Anda di halaman 1dari 8

Comparative Biochemistry and Physiology, Part B 146 (2007) 299 306 www.elsevier.

com/locate/cbpb

Amino acid sequence and characterization of C-type lectin purified from the snake venom of Crotalus ruber
Jiharu Hamako a , Yukiyo Suzuki b , Nobuhiro Hayashi c , Mina Kimura c , Yasuhiro Ozeki d , Keiichiro Hashimoto c , Taei Matsui b,
c a Course of Medical Communication, Fujita Health University College Toyoake, Aichi 470-1192, Japan Department of Biology, Fujita Health University School of Health Sciences, Toyoake, Aichi 470-1192, Japan Division of Medical Polymer Science, Institute for Comprehensive Medical Science, Fujita Health University, Toyoake, Aichi 470-1192, Japan d Division of Biochemistry, Environmental Biosciences Course, International Graduate School of Arts and Sciences, Yokohama City University, 22-2, Seto, Kanazawa-ku, Yokohama Kanagawa 236-0027, Japan b

Received 19 April 2006; received in revised form 12 October 2006; accepted 9 November 2006 Available online 5 December 2006

Abstract Galactoside-binding lectin was purified from the snake venom of Crotalus ruber by affinity chromatography on a lactoseagarose column, and the complete amino acid sequence was determined. The C. ruber venom lectin (CRL) showed a single band of 28 kDa by SDS-polyacrylamide electrophoresis under non-reducing conditions, but it showed a single band of 15 kDa under reducing conditions, indicating that CRL is a disulfide-linked homodimer of 15 kDa subunit. CRL specifically recognized beta-galactosides such as thiodigalactoside followed by Nacetylgalactosamine when examined with their inhibitory effects on CRL-induced hemagglutination. A CRL subunit was composed of 135 residues containing nine Cys residues and showed a high similarity to other C-type galactoside-binding lectins from snake venoms. C. atrox lectin (CAL) showed almost the same sequence except for eight amino acid residues. Neither CRL nor CAL induced platelet aggregation by itself or inhibited platelet aggregation mediated by von Willebrand factor or fibrinogen with agonists. CRL showed a similar oligomeric form and the sugar specificity as CAL, but it showed different divalent cation sensitivity such as Mn2+ and Ni2+. Homology modeling suggested that the amino acid substitution found in CRL does not affect sugar recognition of the lectin but might alter the conformation and influence the sugar binding pocket induced by the metalion binding. 2006 Elsevier Inc. All rights reserved.
Keywords: Crotalus atrox; Crotalus ruber; C-type lectin; Divalent cation; Platelet aggregation; Primary structure; Snake venom; von Willebrand factor

1. Introduction Snake venom contains a variety of bioactive substances such as hemorrhagic or coagulant proteases, phospholipases, growth factors, disintegrins and neuronal toxins (Ogawa et al., 1996; Tsetlin, 1999; Matsui et al., 2000; Fox and Serrano, 2005). Several C-type lectins, which require Ca2+ for their carbohydrate binding activity, have been found in snake venom (Gartner and Ogilvie, 1984). Interestingly, a number of C-type lectin-like proteins, which show neither carbohydrate binding activity nor

Corresponding author. Tel.: +81 562 93 2954; fax: +81 562 93 4595. E-mail address: tmatsui@fujita-hu.ac.jp (T. Matsui). 1096-4959/$ - see front matter 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpb.2006.11.022

divalent cation-dependency but show significant structural similarity to C-type lectins, have also been isolated from snake venoms (Lu et al., 2005). The latter proteins are composed of disulfide-linked heterodimer with similar sequences and express highly variant activities toward platelets (GPIb-BP, mamushigin, etc.), hemostatic plasma proteins such as von Willebrand factor (VWF) (botrocetin and bitiscetin) and coagulation factors (FXBP and FIX-BP) (Matsui and Hamako, 2005; Morita, 2005). In contrast, the former C-type lectins are homodimer and show no remarkable effects other than hemagglutination of erythrocytes or platelets (Ogilvie et al., 1989). Although 3D structures of several C-type lectin-like snake venom proteins have been elucidated (Mizuno et al., 1999; Hirotsu et al., 2001; Sen et al., 2001), 3D structural data have

300

J. Hamako et al. / Comparative Biochemistry and Physiology, Part B 146 (2007) 299306

not been obtained for C-type lectins of snake venoms. Recently, Walker et al. (2004) have firstly reported the crystal structure of Crotalus atrox lectin (western diamond-backed rattlesnake) (RSL: rattle snake lectin has been used commonly but we abbreviate it as CAL in this paper), and they have also shown the structural relationship between the Ca2+-binding site and carbohydrate-recognition domain. Interestingly, C-type lectinlike proteins bind to their ligand molecules at the concave region between two subunits (Fukuda et al., 2002; Maita et al., 2003), whereas CAL uses the limb of both subunits for carbohydrate binding. In this paper, we have purified the C-type lectin from C. ruber venom (CRL), and the complete amino acid sequence was determined by conventional amino acid sequencing. We have comparatively studied whether the difference of the amino acid sequence found between CRL and CAL affects the relative molecular masses, sugar specificity, metal ionic sensitivity and platelet agglutinations. 2. Materials and methods 2.1. Materials Crude C. ruber and C. atrox venoms were purchased from Sigma Aldrich Chemicals (St. Louis, MO, USA). Lactose and other reagents used were of the highest biochemical grade available from Wako Pure Chemicals (Osaka, Japan). Human VWF was purified from factor VIII concentrates as described (Chopek et al., 1986). Botrocetin and bitiscetin were prepared as previously described (Fujimura et al., 1991; Hamako et al., 1996). 2.2. Purification of CRL Crude venom powder (50 mg) was dissolved in 1 mL of chilled TBS (150 mM NaCl, 10 mM TrisHCl, pH 7.5) containing 0.1 mg/mL NaN3 and the supernatant obtained after centrifugation (15,000 rpm, at 4 C for 10 min) was passed through a cartridge filter with a 0.45 m pore size (Millipore Co, Bedford, MA, USA). The filtrate was applied to a lactosyl agarose (EYLaboratories, San Mateo, CA, USA) column (4 mL bed) equilibrated with TBS at 4 C. The column was washed with TBS until the absorbance monitor at 280 nm of eluate was returned to the base line, then TBS containing 100 mM lactose and 0.1 mg/mL NaN3 was added to the column to elute CRL. Eluted peak fractions were concentrated by a Ultrafree centrifugal filter device (Millipore) repeatedly with TBS to remove lactose. CAL was purified from crude venom of C. atrox with lactosylagarose column as above. Protein concentration was determined with BCA protein assay reagent (Pierce, Rockford, IL, USA) using bovine serum albumin (BSA) as a standard. 2.3. Gel filtration CRL was dissolved in TBS containing 1 mM EDTA and 100 mM lactose and subjected to gel filtration on a Superose

12 column using FPLC system (GE Healthcare Bio-science Corp., Piscataway, NJ, USA). The column was equilibrated and eluted with TBS containing 1 mM EDTA and 100 mM lactose. The eluting position of CRL was compared with those of standard proteins of known molecular mass such as human IgG (180 kDa), human transferrin (80 kDa), BSA (66 kDa), ovalbumin (45 kDa), bovine -lactoglobin (18 kDa) and bovine cytochrome c (12 kDa) under the same conditions. 2.4. Hemagglutination assay Lectin activity was monitored by the hemagglutination using trypsinized and glutaraldehyde-fixed rabbit erythrocytes as described previously (Ozeki et al., 1997). CRL was serially diluted with TBS on a V-bottom microtiter plate and hemagglutination was performed in the presence of 1% BSA 0.05% Tween 20 and 0.25% erythrocytes at 25 C for 2 h. The titer was expressed as the reciprocal of the highest dilution giving positive hemagglutination. For the specific sugar analysis, CRL and CAL (the titer was previously adjusted to 16) were mixed with serially diluted sugar solution in the presence of 1% BSA, 0.05% Tween 20 and 0.25% erythrocytes. The most effective minimum sugar concentration to inhibit lectin-induced hemagglutination was calculated. 2.5. Platelet aggregation assay Platelet-rich plasma (PRP) was prepared from freshly donated healthy normal blood with informed consent containing 0.38% Na-citrate by gentle centrifugation (800 rpm, at 25 C for 10 min). For preparing the washed platelets, PRP was mixed with a 15% volume of aciddextrose solution (ACD, 75 mM Na-citrate, 38 mM citric acid, 122 mM glucose) and centrifuged at 2500 rpm at 25 C for 10 min. Pelleted platelets were gently washed with HEPES-buffered saline (HBS: 145 mM NaCl, 5 mM KH2PO4, 1 mM MgSO4, 0.5 mM Na H2PO4, 5 mM glucose and 10 mM HEPES-buffer, pH 7.4) and finally suspended in HBS containing 0.2 mM CaCl2 to obtain 3 105 platelets/l. The PRP or washed platelets (250 L, 3 105 platelets/L) were preincubated with each lectin (1073 g/mL) or RCA120 (Seikagaku-kogyo, Tokyo, Japan, 0.4 g/mL) at 37 C for 2 min with gentle stirring, then agonists (with 100 g/mL of human fibrinogen in the case of the washed platelets) were successively added. Type I collagen (Horm-Chemie, Munchen, Germany, 2 or 5 g/mL), thrombin (0.8 U/mL) or ADP (2.5 or 5 M) was added as an inducer for platelet aggregation. For VWF-induced platelet agglutination, botrocetin, bitiscetin (2 g/mL) and ristocetin (1.5 mg/mL) were used as a VWFmodulator using the washed platelets with VWF (10 g/mL) or PRP. Platelet aggregation was monitored by light transmittance change with a dual channel aggregometer (Mebanix, Yokohama, Japan). Platelet aggregation induced by an agonist in the absence of lectin (addition of the same amounts of TBS) was compared as a control.

J. Hamako et al. / Comparative Biochemistry and Physiology, Part B 146 (2007) 299306

301

2.6. SDS-PAGE and blotting Aliquots of CRL and CAL solution were mixed with an equal volume of SDS-buffer (4% SDS, 125 mM TrisHCl, pH 6.8, 20% glycerol) containing 10% 2-mercaptoethanol or not, and heated at 95 C for 3 min. SDS-PAGE was performed as described (Laemmli, 1970) and proteins were stained with Coomassie brilliant blue. Proteins were electro-transferred to a polyvinyliden difluoride (PVDF) membrane immediately after SDS-PAGE using 10 mM CAPS buffer, pH 11.0, in 10% methanol (0.4 V/cm, at 4 C for 90 min). Protein bands on a PVDF membrane were excised after staining with Coomassie brilliant blue, and were subjected to protein sequencer analysis as described (Matsudaira, 1987). 2.7. Cationic ion requirement assay Lectin solutions (10 g/mL, 0.7 mL) were extensively dialyzed against 500 mL of TBS containing 5 mM EDTA for 15 h at 4 C, then against TBS for 6 h twice and 15 h once. The divalent cation-depleted lectin was serially diluted with TBS and incubated with TBS or TBS containing 10 mM of metal ion solutions (CaCl2, MgCl2, NiSO4, MnCl2 and BaCl2) at pH 7.5 for 1 h at 25 C followed by the addition of erythrocytes (0.25%) in the presence of 1% BSA and 0.05% Tween 20. The hemagglutinating activity (titer) was scored after 2 h. 2.8. S-pyridylethylation and enzymatic cleavage CRL (150 g) was reduced and S-pyridylethylated (PE) with tri-n-butylphosphine and 4-vinylpyridine in 0.3 M TrisHCl buffer, pH 8.3, containing 6 M guanidine hydrochloride at 25 C for 5 h as described (Ito et al., 2001). The PECRL was desalted and purified by HPLC on a SynChropak RP-8 column (4.6 250 mm, SynChrom, Lafayette, IN) with a linear gradient of acetonitrile into aqueous 0.1% trifluoroacetic acid. Aliquots (1801000 pmol) of each PECRL subunit were digested with several specific proteases. Digestion with Acromobacter protease I (AP-I), a gift from Dr. T. Masaki, Ibaraki University, Japan, which specifically cleaves lysyl bonds (Masaki et al., 1981), was performed in 50 mM TrisHCl buffer, pH 9.0, containing 2 M urea at 37 C for 6 h (E/S ratio of 1: 150 by weight). Digestion with endoproteinase Asp-N from Pseudomonas fragi (Takara Shuzo, Kyoto, Japan), which specifically cleaves aspartic acid residues at the N-terminal side (Drapeau, 1980), was performed in 50 mM Na-phosphate buffer, pH 8.0, at 37 C for 18 h (E/S ratio of 1:100 by weight). Digestion with Staphylococcus aureus V8-proteinase (Takara Shuzo) which specifically cleaves glutamic acid residues at the C-terminal side was performed in 50 mM ammonium acetate buffer, pH 5.0, containing 2 M urea at 37 C for 18 h (E/S ratio of 1:100 by weight). Digestion with hog stomach pepsin (KochLight, Suffolk, England) was performed in 10% acetic acid at 37 C for 2 h (E/S ratio of 1:100 by weight). Peptides were separated by HPLC on a SynChropak RP C18 column (4.6 250 mm, SynChrom) with gradients of acetonitrile into dilute aqueous trifluoroacetic acid.

2.9. Amino acid sequence determination and mass spectrometric analysis Amino acid sequence determination was carried out with an Applied Biosystems Procise protein sequencing system (model 494 protein sequencer connected to a phenylthiohydantoin analyzer). Mass determination was performed using an ion-spray mass spectrometer (PE-Sciex API-III biomolecular mass analyzer) and a matrix-assisted laser desorption ionization-time of flight mass spectrometer (MALDI-TOF MS) (PerSeptive Biosystems Voyager RPF). Approximately 530 pmol of peptides were dissolved in an appropriate volume of 33% isopropanol containing 0.1% TFA and 50% acetonitrile and injected into the spectrometer for positive-mode analysis in ion-spray mass analysis. For MALDI-TOF MS, peptides (75180 fmol) were mixed with a saturated matrix solution of -cyano-4-hydroxycinnamic acid in 0.05% TFA and 50% acetonitrile, and analyzed in the linear mode with positive ion detection at an accelerating voltage of 20 kV with mass accuracy of less than 0.1% Da. 2.10. Sequence homology search and homology modeling The amino acid sequence of CRL was examined in the protein sequence data base of SwissProt using BLAST and FASTA programs (Altschul et al., 1990) on Genome Net served by Bioinformatics Center, Institute for Chemical Research, Kyoto University. The 3D structure of CRL was predicted by homology modeling based on the 3D structure of CAL (Protein Data Bank entry 1MUQ) using the homology module of Insight II software (Accelrys Inc., CA, USA). 3. Results and discussion 3.1. Purification of CRL A preliminary experiment using a crude extract of C. ruber venom showed a weak hemagglutinating activity towards trypsinized rabbit erythrocytes. Since this hemagglutination was canceled by addition of lactose, it suggested the presence of the galactoside-recognizing lectin in the venom. When the venom extract was subjected to affinity column chromatography on a lactose-conjugated agarose column, break-through fractions had no hemagglutinability, and the column-bound fraction was obtained by elution with lactose-containing buffer (Fig. 1). This lactose-eluted fraction showed a strong hemagglutinating activity after removal of lactose. This fraction showed a single band of about 28 kDa at SDS-PAGE under nonreducing conditions but a single 15 kDa band under reducing conditions (Fig. 2). This purified protein was thus named C. ruber venom lectin (CRL). In a typical experiment, 350 g of CRL was purified from 50 mg of crude venom. CAL was also purified by the same affinity column, and typically, 420 g of CAL was purified from 50 mg of crude venom. CAL showed a major band with a slightly higher molecular mass compared to CRL under both reducing and non-reducing conditions (Fig. 2). CRL was eluted as a single protein peak upon gel filtration on a Superose 12 column

302

J. Hamako et al. / Comparative Biochemistry and Physiology, Part B 146 (2007) 299306 Table 1 Effects of sugars on CRL- and CAL-induced hemagglutination Saccharides Minimum inhibitory concentration (mM) CRL Thiodigalactoside N-acetyl-D-galactosamine Lactulose Lactose -Methyl-D-galactoside Melibiose -Methyl-D-galactoside Inositol Rhamnose L-Arabinose D-Galactose D-Glucose D-Mannose L-Fucose N-Acetyl-D-glucosamine N-Acetyl-neuraminic acid Sucrose 3.13 6.25 6.25 12.5 12.5 25.0 25.0 25.0 25.0 25.0 50.0b 50.0b 50.0b 50.0b 50.0b 50.0b 50.0b CAL 3.13 12.5 6.25 12.5 12.5 25.0 25.0 25.0 25.0 50.0b 50.0b 50.0b 50.0b 50.0b 50.0b 50.0b 50.0b

Fig. 1. Purification of CRL by affinity column chromatography. Venom solution (50 mg/mL in TBS containing 0.1 mg/mL azide) was applied to a lactose agarose column (4 mL gel) equilibrated with TBS containing 0.1 mg/mL azide. Monitoring the absorbance of eluate at 280 nm, buffer was changed to TBS containing 100 mM lactose at the position indicated by an arrow. The columnbound fractions shown by a bar were collected and concentrated.

between those eluting positions of transferrin and BSA, suggesting that CRL has an apparent molecular mass of 66 80 kDa in the non-denaturing conditions used (data not shown). CAL was eluted at almost the same position as that of CRL. These results together with SDS-PAGE analysis suggest that CRL is a disulfide-linked homodimer composed of 15 kDa subunit and exists as an oligomer in a native condition. Gelpermeation chromatographic analysis of CAL indicated that it forms an aggregate (octamer) under physiological conditions (Hirabayashi et al., 1991) but recent crystal structural analysis of CAL revealed a decameric protein composed of five dimers of the 15 kDa subunit (Walker et al., 2004). An elution profile of CRL on gel filtration was almost identical to that of CAL which suggests that CRL also exists as a large aggregate form. It is not clear why they were eluted at a lower position on a gel filtration column, but some interaction might have been involved between lectin and Superose 12 column since Superose is composed of cross-linked agarose even though EDTA and lactose were added in the elution buffer.

CRL and CAL (the titer was previously adjusted to 16) were mixed with erythrocytes in the presence of serially diluted sugar solutions. The most effective minimum sugar concentration to inhibit lectin-induced hemagglutination was scored.

3.2. Sugar specificity of CRL Hemagglutination induced by CRL was inhibited in the presence of galactosides (Table 1). Thiodigalactoside (TDG, Gal1-1Gal) was the most potent inhibitor followed by Nacetylgalacosamine, lactulose (Gal1-4Fru) and lactose(Gal14Glc).

Fig. 2. SDS-PAGE of CRL and CAL. Aliquots (2 g) of purified CRL and CAL were subjected to SDS-PAGE under reducing (R) or non-reducing (NR) conditions. Numbers with arrows indicate relative molecular mass of marker proteins.

Fig. 3. Effects of divalent cations on CRL- and CAL-induced hemagglutination. CRL and CAL were dialyzed against TBS with or without 5 mM EDTA at 4 C for 15 h and against TBS for 24 h. Lectins treated with EDTA or TBS (control) were incubated with 50 mM TrisHCl (pH 7.5) containing 10 mM metal ion (Ca2+, Mg2+, Ni2+, Mn2+, Ba2+ or TBS (none)) solutions at 25 C for 1 h before mixing with red blood cells. Lectin activity was expressed as % of the control of each lectin dialyzed against TBS.

J. Hamako et al. / Comparative Biochemistry and Physiology, Part B 146 (2007) 299306 Table 2 Mass spectrometric analysis of PECRL and its derived peptides Peptide Residue no. Calculated Intact PECRL K-14 K-15 D-1 D-5 D-7 D-8 D-9 D-10 E-7 E-13 E-15 P-13 P-20 P-24 P-26 1135 114126 3461 120126 8897 8892 7682 98119 98119 125135 109127 5280 5567 81104 6880 110129 16 263.6 2916.0 3377.0 913.4 1291.6 697.3 970.4 2805.4 2819.4 1500.8 3470.0 3573.0 1624.7 3172.4 1651.9 2422.2 MH+ (Da) Observed 17 233.8 2919.2 3385.6 912.0 0.0 1291.0 0.3 697.0 0.0 970.5 0.0 2805.0 0.0 2820.0 0.0 1500.0 0.6 3468.0 0.9 3572.0 0.3 1624.0 0.3 3173.0 1.1 1652.0 0.3 2422.0 0.3

303

metal ions and the reversibility of the lectin activity after iondepletion, several ions were supplemented after EDTA treatment. The hemagglutinating activities of CRL and CAL were recovered to a 100% level by addition of 5 mM Ca2+ or Mg2+ (Fig. 3) indicating that these metal ions interact reversibly. Ni2+ recovered EDTA-treated CAL and CRL to almost 100 and 50% levels of lectin activity, respectively. Mn2+ rescued CAL up to a 50% level but had little effect on CRL. Hemagglutination induced by the intact CRL was inhibited in the presence of Mn2+. Ba 2+ showed no recovering effect on CRL and CAL. Furthermore, Ba2+ was inhibitory to both lectin activities even though lectins were not treated with EDTA. The minimum inhibitory concentration of Ba2+ was 0.25 mM for CAL and CRL. 3.4. Primary structure of CRL The N-terminal 47 amino acid sequence was determined by direct sequencing of the PECRL or CRL on the PVDF after SDS-PAGE. PECRL was digested with AP-I, Asp-N, pepsin and V8 proteinase, and cleaved peptides (designated with prefix K-, D-, P- and E-, respectively) were separated by reversedphase HPLC. Each fractionated peptide was subjected to mass spectroscopic analysis (Table 2) followed by sequencing analysis. Overlaps of the deduced peptide sequence finally resulted in the complete sequence of CRL (Fig. 4). CRL was composed of 135 amino acid residues with 9 Cys residues (amino acid sequence of CRL was registered at UniProt KB with an accession number of P84987). Some microheterogeneity was observed at residues 91 (Asn or Thr), 101 (Glu or Lys) and 113 (Ser or Thr). The amino acid sequence of CRL was compared with other snake venom proteins (Fig. 5). All Cys residues of CRL were preserved as other snake venom C-type lectins. CRL had a high degree of sequence identities with CAL (94%), BAL (Bitis arietans lectin, 89%), APL (Agkistrodon piscivorus lectin, 89%), LML (Lachesis muta stenophrys lectin, 88%), BIL (Bothrops insularis lectin, 89%) and BJL (B. jararacussu lectin, 88%). It also had a significant

Peptides obtained by enzymic digestions followed by HPLC fractionation were subjected to MALDI-TOF MS (Intact PECRL and K peptides) or ion-sprayMS analysis (other peptides). Prefixes K-, D-, E- and P- indicate the peptides obtained by digestion with AP-I, Asp-N, V8 proteinase and pepsin, respectively. Values were calculated for MH + from multiple charged signals observed. The molecular mass of each peptide was calculated as monoisotopic mass from the deduced sequence.

Beta-metylgalactoside was more effective than alpha-metylgalactoside indicating that CRL preferentially recognizes betagalactoside more than alpha-anomer. Monosaccharide galactose, up to 50 mM, showed no inhibitory effect on the hemagglutination induced by both lectins. These sugar specificities of CRL were almost the same as those observed with CAL (Table 1). 3.3. Divalent cation requirement of CRL CRL showed no lectin activity in the presence of EDTA or EGTA indicating that divalent metal cation (Ca2+) is essential for its carbohydrate-recognizing activity, which is the common property of C-type lectins. To elucidate the effects of divalent

Fig. 4. Amino acid sequence of CRL. Determined sequences of intact PECRL (PE-intact) or peptides obtained by limited digestion with AP-I, Asp-N, V8 protease and pepsin, designated with a prefix K, D, E and P, respectively (with fraction number of HPLC), were aligned below the summary sequence (bold face). Lower-case letters indicate tentative identification and x indicates unidentified residues. Residues under asterisk showed microheterogeneity, and the major determination was described as summary sequence.

304

J. Hamako et al. / Comparative Biochemistry and Physiology, Part B 146 (2007) 299306

Fig. 5. Comparison of amino acid sequence of CRL with other snake venom C-type lectins. The amino acid sequence of CRL was aligned with those of other snake venom C-type lectins from C. atrox (CAL) (Hirabayashi et al., 1991), Bitis arietans (BAL) (Nikai et al., 1995), Bothrops insularis cDNA (BIL) (Guimaraes-Gomes et al., 2004), Lachesis muta stenophrys (LML) (Aragon-Ortiz et al., 1996), Agkistrodon piscivorus (APL) (Komori et al., 1999) and B. jararacussu (BJL) (de Carvalho et al., 2002). All Cys residues were conserved among all lectins (box in black). The different residues from CAL are boxed. Numbers on the top indicate the residue number.

homology with C-type lectin-like proteins such as alboaggregin B subunit (36%), botrocetin subunit (33%) and coagulation factor IX/X-binding protein (A chain) (32%). 3.5. Effects of CRL on the human platelet aggregation CRL neither induced platelet aggregation nor showed significant effects on the platelet agglutination mediated by VWF with botrocetin (or bitiscetin or ristocetin) or on platelet aggregation induced by fibrinogen and collagen (or ADP or thrombin) (data not shown). Ogilvie et al. (1989) reported that lactose-specific lectins from Lachesis muta, Ancistrodon piscivorous leukostoma and C. atrox stimulated the aggregation of human platelets at 476 g/mL, and that lectin-induced platelet aggregation was blocked by lactose. In our experiments however, snake venom lectins did not induce platelet aggregation even in the higher concentration (73 g/mL) although betagalactoside specific RCA120 induced platelet aggregation at 0.4 g/mL. It is not clear why our preparation did not induce platelet aggregation. Further experiments might be necessary to elucidate the effect of snake venom lectins on the platelet aggregation. Botrocetin and bitiscetin are platelet agglutination inducers isolated from venom of B. jararaca and Bitis arietans, respectively, and belong to a C-type lectin-like proteins. They bind to VWF and the resulting complex induces platelet agglutination in vitro via platelet glycoprotein Ib (GPIb) (Fukuda et al., 2002; Maita et al., 2003). Further, GPIb-binding proteins from snake venom (which are also C-type lectin-like proteins) inhibit GPIb-mediated platelet aggregation. Our results indicate that CRL and CAL did not interfere these interactions, though many C-type lectin-like proteins modulate VWF-platelet interaction. In the present paper, we have isolated a C-type lectin from C. ruber venom and compared with CAL. C. ruber is one of the 21 species of Crotalus genus living locally in southwestern California and the Pacific coast (Campbell and Lamar, 1989),

while another species C. atrox has a wider habitat ranging from southwestern California, southern Nevada, Arizona, Texas, Oklahoma, Arkansas to south in Mexico (Ernst, 1992). CRL showed almost identical sugar-recognition spectrum and multimeric status as CAL but a different sensitivity against Mn2+ and Ni2+ compared to CAL. The amino acid sequence of CRL was distinct as well, but it showed a high degree of similarity with that of CAL except for eight residues. This differentiation might be attributed to their relative narrow distribution of C. ruber and to the facilitated evolution found in the snake venom proteins (Ogawa et al., 2005). Walker et al. (2004) have recently elucidated the crystal structure of CAL in complex with lactose and TDG. CAL was a

Fig. 6. Homology modeling of CRL compared with the 3D structure of CAL. 3D structure of CRL (right) was predicted by homology modeling based on the structure of CAL (left) (Protein Data Bank entry code 1MUQ) using Insight II software. Amino acid residues, which differ between CAL and CRL, were expressed as CPK model on the main chains. Yellow stick and purple sphere indicate TDG and Ca2+, respectively. Conserved residues forming metalion binding site (Gln96, Asp98, Glu104, Asn119, Asp120) are expressed as light blue on the main chain.

J. Hamako et al. / Comparative Biochemistry and Physiology, Part B 146 (2007) 299306

305

unique decamer composed of two 5-fold symmetric pentamers arranged in a staggered, back-to-back orientation, and the carbohydrate binding sites are located on the rim of the decamer. They reported that Gln96, Asp98, Glu104 and Asn119 make hydrogen bonds with the C3 and C4 hydroxyl groups of galactose, and the side chain of Tyr100 also makes a stacking interaction with the apolar face of the galactose. For the Ca2+ binding, the galactose residue and Asp120 in addition to Gln96, Asp98, Glu104 and Asn119 coordinate the Ca2+ binding site. These amino acid residues are all conserved in CRL, whereas Asp93 and Gln101 in CAL are substituted to Asn and Lys in CRL, respectively (Fig. 5). Since these residues are near the Ca2+ and galactose binding site, substitutions of negativelycharged Asp93 to neutral Asn and neutral Gln to positivelycharged Lys might alter the conformation and influence the sugar binding pocket induced by the metalion binding (Fig. 6). Substitution of Arg71 to Trp brings large side chain moiety, which might alter the formation of the sugar binding pocket constructed by the metalion binding loop indirectly. Alternatively, these substitutions would affect the stability of the sugar binding pocket of CRL against the metalion binding. The difference in the effect of Mn2+ and Ni2+ between CRL and CAL might be attributed to this substitution. The inhibitory effect of Ba2+ might be due to the steric hindrance caused by a large diameter of Ba2+ atom. Further experiments using mutant lectin molecules, in which point mutations are introduced, are necessary to verify these possibilities. The physiological function of snake venom lectin is not clear, however, hemagglutinating activity is harmful to the prey even though lectin did not affect the platelet function. These lectins together with metalloproteinases, serine-proteases, phospholipases and C-type lectin-like proteins (inhibitor or activator for coagulation and hemostasis) act synergistically and effectively deprive the activity of the prey. Acknowledgments We thank Dr. Koiti Titani, Fujita Health University, for helpful suggestions. We are grateful to Ms. Tomomi Matsuura, Ayako Hirai and Mr. Masami Suzuki at Fujita Health University for technical assistance. We thank Mr. Ronald G. Belisle for editing the manuscript. This work was supported by Grants-inAid from Japanese Ministry of Education Culture and Science (to T. M.) and Fujita Health University (to T. M. and J. H.). References
Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. J. Mol. Biol. 215, 403410. Aragon-Ortiz, F., Mentele, R., Auerswald, E.A., 1996. Amino acid sequence of a lectin-like protein from Lachesis muta stenophrys venom. Toxicon 34, 763769. Campbell, J.A., Lamar, W.M., 1989. The venomous reptiles of Latin America. In: (Eds.) Ithaca, Comstock Publ. associates. Chopek, M.W., Girma, J.P., Fujikawa, K., Davie, E.W., Titani, K., 1986. Human von Willebrand factor: a multivalent protein composed of identical subunits. Biochemistry 25, 31463155. de Carvalho, D.D., Marangoni, S., Novello, J.C., 2002. Primary structure characterization of Bothrops jararacussu snake venom lectin. J. Protein Chem. 21, 4350.

Drapeau, G.R., 1980. Substrate specificity of a proteolytic enzyme isolated from a mutant of Pseudomonas fragi. J. Biol. Chem. 255, 839840. Ernst, C.H., 1992. Venomous reptiles of North America. In: (Eds.) Washington, D.C., Smithsonian Inst. Press. Fox, J.W., Serrano, S.M., 2005. Structural considerations of the snake venom metalloproteinases, key members of the M12 reprolysin family of metalloproteinases. Toxicon 45, 969985. Fujimura, Y., Titani, K., Usami, Y., Suzuki, M., Oyama, R., Matsui, T., Fukui, H., Sugimoto, M., Ruggeri, Z.M., 1991. Isolation and chemical characterization of two structurally and functionally distinct forms of botrocetin, the platelet coagglutinin isolated from the venom of Bothrops jararaca. Biochemistry 30, 19571964. Fukuda, K., Doggett, T.A., Bankston, L.A., Cruz, M.A., Diacovo, T.G., Liddington, R.C., 2002. Structural basis of von Willebrand factor activation by the snake toxin botrocetin. Structure 10, 943950. Gartner, T.K., Ogilvie, M.L., 1984. Isolation and characterization of three Ca2+dependent beta-galactoside-specific lectins from snake venoms. Biochem. J. 224, 301307. Guimaraes-Gomes, V., Oliveira-Carvalho, A.L., Junqueira-de-Azevedo Ide, L., DL, S.D., Pujol-Luz, M., Castro, H.C., Ho, P.L., Zingali, R.B., 2004. Cloning, characterization, and structural analysis of a C-type lectin from Bothrops insularis (BiL) venom. Arch. Biochem. Biophys. 432, 111. Hamako, J., Matsui, T., Suzuki, M., Ito, M., Makita, K., Fujimura, Y., Ozeki, Y., Titani, K., 1996. Purification and characterization of bitiscetin, a novel von Willebrand factor modulator protein from Bitis arietans snake venom. Biochem. Biophys. Res. Commun. 226, 273279. Hirabayashi, J., Kusunoki, T., Kasai, K., 1991. Complete primary structure of a galactose-specific lectin from the venom of the rattlesnake Crotalus atrox. Homologies with Ca2(+)-dependent-type lectins. J. Biol. Chem. 266, 23202326. Hirotsu, S., Mizuno, H., Fukuda, K., Qi, M.C., Matsui, T., Hamako, J., Morita, T., Titani, K., 2001. Crystal structure of bitiscetin, a von Willebrand factordependent platelet aggregation inducer. Biochemistry 40, 1359213597. Ito, M., Hamako, J., Sakurai, Y., Matsumoto, M., Fujimura, Y., Suzuki, M., Hashimoto, K., Titani, K., Matsui, T., 2001. Complete amino acid sequence of kaouthiagin, a novel cobra venom metalloproteinase with two disintegrinlike sequences. Biochemistry 40, 45034511. Komori, Y., Nikai, T., Tohkai, T., Sugihara, H., 1999. Primary structure and biological activity of snake venom lectin (APL) from Agkistrodon p. piscivorus (Eastern cottonmouth). Toxicon 37, 10531064. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680685. Lu, Q., Navdaev, A., Clemetson, J.M., Clemetson, K.J., 2005. Snake venom C-type lectins interacting with platelet receptors. Structurefunction relationships and effects on haemostasis. Toxicon 45, 10891098. Maita, N., Nishio, K., Nishimoto, E., Matsui, T., Shikamoto, Y., Morita, T., Sadler, J.E., Mizuno, H., 2003. Crystal structure of von Willebrand factor A1 domain complexed with snake venom, bitiscetin: insight into glycoprotein Ibalpha binding mechanism induced by snake venom proteins. J. Biol. Chem. 278, 3777737781. Masaki, T., Fujihashi, T., Nakamura, K., Soejima, M., 1981. Studies on a new proteolytic enzyme from Achromobacter lyticus M497-1. II. Specificity and inhibition studies of Achromobacter protease I. Biochim. Biophys. Acta 660, 5155. Matsudaira, P., 1987. Sequence from picomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes. J. Biol. Chem. 262, 1003510038. Matsui, T., Hamako, J., 2005. Structure and function of snake venom toxins interacting with human von Willebrand factor. Toxicon 45, 10751087. Matsui, T., Fujimura, Y., Titani, K., 2000. Snake venom proteases affecting hemostasis and thrombosis. Biochim. Biophys. Acta 1477, 146156. Mizuno, H., Fujimoto, Z., Koizumi, M., Kano, H., Atoda, H., Morita, T., 1999. Crystal structure of coagulation factor IX-binding protein from habu snake venom at 2.6 A: implication of central loop swapping based on deletion in the linker region. J. Mol. Biol. 289, 103112. Morita, T., 2005. Structures and functions of snake venom CLPs (C-type lectinlike proteins) with anticoagulant-, procoagulant-, and platelet-modulating activities. Toxicon 45, 10991114.

306

J. Hamako et al. / Comparative Biochemistry and Physiology, Part B 146 (2007) 299306 Ozeki, Y., Tazawa, E., Matsui, T., 1997. D-galactoside-specific lectins from the body wall of an echiuroid (Urechis unicinctus) and two annelids (Neanthes japonica and Marphysa sanguinea). Comp. Biochem. Physiol. B 118, 16. Sen, U., Vasudevan, S., Subbarao, G., McClintock, R.A., Celikel, R., Ruggeri, Z.M., Varughese, K.I., 2001. Crystal structure of the von Willebrand factor modulator botrocetin. Biochemistry 40, 345352. Tsetlin, V., 1999. Snake venom alpha-neurotoxins and other three-finger proteins. Eur. J. Biochem. 264, 281286. Walker, J.R., Nagar, B., Young, N.M., Hirama, T., Rini, J.M., 2004. X-ray crystal structure of a galactose-specific C-type lectin possessing a novel decameric quaternary structure. Biochemistry 43, 37833792.

Nikai, T., Suzuki, J., Komori, Y., Ohkura, M., Ohizumi, Y., Sugihara, H., 1995. Primary structure of the lectin from the venom of Bitis arietans (puff-adder). Biol. Pharm. Bull. 18, 16201622. Ogawa, T., Nakashima, K., Nobuhisa, I., Deshimaru, M., Shimohigashi, Y., Fukumaki, Y., Sakaki, Y., Hattori, S., Ohno, M., 1996. Accelerated evolution of snake venom phospholipase A2 isozymes for acquisition of diverse physiological functions. Toxicon 34, 12291236. Ogawa, T., Chijiwa, T., Oda-Ueda, N., Ohno, M., 2005. Molecular diversity and accelerated evolution of C-type lectin-like proteins from snake venom. Toxicon 45, 114. Ogilvie, M.L., Byl, J.W., Gartner, T.K., 1989. Platelet-aggregation is stimulated by lactose-inhibitable snake venom lectins. Thromb. Haemost. 62, 704707.