- Email: [email protected]
Biochemical characterization of a thrombin-like enzyme and a fibrinolytic serine protease from snake (Agkistrodon saxatilis) venom
Toxicon 39 (2001) 555±560
www.elsevier.com/locate/toxicon
Biochemical characterization of a thrombin-like enzyme and a ®brinolytic serine protease from snake (Agkistrodon saxatilis ) venom You-Seok Koh a, Kwang-Hoe Chung b, Doo-Sik Kim a,* a
Department of Biochemistry, College of Science and Bioproducts Research Center, Yonsei University, Seoul 120-749, South Korea b Cardiovascular Research Institute, Yonsei University College of Medicine, Seoul 120-752, South Korea Received 22 February 2000; accepted 3 May 2000
Abstract A thrombin-like enzyme and a ®brinolytic serine protease were puri®ed to homogeneity from the venom of a Korean snake Agkistrodon saxatilis emelianov. Both the puri®ed enzymes migrated as a single protein band corresponding to 39 kDa in SDS-PAGE. However, the molecular mass was reduced to 28 kDa by enzymatic removal of the N-linked carbohydrates in those two dierent enzyme species. Although the thrombin-like enzyme and the ®brinolytic protease show homologous features in their molecular sizes and N-terminal amino acid sequences, yet they can be clearly distinguished from each other in terms of substrate speci®city, susceptibility to inhibitors and ®brinogen degradation. It is postulated that these two enzymes are capable of functioning in a cooperative manner to eectively remove ®brinogen and consequently to reduce the blood viscosity. 7 2000 Elsevier Science Ltd. All rights reserved. Keywords: Snake venom; Fibrinogen clotting; Fibrinolytic enzyme; Thrombin-like enzyme
1. Introduction It is known that the snake venom contains a family of serine proteases, which act upon the dierent stages of blood coagulation (Stocker, 1990). Thrombin-like enzymes (Esnouf and Tunnah, 1967; Markland and Damus, 1971; Pan et al., 1999) and ®brino(geno)lytic enzymes (Matsui et al., 1998; Siigur and Siigur, 1991) can act on ®brinogen, leading to de®brinogenation of blood and a consequent decrease in blood viscosity.
* Corresponding author. Tel.: +82-2-361-2700; fax: +82-2362-9897. E-mail address: [email protected] (D.-S. Kim).
The ®brin monomers produced by thrombin-like enzymes are unstable and more susceptible to plasmin proteolysis than thrombin-induced ®brin, and are therefore promptly degraded. Consequently, ®brin degradation products increase during the initial phase of de®brinogenation and gradually fall to normal levels after ®brinogen depletion (Chang and Huang, 1995). Fibrino(geno)lytic enzymes also lead to ®brinogen depletion by the direct degradation of ®brin polymer (Siigur et al., 1998) or by the generation of ®brinogenolytic products that are no longer converted to the normal ®brin clots by thrombin (Matsui et al., 1998). It is also known that these enzymes do not aect the plasma coagulation factors, nor do they induce platelet aggregation and the release reaction
0041-0101/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 1 - 0 1 0 1 ( 0 0 ) 0 0 1 6 9 - 0
556
Y.-S. Koh et al. / Toxicon 39 (2001) 555±560
(Stocker, 1990). In this work, we puri®ed a novel thrombin-like and a direct-acting ®brinolytic serine proteases from the venom of Agkistrodon saxatilis emelianov. Molecular properties of the puri®ed enzymes were characterized and their physiological function related to the ®brinogen clotting cascade were investigated.
was determined using an applied biosystems precise protein sequencing system. 2.5. Thrombin clotting assay
2. Materials and methods
Fibrinogen clotting time was measured as described by Hofmann et al. (1983). The clotting time of 0.5 ml human ®brinogen (0.5 mg/ml) in 50 mM Tris±HCl (pH 7.5) was measured after the addition of 0.1 ml of puri®ed enzymes incubated at 378C for 5 min.
2.1. Materials
2.6. Fibrinogenolytic activity assay
Fresh crude venom of A. saxatilis emelianov was obtained directly from a local snake farm in Korea. QSepharose, Superdex 75 and Mono Q columns were purchased from Pharmacia (Uppsala, Sweden). Fibrinogen was obtained from Korea Green Cross (Seoul, Korea). Thrombin, factor Xa, soybean trypsin inhibitor and synthetic substrates were purchased from Sigma (St. Louis, MO).
The ®brinogenolytic activity was determined by incubating 0.1 ml ®brinogen (0.2%) in 50 mM Tris± HCl (pH 7.5) at 378C with 0.1 ml of sample to be tested. Aliquots were taken at 10, 30 and 60 min intervals and separated by SDS-PAGE to examine the cleavage pattern of the ®brinogen.
2.2. Puri®cation of thrombin-like enzyme and ®brinolytic enzyme
Fibrinolytic activity was assayed using the ®brin plate technique of Astrup and Mullertz (1952). The ®brin plate assay was performed with 5 ml ®brinogen (0.5%) clotted with three units of thrombin or the same units of the puri®ed thrombin-like enzyme. Protease samples (10 ml) were placed on the ®brin surface and incubated at 378C for 6 h. Fibrinolytic activity was assessed by examining the reduction of the absorbance of the lysed zone at 600 nm.
A. saxatilis venom (1 ml) was diluted 10-fold with 20 mM Tris±HCl (pH 8.0), loaded onto a Q-Sepharose FPLC column equilibrated with 20 mM Tris±HCl (pH 8.0) and step-eluted with buers containing 50 and 100 mM NaCl. Fractions containing thrombin-like and ®brinolytic activity were collected separately and concentrated. Each of the two concentrated fractions separately applied to a Superdex 75 FPLC gel ®ltration column equilibrated with 20 mM Tris±HCl (pH 8.0) containing 150 mM NaCl. Fractions with thrombinlike activity were subjected to a mono Q FPLC column equilibrated with 20 mM Tris±HCl (pH 8.0) and eluted with a linear gradient of 0±0.2 M NaCl. Fractions showing the ®brinolytic activity were recovered with a linear gradient of 0.1±0.3 M. Homogeneity and molecular weight of the puri®ed proteins were determined by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis. 2.3. Removal of carbohydrate Puri®ed enzymes (1 mg/ml in 20 mM Tris±HCl, pH 8.0) were incubated with N-glycosidase F (2 units/mg protein) at 378C for 24 h and the deglycosylated proteins were subjected to 12% SDS-PAGE. 2.4. Amino terminal amino acid sequence determination The puri®ed enzymes were subjected to 12% SDSPAGE and electrotransferred to a PVDF membrane. Amino terminal amino acid sequence of the protein
2.7. Fibrin plate assay
2.8. Substrate speci®city After mixing 50 ml of chromogenic substrate (3 mM), 50 ml of puri®ed enzymes (1 mM) and 0.9 ml of 50 mM Tris±HCl (pH 7.5), and incubating at 378C for 20 min, the digestion of the substrate was determined by measuring increased absorbance at 405 nm. The chromogenic substrates examined were: N-BenzoylPhe-Val-Arg-p-nitroanilide (B-7632), pyro-Glu-PheLeu-p-nitroanilide (P-3169), D-Val-Leu-Lys-p-nitroanilide (V-7127), N-p-tosyl-Gly-Pro-Lys-p-nitroanilide (T6140), pyro-Glu-Gly-Arg-p-nitroanilide (B-2291), Nbenzoyl-Pro-Phe-Arg-p-nitroanilide (B-2133), N-Methoxy-succinyl-Ala-Ala-Pro-Val-p-nitroanilide (M-4765), N-Benzoyl-Ile-Glu-Gly-Arg-p-nitroanilide (B-2291) and D-Phe-Pip-Arg-p-nitroanilide (S-2238). 2.9. Inhibition of enzyme activity The eect of pepstatin, leupeptin, phenylmethyl-sulfonyl ¯uoride (PMSF), ethylene diamine tetraacetic acid (EDTA), soybean trypsin inhibitor and hirudin were examined by incubation with enzyme in 20 mM Tris±HCI (pH 7.5) at 378C for 5 min. After mixing
Y.-S. Koh et al. / Toxicon 39 (2001) 555±560
557
Table 1 Puri®cation of thrombin-like enzyme and ®brinolytic enzyme Thrombin-like enzyme
Crude venom Q-Sepharose Superdex 75 Mono-Q
Fibrinolytic enzyme
Total protein (mg)
Total activity (NIH unit)a
Speci®c activity (NIH unit/mg)
Total protein (mg)
Total activity (CU)b
Speci®c activity (CU/mg)
190 18.2 3.7 0.62
2660 1750 772 217
14 96.2 257.3 350
190 26 5.1 0.4
532 320 128 27
2.8 12.3 25.1 67.5
a
Fibrinogen clotting activity was expressed in NIH equivalent units of human thrombin. Zone clearance of a standard plasmin at dierent dilutions was used to prepare a standard curve and ®brinolytic activity was expressed in caseinolytic units (CU). b
each of the inhibitors and 10 ml of puri®ed enzyme (10 pmole), the remaining activity was determined by measuring the hydrolysis of synthetic substrate N-benzoyl-Phe-Pip-Arg-p-nitroanilide.
3. Results and discussion 3.1. Puri®cation of enzymes Crude venom was fractionated by anion-exchange chromatography on a Q-Sepharose column. Thrombinlike activity was found in a 0.05 M eluting fraction, and ®brinolytic activity was recovered from 0.1 M eluent. Each fraction was further puri®ed on a column of Superdex 75 and anion-exchange chromatography in a mono-Q column (Table 1). Novel thrombin-like and ®brinolytic enzymes were puri®ed to homogeneity, which were determined by 12% SDS-PAGE analysis. The puri®ed thrombin-like and ®brinolytic enzymes appeared as similar sized bands, that corresponded to 39 kDa under reducing conditions (Fig. 1). Removal of the N-linked carbohydrates from the two enzymes by incubation with N-glycosidase F resulted in products with molecular masses of 28 kDa. The N-terminal amino acid sequences of both the enzymes were compared with those of snake venom serine proteases (Fig. 2). Puri®ed enzymes showed signi®cant homology to ancrod (Au et al., 1993), batroxobin (Itoh et al., 1988) and ¯avoxobin (Shieh et al., 1988). 3.2. Inhibition of enzyme activity When both the enzymes were preincubated with hirudin, EDTA, pepstatin or trypsin inhibitor, they did not reveal any loss of proteolytic activity (Fig. 3). Since hirudin inhibits thrombin by binding to its ®brinogen recognition exosite (Jandrot-Perrus et al., 1991), the hirudin-insensitivity of both the enzymes suggests that they have a ®brinogen recognition site, which dif-
fers from that of thrombin. The proteolytic activity of the two enzymes was inhibited by PMSF in a dose dependent manner, but the thrombin-like enzyme was more strongly inhibited, at any concentration of PMSF, than the ®brinolytic enzyme. However, leupeptin showed an inhibitory eect only with the thrombinlike enzyme. These results indicated that the two enzymes are serine proteases, but have dierent inhibitor susceptibilities not only to thrombin but also to each other. 3.3. Substrate speci®city The puri®ed enzymes demonstrated similar substrate speci®city for chromogenic substrates in spite of their
Fig. 1. SDS-PAGE analysis of puri®ed enzymes. Puri®ed enzymes were subjected to electrophoresis in a 12% sodium dodecylsulfate polyacrylamide gel under reducing condition. Thrombin-like enzyme (lane 1) and ®brinolytic enzyme (lane 3) were deglycosylated (lane 2, lane 4) by peptide-N-glycosidase-F as described in Section 2. Molecular weight markers were phosphorylase B (97,400 Da), serum albumin (66,000 Da), ovalbumin (45,000 Da), carbonic anhydrase (31,000 Da) and trypsin inhibitor (21,500 Da).
558
Y.-S. Koh et al. / Toxicon 39 (2001) 555±560
Fig. 2. Comparison of the N-terminal amino acid sequences of the puri®ed enzymes with snake venom serine proteases. Determined N-terminal sequences of the thrombin-like enzyme and the ®brinolytic enzyme are aligned with those of the ancrod from Calloselasma rhodostoma (Au et al., 1993), the batroxobin from Bothrops atrox (Itoh et al., 1988) and the ¯avoxobin from Trimeresurus ¯avoviridis (Shieh et al., 1988).
opposite physiological actions on ®brinogen clotting (Fig. 4). Thrombin-speci®c substrate, D-Phe-Pip-Arg-pnitroanilide, and plasma kallikrein substrate, D-ProPhe-Arg-p-nitroanilide, were most susceptible to hydrolysis by both enzymes. However, the puri®ed ®brinolytic enzyme degraded the plasminogen activator substrate very rapidly, D-Val-Leu-Lys-p-nitroanilide than the thrombin-like enzyme. Thrombin-like enzyme exhibited low Km value (34±64 mM) compared to the ®brinolytic enzyme (196±324 mM) in hydrolyzing several synthetic substrates. These results imply that the thrombin-like enzyme has a higher substrate speci®city and binding anity than the ®brinolytic enzyme. Furthermore, the two enzymes exhibited signi®cant dierence in leupeptin inhibition (Fig. 3). It is possible to
Fig. 3. Inhibition of enzyme activity by protease inhibitors. Puri®ed thrombin-like enzyme (Q) and ®brinolytic enzyme (q) were preincubated with equimolar hirudin, soybean trypsin inhibitor, pepstatin or leupeptin. Protease activity was measured using synthetic substrate S-2238.
Fig. 4. Substrate speci®cities of the thrombin-like enzyme and the ®brinolytic enzyme. Catalytic activity of the thrombin-like enzyme (Q), the ®brinolytic enzyme (q), human thrombin (-) and human urokinase (K) was measured using synthetic substrates S-2238, B-2133, V-7127, T-6140 and B-2291.
hypothesize that the two serine proteases have characteristic active site structures. The ®brinogenolytic activity of the two enzymes was analyzed by SDS-PAGE, which demonstrated that both the puri®ed enzymes are capable of degrading ®brinogen A and B chains (Fig. 5A). Thrombin-like enzyme initially degraded the ®brinogen A chain and triggered the clotting of ®brinogen. Upon an extended enzyme reaction, digestion of ®brinogen B chain was then detected (Fig. 5A, lane 4). The release of ®brinopeptide A and B by the thrombin-like enzyme was identi®ed by HPLC C18 column fractionation (data not shown), when the enzyme reaction was allowed for 4 h. On the contrary, the ®brinolytic enzyme rapidly degraded the ®brinogen B chain in 10 min, and sub-
Fig. 5. Degradation of ®brinogen and prothrombin by the puri®ed enzymes. (A) Fibrinogen (lane 1) was incubated with thrombin-like enzyme (lanes 2, 3 and 4) or with ®brinolytic enzyme (lanes 5, 6 and 7) for 10 min (lanes 2 and 5), 30 min (lanes 3 and 6), and 60 min (lanes 4 and 7). (B) 10 pmole human prothrombin (lane 1) was incubated with 0.1 pmole of the thrombin-like enzyme (lane 2), the ®brinolytic enzyme (lane 3) or 0.05 unit factor Xa (lane 4) in 50 mM Tris±HCl (pH 7.5) for 2 h at 378C.
Y.-S. Koh et al. / Toxicon 39 (2001) 555±560
sequently hydrolyzed the A chain after 1 h incubation. Furthermore, the ®brinogenolytic products of the ®brinolytic enzyme revealed dierent cleavage pattern compared with those of the thrombin-like enzyme. These two enzymes also demonstrated proteolytic activity on prothrombin in a dierent manner (Fig. 5B). The proteolytic fragments of prothrombin generated by the ®brinolytic enzyme revealed similar cleavage pattern with that of factor Xa on SDS-PAGE analysis. However, the proteolytic products of prothrombin generated by both enzymes did not show thrombin activity. Therefore, such an abnormal prothrombin degradation may lead to the suppression of the thrombin-induced ®brinogen clotting that interrupts ®brinogen depletion in blood stream. 3.4. Fibrinogen and ®brin plate assay The puri®ed thrombin-like enzyme was shown to have 350 NIH unit/mg of clotting activity as determined by the ®brinogen plate assay. The puri®ed ®brinolytic enzyme produced a lysed zone on both the thrombin-clotted and thrombin-like enzyme-induced ®brin plates. However, the lysed zone that was detected on the thrombin-like enzyme-induced ®brin plate was doubly clear than that on the thrombininduced clot (Fig. 6). Therefore, it is possible to assume that the ®brinolytic enzyme is able to eectively remove blood ®brinogen by degrading the thrombin-like enzyme-induced ®brin clot. These results demonstrate that the thrombin-like enzyme and the ®brinolytic protease may act in a cooperative manner to remove ®brinogen and thereby reduce blood viscosity. In conclusion, we puri®ed and characterized a novel thrombin-like enzyme and a direct-acting ®brinolytic
Fig. 6. Clot lysis activity of the ®brinolytic enzyme. Fibrinolytic activity was measured on a ®brin clot induced by human thrombin (.) or by the thrombin-like enzyme (Q).
559
serine protease. Those two enzymes showed a signi®cant similarity in terms of molecular size and N-terminal amino acid sequence, but were distinguished from each other in view of sensitivity to inhibitors, substrate speci®city and ®brinogen hydrolysis. Although these enzymes appear to have biochemical functions corresponding to human thrombin and plasmin, respectively, both the snake venom-derived enzyme activities are capable of contributing to the blood viscosity by reducing ®brinogen concentration. Experimental data obtained in this work will be useful not only for understanding the functional mechanisms of proteases, but also for developing therapeutic agents related to thrombotic disorders.
References Astrup, T., Mullertz, S., 1952. The ®brin plate method for the estimation of ®brinolytic activity. Arch. Biochem. Biophys. 40, 346±351. Au, L.C., Lin, S.B., Chou, J.S., Teh, G.W., Chang, K.J., Shih, C.M., 1993. Molecular cloning and sequence analysis of the cDNA for ancrod, a thrombin-like enzyme from the venom of Calloselasma rhodostoma. Biochem. J. 294, 387± 390. Chang, M.C., Huang, T.F., 1995. The antiplatelet activity of ancrod on administration to rabbits. J. Lab. Clin. Med. 125, 508±516. Esnouf, M.P., Tunnah, G.W., 1967. The isolation and properties of the thrombin-like activity from Ancistrodon rhodostoma venom. Brit. J. Haematol. 13, 581±590. Hofmann, H., Dumarey, C., Bon, C., 1983. Blood coagulation induced by Bothrops atrox venom: identi®cation and properties of a factor X activator. Biochimie 65, 201±210. Itoh, N., Tanaka, N., Funakoshi, I., Kawasaki, T., Mihashi, S., Yamashina, I., 1988. Organization of the gene for batroxobin, a thrombin-like snake venom enzyme. Homology with the trypsin/kallikrein gene family. J. Biol. Chem. 263, 7628±7631. Jandrot-Perrus, M., Huisse, M.G., Krstenansky, J.L., Bezeaud, A., Guillin, M.C., 1991. Eect of the hirudin carboxy-terminal peptide 54±65 on the interaction of thrombin with platelets. Thromb. Haemost. 66, 300±305. Markland, F.S., Damus, P.S., 1971. Puri®cation and properties of a thrombin-like enzyme from the venom of Crotalus adamanteus (Eastern diamond back rattle-snake). J. Biol. Chem. 246, 6460±6473. Matsui, T., Sakurai, Y., Fujimura, Y., Hayashi, I., Oh-Ishi, S., Suzuki, M., Hamako, J., Yamamoto, Y., Yamazaki, J., Kinoshita, M., Titani, K., 1998. Puri®cation and amino acid sequence of halystase from snake venom of Agkistrodon halys blomhoi, a serine protease that cleaves speci®cally ®brinogen and kininogen. Eur. J. Biochem. 252, 569±575. Pan, H., Du, X., Yang, G., Zhou, Y., Wu, X., 1999. cDNA cloning and expression of acutin. Biochem. Biophys. Res. Commun. 255, 412±415. Shieh, T.C., Kawabata, S.I., Kihara, H., Ohno, M., Iwanaga,
560
Y.-S. Koh et al. / Toxicon 39 (2001) 555±560
S., 1988. Amino acid sequence of a coagulant enzyme, ¯avoxobin, from Trimeresurus ¯avoviridis venom. J. Biochem. 103, 596±605. Siigur, E., Siigur, J., 1991. Puri®cation and characterization of lebetase, a ®brinolytic enzyme from Vipera lebetina (snake) venom. Biochem. Biophys. Acta 1074, 223±229. Siigur, J., Samel, M., Tonismagi, K., Subbi, J., Siigur, E., Tu,
A.T., 1998. Biochemical characterization of lebetase, a direct-acting ®brinolytic enzyme from Vipera lebetina snake venom. Thromb. Res. 90, 39±49. Stocker, K.F., 1990. Snake venom proteins aecting hemostasis and ®brinolysis. In: Stocker, K.F. (Ed.), Medical Use of Snake Venom Proteins. CRC Press, Boca Raton, pp. 97±160.
www.elsevier.com/locate/toxicon
Biochemical characterization of a thrombin-like enzyme and a ®brinolytic serine protease from snake (Agkistrodon saxatilis ) venom You-Seok Koh a, Kwang-Hoe Chung b, Doo-Sik Kim a,* a
Department of Biochemistry, College of Science and Bioproducts Research Center, Yonsei University, Seoul 120-749, South Korea b Cardiovascular Research Institute, Yonsei University College of Medicine, Seoul 120-752, South Korea Received 22 February 2000; accepted 3 May 2000
Abstract A thrombin-like enzyme and a ®brinolytic serine protease were puri®ed to homogeneity from the venom of a Korean snake Agkistrodon saxatilis emelianov. Both the puri®ed enzymes migrated as a single protein band corresponding to 39 kDa in SDS-PAGE. However, the molecular mass was reduced to 28 kDa by enzymatic removal of the N-linked carbohydrates in those two dierent enzyme species. Although the thrombin-like enzyme and the ®brinolytic protease show homologous features in their molecular sizes and N-terminal amino acid sequences, yet they can be clearly distinguished from each other in terms of substrate speci®city, susceptibility to inhibitors and ®brinogen degradation. It is postulated that these two enzymes are capable of functioning in a cooperative manner to eectively remove ®brinogen and consequently to reduce the blood viscosity. 7 2000 Elsevier Science Ltd. All rights reserved. Keywords: Snake venom; Fibrinogen clotting; Fibrinolytic enzyme; Thrombin-like enzyme
1. Introduction It is known that the snake venom contains a family of serine proteases, which act upon the dierent stages of blood coagulation (Stocker, 1990). Thrombin-like enzymes (Esnouf and Tunnah, 1967; Markland and Damus, 1971; Pan et al., 1999) and ®brino(geno)lytic enzymes (Matsui et al., 1998; Siigur and Siigur, 1991) can act on ®brinogen, leading to de®brinogenation of blood and a consequent decrease in blood viscosity.
* Corresponding author. Tel.: +82-2-361-2700; fax: +82-2362-9897. E-mail address: [email protected] (D.-S. Kim).
The ®brin monomers produced by thrombin-like enzymes are unstable and more susceptible to plasmin proteolysis than thrombin-induced ®brin, and are therefore promptly degraded. Consequently, ®brin degradation products increase during the initial phase of de®brinogenation and gradually fall to normal levels after ®brinogen depletion (Chang and Huang, 1995). Fibrino(geno)lytic enzymes also lead to ®brinogen depletion by the direct degradation of ®brin polymer (Siigur et al., 1998) or by the generation of ®brinogenolytic products that are no longer converted to the normal ®brin clots by thrombin (Matsui et al., 1998). It is also known that these enzymes do not aect the plasma coagulation factors, nor do they induce platelet aggregation and the release reaction
0041-0101/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 1 - 0 1 0 1 ( 0 0 ) 0 0 1 6 9 - 0
556
Y.-S. Koh et al. / Toxicon 39 (2001) 555±560
(Stocker, 1990). In this work, we puri®ed a novel thrombin-like and a direct-acting ®brinolytic serine proteases from the venom of Agkistrodon saxatilis emelianov. Molecular properties of the puri®ed enzymes were characterized and their physiological function related to the ®brinogen clotting cascade were investigated.
was determined using an applied biosystems precise protein sequencing system. 2.5. Thrombin clotting assay
2. Materials and methods
Fibrinogen clotting time was measured as described by Hofmann et al. (1983). The clotting time of 0.5 ml human ®brinogen (0.5 mg/ml) in 50 mM Tris±HCl (pH 7.5) was measured after the addition of 0.1 ml of puri®ed enzymes incubated at 378C for 5 min.
2.1. Materials
2.6. Fibrinogenolytic activity assay
Fresh crude venom of A. saxatilis emelianov was obtained directly from a local snake farm in Korea. QSepharose, Superdex 75 and Mono Q columns were purchased from Pharmacia (Uppsala, Sweden). Fibrinogen was obtained from Korea Green Cross (Seoul, Korea). Thrombin, factor Xa, soybean trypsin inhibitor and synthetic substrates were purchased from Sigma (St. Louis, MO).
The ®brinogenolytic activity was determined by incubating 0.1 ml ®brinogen (0.2%) in 50 mM Tris± HCl (pH 7.5) at 378C with 0.1 ml of sample to be tested. Aliquots were taken at 10, 30 and 60 min intervals and separated by SDS-PAGE to examine the cleavage pattern of the ®brinogen.
2.2. Puri®cation of thrombin-like enzyme and ®brinolytic enzyme
Fibrinolytic activity was assayed using the ®brin plate technique of Astrup and Mullertz (1952). The ®brin plate assay was performed with 5 ml ®brinogen (0.5%) clotted with three units of thrombin or the same units of the puri®ed thrombin-like enzyme. Protease samples (10 ml) were placed on the ®brin surface and incubated at 378C for 6 h. Fibrinolytic activity was assessed by examining the reduction of the absorbance of the lysed zone at 600 nm.
A. saxatilis venom (1 ml) was diluted 10-fold with 20 mM Tris±HCl (pH 8.0), loaded onto a Q-Sepharose FPLC column equilibrated with 20 mM Tris±HCl (pH 8.0) and step-eluted with buers containing 50 and 100 mM NaCl. Fractions containing thrombin-like and ®brinolytic activity were collected separately and concentrated. Each of the two concentrated fractions separately applied to a Superdex 75 FPLC gel ®ltration column equilibrated with 20 mM Tris±HCl (pH 8.0) containing 150 mM NaCl. Fractions with thrombinlike activity were subjected to a mono Q FPLC column equilibrated with 20 mM Tris±HCl (pH 8.0) and eluted with a linear gradient of 0±0.2 M NaCl. Fractions showing the ®brinolytic activity were recovered with a linear gradient of 0.1±0.3 M. Homogeneity and molecular weight of the puri®ed proteins were determined by 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis. 2.3. Removal of carbohydrate Puri®ed enzymes (1 mg/ml in 20 mM Tris±HCl, pH 8.0) were incubated with N-glycosidase F (2 units/mg protein) at 378C for 24 h and the deglycosylated proteins were subjected to 12% SDS-PAGE. 2.4. Amino terminal amino acid sequence determination The puri®ed enzymes were subjected to 12% SDSPAGE and electrotransferred to a PVDF membrane. Amino terminal amino acid sequence of the protein
2.7. Fibrin plate assay
2.8. Substrate speci®city After mixing 50 ml of chromogenic substrate (3 mM), 50 ml of puri®ed enzymes (1 mM) and 0.9 ml of 50 mM Tris±HCl (pH 7.5), and incubating at 378C for 20 min, the digestion of the substrate was determined by measuring increased absorbance at 405 nm. The chromogenic substrates examined were: N-BenzoylPhe-Val-Arg-p-nitroanilide (B-7632), pyro-Glu-PheLeu-p-nitroanilide (P-3169), D-Val-Leu-Lys-p-nitroanilide (V-7127), N-p-tosyl-Gly-Pro-Lys-p-nitroanilide (T6140), pyro-Glu-Gly-Arg-p-nitroanilide (B-2291), Nbenzoyl-Pro-Phe-Arg-p-nitroanilide (B-2133), N-Methoxy-succinyl-Ala-Ala-Pro-Val-p-nitroanilide (M-4765), N-Benzoyl-Ile-Glu-Gly-Arg-p-nitroanilide (B-2291) and D-Phe-Pip-Arg-p-nitroanilide (S-2238). 2.9. Inhibition of enzyme activity The eect of pepstatin, leupeptin, phenylmethyl-sulfonyl ¯uoride (PMSF), ethylene diamine tetraacetic acid (EDTA), soybean trypsin inhibitor and hirudin were examined by incubation with enzyme in 20 mM Tris±HCI (pH 7.5) at 378C for 5 min. After mixing
Y.-S. Koh et al. / Toxicon 39 (2001) 555±560
557
Table 1 Puri®cation of thrombin-like enzyme and ®brinolytic enzyme Thrombin-like enzyme
Crude venom Q-Sepharose Superdex 75 Mono-Q
Fibrinolytic enzyme
Total protein (mg)
Total activity (NIH unit)a
Speci®c activity (NIH unit/mg)
Total protein (mg)
Total activity (CU)b
Speci®c activity (CU/mg)
190 18.2 3.7 0.62
2660 1750 772 217
14 96.2 257.3 350
190 26 5.1 0.4
532 320 128 27
2.8 12.3 25.1 67.5
a
Fibrinogen clotting activity was expressed in NIH equivalent units of human thrombin. Zone clearance of a standard plasmin at dierent dilutions was used to prepare a standard curve and ®brinolytic activity was expressed in caseinolytic units (CU). b
each of the inhibitors and 10 ml of puri®ed enzyme (10 pmole), the remaining activity was determined by measuring the hydrolysis of synthetic substrate N-benzoyl-Phe-Pip-Arg-p-nitroanilide.
3. Results and discussion 3.1. Puri®cation of enzymes Crude venom was fractionated by anion-exchange chromatography on a Q-Sepharose column. Thrombinlike activity was found in a 0.05 M eluting fraction, and ®brinolytic activity was recovered from 0.1 M eluent. Each fraction was further puri®ed on a column of Superdex 75 and anion-exchange chromatography in a mono-Q column (Table 1). Novel thrombin-like and ®brinolytic enzymes were puri®ed to homogeneity, which were determined by 12% SDS-PAGE analysis. The puri®ed thrombin-like and ®brinolytic enzymes appeared as similar sized bands, that corresponded to 39 kDa under reducing conditions (Fig. 1). Removal of the N-linked carbohydrates from the two enzymes by incubation with N-glycosidase F resulted in products with molecular masses of 28 kDa. The N-terminal amino acid sequences of both the enzymes were compared with those of snake venom serine proteases (Fig. 2). Puri®ed enzymes showed signi®cant homology to ancrod (Au et al., 1993), batroxobin (Itoh et al., 1988) and ¯avoxobin (Shieh et al., 1988). 3.2. Inhibition of enzyme activity When both the enzymes were preincubated with hirudin, EDTA, pepstatin or trypsin inhibitor, they did not reveal any loss of proteolytic activity (Fig. 3). Since hirudin inhibits thrombin by binding to its ®brinogen recognition exosite (Jandrot-Perrus et al., 1991), the hirudin-insensitivity of both the enzymes suggests that they have a ®brinogen recognition site, which dif-
fers from that of thrombin. The proteolytic activity of the two enzymes was inhibited by PMSF in a dose dependent manner, but the thrombin-like enzyme was more strongly inhibited, at any concentration of PMSF, than the ®brinolytic enzyme. However, leupeptin showed an inhibitory eect only with the thrombinlike enzyme. These results indicated that the two enzymes are serine proteases, but have dierent inhibitor susceptibilities not only to thrombin but also to each other. 3.3. Substrate speci®city The puri®ed enzymes demonstrated similar substrate speci®city for chromogenic substrates in spite of their
Fig. 1. SDS-PAGE analysis of puri®ed enzymes. Puri®ed enzymes were subjected to electrophoresis in a 12% sodium dodecylsulfate polyacrylamide gel under reducing condition. Thrombin-like enzyme (lane 1) and ®brinolytic enzyme (lane 3) were deglycosylated (lane 2, lane 4) by peptide-N-glycosidase-F as described in Section 2. Molecular weight markers were phosphorylase B (97,400 Da), serum albumin (66,000 Da), ovalbumin (45,000 Da), carbonic anhydrase (31,000 Da) and trypsin inhibitor (21,500 Da).
558
Y.-S. Koh et al. / Toxicon 39 (2001) 555±560
Fig. 2. Comparison of the N-terminal amino acid sequences of the puri®ed enzymes with snake venom serine proteases. Determined N-terminal sequences of the thrombin-like enzyme and the ®brinolytic enzyme are aligned with those of the ancrod from Calloselasma rhodostoma (Au et al., 1993), the batroxobin from Bothrops atrox (Itoh et al., 1988) and the ¯avoxobin from Trimeresurus ¯avoviridis (Shieh et al., 1988).
opposite physiological actions on ®brinogen clotting (Fig. 4). Thrombin-speci®c substrate, D-Phe-Pip-Arg-pnitroanilide, and plasma kallikrein substrate, D-ProPhe-Arg-p-nitroanilide, were most susceptible to hydrolysis by both enzymes. However, the puri®ed ®brinolytic enzyme degraded the plasminogen activator substrate very rapidly, D-Val-Leu-Lys-p-nitroanilide than the thrombin-like enzyme. Thrombin-like enzyme exhibited low Km value (34±64 mM) compared to the ®brinolytic enzyme (196±324 mM) in hydrolyzing several synthetic substrates. These results imply that the thrombin-like enzyme has a higher substrate speci®city and binding anity than the ®brinolytic enzyme. Furthermore, the two enzymes exhibited signi®cant dierence in leupeptin inhibition (Fig. 3). It is possible to
Fig. 3. Inhibition of enzyme activity by protease inhibitors. Puri®ed thrombin-like enzyme (Q) and ®brinolytic enzyme (q) were preincubated with equimolar hirudin, soybean trypsin inhibitor, pepstatin or leupeptin. Protease activity was measured using synthetic substrate S-2238.
Fig. 4. Substrate speci®cities of the thrombin-like enzyme and the ®brinolytic enzyme. Catalytic activity of the thrombin-like enzyme (Q), the ®brinolytic enzyme (q), human thrombin (-) and human urokinase (K) was measured using synthetic substrates S-2238, B-2133, V-7127, T-6140 and B-2291.
hypothesize that the two serine proteases have characteristic active site structures. The ®brinogenolytic activity of the two enzymes was analyzed by SDS-PAGE, which demonstrated that both the puri®ed enzymes are capable of degrading ®brinogen A and B chains (Fig. 5A). Thrombin-like enzyme initially degraded the ®brinogen A chain and triggered the clotting of ®brinogen. Upon an extended enzyme reaction, digestion of ®brinogen B chain was then detected (Fig. 5A, lane 4). The release of ®brinopeptide A and B by the thrombin-like enzyme was identi®ed by HPLC C18 column fractionation (data not shown), when the enzyme reaction was allowed for 4 h. On the contrary, the ®brinolytic enzyme rapidly degraded the ®brinogen B chain in 10 min, and sub-
Fig. 5. Degradation of ®brinogen and prothrombin by the puri®ed enzymes. (A) Fibrinogen (lane 1) was incubated with thrombin-like enzyme (lanes 2, 3 and 4) or with ®brinolytic enzyme (lanes 5, 6 and 7) for 10 min (lanes 2 and 5), 30 min (lanes 3 and 6), and 60 min (lanes 4 and 7). (B) 10 pmole human prothrombin (lane 1) was incubated with 0.1 pmole of the thrombin-like enzyme (lane 2), the ®brinolytic enzyme (lane 3) or 0.05 unit factor Xa (lane 4) in 50 mM Tris±HCl (pH 7.5) for 2 h at 378C.
Y.-S. Koh et al. / Toxicon 39 (2001) 555±560
sequently hydrolyzed the A chain after 1 h incubation. Furthermore, the ®brinogenolytic products of the ®brinolytic enzyme revealed dierent cleavage pattern compared with those of the thrombin-like enzyme. These two enzymes also demonstrated proteolytic activity on prothrombin in a dierent manner (Fig. 5B). The proteolytic fragments of prothrombin generated by the ®brinolytic enzyme revealed similar cleavage pattern with that of factor Xa on SDS-PAGE analysis. However, the proteolytic products of prothrombin generated by both enzymes did not show thrombin activity. Therefore, such an abnormal prothrombin degradation may lead to the suppression of the thrombin-induced ®brinogen clotting that interrupts ®brinogen depletion in blood stream. 3.4. Fibrinogen and ®brin plate assay The puri®ed thrombin-like enzyme was shown to have 350 NIH unit/mg of clotting activity as determined by the ®brinogen plate assay. The puri®ed ®brinolytic enzyme produced a lysed zone on both the thrombin-clotted and thrombin-like enzyme-induced ®brin plates. However, the lysed zone that was detected on the thrombin-like enzyme-induced ®brin plate was doubly clear than that on the thrombininduced clot (Fig. 6). Therefore, it is possible to assume that the ®brinolytic enzyme is able to eectively remove blood ®brinogen by degrading the thrombin-like enzyme-induced ®brin clot. These results demonstrate that the thrombin-like enzyme and the ®brinolytic protease may act in a cooperative manner to remove ®brinogen and thereby reduce blood viscosity. In conclusion, we puri®ed and characterized a novel thrombin-like enzyme and a direct-acting ®brinolytic
Fig. 6. Clot lysis activity of the ®brinolytic enzyme. Fibrinolytic activity was measured on a ®brin clot induced by human thrombin (.) or by the thrombin-like enzyme (Q).
559
serine protease. Those two enzymes showed a signi®cant similarity in terms of molecular size and N-terminal amino acid sequence, but were distinguished from each other in view of sensitivity to inhibitors, substrate speci®city and ®brinogen hydrolysis. Although these enzymes appear to have biochemical functions corresponding to human thrombin and plasmin, respectively, both the snake venom-derived enzyme activities are capable of contributing to the blood viscosity by reducing ®brinogen concentration. Experimental data obtained in this work will be useful not only for understanding the functional mechanisms of proteases, but also for developing therapeutic agents related to thrombotic disorders.
References Astrup, T., Mullertz, S., 1952. The ®brin plate method for the estimation of ®brinolytic activity. Arch. Biochem. Biophys. 40, 346±351. Au, L.C., Lin, S.B., Chou, J.S., Teh, G.W., Chang, K.J., Shih, C.M., 1993. Molecular cloning and sequence analysis of the cDNA for ancrod, a thrombin-like enzyme from the venom of Calloselasma rhodostoma. Biochem. J. 294, 387± 390. Chang, M.C., Huang, T.F., 1995. The antiplatelet activity of ancrod on administration to rabbits. J. Lab. Clin. Med. 125, 508±516. Esnouf, M.P., Tunnah, G.W., 1967. The isolation and properties of the thrombin-like activity from Ancistrodon rhodostoma venom. Brit. J. Haematol. 13, 581±590. Hofmann, H., Dumarey, C., Bon, C., 1983. Blood coagulation induced by Bothrops atrox venom: identi®cation and properties of a factor X activator. Biochimie 65, 201±210. Itoh, N., Tanaka, N., Funakoshi, I., Kawasaki, T., Mihashi, S., Yamashina, I., 1988. Organization of the gene for batroxobin, a thrombin-like snake venom enzyme. Homology with the trypsin/kallikrein gene family. J. Biol. Chem. 263, 7628±7631. Jandrot-Perrus, M., Huisse, M.G., Krstenansky, J.L., Bezeaud, A., Guillin, M.C., 1991. Eect of the hirudin carboxy-terminal peptide 54±65 on the interaction of thrombin with platelets. Thromb. Haemost. 66, 300±305. Markland, F.S., Damus, P.S., 1971. Puri®cation and properties of a thrombin-like enzyme from the venom of Crotalus adamanteus (Eastern diamond back rattle-snake). J. Biol. Chem. 246, 6460±6473. Matsui, T., Sakurai, Y., Fujimura, Y., Hayashi, I., Oh-Ishi, S., Suzuki, M., Hamako, J., Yamamoto, Y., Yamazaki, J., Kinoshita, M., Titani, K., 1998. Puri®cation and amino acid sequence of halystase from snake venom of Agkistrodon halys blomhoi, a serine protease that cleaves speci®cally ®brinogen and kininogen. Eur. J. Biochem. 252, 569±575. Pan, H., Du, X., Yang, G., Zhou, Y., Wu, X., 1999. cDNA cloning and expression of acutin. Biochem. Biophys. Res. Commun. 255, 412±415. Shieh, T.C., Kawabata, S.I., Kihara, H., Ohno, M., Iwanaga,
560
Y.-S. Koh et al. / Toxicon 39 (2001) 555±560
S., 1988. Amino acid sequence of a coagulant enzyme, ¯avoxobin, from Trimeresurus ¯avoviridis venom. J. Biochem. 103, 596±605. Siigur, E., Siigur, J., 1991. Puri®cation and characterization of lebetase, a ®brinolytic enzyme from Vipera lebetina (snake) venom. Biochem. Biophys. Acta 1074, 223±229. Siigur, J., Samel, M., Tonismagi, K., Subbi, J., Siigur, E., Tu,
A.T., 1998. Biochemical characterization of lebetase, a direct-acting ®brinolytic enzyme from Vipera lebetina snake venom. Thromb. Res. 90, 39±49. Stocker, K.F., 1990. Snake venom proteins aecting hemostasis and ®brinolysis. In: Stocker, K.F. (Ed.), Medical Use of Snake Venom Proteins. CRC Press, Boca Raton, pp. 97±160.