Functional proteomic approach to discover geographic variations of king cobra venoms from Southeast Asia and China

J O U RN A L OF P ROT EO M IC S 8 9 ( 2 01 3 ) 1 4 1 –1 53

Available online at www.sciencedirect.com

ScienceDirect www.elsevier.com/locate/jprot

Functional proteomic approach to discover geographic variations of king cobra venoms from Southeast Asia and China Hui-Ching Changa,b,1 , Tein-Shun Tsaic , Inn-Ho Tsaia,b,⁎,1 a

Institute of Biological Chemistry, Academia Sinica, Taipei, Taiwan Institute of Biochemical Sciences, National Taiwan University, Taipei, Taiwan c Department of Biological Science and Technology, National Pingtung University of Science and Technology, Pingtung, Taiwan b

AR TIC LE I N FO

ABS TR ACT

Article history:

This study deciphers the geographic variations of king cobra (Ophiophagus hannah) venom

Received 20 May 2013

using functional proteomics. Pooled samples of king cobra venom (abbreviated as Ohv)

Accepted 13 June 2013

were obtained from Indonesia, Malaysia, Thailand, and two provinces of China, namely

Available online 21 June 2013

Guangxi and Hainan. Using two animal models to test and compare the lethal effects, we found that the Chinese Ohvs were more fatal to mice, while the Southeast Asian Ohvs were

Keywords:

more fatal to lizards (Eutropis multifasciata). Various phospholipases A2 (PLA2s), three-finger

King cobra

toxins (3FTxs) and Kunitz-type inhibitors were purified from these Ohvs and compared.

Ophiophagus hannah

Besides the two Chinese Ohv PLA2s with known sequences, eight novel PLA2s were identified

Venom geographic variations

from the five Ohv samples and their antiplatelet activities were compared. While two 3FTxs

Phospholipases A2

(namely oh-55 and oh-27) were common in all the Ohvs, different sets of 3FTx markers were

Three-finger toxins

present in the Chinese and Southeast Asian Ohvs. All the Ohvs contain the Kunitz inhibitor,

Taxa-dependent lethality

OH-TCI, while only the Chinese Ohvs contain the inhibitor variant, Oh11-1. Relative to the Chinese Ohvs which contained more phospholipases, the Southeast Asian Ohvs had higher metalloproteinase, acetylcholine esterase, and alkaline phosphatase activities. Biological significance Remarkable variations in five king cobra geographic samples reveal fast evolution and dynamic translational regulation of the venom which probably adapted to different prey ecology as testified by the lethal tests on mice and lizards. Our results predict possible variations of the king cobra envenoming to human and the importance of using local antivenin for snakebite treatment. © 2013 Elsevier B.V. All rights reserved.

Abbreviations: Ohv, Ophiophagus hannah venom; BCA, bicinchoninic acid; dPPC, L-dipalmitoyl phosphatidylcholine; ALP, alkaline phosphatase; LAAO, L-amino acid oxidase; PLA2, phospholipase A2; 3FTx, three-finger toxin. ⁎ Corresponding author at: Institute of Biological Chemistry, Academia Sinica, P. O. Box 23-106, Taipei, Taiwan. Tel.: +886 2 23665521; fax: +886 2 23635038. E-mail address: [email protected] (I.-H. Tsai). 1 The first and the third authors contributed equally. 1874-3919/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jprot.2013.06.012

142

1.

J O U RN A L OF P ROTE O M IC S 8 9 ( 2 01 3 ) 1 4 1 –15 3

Introduction

The king cobra (Ophiophagus hannah) currently is the sole member of the Ophiophagus genus, and is the largest venomous snake in the world. This snake inhabits the highland forests from India, through Southeast Asia, to South China and the Philippines [1]. As implied in its Latin name, the king cobra's major food source is other snakes, but they also feed on other small vertebrates when snakes are scarce. Many venom proteins have been isolated and characterized from O. hannah venom (hereafter abbreviated as Ohv), including three finger toxins (3FTxs) [2,3], phospholipases A2 (PLA2) [4–7], L-amino acid oxidase [8], metalloproteinase [9], Kunitz type protease inhibitors [10], and Factor X activator [11]. Dozens of 3FTx cDNAs from the venom glands of Chinese king cobra have been cloned and sequenced, and about 20 of the 3FTxs have been isolated from the venom [2,3]. Additionally, two Chinese Ohv PLA2s have been fully sequences [7]. Recently, a genomic draft (1.36–1.59 Gbp) of an Indonesian king cobra has revealed evidence of duplication and recruitment of many toxin and enzyme genes to the venom glands, and massive expansion of both the PLA2 and the 3FTx gene families [12]. We know that intra-species variations in snake venom composition may exist, and are linked to variables such as diet, geographical distribution, ontogeny and others. Previous comparisons of nine enzymatic activities in four Southeast Asian Ohv samples did not unravel significant differences in their compositions [13]. However, the vast range of king cobra distribution and the lack of proteomic data of southeastern and southern Asian Ohvs warrant a closer look at the Ohv biodiversity. To better understand the geographic variations, we herein studied the proteomes of pooled Ohvs from Indonesia, Malaysia, Thailand, and the two Chinese provinces, Guangxi and Hainan. We show that there are significant differences in the enzymatic activities and taxon-specific lethal potencies (toward mice and lizards) between the Chinese and the Southeast Asian Ohv samples. Furthermore, we have purified various isoforms of PLA2s, 3FTxs, and Kunitz-type protease inhibitors from these Ohvs, and analyzed their molecular masses, N-terminal sequences and biological functions. The results are then discussed in the light of existing Ohv transcriptomic and proteomic data to render a broader perspective on Ohv's geographic variations and evolution.

2.

Materials and methods

2.1.

Venoms and reagents

Lyophilized Ohv samples from Indonesia, Malaysia and Hainan were purchased from Latoxan (Valence, France), Medtoxin Venom Laboratories (DeLand, Florida, U.S.A.), and Xinyuan Snake Venom Co. (Guangzhou, China), respectively. Pooled Ohv samples from Guangxi and Thailand were gifts from Dr. Hai Lan (Nanning, China) and Prof. N.H. Tan (University of Malaya, Kuala Lumpa, Malaysia), respectively. The Guangxi Ohv sample was collected from five king cobras in 2011. However, we could not discern how many snakes had been milked or how broad the distributions were for the other pooled samples.

L-dipalmitoyl

phosphatidylcholine (dPPC) was from Avanti Polar Lipids (Alabaster, AL, U.S.A.). Sequencing-grade modified trypsin was purchased from Promega Corp. (Madison, WI, USA). Sodium deoxycholate, collagen, o-dianisidine and other biochemicals were from Sigma Chemical (St. Louis, MO, U.S.A.). Sequencing-grade Lys-C, p-nitrophenyl phosphate, reagent grade buffers and solvents were from Merck Co. (Germany).

2.2.

Protein determination and SDS-PAGE

Soluble crude venom and isolated proteins were quantified by BCA protein assay (Pierce Chemical Co., Rockford, IL, U.S.A.) using bovine serum albumin as the standard. To prepare samples for SDS-PAGE analyses, 25 μg of venom proteins was dissolved in buffer containing 50 mM dithiothreitol and incubated at 95 °C for 5 min, or in buffer without the reducing agent and incubated at room temperature. The samples were loaded onto a 4–12% acrylamide gel (NuPAGE Bis-Tris gel, Invitrogen, USA) prepared with MES buffer. Electrophoresis was carried out using XCell SureLock™ system (Invitrogen, USA) at a constant voltage of 180 V. After electrophoresis, the gel was stained with Coomassie brilliant blue.

2.3.

Lethal potency of Ohv toward two model animals

Male ICR mice were purchased from BioLASCO, Ltd. (Taipei, Taiwan). Wild lizards (Eutropis multifasciata) were collected in May–December 2012 at Pingtung, Taiwan. The body weights of the mice were 25–35 g, while those of the lizards were 1.50– 49.2 g (mean ± SE = 20.8 g ± 1.49; N = 81). The animals were kept in a 12/12 h light/dark cycle at 25 °C with water and food ad libitum in accordance with the World Health Organization's International Guiding Principles for Animal Research (WHO Chronicle, 1985). Each experimental group contained four animals. A designed dose of each geographic Ohv sample was dissolved in 100 μl (for mice) or 40 μl (for lizards) PBS buffer and injected intraperitoneally into the animals. The animals were observed for up to 24 h after injection, and the number of death was recorded. The apparent median lethal doses were estimated from the lowest dose which resulted in the death of at least two injected animals, or by interpolation from the data correlating the number of death and dosage.

2.4.

In vitro assays of four enzymes

L-amino acid oxidase (LAAO) activity was determined as previously described [14] with slight modifications. Reaction mixtures containing 0.1 M Tris–HCl buffer (pH 7.5), 150 mM NaCl, 1.0 mM L-Leu, 400 mU of horseradish peroxidase, and 10 mM o-dianisidine (substrate for peroxidase) were thermostated at 25 °C and the venom samples were added to initiate the reaction. The initial rates were followed spectrophotometrically at 436 nm for 5 min. One unit (U) of activity is defined as the amount of enzyme that oxidizes 1 nmol of substrate per min using a molar extinction coefficient of 8.3 × 103 M− 1 cm− 1. Alkaline phosphomonoesterase or alkaline phosphatase (ALP) activity was determined using 96-well microplates [15]. Venom (5 μg in 5 μl) was added to a mixture containing 25 μl of 0.01 M p-nitrophenyl phosphate, 15 μl of 0.01 M MgSO4 and

J O U RN A L OF P ROT EO M IC S 8 9 ( 2 01 3 ) 1 4 1 –1 53

25 μl of 0.5 M glycine buffer (pH 8.5) and incubated at 37 °C for 15 min. 100 μl of 0.2 N NaOH was added and the absorbance at 400 nm was measured. Specific activity was calculated based on a molar absorption coefficient of 1.8 × 104 M−1 cm−1 for p-nitrophenolate. Acetylcholine esterase (AChE) activity of Ohv was assayed in 96-well microplates using 0.01–3 mM acetylthiocholine iodide as substrate in 200 μl of 10 mM sodium phosphate (pH 7.5) at 25 °C in the presence of 0.04 M MgCl2 and 0.2 mM 5,5′-dithiobis(2-nitrobenzoic acid) (DTNB) [16]. The optimal substrate concentration was found to be 1.5 ± 0.5 mM. The reaction was monitored at 405 nm using 1.5 mM of acetylthiocholine as substrate, and the specific activity was calculated based on the molar absorption coefficient of 1.4 × 104 M− 1 cm− 1 for nitrobenzoate. The protease activity towards azocasein was measured by a colorimetric assay [17]. 100 μl of reaction mixture containing 0.1 μg of crude venom and 0.5 mg of azocasein in 50 mM Tris– HCl buffer (pH 8.0) was incubated at 37 °C for 90 min. Reaction was halted by adding 200 μl of 5% trichloroacetic acid. After spinning at 1000 g for 5 min, 150 μl of the supernatant was neutralized by an equal volume of 0.5 M NaOH, and the absorbance at 440 nm was measured. One unit of the proteolytic activity is defined as the amount of enzyme required to cause a 1.0 OD440 nm increase per min.

2.5.

Procoagulating activities

Blood (12 mL) was collected from the ear vein of adult rabbits and dispensed into a tube containing 0.15 M sodium citrate (9:1 v/v). After 3000 rpm centrifugation for 10 min, the supernatant plasma was harvested. Activated partial thromboplastin time (aPTT) was performed by adding 0.75–3.0 μg of Ohvs to 50 μL of the rabbit plasma, followed by 50 μL of TriniCLOT HS reagent (Tcoag Ireland Ltd., Ireland), and incubated at 37 °C for 1 min. Finally, 50 μL of pre-warmed CaCl2 was added and the time to clot was measured by a Hemostasis Analyzer (KC-1; Sigma Diagnostics, St. Louis, USA).

2.6.

PLA2 purification and assay

About 10 mg of the crude venom was dissolved in 220 μl of double-distilled water. After repeating centrifugations at 20,000 g for 5 min, 200 μl of the supernatant was injected into a gel-filtration column (Superdex 75, 10/300 GL; GE healthcare, Germany) on an AKTA FPLC system (GE healthcare, Germany). Proteins were eluted with 0.1 M ammonium acetate (pH 6.5) at a flow rate of 1.0 ml/min at room temperature, and fractions of 0.5 ml were collected. PLA2 activity toward mixed micelles of deoxycholate and dPPC was assayed using a pH-stat apparatus (Radiometer, Copenhagen, Denmark) at pH 7.4 and 37 °C [18]. The fractions containing PLA2 activities were pooled, lyophilized, and re-dissolved for further purification by reversed-phase HPLC. A silica gel column (Vydac C18, 4.6 mm × 250 mm, 5 μm particle size, 300 Å pore size) was equilibrated with solvent A (0.07% TFA), and eluted at a flow rate of 1 ml/min by a stepwise linear gradient of solvent B (0.07% TFA in CH3CN): 0–30% for 5 min, 30–45% for 15 min, and 45–55% for 20 min. Collected protein peaks were dried by SpeedVac (Labconco, U.S.A.) and subjected to biochemical analyses.

2.7.

143

Effects of Ohv PLA2s on platelet aggregation

Blood was collected from healthy adult volunteers and dispensed into a tube containing 3.8% sodium citrate (1:9 v/v with blood). After centrifugation at 1000 rpm for 10 min, the supernatant was harvested as platelet-rich plasma (PRP). Platelet aggregation was measured on an aggregometer (Payton, module 600B, Canada) under continuous stirring at 37 °C. The PRP (400 μl) was incubated with the indicated concentration of purified PLA2 or equal volume of PBS (as the control) for 3 min, before collagen was added to a final concentration of 5 μg/ml. Percent inhibition was determined as (1 − residual aggregation / baseline aggregation) × 100%.

2.8. Purification and identification of 3FTxs and Kunitz inhibitors The major peak eluted from the Superdex 75 column contained 3FTxs and Kunitz-type protease inhibitors. This peak was pooled, lyophilized and further purified by ion-exchange chromatography on a Mono S 5/50 GL column (GE healthcare, Germany). The column was first equilibrated with 50 mM MES buffer (pH 6.0) at a flow rate of 1 ml/min. After sample injection, the column was washed for 5 min, followed by the two steps of linear gradients of NaCl in the same buffer: 0 to 0.1 M for 30 min and 0.1 to1 M for 15 min. Major protein peaks were concentrated by a membrane filter (3000 MWCO, Amicon, U.S.A.) before being further purified by reversed-phase HPLC on a silica gel column (Vydac C18, 4.6 mm × 250 mm). The column was first equilibrated with 85% solvent A (0.1% TFA) and 15% solvent B (0.1% TFA in 80% acetonitrile). Elution (1 ml/min) was carried out using a linear gradient of 15% to 50% solvent B.

2.9. Protein identification by mass analyses and N-terminal sequencing Purified proteins were subjected to mass analyses on an ESI-MS spectrometer (MicrOTOF; Bruker Daltonics, Bremen, Germany). The sample was dissolved in 50% (v/v) acetonitrile with 0.1% formic acid and premixed with a 5 mg/ml matrix solution of sinapic acid in 70% (v/v) acetonitrile with 0.1% formic acid for spotting onto the target plate. In certain cases, the protein mass was analyzed with Q-TOF Ultima MALDI (Micromass, Manchester, UK) at the Protein Core of Academia Sinica, Taiwan. For protein identification, LC-MS/MS peptide sequencing and fingerprinting were performed. In-gel digestion with sequencing grade Lys-C (for 3FTx identification) or trypsin (for venom components other than 3FTx) was carried out as previously described [18]. The resultant peptides were analyzed by an ESI-QUAD-TOF instrument (Waters Synapt G2 HDMS, Manchester, UK) operating in a data-dependent mode of acquisition. Peak list file was created and submitted to Mascot (Server v2.2, Swiss Prot) using MS/MS Ion Search to identify the venom proteins. Search parameters included “up to two missed cleavage” and various side chain modifications including carbamidomethyl (C), oxidation (M), oxidation (HW), and propionamide (C) at 25 ppm peptide tolerances and 0.05 Da mass tolerances. To determine the N-terminal sequences, purified proteins were desalted, vacuum dried and sequenced by an automatic sequencer (Procise 492; applied Biosystems, USA). Alternatively,

144

J O U RN A L OF P ROTE O M IC S 8 9 ( 2 01 3 ) 1 4 1 –15 3

protein samples were separated by SDS-PAGE under reducing conditions. After the electrophoresis, protein bands were blotted onto a PVDF membrane and stained with Amido Black (0.2% in 7% acetic acid). Specific protein bands were cut out and subjected to sequence analyses.

3.

Results and discussion

3.1.

SDS-PAGE analyses of Ohv geographic samples

As shown by SDS-PAGE under both reducing and non-reducing conditions (Fig. 1), high (>50 kDa) and low (<14 kDa) molecular weight proteins are abundant in Ohvs. All of the Ohvs showed protein bands of 65, 53, 27, 14, and 7 kDa (Fig. 1), which might represent the major venom families, namely LAAO, metalloprotease (MP), cysteine-rich secretory protein (CRISP), PLA2s, 3FTxs and Kunitz inhibitors [2,7–10,19], respectively. The LAAO and the metalloprotease bands were upper-shifted under the reducing conditions, as previously reported by another research group [8,9]. The electrophoretic patterns of Malaysian and Thai Ohvs are very similar, but different from that of the two Chinese Ohvs, which are also similar to each other. Relative to the Southeast Asian Ohvs, the Chinese Ohvs contain less CRISP but more PLA2s and long chain 3FTxs. In addition, natriuretic peptides (1.9 kDa), VEGF, NGF, AChE, hyaluronidase [12], and complement-depleting factor [20] have been isolated or identified in Ohv before. By tryptic peptide sequencing on LC-MS/MS, we found that the 140 kDa, 110 kDa, and 27 kDa protein bands in Fig. 1 (non-reducing conditions) could be identified as complement-depleting factor (previously named OVF [20]), Ohv LAAO [8], and Ohv CRISP [19], respectively. It is known that LAAO is present as non-covalent homodimer in crude venom [21,22], and OVF contains three subunits with respective masses of 72 kDa, 45 kDa and 32 kDa.

Fig. 1 – Comparison of five geographic samples of Ohvs by SDSPAGE. The geographic origins are shown at the top using the first two letters of the regions. Number 1–7 arrows indicate the predicted bands of OVF (Ophiophagus venom factor), homodimeric LAAO, monomeric LAAO, MP, CRISP, PLA2, and 3FTx, respectively.

3.2.

Taxon-dependent venom fatality

King cobras mainly prey on snakes [1]. Although testing venom activity on natural prey is important for our understanding of the evolution of venom, natural prey species are often difficult to obtain in sufficient numbers. As shown in Fig. 2A, the apparent median lethal doses for mice of the Ohvs from Indonesia, Malaysia, Thailand, Guangxi, and Hainan were estimated to be 1.1, 2.9, 3.5, 0.6, and 0.5 μg/g, respectively. These values are similar to the previously reported Ohv lethal doses [1,13]. Thus, the Chinese Ohvs are more fatal to mice compared to the Southeast Asian Ohvs. We noticed that the injected ICR mice showed typical symptoms of peripheral neurotoxicity that are similar to the symptoms of drowsiness, muscle paralysis, and respiratory failure in human patients [23]. E. multifasciata has been a recent invading lizard in Taiwan [24] but it is widely distributed in India, Southeast Asia, and southern China [25]. The lizard is thus a possible prey for king cobras. The envenomed lizards showed signs of neurotoxicity, e.g. bilateral ptosis, relaxed mouths, limb quivering, and paralysis. For E. multifasciata, the apparent median lethal doses of Ohvs from Indonesia, Malaysia, and Guangxi were estimated to be 17.6, 13.9 and 30.2 μg/g, respectively (Fig. 2B). Although all of the Ohvs are in general more fatal to mice than to lizards, the Southeast Asian Ohvs are more fatal to lizards compared to the Chinese Ohvs. The Chinese Ohv is 60-fold more fatal to mice than to lizards, however, for the Malaysian Ohv, the specificity difference is only 5-fold. Taxa-dependent toxicities of natural toxins could be resulted from arsenal races between predators and preys. For example,

Fig. 2 – Comparison of the lethal potencies of different geographic samples of Ohv in mice (A) and E. multifasciata lizards (B). Number of deaths in each group was examined after 24 h.

J O U RN A L OF P ROT EO M IC S 8 9 ( 2 01 3 ) 1 4 1 –1 53

the preferred binding of Denmotoxin (a 3FTx from bird-hunting Colubrid snake) to the chick acetylcholine receptors rather than the mouse receptors has been attributed to special electrostatic interactions between Asp41 of the toxin and Arg193 in the δsubunit of the chick α1βγδ-nAChR [26]. Additionally, the LD50 (34.8 ± 5.2 μg/g) of Daboia siamensis venom in frog (Hoplobatrachus rugulosus) [27] was 22-fold higher than that in mice [28]. The much lower lethality of viperid venoms towards amphibians compared to rodents could be due to the lower activities of venom enzymes or toxins under the ambient body temperature of the amphibians.

3.3.

Variations in the specific activities of Ohv enzymes

Recent study revealed that metalloprotease, hyaluronidase, and LAAO are expressed in king cobra venom glands [12]. Apparently, high LAAO activities were detected in all of the Ohv samples, among which the Indonesian Ohv contained the highest activity (Fig. 3A). This oxidase is known to inhibit platelet aggregation, induce local inflammation, and inhibit

145

bacterial growth [8]. As shown in Fig. 3B, the PLA2 specific activities of Chinese Ohvs were higher than those of Southeast Asian Ohvs. In agreement with the results, the PLA2 yields of Guangxi and Hainan Ohvs were estimated to be 2.6%, and 2.0% of the total venom proteins, while the PLA2 yields of Indonesian and Malaysian Ohvs were 0.9% and 0.7%, respectively (Table 1). The differences in PLA2 contents are in agreement with the relative intensities of 14 kDa bands shown on the protein gel (Fig. 1). It is known that Ohv contains metalloprotease as the major protease [12,13]. It is thus not surprising that the Ohv caseinolytic activities were significantly inhibited by 5 mM EDTA (Fig. 3C). The Ohv metalloproteases showed dermal hemorrhagic effects on rabbit but not on other mammals tested [13,29]. We found that the Southeast Asian Ohvs contained more metalloproteinase activities than the Chinese Ohvs. The Southeast Asian Ohvs also have higher ALP activities (Fig. 3D), especially the Indonesian Ohvs. ALP is commonly detected in crotalid and elapid venoms. The enzyme plays a central role in liberating purines (especially adenosine) which may inhibit platelet aggregation, increase

Fig. 3 – Comparison of various enzymatic activities and procoagulant effects of the Ohv geographic samples. The geographic regions are abbreviated using the first two letters. L-amino acid oxidase (A), phospholipase A2 (B), caseinolytic with or without EDTA (C), alkaline phosphatase (D), acetylcholine esterase (E) and aPTT (F). Results shown are means ± S.D. of triplicate experiments.

146

J O U RN A L OF P ROTE O M IC S 8 9 ( 2 01 3 ) 1 4 1 –15 3

Table 1 – Geographic variations of PLA2 in the Ohv samples. Group IB PLA2s are underlined. The PLA2 mass was determined by ESI-MS. The specific activities were calculated based on the initial rate and the PLA2 concentration. Amino acid sequences are shown in single-letter code (n.d.: not determined). Geographic origins of the Ohvs are abbreviated by the first two letters, and %yields of the PLA2s relative to the total Ohv proteins (w/w) are shown in parentheses. PLA2 variants

Mass (Da)

OH APLA2 OH APLA2II PLA-1 PLA-2 PLA-3 PLA-4 PLA-5 PLA-6 PLA-7 PLA-8

13,720 13,176 13,263 13,840 13,275 13,222 14,011 13,874 14,036 13,868

Specific activity (μmol/mg/min) 1709 1725 864 11 1300 776 187 14 119 81

± ± ± ± ± ± ± ± ± ±

258 343 132 7.9 172 102 46 5.3 29 10

N-terminal sequences

Ohv origin (%yield of the PLA2)

HLIQFGNM HLVQFNGM n.d. NLLQFNYM HLVQFNGM HLVQFNGM NLLQFNGM n.d. n.d. n.d.

Gu (1.6), Ha (1.3) Gu (0.34), Ha (0.44) Gu (0.69) Ha (0.27), In (0.06) In (0.32), Ma (0.31) In (0.22), Ma (0.21), Th (0.82) In (0.25) Ma (0.16) Th (0.07) Th (0.19)

Fig. 4 – Purification of PLA2s from four Ohv samples. (A) Gel-filtration of crude Ohv on a Superdex 75 column. The PLA2-activities (dashed line) were assayed for each fraction. Fractions containing 3FTxs are indicated by a horizontal bar. (B) The PLA2 fractions pooled from (A) were lyophilized and further purified by RP-HPLC. Novel PLA2s are designated as PLA-1 to 6.

147

J O U RN A L OF P ROT EO M IC S 8 9 ( 2 01 3 ) 1 4 1 –1 53

Table 2 – Peptide mass fingerprint analyses. PLA-3 and Oha-III were hydrolyzed by trypsin and the Lys-C, respectively. Mass-determined for the resultant peptides was matched to the theoretical peptide masses calculated based on the amino acid sequences of OH-APLA2II and LNTX-2H, respectively. Protein

Experiment Mass, Da

Theoretical Mass, Da

PLA-3

2876.2 1738.7 1816.6 1308.7 1175.6 1335.6 2579.1 1487.6 1503.6 1360.7 2197.2 854.5 1785.7

2876.2 1738.7 1816.6 1308.7 1175.6 1335.6 2579.1 1487.6 1503.6 1360.7 2197.2 854.5 1785.7

Oha-III

a b

Modification 4 2 3 1

SCMC a SCMC a SCMC a SCMC a

1 3 2 2 2 2

SCMC a SCMC a SCMC a SCMC a + 1 HTRP b SCMC a SCMC a

Residue positions 44–66 67–80 81–95 96–108 109–118 109–119 3–24 25–36 25–36 37–48 37–56 49–56 57–70

3 SCMC a

Predicted amino acid sequence CCQVHDNCYTQAQQLTECSPYSK RYSYDCSEGTLTCK ADNDECAAFVCDCDR VAAICFAGAPYNK ENINIDTTTR ENINIDTTTRC CYVTPDATSQTCPDGENICYTK SWCDVFCSSRGK SWCDVFCSSRGK RIDLGCAATCPK RIDLGCAATCPKVKPGVDIK VKPGVDIK CCSTDNCNPFTPWK

SCMC: S-carbamidomethyl Cys resulted from reductive alkylation of Cys. HTRP: hydroxytryptophan derived from tryptophan oxidation.

vascular permeability, and induce hypotension [30]. The phosphodiesterase activities present in many elapid venom species including Ohv [13] have not been compared here. The venoms of krait, cobra, and king cobra are known to be relatively rich in AChE activities among elapid venoms [13]; the Ohv AChE could cross react with the monoclonal antibodies of krait's AChE [31]. As shown in Fig. 3E, the AChE specific activities of both Malaysian and Thai Ohvs were several-fold higher than those of the Chinese Ohvs. Interestingly, neurotoxicity seems to be inversely associated with higher AChE activity of the elapid venoms. Most likely, the immediate hydrolysis of acetylcholine released from the synaptic vesicles may avoid the competition between the neurotransmitter and the neurotoxic 3FTxs and thus potentiates their toxicities.

3.4.

Pro-coagulant effects of Ohv on rabbit plasma

A recent report described the prolongation of thrombin clotting time and aPTT by Indian Ohv [32]. A serine protease with factor X-activating activity has also been found in the Chinese Ohv [11]. As shown in Fig. 3F, all of the Ohvs showed procoagulant

Table 3 – Inhibition of collagen-induced human PRP aggregation by Ohv PLA2. The experiments were repeated two or three times with reproducible results. PLA2 variant OH-APLA2 OH-APLA2 II PLA-1 PLA-2 PLA-3 PLA-4 PLA-5 PLA-6

Dose (nM)

Inhibition (%)

200 400 200 400 400 400 400 400 100 200 400

30 83 3 5 6 8 29 55 73 89 6

effects at a final concentration of 20 μg/ml or above, and the Southeast Asian Ohvs were slightly more potent than the Chinese Ohvs. However, human plasma was relatively insensitive to Ohv (even at a final concentration of 50 μg/ml) by aPTT measurement (data not shown).

3.5.

Geographic variations of PLA2 isoforms

The Superdex 75 elution patterns of the three Southeast Asian Ohvs are very similar but slightly different from those of the Chinese Ohvs (Fig. 4A and Supplemental Fig. 1A). The Ohv PLA2 activities were eluted from the column at around 10–12 ml, indicating that the PLA2s possibly present as non-covalent dimers in pH 6.5 buffer. After gel filtration and RP-HPLC, some of the PLA2 peaks were subjected to the second round RP-HPLC for purification. PLA-2, PLA-6 and PLA-8 were eluted at similar retention time, and so were PLA-1 and PLA-4, or PLA-3 and PLA-7 (Fig. 4B and Supplemental Fig. 1B). The masses of purified PLA2s were determined and matched to those of known Ohv PLA2s. Most of the Ohv samples contained three PLA2 isoforms except for the Indonesian Ohv which contained four PLA2s. A total of eight novel PLA2 isoforms (designated as PLA-1 to PLA-8) were purified from the five Ohv samples. Although their full amino acid sequences are not solved, their masses, specific activities, N-terminal sequences, and occurrences are compiled in Table 1. Group IB PLA2s (with pancreatic loop [6] and mass > 13.7 kDa) could be easily distinguished from Group IA PLA2s (mass < 13.4 kDa). The Ohv PLA2s showed a wide range of specific activities towards the lecithin micelles, and their enzymatic activities and anti-platelet effects served as references for their identification. OH-APLA2 and OH-APLA2 II [7] were abundant only in Chinese Ohvs. In contrast, PLA-3 and PLA-4 were the major isoforms present in Southeast Asian Ohvs and they were probably the two isoforms previously isolated from Malayan Ohv [4]. The tryptic MS/MS profiles of PLA-7 (data not shown), PLA-3 (Table 2 upper part), PLA-1, PLA-4, and PLA-5 (data not shown) revealed a 81%, 64%, 64%, 44%, and 35% match, respectively, to the sequence of OH-APLA2 II, but not to that of

148

J O U RN A L OF P ROTE O M IC S 8 9 ( 2 01 3 ) 1 4 1 –15 3

Fig. 5 – Separation of 3FTxs in four Ohv geographic samples by Mono S ion-exchanger. Fractions containing 3FTxs (Fig. 4A) were isolated by Mono S column on an FPCL system. The toxins were eluted at flow-rate of 1.0 ml/min using two-steps of NaCl gradient as shown. Major peaks were harvested and annotated with numbers.

OH-APLA2. We suspect that PLA-7 of Thai Ohv and PLA-3 of Indonesian and Malaysian Ohvs are orthologous to the Chinese OH-APLA2 II. Additionally, the tryptic MS/MS results revealed that 9 peptides of PLA-4 match to the 11 peptides of PLA-1, suggesting their structural similarity although their antiplatelet activities differed (Table 3), while the tryptic MS/MS profiles of PLA-3 and PLA-5 were less similar to those of PLA-1 and PLA-4. The tryptic MS/MS results of PLA-2, PLA-6 and PLA-8 (data not shown) revealed partial peptide-matches to one another, as well as sequence similarities to a Group IB PLA2 previously cloned from a Chinese king cobra (accession no. AAG17443). The N-terminal sequence of PLA-2 was identical to that of AAG17443-encoded PLA2 (Table 1). PLA-2, PLA-6 and PLA-8 all showed very low catalytic activities and were isolated from different Ohvs; thus, they are possibly orthologs expressed in different Ohvs. The AAG17443-PLA2 contains unusual amino acid substitutions Thr31 (part of the interface-recognition site) and Val33 (normally Gly33 of the Ca+ 2-binding loop), which possibly result in the low catalytic activity of these orthologous PLA2s [33]. While OH-APLA2 of the Chinese Ohvs is highly lethal to mice [34], most of the PLA2s in Southeast Asian Ohvs are similar to either OH-APLA2 II or AAG17443-encoded PLA2, and possibly are less lethal to mice [4]. The toxic site of OH-APLA2 has been attributed to the basic residues, Lys21, Lys40, Lys57, and Arg80 [35] while the corresponding four residues in OH-APLA2 II and the AAG17443-encoded PLA2 are either acidic or neutral. Notably, most Ohv PLA2s contain a characteristic Asn6 except that OH-APLA2 contains a unique Gly6 substitution (Table 1), while Lys6 substitution is usually observed in the PLA2s of cobras and kraits and their amino acid sequences are only about 55–70%

similar to OH-APLA2 and OH-APLA2 II [36]. Recent study of the PLA2 exon-2 sequences of Indonesian Ohv reported seven distinct PLA2 genes [Chapter 5 in ref. 12]. The N-terminal sequences of the purified Indonesian Ohv PLA2 are either NLLQFN/ or HLVQFN/ (Table 1) which seem to match the sequences of the PLA2 genomic hit No. 4, 7, or 8, and hit No. 9, respectively [12]. The fact that we only isolated four PLA2s from the Indonesian Ohv (Table 1) could possibly be explained by negative translational regulation of these PLA2s.

3.6.

The effects of Ohv PLA2s on platelets

Among more than a dozen elapid venom species that have been analyzed, Ohvs showed relatively high antiplatelet activities [37], which could be attributed to the LAAO, ALP, and PLA2 of the venom. We thus compared the antiplatelet effects of the purified Ohv PLA2 on human PRP (shown in Table 3) and rabbit PRP (not shown). Using collagen as the agonist, only three isoforms, OH-APLA2, PLA-4 and PLA-5, showed strong antiplatelet effects. At 400 nM, OH-APLA2 was at least four-times more potent than OH-APLA2II toward both PRPs. The antiplatelet effect of PLA-1 was much weaker than PLA-4. PLA-5 (found only in Indonesian Ohv) showed exceptionally high antiplatelet effect at a concentration of ≤100 nM, while the less catalytic PLA-2 and PLA-6 showed very low antiplatelet activities. Although the antiplatelet effects of PLA-7 and PLA-8 of the Thai Ohv were not studied, they have been shown by tryptic MS/MS analyses to be highly similar to OH-APLA2 II and PLA-2, respectively, which were found to have low antiplatelet activities (Table 3). Our results support that the antiplatelet effects of Ohv PLA2s were not closely related to their lipolytic activities, in agreement with a previous report [38].

J O U RN A L OF P ROT EO M IC S 8 9 ( 2 01 3 ) 1 4 1 –1 53

149

Fig. 6 – Further purification of Ohv 3FTxs by RP-HPLC. Only the elution profiles of the major four Mono S peaks for each sample (annotated using the first two letters of the region) are shown, and the rest are in Supplemental Fig. 2. Each of the Mono S peak (see numbering in Fig. 5) was concentrated before loaded onto a C18-column. Elution was carried out using a linear gradient of solvent B: 15–50% for 50 min. Gradient line is omitted for clarity. The 3FTxs are annotated by matching their masses with those of the known toxins in databases. Superscripts “a–c” follow those in Table 4 footnotes.

3.7.

Purification and proteomic analyses of 3FTxs

About 5–10 3FTx peaks for each Ohv sample could be separated by Mono S (5/50 GL) column. When injected at the dosage of 2.0 μg/g, all of the 3FTx fractions were lethal to mice except for the weakly-bound or the proteins eluted during the first 5 min. The Southeast Asian Ohvs had more weakly-bound 3FTxs compared to the Chinese Ohvs (Fig. 5). During purification by RPHPLC, the 3FTxs were eluted out at 26–46% solvent B (Fig. 6 and Supplemental Fig. 2). Seven or more 3FTx isoforms were purified from each Ohv sample and they could be classified as

either long-chain or short-chain neurotoxins [2,3,39] according to their masses and disulfide bond numbers (>7090 Da with 5 disulfide bonds, or <7060 Da with 4 disulfide bonds, respectively). We identified a total of 12 long-chain 3FTxs and 7 short-chain 3FTxs in all the samples (Table 4), confirming a high diversity of Ohv 3FTxs [2,12]. Oxidation of Trp or Met residues in some of the 3FTxs resulted in a mass increase of 16 or 32 Da [40]. This phenomenon could be attributed to the hydrogen peroxide generated by the abundant L-amino acid oxidase in the Ohvs [8]. Remarkably, a long chain toxin oh-55 and a short chain toxin oh-27 were isolated from all the four Ohv samples we

150

J O U RN A L OF P ROTE O M IC S 8 9 ( 2 01 3 ) 1 4 1 –15 3

Table 4 – Variations of Ohv 3FTxs in four geographic samples. Long-chain toxins are listed above the short-chain toxins. Protein masses were determined by ESI-MS spectrometry. The yield of each purified 3FTx relative to the crude venom was quantified by BCA protein determination. A hyphen indicates absence in the venom. UniProt or Genbank accession numbers of the toxins are shown in parentheses: oh-55 (Q53B58), Toxin b (P01386), oh-17 (Q53B54), oh-57 (Q53B56), Oh-4 (AAB25587), oh-3/Oh-6 (P82662), oh-34 (Q53B53), LNTX-2H (DQ902574), oh-37 (Q53B59), oh-56 (Q53B57), oh-27 (Q69CK0), oh-84 (Q53B46), wntx33 (ABB83636), Oh9-1 (P83302), SNTX14 (Q2VBP0), and wtDE-1 (ABK41956). 3FTx

oh-55 Toxin b oh-17 oh-56 oh-57 Oh-4 oh-3/Oh-6 oh-34 oh-37 LNTX-2H Oha-II Oha-III oh-27 oh-84 wntx33 Oh9-1 SNTX14 wtDE-1 Oha-I a b c

Mass determined (Da)

7941 a 8045 8028 7583 7757 8004 7523 7682 7532 7976 a 7096 7555 7012 7054 6998 6508 6562 7019 7001

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

%yield (w/w) in the Ohv sample Indonesia

Malaysia

Guangxi

Hainan

1.5 b + 1.3 a -

1.7 a -

1.6 a 9.3 0.7 0.1 + 0.8 a 0.3 + 0.8 + 0.5 + 2.0 2.4 2.1 1.6 1.0 0.2 -

2.0 a 3.4 + 1.0 a 0.4 + 1.4 + 0.8 + 0.2 + 0.8 0.7 4.0 -

0.8 0.2 2.1 a 0.5 2.3 0.7 29.5 0.8

0.005 c 1.1 0.2 0.2 40.3 0.6

0.2 a 0.3 c 1.1 a 0.1 a

5.4 a 0.8 a 6.3 c 3.6 a 0.5 a

Apparent mass increase of 32 Da, attributable to two Trp oxidations. Apparent mass increase of 16 Da, attributable to one Trp oxidation. Apparent mass increase of 32 Da, attributable to one Met and one Trp oxidations.

analyzed (Table 4). Because of its potent analgesic action on mice, oh-55 has also been named hannalgesin [41] and it is lethal to mice (LD50 = 80–120 ng/g). oh-27 has also been named β-cardiotoxin because of its beta-blocker effects and low lethal potency in mice [42]. The toxin oh-84 is relatively abundant in the Chinese Ohvs and it differs from oh-27 by only two amino acid substitutions (Fig. 7). We also confirm the presence of Oh9-1 [43], oh-56, SNTX14 and wntx33 in our Guangxi Ohv. The amino acid sequence of oh-56 is 84% identical to that of oh-55 [2,3]. Notably, the Chinese Ohvs are rich in either Toxin b or oh-17, which differ from each other by four substitutions at loop I [2]. Taken together the results in Table 4 and previous reports [2,3,39,43], long chain toxins Oh-4, oh-17, oh-56, oh-57 and short chain wntx33 are the 3FTx markers of Chinese Ohvs and are absent in Southeast Asian Ohvs. Notably, the lethal toxins oh-3 (also named Oh-6 [39]), Oh-4 and oh-17 are more abundant in the Hainan Ohv than in the Guangxi Ohv (Table 5). Both oh-3 and Oh-4 showed significant cytolytic activities in addition to neurotoxic activities [3] and they possibly could synergize with the cardiotoxic OH-APLA2 [34,44]. Thus, Hainan Ohv might be more cytotoxic and cardiotoxic than Guangxi Ohv, agreeing with the results obtained for mouse lethality tests (Fig. 2A). In contrast to the abundance of long chain 3FTxs in Chinese Ohvs, the most abundant 3FTx in Southeast Asian Ohvs is the short chain toxin wtDE-1 (Table 4). The Indonesian Ohv contains more long chain toxins than the Malaysian and Thai Ohvs, including LNTX-2H (7943 Da) [42] and a novel toxin Oha-III (7555 Da). The amino acid sequence of LNTX-2H is very similar to those of Toxin b and oh-17 of Chinese Ohvs

(Fig. 7), and these orthologous toxins are all fatal to mice. The MS/MS data of Lys-C-digested Oha-III (Table 2 lower part) suggest that it is possibly a truncated form of LNTX-2H. Some of the long-chain 3FTxs are known to be more toxic to mice than the short-chain 3FTxs [39,45,46], but their contents are low in the Malaysian and Thai Ohvs (Table 5). We also examined the 20 partial 3FTx sequences of the Indonesian king cobra genomic draft [Chapter 5 of ref. 12] and identified eight known 3FTxs, namely, oh-55, oh-84, oh-57, Oh9-1, oh-34, LNTX-2H, wtDE-1, and haditoxin [47]. However, oh-84, oh-57 and Oh9-1 are missing in our Indonesia Ohv sample, although they have been isolated from the Chinese Ohvs [2,43]. This may reflect a strong translational regulation and selected expression of the 3FTxs which result in discrepancy between the proteome and the venom transcriptome. It was reported that Thai Ohv contained two lethal 3FTxs (Toxin b and its homolog) [45] and a non-lethal 3FTx named DE-1 [48]. We have also characterized 3FTxs in the major peaks of Thai Ohv eluted from the Mono S column, and found toxins with masses matching to those of wtDE-1, Oha-II and an oxidized form of oh-55, respectively (Supplemental Fig. 1C). We could not, however, find any 3FTx with a mass identical to that of DE-1. In fact, wtDE-1 could be an ortholog of wntx33 since their amino acid sequences differ by only two conservative substitutions (Fig. 7). The N-terminal seven residues of the newly identified Oha-I and Oha-II are identical to those of wtDE-1 and oh-56 (or oh-55 and Toxin b) [2], respectively. The Lys-C-digested peptide MS/MS results (data not shown) also supported sequence similarities between Oha-I and wtDE-1 (37% match), and between Oha-II and oh-56 (33% match). Taken together, long

J O U RN A L OF P ROT EO M IC S 8 9 ( 2 01 3 ) 1 4 1 –1 53

151

Fig. 7 – Amino acid sequence alignment of the major Ohv-3FTxs. UniProt or Genbank accession numbers are: oh-55 (Q53B58), Toxin b (P01386), oh-17 (Q53B54), LNTX-2H (DQ902574), Oh-4 (AAB25587), oh-3/Oh-6 (P82662), wtDE-1 (ABK41956), wntx33 (ABB83636), Oh9-1 (P83302), SNTX14 (Q2VBP0), oh-84 (Q53B46), oh-27 (Q69CK0), Wa-III (ABX82864). Conserved residues are shown as white letters against a black background. chain toxins LNTX-2H, Oha-II and Oha-III, and short chain wtDE-1 and Oha-I could be the 3FTx markers of Southeast Asian Ohvs (Table 4). Both wtDE-1 and wntx33 contain Asp32 substitutions corresponding to the Asp41 of Denmotoxin, which was shown to bind the nAChRs of chick 100-fold stronger than those of mouse [26], while Oh9-1 and oh-26 with hydrophobic residues at this positions are more lethal to mouse (Table 5 and ref. [2]). It remains to be verified that the presence of this specific Asp residue in loop II of a 3FTx could confer a better binding specificity not only for the bird receptors [26] but also for the reptile receptors. The abundance of short chain wtDE-1 in all Southeast Asian Ohvs in contrast to the high contents of long Table 5 – Masses and lethalities of major Ohv 3FTxs. Theoretical mass of each toxin was calculated based on the amino acid sequence. The LD50 values in mouse (by intraperitoneal injection) were based on literatures or the present study. 3FTx

Mass, Da

LD50 (μg/g mice) [ref.]

Oh-4 oh-3 oh-17 oh-34 oh-37 oh-55 Toxin b oh-26 oh-27 Oh9-1 wtDE-1 wntx33 LNTX-2H a oh-84 SNTX-14 oh-57 a Oha-I Oha-II Oha-III

8004 7524 8027 7683 7534 7909 8043 6397 7012 6508 7019 6998 7976 7054 6562 7788 7001 7096 7555

0.25 [39] 0.10–0.17 [2,39] 0.21 [2] >0.30 [2] 0.11 [2] 0.08–0.12 [2,41] 0.11 [45] 0.16 [2] >10 [42] 2 [43] >10 [48 and this study] >0.50 [this study] 0.21 [this study] >0.50 [this study] >0.50 [this study] >0.50 [this study] >0.50 [this study] >0.50 [this study] >0.50 [this study]

a

Apparent mass was 32 Da higher than expected, possibly due to two Trp oxidations.

chain lethal toxins for mice in the Chinese Ohvs possibly is responsible for the dramatic shift of the fatality between taxon type (or between reptiles and mammals) (Fig. 2). Remarkably, many of the Ohv 3FTxs belong to the weak toxin group and orphan group X and XVIII [37], which are different from the 3FTxs expressed in the Asian cobra and krait venoms [36,49]. Structural plasticity and subtle deviations between the 3FTxs presumably result in distinction of their molecular targets [49]. Moreover, functional complexity could be increased by synergism between certain 3FTxs to generate new toxicity [3,50].

3.8.

Variations of Ohv Kunitz inhibitors

From all of the Ohv samples, we isolated a protein of 6339 Da during the purification of 3FTxs. This protein was identified as the previously reported trypsin/chymotrypsin inhibitor (OH-TCI) [10]. We also purified another relatively abundant Kunitz-inhibitor of 6493 Da (identical to the mass of Oh11-1 [51]) from the Chinese Ohvs. Oh11-1 is a weak chymotrypsin inhibitor with Asn substitution at the active site P1 position, and its amino acid sequence differs from that of OH-TCI mainly at residues 10–29 [10]. OH-TCI showed much higher affinity for trypsin than Oh11-1 [10,51]. However, the true protease targets and physiological roles of both inhibitors remain to be elucidated. Interestingly, both Ohv inhibitors are structurally similar to the two Kunitz inhibitors Wa KIn-I and Wa KIn-II of W. aegyptia venom [36], respectively.

4.

Conclusion and perspective

Different regional distribution and prey ecology should have a great impact on the evolution of snake venom [52]. In addition, venom differences perhaps are reflective of the organismal diversities and individual variations. The present study has used pooled Ohvs from five regions to explore and consolidate the geographic variations of king cobra. We were not able to obtain Ohvs from India and Myanmar for a more comprehensive study. However, advanced mass spectrometry and

152

J O U RN A L OF P ROTE O M IC S 8 9 ( 2 01 3 ) 1 4 1 –15 3

powerful combination of Ohv proteomics with existing Ohv transcriptomics have led to the discovery of eight novel PLA2s and several new 3FTxs, and confirmed the molecular evolution of orthologous venom proteins. Relative to the Southeast Asian Ohvs, the Chinese Ohvs are more fatal to mice, probably because the Chinese Ohvs contain more toxic PLA2s and lethal long-chain 3FTxs. However, the Chinese Ohvs contain less wtDE-1 like 3FTxs and CRISP, and have lower specific activities of ALP, AChE and metalloprotease than the Southeast Asian Ohvs, and elicit much lower fatality in lizards. This might imply different ecological conditions and adaptations of the king cobra subpopulations, and distinct clinical phenotypes of human envenoming in different regions. The venom variations observed hopefully would provide valuable information for better antivenom production and snakebite treatment, and support the conservation of biodiversity of this species.

Conflict of interest statement The authors declare no conflicts of interest.

Acknowledgments We are grateful for the generous gifts of the Guangxi king cobra venom from Dr. Lan, Hai and of the Thai king cobra venom from Prof. Tan, Nget Hong. Great appreciation also goes to Yu-Qing Lin and Yi-Ling Huang (National Pingtung University of Science and Technology) for assistance in the lizard experiments. This work was supported by the research grants funded by the Academia Sinica, Taiwan and the National Science Council of Taiwan.

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jprot.2013.06.012.

REFERENCES [1] Veto T, Price R, Silsby JF, Carter JA. Treatment of the first known case of king cobra envenomation in the United Kingdom, complicated by severe anaphylaxis. Anaesthesia 2007;62:75–8. [2] He YY, Lee WH, Zhang Y. Cloning and purification of alpha-neurotoxins from king cobra (Ophiophagus hannah). Toxicon 2004;44:295–303. [3] Li J, Zhang HY, Liu J, Xu KS. Novel genes encoding six kinds of three-finger toxins in Ophiophagus hannah (king cobra) and function characterization of two recombinant long-chain neurotoxins. Biochem J 2006;398:233–42. [4] Tan NH, Saifuddin MN. Purification and characterization of two acidic phospholipase A2 enzymes from king cobra (Ophiophagus hannah) snake venom. Int J Biochem 1990;22:481–7. [5] Chiou JY, Chang LS, Chen LN, Chang CC. Purification and characterization of a novel phospholipase A2 from king cobra (Ophiophagus hannah) venom. J Protein Chem 1995;14:451–6.

[6] Huang MZ, Gopalakrishnakone P, Chung MCM, Kini RM. Complete amino acid sequence of an acidic, cardiotoxic phospholipase A2 from the venom of Ophiophagus hannah (king cobra): a novel venom enzyme with “pancreatic loop”. Arch Biochem Biophys 1997;338:150–6. [7] Wang QY, Shu YY, Zhuang MX, Lin ZJ. Cloning and sequence analysis of cDNAs encoding two acidic PLA2 from venom of Ophiophagus hannah (king cobra), Guangxi species. Acta Biochim Biophys Sin 2001;33:340–4. [8] Jin Y, Lee WH, Zeng L, Zhang Y. Molecular characterization of L-amino acid oxidase from king cobra venom. Toxicon 2007;50:479–89. [9] Guo XX, Zeng L, Lee WH, Zhang Y, Jin Y. Isolation and cloning of a metalloproteinase from king cobra snake venom. Toxicon 2007;49:954–65. [10] He YY, Liu SB, Lee WH, Qian JQ, Zhang Y. Isolation, expression and characterization of a novel dual serine protease inhibitor, OH-TCI, from king cobra venom. Peptides 2008;29:1692–9. [11] Lee WH, Zhang Y, Wang WY, Xiong YL, Gao R. Isolation and properties of a blood coagulation factor X activator from the venom of king cobra (Ophiophagus hannah). Toxicon 1995;33: 1263–76. [12] Vonk FJ. Snake evolution and prospecting of snake venom. Ph.D. thesis Netherland: Leiden University; July, 2012 . [13] Tan NH, Hj MNS. Enzymatic and toxic properties of Ophiophagus hannah (king Cobra) venom and venom fractions. Toxicon 1989;27:689–95. [14] Bergmeyer HU. L-Amino acid oxidase. In: Bergmeyer HU, editor. Methods of enzymatic analysis. , Samples, reagents, assessment of resultsWeinheim: Verlag Chemie GmbH; 1983. p. 149–50. [15] Yap MKK, Tan NH, Fung SY. Biochemical and toxinological characterization of Naja sumatrana (Equatorial spitting cobra) venom. J Venom Anim Toxins 2011;17:451–9. [16] Cousin X, Bon S, Duval N, Massoulie J, Bon C. Cloning and expression of acetylcholinesterase from Bungarus fasciatus venom. J Biol Chem 1996;271:15099–108. [17] Perez-Llorens JL, Benitez E, Vergara JJ, Berges JA. Characterization of proteolytic enzyme activities in macroalgae. Eur J Phycol 2003;38:55–64. [18] Tsai IH, Chang HC, Chen JM, Cheng AC, Khoo KH. Glycan structures and intrageneric variations of venom acidic phospholipase A2 from Tropidolaemus pitvipers. FEBS J 2012;279:2672–82. [19] Yamazaki Y, Hyodo F, Morita T. Wide distribution of cysteine-rich secretory proteins in snake venoms: isolation and cloning of novel snake venom cysteine-rich secretory proteins. Arch Biochem Biophys 2003;412:133–41. [20] Zeng L, Sun QY, Jin Y, Zhang Y, Lee WH, Zhang Y. Molecular cloning and characterization of a complement-depleting factor from king cobra, Ophiophagus hannah. Toxicon 2012;60: 290–301. [21] Tan NH, Saifuddin MN. Isolation and characterization of an unusual form of L-amino acid oxidase from King cobra (Ophiophagus hannah) venom. Biochem Int 1989;19: 937–44. [22] Chen HS, Wang YM, Huang WT, Huang KF, Tsai IH. Cloning, characterization and mutagenesis of Russell's viper venom L-amino acid oxidase: insights into its catalytic mechanism. Biochimie 2012;94:335–44. [23] Gold BS, Pyle P. Successful treatment of neurotoxic king cobra envenomation in Myrtle Beach, South Carolina. Ann Emerg Med 1998;32:736–8. [24] Shiau TW, Hou PC, Wu SH, Tu MC. A Survey on alien pet reptiles in Taiwan. Taiwania 2006;51:71–80. [25] Goldberg SR. Reproduction in the Many-Lined Sun Skink, Eutropis multifasciata (Squamata: Scincidae) from Sarawak, Malaysia. Curr Herpetol 2013;32:61–5.

J O U RN A L OF P ROT EO M IC S 8 9 ( 2 01 3 ) 1 4 1 –1 53

[26] Pawlak J, Mackessy SP, Fry BG, Bhatia M, Mourier G, Fruchart-Gaillard C, et al. Denmotoxin: a three-finger toxin from colubrid snake Boiga dendrophila (mangrove catsnake) with bird-specific activity. J Biol Chem 2006;39:29030–41. [27] Lee CY. Toxicological studies on the venom of Vipera russelli formosensis Maki: Part I. Toxicity and pharmacological properties. J Formos Med Assoc 1948;47:65–98. [28] Lee CY, Chang CC, Su C, Chen YW. The toxicity and thermostability of Formosa snake venoms. J Formos Med Assoc 1962;61:239–44. [29] Weissenberg S, Ovadia M, Kochva E. Species specific sensitivity towards the hemorrhagin of Ophiophagus hannah (Elapidae). Toxicon 1987;25:475–81. [30] Dhananjaya BL, D'Souza CJ. The pharmacological role of phosphatases (acid and alkaline phosphomonoesterases) in snake venoms related to release of purines—a multitoxin. Basic Clin Pharmacol Toxicol 2011;108:79–83. [31] Frobert Y, Creminon C, Cousin X, Remy MH, Chatel JM, Bon S, et al. Acetylcholinesterases from Elapidae snake venoms: biochemical, immunological and enzymatic characterization. Biochim Biophys Acta 1997;1339:253–67. [32] Gowtham YJ, Kumar MS, Girish KS, Kemparaju K. Hemostatic interference of India king cobra (Ophiophagus hannah) venom: comparison with three other snake venoms of the subcontinent. Biochemistry (Mosc) 2012;77:639–47. [33] Berg OG, Gelb MH, Tsai MD, Jain MK. Interfacial enzymology: the secreted phospholipase A2-paradigm. Chem Rev 2001;101: 2613–53. [34] Huang MZ, Gopalakrishnakone P. Pathological changes induced by an acidic phospholipase A2 from Ophiophagus hannah venom on heart and skeletal muscle of mice after systemic injection. Toxicon 1996;34:201–11. [35] Xu SJ, Gu L, Wang QY, Shu YY, Song SY, Lin ZJ. Structure of a king cobra phospholipase A2 determined from a hemihedrally twinned crystal. Acta Crystallogr D Biol Crystallogr 2003;59: 1574–81. [36] Tsai HY, Wang YM, Tsai IH. Cloning, characterization and phylogenetic analyses of members of three major venom families from a single specimen of Walterinnesia aegyptia. Toxicon 2008;51:1245–54. [37] Oyama E, Takahashi H. Distribution of low molecular weight platelet aggregation inhibitors from snake venoms. Toxicon 2007;49:293–8. [38] Huang MZ, Gopalakrishnakone P, Kini RM. Role of enzymatic activity in the antiplatelet effects of a phospholipase A2 from Ophiophagus hannah snake venom. Life Sci 1997;61:2211–7. [39] Chang CC, Huang TY, Kuo KW, Chen SW, Huang KF, Chiou SH. Sequence characterization of a novel alpha-neurotoxin from the king cobra (Ophiophagus hannah) venom. Biochem Biophys Res Commun 1993;191:214–23.

153

[40] Finley EL, Dillon J, Crouch RK, Schey KL. Identification of tryptophan oxidation products in bovine alpha-crystallin. Protein Sci 1998;7:2391–7. [41] Pu XC, Wong PTH, Gopalakrishnakone P. A novel analgesic toxin (hannalgesin) from the venom of king cobra (Ophiophagus hannah). Toxicon 1995;33:1425–31. [42] Rajagopalan N, Pung YF, Zhu YZ, Wong PT, Kumar PP, Kini RM. Beta-cardiotoxin: a new three-finger toxin from Ophiophagus hannah (king cobra) venom with beta-blocker activity. FASEB J 2007;21:3685–95. [43] Chang LS, Liou JC, Lin SR, Huang HB. Purification and characterization of a neurotoxin from the venom of Ophiophagus hannah (king cobra). Biochem Biophys Res Commun 2002;294:574–8. [44] Huang MZ, Wang QC, Liu GF. Effects of an acidic phospholipase A2 purified from Ophiophagus hannah (king cobra) venom on rat-heart. Toxicon 1993;31:627–35. [45] Joubert FJ. Snake venom toxins the amino acid sequences of two toxins from Ophiophagus hannah (king cobra) venom. Biochim Biophys Acta 1973;317:85–98. [46] Servent D, Winckler-Dietrich V, Hu HY, Kessler P, Drevet P, Bertrand D, et al. Only snake curaremimetic toxins with a fifth disulfide bond have high affinity for the neuronal alpha-7 nicotinic receptor. J Biol Chem 1997;272: 24279–86. [47] Roy A, Shou X, Chong MZ, D'hoedt D, Foo CS, Rajagopalan N, et al. Structural and functional characterization of a novel homodimeric three-finger neurotoxin from the venom of Ophiophagus hannah (king cobra). J Biol Chem 2010;285: 8302–15. [48] Joubert FJ. The amino-acid sequence of polypeptide DE-1 from Ophiophagus hannah (king cobra) venom. Hoppe Seylers Z Physiol Chem 1977;358:565–74. [49] Fry BG, Wuster W, Kini RM, Brusic V, Khan A, Venkataraman D, et al. Molecular evolution and phylogeny of elapid snake venom three-finger toxins. J Mol Evol 2003;57:110–29. [50] Zhou H, Meng QX, Xie LP, Zhang RQ. A weak neurotoxin exhibits synergic effect with cardiotoxins when co-applied to two non-neural cell lines. Biol Pharm Bull 2004;27: 1241–4. [51] Chang LS, Chung C, Huang HB, Lin SR. Purification and characterization of a chymotrypsin inhibitor from the venom of Ophiophagus hannah (king Cobra). Biochem Biophys Res Commun 2001;283:862–7. [52] Fry BG, Scheib H, van der Weerd L, Young B, McNaughtan J, Ramjan SF, et al. Norman JA: evolution of an arsenal: structural and functional diversification of the venom system in the advanced snakes (Caenophidia). Mol Cell Proteomics 2008;7:215–46.