Purification and characterization of piscivorase i and ii, the fibrinolytic enzymes from eastern cottonmouth moccasin venom (Agkistrodon piscivorus piscivorus)

Toxicon, Vol. 33. No. 7, pp. 929-941. 1995 Copyright 0 1995 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0041~)101/95 $9.50 + 0.00

0041-0101(95)00008-9

PURIFICATION AND CHARACTERIZATION OF PISCIVORASE I AND II, THE FIBRINOLYTIC ENZYMES FROM EASTERN COTTONMOUTH MOCCASIN VENOM (AGKISTRODON PISCIVORUS BUM-SO0 Natural

Products

PISCIVORUS)

HAHN, IL-MOO CHANG and YEONG-SHIK Research

(Received

Institute, Seoul National University, Seoul 110-460. South Korea 1 I October

28 Yungun-Dong,

1994; accepted 9 December

KIM* Jongro-Ku.

1994)

B.-S. Hahn, I.-M. Chang and Y.-S. Kim. Purification and characterization of piscivorase I and II, the fibrinolytic enzymes from eastern cottonmouth moccasin venom (Agkistrodon piscivorus piscivorus). Toxicon 33, 929-941, 1995.-Fibrinolytic enzymes, piscivorase I and II, were isolated from Agkistrodon piscivorus piscivorus (eastern cottonmouth moccasin) venom using gel filtration on Bio-Gel P-100 and ion-exchange chromatography on CM-Sepharose CL-6B. The mol. wts of these proteases, piscivorase I and II, are 23,400 and 29,000 and isoelectric points are 6.6 and 8.5, respectively. These fibrinolytic enzymes were homogeneous by SDS-polyacrylamide gel electrophoresis. Piscivorase I readily cleaved the Act- and BP-chain of fibrinogen, but piscivorase II cleaved readily the Am-chain and more slowly the BP-chain. These fibrinolytic enzymes were activated by Ca2+, Mg2+ and Ba’+, but inhibited by Zn’+, Cu2+ and Mn2+. Both fibrinolytic enzymes were also inhibited by cysteine, fl-mercaptoethanol, and by metal chelators such as EDTA and EGTA, but not by benzamidine, phenylmethanesulfonyl fluoride (PMSF), soybean trypsin inhibitor and aprotinin. These fibrinolytic enzymes did not act like thrombin, plasmin and kallikrein, using specific chromogenic substrates. Neither fibrinolytic enzyme induced platelet aggregation, and piscivorase I showed low haemorrhagic activity at dosages of 55 pg.

INTRODUCTION

Snake venoms, particularly those belonging to Crotalidae and Viperidae families, are known to affect strongly the blood coagulation system. Fibrinogen clotting enzymes (thrombin-like enzymes) have been isolated and characterized from the venoms of Agkistrodon rhodostoma (Nolan et al., 1976), Bothrops atrox (Stocker and Barlow, 1976) Agkistrodon contortrix contortrix (Herzig et al., 1970) and Lachesis muta muta (Yarleque * Author

to whom

correspondence

should

be addressed. 929

930

B.-S. HAHN et al.

et al., 1989). Several venoms also contain

proteases capable of solubilizing fibrin or rendering fibrinogen incoagulable. Fibrino(geno)lytic enzymes have been reported from the venoms of A. acutus (Ouyang and Huang, 1976), A. contortrix contortrix (Ahmed et al., 1990; Guan et al., 1991), A. rhodostoma (Ouyang et al., 1983), A. halys brevicaudus (Xiaohong and Liren, 1991), A. piscivorus piscivorus (Nikai et al., 1988) A. piscivorus conanti (Retzios and Markland, 1990) and Crotalus atrox (Willis and Tu, 1988). The fibrino(geno)lytic enzymes fall into two groups, based on their mol. wts. They have mol. wts of approximately 25,000 and 60,000. The most characteristic aspect of the amino acid composition of these proteases is the very high levels of Asx and Glx residues. Their potential use as thrombolytic agents is being investigated in various fields. In this study we will focus on the purification and characterization of two fibrinolytic enzymes from A. piscivorus piscivorus (eastern cottonmouth moccasin).

MATERIALS

AND METHODS

The venom of A. pisciuorus piscivorus was purchased from Sigma (St Louis, MO, U.S.A.). Bovine fibrinogen, thrombin, plasmin, azocasein, chromogenic substrates (T-1637, T-6140 and B-2133) and CM-Sepharose CL-6B were also obtained from Sigma. Bio-Gel P-100 and Coomassie blue G-250 protein assay reagent were purchased from Bio-Rad (New York, U.S.A.).

Venom fractionation

Crude venom (200 mg) was dissolved in 1 ml of starting buffer, 40 mM Tris-HCl (PH 7.1) containing 0.1 M NaCl, and centrifuged at 1600 x g for 10 min to remove insoluble material. The supernatant was applied to a Bio-Gel P-100 column (1.5 x 110 cm) equilibrated and eluted at 2.4 ml/hr with starting buffer. The eluant was monitored by spectrophotometry at 280 nm. The fractions eluted by Bio-Gel P-100 were collected and dialysed in 10 mM Tris-HCl @H 8.3) containing 2 mM CaCl, and 5 mM NaCl. The fractions dialysed against 10 mM Tris-HCl (pH 8.3) containing 2 mM CaCl, and 5 mM NaCl were applied to a CM-Sepharose CLdB column (1.5 x 8.5 cm) and eluted at a Bow rate of 15 ml/hr. Bound proteins were eluted using a linear salt gradient from 0.005 M to 0.2 M and step gradient at 0.3 M with a total volume of 130 ml.

Proteolytic

activity

Proteolytic activity was measured with azocasein according to the previous method with slight modifications (Beynon and Kay, 1978). The reaction mixture contained 1 ml of azocasein (2 mg/ml in 0.2 M borate buffer, pH 7.8). The digestion was initiated by adding crude venom, piscivorase I or piscivorase II. The mixtures were incubated at 37°C for 1 hr. Immediately, 0.25 ml of the mixtures was transferred into a 1.5 ml capped microfuge tube containing 1 ml of 5% (w/v) trichloroacetic acid and mixed. The tubes were centrifuged at 11,000 x g for 5 min and the absorbance of the supematant was measured at 340 nm.

Fibrinolyric activity

Fibrinolytic activity was estimated by a modified fibrin plate assay method (Astrup and Mullertz, 1952) using plasmin as a standard (0.002~.015 U). The fibrinogen solution (10 ml of 0.8% bovine fibrinogen in 0.2 M borate buffer, pH 7.8) was pipetted into a petri dish (1.5 x 9 cm) and mixed with 0.5 ml of 10 NIH U/ml thrombin in 0.9% saline. The dishes were allowed to stand for 30 min at room temperature to obtain a fibrin clot layer. Fifteen microlitres of piscivorase I (PI), piscivorase II (PII) and crude venom were spotted on the fibrin layer and incubated at 37°C for 6 hr. Fibrin clearance was then determined by measuring two perpendicular diameters of the lysed zone. Incubations were carried out in triplicate and the mean of six diameters was taken as the diameter of the zone of fibrin plate clearance. Fibrinolytic activity was also measured by SDS-polyacrylamide gel electrophoresis (PAGE) according to the modified method (Ahmed ez al., 1990). A fibrin clot was formed by incubating 500 ~1 of 0.8 mg/ml fibrinogen solution in 5 mM imidazole-saline buffer, pH 7.4, containing 5 mM CaCl, and 10 ~1 of 1000 U/ml human thrombin for 1 hr at 37°C. The fibrin clot was centrifuged at 11,000 x g for 5 min, followed by three washes (3 ml each) with 5 mM imidazole-saline buffer. The clot was resuspended in 500 ~1 of imidazole-saline buffer and sonicated on ice to disperse the clot. A time course study was initiated by addition of 13 pg of PI and PI1 to fibrin clot and the suspension was incubated at 37°C. Aliquots were removed after 0.5, 1, 3, 5 and 24 hr of incubation, mixed with an equivalent volume of SDS-PAGE sample buffer, and electrophoresed on SDS-polyacrylamide gel.

931

Fibrinolytic Enzymes from A. p. piscivorus Fibrinogenolyiic

activity

Fibrogenolytic activity was measured by incubating 200 ,ul of human fibrinogen (1 mg/ml in 40 mM Tris-HCl, pH 7.4, containing 0.1 M NaCl) with 6 pg of PI and PII at 37°C. At various time intervals (0.5, 1, 3,5, and 24 hr), aliquots of the incubation mixture were withdrawn and mixed with an equal volume of SDS-PAGE sample buffer, boiled, and electrophoresed by SDS-PAGE. Fibrinogenolytic activities were determined by a modified method of Maeno and Mitsuhashi (1960). Crude venom, PI or PII (50 ~1 each) in 5 mM imidazole-saline (pH 7.4) and 1% bovine fibrinogen solution (100 ~1) were mixed and incubated at 37°C for 30 min. Trichloroacetic acid (0.4 M, 200 ~1) was added to stop the reaction. After centrifugation at 11,000 x g for lOmitt, Na,CO, (20%, 200 ~1) and phenol reagent (50 ~1) were added to the supernatant, and the mixture was left for 20 min at room temperature. The absorbance was measured at 540 nm. According to a standard plasmin (0.00220.01 U) and tyrosine curve, the fibrinogenolytic activity could be calculated. Effect of pH

The effect of pH on PI and PI1 was measured by alteration of proteolytic activity. PI and PI1 (6 pg each) were dissolved in citrate buffer (50 mM, pH 336), Tris-HCI buffer (50 mM, pH 7-9) or phosphate buffer (50 mM, pH 10-l 1). Then, both proteases were incubated at room temperature for 2 hr. The effect of pH on the proteolytic activity was determined using azocasein as a substrate. E#ect of temperature

PI, PI1 (6 pg each) and crude venom (100 pg) in 40 mM Tris-HCl buffer (pH 8.5) were incubated for 15 min at 10, 20, 30, 40, 50, 60, 70 and 80°C. The samples were then cooled on ice and the effect of temperature on the proteolytic activity was determined using azocasein as described above. Inhibition studies

PI and PI1 (6 fig each) in 10 mM Tris-HCl (pH 8.3) were incubated at 37°C for 1 hr with several inhibitors (2-10 ~1) at 100 ~1 of final volume in 10 mM Tri-HCl (pH 8.3) and the proteolytic activity was determined using azocasein as a substrate. The inhibitors used for inhibitory effects were 10 mM bcnzamidine, 5 mM EDTA, 5 mM EGTA, 5 mM cysteine, 2 mM PMSF, 10 mM j?-mercaptoethanol, 50 FM soybean trypsin inhibitor and 50 pM aprotinin. The activity was expressed as a percentage of control (the activities of PI or PI1 alone with no inhibitors).

1

-

0.8

-

53

0.6

-

e 2 2

0.4

-

g Z c!

i

0

20

40

60

80

100

Tube No. Fig. 1. Gel filtration on Bio-Gel P-100. The column (1.5 x 110 cm) of Bio-Gel P- 100 was equilibrated with 40 mM Tris-HCl @H 7. I) containing 0.1 M NaCl. Crude venom (200 mg) was applied to the column. Elution was performed with starting buffer at 4°C at a flow rate of 2.4 ml/hr. The eluant was monitored by spectrophotometry at 280 nm (-•-). Fibrinolytic activity (-_O-_) was measured and fractions were pooled (ti).

932

B.-S. HAHN et al.

E

1 0.8

g Y

ii 0.6 s g 0.4 (II 0.2

2

0 0

10

20

30

40 Tube

50 No.

60

70

60

Fig. 2. Ion-exchange on CM-Sepharose CL4B. The column (1.5 x 8.5cm) of CM-Sepharose CL4B was equilibrated with 1OmM Tris-HCI @H 8.3) containing 2 mM CaCl, and 5 mM NaCl. Fifteen milligrams of material from the previous step (peak 4) was applied to the column. Elution was performed with a linear salt gradient from 0.005 M to 0.2 M and step gradient at 0.3 M with total volume of 130 ml at a Row rate of 15 ml/hr. The eluant was monitored by spectrophotometry at 280nm (-a--). Fibrinolytic activity was measured (-I-J-).

Eflect of metal ions

PI and PI1 (6pg each) in IO mM Tris-HCl (pH 8.3) were preincubated with some divalent cations at 37°C for 2 hr. Proteolytic activity was determined using azocasein as a substrate. The following metal ions were tested: CaCl,, MgCl,, BaCl,, CuSO,, ZnCI, and MnCl,, each at a final concentration of 5 mM. PIatelef aggregation

This was determined according to the method of Willis and Tu (1988). Blood from Sprague-Dawley rats (280 g) was collected in 3.8% trisodium citrate (9 vol. of blood/l vol. of citrate) and immediately centrifuged at room temperature for 10 min at 220 x g to obtain PRP (platelet rich plasma). PPP (platelet poor plasma) was prepared by recentrifugation of precipitate at 1600 x g for IO min. Platelet aggregation was recorded with a Chrono-Log aggregometer (Model SOO-VS)at 37°C under continuous stirring at 900 rpm. The reaction mixture consisted of 450 ~1 of PRP which was preincubated for 1min at 37°C prior to the addition of 50 ~1 of PI, PII (30 pg each), crude venom (30-50 pg) or 0.2 mM ADP: The maximum aggregation response obtained from the addition of ADP gave a value of 100% aggregation. Table 1. Purification of two proteases (piscivorase I and II) from A. p. piscivorus Protein (mg)

Purification step PI Crude Bio-Gel P- 100 CM-Sepharose CL-6B

PI1 200 15

9.62

1.14

Azocaseinolytic activity (U/mg)* PI PII 6.11 20.3 9.25 18.0

Fibrinolytic activity (U/mg)t PI PII 0.036 0.082 0.092 0.027

Fibrinogenolytic activity (U/mg)S PI PI1 0.26 0.32

0.058

*One unit of azocaseinolytic activity is defined as the amount of enzyme which causes a net increase of 1 of absorbance at 340 nm in 1 hr. t$One unit is U of the plasmin (one unit produces a change in A 275of 1.O in 20 min at pH 7.5 at 37°C when measuring perchloric acid soluble products from #-casein in a final volume of 5 ml). Values represent means of triplicate experiments, which produced similar results.

Fibrinolytic Enzymes from A. p. piscivorus Haemorrhagic

933

activity

Haemorrhagic activity was determined by the method of Bjamason and Fox (1983). Three ICR mice (22-24 g) were each injected subcutaneously in the back with the test samples dissolved in 0.1 ml of 0.9% saline at various concentrations. The minimum haemorrhagic dose (MHD) was defined as the least amount of protein that caused a haemorrhagic reaction 5 mm in diameter 6 hr after injection. Activity

on chromogenic

substrates

Activities were measured using chromogenic substrates, N-benzoyl-Pro-Phe-Arg-p-nitroanilide hydrochloride (B-2133), N-p-tosyl-Gly-Pro-Arg-p-nitroanilide acetate (T-1637) and N-p-tosyl-Gly-Pro-Lys-p-nitroanilide acetate (T-6140), specific for kallikrein, thrombin and plasmin, respectively. Activities were tested by mixing PI (6 pg/200 pl), PII (6 pg/200 ~1) or crude venom (lOpg/200 ~1) in 10 mM Tri-HCl (PH 8.3) and 200~1 of 0.5 mM substrates. The mixtures were incubated at 37°C for 1 min. The reaction was stopped by the addition of 100 ~1 of 50% acetic acid to solution. Absorbance was measured at 405 nm. Plasmin (0.01 U) and thrombin (0.135 NIH U) in 100 pl of 10mM Tris-HCl (pH 8.3) were used as controls. Amino acid analysis This was carried out on a Waters PICO.TAG amino acid analyser, using a PICO.TAG column. Salt-free PI and

PI1 were hydrolysed with constant-boiling hydrochloric acid at 110°C under vacuum for 24 hr. The sample was derivatized with phenylisothiocyanate and the amino acids were calculated from Pierce amino acid standard solution. RESULTS

Venom fractionation Two fibrinolytic enzymes were isolated using two-step fractionation (Figs 1 and 2). The first chromatography of crude venom using a column of Bio-Gel P-100 yielded four major

Fig. 3. SDS-PAGE analysis of piscivorase I and II. Lanes 1 and 6, A mixture of bovine serum albumin (66,000), ovalbumin (4S,OOO),glyceraldehyde-3phosphate dehydrogenase (36,000), carbonic anhydrase (29,000), trypsinogen (24,000), soybean trypsin inhibitor (20,100) and a-lactalbumin (14,200); lane 2, crude venom; lane 3, Bio-Gel P-100 (peak 4); lane 4, piscivorase I; lane 5, piscivorase II.

934

B.-S. HAHN

et al.

fractions, Pl, P2, P3, and P4 (Fig. 1). The fibrinolytic activities were found in the fractions Pl and P4 when tested on fibrin plate. The activities of Pl and P2 on fibrin clot were 0.045 and 0.082 U/mg, respectively. Further chromatography of fraction P4 using a column of CM-Sepharose CL-6B resulted in four fractions, Cl, C2, C3 and C4 (Fig. 2). The fibrino(geno)lytic and azocaseinolytic activities were observed in fractions Cl and C4. The results and recovery are shown in Table 1. Both PI and PII were found to be homogeneous by SDS-PAGE. The mol. wts of these fibrinolytic enzymes, PI and PII, are 23,400 and 29,000 (Fig. 3) and isoelectric points are 6.6 and 8.5, respectively (data not shown).

14

Fig. 4. SDS-PAGE analysis of reduced human fibrin after digestion with piscivorase I and II. (A) Lane 1, Fibrin control (no piscivorase I) after 0 hr; lanes 2-6, fibrin and protease (piscivorase I) after 0.5, I, 3, 5 and 24 hr, respectively; lane 7, a mixture of myosin (200,000), p-galactosidase (116,250), phosphorylase B (97,400), bovine serum albumin, ovalbumin, glyceraldehyde-3-phosphate dehydrogenase, carbonic anhydrase, trypsinogen, soybean trypsin inhibitor and u-lactalbumin. (B) Lane 1, Fibrin control (no piscivorase II) after 0 hr; lanes 2-6, fibrin and protease (piscivorase II) after 0.5, 1, 3, 5 and 24 hr, respectively; lane 7, mol. wt markers.

Fibrinolytic

Enzymes

from

A. p. pisciuorus

935

Fibrinogenolytic and jibrinolytic activities

The activities of PI and PI1 on fibrinogen and fibrin clot were determined by SDS-PAGE (Figs 4 and 5). PI readily cleaved the Acr- and B/3-chains of fibrinogen, but PI1 cleaved the Aa-chain readily and the BP-chain more slowly. Neither PI nor PI1 digested or had any effect on the y-chain after 24 hr. Hydrolysis of the fibrin clot of PI and PII resulted in the degradation of same M- and j-chains as in fibrinogen hydrolysis, while the y-chain

A

Fig. 5. SDS-PAGE analysis of reduced human fibrinogen after digestion with piscivorase I and II. (A) Lanes 1 and 2, Fibrinogen control (no piscivorase I) after 0 hr and 24 hr; lanes 3-7, fibrinogen and protease (piscivorase I) after 0.5, 1, 3, 5 and 24 hr, respectively; lane 8, a mixture of bovine serum albumin, ovalbumin, glyceraldehyde-3-phosphate dehydrogenase, carbonic anhydrase, trypsinogen, soybean trypsin inhibitor and a-lactalbumin. (B) Lanes 1 and 2, Fibrinogen control (no piscivorase II) after 0 hr and 24 hr; lanes 3-7, fibrinogen and protease (piscivorase II) after 0.5, 1, 3, 5 and 24 hr, respectively; lane 8, mol. wt markers.

936

B.-S. HAHN et al. 120 100

20

0 7

5

3

11

9 PH

Fig. 6. Effect of pH on piscivorase I and II. Piscivorase I and II (6 pg each) were incubated at room temperature for 2 hr at various pH values. The azocaseinolytic activity was determined as described in Materials and Methods: 0, piscivorase I; 0, piscivorase II. Values represent the means of triplicate tests, which produced similar results.

appeared unaffected (Figs 4 and 5). The fibrinogenolytic PI1 were 0.20, 0.32 and 0.058 U/mg, respectively. Eflect

activities of crude venom, PI and

ofpH

PI is stable in the pH range of 7-10, whereas PI1 is stable between pH 6 and 10. Two fibrinolytic enzymes at pH below 4 and above 10 resulted in an abrupt decrease in proteolytic activity. PI had about 14% proteolytic activity at pH 3 and PI1 was completely inactivated at pH 4 and 11 (Fig. 6).

20 0

i 0

20

60

80

Tanl;:rat”re(%)

Fig. 7. Effect of temperature on crude venom, and piscivorase I and II. The crude venom, piscivorase I and II (6ng each) were incubated at various temperatures for 15 min. The azocaseinolytic activity was determined as described in Materials and Methods: 0, crude venom; 0, piscivorase I; 0, piscivorase II. Values represent means of triplicate tests, which produced similar results.

931

Fibrinolytic Enzymes from A. p. pisciuorus Table 2. Effects of some inhibitors on the azocaseinolytic activity of two proteases (piscivorase I and II)

Inhibitor None Benzamidine EDTA EGTA Cysteine PMSF 2-Mercaptoethanol Soybean trypsin inhibitor Aprotinin

Concentration (mM)

Relative activity W) PI PI1 100

10 5 5 5 2 10

100 8 5 31 108 30

116 3 I 8 116 9

50* 50*

110 108

120 113

Piscivorase I (PI) and II (PII), 6 pg each in IO mM Tris-HCl (pH 8.3) were incubated at 37°C for I hr with several inhibitors (2-10 ~1) at 100 ~1 of final volume in 10 mM Tris-HCl (pH 8.3). Azocaseinolytic activity was determined as described in Materials and Methods. Values represent % of control and are means of triplicate experiments, which produced similar results. */I’M.

Eflect of temperature

PI and PI1 were stable in the temperature range of 10--6O’C, gradually losing proteolytic activity above 60°C. Azocaseinolytic activities of both fibrinolytic enzymes remained at 13% (PI) and 7% (PII) at 80°C (Fig. 7).

Inhibition

studies

The azocaseinolytic activities of PI and PI1 were inhibited by EDTA, EGTA, cysteine and /I-mercaptoethanol but not by benzamidine, PMSF and soybean trypsin inhibitor. Thus these studies suggest that both fibrinolytic enzymes should be metalloproteases (Table 2).

Table 3. Effects of divalent cations on the azocaseinolytic activity of two proteases (piscivorase I and II)

Divalent cation None CaCl, MgCl, BaCI, cuso, ZnCI, MnCl,

Concentration (mM)

Relative activity (%) PI PI1 100

5 5 5 5 5 5

134 127 133 15 40 7

158 151 157 58 1 21

Piscivorase I (PI) and II (PII), 6 pg each in 10 mM Tris-HCI (pH 8.3) were preincubated with some divalent cations at 37°C for 2 hr. Azocaseinolytic activity was determined as described in Materials and Methods. These values represent % of control and are means of triplicate experiments, which produced similar results.

938

B.-S. HAHN ef al.

Eflect of metal ions

The azocaseinolysis of PI and PI1 was activated by Ca’+, Mg2+, Ba2+ but inhibited by Zn2+, Cu2+ and Mn2+ (Table 3). Platelet aggregation

The effect of crude venom, PI and PI1 on platelet aggregation was compared to the aggregation response induced by ADP (Fig. 8). Aggregation response of the crude venom to PRP was about half of that induced by ADP. The addition of PI and PII to PRP did not induce platelet aggregation. The addition of ADP after a 2.5 min incubation of PRP with PI or PI1 resulted in platelet aggregation approximately 95% (PI) and 91% (PII) of that obtained by ADP (Fig. 8). Haemorrhagic

activity

The haemorrhagic activity of the crude venom and PI was determined as described above, and the activity of PI1 is currently being studied. PI did not show haemorrhage at a low dose (11 pg) but haemorrhage occurred at 55 pg per mouse. The MHD of crude venom and PI were 2.0 pg and 23 ,ug, respectively.

Tlme(mml

2.5

0

2.5

0

2.5

0

2.5

0

Fig. 8. Effect of crude venom, and piscivorase I and II on platelet aggregation. (a) Platelet aggregation induced by 50~1 of crude venom (3Opg). (b) Platelet aggregation in the presence of 504 of crude venom (SOpg) with 20 ~1 of 0.2 mM ADP added after 2.5 min. (c) Platelet aggregation in the presence of 50 ~1 of piscivorase I (30 fig) with 20 ~1 of 0.2 mM ADP added after 2.5 min. (d) Platelet aggregation in the presence of 50 ~1 of piscivorase II (30 pg) with 20 ~1 of 0.2 mM ADP added after 2.5 min. (e) Induced platelet aggregation using 20 ~1 of 0.2 mM ADP, without protease, was considered to be 100% aggregation.

Fibrinolytic Enzymes from A. p. pisciuorus

939

Table 4. Amidolytic activities of two proteases from A. D. Discivorus

Chromogenic substrates

Amidolytic activity (A,,lmin) Crude venom PII

PI

T-1637* T-6140? B-2133t

0.016 0.000 0.005

0.000 0.000 0.013

0.442 0.011 0.441

All data are means of triplicate experiments. Thrombin (0.135 NIH U) and plasmin (0.01 U) were used in the test, and had values are 0.35 and 0.295 (A,,/min), respectively. PI and PII: Piscivorase I and II. *N-p- tosyl-Gly-Pro- Arg-p-nitroanilide acetate (for thrombin). TN-p-tosyl-Gly-Pro-Lys-p-nitroanilide acetate (for plasmin). IN-benzoyl-Pro-Phe-Arg-p-nitroanilide hydrochloride (for kallikrein).

Activity on chromogenic substrates Crude venom, PI and PI1 were tested on chromogenic substrates, B-2133, T-1637 and T-6140. The crude venom had activity on T-1637 and B-2133 but PI and PI1 had negligible activity on these substrates (Table 4).

Table 5. Amino acid composition of piscivorase I and II Amino acid

PI*

PIIT

Accfibt

Apcfib§

Qa

2 22 28 27 21 6 8 13 13 10 6 12 2 8 17 9 3 7

11 35 23 18 18 8 10 15 14 23 2 14 5 16 22 7 17 6

6 31 20 13 14 IO 9 13 11 6 5 12 7 12 22 7 8 ND

3 35 21 14 14 9 8 14 12 6 6 14 6 I1 22 7 9 ND

Asx Glx Ser Gly His Arg Thr Ala Pro Tyr Val Met Ile Leu Phe LYS

Tru

Cya is the sum of cysteic acid and oxidized cystein. Asx and Glx are the sum of asparagine and aspartic acid, and glutamine and glutamic acid, respectively. *Piscivorase I. tPiscivorase II. $From A. contortrix contortrix (Retzios and Markland, 1990). §From A. pirciuorus conanti (Retzios and Markland, 1990). ND, Not determined.

940

B.-S. HAHN

et al.

Amino acid analysis

The amino acid compositions were calculated on the basis of the mol. wts of 23,400 (PI) and 29,000 (PII). Table 5 shows a comparison with other fibrinolytic enzymes. DISCUSSION

The fibrinolytic activities of A. piscivoruspiscivorus venom have been reported previously (Bajwa et al., 1982; Nikai et al., 1988). We purified two fibrinolytic enzymes from this venom, with mol. wts of 23,400 and 29,000 and isoelectric points of 6.6 and 8.5. Piscivorase I of the two fibrinolytic enzymes easily digests the Aa- and BP-chains of fibrinogen and fibrin clots but piscivorase II preferentially digests the Aa-chain, and digests the BP-chain more slowly. Therefore, piscivorase I and II belong to the class of a, B-type fibrinogenase as named by Huang et al. (1992), in agreement with the results reported previously (Ahmed et al., 1990; Siigur and Siigur, 1991). Proteolysis of piscivorase I and II was activated by Ca*+, Mg*+ and Ba2+ but inhibited by Zn2+ and Ca’+. These properties are close to those of the LHFII isolated from L. muta muta (Sanchez et al., 1991). Both fibrinolytic enzymes are inhibited by cysteine and P-mercaptoethanol and metal chelating agents such as EDTA and EGTA, but not by serine protease inhibitors such as PMSF, benzamidine and soybean trypsin inhibitor. These results suggest that both fibrinolytic enzymes belong to the metalloproteases and disulfide bonds may be important for the fibrinolytic activities of these enzymes. Piscivorase I induced haemorrhage at a dose of 55 pg but not at 11 pg, and its MHD is 23 pg, whereas HT-b and HT-e purified from C. atrox (Bjarnason and Tu, 1978) have MHD of 3 and 1 pg. Our results are similar to LHFII (Sanchez et al., 1991). Therefore piscivorase I is a low haemorrhagic metalloprotease. Piscivorase I and II are heat-labile proteins. The proteolytic activities of these proteins were nearly destroyed after incubation at 80°C for 15 min. Both proteases were stable in the temperature range of 1060°C and in the pH range of 7-10. The proteolytic activity of PI is decreased at pH below 7 and above 10, whereas high proteolytic activity of PI1 remains at pH 6 and is completely inactivated at pH 11. Both proteases contain relatively high amounts of Asx and Glx. These results are close to fibrinolytic proteases reported previously (Willis and Tu, 1988; Retzios and Markland, 1990). Neither protease showed serine protease-like activity using chromogenic substrates. Thrombin-like activity, which was often observed in Viperidae species, was shown in the purified fractions. The effect on fibrinogen indicated that PI is a haemorrhagic protease but does not have anticoagulant properties, as the clotting time was not increased. PI1 hydrolysed the Aa-chain readily, but the BP-chain only slowly. This is true for the nonhaemorrhagic protease. It is not clear whether PI1 is nonhaemorrhagic. Acknowledgement-This

study was supported

by a Genetic

Engineering

Grant

from the Ministry

of Education.

REFERENCES Ahmed, N. K., Tennant, K. D., Markland, F. S. and Lacz, J. P. (1990) Biochemical characteristics of fibrolase, a fibrinolytic protease from snake venom. Hemostusis 20, 147. Astrup, T. and Mullertz, S. (1952) The fibrin plate method for estimating of fibrinolytic activity. A&s Biochem. Biophys. 40, 346. Bajwa, S. S., Kirakossian, H., Reddy, K. N. N. and Markland, F. S. (1982) Thrombin-like and fibrinolytic enzymes in the venoms from the gaboon viper (Bifis gubonica), eastern cottonmouth moccasin (Agkisrrodon p. piscivorus) and southern copperhead (Agkistrodon c. contortrix) snakes. Toxicon 20, 427.

Fibrinolytic Enzymes from A. p. piscivorus

941

Beynon, R. J. and Kay, J. (1978) The inactivation of native enzymes by a neutral proteinase from rat intestinal muscle. Biochem. J. 173, 29 I. Bjamason, J. B. and Fox, J. W. (1983) Proteolytic specificity and cobalt exchange of hemorrhagic toxin e, a zinc protease isolated from the venom of the western diamondback rattlesnake (Crotalus atrox). Biochemistry 22, 3770.

Bjarnason, J. B. and Tu, A. T. (1978) Hemorrhagic toxins from western diamondback rattlesnake (Crofalus utrox) venom: isolation and characterization of five toxins and the role of zinc in hemorrhagic toxin e. Biochemistry

17, 3395.

Guan, A. L., Retzios, A. D., Henderson, G. N. and Markland, F. S. (1991) Purification and characterization of a fibrinolytic enzyme from venom of the southern copperhead snake (Agkistrodon contortrix contortrix). Archs Biochem. Biophys. 289, 197.

Herzig, R. H., Ratnoff, 0. D. and Shainoff, J. R. (1970) Studies on a procoagulant fraction of southern copperhead snake venom: the preferential release of fibrinopeptide B. J. Lab. clin. Med. 76, 45 I. Huang, T. F., Chang, M. C., Peng, H. C. and Teng, C. M. (1992) A novel a-type fibrinogenase from Agkistrodon rhodostoma snake venom. Biochim. biophys. Actn 1160, 262. Maeno, H. and Mitsuhashi, S. (f960) Studies on habu snake venom 2C. Studies on H-proteinase of habu venom. Jpn. J. Microbial.

4, 173.

Nikai, T., Katano, E., Komori, Y. and Sugihara, H. (1988) b-Fibrinogenase p. piscivorus.

from the venom of Agkisrrodon

Comp. Biochem. Physiol. 89B, 509.

Nolan, C., Hall, L. S. and Barlow, G. H. (1976) Ancrod, the coagulating enzyme from Malayan pit viper (Agkistrodon rhodostoma) venom. In: Methods in Enzymology, Vol. 45, p. 205 (Lorand, L., Ed.). New York: Academic Press. Ouyang, C. and Huang, T. F. (1976) Purification and characterization of the fibrinolytic principle of Agkistrodon acutus venom. Biochim. biophys. Acta 439, 146. Ouyang, C., Hwang, L. J. and Huang, T. F. (1983) a-Fibrinogenase from Agkistrodon rhodostoma (Malayan pit viper) snake venom. Toxicon 21, 25. Retzios, A. D. and Markland, F. S. (1990) HPLC-based two-step purification of fibrinolytic enzymes from the venom of Agkistrodon contortrix contortrix and Agkistrodon piscivorus conanti. Prot. Express. Purif: 1, 33. Sanchez, E. F., Magalhaes, A., Mandelabaum, F. R. and Diniz, C. R. (1991) Purification and characterization of the hemorrhagic factor II from the venom of the bushmaster snake (Lachesis muta mum). Biochim. biophys. Acta 1074, 347.

Siigur, E. and Siigur, J. (1991) Purification and characterization lebetina (snake) venom.

of lebetase, a fibrinolytic enzyme from Vipera

Biochim. biophys. Acta 1074, 223.

Stocker, K. and Barlow, G. H. (1976) The coagulant enzyme from Bothrops atrox venom (batroxobin). In: Methods in Enzymology, Vol. 45, p. 214 (Lorand, L., Ed,). New York: Academic Press. Willis, T. W. and Tu, A. T. (1988) Purification and biochemical characterization of atroxase, a nonhemorrhagic fibrinolytic protease from western diamondback rattlesnake venom. Biochemistry 27, 4769. Xiaohong, Q* and Liren, M. (1991) Fibrinolytic enzyme from Agkistrodon hatys breuicaudus (Korean mamushi) snake venom. Toxicon 29, 1381. Yarleque, A., Campos, S., Escobar, E., Lazo, F., Sanchez, N., Hyslop, S., Marsh, N. A., Butterworth, P. J. and Price, R. G. (1989) Isolation and characterization of a fibrinogen-clotting enzyme from venom of the snake, Lachesis muta muta (Peruvian bushmaster). Toxicon 27, 1189.