Purification and partial characterization of a coagulant serine protease from the venom of the Iranian snake Agkistrodon halys

ORIGINAL PAPER

Purification and partial characterization of a coagulant serine protease from the venom of the Iranian snake Agkistrodon halys

Ghorbanpur M^I; Zare Mirakabadi A^II; Zokaee F^I; Zolfagarrian H^II; Rabiei H^II

^IChemical Engineering Department, Amirkabir University, Tehran, Iran

^IIDepartment of Venomous Animals and Antivenom Production, Razi Vaccine and Serum Research Institute, Karaj, Iran

^{Correspondence to}

ABSTRACT

Agkistrodon halys is one of several dangerous snake species in Iran. Among the most important signs and symptoms in patients envenomated by this snake is disseminated intravascular coagulation. A thrombin-like enzyme, called AH143, was isolated from Agkistrodon halys venom by gel filtration on a Sephadex G-50 column, ion-exchange chromatography on a DEAE-Sepharose and high performance liquid chromatography (HPLC) on a C18 column. In the final stage of purification, 0.82 mg of purified enzyme was obtained from 182.5 mg of venom. The purified enzyme showed a single protein band by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), under reducing conditions, and its molecular mass was found to be about 30 kDa. AH143 revealed clotting activity in human plasma, which was not inhibited by EDTA or heparin. This enzyme still demonstrated coagulation activity when exposed to variations in temperature and pH ranging, respectively, from 30 to 40°C and from 7.0 to 8.0. It also displayed proteolytic activities on synthetic substrate. The purified enzyme did not show any effect on casein. We concluded that the venom of the Iranian snake Agkistrodon halys contains about 0.45% single procoagulant protein which appears to be a thrombin-like enzyme.

Key words: Iranian snake, venom, Agkistrodon halys, serine protease, chromatography, coagulant activity.

INTRODUCTION

Snake venoms contain numerous enzymes such as glucosidases and proteases, with the latter group including metalloproteases and serine proteases. Several venom serine proteases present important fibrino(geno)lytic, kininogenase and thrombin-like activities (1, 2).

Thrombin-like enzymes (TLE) from snake venoms belong to a class of serine proteases that can result in blood clotting in vitro, a feature exhibited by numerous snake toxins. These enzymes render the blood uncoagulable when acting in vivo, apparently by depleting the circulating fibrinogen (3, 4). Most of thrombin-like enzymes cleave fibrinogen to release fibrinopeptide A and cannot activate factor XIII (5, 6). Therefore, thrombin-like enzymes hydrolyze fibrinogen to produce non-cross-linked fibrins that are more susceptible to the lytic action of plasmin than thrombin-induced cross-linked clots. These clots formed by TLE action are removed from the circulation either by fibrinolysis or by the reticuloendothelial system (5, 6).

Several venoms from the families Crotalidae and Viperidae contain proteolytic enzymes that exercise some effect on the blood coagulation process (7, 8). Since fibrinogen-clotting enzyme was first discovered in snake venom in the early years of the 20^th century, numerous TLE have been isolated from different snake species (5). These enzymes are widely distributed among several pit viper genera (Agkistrodon, Bothrops, Crotalus, Lachesis and Trimeresurus), as well as in true vipers (Bitis and Cerastes) and colubrids (Dispholidus typus) (9). Some TLE from snake venoms with known complete amino acid sequences are: ancrod (Calloselasma rhodostoma, formerly Agkistrodon rhodostoma) batroxobin (Bothrops atrox moojeni), bilineobin (A. binlineatus) and crotalase (C. adamanteus) (10-14).

Thrombin-like enzymes are medically promising and act as defibrinogenating agents, such as ancrod and batroxobin, that have been extensively used in victims of thrombosis, myocardial infarction, peripheral vascular disease, acute ischemia and renal transplant rejection (15). Additionally, these enzymes are employed in deep corneal ulcer repair and non-pregnant canine uterus (16, 17). A study on the efficacy of fibrin glue derived from snake venom in colon anastomosis of rats was carried out by Leite et al. (14).

To the best of our knowledge, no prior work has been published on purification of this enzyme from A. halys venom; hence, in this study, we attempted to it and report its partial characterization.

MATERIALS AND METHODS

Material

Fresh crude venom of A. halys was obtained directly from local Iranian snakes, lyophilized and stored at -20°C. DEAE-Sepharose, Sephadex G-50 and C18 columns were purchased from Pharmacia (Sweden). Bovine serum albumin BApNA (N2-benzoyl-DL-arginine-p-nitroanilide), standard protein marker kits and other reagents for enzymatic and biochemical assays were purchased from Sigma. All chemicals were of analytical reagent grade.

Isolation of Thrombin-Like Enzyme

Lyophilized crude venom of A. halys (200 mg) was dissolved in 8 mL of 50 mM ammonium acetate buffer (pH 7.4) and centrifuged at 5,000 rpm for 15 minutes at 4°C. Subsequently, the supernatant was filtered on a 0.45 microfilter to remove insoluble materials.

Clear supernatant was submitted to molecular exclusion on a Sephadex G-50 chromatographic column (150 x 3 cm), previously equilibrated with ammonium acetate buffer (pH 7.4) and, then, eluted with the same buffer. Fractions of 9 mL/tube were collected at a flow rate of 60 mL/hour at 4°C. Obtained fractions were denominated consecutively AH1 to AH5 to indicate A. halys fractions 1 to 5.

Fractions revealed clotting activity during the gel chromatography step. AH1 was pooled and dialyzed overnight at 4°C against distilled water and applied on DEAE-Sepharose­ column (2.5 x 20 cm), equilibrated with 20 mM Tris buffer, at pH 8.2. In this step, proteins were eluted with a linear gradient of NaCl from 0.0 to 0.5 mM. Flow rate was 17 mL/hour and 5-mL fractions were collected at 4°C.

The fraction that had shown clotting activity in the previous step was pooled and dialyzed overnight at 4°C and applied on an HPLC column, C18 (H₂O, 0.1% trifluoroacetic acid), and eluted with a concentration gradient of solvent B (acetonitrile, 0.1% trifluoroacetic acid) from 0 to 30%, at a flow rate of 0.5 mL/minute during 30 minutes. The peaks were monitored at 280 nm.

Blood Collection

Normal pooled plasma comprised samples from ten healthy donors, without history of bleeding or thrombosis. Blood was centrifuged for 20 minutes at 2,000 rpm, and the plasma was freshly used for test.

Determination of Molecular Mass

SDS-PAGE (12%) was performed according to Laemmli (18), utilizing a standard of low molecular masses ranging from 6.5 to 66 kDa.

Purity Analysis

Analysis of purity was carried out by HPLC on a C18 column as previously described.

Prothrombin Time (PT) Assay

Prothrombin time reagent (200 µL) and sample aliquots (200 µL) were preincubated for ten minutes at 37°C and mixed. Then, 100 µL of plasma was added and clotting time was recorded (19).

Amidolytic Activity

Sample proteolytic activity was assayed using the method of Laemmli (18). The substrate solution was prepared by dissolving BApNA in 5 mL of dimethyl methyl sulfoxide (DMSO) and adding 95 mL of 0.05 M Tris-HCl buffer (pH 8.2). Proteolytic activity was monitored through the expression of BApNA hydrolysis by mixing 50 µL of sample with 100 µL of the substrate solution. After ten minutes of incubation at 37°C, the absorbance was measured at 410 nm. One unit of protease activity corresponds to an increase of A410 = 0.001/min (18).

Coagulant Activity

Plasma (200 µL) and sample aliquots with different amounts of enzyme (2.5, 5.0, 10.0, 20.0 or 50.0 µg) were briefly preincubated at 37°C. After that, they were mixed and shaken and then clotting times were recorded (15, 20, 21).

Proteolytic Activity on Casein

Proteolytic activity was determined by the Sant'Ana method (22). Various amounts (10, 20, 30, 40, 50 or 60 mg) of the enzyme were added to the buffer solution 0.1M Tris-HCl, pH 9.0, and the final volume was adjusted to 250 µL, followed by 750 µL of 1% (m/v) casein and incubated for 15 minutes at 37°C. The reaction was stopped by addition of 1.5 mL of 30% TCA. Resulting proteolysis products in the supernatant solution were evaluated spectrophotometrically after centrifugation at 1,600 g for 10 minutes at 280 nm. One unit of caseinolytic activity corresponds to an increase of A280 = 0.001/min (2).

Inhibition of Enzyme Activity

The effect of ethylenediaminetetraacetic acid (EDTA) and heparin were examined by incubation with enzymes in 180 µL of 20 mM Tris-HCI, with optimum pH at 37°C for five minutes. After mixing each of the inhibitors and 10 mL of purified enzyme, the remaining activity was determined by measuring the hydrolysis of synthetic BApNA. The remaining coagulating activity was calculated as percent activity in relation to the control preparation incubated without inhibitors, which was considered 100% (18).

Effect of pH and Temperature

The effect of pH on enzyme activity was evaluated by measuring residual enzyme activity after incubation at different pH, ranging from 4.0 to 6.0 with 0.1 M citrate-NaOH buffer while between 7.0 and 10 with 0.1 M Tris-HCl buffer, at 37°C for ten minutes.

Additionally, enzyme activity was assayed at diverse temperatures ranging from 20 to 70°C using BApNA as a substrate. The assay was conducted at pH 7.5 with 0.05 M Tris-HCl buffer containing 0.02 M CaCl2 for five minutes in a temperature-controlled water bath. Thereafter, heat-treated samples were rapidly cooled in an ice bath and residual activity was measured using BApNA as substrate at pH 7.5 and 25°C for five minutes, as previously described (18).

Protein Determination

Protein concentration was measured according to Lowry et al. (23) using BSA as a standard.

RESULTS

Isolation of the Thrombin-Like Enzyme

In the Sephadex G-50 fractionation of the crude venom, five peaks with absorbance at 280 nm were obtained (AH1 to AH5) as shown in Figure 1. All fractions were tested for coagulation, it was found that AH1 fraction was positive for procoagulation. PT assay revealed that AH1 fraction had a 10 µmol/min/mg activity. The yield of procoagulant enzyme fraction (AH1) was calculated and found to be 36% (Table 1).

Table 1.
Purification of a thrombin-like enzyme (AH143) from A. halys venom

Step

Protein

(mg)

Specific activity*

( µmol/min/mg)

Total amidase activity

(µmol/min)

Yield

%

Venom

182.5

10

1801.9

100

AH1

48.12

32.39

1558.6

91

AH14

1.8

622.2

1120

65

AH143

0.82

469.7

328.8

36

* Activity was determined using BApNA as substrate.

Further purification was carried out by ion exchange chromatography on DEAE-Sepharose. In this step, five fractions were obtained (AH11 to AH15), out of which AH14 fraction revealed procoagulant activity (Figure 2). The total yield of this fraction was 56% that accounted for 0.98% of total venom. AH14 fraction was pooled, dialyzed and applied to a C18 reversed-phase HPLC column. Four peaks (AH141 to AH144) were obtained (Figure 3) and AH143 presented coagulant proteolytic activity. The PT-test reached its elution peaks at 9 and 12 minutes (Figure 3). The purification procedure of AH143 is summarized in Table1. Through this technique, about 0.82 mg of purified enzyme was obtained from 182.5 mg of venom.

Purity and Determination of Molecular Mass

The homogeneity of purified enzymes was confirmed by SDS-PAGE and HPLC as shown in Figures 4 and 5, respectively. Isolated AH143 revealed higher purity, analyzed by C18 reversed-phase HPLC (Figure 4). The chromatographic profile of AH143 at 280 nm did not show the contaminating peptide. Furthermore, this enzyme showed a single band by SDS-PAGE (Figure 6) whereas its molecular mass was estimated to be about 30 kDa under reducing conditions.

Caseinolytic, Clotting and Amidolytic Activity Assays

AH143 showed no proteolytic activity on casein while coagulant activity assay demonstrated that it presented serine protease activity on BAPNA. AH143 also showed clotting activity on human plasma (Figure 6) and at a concentration of 15 µg/mL it was able to coagulate plasma in 20 seconds. However, at lower concentrations (less than 15 µg/mL), this enzyme did not show any coagulant activity. The time of plasma coagulation decreased with increased concentrations of AH143.

Inhibition of Enzyme Activity

The amidolytic activity of AH143 was not affected by metal chelator (EDTA) or heparin.

Effect of pH and Temperature on Activity

The optimum temperature for AH143 catalysis was evaluated in terms of amidolytic activity on BApNA. AH143 activity after incubation at different pH, ranging from 7 to 8, was high (Figure 7A) and was stable at temperatures ranging from 20 to 40°C (Figure 7B). AH143 maximum activity was found to be at 7.5. At pH 6.0, AH143 activity was about 50% of the maximum; at pH 9.0, it was about 65%; while almost 95% of the activity was observed at pH 8.0.

AH143 displayed progressively higher hydrolysis rates from 20 up to 40°C. The optimum temperature for its activity was 37°C. The enzyme lost almost 60% of its maximum activity at temperatures above 50°C, being completely inactivated at 60°C.

DISCUSSION

Coagulant enzymes that can convert fibrinogen into fibrin gel are widely distributed in Crotalidae snake venoms (8, 22, 24). The selected venom for this study was from the same family group.

This article reports an efficient and relatively simple procedure for the isolation of a highly purified thrombin-like serine protease from the venom of Iranian Agkistrodon halys, called AH143. This enzyme was isolated and purified by a combination of gel filtration on Sephadex G-50 (Figure 1), ion-exchange chromatography on DEAE-Sepharose (Figure 2) and HPLC on C18 column. However, the total yield of the enzyme accounted only for 36% (Table 1). This may be due to the partial loss of the biological activity of the enzyme. On the other hand, HPLC was able to isolate a protein with same type of activity, but with different molecular mass (results unpublished) which could be the reason for reduced specific activity in the purification final step.

In conventional methods, the thrombin-like protein represent around 1% of the venom (2), but through the method herein described, it accounted for approximately 0.83% of the venom protein, out of which AH143 was found to be 0.45% (Table 1). The amount of TLE in snake venoms from diverse species or in the same species from various places can be different (6, 7). Analyses of AH143 activity in various conditions of temperature and pH were carried out in accordance with Nikai et al. (13). The temperature under which the purification was performed was based on the observation that the enzyme solution was stable at 4°C (4, 7). Several reports indicate that the optimum pH for TLE activity is in the range of 7.0 to 8.5, being practically inactive at 5 (4).

The molecular mass of AH143 is similar to other thrombin-like enzymes already studied (6, 13). In addition, molecular mass and enzymatic activities - including plasma clotting, amidolytic activity and not inhibition by heparin - characterize AH143 as a thrombin-like enzyme isolated from Agkistrodon halys venom (Figures 5, 6 and 7) (4, 15, 20, 21, 25, 26).

Since AH143 is not inhibited by heparin, it can be used in quantitative determination of fibrinogen as well as in plasma of patients under heparin treatment and in the laboratory for routine assays of coagulation factors (4).

In conclusion, the venom of Iranian snake (Agkistrodon halys) contain about 0.45% of single procoagulant protein with molecular mass of about 30 kDa which it seems to be a thrombin-like enzyme.

Received: September 4, 2008

Accepted: February 13, 2009

Abstract published online: March 5, 2009

Full paper published online: August 31, 2009

CONFLICTS OF INTEREST: There is no conflict.

• 1. Kini RM. Anticoagulant proteins from snake venoms: structure, function and mechanism. Biochem J. 2006;397(3):377-87.
• 2. Ponce-Soto LA, Bonfim Novello VL, Navarro Oviedo R, Yarleque Chocas A, Marangoni S. Isolation and characterization of a serine protease, Ba III-4, from Peruvian Bothrops atrox venom. Protein J. 2007;26(6):387-94
• 3. Bell WR. Clinical trials with ancrod. In: Haemostasis and animal venoms. New York: Marcel Dekker; 1988. p. 544-51.
• 4. Magalhăes A. Purification and properties of a coagulant thrombin-like enzyme from the venom of Bothrops leucurus Comp Biochem Physiol. 2007;146(4):565-75.
• 5. Choa SY, Hahna BS, Yangb KY, Kima YS. Purification and characterization of calobin II, a second type of thrombin-like enzyme from Agkistrodon caliginosus (Korean viper). Toxicon. 2001;39(4):499-506.
• 6. Furukawa K, Ishimara S. Use of thrombin-like snake venom enzymes in the treatment of vascular occlusive diseases. In: Medical use of snake venom proteins. Boca Raton: CRC Press; 1990. p. 161-99.
• 7. Etsuko O, Hidenobu T. Purification and characterization of a thrombin like enzyme, elegaxobin II, with lys-bradykinin releasing activity from the venom of Trimeresurus elegans (Sakishima-Habu). Toxicon. 2003;41(5):559-68.
• 8. Etsuko O, Hidenobu T. Purification and characterization of a thrombin-like enzyme, elegaxobin, from the venom of Trimeresurus elegans (Sakishima-habu). Toxicon. 2000;38(8):1087-100.
• 9. Kini RM. Serine proteases affecting blood coagulation and fibrinolysis from snake venoms. Haemost Thromb. 2005;34(4-5):200-4.
• 10. Burkhart W, Smith GFH, Su JL, Parikh I, Le Vine H. Amino acid sequence determination of ancrod, the thrombin-like fibrinogenase from the venom of Agkistrodon rhodostoma FEBS Lett. 1992;297(3):297-301.
• 11. Itoh N, Tanaka N, Mihashi S, Yamashina I. Molecular cloning and sequence analysis of cDNA for batroxobin, a thrombin-like snake venom enzyme. J Biol Chem. 1987;262(7):3132-5.
• 12. Au LC, Liu SB, Chou JS, Teh GW, Chang KJ, Shih CM. Molecular cloning and sequence analysis of the cDNA for ancrod, a thrombin-like enzyme from the venom of Calloselasma rhodostoma Biochem J. 1993;294(Pt2):387-90.
• 13. Nikai T, Ohara A, Komori Y, Fox JW, Sugihara H. Primary structure of a coagulant enzyme, bilineobin, from Agkistrodon bilineatus venom. Arch Biochem Biophys. 1995; 318(1):89-96.
• 14. Leite CVS, Naresse E, Arantes HL, Lopes AF, Thomazini IA, Giannini MJS, Mercadante MC, et al. An evaluation by rat colon anastomosis of the efficacy of fibrin glue derived from snake venom. J Venom Anim Toxins. 2000;6(2):180-93.
• 15. Castro HC, Zingali RB, Albuquerque MG, Pujol-Luza M, Rodrigues CR. Review -snake venom thrombin-like enzymes: from reptilase to now. CMLS Cell Mol Life Sci. 2004;61:843-56.
• 16. Moraes JRE, Correia PHA, Camplesi AC, Moraes FR. Experimental use of fibrin glue derived from snake venom in non-pregnant canine uterus. J Venom Anim Toxins incl Trop Dis. 2004;10(2):133-43.
• 17. Sampaio RL, Ranzani JJT, Brandăo CVS, Thomazini-Santos IA, Barraviera B, Barraviera SRCS, et al. Use of fibrin glue derived from snake venom in the repair of deep corneal ulcers - experimental study in dogs. J Venom Anim Toxins incl Trop Dis. 2007;13(4):857-73.
• 18. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage, T4. Nature.1970;227(5259):680-5.
• 19. Athukorala Y, Jung WK, Vasanthan T, Jeon YJ. An anticoagulative polysaccharide from an enzymatic hydrolysate of Ecklonia cava carbohydrate polymers. 2006;66(2):184-91.
• 20. Yu X, Li Z, Xia X, Fang H, Zhou C, Chen H. Expression and purification of ancrod, an anticoagulant drug, in Pichia pastoris Protein Expr Purif. 2007;55(2):257-61.
• 21. Silva-Junior FP, Guedes HLM, Garvey LC, Bourguignon C, Di Cera Enrico, Giovanni-De-Simone Salvatore. BJ-48, a novel thrombin-like enzyme from the Bothrops jararacussu venom with high selectivity for Arg over Lys in P1: Role of N-glycosylation in thermostability and active site accessibility. Toxicon. 2007;50(1):18-31.
• 22. Sant'Ana CD, Ticli FK, Oliveira LL. BjussuSP-I: A new thrombin-like enzyme isolated from Bothrops jararacussu snake venom. Comp Biochem Physiol. 2008;151(3):443-54.
• 23. Lowry, OH, NJ Rosebrough, AL Farr, and RJ Randall. Protein measurement with the folin phenol reagent. J. Biol. Chem. 1951;193(1):265-75.
• 24. Takashi S, Takashi H. Purification of two thrombin-like enzyme from the venom of Agkistrodon caliginosus (Kankoku mamushi). Toxicon. 1984;22(1):29-38.
• 25. Koh YS, Chung KH, Kim DS. Biochemical characterization of a thrombin-like enzyme and a fibrinolytic serine protease from snake (Agkistrodon saxatilis) venom, Toxicon.  2001;39(4):555-60.
• 26. Matsui T, Fujimura Y, Titani K. Snake venom proteases affecting homeostasis and thrombosis. Biochim Biophys Acta. 2000;1477(1-2):146-56.

Publication Dates

• Publication in this collection
  14 Sept 2009
• Date of issue
  2009

History

• Accepted
  13 Feb 2009
• Received
  04 Sept 2008