Studies on a Proteolytic Enzyme System of the Blood. Ii. Fibrinolysokinase Activators for Profibrinolysin

STUDIES ON A PROTEOLYTIC ENZYME SYSTEM OF THE BLOOD. II. FIBRINOLYSOKINASE ACTIVATORS FOR PROFIBRINOLYSIN

Jessica H. Lewis, … , Ann C. Howe, Jane Rogers

J Clin Invest. 1950;29(8):1059-1068. https://doi.org/10.1172/JCI102337.

Research Article

Find the latest version: https://jci.me/102337/pdf STUDIES ON A PROTEOLYTIC ENZYME SYSTEM OF THE BLOOD. II. FIBRINOLYSOKINASE ACTIVATORS FOR PROFIBRINOLYSIN 1

By JESSICA H. LEWIS AND JOHN H. FERGUSON WITH THE TECHNICAL ASSISTANCE OF ANN C. HOWE AND JANE ROGERS (From the Department of Physiology, University of North Carolina, Chapel Hill) (Submitted for publication March 3, 1950; accepted, May 8, 1950) Fleisher and Loeb (1), in 1915, described assay method have needed investigation. Since fibrinolysis of plasma clots (rabbit, cat, guinea- factors, in particular the thromboplastic protein pig) during aseptic contact with fragments of vari- described by Chargaff (14), modifying blood- ous animal tissues. Their assumption that this clotting reactions are noted in tissue preparations, fibrinolysis was simply due to a tissue proteolytic some preliminary tests of these have been made enzyme may be questioned in the light of more with our tissue fractions, including an experi- modern work on fibrinolytic mechanisms. In mental separation of fibrinolysokinase from throm- particular, Christensen (2, 3), following the earlier boplastic activity. lead of Milstone (4), clearly showed that plasma contains a precursor (plasminogen or profibrinoly- MATERIALS sin) which, in the human, can be activated to a Materials used in our standard assay procedures for fibrinolytic enzyme by streptokinase. "Fibrino- fibrinolysin and antifibrinolysin, viz., borate buffer, pH = kinase" substances from various tissues have 7.7; thrombin; "fibrinogen" (BF: Armour's bovine been described by Astrup and Permin (5) and plasma fraction-I); and dog serum fibrinolysin, are fully Permin (6). Tagnon and Petermann (7) described in the first paper of this series (15). recently Profibrinolysin (prolysin): Recalcified hemoglobin-free fractionated rat lung tissue by differential centrifu- oxalated plasma, defibrinated in bottles with the aid of gation, following the principles established by glass rods and a mechanical shaker, gave a good yield of Claude (8), and found kinase activity in the par- clear serum, which was cooled to O° C., brought to one- ticulate (notably "microsome") fraction and in- third saturation with ammonium sulfate (saturated solu- hibitor in the supernate. tion at 00 C.) and kept at zero temperature for 30-60 minutes. The precipitate, recovered by centrifugation, Our own studies employ similar methods of was dissolved in distilled water and dialyzed 18 hours tissue fractionation, applied, in the first place, to against cold running tap water (40 C.). The final an extensive survey of dog tissue fractions, with precipitate, redissolved in borate buffer to one-fourth testing for fibrinolysokinase activity on dog serum the original serum volume (corrected for diluents), con- profibrinolysin. Previous studies of bacterial ac- tained most of the prolysin, very little active lysin, and no demonstrable antilysin. It was used immediately or tivators (9), streptokinase (10) and staphyloki- stored at -20° C. nase (11, 12) emphasize species variability in the Saline-phosphate buffer: 0.85 per cent sodium chloride, susceptibility of prolysins to bacterial lysokinases, buffered at pH = 7.3 with 0.0025 M phosphate, was used suggesting the need for similar comparative stud- for tissue extractions. Fibrinolysokinase (lysokinase), tissue fractions: Under ies with tissue lysokinases. Having found certain nembutal anesthesia, animals were exsanguinated and the lung preparations among the most active (cf. Tag- various tissues dissected out as free as possible from fat non and Petermann [13]) for dog prolysin, we and connective tissue, washed with saline-buffer, drained have, in this study, tested lung fractions from eight on filter paper, and the (wet) weight recorded. Per- fusion with cold saline-buffer in situ (brain) or isolated species on prolysins from three (dog, human, (other organs) removed practically all the blood from bovine). In testing for fibrinolysokinase activity brain, lungs, liver, and kidney, but left some residual some important technical considerations of the blood in the spleen. A 10 gm piece of the larger organs, or as much as possible of the smaller tissues, was rapidly 1 This investigation was supported by a research macerated in a Waring blendor with frozen saline-buffer, grant from the Division of Research Grants and Fellow- then passed through a 1 mm sieve, and thoroughly ships of the National Institutes of Health, U. S. Public reground with pestle and mortar. The smallest speci- H!ealth Service. mens were cut up and ground directly in the chilled 1059 1060 JESSICA H. LEWIS AND JOHN H. FERGUSON mortar. The mixture of tissue pulp plus saline-buffer constant and the fibrinogen (fibrin) concentration varied was further diluted to 40 ml (or proportionately) and to give lysis times which are inversely proportional to subjected to differential centrifugation (see below). the fibrinogen concentration (15, 16). We have recently Each sedimented fraction was washed three times with employed this principle to extend the range of sensitivity saline-buffer, recentrifuging as before, and finally sus- of our lysin assay method (15) by simply varying the pended in the diluent to volume (ml) equal to original fibrin concentration in the standard clot. Choosing 0.2 specimen weight (gm). and 1.0 per cent clots ("per cent" referring to the dry As 2 ml of each fraction were needed for testing, it weight of BF [see reagents] in the 0.5 ml "fibrinogen" was necessary to dilute fractions of pituitary (0.044 gm), solution used in the standard clot), permits an assayable adrenal cortex (1.0 gm), adrenal medulla (scraped from range of lysin units from 1-300. If a 100 unit lysin, two dog adrenals), rib bone-marrow (1.86 gm), ureter which completes dissolution of the 1.0 per cent clot in (1.0 gm), leucocytes (0.25 ml "cream" packed from 800 five minutes, is diluted to 20 units, it will lyse the 0.2 ml blood), and platelets (packed from 100 ml plasma and per cent clot in five minutes also, and the points of a microscopically free from corpuscles). "Fractionation" single standard reference curve (cf. 15), plotting lytic of some of the tissues (see Table I) frequently produced units against lysis times, hold both for a dilution series only certain of the sedimented fractions. of original lysin tested on 1.0 per cent clots and for a Differential centrifugation employed 1.) refrigerated parallel series of diluted lysin tested on 0.2 per cent clots, International Centrifuge, Model PR-1; 2.) Servall High as shown by the data in Figure 1. However, when a Speed Vacuum Centrifuge, Type SS-2. similar comparison was made between the two strengths Fraction I: the diluted tissue pulp (see above) was of clot using prolysin-fibrinolysokinase mixtures instead spun in 50 ml tubes at 2500 rpm (1900 g) for three of fully activated fibrinolysin, some deviations occurred, minutes, typically separating two layers of sediment, the in all probability (see below) due to activation of a bottom (Fraction I) consisting of gross particles, con- prolysin contaminant of the BF, which differs, of course, nective tissue, and large cellular fragments. with the clot dilution. Hence, for comparative assays of Fraction II: the fine layer on the surface of I, de- various tissue preparations, the same substrate, usually scribed by Claude (8) as consisting largely of cell nuclei. 0.2 per cent clots, was used throughout. Fractions III-A and III-B: the respective sediments Fibrinolysokinase (lysokinase) assay: In order to ob- in two successive recentrifugations of the supernatant, tain maximum fibrinolysokinase effects, it has been found again at 2500 rpm for three minutes each. These were necessary to employ as the profibrinolysin a dog serum small in volume. fraction (see Materials) free from antifibrinolysin and Fraction IV: corresponding to Claude's "large gran- with minimal spontaneous activity, under the test con- was the obtained from the ules," granular sediment pre- ditions, but able to yield a potent fibrinolysin in preceding supernatant by 30 minutes centrifugation at 3500 liminary tests with staphylokinase activator. Activation rpm (2700 g). of prolysin by tissue kinases requires considerable time Fraction V: was a small additional precipitate ob- and is complicated by deterioration of the formed lysin. tained on repeating the preceding centrifugation on the supernatant from IV. At 370 C., the rapidity of activation may be offset by the Fraction VI: corresponding to Claude's "microsomes," subsequent lysin deterioration, while at 40 C., activation was the small sediment, usually homogenous in ap- pearance and often amber-colored, obtained by very high- FRACTIONI I% 02% speed centrifugation (up to 17,000 rpm: 48,000 g) of the preceding supernatant, for 90 minutes. u 300 60l Fraction VII: the final supernatant, was typically a z 250 50 clear fluid, sometimes colored perhaps with a trace of z hemoglobin. 0200 40I A single lung fraction, called (IV + VI), was used in o-j Z 150 30. later experiments. It was prepared by macerating the 21lo 20 perfused lung in a Waring blendor, straining to remove fibrous tissue, cartilage, etc., and centrifuging for 15 5010 minutes at 2500 rpm, discarding the sediment. The super- 4812 I161 20 1 2 3 4 5 natant was then recentrifuged at 20,000 g for 90 minutes, MINUTE HOURS the sediment collected as a single fraction, washed, and, LYSIS TIME employing the Waring blendor, suspended in a volume (ml) of saline buffer equal to one-half the original lung FIG. 1. LYsis TIMES OF 1 PER CENT AND 0.2 weight (gm). PER CENT CLOTS 2.2 ml of clotting mixture contained 1) 0.5 ml 1 per METHODS cent bovine fraction I (x-x) and 0.5 ml dilutions of full-strength dog fibrinolysin; 2) 0.5 ml 0.2 per cent BF Fibrinolysin (lysin) assay: Within definable experi- (0-*) and 0.5 ml dilutions of one-fifth strength mental limits, the fibrinolysin concentration may be kept enzyme. STUDIES ON A PROTEOLYTIC ENZYME SYSTEM OF THE BLOOD. II 1061

tion. The second possibility will be considered more 600 ACTIVATION OF DOG PROLYSIN fully in a later communication, but it may be stated that BY LUNG I31 -MI , 0-e 371C a number of experiments have failed to yield any evidence cn500- x- -x 40C tt) 500 °---° CONTROLS - of "autocatalytic" lysin formation in mixtures contain- Z 400_ ing both active enzyme and the prolysin. The third possibility is considered the most likely and novel and z I -300 , receives support from the following (Figure 3): A 4 per cent solution of BF was clotted with 100 units of bovine thrombin and the "serum" (BFS) ir o expressed and used in the following tests: I) BFS + fraction i100 . * (IV + VI) (dog lung); II) BFS + saline-buffer; III) _------_-_ ------fraction (IV+ VI) (dog lung) + borate buffer. These 0 1 2 3 4 mixtures were incubated at 370 C. and sampled at in- INCUBATION PERIOD (HOURS AT 3rC; WEEKS AT 4°C) tervals for the usual tests of fibrinolytic activity.

FIG. 2. TEMPERATURE EFFECTS ON ACTIVATION OF DoG PROLYSIN BY DoG LUNG FIBRINOLYSOKINASE 35 37°C is very much slower (see Figure 2) but deterioration cl) 30- is almost completely prevented, resulting in larger yields Z 25* of lysin and more significant "maxima." These con- z 20- siderations led us to accept a compromise most suited to the particular experiment and the incubation conditions are therefore explicitly stated in each case. In actual practice, three mixtures: 1) 0.5 ml prolysin + In -I115 0.5 ml tissue fraction, 2) 0.5 ml prolysin + 0.5 ml saline- z 5-0 it buffer, 3) 0.5 ml tissue fraction+0.5 ml borate buffer, 0 N O were incubated (as stated) and then tested for time of lysis when added to the standard clot. Controls 2) were INCUBATION PERIOD (HOURS) negative or negligible (< 0.5 unit), while controls 3) frequently showed some minor lytic activity especially FIG. 3. ACTIVATION OF BOVINE FRACTION I "SERUM"'BY with the more active tissue kinases (see below). DoG FIBRINOLYSOKINASE (see text) Cause of lytic activity in tissue fractions: The traces of lytic activity in the tissue controls (without added The data (Figure 3) clearly show a progressive for- prolysin) might suggest three possibilities: 1) that the mation of lysin in I, as compared with the negative tissue fractions contain active lysin as well as lysokinase, results in II and III, strongly suggesting the presence 2) that active lysin (apart from kinase) might activate of some prolysin in the "serum" (expressed fluid) from prolysin through some "autocatalytic" mechanism, or 3) the clotted BF. that the bovine "fibrinogen," used for the test clots, may be contaminated with some prolysin which may react RESULTS with tissue lysokinases even though bacterial fibrinolyso- I. Survey of fibrinolysokinase activity in various kinases fail (9). The first possibility cannot be wholly ruled out pending the preparation of a clot substrate tissue fractions (Dog): which is completely free from prolysin, as shown by a Table I presents the results of an extensive negative test becoming positive only when a known survey of the various fractions of tissue obtained prolysin is added. There is no doubt about the inability of bacterial kinases to act in the absence of prolysin. from dogs. Each tissue fraction was incubated We have therefore taken plasma from dogs (a suitable with a) dog profibrinolysin and b) buffer for one species) and attempted to prepare a "fibrinogen" which hour at 290 C. with subsequent assay of lysin. shows no fibrinolysin formation on treatment with Kinase units were obtained by subtracting from staphylokinase. However, the following procedures were the total units found in a) the sum of the units in unsuccessful: 1) Ten to 12 reprecipitations at 25 per cent saturation with ammonium sulfate (4); 2) freezing and b) and in the prolysin control. Subsequent ex- recovery of the cold-precipitable fibrinogen (17); 3) periments demonstrated that the one hour incuba- adsorption with BaSO4, BaCOs, CaaP208, asbestos, and tion at 290 C. was not optimal but even the low charcoal. These studies are continuing but, until we kinase titers obtained suggest which tissues are the obtain a more satisfactory substrate, we are proceeding most potent sources of fibrinolysokinase. with the use of bovine plasma fraction I (partially purified fibrinogen) in our investigations on fibrinolysis , Lung and uterus (esp. during pregnancy) pro- and deferring the final answer to the tissue protease ques- vide the most active fractions. One or more 1062 JESSICA H. LEWIS AND JOHN H. FERGUSON

TABLE I other fractions but actually contains less organic Survey of fibrinolysokinase activity of dog tissue fractions* material (dry weight less ash weight) than frac- Kinase units per ml tion I. Fraction I I_ Ill-A III-B IV V VI VII II. Antilysin (antifibrinolysin) and prolysin (pro- Adrenal cortex 1.4 1.4 0.8 0.6 1.0 0 0.6 0 Adrenal medulla 0.8 0.8 0.6 0.2 0.8 0.2 0 0 fibrinolysin) titers in tissue fractions: Artery 0.6 0 0 0 0 0 0.2 0 Bone marrow 0.8 - 0.8 - 2.6 0 0.8 0 Brain 2.0 2.6 0.4 0 1.2 0.6 0.8 0.4 Antilysin titers of all fractions were determined, Gall bladder 2.6 - 2.0 1.4 3.0 1.6 4.0 0 Heart 0 0 0.4 0.2 0.8 0.4 0.6 0 against dog serum lysin, using the technic previ- Intestine 1.8 1.0 - - 0.4 0.4 1.0 0 Kidney 0.8 0.2 0 0 0.8 0.2 0.6 0 ously described (15). Fractions VII of all tis- Liver 0.8 0.8 0 0 1.0 0 0.6 0 Lung I 2.2 2.6 3.0 4.0 6.0 3.0 3.4 0.8 sues, even those thoroughly perfused, showed II 4.8 2.4 1.2 0.8 7.8 5.2 8.8 0 III 4.0 1.4 0.8 - 8.0 5.4 7.4 0 some antilysin (100-400 units/ml). Fractions I Lymph gland 4.8 0.4 - - 2.6 0.6 3.4 0 Mennges 2.0 - 1.0 - 1.4 0.6 1.0 0.8 and VI frequently showed significant antilysin Muscle 0 0 0 0 0 0 0 0 Omentum 0.6 - 0.4 0 - 0.4 0.8 0 titers (80-200 units/ml), even in materials that Ovary I 1.4 - 0.8 0.8 0.8 1.0 0.8 0 (Pregnant) II 3.0 0.9 - - 1.2 0.6 1.2 0 were also active lysokinases. Pancreas 3.3 3.4 3.4 3.2 3.2 0 0 0 Placenta 2.6 1.0 0 0 0 0 0 0 Prolysin could not be detected in any of the Pericardium 1.2 - 1.2 0.4 0.8 1.0 0.6 0 Pituitary 1.2 - 1.2 0.6 1.6 0.2 0.4 0.4 fractions after treatment with staphylokinase. Seminal vesicles 1.2 0.6 0.8 0.4 0.8 0.4 0.8 0 Spinal cord 2.0 - 1.0 0.8 1.2 1.0 1.0 0.2 Spleen 0.6 0.2 1.6 0.6 1.0 0.8 0 0 Stomach 1.0 1.0 - - - 0.8 1.0 0.4 TABLE III Testis 1.0 0.8 0.8 0.4 0.8 0.8 1.0 0 Thyroid 1.4 0.8 0.6 0.6 0.4 0.6 0.6 0 Lysin units/ml obtained after incubation for one hour at 370 C Ureter 2.2 - 0.8 0 0.8 0.2 0.8 0 of the three prolysins with staphylokinase, streptokinase and Urinary bladder 2.0 1.4 - - - 1.0 0.8 0 Uterus I 4.0 5.0 2.0 1.0 2.6 1.0 1.0 1.0 lung fractions VI obtained from various species (Pregnant) II 5.8 2.2 0.8 - 1.2 0.6 1.8 0 Vein 3.0 - - 0 0.2 0 0.2 0 Erythrocytes 0.6 - 0.6 - - - 0 0 Activators Human Dog Bovine Borate Leukocytes 1.0 - - 0 0.8 - 0.6 0 prolysin prolysin prolysin buffer Platelets 0.4 - - 0 0 - 0 0 Staphylokinase 320.0 640.0 0 0 *- No fraction. Streptokinase 252.0 tr. 0 0 0 No Mouse VI 26.0 56.0 4.0 2.7 activity. Rat VI 3.2 19.0 3.2 2.7 Guinea pig VI* 3.3 14.0 2.8 tr. "good" fractions are noted with pancreas, gall- Rabbit VI* tr. 10.8 tr. 0 Cat VI* 7.8 42.4 5.3 2.2 bladder, ovary (with corpus luteum of pregnancy), Dog VI* 5.4 42.4 5.7 2.2 lymph gland, placenta, urinary bladder, ureter, Human VIt 18.0 40.0 11.8 5.2 Bovine VI 0 tr. tr. 0 vein, bone-marrow, brain, spinal cord, and menin- Buff. saline 0 tr. 0 0 ges. In general, fraction I was most active, with fraction IV and VI not far behind. Fraction VII * Perfused free from blood. t Obtained at operation through the courtesy of Dr. was usually inactive. William Sealy, Duke Medical School. Table II shows the fibrinolysokinase units ob- Trace (tr.) = <0.5 units/ml. tained from the various dog lung fractions, and the single lung fraction (IV + VI), prepared from III. Fibrinolysokinase activity of fractions pre- the same lung, and compared to staphylokinase. pared from lungs of various animal species, as In each case the kinase was allowed to react with tested on human, dog, and bovine prolysins: the prolysin for one hour at 37° C. The increased A. The centrifugally prepared tissue fractions temperature of incubation has increased the lysin (I-VII) from lungs of six mice, two rats, and yield. The single fraction (IV + VI), which one each of the other six animals listed in Table III produces very marked activation, is, in relation to were tested, under the experimental conditions original lung weight, double the strength of the stated, against prolysins prepared in the same man-

TABLE II Fibrinolysokinase units/ml of dog lung fractions and staphylokinase

Fraction: I II Ill-A Ill-B IV V VI VII (IV + VI) Staph. Units/ml 43.6 3.0 2.4 2.9 19.2 5.0 27.6 2.1 317.6 900 STUDIES ON A PROTEOLYTIC ENZYME SYSTEM OF THE BLOOD. II 1063 ner from sera of human, dog, and bovine origin. ing recorded. At the 40 C. temperature the ac- Streptokinase and staphylokinase tests and con- tivation was followed over four weeks. At the trols are included for comparison. The lysin end of this period the chloroformed and control units (per ml) are recorded only for fractions VI, (saline) human and dog prolysins were treated which were typical and usually the most active. with staphylokinase and warmed to 370 C., for Streptokinase activated only human prolysin 20 minutes, demonstrating very considerable quan- and the lower unitage was probably due to the tities of persisting prolysin as seen from the uni- deterioration after one hour at 370 C. of the lysin, tage (in parentheses) included in the tests men- which was formed very rapidly with this activator. tioned. Staphylokinase activated human and dog proly- Very high yields of lysin were obtained from sins. Bovine prolysin was not affected by the the human prolysin with streptokinase and from bacterial kinases. These facts were confirmatory the human and dog prolysins with staphylokinase, of earlier work. especially at the cold temperature. The dog lung Mouse, cat, dog, human, and rat lung fractions lysokinase was active on all three prolysins at 370 showed good to fair kinase activity on all three C. but the titers were increased markedly (espe- prolysins; guinea-pig and rabbit lung were less cially with human and dog prolysins) after four active, and bovine lung fractions were inexplicably weeks at 40 C. However, at this time the values inactive. The controls (borate buffer instead of were still short of the staphylokinase maxima. prolysin) showed slight but significant lytic ac- No significance can be attached to the mere traces tivity, in tests with those lung fractions which of lysis detected overnight in the other tests (in- showed most lysokinase activity. cluding the controls). B. The effects of temperature on the activation of the three prolysins (human, dog, bovine) are IV. Do tissue kinases and bacterial kinases act on recorded in Table IV. The "activators" tested the same precursor of fibrinolysin? were 1) dog lung (IV + VI), 2) streptokinase, Dog prolysin, as prepared in the above experi- 3) staphylokinase, 4) chloroform treatment (15). ments, does not activate "spontaneously" even af- The lysin unitage was determined at intervals ter four weeks at 40 C. (cf. Fig. 2). Other experi- over four hours at 370 C., the maximum yield be- ments, to be published later, show that prolysin, prepared by different techniques, e.g., isoelectric TABLE IV precipitation at low ionic strength, higher satura- Lysin units/ml obtained from human, dog, and bovine prolysins after incubation with the various activators for tion with ammonium sulfate, etc., will activate optimal period at 370 C. and for four weeks at 40 C. "spontaneously," reaching maximum titer in about four weeks at 40 C. This maximum titer is not Human Dog Bovine Activator Temp. prolysin prolysin prolysin increased by staphylokinase treatment, and it is 370 C. 16.0 108.0 6.8 similar to the titer obtained by staphylokinase ac- Dog lung (IV+VI) tivation of the fresh prolysin. From these obser- 40 C. 225.0 540.0 28.0 vations we concluded that the same prolysin was 370 C. 280.0 600+ tr. activated by staphylokinase and by the "spontane- Staphylokinase 40 C. 600+ 600+ tr. ous" mechanism. After dog prolysin was maximally activated by 370 C. 600+ tr. tr. Streptokinase staphylokinase, no increase in titer could be ob- 40 C. 600+ tr. tr. tained by treatment with lung kinase. On the 370 C. tr. tr. tr. other hand, after "maximal" activation with lung Chloroform kinase, subsequent 40C. tr. (234)* tr. (600 +)* tr. treatment with staphylokinase usually did produce some increase in lysin titer, 370 C. tr. tr. tr. Saline but never to a value higher than that obtained by 40 C. tr. (220)* tr. (600 +)* tr. staphylokinase treatment alone. It was therefore concluded that a single pre- * Figures in parentheses indicate values obtained after staphylokinase activation. cursor (prolysin) would suffice to explain lysin 1064 JESSICA H. LEWIS AND JOHN H. FERGUSON formation by the three activation mechanisms: 1) ration (concentrated in one quarter original serum "spontaneous" ( ? serum fibrinolysokinase), 2) value) obtained from the same dog serum. bacterial fibrinolysokinase (e.g., staphylokinase), That prolysin in whole plasma is also poorly and 3) tissue fibrinolysokinase (e.g., lung [IV + activated by tissue lysokinases is illustrated, inci- VI]); complete activation being achieved by the dentally, in the tests of Table VI. As stated in first two (under appropriate experimental condi- the footnote, the plasma clots containing dog lung tions) but only partial activation by the present fraction (IV + VI) lysed within two and a half preparations of lung tissue. hours, whereas the control clots were solid after 24 hours. V. Prolysin activation at varying dilutions of fibrinolysokinase: VII. Extraction experiments on lung fibrinoly- Various dilutions of dog lung fraction (IV + sokinase: VI) were incubated at 370 C. with dog prolysin Dog lung fraction (IV + VI) is a complex mix- and the mixtures sampled at intervals for the usual ture in which the lysokinase factor is present in fibrinolysin assay. The results are presented the suspended material and not demonstrable in graphically in Figure 4. Lysin deterioration is the soluble extractive obtainable with saline. clearly evident beyond the peaks of the curves and Subjecting the preparation to lipid solvents, e.g., undoubtedly modifies the kinetics of the activation alcohol, acetone, ether, chloroform, carbon tetra- reaction. Nevertheless, the curves differ greatly chloride, failed to yield any active extracts and and suggest that rate, certainly, and amount, prob- the residue always retained most of its activity. ably, of lysin formation are directly dependent The tissue fraction dried with alcohol and ether upon the concentration of fibrinolysokinase. suffered only partial diminution of lysokinase potency. VI. Tissue lysokinases and lysin formation in whole serum or plasma: VIII. Stability of fibrinolysokinase: The activation of prolysin in whole (dog) serum The lysokinase activity of dog lung fraction (IV by lung fibrinolysokinase, at 370 C., has proved + VI) was destroyed by five minutes heating at disappointing, only traces of fibrinolytic activity 800 C. and only slight activity remained after 30 being detectible. When serum was diluted ( x 5) minutes warming at 560 C. At 370 C. and 29° C. with saline, a fibrinolysin unitage of about 6.5 per potency remained unchanged for at least four ml was obtained. Considering relative dilutions hours. At 00 C. no detectible alteration was noted (1: 20) this is a significant amount of the 230 in 48 hours and only slight deterioration after two units/ml simultaneously found in a prolysin prepa- weeks. Freezing and storage of the more concentrated tissue fractions at - 200 C. showed unchanged unitage data over at least a month. Un- der all the above conditions, dilution appeared to be a factor in lessening the kinase stability. IX. Blood clotting factors in tissue fibrinolysokinase preparations: In connection with the survey of various tissue fractions for lysokinase activity, many tests were run to afford at least a preliminary idea of the presence or absence of a number of factors which might be expected to participate in or modify blood- 1 2 3 4 reactions. INCUBATION PERIOD (HOURS) clotting Fibrinogen: The absence of clotting on the ad- FIG. 4. ACTIVATION OF DoG PROLYSIN BY VARIOUS dition of thrombin showed that no fibrinogen ap- DILUTIONS OF LUNG FIBRINOLYSOKINASE peared in any of the tissue fractions. STUDIES ON A PROTEOLYTIC ENZYME SYSTEM OF THE BLOOD. II 1065

Thrombin: tested for by adding 0.1 ml of tis- TABLE VI sue fraction to 0.1 ml fibrinogen (free from pro- Separation of fibrinolysokinase and thromboplastin thrombin after BaSO4 adsorption). The only Plasma clotting Prolysin activation seemingly significant result was a 26 minute coagu- Reagents time lysis timet lation by seminal vesicle fraction I, although ques- seconds minutes tionable coagulation after 12 hours was seen in a Buffered saline 127.0 No lysis five hours Thromboplastin 8.4 No lysis five hours small number of other tests. Staphylokinase 200.0* i Prothrombin was considered detectible, if pres- Lung IV + VI 13.0* 3 ent (particularly in view of the thromboplastic Lung IV + VI, 18.5 No lysis five hours activity of many of the tissue preparations) by boiled 3 min. simply repeating the foregoing test with the ad- Lung IV + VI dition of 0.1 ml 0.02 M CaCl2. Again only the + 13.4* 4j buffered saline seminal vesicle I clotted the fibrinogen, but the Lung IV + VI 22 minutes was not significantly different from + t 7 the preceding test. In other occasional instances, heparin 0.25 mg/ml some clots were observed in 12-24 hours. * Plasma clots lysed by two and a half hours. That fibrinogenolysis was a factor operating to t At two and a half hours no clot after addition strong thrombin, suggesting fibrinogenolysis. mask possible traces of thrombic (or prothrom- $ Lysis time of 1 per cent fibrin clot containing prolysin bic?) activity was evident from the finding that + activator (preincubated one hr. at 370 C). thrombin would no longer clot the fibrinogen in a number of the mixtures tested after a number of (see below) activities in the same tissue prepara- hours incubation with tissue fractions. tion must be emphasized. Antithromboplastin: Observations that dilution Antithrombin: tested (with appropriate con- of trols) by incubation of the tissue fraction with a lung (IV + VI) decreased rather than in- potent (bovine) thrombin. Table V records the creased plasma clotting times argue against any readily demonstrable "antithromboplastin." results of some tests on dog lung fraction (IV + Thromboplastin: The anticipated finding of con- VI) which is seen markedly to weaken the thrombic activity both immediately (15 seconds) and siderable thromboplastic activity in many of the increasingly with incubation (290 C.). In the tissue fractions was simply demonstrated by the controls, with saline, and with ordinary thrombo- plasma prothrombin time test of Quick (18), plastin (which shows no fibrinolysokinase activity merely substituting the various tissue fractions for usual -see Table VI), the thrombin is only slightly un- the thromboplastin. A single specimen of - stable over the 30 minute incubation period. The oxalated dog plasma, kept frozen at 200 C. significance of antithrombic and thromboplastic between test runs, was used throughout the comparative survey. The simple recalcification time TABLE V of this plasma was 120 seconds and the pro- Antithrombic action of dog lung IV + VI thrombin clotting time, with Schieffelin's "horse tissue" thromboplastin, was 8.4 sec. Tissue frac- Clotting time: 0.25 ml. thrombin mix- tions giving prothrombin times (in parentheses) ture + 0.5 ml. fibrinogen Incubation Period: (290 C) of 15 seconds or less include: brain I (12.5"), II ( 1 1.9"), IV ( 1 5") ; spinal cord I ( 10.9"); 15"1 15' 30' lymph gland I (12.4"), II (14.0"), IV (12.5"), Thrombin V (14.0") ; intestine I (12.2"), II (14.3"), IV 1. + 26.5" 32.8" 38.3" Buffered saline (14.0"), VI (11.0"); stomach VI (13.2") ; testis I (12.0"'), II (15.0"); kidney IV (14.5"); lung I Thrombin 2. + 119.6" 216.0" 250.0" (12.8"), IV (15.0"); uterus VI (15.0"). Tests Lung IV + VI were not run on placenta. Thrombin The chosen method designedly demonstrates 3. + 24.0" 28.6" 31.6" Thromboplastin only the more powerful thromboplastins, sometimes acting despite the presence also of consider- 1066 JESSICA H. LEWIS AND JOHN H. FERGUSON able inhibitors (see antithrombin). For purposes lysokinase and slightly by guinea-pig and rabbit of the survey, a prothrombin time of longer than preparations. The lack of activity of bovine lung 40 seconds was regarded as indicating no significant fractions for all three prolysins is unexplained. amount of thromboplastic activity. This was the Although tissue lysokinase, in the relatively im- finding with: all fractions VII; all fractions of pure form studied, does not activate (dog) pro- thyroid, muscle, liver, bone-marrow, platelets, fibrinolysin to as great an extent as does staphylo- leucocytes, erythrocytes. Pancreas fractions VI kinase, the evidence (Section IV) indicates a com- and VII, although not markedly fibrinolytic, pro- mon prolysin able to yield a similar fibrinolysin longed the clotting times to many minutes in these with the several types of activator. The prolysin tests. in the present experiments is not subject to "spon- Experimental separation of thromboplastic and taneous" activation or to activation by treatment fibrinolysokinase activity of tissue fractions: with chloroform (Table IV). This is distinctly Since the potent thromboplastin preparation used different from other prolysin preparations (Sec- in these studies showed no measurable fibrinoly- tion IV) and raises new questions as to a probable sokinase activity, while the staphylokinase lacked "serum lysokinase" which has apparently been thromboplastic activity, it is probable that the fre- removed in the present method of fractionation. quent occurrence of the two kinds of activators in Further experiments are in progress on this the same tissue fraction is merely coincidental, in question. which case an experimental separation should be Tissue lysokinases, like staphylokinase (but dif- possible. The experiments summarized in Table fering from streptokinase [12]), require a con- VI show that this may be accomplished by tak- siderable time to effect maximal activation of ing advantage of 1) the greater thermolability of prolysin. This time factor is modified by a num- the lysokinase and 2) the sensitivity of the throm- ber of variables, including temperature, strength boplastin to inhibition by heparin. All the above of kinase, etc., and the fibrinolysin yield is altered points are clearly brought out in the data of the also by deterioration and by the action of anti- Table. fibrinolysin, if present. The last may account for the difficulties encountered in using tissue kinase DISCUSSION on whole serum or plasma, especially undiluted, as The evidence of the survey tests reported in this compared with the good lysin production from our study supports the conclusion that a fibrinoly- purified (stable and antilysin-free) prolysins. sokinase (factor which converts serum profibri- Dog lung fraction (IV + VI) is found to be nolysin into active fibrinolysin) is quantitatively relatively heat labile. It is well preserved in the demonstrable in materials obtained by differential frozen state at - 20° C. and reasonably stable at centrifugation from a wide variety of tissues. refrigerator temperatures. At room or body tem- Lung is a potent source and convenient as to peratures it activates prolysin rapidly, while at 40 amount available and ease of exsanguinating by C. the activation, though slow, continues for sev- perfusion. The data of Table I show fibrinoly- eral weeks. The tissue fraction is complex and has sokinase to be ubiquitous, in very varying amounts so far resisted efforts to purify the lysokinase by in a great many tissues and fractions. Only frac- extraction with saline and phosphate buffers (pH: tions VII usually give a negative result. The 4.85-9.5), etc. most significant concentration appears to be in Antifibrinolysin ( ? from residual blood) oc- the small particle fractions IV and VI, corre- curs in many of the tissue fractions, although sponding to Claude's "large granules" and "mi- fibrinolysokinase may also be detected, e.g., frac- crosomes," and confirming, to some extent, Tag- tions VI. No evidence for any antifibrinolyso- non's findings on rat lung. kinase could be obtained by dilution of dog lung Lack of real species specificity is indicated in fraction (IV + VI). the data on lung fractions VI in Table III, in The crude tissue fractions contain factors which which tests were made with prolysins of human, participate in or modify blood-clotting reactions dog, and bovine origin, all of which were moder- in experimental systems. Fibrinogenolysis and ately activated by mouse, cat, dog, and human lung fibrinolysis are encountered in mixtures when ac- STUDIES ON A PROTEOLYTIC ENZYME SYSTEM OF THE BLOOD. II 1067 tive fibrinolysin appears and the extent to which Lung fractions from eight animal species this protease may interfere with the various pro- showed quantitative differences but no real species tein factors of the clotting system raises a number specificity for prolysins of three (human, dog, of questions which we shall not go into here. An- bovine). The successful activation ( ? partial) of tithrombic activity may interfere, to a minor ex- bovine profibrinolysin is noteworthy, in view of tent, by causing difficulties in obtaining a "stand- previous inability to find a bacterial activator. ard" clot for fibrinolysin assay. In lung fraction The crude tissue preparations often contain fac- (IV + VI) a potent antithrombin is demonstrable tors which act on blood-clotting mechanisms, es- although the tissue preparation is also strongly pecially thromboplastin and antithrombin. Throm- thromboplastic. The thromboplastic activity boplastic and fibrinolysokinase activities are ex- found in the tissue survey was almost completely perimentally separable. restricted to the particulate fraction, although not confined to any one fraction (cf. "thromboplastic BIBLIOGRAPHY protein" of Chargaff [ 14] ). This thromboplastic 1. Fleisher, M. S., and Loeb, L., On tissue fibrinolysins. activity is almost as ubiquitous and variable (quan- J. Biol. Chem., 1915, 21, 477. titatively) as the fibrinolysokinase activity but no 2. Christensen, L. R., Streptococcal fibrinolysis: a significant correlation in distribution could be proteolytic reaction due to a serum enzyme activated by streptococcal fibrinolysin. J. Gen. found between the two. The experiments of Table Physiol., 1945, 28, 363. VI confirm the expectation that thromboplastic 3. Christensen, L. R., and MacLeod, C. M., A proteolytic and fibrinolysokinase activities are probably inde- enzyme of serum: characterization, activation, and pendent, despite frequent co-existence of these reaction with inhibitors. J. Gen. Physiol., 1945, two factors in crude tissue fractions. 28, 559. 4. Milstone, H., A factor in normal blood which par- CONCLUSIONS ticipates in streptococcal fibrinolysis. J. Immunol., 1941, 42, 109. A fibrinolytic technic, previously employed for 5. Astrup, T., and Permin, P. M., Fibrinolysis in the assay of fibrinolysin and antifibrinolysin, is slightly animal organism. Nature, 1947, 159, 681. modified to increase its range of sensitivity and to 6. Permin, P. M., Properties of fibrinokinase-fibrino- permit adaptation to a quantitative study of the lysin system. Nature, 1947, 160, 571. fibrinolysokinase activity of certain tissue extracts. 7. Tagnon, H. J., and Petermann, M. L., Activation of proplasmin by a tissue fraction. Proc. Soc. Exper. The present data offer a survey of a large number Biol. & Med., 1949, 70, 359. of animal tissues, fractionated by differential cen- 8. Claude, A., Fractionation of mammalian liver cells by trifugation, and indicate a widespread distribution differential centrifugation; problems, methods and of this factor, with lung, uterus, and a number preparation of extract. J. Exper. Med., 1946, 84, of other tissues giving good yields of kinase ac- 51. Fractionation of mammalian liver cells by dif- tivity. The "large granule" and "microsome" ferential centrifugation; experimental procedures (Claude) fractions particularly show this activity, and results. Ibid., 61. 9. Lewis, J. H., Ferguson, J. H., and Jackson, B. G., and it is not extractible with ordinary solvents. Bacterial activators (lysokinases) of the fibrino- A lung fraction (IV + VI) was selected for some lytic enzyme system of serum. Proc. Soc. Exper. detailed studies. Biol. & Med., 1949, 72, 703. The tissue lysokinase is relatively heat labile and 10. Ferguson, J. H., Travis, B. L., and Gerheim, E. B., best preserved frozen at - 200 C. It requires time Fibrinogenolytic demonstration of activation and inhibition of tryptase in plasma protein fraction-I for maximal effect on profibrinolysin and the rate ("antihemophilic globulin"). Proc. Soc. Exper. of activation and final fibrinolysin yield are modi- Biol. & Med., 1947, 64, 285. fied by temperature, strength of kinase, deteriora- 11. Gerheim, E. B., Ferguson, J. H., Travis, B. L., tion of formed lysin, and by antilysin, if present. Johnston, C. L., and Boyles, P. W., Staphylococcal Serum and plasma are activated by tissue lysoki- fibrinolysis. Proc. Soc. Exper. Biol. & Med., 1948, nase with difficulty but dilution lessens the "anti- 68, 246. 12. Gerheim, E. B., and Ferguson, J. H., Species reactiv- lytic" effect. Antilysin is also demonstrable in ity to staphylokinase. Proc. Soc. Exper. Biol. & some tissue fractions. Med., 1949, 71, 261. 1068 JESSICA H. LEWIS AND JOHN H. FERGUSON 13. Tagnon, H. J., and Petermann, M. L., A proplasmin plasma: increase in antifibrinolysin activity during activator from mammalian tissue. J. Clin. Invest., pteroylglutamic acid deficiency. Am. J. Physiol., 1949, 28, 814. 1947, 150, 661. 14. Chargaff, E., Cell structure and the problem of blood 17. Ware, A. G., Guest, M. M., and Seegers, W. H., coagulation. J. Biol. Chem., 1945, 160, 351. Fibrinogen: with special reference to its prepara- 15. Lewis, J. H., and Ferguson, J. H., Studies on a tion and certain properties of the product. Arch. proteolytic enzyme system of the blood. I. Inhi- Biochem., 1948, 13, 231. bition of fibrinolysin. J. Clin. Invest., 1950, 29, 486. 18. Quick, A. J., The Hemorrhagic Diseases and the 16. Guest, M. M., Ware, A. G., and Seegers, W. H., A Physiology of Hemostasis. Charles C. Thomas, quantitative study of antifibrinolysin in chick Springfield, Ill., 1942.