, the SK amount of bovine plasminogen present, they were plasmin assayed at only 820 units per mg. N, or qualitatively similar. approximately one four-hundredth the original That these findings are attributable to a con- assay. tamination of the fibrinogen with bovine plasmino- Plasminogen-deficient bovine fibrinogen could gen rather than associated with bovine fibrinogen be readily prepared by subjecting the Armours itself was demonstrated, by studying the digestion Fraction I to a three-fold repurification according of casein, or the acceleration of fibrinolysis, with to the Laki technique (15). When this prepara- the fluid expressed after the bovine fibrinogen was tion was used as a substrate in the fibrinolytic clotted with thrombin. Although some of the bo- assay, the assay of SK activated human plasmin vine plasminogen was adsorbed on the clot, suffi- was similar to that found with purified human cient amounts were present in the supernatant to fibrinogen. confirm these observations. A number of at- The bovine thrombin preparations used in the tempts have been made to fractionate and con- fibrinolytic assay can also be readily demonstrated centrate the bovine plasminogen in the fibrinogen, to contain bovine plasminogen. However, in the by the procedure used for human plasminogen. amounts used in the assay (1 thrombin unit) in- The bovine plasminogen, however, appears to be significant amounts of the contaminating plasmino- much more labile, readily denaturing at acid pH's. gen are added, and do not affect the assay. How- We are pursuing this phase of the investigation, ever, if ten or more units of thrombin are added to however, in order to determine whether the bo- the clot, a significant effect on the assay becomes vine plasminogen has similar characteristics to the apparent. human enzyme. Further evidence for the contamination of bo- V. The influence of the bo7Ane plasminogen con- vine fibrinogen with bovine plasminogen was ob- taminant on the fibrinolytic assay for strepto- tained by purification of the bovine fibrinogen. kinase activated plasmin When more highly purified bovine fibrinogen It should be re-emphasized that the fibrinolytic preparations were used as substrates in the fibrino- assay for streptokinase activated plasmin is car- lytic assay, there was a sharp reduction in the ried out by serially diluting the human proenzyme measurable fibrinolytic activity of streptokinase preparation and then adding an equal amount of activated human plasmin. For this experiment two bovine fibrinogen, SK and bovine thrombin to samples of bovine fibrinogen (Armours Fraction I, each of the assay tubes. Since the rapid activation and a sample of fibrinogen obtained through the of the bovine plasminogen contaminant of bovine courtesy of Dr. Walter Seegers) were further fibrinogen requires only traces of human plasmin purified by ammonium sulfate fractionation ac- in the presence of SK, one would predict that cording to the technique described by Laki (15). while the addition of SK to a fully activated human A purified human plasminogen preparation ' was plasmin preparation would not enhance the pro- assayed for its streptokinase activatable fibrinolytic teolytic activity, it should strikingly enhance the activity utilizing each of the four different fibrino- conventional fibrinolytic assay. In order to test gens at the same clottable protein concentration this hypothesis, the spontaneously active human (0.15 per cent fibrinogen in the final clot). The plasmin preparation, Fraction III., was studied in fibrinolytic assay of the SK activated human plas- the presence and absence of SK. This preparation min with the original Armours Fraction I was had a spontaneous proteolytic activity of 4.3 casein 175,000 units per mg. N. With the purified Frac- units per mg. N, which did not increase upon the tion I, the assay fell to 4,400 units per mg. N, or addition of 1,000 units of streptokinase to each one-fortieth of the original assay. With the origi- ml. of the enzyme preparation. However, the nal Seegers fibrinogen, the SK activated human fibrinolytic assay of this preparation before the ad- 9 This preparation, following SK activation, had 60 dition of SK was 32 units per mg. N, and was in- casein units per mg. N. creased to 16,000 units per mg. N following the ad- 1060 SOL SHERRY

50 2 I 3 I 0 0- 45 I c I %._ 40

0 35 30 C I ._0 25 0 I m 20 I

20 15- 10.

._0 5. ._ ~~~~~~~~~~ ~~~~I I _ _ 506 250 12563 31 16 8 4 2 1 .5 .25 .12 .06 Gamma of Human Plasmin N FIG. 3. LYsIs TnME OF BOVINE CLOT AS A FUNCTION OF HUMAN PLASMIN CONCENTRATION (IN THE PRESENCE AND ABSENCE OF SK) Note that the abscissa plots successive serial dilutions of human plasmin N. See text for details. dition of SK. The lysis times of the bovine clots plasmin plus significant amounts of proteolytic with each serial dilution of the human plasmin activity derived from the human plasmin. In zone preparation, in the presence and absence of SK, two, the amounts of human plasmin present do not is shown in Figure 3. significantly affect fibrinolysis, but are adequate, Observations on the lysis time of the bovine in the presence of SK, for complete activation of clot, by the human plasmin preparation alone, re- the bovine plasminogen. In zone three, the con- vealed the expected relation of fibrinolysis time to centration of human plasmin has finally become enzyme concentration as seen with other proteo- a limiting factor in the activation of bovine plas- lytic enzymes. Upon the addition of SK however, minogen, resulting in a progressive prolongation the pattern of fibrinolysis time to enzyme concen- of the clot lysis time as a function of human plas- tration was distinctly different. There was an min concentration. initial zone where the lysis times were very rapid, Note that the activating function of human but related to enzyme concentration; followed by plasmin in the presence of SK, on bovine plas- a second zone where the lysis time was constant minogen, in this experiment, is 500 times the actual even though the human plasmin concentration was fibrinolytic activity of the human plasmin. reduced 16-fold; followed by a third and final zone where concentration was again related to lysis time. VI. Streptokinase as an essential component of The kinetics of these observations on clot lysis the activating system may be explained by the presence of the contami- Although not shown in Figure 3, and as would nating bovine plasminogen in constant amounts in be expected, the lysis time of human clots, by all the assay tubes, and its activation by the human various concentrations of the spontaneously acti- plasmin plus SK. The horizontal broken line in vated human plasmin preparation Fraction III,, Figure 3 may be interpreted as representing the was not effected by the addition of 1,000 units of lysis time of the clot which would result from the SK, and was similar to that observed for the bovine maximal activation of the contaminating bovine clots in the absence of SK. These latter observa- plasminogen present. The first zone represents tions suggest that human plasmin per se does not the fibrinolytic activity of the activated bovine rapidly activate bovine plasminogen, and that SK FIBRINOLYTIC ACTIVITY OF SK ACTIVATED HUMAN PLASMIN 1061 is an essential component of the activating system. plasma is not restricted to fibrinogen and fibrin This was more effectively demonstrated by study- is evident from the recent studies of Pillemer, Rat- ing a preparation of purified human plasminogen noff, Blum, and Lepow on complement destruc- which had ben activated to plasmin by SK; then tion (20), and of Ferguson, on the disappearance inactivating the SK without destroying the plas- of Ac globulin (21). min; and finally adding fresh SK to restore that It has recently been suggested that the intra- lost by the inactivating procedure. The inactiva- venous injection of purified human plasminogen, tion of SK in the presence of purified human which has been activated in vitro by SK, may prove plasmin was accomplished by heating the prepara- useful as a clot dissolving agent in man (5). The tion in a boiling water bath for 15 minutes at pH basis for this view was the observation of signifi- 2.0. Preliminary observations had shown that cant circulating proteolytic and fibrinolytic activity plasmin was stable at pH 2.0 under these condi- following the injection of SK activated human tion,10 whereas SK, in the presence of human plas- plasmin into dogs. These observations however, min, was fairly rapidly inactivated by this pro- have not taken into account, that this type of mix- cedure. ture activates the dogs plasminogen in vivo. We A human plasminogen preparation containing have used this procedure to activate the animals 0.15 mg. N. per ml., and to which 17,000 units of own plasminogen in vivo, and to evaluate plasmin SK had been added, was allowed to activate at room as a clot dissolver (3). The injection of previ- temperature for 10 minutes. Assay of the plasmin ously activated purified human plasminogen for formed revealed 90,000 fibrinolytic units (bovine clot dissolution in man, although deserving trial, clot) and 28.9 casein units per mg. N. Following may not be as promising as suspected, because of 15 minutes heating at pH 2.0 in a boiling bath, the the low order of activity of the present prepara- proteolytic assay of this preparation, when subse- tions, as well as the obvious limitations on the quently adjusted to the onrginal pH, was not sig- amount of material available. However, since nificantly altered, but the fibrinolytic activity had plasminogen is the naturally occurring proteolytic fallen to 1,000 units per mg. N. Readdition of enzyme precursor of plasma, it would still appear 17,000 units of fresh SK restored the fibnrnolytic that the in tivo activation of this enzyme in man, assay to its original 90,000 units per mg. N. Since by the introduction of SK, may be the most desir- the fibnnolytic assay, under the conditions of the able method for directly effecting the enzymatic experiment cited, was primarily a measure of the dissolution of thrombi and emboli. activating function of the preparation studied, SK The observations described in this article reveal appears to be a necessary component of this acti- that the present assay using bovine fibrinogen and vating system. thrombin for the fibrinolytic activity of strepto- DISCUSSION kinase activated human plasmin (14) is not a measure of the actual fibrinolytic activity of plas- On the basis of the observations cited in this mn. This assay measures the activator function report, it appears that the fibrinolytic activity of of human plasmin rather than its fibrinolytic ac- human plasmin is comparable to that which has tivity. The activator function in the presence of been demonstrated for a number of other proteo- SK can be measured with approximately 1/500 of lytic enzymes. In plasma, plasmin appears to at- the amount of enzyme actually required to lyse a tack fibrin more readily hn fibrinogen but this clot in thirty minutes. The measurement of the may be due, as Ratnoff has suggested, to its ability actual fibrinolytic activity of SK activated human to activate more readily in the presence of fibrin plasmin requires more purified fibrinogen than (19) or to local factors existing on a plasma clot. those commercially available. The data presented indicate that the fibrinolytic Similarly the present assay for streptokinase activity of plasmin, in a more purified system, is should be re-evaluated. The unit of streptokinase predictable on the basis of its general proteolytic is defined as the smallest amount of SK, which in properties. That the action of this enzyme in the presence of human plasminogen will produce 1O True for purified plasmin preparations only, and not lysis of a bovine clot in 10 minutes or less, and for plasma. was conceived as representing the amount of 1062 SOL SHERRY kinase needed to rapidly activate sufficient human tor now converted human and all other animal plasminogen to lyse a clot in the period noted (14). plasminogens into plasmin. We are inclined to Our data suggest that the unit of SK may repre- identify this activator function of human plasma, sent the amount of SK required in the presence in the presence of SK, to human plasminogen or of human plasminogen to rapidly activate the ma- plasmin itself. This viewpoint is based on a) jor portion of the bovine plasminogen contaminant purification data (12); b) the trace quantities of present in the bovine reagents. purified human plasminogen or plasmin prepara- The observations reported on in this article are tions required; c) plasmin or plasminogen inacti- of some interest in relation to the recent literature vation causing the loss of activator function; and concerning a) the mechanism of activation of d) inactivation of streptokinase, following its ad- plasminogen by streptokinase, b) the presence of dition to human plasminogen or plasmin, results a proactivator system in blood for plasminogen in an immediate loss of activator function; and is activation, and c) the participation of a comple- promptly restored by the addition of fresh SK. ment-like factor in the streptokinase activation of The recent suggestion of Geiger that a comple- human plasminogen: ment-like factor participates in the activation of Recently, the original studies of Christensen and human plasminogen (25) was based on the marked MacLoed (22) on the kinase nature of activation increment of proteolytic activity which followed of plasminogen by SK, have been seriously ques- the addition of fresh animal sera to a mixture of tioned (23). Wasserman, employing the fibrino- SK and heated human globulin preparations. The lytic assay, showed that the kinetics of the activa- heating of the human globulin was shown to in- tion of plasminogen by SK, is compatible with the activate most, but not all, of the human plasmino- interpretation of a stoichiometric reaction, and gen. The observations cited in this report suggest that the formation of plasmin does not conform to that Geiger's results may be alternatively explained the kinetics of an enzymatic reaction. The amount as a demonstration of the activation of the plas- of plasmin formed, in the presence of an excess of minogen of the various animal sera by the mix- plasminogen, was found to be dependent on the ture of small amounts of human plasminogen and SK concentration, and that at all levels of strepto- SK, without invoking the participation of another kinase, maximal plasmin formation occurred al- factor. most instantaneously and did not increase with The mechanism for the activation of animal prolonged incubation. plasminogen by streptokinase in the presence of Although we have been able to confirm Wasser- human plasminogen or plasmin is obscure. It can man's observations, they may not yield information be shown that the presence of animal plasminogens on the kinetics of human plasminogen activation does not enhance the ability of SK to activate hu- by SK. In the fibrinolytic assay, these kinetics man plasminogen. Our working hypothesis is may be masked by the activation of the bovine that the plasma inhibitor to the activation of the plasminogen contaminating the reagents. Studies various animal plasminogens by streptokinase has on the kinetics of plasminogen activation by strep- a much greater affinity for human plasminogen or tokinase are also hampered by the impurities which plasmin than it does for its respective plasminogen. are present in the available SK preparations. Pre- The addition of the human factor results in a re- liminary data obtained with plasminogen-deficient lease of the animal plasminogen for activation by fibrinogen are still in agreement with the initial streptokinase. An alternative view, also under observations of Wasserman, and suggest that the exploration, is that the rapid activation of plasmi- mechanism by which streptokinase activates hu- nogen is carried out by a SK-plasmin complex, and man plasminogen may require further clarification. that human plasmin combines more readily with The demonstration of bovine plasminogen acti- SK in this respect than the plasmin of other ani- vation by SK in the presence of human plasma, mals. by Mullertz and Lassen (24) led him to hypothe- SUMMARY size the presence of a proactivator in human plasma 1. The ratio of fibrinolytic to proteolytic ac- which was activated by SK, and that this activa- tivity of human plasmin, in more purified systems, FIBRINOLYTIC ACTIVITY OF SK ACTIVATED HUMAN PLASMIN 1063 is of the same order as that shown by trypsin 10. Remmert, LeM., and Cohen, P. P., Partial purification and chymotrypsin. and properties of a proteolytic enzyme of human serum. J. Biol. Chem., 1949, 181, 431. 2. Streptokinase, in the presence of minute 11. Todd, E. W., Quantitative studies on the total plasmin amounts of human plasminogen or human plasmin, and the trypsin inhibitor of human blood serum. activates the bovine plasminogen contaminating I. Methods for the titration of total plasmin and the bovine fibrinogen in the fibrinolytic assay. of trypsin inhibitor. J. Exper. Med., 1949, 89, 295. 12. Troll, W., Sherry, S., and Wachman, J., The action 3. The fibrinolytic assay for streptokinase acti- of plasmin on synthetic substrates. J. Biol. Chem., vated human plasmin, as presently employed with 1954, 208, 85. bovine fibrinogen and thrombin, measures the ac- 13. Loomis, E. C., George, C., Jr., and Ryder, A., Fibri- human rather than nolysin: Nomenclature, unit, assay, preparation and tivator function of the enzyme properties. Arch. Biochem., 1947, 12, 1. the fibrinolytic activity of the human plasmin. 14. Christensen, L. R., Methods for measuring the activity of components of the streptococcal fibrinolytic ACKNOWLEDGMENT system, and streptococcal desoxyribonuclease. J. Clin. Invest., 1949, 28, 163. The author acknowledges with gratitude the many ex- 15. Laki, K., The transition of fibrinogen to fibrin in cellent suggestions of Dr. Walter Troll, and the techni- Blood Clotting and Allied Problems, Transactions cal assistance of Miss Mary E. Schmidt in this study. of the Fourth Conference, 1951, ed. by J. E. Flynn, Josiah Macy, Jr. Foundation, New York, 1951, REFERENCES p. 217. 16. Bergmann, M., Fruton, J. S., and Pollok, H., The 1. Tillett, W. S., and Sherry, S., The effect in patients specificity of trypsin. J. Biol. Chem., 1939, 127, of streptococcal fibrinolysin (streptokinase) and 643. streptococcal desoxyribonuclease on fibrinous, puru- 17. Christensen, L. R., and Smith, D. H., Jr., Plasmino- lent, and sanguineous pleural exudations. J. Gin. gen purification by acid extraction. Proc. Soc. Invest., 1949, 28, 173. Exper. Biol. & Med., 1950, 74, 840. 2. Johnson, A. J., and Tillett, W. S., The lysis in rab- 18. Milstone, H., A factor in normal human blood which bits of intravascular blood clots by the streptococcal participates in streptococcal fibrinolysis. J. Im- fibrinolytic system (streptokinase). J. Exper. Med., munol., 1941, 42, 109. 1952, 95, 449. 19. Ratnoff, 0. D., Studies on a proteolytic enzyme in 3. Sherry, S., Titchener, A., Gottesman, L., Wasserman, human plasma. IX. Fibrinogen and fibrin as sub- P., and Troll, W., The enzymatic dissolution of strates for the proteolytic enzyme of plasma. J. intravascular clots. J. Lab. & Clin. Med., 1953, 42, Cin. Invest., 1953, 32, 473. 952 (Abst.). 20. Pillemer, L., Ratnoff, 0. D., Blum, L., and Lepow, 4. Innerfield, I., Schwarz, A., and Angrist, A., Intra- I. H., The inactivation of complement and its venous trypsin: its anticoagulant, fibrinolytic, and components by plasmin. J. Exper. Med., 1953, thrombolytic effects. J. Gin. Invest, 1952, 31, 97, 573. 1049. 21. Ferguson, J. H., in Flynn, J. E., and Standley, E. T., 5. Cliffton, E. E., Cannamela, D. A., and Grossi, C., In On the conversion of prothrombin to thrombin in vivo studies of human plasmin. Intravenous in- Blood Clotting and Allied Problems, Transactions jection in dogs and rabbits. J. Applied Physiol., of the Second Conference, 1949, ed. by J. E. Flynn, 1953, 6, 143. Josiah Macy, Jr. Foundation, New York, 1949, p. 174. 6. Biggs, R., and MacFarlane, R. G., Human Blood 22. Christensen, L. R., and MacLoed, C. M., A proteo- Coagulation and Its Disorders. Springfield, Ill., lytic enzyme of serum: Characterization, activa- Charles C Thomas, 1953. tion, and reaction with inhibitors. J. Gen. Physiol., 7. Cliffton, E. C., and Cannamela, D. A., Fibrinolytic 1945, 28, 559. and proteolytic activity of a human 'plasminogen,' 23. Wasserman, A. E., Streptokinase activation of a pro- prepared from fraction III of human plasma. J. teolytic enzyme in human blood. Arch. Biochem. Applied Physiol., 1953, 6, 42. & Biophys., 1952, 41, 158. 8. Kline, D. L., The purification and crystallization of 24. Mullertz, S., and Lassen, M., An activator system in plasminogen (profibrinolysin). J. Biol. Chem., blood indispensable for formation of plasmin by 1953, 204, 949. streptokinase. Proc. Soc. Exper. Biol. & Med., 9. Christensen, L. R., Streptococcal fibrinolysis: A 1953, 82, 264. proteolytic reaction due to a serum enzyme acti- 25. Geiger, W. B., Involvement of a complement-like vated by streptococcal fibrinolysin. J. Gen. Physiol., factor in the activation of blood protease. J. Im- 1945, 28, 363. munoI., 1952, 69, 597.