THE JOURNAL OF BIOLOGICAL CHEMISTRY Val. 258, No. 1, Issue of January 10, pp. 622-628, 1983 Printed in U.S.A.

Purification and Characterization of the Biliary Plasminogen Activator Bilokinase*

(Received for publication, August 6, 1982)

Susumu Oshiba and Toyohiko Ariga From the Department of Physiology, Nihon University School of Medicine, Ztabashi-Ku, Tokyo 173, Japan

The aim of the present study is to elucidate the en- zymological and chemical properties of the plasmino- gen activator in bile, bilokinase. A bilokinase prepa- ration was obtained from 94.2 liters of bovine gall bladder bile through seven purification steps, two of which employed a precipitation method in which am- monium sulfate and acetone were used sequentially. Twenty mg of bilokinase preparation with a specific activity of 5,900 IU/mg were obtained, for a 9% yield with 6,556-fold purification. This preparation revealed a single band upon disc electrophoresis. The bilokinase was a 3.32 S protein and its molecular weight was found to be 57,000. Isoelectric focusing showed that the bilo- kinase was separable into 5 fractions having different isoelectric points ranging from pH 7.4 to 9.0. The prop- erties of the individual fractions have not yet been determined. The enzymatic activity of bilokinase was recognized to be heat-labile and to be stable at pH 4.0. The activation of plasminogen by bilokinase took place most effectively at pH 7.8 in a manner similar to that of urokinase. In its hydrolysis of both W-acetylglycy1-L- lysine-methyl ester-acetate and H-D-Glu-Gly-Arg-pNA (S-2227), bilokinase showed similar K,,, values to those of urokinase; however, they were quite distinct from those of plasmin. It was concluded therefore that bilo- kinase is a plasminogen activator with enzymatic prop- erties which are quite similar to those of the urinary plasminogen activator urokinase. The origin of biloki- nase and its relation to liver function are now under investigation.

Bilokinase, a plasminogen activator occurring in human and mammalian bile, was initially identified and named by Oshiba et al. (1). The enzymatic nature of bilokinase was investigated using a partially purified preparation. However, the enzymo- logical and chemical characteristics of this enzyme in a highly purified state were not examined.

In 1973, Ariga (2) developed a specific precipitation method for obtaining an active fraction from bile and separating it from the bile acids, bile pigments, and inhibitory materials. This permitted a 10-fold elevation of the plasminogen activa- tor activity and so provided suitable material for the further purification of bilokinase. In the present study, bilokinase was purified in seven steps including the so-called Ariga method (A-A method for short, since ammonium sulfate and acetone are used in it). The characteristics of the bilokinase were assessed using this bilokinase preparation. The properties

* The costs of publication of this article were defrayed in part by

marked "advertisement" in accordance with 18 U.S.C. Section 1734 the payment of page charges. This article must therefore be hereby

solely to indicate this fact.

were compared with those of urokinase and similarities and differences are discussed hereafter.

MATERIALS AND METHODS

Collection of Bile-Bovine gall bladders were obtained from a slaughter house, and bacteria-free bile was selected after bacteriolog- ical tests which were performed using a nutrient agar broth containing 5 g of meat extract, 10 g of peptone, 10 g of lactose, 80 mg of bromthymol blue, and 15 g of agar in 1 liter of water, pH 7.4. About 0.5 ml of bile and 10 ml of the broth were placed in a sterilized dish (9-cm diameter). This was incubated for 18 h at 37 "C.

Further Examination of the Bile for Fibrinolytic Activity-The bilokinase activity was checked by the method developed by us in 1973 (2), and bile with low activity was rejected. The bilokinase fraction was obtained from the bile as a precipitate, separating it from the bile pigments, bile acids, and other inhibitory materials. The precipitation was accomplished by the sequential addition of ammo- nium sulfate and acetone as follows. 0.5 ml of bile obtained with a syringe from the ligated bladder was transferred to a centrifuge tube (15 ml), and 4.5 ml of 0.12 M ammonium sulfate was added. The temperature of this combined solution was maintained at 1-3 " C , and an equal volume (5 ml) of acetone a t about 2 "C was then added and stirred well. Immediately after, the combined solution was centrifuged at 3000 X g for 5 min at 4 "C. The precipitate was obtained at the midpoint of the two solution phases. In the lower phase, the ammo-

was hydrated. The precipitate was then dissolved in 0.5 ml of 50 mM nium sulfate solution was condensed; and, in the upper, the acetone

Tris buffer, pH 8.0 (containing 0.15 M NaCl), and the insoluble materials were removed by centrifugation. The supernatant was used as the bilokinase extract. In the present experiments, the above procedure was performed with each bile sample in order to examine its bilokinase activity. The activity was measured by spotting 30 p1 of the bilokinase extract onto a fibrin plate containing calcium as de- scribed below. Bile which showed an activity of less than 100 mm2 in the lysis zone was discarded.

Employing the above bacterial and activity tests, 94.2 liters of bile were finally pooled. This amount was 60% of the total bile collected.

Determination of Fibrinolytic Activity-The fibrinolytic activity of bilokinase was determined using a slight modification of the fibrin plate method of Astrup and Mullerz (3). Thirty-four mg of bovine fibrinogen (65% clottable preparation, Miles Laboratories Inc., Elk- hart, IN) were dissolved in 10 ml of borate buffer, pH 7.8, and filtered. Eight ml of this filtrate was transferred into a Petri dish. One-half ml of 0.18 M calcium chloride and 0.5 ml of 20 unit/ml of thrombin solution (Mochida Pharmaceutical Co. Ltd., Tokyo) were then gently added and mixed well by moving the dish back and forth a few times. The mixture was allowed to stand at room temperature for about 30 min before use so that fibrin plate formation could be completed. Thirty p1 of the enzyme solution to be tested were applied to the plate and incubated at 37 "C for 18 h. The amount of activity was assessed from the perpendicular square of the lysis zone.

Electrophoresis-The method developed by Reisfeld et al. (4) was applied for bilokinase. Polyacrylamide gels, 7.5%, pH 4.3, prepared in glass tubes with an inner diameter of 5 mm and a length of 60 mm were used. Electrophoresis was performed with 8 mA/tube for 50 min a t 4 "C. After the run, the gels were fixed and stained overnight with a 10% acetic acid solution containing 0.02% Coomassie brilliant blue (R-250).

Isoelectric Focusing-This was carried out using the method of Versterberg and Svensson (5) using an ampholyte (LKB) of pH 2-10. Focusing was performed in a sucrose concentration gradient column

622

Biliary Plasminogen Activator

TABLE I Purification of bilokinase from bovine bile

94.2 liters of bovine gall bladder bile selected for its bacteria-free and fibrinolytic activity properties were subjected to the 7-step purification method for obtaining the fibrinolytic enzyme bilokinase. Note that the total activity shown by the original bile markedly increases in the first step.

Steo Total activitv" Total Droteinh Suecific activitv Purification factor Recover'

623

Bovine bile 1. A - A ~ (1) 2. A - A ~ (2) 3. SP-Sephadex batch 4. CM-Sephadex column 5. Sephacryl column 6. DEAE-Sephadex column 7. SP-Sephadex column

x104 IU 33.1

130.4 110.5 51.0 21.4 18.9 16.0 11.8

mg 367,400 85,100 59,700 2,700

758 339 150 20

I U / w 0.9

15.3 18.5

188.9 282.3 557.5

1,066.7 5,900.0

1 17 21

210 314 619

1,185 6,556

90

100 84.7 39.1 16.4 14.5 12.3 9.0

Determined by the fibrin plate technique and listed as equivalent to urokinase units (standard human preparation obtained from Mochida

Determined by the method of Lowry et al. (10). Pharmaceutical Co. Ltd., Tokyo).

' Recovery of the bilokinase activity was calculated considering the activity obtained in the first step to be 10096, since this elevated the activity of the original bile by 390%.

The method of extracting bilokinase from bile by sequential addition of ammonium sulfat,e and acetone developed by Ariga (2).

(120 ml) at 900 V for 48 h at 2 "C. The contents were then collected in 1-ml portions, and pH and optical density at 280 nm were measured with a pH meter (HM-18B with an electrode GST-155C, TOA Elec- tronic Ltd., Tokyo) and a spectrophotometer (UV-BlOA, Shimadzu, Tokyo), respectively. The fibrinolytic activity was assayed on a fibrin plate after bringing to neutural pH with solid Tris.

Ultracentrifugal Analysis-The sedimentation velocity method and the sedimentation equilibrium method were used to determine the S value and molecular weight of bilokinase. Lyophilized bilokinase was reconstituted by adding 50 m sodium acetate buffer, pH 4.5, containing 0.35 M NaC1, dialyzed against the same buffer for 5 h, and then made into concentrations of 0.4,0.3,0.2, and 0.1% by adding the buffer. These 0.45-ml portions were centrifuged in a double-sector cell at 55,000 rpm for 67 min at 20 "C using a UCA-1 ultracentrifuge (Hitachi Co., Tokyo). The sedimentation coefficient was estimated by assuming a partial specific volume of 0.75 (6). The sedimentation equilibrium method was performed by the Yphantis method (7), using the same centrifuge and a multichannel cell in which the bilokinase was centrifuged at 18,500 rpm for 150 min. The molecular weight of the bilokinase solutions of different concentrations as described in the sedimentation velocity method was determined from the values of the hinge points formed in the equilibrium states and from the initial concentration of the bilokinase protein which had been subjected to double-sector cell analysis, as described above.

Esterolytic Activity-Hydrolysis of the N"-acetylglycyl-L-lysine methyl ester-acetate substrate of Walton (8) was carried out according to the titration method (8). Ne-Acetylglycyl-L-lysine methyl ester- acetate (Cyclo Chemical Corp.) was dissolved in 0.15 M KC1 solution to make a stock solution of 36 m. The N"-acetylglycyl-L-lysine methyl ester-acetate solution diluted serially with 0.15 M KCl, and 0.5 ml of it was put into a titration vessel together with 0.9 ml of gelatin diluent, pH 7.8, which was made of 0.5% gelatin and 0.15 M KC1 in 5 X M Tris, and prewarmed at 35 "C for 3 rnin. Bilokinase or urokinase was then added to the vessel in a volume of 0.1 ml of 150- IU concentration. Titration was performed with 50 nm KOH solution at 35 "C for 15 min by using a pH s ta t pH meter E 512, electrode EA- 125, Dosimat syringe burette assembly and Impulsomat E 473 micro- titration assembly (Metrohm Herisau, Switzerland). The gelatin dil- uent contained 33 m~ trans-(aminomethy1)cyclohexanecarboxylic acid (Daiichi Pharmaceutical Co., Tokyo).

Amidolytic Activity-Synthetic peptide substrate S-2227 (H-D- Glu-Gly-Arg-pNa) developed by AB KABI (Stockholm, Sweden) for detecting urokinase activity (9) was used in this study to compare bilokinase, urokinase, and plasmin. 0.5 ml of one of these enzymes in 50 mM Tris-HC1 buffer at pH 9.0 for urokinase and plasmin and at pH 8.5 for bilokinase was put into a cuvette (2 x 10 mm) and prewarmed at 37 "C for 3 min. 120 p1 of freshly prepared 2 mM substrate was dissolved in distilled water and then added into the cuvette and mixed well. The increase of absorbance at 405 nm was recorded for 10 min at 37 "C by using a Hitachi 124-spectrophotom- eter equipped with a circulation system to maintain a constant temperature and a QPD 54 Recorder (Hitachi, Tokyo). A Lineweaver- Burk plot was drawn at several concentrations of substrate less than

0.39 mM. Amounts of the enzymes used in this study are shown in the legend of Table I.

RESULTS

Procedure for Purification of Bilokinase Steps 1 and 2-All procedures were carried out at 4 "C

unless otherwise stated. Each 850-ml portion of the bile was transferred to a 5-liter glass beaker and diluted with 1,650 ml of 0.16 M ammonium sulfate solution. To this well mixed solution, an equal volume of acetone was added and the mixture was stirred. I t was then allowed to stand for about 5 min until the precipitates floated to the top. The solution was siphoned off and discarded. The precipitates remaining were then centrifuged at 4,000 X g for 5 min and dissolved in Tris buffer. The insoluble materials were removed by centrifuga- tion a t 10,000 X g for 30 min and the supernatant retained. This extract was designated Step 1-bilokinase. At this stage, the amount of protein was reduced to one-fourth of that in the bile, and the total activity was increased about 4 times. The elevation in activity was recognized to be due to removal of inhibitory factors in the bile. The A-A method was em- ployed again in the next step. Solid ammonium sulfate was dissolved in the Step 1-bilokinase extract to 0.1 M. An equal volume of acetone was added, and precipitates were obtained. These were dissolved in 50 m~ sodium acetate buffer, pH 4.3, containing 0.15 M NaC1, and the resultant solution, designated Step 2-bilokinase, showed a total activity of 1.1 X 10" IU.'

Step 3: Batch Method Using Sulfopropyl-Sephadex-To 5.1 liters of Step 8-bilokinase, 1 liter of SP-Sephadex C-25 equilibrated with the above-mentioned acetate buffer, pH 4.3, was added. After stirring overnight, the mixture was allowed to stand for 3 h. An additional 1 liter of SP-Sephadex gel was added to the decanted supernatant, and this mixture was stirred for 1 h. The resultant gels and the gels obtained in the fiist adsorption were pooled, packed into a column (9 X 30 cm), and washed with about 40 liters of 50 mM sodium acetate buffer, pH 4.3, containing 0.1 M NaCl until absorption of the effluent at 280 nm fell to less than 0.1. Bilokinase was eluted with 4.2 liters of 1% ammonium hydroxide containing 0.2 M NaCl. The pooled effluents were brought to neutral pH with diluted sulfuric acid, and the resultant sample was designated

' For quantitation of urokinase activity, three types of units have been employed IU, CTA (Committee of Thrombolytic Agents) unit, and Ploug units. The IU is currently preferred. Equivalent activity, 1.00 IU = 0.96 CTA units = 0.67 Ploug units.

624 Biliary Plasminogen Activator

Step 3-bilokinase. After this step, the total fibrinolytic activity of the sample fell to about half that of step 2-bilokinase; 500. -0.5 however, the amount of protein present fell to %oth of that in - Step 2-bilokinase, so that the specific activity of Step % 400- I P-1 + P - 2 4 -0.4 f l kinase was 10 times higher than that of Step 2-bilokinase, as - B F1

indicated in Table I. Step 4: Column Chromatography with Carboxymethyl- .?

Sephadex-Step 3-bilokinase was dialyzed against 50 liters of 200- 50 mM sodium acetate buffer, pH 5.8, containing 50 mM NaCl 4 for 12 h, and each 1.4-liter portion of the total 4.2 liters of 100- dialyzed solution was applied to a CM-Sephadex column equilibrated with 50 mM sodium acetate buffer, pH 4.3, con- taining 50 m~ NaC1. The column was washed with 20 liters of 50 100 150 200 the same buffer. Elution of the adsorbed bilokinase was per- formed with an NaCl gradient. The elution profiie is shown in ~ i ~ . 1. The active fraction was eluted at a concentration of FIG. 2. Gel filtration of Step 4-bilokinase obtained from bo-

above 0.4 M NaCL separating it from the largest inert protein from Step 3-bdobase as shown in Fig. and contained 2.1 x 105 Iu vine bile. Step 4-bilokinase was obtained at an intermediate punty

Peak, which was eluted at a lower concentration of NaCl with 758 mg of protein in a volume of 100 ml. A portion of 12.5 ml was between 0.2 and 0.4 M. The active fraction was collected as applied to a Sephacryl S-200 column (2.6 x 90 cm), and the column indicated in Fig. 1 and concentrated using polyvinylpyrroli- was eluted with Tris-saline buffer (50 n m , 0.15 M NaCI, pH 7.5). done K-90 (Nakarai Chemical CO. Ltd., Kyoto). For this Eight-ml fractions were collected. There were two major fibrinolyti- purpose, the effluents were transferred to a cellulose tube and CallY active Peaks (H); however, O d Y fractions at the second

peak (p-2) were collected for further purification. The details are

0

4 300- m

-0.3 .d

Tube number

covered with polyvinylpyrrolidone powder overnight a t 4 "C. under -, nm (o,D. zsonm). The concentrates were neutralized with solid Tris, dialyzed against 50 m~ Tris buffer, pH 7.5, containing 0.15 M NaC1, overnight, and then centrifuged to remove insoluble materials 50,000-60,000. The p-1 activity amounted to only 10% of the at 5000 X g for 10 min. The resultant supernatant was desig- total, so that this material was not used for further purifica- nated Step 4-bilokinase. The total activity of this material tion. The pooled p-2 fraction had a total activity of 1.89 X lo5 was 2.1 X lo5 IU, and the protein content was 758 mg. IU and was designated Step 5-bilokinase. Three liters of step

Step 5: Gel Filtration-Step 4-bilokinase was applied to a 5-bilokinase were next applied to a DEAE-Sephadex column Sephacryl' S-200 column. The chromatogram obtained is to remove the acid proteins. shown in Fig. 2. There were two major active peaks. The early Step 6: DEAE-Sephadex Column Chromatography-The one, p-1, was eluted together with a large amount of protein whole of Step 5-bilokinase was applied to a DEAE-column (5 of which the molecular weight was assumed to be about ,X 25 cm) which had been equilibrated with the same buffer as 100,000. The later one, p-2, was eluted as the major active that used for the gel fitration. The flow-through effluents (3.2 peak with a small amount of protein at a molecular weight of liters) were then collected at a flow rate of 5 ml/min and were

designated Step 6-bilokinase. At this step, the specific activity was elevated to twice that in the previous step, producing a

-1.2 reduction in the protein content without a loss of activity, as shown in Table I.

Step 7: SP-Sephadex Column Chromatography-Step 6- bilokinase was concentrated, dialyzed, and applied to a SP- Sephadex column. The details are shown in the legend of Fig.

0.4"0.8 3. The bilokinase activity was found as a single peak which E j appeared at an NaCl concentration of more than 0.3 M at pH

3-.0.6 2 4.8 (Fig. 3). The active fraction obtained from 5 runs through .4 the columns amounted to 1.5 liters in volume. This was

0 2 - 0 . 4 dialyzed against 5 liters of ammonium acetate buffer for 24 h, changing the dialyzates every 8 h, and then concentrated to

0.1..0.2 about 100 ml using a Diaflo membrane. The dialyzed and concentrated material was designated Step 7-bilokinase. This represented the final bilokinase preparation (referred to below

0 50 100 150 200 250 300 as bilokinase) which was used in subsequent experiments. Tube number Lyophilized bilokinase containing 20 mg of protein with an

F ~ ~ ; , 1, column chromatography of Step 3-bilokinase on C" activity of 1.2 X io5 I u was finally obtained and stored at -80 Sephadex C-25. One-third of the volume (1.4 liters) of Step 3- "C. The purity and recovery at each step of the purification bilokinase (2,700 mg of protein, 5.1 X IO" IU in 4.2 liters) was applied process are summarked in Table I. to a CM-Sephadex column (5 X 30 cm). Elution of bilokinase was performed with an NaCl gradient system at a constant pH of 4.3. Fifteen-ml fractions were collected. -, absorption at 280 nm ( O B . 280nm); M, fibrinolytic activity assayed by using 30 pI of each The results of disc electrophoresis are shown in Fig. 4. effluent with plasminogen-rich fibrin plates as described under "Ma- Bilokinase yielded a single band which migrated towards the terials and Methods." - - -, NaCl concentration measured by a chlo- cathodic side of an acidic gel, pH 4.3 (Fig. 4A). Densitometric ride ion electrode (TOA Electronic Inc., Tokyo). The bilokinase analysis and fibrinolytic activity measurements were per- activity peak was detectable in the sohtion eluted at a concentration formed simu~taneous.y. T~ the latter, unstained gel of above 0.4 M NaCl and was equal to hut did not coincide with the

nating the active fraction. The active portion found in tubes 100-200 and Methods" and was sliced into l-mm wide strips and was pooled and subjected to further purification in Step 5. placed on a fibrin plate. The results are shown in Fig. 4, B and

,,"---------- 0.5"l.O

0

Homogeneity

peak for protein, suggesting that some inert proteins were contami. was prepared under the conditions described under

Biliary Plasminogen Activator 625

- 0.8

600 - 6

- 0.6

a 0 . 5 - - - z Y

A .% 400 ' % 0.4.-0.4

0

U .- c

G 0.2.-0.2

0.1"

20 40 60 80 100

Tube number FIG. 3. SP-Sephadex column chromatography of Step 6-bi-

lokinase. This obtained as a 3.2-liter fraction through a DEAE- Sephadex column was concentrated to 150 ml using polyvinylpyrrol- idone as described in Step 4 and dialyzed against 5 liters of sodium acetate buffer ( 5 0 m ~ , pH 4.8, containing 0.1 M NaCI). Thirty-ml portions of the dialyzed concentrate were applied to an SP-Sephadex column (2.6 X 40 cm) equilibrated with the dialysis buffer. The column was eluted using an NaCl concentration gradient from 0.1-0.8 M (- - -). Six-ml fractions were collected, and 30 pl of each fraction were used for determining the fibrinolytic activity. (W) on fibrin plates. The solid line shows Am ,,,,, (O.D. 28Onm).

Heterogeneity in Isoelectric Points Isoelectric focusing yielded five peaks. All of these were

shown to be due to basic proteins (Fig. 6 ) .

Molecular Weight The molecular weight of bilokinase was estimated from the

VJV, ratio on Sephadex G-100 to be 58,000 (Fig. 7). Molecular weight estimation by sedimentation equilibrium gave a mo- lecular weight of 57,000 that was 1000 lower than that obtained by gel filtration. Since the multichannel cell measurement was considered to be the more precise, the molecular weight of bilokinase was taken to be 57,000.

Fibrinolytic Activity The optimal pH was found by the fibrin plate method to be

7.8. A t pH 7.5-8.0, bilokinase demonstrated a uniform maxi- mum activity. No fibrinolytic activity was shown by bilokinase on plates made of plasminogen-free fibrin and of fibrin with trans-(aminomethy1)cyclohexanecarboxylic acid added at a concentration of 10 mM.

Thermostability of Bilokinase Bilokinase at pH 7.8 was found to be stable for 3 months

when frozen at -80 "C, for 2 weeks at -20 "C, and for several

1 10 20 30 40

Slice number

FIG. 5. Sedimentation pattern of bilokinase. Lyophilized bilo- kinase was dissolved in 50 m~ acetate buffer (containing 0.35 M NaCI). The photograph was taken 67 min after a speed of 55,500 rpm was attained. The details are described under "Materials and Meth- ods".

A Scan number (mm)

FIG. 4. Polyacrylamide gel disc electrophoresis of bilokinase at pH 4.5. Electrophoresis was carried out in 7.5% gel a t 4 "C for 40 min using 8 mA/tube. The amounts of protein applied were about 20 pg. Staining was with 0.02% Coomassie Blue in 10% acetic acid solution. A, stained gel; B, a pattern measured with an 02-804 densitometer (Asuka Manufacturers Ltd., Tokyo, Japan) at 500 nm; C, fibrinolytic activity shown on gel slices of unstained gel. Slices about 1-mm thick were placed on fibrin plates, and the lysis zones were measured after incubation for 18 h at 37 "C.

C, respectively. The protein and activity migration distances were assumed to be coincident.

Fig. 5 illustrates the sedimentation pattern. Bilokinase yielded a single peak. The S P O . ~ was 3.32.

100- n

2 80- E

h

's 60. cr

.I

2 40-

0

20.

1.2 12 I

40 BO 80 100

Tube number FIG. 6. Isoelectric focusing of bilokinase. Bilokinase (2000 IU)

was submitted to electrophoresis in a column containing 0.5% (pH 2- 10) ampholyte. Electrophoresis was carried out a t 4 "C for 48 h at 900 V/1 mA. The effluent was collected in I-ml portions. - - -, measured pH; -, absorbance at 280 nm (O.D. 280nm); M and shadow, fibrinolytic activity.

Biliary Plasminogen Activator 626

2.

1.1

1.1

8 >" 1.4

\

1.:

1 .1

0 -

B -

6 -

4 -

2 -

D - 1

\\* Soybean trypsin inhibitor

( 21500 )

( 34000 ) Ovalbumin ( 45000 )

Molecular weight x1O" FIG. 7. Estimation of the molecular weight of bilokinase by

gel filtration. A column (2.6 X 100 cm) was packed with Sephadex G-100 equilibrated with Tris buffer, pH 7.8 (50 m ~ , containing 0.15 M NaC1). Two-ml samples of bilokinase (BK) (Step 7-bilokinase) and low molecular weight urokinase (UK) preparation obtained from Mochida Pharmaceutical Co. Ltd. (Tokyo) were applied to the column as a solution containing about 1,000 IU each, and their elution volumes (V,) were estimated by measuring fibrinolytic activity. The other standard protein preparations, purchased from Boehringer Mannheim GmbH were applied as a solution of 30 mg each dissolved in 5 ml of Tris buffer. The void volumes (V,) were determined by noting the appearance of blue dextran 2,000 (Pharmacia, Uppsala) which was dissolved in the enzyme solutions. Bilokinase is shown by the arrow, and the VJV, value was determined to be 1.35, correspond- ing to a molecular weight of 58,000.

days at 0 "C. Bilokinase was unstable, however, a t elevated temperatures. Exposure to 25 "C for 100 min caused it to lose 20% of its initial activity, and exposure to 60 "C for 100 min caused a 40% loss. I t was denatured immediately a t 100 "C.

Stability in Relation to p H Bilokinase was stable in an acidic solution, especially around

pH 4, at which 83% of its initial activity was retained, even when kept a t 26 "C for 5 h. However, at pH 6.0 or more, over 50% of the activity was lost.

Concentration and Activity Relationship The effect of bilokinase concentration on fibrinolytic activ-

ity was assessed by the fibrin clot lysis technique, and a comparison was made with urokinase. The concentration/ activity curves obtained for bilokinase and urokinase showed almost the same slopes on a log-to-log scale (Fig. 8). It is therefore assumed that bilokinase activates plasminogen in a way similar to urokinase.

Hydrolysis of Synthetic Substrates The Lineweaver-Burk plot for bilokinase and urokinase

using N"-acetylglycyl-L-lysine methyl ester-acetate as sub- strate showed the K, and V,,, values indicated in Table 11. Only a moderate K,,, value was found with bilokinase, similar to urokinase. From the slight differences in K, values, it cannot be assumed that the enzyme are distinct as regards to their affinities for the substrate. The V,,, values were almost identical. The Ki values were obtained by adding trans-(ami- nomethy1)cyclohexanecarboxylic acid to the reaction mix-

tures. trans-(Aminomethy1)cyclohexanecarboxylic acid acted as a competitive inhibitor on both the bilokinase- and uroki- nase-catalyzed hydrolysis of Ne-acetylglycyl-L-lysine methyl ester-acetate and yielded simiiar K, values. The amidolytic activities of bilokinase, urokinase, and plasmin were also com- pared with S-2227 as a substrate. With substrates, bilokinase

2000 -

1000 -

3 ?

600 - \ - 260 -

100 -

O L ' I.

I 1 I 2 6 10 20 Lysis time (100/min)

FIG. 8. Concentration/activity curves for bilokinase and urokinase. The relation between the units and fibrin clot lysis time was determined. The clots were prepared from 0.4 ml of 0.14% bovine fibrinogen solution, 0.1 ml of bilokinase or urokinase, and 0.05 ml of 20 units/& of bovine thrombin. Borate buffer, pH 7.8 (see "Materials and Methods"), was used as a solvent. Both bilokinase and urokinase were diluted to give a lysis time from 10-50 min. The curve for bilokinase (M) was drawn to intersect at the 12.5-min point (600 IU) on the curve obtained from urokinase (M), which was drawn using standard known units.

TABLE 11 Hydrolysis of N"-acetylglycyl-L-lysine methyl ester-acetate by

bilokinase and urokinase Bilokinase or urokinase containing 150 IU was present in the 1.5

ml of 0.15 M KC1 solution to be titrated. The titrant was 50 mM KOH. Titrations were performed at pH 7.8, 35 "C. To obtain K,, trans- (aminomethv1)cvclohexanecarboxvlic acid was used at a concentra-

~~

M pmol/min/IU M

Bilokinase 1.2. 1.58. IO"' 1.02. Urokinase 8.0. 1.45. lo-' 8.04. lo-"

TABLE I11 Amidolysis of S-2227 (H- L-Glu-Gly-Arg-pNA) by bilokinase,

urokinase, and plasmin K , V*W. M

Bilokinase" 8.3. 5.6. pmol/min/100 IU Urokinase* 2.6. 7.0. IO-' pmol/min/100 IU Plasmin' 2.0. 5.7.10" pmol/min/100 CU

a A mixture of 0.5 ml of bilokinase (60 IU) and 0.12 ml of serially diluted substrate of less than 4.8 m~ (pH 8.5, ionic strength = 0.05) was incubated at 37 "C for 10 min. During the incubation, the amounts of pNA released were recorded continuously at 405 nm.

" A mixture of 0.5 ml of urokinase (60 IU) and 0.12 ml of the substrate solution of the same concentration as that used for biloki- nase was incubated, however, the assay was made at pH 9.0 at which urokinase showed the highest splitting activity.

e A mixture of 0.5 ml of plasmin (AB KABI) of 1 casein unit (CU) and 0.12 ml of substrate, pH 9.0, was incubated in the same manner as described above.

Biliary Plasminogen Activator 627

and urokinase gave the same K m values of the order of The V,,,, assessed using identical fibrinolytic activity (60 Iu in the reaction mixtures) with both enzymes, showed exactly the same values, as indicated in Table 111. Plasmin, however, gave a Km value which was as much as 2 orders larger than that of bilokinase and urokinase (Table 111). It is apparent, therefore, that bilokinase constitutes an enzyme which is similar to urokinase but clearly distinct from plasmin.

DISCUSSION

Many body fluids have been found to contain plasminogen activators, such as urine (11, 12), blood (13, 14), tears (15), milk (16), amniotic fluid (17), and abdominal ascitic fluid (18). Bile, however, was long believed to be one of the exceptions until the discovery of bilokinase by Oshiba et al. (1). The fibrinolytic activity caused by bile acids reported by Olesen (19) and Fleming and Norman (20) turned negative on using fibrin plates with calcium added. Such fibrinolytic activity is thus clearly different from the enzymatic activity caused by bilokinase (1). Gibinski and Gasinski (21) reported that bile obtained from gallstone patients and then diluted shortened euglobulin lysis time; however, they concluded that this action was not the result of plasminogen activation. Kulapongs and Bachmann (22), using a rat liver perfusion, reported that the addition of urokinase to the perfusate caused it to be excreted into the bile. However, the activity of the bile was quite weak, only 2-576 of the activity of the urokinase. Since the circula- tory blood did not show any demonstrable activator activity, bile activity does not present positive evidence of the existence of urokinase in the bile. Urokinase excretion into the bile was also noted by Jedrychowski et al. (23) in their experiments using an extracorporeal perfusion of pig liver. At present, it can be assumed that the biliary plasminogen activator is not urokinase but that urokinase must act as a stimulus for its excretion. King (24) found a protein of low molecular weight (around 15,000) in mammalian bile which he termed cholely- sin. It showed fibrinolytic activity, but no plasminogen acti- vator activity, and is considered to be quite a different enzyme from bilokinase. As positive evidence of the existence of a bile activator, Noren et al. (25) found a distinct activity in human bile using their newly developed fibrin plate method.

One of the reasons for the difficulty in establishing the presence of a biliary plasminogen activator may be its low activity in bile. Oshiba et al. (1) initially assessed the utility of lysis using the fibrin plate method with gall bladder bile obtained from a dog and subsequently published a compara- tive study on the activity of hepatic bile and gall bladder bile in nine mammals (26). The bile used in the present study was bovine bile, which ranked medium in its activity among the various mammalian biles. In the original purification of bilo- kinase, Oshiba et al. (1) performed gel fitration of dog bile using a Sephadex G-75 column and noted an activity in the flow-through fraction. It was shown by Ariga (2) that the active factor in bile could be precipitated and separated from the large amounts of precipitates obtained by conventional precipitation methods. A novel precipitation method was then developed which permitted a fraction of higher activity to be obtained from the bile (2). This method was applied in the present work.

The final preparation showed a purity of 6,556-fold Over that of the bile; however, the specific activity remained rather low (5,900 IU/mg of protein) compared to the activity of urokinase obtained by Ogawa et al. (27), which was as high as 218,000 CTA (Committee of Thrombolytic Agents) units/mg of protein. The reason for this difference is still not known. The possibility of a high degree of inert protein contamination in the bilokinase preparation is not suspected, due to the

single protein band obtained in association with the bilokinase activity on disc electrophoresis and because a single peak was shown by sedimentation velocity. Contamination by a higher molecular weight albumin (67,000) or globulin (150,000) there- fore appears unlikely. Bilokinase also showed similar K , and K , values to urokinase in its hydrolysis of synthetic substrates, indicating a similarity between these two enzymes. Thus, the major cause of any lowering of the specific activity of biloki- nase would be denaturation of the bilokinase protein, and the denatured protein may not be removed from the intact bilo- kinase at any point throughout the purification procedures.

In bilokinase, there were 5 active fractions which migrated to distinct isoelectric points. All of the fractions migrated towards the alkaline side, but a majority of 52% were grouped at pH 8. Although the heterogeneity of bilokinase in respect to its isoelectric migration points agreed with that of urokinase as described by Soberano et al. (28) and Ogawa et al. (27), no identical points were noted. The detailed properties of each individual component, such as the molecular weight and spe- cific activity, remain unknown. Bilokinase is a heat-labile and an acid-stable protein. The fibrinolytic activity of bilokinase is quite similar to that of urokinase. The optimal pH at which bilokinase acts as a plasminogen activator is 8.5, which is identical with that of urokinase as estimated by Ploug and Kjeldgaard (29), Johnson et al. (30), Loland and Condit (31), and Ogawa et al. (27).

The physiological function and origin of bilokinase are not well known at present. Oshiba and Schoenfield (32) reported that on isolated hamster liver, perfusion produced a parallel excretion of bile acids and bilokinase activity, but there was no relation between the activity and bile flow. They postu- lated, therefore, that bilokinase may be produced in the liver and excreted into the bile upon receiving bile acid (taurocho- late) stimulation.

Fletcher et al. (33) and Kulapongs and Bachmann (22) found that the liver acts as a clearance system for urokinase and other plasminogen activators. They considered bile to play a role as a major excretion route for injected urokinase. However, only a trace of urokinase was observed in the bile even though extremely large amounts of urokinase were ad- ministered. Their findings thus do not seem to be pertinent to the true excretion or production of bilokinase. It was proposed that the liver acted to reduce the presence of blood plasmin- ogen activators by Kwaan et al. (34), von Kaulla (35), and Tytgat et al. (36), and so a hyperfibrinolytic state might occur in association with cirrhosis of the liver. It should be pointed out that another possibility is the existence of bilokinase in the bile, separating these activators upon an impairment of liver function. The physiological function of bilokinase may be regarded as broadly similar to that of the large numbers of other plasminogen activators. The relation between liver func- tion and bilokinase activity remains unclear, and the role of bilokinase in coagulation and homeostasis or in gallstone formation is still to be determined.

REFERENCES 1. Oshiba, S., Hata, S., and Okamoto, S. (1969) Jpn. J. Physiol. 19,

2. Ariga, T. (1973) Nihon Uniu. Med. J . 15, 209-223 3. Astrup, T., and Mullerz, S. (1951) Arch. Biochern. Biophys. 40,

4. Reisfeld, R. A., Lewis, U. J., and Williams, D. E. (1962) Nature

5. Vesterberg, O., and Svensson, H. (1966) Acta Chern. Scand. 20,

6 . Svedberg, T., and Pedersen, K. 0. (1940) The Ultracentrifuge,

7. Yphantis, D. A. (1960) Ann. N . Y. Acad. Sci. 88, 586-601 8. Wdton, P. L. (1967) Biochim. Biophys. Acta 132, 104-114

212-219

346-351

(Lond.) 195,281-283

820-834

Oxford University Press, London

628 Biliary Plasminogen Activator

9. Pamphlet of AB KABI (1977) Stockholm, Sweden 10. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.

11. Astrup, T., and Sterndorff, I. (1952) Proc. Soc. Exp. Biol. Med.

12. Sobel, G. W., Mohler, S. R., Jones, N. W., Dowdy, A. B., and Guest, M. M. (1952) Am. J. Physiol. 171, 768-769

13. Olesen, E. S. (1960) Thromb. Diath. Haemorrh. 5,97-107 14. Semar, M., Skoza, L., and Johnson, A. J. (1969) J. Clin. Inuest.

15. Storm, 0. (1955) Scand. J. Clin. Lab. Inuest. 7,55-58 16. Astrup, T., and SterndorfY, I. (1953) Proc. Soc. Exp. Biol. Med.

17. Albrechen, 0. K., Storm, O., and Classen, M. (1958) Scand. J.

18. Albrechen, 0. K., and Troll, E. D. (1955) Acta Haematol. (Basel),

19. Olesen, E. S. (1960) Thromb. Diath. Haemorrh. 4,473-481 20. Fleming, W. H., and Norman, P. S. (1960) Proc. Soc. Erp. Biol.

21. Gibinski, K., and Gasinski, J . (1972) Acta Med. Pol. 13, 199-204 22. Kulapongs, P., and Bachmann, F. (1968) Fed. Proc. 27,569 23. Jedrychowski, A,, Rarbhov, S. P., and Hillenbrand, P. (1972) Gut

(1951) J. Biol. Chem. 193,265-275

81,675-678

48, 1777-1785

84,605-608

Clin. Lab. Inuest. 10, 310-318

14,376-382

Med. 104,636-638

13.54-57

24. King, J. B. (1972) Thromb. Diath. Haemorrh. 28, 299-305 25. Noren, I., Ramstrom, G., and Wallen, P. (1975) Haemostasis 4,

26. Oshiba, S., Sawai, H., Ariga, T., Imai, H., and Kato, J . (1976) The

27. Ogawa, N., Yamamoto, H., Katamine, T., and Tajima, H. (1975)

28. Soberano, M. E., Ong, E. B., Johnson, A. J., Levy, M., and

29. Ploug, J., and Kjeldgaard, N. 0. (1957) Biochim. Biophys. Acta

30. Johnson, A. J., Kline, D. L., and Alkjaersig, N. (1969) Thrornb.

31. Loland, L., and Condit, E. V. (1965) Biochemistry 4,265-270 32. Oshiba, S., and Schoenfield, L. J. (1970) Proc. SOC. Exp. Biol.

33. Fletcher, A. P., Biederman, O., Moore, D., Alkjaersig, N., and

34. Kwaan, H. C., Lo, R., and McFadzean, A. J. S. (1959) Clin. Sci.

35. von Kaulla, K. N. (1964) Lancet 1, 1046-1047 36. Tytgat, G., Collen, D., DeVreker, R., and Verstraete, M. (1968)

Acta Haematol. (Basel) 40,265-274

110-124

16th International Congress of Hematology pp. 349 (abstr.)

Thromb. Diath. Haemorrh. 34, 194-209

Schoellmann, G. (1976) Biochim. Biophys. Acta 445,763-773

24,278-282

Diath. Haemorrh. 21,259-272

Med. 133,89-92

Sherry, S. (1964) J. Clin. Inuest. 43, 681-695

(Lond.) 18,251-261