Purification and Properties of a Plasminogen Activator from Pig Heart*

An improved procedure is described for the purification of plasminogen activator from pig heart. The initial purification steps were similar to these described previously (Bachmann, F., Fletcher, A. P., Alkjaersig, N., and Sherry, S. (1964) Biochemistry 3, 1578--1585). Use of a novel extraction medium containing EDTA, cysteine, and 2,3-dimercaptopropanol-1 facilitated the removal of large amounts of inert proteins prior to gel filtration on Bio-Gel P-150. The final product had a specific activity of 120,000 to 160,000 CTA units/mg of protein (CTA, Committee on Thrombolytic Agents of the National Heart Institute). Total purification over pig heart was 25,000 to 30,000-fold, average recovery compared to the initial extract was 6 to 8%. Polyacrylamide gel electrophoresis revealed a major and two minor components. The molecular weight of the activator determined by gel filtration was 51,500 +/- 3,400 for the major activity component and 48,000 for a minor component which was partially separated from the major peak in eight of nine chromatography runs. A gamma-globulin fraction of antiserum against purified activator neutralized the biological activity of the activator on fibrin plates. Immunoelectrophoresis of gel-filtered activator revealed only one anodic component.

An improved procedure is described for the purification of plasminogen activator from pig heart. The initial purification steps were similar to those described previously (Bachmann, F., Fletcher, A. I'., Alkjaersig, h'., and Sherry, S. (1964) Hiochemisby 3, 1578-1585). 1Jse of a novel extraction medium containing EDTA, cysteine, and 2,3-dimercaptopropanol-1 facilitated the removal of large amounts of inert proteins prior to gel filtration on Rio-Gel P-150. The final product had a specific activity of 120,000 to 160,000 CTA unitslmg of protein (CTA, Committee on Thrombolytic Agents of the National Heart Institute).
Total purification over pig heart was 25,000 to 39,090-fold, average recovery compared to the initial extract was 6 to 8%. I'olyacrylamide gel electrophoresis revealed a major and two minor components. The molecular weight of the activator determined by gel filtration was 51,500 + 3,190 for the major activity component and 48,900 for a minor component which was partially separated from the major peak in eight of nine chromatography runs. A -y-globulin fraction of antiserum against purified activator neutralized the biological activity of the activator on fibrin plates. Immunoelectrophoresis of gel-filtered activator revealed only one anodic component. were found to extract tissue activator from pig heart. Although all these agents solubilized not only tissue activator but other cellular components as well, they have served well to produce initial extracts for studies on the purification of tissue activator of plasminogen (7)(8)(9). Kok  Activator antiserum was produced in rabbits by injecting intramuscularily 0.5 mg of chromatographed tissue activator (> 100,000 CTA unitsimg) in complete Freund's adjuvant. A booster injection was given 1 month and 5 months after the initial injection, and 5 days later blood was obtained by heart puncture. Serum was harvested after spontaneous clotting and incubation for 18 h at 37", and crude y-globulin was prepared by 33% ammonium sulfate precipitation. A -y-globulin fraction was similarly prepared from serum obtained from a nonimmunized rabbit.
After overnight dialysis of the y-globulin preparations against cold 0.9% NaCl solution, the preparations were adjusted with saline solution to the same protein concentration, 16 (13). The stained gels were photographed as described by Oliver and Chalkley (14). To detect tissue activator activity on unstained gels, the gels were sliced into 2-mm segments with a gel slicer. Each slice was dipped into 0.1 M sodium borate buffer, pH 8.0, for 5 s and placed on a fibrin plate which was then incubated at 37". The lysis zone was calculated as the difference between the product of the diameters of the total lysis zone and of the gel slice.

Enzyme Purification
Step 1: Acetone Drying-Four to five kilograms of partially thawed pig heart were ground in an electric meat grinder. Onekilogram portions were suspended in 4 liters of -20" acetone in a 6liter Waring Blendor and further homogenized.
Pieces of dry ice were added to maintain low temperature. Filtrations were performed in Buchner funnels using Whatman No. 541 paper and suction. The filter cake was then added back to the Waring Blendor and the homogenizing process was repeated four times. The final filter cake was fragmented and spread on large sheets of filter paper to dry at room temperature.
The fine tan-colored powder weighed 600 to 700 g, or about 15% of the weight of wet tissue. Chromatography eluates were continuously monitored by an LKB Uvicord II detector and recorder unit at 280 nm, and absorbance of each fraction was also measured at 280 nm in a Beckman DU spectrophotometer.
Step 2: Extraction by Acetate Buffer-For each 100 g of acetonedried pig heart powder, 800 ml of 0.3 M potassium acetate, pH 4.2, was added and the suspension was stirred continuously for 6 h at 5". The supernatant was recovered by centrifugation i n the cold at 5,000 x g for 30 min, and the sediment re-extracted with 400 ml of 0.3 M potassium acetate, pH 4.2, for at least 3 h, again recovering the supernatant by centrifugation.
Mixtures of 0.05 M TrisiHCl buffer, pH 7.4, 0.01 M EDTA, freshly prepared 0.05 M cysteine hydrochloride, and 0.03 M 2,3-dimercaptopropanol-l in the ratio of 70:10:10:0.1 volumes, respectively, were prepared immediately before use and the pH adjusted to 7.4. Potassium acetate buffers were prepared from glacial acetic acid and the pH adjusted with KOH. Sodium chloride was added prior to diluting to volume.
The P-naphthylamides of L-lysine, L-arginine, and L-histidine were products of Schwarz'Mann, Orangeburg, N. Y. Amidase activity of tissue activator preparations was determined by the method of Blackwood and Mandl (16). Bio-Gel P-150 (100 to 200 mesh) was a product of Bio-Rad Laboratories, Richmond, Calif., and Sephadex G-200 was obtained from Pharmacia Fine Chemicals, Piscataway, N. J. Molecular weight marker proteins were products of SchwarziMann and Pharmacia Fine Chemicals.
Step 3: First Ammoniwn Sulfate Precipitation- The combined supernatant solutions were continuously stirred while 300 g of ammonium sulfate/liter of extract were added over a 4-h period to give 50% saturation at 2". After overnight settling, the settled precipitate was centrifuged and worked into a smooth paste and dispersed in 300 ml of cold distilled water/l00 g of original dried product. The pH was adjusted to pH 4.2 with 1 M acetic acid and the suspension stirred for 2 h and centrifuged.
The sediment was re-extracted overnight with 150 ml of 0. A sample volume of 2 ml was applied and the column developed at a flow rate of 11.5 ml/h under a hydrostatic head of 16 cm of HLO. Fractions of 1.5 to 2.0 ml were collected and subjected to Lowry protein assay for marker proteins or for activator activity on bovine fibrin plates. Molecular weight also was estimated from elution volumes of tissue Step 4: Second Ammonium Sulfate Precipitation-The eluates were adjusted to pH 8.2 with solid Tris and finely powdered (NH&SO, was added slowly with stirring (200 g/liter, 35% saturation at 2"). One and one-half hours later, the solution was centrifuged and the precipitate, dissolved in 75 ml of 0.05 M acetic acid, was dialyzed overnight against two changes of 2.5 liters of cold distilled water to a resistance greater than 1,000 ohms.
Step 5: Zn'+ Precipitation at Low Ionic Strength-The dialyzed preparation was adjusted to pH 4.2 with acetic acid and the sample was cleared by centrifugation.
The pH of the supernatant was adjusted to 6.0 with 1 N NaOH and a solution of 10 rnM zinc acetate was added slowly to give a final Zn" concentration of 0.3 mM. The pH was adjusted to 6.5 with 0.1 M sodium barbital and stirred for 15 min. After settling for 1 h, the precipitate was recovered by centrifugation, dispersed into 20 ml of 0.5 M acetic acid, and the pH of the solution adjusted to 4.2, the solution was stirred until the solid dissolved and then lyophilized.
Six hundred to seven hundred grams of acetone-dried pig heart yielded 2 to 2.5 g of lyophilized Step 5 product. 2 We are indebted to Abbott Laboratories, North Chicago, Ill., for Step 6 propanol-1, pH 7.4, and homogenized with a tissue homogenizer. The pH was adjusted to 7.4 with 0.1 M NaOH and the suspension was stirred for 20 rain at room temperature.
It was then centrifuged at 9,000 x g for 30 min at 5" and the supernatant was decanted. The residue was then taken up in 180 ml of 7O:ZO Tris:water, homogenized, and the pH was adjusted to 7.4. After stirring for 20 min at room temperature, the suspension was again centrifuged at 9,000 x g for 30 ruin and the supernatant was decanted.
The residue was taken up to 20 ml of 0.075 M potassium acetate, 0.3 M NaCl, pH 4.2, the pH of the suspension was adjusted to 4.2 and it was stirred for 30 ruin at room temperature and centrifuged at 5" at 13,500 x g for 30 min. The supernatant, which contained the bulk of the tissue activator activity, was chromatographed on Bio-Gel P-150.
Step 7: Gel Filtration on Bio-Gel P-150-A glass column (2.5 x 100 cm) was packed to a height of 90 cm with Bio-Gel P-150 and equilibrated with 0.075 M potassium acetate, 0.3 M NaCl, pH 4.2, at 5". The tissue activator preparation (Fraction 6) was layered over the top of the gel and allowed to enter the gel. fractionation procedure resulted in a significant reduction of the first major protein peak seen in Fig. 1, and of the proteins associated with the activator activity peak (Fig. 2) and separation of activator from inert proteins by gel filtration has been noted previously (7,9). Assay Variations -Because of the discrepancy between bo- Step 2 material, only 38% was recovered when the activity was measured by the bovine fibrin plate assay. In the subsequent four purification steps, the recovery data as measured by both assays were comparable, resulting in a yield of 28 to 29% from Step 2 to Step 5. Activity ratios of activator in purification steps beyond Step 5 were dependent on the medium in which the preparation was suspended.
As seen in Tables I and II, human clot lysisibovine fibrin plates and human fibrin plate/bovine fibrin plate ratios were near 3 and 2, respectively, when Step 5  Tissue activator activity consistently appeared at the same elution volumes and was not associated with a well defined 280 nm absorbance. As seen in Fig. 4, there appear to be two incompletely separated activity peaks. Initially, this was attributed to assay errors, but it is significant that a double activity peak was seen in eight of nine chromatograms, although the relative activities of the two peaks varied from experiment to experiment.  phoretic gels run in the acid system of Neville (12), the alkaline system of Davis (18) did not reveal distinct bands. Diffuse staining of the top one-half of the gel was observed. When gel slices were placed on bovine and human fibrin plates, tissue activator was distributed over the same part of the gel.
Estimation of the molecular weight of tissue activator by gel filtration is shown in Fig. 7. The chromatogram was obtained by gel filtration of Fraction 6 on a preparative Bio-Gel P-150 column, calibrated with marker proteins. The estimated molecular weight of the major activator peak was 52,500, that of the minor peak was 48,000. Gel filtration of an acid/acetate extract of Step 5 activator on columns (1.5 x 56 cm) of Bio-Gel P-150 gave an average molecular weight of 51,500 -+ 3,400 for the major component; because of the smaller size of the column, separation was not sufficient to accurately estimate the molecular weight of the smaller species. Immunological Studies-Quenching of the bovine fibrin plate activity of tissue activator by the y-globulin fraction of antiserum against activator is illustrated in Fig. 8. Incorporation of y-globulin in the plate at a final dilution of 1:lOOO completely inhibited lysis at all activator concentrations up to 14.5 CTA units/ml. Quenching of activity was observed at all y-globulin dilutions up to 1:25,000. However, y-globulin from serum of nonimmunized rabbit at the same dilutions had no effect on activator activity. Immunoelectrophoretic patterns of Fraction 6 and chromatographed tissue activator are shown in Fig. 9. Three precipitin arcs, representing one cathodic component and two anodic 9. Immunoelectrophoresis of activator obtained by Bio-Gel P-150 chromatography (top) and of Fraction 6 (bottom). Each preparation, containing approximately 10,000 CTA units/ml, was applied as a l-p1 spot (arrow) to thin gel agarose film, and electrophoresis was conducted at 90 V for 45 min. The trough contained 40 ~1 of yglobulin fraction of antiserum to activator. Diffusion was carried out for 24 h, the film washed in 0.9% sodium chloride solution for 3 h and in water for 1 h before staining with 0.125% Coomassie brilliant blue R-250 (18). Destaining was carried out with a 87.5:7.5:5 mixture of water:glacial acetic acid:methanol. The anode is to the right.
components, are commonly seen in Fraction 6 patterns. Only one anodic arc was seen in the most purified activator preparation. Similar immunoelectrophoretic studies in fibrimagarose media established that the anodic band is the active component.

DISCUSSION
Tissue activator of plasminogen is an activity which has been demonstrated in a variety of animal and human tissues (19). Conversion of the zymogen, plasminogen, to the proteolytic form, plasmin, by tissue activator is accomplished by means as yet not elucidated, although it is probably achieved, as with urokinase, through a proteolytic process. The main function of plasmin, a proteolytic enzyme with a high affinity for polymerized fibrin, is the removal of intravascular thrombi. The mechanism for release of tissue activator from tissues has not been well established. Part of the problem has been the difficulty in obtaining sufficient amounts of tissue activator in highly purified form to study its physicochemical properties and to produce specific antisera for localization studies.
The purification procedure described in this report results in the isolation of highly purified activator, representing a 25,000-to 30,000-fold increase in specific activity from that of acetone-dried pig heart. Initial steps of purification, extraction in acid/acetate buffer, ammonium sulfate precipitations at