Venomics, lethality and Neutralization of Naja kaouthia (monocled cobra) venoms from three different geographical regions of Southeast Asia

Accepted Manuscript Venomics, lethality and Neutralization of Naja kaouthia (monocled cobra) venoms from three different geographical regions of Southeast Asia Kae Yi Tan, Choo Hock Tan, Shin Yee Fung, Nget Hong Tan PII: S1874-3919(15)00076-7 DOI: doi: 10.1016/j.jprot.2015.02.012 Reference: JPROT 2064 To appear in: Journal of Proteomics Received date: 22 January 2015 Accepted date: 24 February 2015 Please cite this article as: Tan Kae Yi, Tan Choo Hock, Fung Shin Yee, Tan Nget Hong, Venomics, lethality and Neutralization of Naja kaouthia (monocled cobra) venoms from three different geographical regions of Southeast Asia, Journal of Proteomics (2015), doi: 10.1016/j.jprot.2015.02.012 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Venomics, lethality and Neutralization of Naja kaouthia (monocled cobra) venoms from three different geographical regions of Southeast Asia Kae Yi Tan a, Choo Hock Tan b, *, Shin Yee Fung a, Nget Hong Tan a a Department of Molecular Medicine, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia b Department of Pharmacology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia * Corresponding author at: Department of Pharmacology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia E-email address: tanchoohock@gmail.com; tanch@um.edu.my Abstract Previous studies showed that venoms of the monocled cobra, Naja kaouthia from Thailand and Malaysia are substantially different in their median lethal doses. The intraspecific venom variations of N. kaouthia, however, have not been fully elucidated. Here we investigated the venom proteomes of N. kaouthia from Malaysia (NK-M), Thailand (NK-T) and Vietnam (NK-V) through reverse-phase HPLC, SDS-PAGE and tandem mass spectrometry. The venom proteins comprise 13 toxin families, with three-finger toxins being the most abundant (63-77%) and the most varied (11-18 isoforms) among the three populations. NK-T has the highest content of neurotoxins (50%, predominantly long neurotoxins), followed by NK-V (29%, predominantly weak neurotoxins and some short neurotoxins), while NK-M has the least (18%, some weak neurotoxins but less short and long neurotoxins). On the other hand, cytotoxins constitute the main bulk of toxins in NK-M and NK-V venoms (up to 45% each), but less in NK-T venom (27%). The three venoms show different lethal potencies that generally reflect the proteomic findings. Despite the proteomic variations, the use of Thai monovalent and Neuro polyvalent antivenoms for N. kaouthia envenomation in the three regions is appropriate as the different venoms were neutralized by the antivenoms albeit at different degrees of effectiveness. 1

Significance Biogeographical variations were observed in the venom proteome of monocled cobra (Naja kaouthia) from Malaysia, Thailand and Vietnam. The Thai N. kaouthia venom is particularly rich in long neurotoxins, while the Malaysian and Vietnamese specimens were predominated with cytotoxins. The differentially expressed toxin profile accounts for the discrepancy in the lethal dose of the venom from different populations. Commercially available Thai antivenoms (monovalent and polyvalent) were able to neutralize the three venoms at different effective doses, hence supporting their uses in the three regions. While dose adjustment according to geographical region seems possible, changes to standard recommended dosage should only be made if further study validates that the monocled cobras within a population do not exhibit remarkable inter-individual venom variation. Keywords: Naja kaouthia; monocled cobra; venom proteome; three-finger toxins; geographical variation; antivenom neutralization * Correspondence to: Choo Hock Tan, Department of Pharmacology, Faculty of Medicine, University of Malaya, 50603 Kuala Lumpur, Malaysia 2

1 Introduction Snakebite envenomation remains as a neglected tropical disease and a disease of poverty [1-3]. It poses serious public health threats in many parts of the world, particularly developing and under-developed countries in the tropics and subtropics [4-6]. Approximately 5.5 million snakebite cases occur annually, resulting in close to 2 million envenomations worldwide with close to a hundred thousand fatalities [7], besides an unknown percentage of survivors who continue to suffer permanent physical disability due to local tissue destruction. In Southeast Asia, snake envenomation affect not only rural and agricultural populations, but also those living near the cities as humans are encroaching into the habitats of snakes following rapid urbanization [8]. Among the many venomous snakes in this region, cobra (Naja sp.) is one of the most common biters capable of delivering large quantity of lethal venom hence it is classified as a Category I medically dangerous snake by WHO [9]. Various species of cobras inhabit different parts of Asia, Middle East and Africa; among these, the monocled cobra (Naja kaouthia) is a widely distributed species in the Indochina subcontinent and the northern Malayan Peninsula, as well as north-eastern India and southern China. There is a unique O -shaped mark on the hood that easily distinguishes it from the other spectacled cobras [10, 11]. This species adapts well to a range of habitats from natural to anthropogenically impacted environments. In recent years, it has also been found not limited to the northern region of Malayan Peninsula. N. kaouthia envenoming is capable of causing rapid onset of neuromuscular paralysis in victims, and delayed or inadequate treatment can lead to respiratory failure and death [12]. Other accompanying toxic effects of N. kaouthia envenomation include extensive tissue necrosis that often results in crippling disability, adding to the toll of sufferings by the victim s family. Snake venoms consist mainly of proteins and peptides that exhibit diverse biochemical and pharmacological activities [13]. Venom represents a trophic adaptive trait, and is unique between species. The complexity of venom develops through a series of evolutionary events that include repeated gene duplication and molecular adaptation, leading to protein neofunctionalization to suit the toxin roles in predation, digestion and defence [14, 15]. The major contents and biochemical activities of venoms from phylogenetically closely related species generally share a similar pattern; for example the predominance of muscleparalysing neurotoxins in the venoms of most elapid snakes. Nevertheless, venom 3

composition (toxin subtypes and relative abundances) can vary remarkably between congeneric or even intraspecific species as a result of differences in their ecological niche and the consequent genetic adaptation [13]. The implication of this phenomenon is medically relevant, as diverse toxin composition can lead to varied envenoming effects and treatment outcome [16]. It is known that in Southeast Asia, many countries depend on antivenom supply from non-domestic manufacturers that use immunogens from species non-native to the importing countries. This poses a question of how appropriate or effective the antivenoms are for heterologous or non-native species, considering the various reports on geographical venom variations [17, 18]. Since antivenom is the only definitive treatment for snake envenomation, essentially, the effectiveness of venom neutralization relies on the molecular characteristics and antigenic determinants of the venom toxins [13]. Considerable compositional, syndromic and immunological variations have been reported for the venoms of several cobra taxa, including those which are sympatric [19, 20]. Earlier, a preclinical assessment has indicated that the venoms of N. kaouthia from Malaysia and Thailand exhibited substantially different degree of lethality (LD 50 ) and response to antivenom neutralization [21, 22], indicating the occurrence of geographical variation in the toxin composition of this wide-ranging species. Differences in the venom neurotoxins between Thai and Chinese N. kaouthia were reported previously [23], and 5 toxin families have been shown present in the Thai N. kaouthia venom (using 2D electrophoresis) [24], however, to date, there has no in-depth study on the biogeographical variations of N. kaouthia venom, especially where details of subtype composition and relative abundance are concerned. For years, meticulous profiling of venom toxins was challenging as venoms are complex mixtures of proteins, and the subtle yet important variability resulted from molecular evolution makes comparison of venom composition between species or population a difficult task. Nonetheless, this has been greatly overcome by recent breakthroughs in - omics technologies that are increasingly incorporated in proteomic research of snake venom. The combined use of high performance liquid chromatography with high resolution mass spectrometry and powerful data mining programme has enabled toxinologists to gain deeper insights into the compositional variation of venom toxins, hence improving the understanding of their biodiversity and medical importance, particularly on pathogenesis, treatment optimization and for drug discovery [25-27]. In the present study, we applied the proteomic approach to investigate the geographical variations of N. kaouthia venoms sourced from 4

Malaysia, Thailand and Vietnam, where the countries share similar concerns of having this wide-ranging species as a source of envenomation. In addition, the venoms lethal toxicities and therapeutic responses to two cobra antivenoms (monovalent and polyvalent antivenoms) were also evaluated and correlated to the proteomic findings. 2 Materials and Methods 2.1 Venoms and antivenoms Venom of Malaysian Naja kaouthia (NK-T, identified by author CHT) was collected from specimens in the northern region of the Malayan Peninsula. Venoms of Thai (NK-T) and Vietnam (NK-V) were gifts from Professor Kavi Ratanabangkoon of the Chulabhorn Graduate Institute, Bangkok. All venoms were pooled samples and were lyophilized products stored at -20 C until use. The two antivenoms used in the study are products of Queen Saovabha Memorial Institute (QSMI), Thai Red Cross Society from Bangkok, Thailand: (a) N. kaouthia monovalent antivenom (NKMAV; Cobra antivenin; Lyophilized; Batch no.0080210; Exp. Date Aug 9 th, 2015), a purified F(ab ) 2 obtained from equine serum hyperimmunized specifically against the venom of N. kaouthia (Thai monocled cobra); (b) Neuro polyvalent antivenom (NPAV; Lyophilized; Batch no.0020208; Exp. Date Oct 5 th, 2014), a purified F(ab ) 2 obtained from equine serum hyperimmunized against a mixture of four venoms: N. kaouthia (Thai monocled cobra), Ophiophagus hannah (king cobra), Bungarus candidus (Malayan krait) and Bungarus fasciatus (banded krait). The antivenoms were reconstituted according to the attached product leaflet: 10 ml of normal saline was added to one vial of the lyophilized antivenom. According to the product leaflet, 1 ml of NKMAV antivenom is able to neutralise 0.6 mg of N. kaouthia venom; while 1 ml of NPAV antivenom can neutralise the following amount of snake venoms: 0.6 mg each of N. kaouthia and B. fasciatus venoms, 0.4 mg of B. candidus venom and 0.8 mg of O. hannah venom. 2.2 Animals Albino mice (ICR strain, 20-30 g) were supplied by the Animal Experimental Unit, Faculty of Medicine, University of Malaya. The animals were handled according to the guideline given by CIOMS on animal experimentation [28]. All animal experiments were approved by 5

the International Animal Care and Use Committee (IACUC) of University of Malaya (Ethical clearance letter No. 2013-06-07/MOL/R/FSY). 2.3 Chemicals and materials All chemicals and reagents used in the studies were analytical grade. Ammonium bicarbonate, dithiothreitol (DTT) and iodoacetamide were purchased from Sigma-Aldrich (USA). MS grade trypsin protease, Spectra Multicolor Broad Range Protein Ladder (10 to 260 kda), and HPLC grade solvents used in the studies were purchased from Thermo Scientific Pierce (USA). LiChroCART 250-4 LiChrospher WP 300 RP-18 (5 µm) HPLC cartridge and Milipore ZipTip C 18 Pipette Tips were purchased from Merck (USA). 2.4 C 18 reverse-phase HPLC fractionation Crude venoms (3 mg) were reconstituted in ultrapure water and centrifuged at 10,000 g for 5 min. The supernatants were subjected to LiChrospher WP 300 C 18 reverse-phase column (5 µm) using a Shimadzu LC-20AD HPLC system (Japan). The venom components were eluted at 1 ml/min with a linear gradient of 0.1% TFA in water (Solvent A) and 0.1% TFA in 100% ACN (Solvent B) (0-5% B for 10 min, followed by 5-15% B over 20 min, 15-45% B over 120 min and 45-70% B over 20 min) [29]. Protein was detected at 215 nm and peaks were collected manually and lyophilized. 2.5 SDS-PAGE and tryptic digestion of protein The concentrated fractions obtained from reverse-phase HPLC were reconstituted in ultrapure water and subjected to 15% sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) according to the procedure of Laemmli (1970) [30]. Spectra Multicolor Broad Range Protein Ladder (10 to 260 kda) (Thermo Scientific Pierce, USA) was used for molecular weight calibration. Protein bands of interest were excised from Commassie Brilliant Blue-stained SDS-PAGE gel and subjected to the standard procedure of reduction with DTT, alkylation with iodoacetamide, and in-gel digestion with MS grade Pierce trypsin protease according to manufacturer s protocol (Thermo Scientific Pierce, Rockford, IL, 6

USA). The tryptic digested peptides were desalted using standard bed Milipore ZipTip C 18 Pipette Tips (Merck, USA) according to manufacturer s protocol to enhance the performance of mass spectrometry. 2.6 High resolution mass spectrometry analysis The tryptic digested peptide samples (0.5 µl) and 0.5 µl α-cyano-4-hydroxycinnamic acid matrix was mixed and spotted on OPTI-TOF LC/MALDI insert plate (123 mm x 81 mm). MALDI-TOF/TOF was performed using AB SCIEX 5800 Plus Analyzer equipped with a neodymium: yttrium-aluminium-garnet laser (laser wavelength was 349 nm). The TOF/TOF calibration mixtures (AB SCIEX, USA) were used to calibrate the spectrum to a mass tolerance within 10 ppm. For MS mode, peptides mass maps were acquired in positive reflection mode and 800-4000 m/z mass range were used with 100 laser shots per spectrum. The MS/MS peak detection criteria used were a minimum signal-to-noise (S/N) of 100. The raw mass spectra acquired were exported to AB SCIEX ProteinPilot Software search against all non-redundant NCBI Serpentes database (taxid: 8570) (updated August, 2014). MS peak filter mass range 800-4000 m/z was applied. Precursor and fragment mass tolerances were set to 100 ppm and 0.2 Da respectively and allowing one missed cleavage. Oxidation (M) was set as a variable modification and carbamidomethylation (C) was set as a fixed modification. The protein score intervals (C.I.) above 95% were considered as confident identification. The unidentified samples from MALDI-TOF/TOF were subjected to Thermo Scientific Pierce Orbitrap Fusion Tribrid with an Easy-nLC 1000 ultra-high pressure LC on a Thermo Scientific Pierce EASY-Spray PepMap C 18 column (2 µm, 75 µm x 25 cm). The peptides were separated over 44 min gradient eluted at 300 nl/ml with 0.1% FA in water (Solvent A) and 0.1% FA in 100% ACN (Solvent B) (0-5% B in 5 min, followed by 5-50% B over 30 min and 50-95% B over 2 min). The run finished by holding a 95% B for 7 min. MS1 data was acquired on a Orbitrap Fusion mass spectrometry using full scan method according to the following parameters: scan range 300-2000 m/z, orbitrap resolution 240,000; AGC target 200,000; maximum injection time of 50 ms. MS2 data were collected using the following parameters: rapid scan rate, CID collision energy 30%, 2 m/z isolation window, AGC 10000 and maximum injection time of 35 ms. MS2 precursors were selected for a 3 sec cycle. The precursors with an assigned monoisotopic m/z and a charge 7

state of 2-6 were interrogated. The precursors were filtered using a 70 sec dynamic exclusion window. MS/MS spectra were searched using Thermo Scientific Pierce Proteome Discoverer Software Version 1.4 with SEQUEST against the non-redundant NCBI Serpentes database (taxid: 8570) (updated August, 2014) downloaded from NCBI database (http://www.ncbi.nlm.nih.gov/protein/?term=serpentes). Precursor and fragment mass tolerances were set to 10 ppm and 0.8 Da respectively and allowing up to two missed cleavages. Static modification used was carbamidomethylation (C). The high confidence level filter with false discovery rate (FDR) of 1% was applied to the peptides and the protein ID with highest protein score > 10 was considered as confident identification. 2.7 Protein relative abundance The protein relative abundance was estimated according to the method used in previous venomic studies [20, 25]. The relative abundance of venom fractions were estimated by peak area measurement using Shimazu LCsolution Software (Japan). The fractions showing a protein band in SDS-PAGE were directly implemented with the relative abundance obtained from peak area measurement; while the fractions with various protein bands in SDS-PAGE were estimated by densitometry using Thermo Scientific Pierce myimageanalysis Software (USA). The relative abundances in percentage were then accumulated according to the protein family. 2.8 Determination of medium lethal dose (LD 50 ) of N. kaouthia venoms The median lethal dose (LD 50 ) of the venom was determined by intravenous (i.v., via caudal vein) and subcutaneous (s.c., under the loose skin over the neck) routes of administration into ICR mice (20-25 g, n=4). The survival ratio was recorded after 48 h and LD 50 was calculated using the Probit analysis method [31]. 2.9 In vitro immunocomplexation and in vivo protective effect of antivenoms against lethality of N. kaouthia venoms 8

In vitro immunocomplexation was conducted as described in Ramos-Cerrillo et al. (2008) [32]. A challenge dose of 5 LD 50 of the venom dissolved in 50µL normal saline was preincubated at 37º C for 30 min with various dilutions of reconstituted antivenom (Naja kaouthia monovalent antivenom, NKMAV or Neuro polyvalent antivenom, NPAV) in normal saline to give a total volume of 250 µl. The venom-antivenom mixture was subsequently centrifuged at 10,000 g for 10 min and the supernatant was injected intravenously (i.v.) into the caudal vein of the mice model (20-25 g, n = 4). In vivo protective experiment was carried out with independent administration of venom and antivenom by subcutaneous (s.c.) injection of 5 LD 50 of the venom into mice model (20-25g, n=4), followed by intravenous (i.v.) injection of 200 µl appropriately diluted reconstituted antivenom (NKMAV or NPAV) 10 min post-envenomation. Similarly, the dilutions of injected antivenoms were moved one fold higher or lower from the dosage tested until a series of dosages showing a full survival to without survival were obtained. The survival ratio was recorded after 48 hours for both experiments and the neutralizing potency was calculated using the Probit analysis method [31]. Neutralizing potency of the antivenom was expressed as ED 50 (amount of reconstituted antivenom in µl), ER 50 (ratio of mg/ml reconstituted antivenom that gives 50% survival of animal) and Potency, P (the amount of venom that is completely neutralised by unit volume of antivenom) as decscribed by Morais et al. (2010) [33]. 9

3 Results and Discussions 3.1 Proteomic characterization of N. kaouthia venoms from Malaysia, Thailand and Vietnam Fractionations of the N. kaouthia venoms from Malaysia (NK-M), Thailand (NK-T) and Vietnam (NK-V) for proteomic characterization were carried out using reverse-phase HPLC. Separation by reverse-phase HPLC of the venoms yielded 28 (NK-M), 28 (NK-T) and 29 (NK-V) fractions, respectively. The chromatograms revealed distinct differences in the composition among the three venoms, notably the prominent fraction 6 of NK-T (75-80 min) and fractions 17-19 of NK-V (105-125 min) (Figure 1). SDS-PAGE of the obtained fractions indicated that low molecular mass proteins (5-15 kda) were the most abundant and were eluted over the initial 125 min of the chromatographic fractionation. This was followed by the moderate to high molecular mass proteins (20-150 kda). The fractions generally contained one to several proteins (Coomasie-blue stained) each. The gels were subsequently processed for tryptic digestion and peptide identification using high resolution mass spectrometry. A total of 61 (NK-M), 51 (NK-T) and 68 (NK-V) proteins were identified from each of these venoms, using the combination of RP-HPLC and SDS-PAGE (Figure 1; Table 1(a); Supplementary Table 1(b), Table 1(c)). The large number (51-68) of proteins identified was clustered to 16 protein families. Among these, 13 different toxin families were identified: three-finger toxin (3FTx), phospholipase A 2 (PLA 2 ), cysteine-rich secretory protein (CRISP), snake venom metalloproteinase (SVMP), L-amino acid oxidase (LAAO), cobra venom factor (CVF), Kunitz-type inhibitor (KUN), natriuretic peptide (NP), phosphodiesterase (PDE), 5 nucleotidase (5 NUC), vespryn, c-type lectin (CTL) and nerve growth factor (NGF) (Figure 2; Table 1(a); Supplementary Table 1(b) and 1(c)). Very minute amount of three types of non-toxin proteins i.e. glutathione peroxidases (GPX), endonuclease domain-containing protein (EDCP) and tissue-type plasminogen activator (TTPA), were also detected, however, their roles in the venom are unclear. The occurrence of multiple toxin isoforms within a toxin family in these venoms reflects the redundancy of toxin genes that evolved from the recruitment of physiological proteins into a toxin arsenal of venom during the diversification of advanced snakes [34, 35]. This underlies the basis of venom toxin complexity despite the similarity of protein families across many species. 10

Table 1(a) - Assignment of the reverse-phase isolated fractions of N. kaouthia (Malaysia) venom to protein families by MALDI-TOF/TOF and Orbitrap Fusion analysis of selected peptides ions from in-gel digested proteins. Protein fraction % Protein family/subtype MS/MS derived sequence Matched MH+ Mass Δ (ppm) Accession/Species Matched peptide Protein score 1 1.0 3FTx-SNTX (cobrotoxin)* LECHNQQSSQTPTTTGCSGGETNCYK 2945.2082-0.45 P60771 (N. kaouthia) 1 51 2 1.6 3FTx-SNTX (cobrotoxin-c) LECHNQQSSQAPTTK 1727.7897 41 P59276 (N. kaouthia) 2 64 KWWSDHR 1013.4831 51 3a 1.0 3FTx-SNTX (cobrotoxin)* LECHNQQSSQTPTTTGCSGGETNCYK 2945.2002-3.18 P60771 (N. kaouthia) 4 452 LECHNQQSSQTPTTTGCSGGETNCYKK 3073.2905-4.56 NGIEINCCTTDR 1452.6174-2.27 NGIEINCCTTDRCNN 1840.7310-3.35 3b 0.3 3FTx-SNTX (cobrotoxin-b)* LECHNQQSSQTPTTK 1758.8016-3.40 P59275 (N. kaouthia) 5 62 VKPGVNLNCCR 1316.6531-2.41 TCSGETNCYKK 1347.5627-3.06 KWWSDHR 1014.4858-4.64 TCSGETNCYK 1219.4677-3.39 4 < 0.1 3FTx-SNTX (cobrotoxin)* LECHNQQSSQTPTTTGCSGGETNCYK 2945.1895-6.79 P60771 (N. kaouthia) 2 11 NGIEINCCTTDR 1452.6127-5.46 5a 0.2 3FTx-SNTX (cobrotoxin)* LECHNQQSSQTPTTTGCSGGETNCYK 2945.1943-5.17 P60771 (N. kaouthia) 4 91 LECHNQQSSQTPTTTGCSGGETNCYKK 3073.2916-4.20 LECHNQQSSQTPTTTGCSGGETNCYKKR 3229.3891-5.13 NGIEINCCTTDR 1452.6143-4.37 5b 0.2 Kunitz-type inhibitor FIYGGCGGNANR 1284.5669 26 P20229 (N. naja) 1 99 6 0.7 3FTx-WTX (weak neurotoxin 6)* LTCLICPEKYCNK 1698.7934-4.63 P29180 (N. naja) 9 114 YCNKVHTCLNGEK 1622.7345-4.24 LTCLICPEK 1133.5643-4.52 YIRGCADTCPVR 1467.6771-4.15 GCADTCPVR 1035.4301-4.42 YIRGCADTCPVRKPR 1848.9234-4.69 GCADTCPVRKPR 1416.6775-4.24 EIVQCCSTDK 1239.5302-3.43 KLLGKR 714.4953-4.51 7 0.6 3FTx-LNTX (alpha-elapitoxin- TGVDIQCCSTDNCNPFPTR 2241.9258-2.90 P01391 (N. kaouthia) 10 215 Nk2a)* VDLGCAATCPTVK 1391.6609-3.51 RVDLGCAATCPTVK 1547.7612-3.74 GKRVDLGCAATCPTVK 1732.8781-3.05 TWCDAFCSIR 1315.5519-2.99 IRCFITPDITSK 1450.7675-3.32 TGVDIQCCSTDNCNPFPTRK 2370.0203-2.92 CFITPDITSK 1181.5833-3.25 CFITPDITSKDCPNGHVCYTK 2513.1152-4.24 8a 0.4 3FTx-LNTX (alpha-elapitoxin- Nk2a) DCPNGHVCYTK 1350.5539-1.99 TWCDAFCSIR 1314.5485 25 P01391 (N. kaouthia) 3 206 RVDLGCAATCPTVK 1546.7596 20 TGVDIQCCSTDNCNPFPTR 2240.9249 24 8b 0.3 Kunitz-type inhibitor* TIDECNR 907.3939 0.07 P20229 (N. naja) 2 40 TIDECNRTCVG 1324.5622 0.08 9a 0.1 Not determined - - - - - - 9b 0.5 3FTx-MTLP (muscarinic toxinlike protein 2) 9c 0.1 3FTx-LNTX (alpha-elapitoxin- Nk2a - deduced 8a) 10 2.9 3FTx-LNTX (alpha-elapitoxin- Nk2a)* SIFGVTTEDCPDGQNLCFK 2186.9612-74 P82463 (N. kaouthia) 1 76 - - - - - - TGVDIQCCSTDNCNPFPTR 2241.9268-2.43 P01391 (N. kaouthia) 4 42 VDLGCAATCPTVK 1391.6630-2.02 TWCDAFCSIR 1315.5535-1.78 RVDLGCAATCPTVK 1547.7643-1.69 11a 0.2 vngf GIDSSHWNSYCTETDTFIK 2259.9743-4 P61899 (N. kaouthia) 3 142 ALTMEGNQASWR 1362.6350 3 ALTMEGNQASWR 1378.6299 3 11b 0.3 3FTx-MTLP (muscarinic toxin-like protein 3) TSETTEICPDSWYFCYK 2185.8972-69 P82464 (N. kaouthia) 2 119 ISLADGNDVR 1058.5356-69 11c 0.1 3FTx-CTX (cytotoxin 2)* GCIDVCPK 948.4239-4.14 P01445 (N. kaouthia) 6 26 YVCCNTDR 1087.4262-3.10 MFMVSNK 856.4023-3.83 NSLLVK 673.4217-3.94 NLCYK 697.3309-4.11 TVPVKR 699.4482-4.37 12a 8.4 3FTx-WTX (weak toxin CM-9a)* GCADTCPVGYPKEMIECCSTDK 2578.0270-4.57 P25679 (N. kaouthia) 8 620 LTCLNCPEMFCGK 1629.6835-3.54 GCADTCPVGYPKEMIECCSTDKCNR 3008.2066-2.30 LTCLNCPEMFCGKFQICR 2334.0221-4.28 YIRGCADTCPVGYPKEMIECCSTDK 3010.2753-3.99 GCADTCPVGYPK 1324.5618-3.23 KLHQR 681.4125-4.45 NGEKICFK 995.4935-4.49 12b 1.5 3FTx-CTX (cytotoxin 2)* YVCCNTDRCN 1361.4989-3.09 P01445 (N. kaouthia) 6 106 GCIDVCPK 948.4245-3.49 YVCCNTDR 1087.4261-3.22 MFMVSNK 856.4043-1.55 NSLLVK 673.4221-3.30 11

RGCIDVCPK 1104.5276-1.16 13a 0.7 PLA 2 (acidic 2) SWWDFADYGCYCGR 1784.6711-48 Q91133 (N. atra) 3 128 SWWDFADYGCYCGR 1841.6926-44 LAAICFAGAPYNNNNYNIDLK 2355.1317-42 13b 0.6 PLA 2 (acidic 2) SWWDFADYGCYCGR 1841.6926-52 Q91133 (N. atra) 1 74 13c 12.1 PLA 2 (acidic 1) NMIQCTVPNR 1247.5751-2 P00596 (N. kaouthia) 2 171 SWWDFADYGCYCGR 1841.6926 15 13d 4.8 3FTx-CTX (cytotoxin 2)* YVCCNTDRCN 1361.4988-3.18 P01445 (N. kaouthia) 5 41 GCIDVCPK 948.4237-4.33 YVCCNTDR 1087.4257-3.55 NSLLVK 673.4218-3.76 MFMVSNK 856.4019-4.33 13e 13.8 3FTx-CTX (cytotoxin 2) MFMVSNK 887.3881 15 P01445 (N. kaouthia) 3 124 GCIDVCPK 947.4205 31 YVCCNTDR 1086.4223 20 14a 0.7 PLA 2 (acidic 2 - deduced 13a) - - - - - - 14b 6.7 PLA 2 (acidic 2) SWWDFADYGCYCGR 1841.6926 26 P15445 (N. naja) 3 274 ISGCWPYFK 1156.5375 15 LAAICFAGAPYNDNNYNIDLK 2356.1157 24 14c 5.4 3FTx-CTX (cytotoxin NK-CT1) LVPLFYKTCPAGK 1492.8112-44 P0CH80 (N. kaouthia) 3 110 NSLVLKYVCCNTDR 1740.8287 11 YVCCNTDR 1086.4223 14 15a 0.7 PLA 2 (acidic 1) SWWDFADYGCYCGR 1784.6711-75 P00598 (N. atra) 2 126 SWWDFADYGCYCGR 1841.6926-79 15b 1.1 3FTx-CTX (cytotoxin homolog)* LKCHNTQLPFIYK 1661.8777-3.37 P14541 (N. kaouthia) 12 894 CHNTQLPFIYK 1420.7006-2.54 KFPLKIPIK 1083.7261-2.62 NSALLKYVCCSTDKCN 1932.8578-1.79 NSALLKYVCCSTDK 1658.7800-4.67 GCADNCPKNSALLK 1547.7236-4.52 YVCCSTDKCN 1306.4825-2.78 FPLKIPIK 955.6303-3.87 FPLKIPIKR 1111.7322-2.56 NLCFKATLK 1094.5998-2.69 YVCCSTDK 1032.4097-2.72 NSALLK 645.3910-3.18 15c 0.5 3FTx-CTX (cytotoxin 1)* NSLLVKYVCCNTDRCN 2015.9048-2.42 P60305 (N. kaouthia) 11 304 YVCCNTDRCN 1361.5004-2.01 MFMMSDLTIPVK 1412.6967-1.43 GCIDVCPKNSLLVK 1602.8310-2.06 GCIDVCPK 948.4257-2.21 MFMMSDLTIPVKR 1568.7958-2.57 CNKLIPIASK 1143.6529-2.29 NSLLVKYVCCNTDR 1741.8333-1.58 YVCCNTDR 1087.4269-2.43 NSLLVK 673.4229-2.13 RGCIDVCPK 1104.5252-3.37 16a 1.1 PLA 2 (acidic 2) SWWDFADYGCYCGR 1841.6926-75 Q91133 (N. atra) 1 118 16b 0.9 PLA 2 (acidic 1)* TYSYECSQGTLTCK 1697.7239 5.50 P00596 (N. kaouthia) 5 52 GDNDACAAAVCDCDR 1669.6087 5.19 CCQVHDNCYNEAEK 1826.7008 6.38 NMIQCTVPNR 1232.5931 4.52 GGSGTPVDDLDR 1188.5552 5.08 16c 10.1 3FTx-CTX (cytotoxin 1)* NSLLVKYVCCNTDR 1741.8315-2.64 P60305 (N. kaouthia) 15 920 MFMMSDLTIPVKR 1568.7937-3.88 NLCYKMFMMSDLTIPVKR 2247.1058-4.44 NSLLVKYVCCNTDRCN 2015.9055-2.06 MFMMSDLTIPVK 1412.6954-2.38 LKCNKLIPIASK 1384.8306-2.86 YVCCNTDRCN 1361.4999-2.37 LIPIASKTCPAGK 1355.7675-3.03 RGCIDVCPKNSLLVK 1758.9305-2.78 GCIDVCPKNSLLVK 1602.8315-1.76 GCIDVCPK 948.4253-2.59 YVCCNTDR 1087.4269-2.43 NSLLVK 673.4223-3.03 RGCIDVCPK 1104.5259-2.71 17 8.4 3FTx-CTX (cardiotoxin 2A precursor)* CNKLIPIASK 1143.6524-2.71 YVCCNTDRCN 1361.5037 0.41 Q9PST4 (N. kaouthia) 14 792 CNKLVPLFYK 1281.7033 0.67 LVPLFYKTCPAGK 1493.8218 2.20 MYMVATPK 940.4652 2.21 SSLLVKYVCCNTDRCN 1988.8976-0.57 NLCYKMYMVATPK 1618.7793 0.17 GCIDVCPK 948.4280 0.17 SSLLVKYVCCNTDR 1714.8259 0.43 GCIDVCPKSSLLVKYVCCNTDR 2644.2363 0.45 RGCIDVCPK 1104.5294 0.46 SSLLVK 646.4131-0.56 YVCCNTDR 1087.4296 0.04 VPVKRGCIDVCPK 1527.8155 1.29 GCIDVCPKSSLLVK 1575.8250 1.02 18a 0.3 Vespryn (Thaicobrin) FDGSPCVLGSPGFR 1437.6710-31 P82885 (N. kaouthia) 2 111 FDGSPCVLGSPGFR 1494.6925-41 18b 0.1 3FTx-CTX (cytotoxin 2)* YVCCNTDR 1087.4312 1.50 Q98965 (N. kaouthia) 1 18 12

GCIDVCPK 948.4288 1.08 SSLLVK 646.4139 0.77 FPVKR 646.4040 0.77 19a 0.4 SVMP (atrase-a) EHQEYLLR 1086.5458 19 D5LMJ3 (N. atra) 2 118 ERPQCILNKPSR 1496.7881 19 19b 0.1 Not determined - - - - - - 19c 0.3 CRISP (natrin-1) MEWYPEAASNAER 1552.6616-35 Q7T1K6 (N. atra) 3 99 VLEGIQCGESIYMSSNAR 2012.9295-33 TWTEIIHLWHDEYK 1869.9050-36 19d 0.6 CRISP (natrin-2) IGCGENLFMSSQPYAWSR 2117.9298 5 Q7ZZN8 (N. atra) 1 70 20a 0.5 CRISP (natrin-1) NVDFNSESTR 1167.5156 32 Q7T1K6 (N. atra) 9 349 QKEIVDLHNSLR 1450.7892 30 EIVDLHNSLR 1194.6357 31 MEWYPEAASNAER 1552.6616 29 MEWYPEAASNAER 1568.6565 29 WANTCSLNHSPDNLR 1783.8060 30 VLEGIQCGESIYMSSNAR 2012.9295 30 VLEGIQCGESIYMSSNAR 2028.9245 33 SNCPASCFCR 1257.4689 30 20b 2.9 CRISP (natrin-1) QKEIVDLHNSLR 1450.7892 28 Q7T1K6 (N. atra) 8 609 EIVDLHNSLR 1194.6357 28 MEWYPEAASNAER 1552.6616 43 MEWYPEAASNAER 1568.6565 37 WANTCSLNHSPDNLR 1783.8060 36 VLEGIQCGESIYMSSNAR 2012.9295 39 VLEGIQCGESIYMSSNAR 2028.9244 34 SNCPASCFCR 1257.4689 39 20c 0.2 Not determined - - - - - - 20d 0.1 Not determined - - - - - - 20e 0.2 CTL (BFL-1) KYIWEWTDR 1295.6299 44 Q90WI8 (B. fasciatus) 2 138 YIWEWTDR 1167.5349 35 21 < 0.1 EDCP* GHLNPNGHQPDYSAK 1634.7662-0.54 U3FCT9 (M. fulvius) 2 27 NDQNVVQK 944.4786-1.12 23 1.0 SVMP (kaouthiagin) YIEFYVIVDNR 1429.7241-38 P82942 (N. kaouthia) 1 130 24a 0.4 PDE* NLHNCVNLILLADHGMEAISCNR 2664.2809 0.16 U3FAB3 (M. fulvius) 22 392 MANVLCSCSEDCLTK 1787.7406-1.41 AEYLETWDTLMPNINK 1937.9311-0.18 RPDFSTLYIEEPDTTGHK 2106.0128-0.53 YCSGGTHGYDNEFK 1634.6545 0.26 LWNYFHSTLLPK 1518.8096-0.52 DFYTFDSEAIVK 1434.6783-0.36 CSSITDLEAVNQR 1492.7065 0.25 AKRPDFSTLYIEEPDTTGHK 2305.1422-1.65 VLSFILPHRPDNSESCADK 2185.0704-0.15 IDKVNLMVDR 1202.6554-0.66 AATYFWPGSEVK 1355.6624-0.50 NPFYNPSPAK 1134.5579 0.03 YCLLHQTK 1062.5396-0.47 TLGMLMEGLK 1092.5785-0.64 YISAYSQDILMPLWNSYTISK 2493.2374 0.10 AYLAKDLPK 1018.5914-1.75 NVPKDFYTFDSEAIVK 1872.9366-0.71 VNLMVDR 846.4493-1.09 DCCTDYK 961.3382-0.81 YGPVSGQVIK 1047.5832-0.18 SLQMADR 820.3973-1.15 24b 0.1 Not determined - - - - - - 24c 0.1 Not determined - - - - - - 24d 0.2 SVMP (kaouthiagin) TAPAFQFSSCSIR 1470.6926 38 P82942 (N. kaouthia) 3 140 DYQEYLLR 1098.5345 40 GCFDLNMR 1011.4266 39 24e < 0.1 5'-NUC ETPVLSNPGPYLEFR 1717.8675 24 B6EWW8 2 64 ETPVLSNPGPYLEFRDEVEELQK 2688.3282 23 (G. brevicaudus) 24f 0.2 GPX* AKVDCYDSVK 1184.5562-4.58 V8P395 (O. hannah) 4 40 VDCYDSVK 985.4254-4.22 LVILGFPCNQFGK 1492.7928-3.61 FLVNPQGKPVMR 1385.7668-3.98 25a 0.1 Not determined - - - - - - 25b 0.1 CVF FFYIDGNENFHVSITAR 2028.9693-24 Q91132 (N. kaouthia) 2 77 FVAYYQVGNNEIVADSVWVDVK 2514.2430-26 25c 0.3 5'-NUC ETPVLSNPGPYLEFR 1717.8675 17 B6EWW8 4 124 ETPVLSNPGPYLEFRDEVEELQK 2688.3282 15 (G. brevicaudus) FHECNLGNLICDAVIYNNVR 2420.1365 14 VVSLNVLCTECR 1448.7116 18 25d 0.3 CVF INYENALLAR 1175.6298 14 Q91132 (N. kaouthia) 6 341 HFEVGFIQPGSVK 1443.7511 10 CAGETCSSLNHQER 1647.6729 6 IDVPLQIEK 1053.6070 16 THQYISQR 1031.5148 18 VNDDYLIWGSR 1336.6412 14 25e 0.1 CVF DSITTWVVLAVSFTPTK 1863.9982-23 Q91132 (N. kaouthia) 4 161 GICVAEPYEIR 1305.6387-53 AVPFVIVPLEQGLHDVEIK 2102.1775-17 ASVQEALWSDGVR 1416.6997-38 25f 0.1 CVF GICVAEPYEIR 1305.6387 13 Q91132 (N. kaouthia) 3 206 AILHNYVNEDIYVR 1717.8787 12 13

ASVQEALWSDGVR 1416.6997 13 25g 0.2 Not determined - - - - - - 26a 0.1 CVF QLDIFVHDFPR 1385.7092 9 Q91132 (N. kaouthia) 8 343 QNQYVVVQVTGPQVR 1713.9162 8 GIYTPGSPVLYR 1321.7030 8 KYVLPSFEVR 1236.6866 8 YVLPSFEVR 1108.5917 7 FFYIDGNENFHVSITAR 2028.9694 6 RDGQNLVTMNLHITPDLIPSFR 2536.3220 10 DGQNLVTMNLHITPDLIPSFR 2380.2209 5 26b 1.1 LAAO EGWYVNMGPMR 1338.5849-1 A8QL58 (N. atra) 3 158 TFVTADYVIVCSTSR 1717.8345-3 RIYFEPPLPPK 1355.7601-2 26c 0.3 Not determined - - - - - - 27a < 0.1 CVF QLDIFVHDFPR 1385.7092 15 Q91132 (N. kaouthia) 7 344 QNQYVVVQVTGPQVR 1713.9162 16 GIYTPGSPVLYR 1321.7030 17 KYVLPSFEVR 1236.6866 17 YVLPSFEVR 1108.5917 15 TNHGDLPR 908.4464 16 IKLEGDPGAR 1054.5771 18 27b 0.1 Not determined - - - - - - 27c 0.5 SVMP (mocarhagin-1) VYEMVNALNTMYR 1602.7534 8 Q10749 2 74 VYEMVNALNTMYR 1618.7483 5 (N. mossambica) 28a < 0.1 CVF QLDIFVHDFPR 1385.7092 17 Q91132(N. kaouthia) 10 328 QNQYVVVQVTGPQVR 1713.9162 17 GIYTPGSPVLYR 1321.7030 19 KYVLPSFEVR 1236.6866 18 YVLPSFEVR 1108.5917 18 FFYIDGNENFHVSITAR 2028.9694 13 RDGQNLVTMNLHITPDLIPSFR 2536.3220 18 DGQNLVTMNLHITPDLIPSFR 2380.2209 14 DGQNLVTMNLHITPDLIPSFR 2396.2158 12 IKLEGDPGAR 1054.5771 20 28b 0.1 Not determined - - - - - - 28c 0.9 SVMP (cobrin) ATLDLFGEWR 1206.6033-7 Q9PVK7(N. naja) 1 68 28d 0.1 SVMP (cobrin) VYEMINTMNMIYR 1676.7724 17 Q9PVK7(N. naja) 6 161 VYEMINTMNMIYR 1692.7673 20 ATLDLFGEWR 1206.6033 24 TKPAYQFSSCSVR 1529.7297 20 DDCDLPELCTGQSAECPTDVFQR 2712.1102 15 DSCFTLNQR 1139.5030 35 28e 0.1 SVMP (cobrin) ATLDLFGEWR 1206.6033 41 Q9PVK7(N. naja) 3 65 TKPAYQFSSCSVR 1529.7297 35 DSCFTLNQR 1139.5030 51 28f 0.1 CVF DDNEDGFIADSDIISR 1780.7751 6 Q91132(N. kaouthia) 5 250 GICVAEPYEIR 1305.6387 11 AILHNYVNEDIYVR 1717.8787 8 AVPFVIVPLEQGLHDVEIK 2102.1776 5 ASVQEALWSDGVR 1416.6997 9 28g < 0.1 CVF TMSFYLR 916.4477 23 Q91132(N. kaouthia) 6 295 TMSFYLR 932.4426 21 GICVAEPYEIR 1305.6387 19 AILHNYVNEDIYVR 1717.8787 18 AVPFVIVPLEQGLHDVEIK 2102.1776 14 ASVQEALWSDGVR 1416.6997 19 Cysteine residues determined in MS/MS analysis are carbamidomethylated. Abbreviations: 3FTx; three-finger toxin, LNTX; long neurotoxin, SNTX; short neurotoxin, CTX; cytotoxin, WTX: weak neurotoxin/toxin, MTLP; muscarinic toxin-like protein, vngf; venom nerve growth factor, PLA 2 ; phospholipase A 2, SVMP; snake venom metalloproteinase, CRISP; cysteine-rich secretory protein, CTL; C-type lectin, PDE; phosphodiesterase, 5 NUC; 5 nucleotidase, CVF; cobra venom factor, LAAO; L-amino acid oxidase, EDCP; endonuclease domain-containing protein, GPX; glutathione peroxidase. *Results obtained from Orbitrap Fusion analysis Compared to the earlier reports, the present study unmasked the proteome of N. kaouthia venom to a deeper level. There are more protein classes, subtypes and relative abundances of the venom toxins revealed as shown in Figure 2. Previous 2D PAGE proteomic studies on N. kaouthia venoms from Thailand [24] and Malaysia [36] reported the presence of only 5 toxin classes respectively, without investigating the relative abundances of the various toxins. Presumably, the combination of reverse-phase HPLC and SDS-PAGE yielded better protein separation compared to 2D PAGE used in the previous studies. Also, the highly sensitive MALDI-TOF/TOF and/or Orbitrap Fusion analysis made it possible to 14

identify toxin isoforms at low abundances, hence adding values to the elucidation of N. kaouthia venom complexity. Most of the toxins identified through homologous sequence matching were annotated by N. kaouthia species as well as other closely related species under the genus Naja. The combination of RP-HPLC and SDS-PAGE analysis has also successfully revealed the presence of several novel toxin families (KUN, NP, PDE, 5 NUC and CTL) in N. kaouthia venoms that have not been isolated or reported in proteomic studies of this species, hence supporting some earlier findings reported for enzymatic activities of N. kaouthia venom [19, 37]. The major cobra venom protein families e.g. 3FTx, PLA 2, CRISP, SVMP and CVF were present in all three venom samples. There are, however, remarkable variations in the composition of the 3FTx, the principal lethal toxins of the venom. It is also interesting to note that some low abundance proteins were not detected in all three venoms: KUN was detected in NK-M and NK-V; NP was detected in NK-T; while CTL was detected in NK-M and NK-T only (Figure 2; Table 2). Of note, the absence of some of these minor proteins in certain venoms may be due to concentrations below the detection limit of the analytical assay. Overall, the major differences in the proteome of the N. kaouthia venoms from the three different geographical regions demonstrated again the existence of substantial geographical variations of snake venoms [13]. The phenomenon has profound medical significance as it can lead to variations in the clinical presentation of envenomation and differences in antivenom potency. This would imply that antivenom dosage may need to be tailored and optimised for the same species from different geographical locations. Figure 1 RP-HPLC liquid chromatogram obtained from C 18 reverse-phase fractionation of N. kaouthia venoms (3 mg) sourced from three different geographical locations and SDS-PAGE profiles of the individual fractions under reducing conditions. A broad range protein ladder (10 to 260 kda) was used for calibration. A) N. kaouthia from Malaysia. B) N. kaouthia from Thailand. C) N. kaouthia from Vietnam. Our results show that for N. kaouthia venoms of all three geographical sources, 3FTx and PLA 2 together constitute >85% of the venom proteins. It is interesting to note that 3FTx was highly expressed in NK-T (78.3%) and NK-V (76.4%), but slightly less in NK-M (63.7%). The PLA 2, on the other hand, was highly expressed in NK-M (23.5%), followed by 15

NK-V (17.4%) and lowest in NK-T (12.2%). The other toxin families generally exist in smaller amount, each with a relative abundance of less than 3%. Figure 2 Pie charts showing the relative abundance of venom protein families identified by mass spectrometry following reverse-phase HPLC and SDS-PAGE of N. kaouthia venoms. A) N. kaouthia from Malaysia. B) N. kaouthia from Thailand. C) N. kaouthia from Vietnam. Abbreviations: 3FTx; three-finger toxin, LNTX; long-chain neurotoxin, SNTX; short-chain neurotoxin, CTX; cytotoxin/cardiotoxin, MTLP; muscarinic toxinlike protein, WTX; weak neurotoxin/toxin, PLA 2 ; phospholipase A 2, CRISP; cysteine-rich secretory protein, SVMP; snake venom metalloproteinase, LAAO; L-amino acid oxidase, CVF; cobra venom factor, KUN; Kunitz-type protease inhibitor, NP; natriuretic peptide, PDE; phosphodiesterase, 5 NUC; 5 nucleotidase, CTL; c-type lectin, and NGF; nerve growth factor. Left: 3FTx subtypes; Right: Others protein families in the venoms. 3.2 Comparison of 3FTx composition of N. kaouthia venoms from different geographical regions Three-finger toxins (3FTxs) are virtually the signature lethal toxins of most elapid venoms. In the N. kaouthia venoms examined, they constitute 63.7-78.3% of total venom proteins, indicating that these toxins play a key role in the pathogenesis of the envenomation. 3FTxs share a similar pattern of protein folding: three β-stranded loops extending from a central core containing four conserved disulfide bridges [38-40]. In spite of being structurally similar, 3FTx subtypes exhibit a wide spectrum of pharmacological activities, including neurotoxicity, cytotoxicity/cardiotoxicity, anticoagulant and antiplatelet effects [41, 42]. An examination of the 3FTx composition of the three N. kaouthia venoms revealed substantial differences in terms of subtypes and relative abundances of the toxins (Figure 2; Table 2). The 3FTxs of N. kaouthia venom can be further classified into 5 different subtypes, namely long-chain neurotoxin (LNTX), short-chain neurotoxin (SNTX), weak toxin/neurotoxin (WTX), muscarinic toxin-like protein (MTLP) and cytotoxin/cardiotoxin (CTX) (Figure 2; Table 2). Table 2 Overview of the toxin families subtypes and relative abundance (%) of N. kaouthia Malaysia (NK-M), Thailand (NK-T) and Vietnam (NK-V) venoms. NK-M NK-T NK-V Protein family Subtype Accession/species (%) (%) (%) 16

3FTx 63.7 (14) 78.3 (11) 76.4 (18) LNTX 3.9 (1) 33.3 (1) - alpha-elapitoxin-nk2a P01391 (N. kaouthia) 3.9 33.3 - SNTX 4.2 (3) 7.7 (2) 9.2 (3) cobrotoxin P60771 (N. kaouthia) 2.3 3.6 6.6 cobrotoxin-b P59275 (N. kaouthia) 0.3-0.1 cobrotoxin-c P59276 (N. kaouthia) 1.6 4.1 2.5 CTX 45.7 (6) 27.6 (6) 44.9 (6) cardiotoxin 2A Q9PST4 (N. sputatrix) 8.4 0.2 0.1 cardiotoxin-1f P85429 (N. atra) - - 0.1 cytotoxin 1 P60305 (N. kaouthia) 10.6 - - cytotoxin 2 Q98965 (N. kaouthia) 0.1 - - cytotoxin 2 P01445 (N. kaouthia) 20.1 - - cytotoxin 3 P01446 (N. atra) - 17.1 - cytotoxin 3 P60301 (N. atra) - 0.1 23.0 cytotoxin 4N Q9W6W9 (N. atra) - - 2.0 cytotoxin 5a O73857 (N. sputatrix) - 8.6 15.8 cytotoxin homolog P14541 (N. kaouthia) 1.1 1.2 3.9 cytotoxin I-like T-15 Q91136 (N. atra) - 0.4 - cytotoxin NK-CT1 P0CH80 (N. kaouthia) 5.4 - - MTLP 0.8 (2) 0.8 (1) 3.0 (3) muscarinic toxin-like protein Q9W727 (B. multicinctus) - - 1.4 muscarinic toxin-like protein 1 P82462 (N. kaouthia) - - 0.5 muscarinic toxin-like protein 2 P82463 (N. kaouthia) 0.5 0.5 1.1 muscarinic toxin-like protein 3 P82464 (N. kaouthia) 0.3 0.3 - WTX 9.1 (2) 8.9 (1) 19.3 (6) probable weak neurotoxin NNAM2 Q9YGI4 (N. atra) - - 0.1 weak neurotoxin 6 P29180 (N. naja) 0.7-0.3 weak neurotoxin 6 O42256 (N. sputatrix) - - 11.2 weak neurotoxin 7 P29181 (N. naja) - - 0.4 weak toxin CM-9a P25679 (N. kaouthia) 8.4 8.9 7.0 weak toxin S4C11 P01400 (N. melanoleuca) - - 0.3 PLA 2 23.5 (4) 12.2 (2) 17.4 (2) acidic phospholipase A 2 1 P00596 (N. kaouthia) 13.0 0.3 - acidic phospholipase A 2 1 P00598 (N. atra) 0.7-16.0 acidic phospholipase A 2 2 P15445 (N. naja) 6.7-1.4 acidic phospholipase A 2 2 Q91133 (N. atra) 3.1 11.9 - CRISP 4.3 (2) 2.3 (2) 0.8 (2) natrin-1 Q7T1K6 (N. atra) 3.7 1.6 0.6 natrin-2 Q7ZZN8 (N. atra) 0.6 0.7 0.2 SVMP 3.3 (4) 2.5 (3) 1.6 (3) atragin D3TTC2 (N. atra) - - 0.7 atrase-a D5LMJ3 (N. atra) 0.4-0.3 atrase-b D6PXE8 (N. atra) - 1.0 - cobrin Q9PVK7 (N. naja) 1.1 1.0 - kaouthiagin P82942 (N. kaouthia) 1.3 0.5 0.6 mocarhagin-1 Q10749 (N. mossambica) 0.5 - - LAAO L-amino acid oxidase A8QL58 (N. atra) 1.1 (1) 1.0 (1) 0.5 (1) CVF cobra venom factor Q91132 (N. kaouthia) 0.8 (1) 1.1 (1) 0.7 (1) KUN Kunitz-type inhibitor P20229 (N. naja) 0.5 (1) - < 0.1 (1) NP natriuretic peptide Na-NP D9IX97 (N. atra) - 0.2 (1) - PDE phosphodiesterase 1 U3FAB3 (M. fulvius) 0.4 (1) 0.3 (1) 0.4 (1) 5'-NUC snake venom 5'-nucleotidase B6EWW8 (G. brevicaudus) 0.3 (1) 0.2 (1) 0.2 (1) Vespryn Thaicobrin P82885 (N. kaouthia) 0.3 (1) 0.7 (1) 0.2 (1) CTL BFL-1 Q90WI8 (B. fasciatus) 0.2 (1) 0.4 (1) - NGF 0.2 (1) 0.5 (2) 1.3 (2) venom nerve growth factor P61899 (N. kaouthia) 0.2-0.7 venom nerve growth factor 2 Q5YF89 (N. sputatrix) - 0.4 - venom nerve growth factor P01140 (N. naja) - 0.1 - nerve growth factor beta chain precursor A59218 (N. kaouthia) - - 0.6 Non-toxin 0.2 0.2 0.4 Not determined 1.2 0.1 0.4 Parentheses indicated the number of protein subtypes detected. Abbreviations: 3FTx; three-finger toxin, LNTX; long-chain neurotoxin, SNTX; short-chain neurotoxin, CTX; cytotoxin/cardiotoxin, MTLP; muscarinic toxin-like protein, WTX; weak neurotoxin/toxin, PLA 2 ; phospholipase A 2, CRISP; cysteinerich secretory protein, SVMP; snake venom metalloproteinase, LAAO; L-amino acid oxidase, CVF; cobra venom factor, KUN; Kunitz-type protease inhibitor, NP; natriuretic peptide, PDE; phosphodiesterase, 5 NUC; 5 nucleotidase, CTL; c-type lectin, and NGF; nerve growth factor. 17

Table 2 shows that the three N. kaouthia venoms differ vastly in their relative abundance of long-chain (LNTX) and short-chain (SNTX) α neurotoxins. Both the LNTX (65-72 residues with 5 disulfide bridges) and SNTX (60-62 residues with 4 disulfide bridges) are antagonists of muscular nicotinic cholinoceptor (nachr); though LNTX is also highly specific towards neuronal-type nachr (α7, α8 and α9) [43-45]. α-ntxs are rapid-acting; the toxins can cause flaccid paralysis, respiratory failure and consequently death in envenomed victims especially when treatment is delayed or inadequate, as seen in some cases of N. kaouthia envenomation in Thailand [46]. The short and long α-ntxs of elapid venom differ in their amino acid sequence length, conserved cysteine residues and number of disulfide bridges. In general they also vary in the degree of receptor binding reversibility, as the long α-ntxs are generally less reversible in the neuromuscular blockade compared to short α- NTXs [47]. The only LNTX subtype identified i.e. alpha-elapitoxin-nk2a (also known as the alpha-cobratoxin) is indeed the major neurotoxin present in NK-T venom, with an exceptionally high abundance of 33.3% in contrast to NK-M sample (3.9%) and NK-V sample (not detectable). The occurrence of the major LNTX, with 71 amino acids in a single peptide chain cross-linked by five disulfide bridges, was first described by [48] as the principal neurotoxin (comprising one fourth of crude venom weight) from the venom of Thai cobra, which was named as Naja naja siamensis back then. SNTX in the above mentioned Thai cobra venom was also identified later in three isoforms but with very low content compared to the LNTX [48, 49]. This is in agreement with our current findings on the LNTX and SNTX profile of Thai N. kaouthia venom. The current study demonstrated the presence of SNTX in all three N. kaouthia venoms: 4.2% in NK-M, 7.7% in NK-T and 9.2% in NK-V samples. There are three different subtypes of SNTX and their relative abundance also differs among the three venoms. The SNTX subtypes (cobrotoxin, cobrotoxin-b, cobrotoxic-c) reported here for the Southeast Asian N. kaouthia venoms were consistent with previous findings of three SNTX isolated from the Chinese N. kaouthia venom (Yunnan, China) [50, 51]. However, the Chinese (Yunnan) N. kaouthia venom was known to differ distinctively from the Thai N. kaouthia venom by the absence of LNTX. The neurotoxin profile of Chinese N. kaouthia also appeared similar to that of N. atra (Chinese cobra) which is dominated by SNTX [23]. The authors hypothesized that sharing of remarkably similar climate, habitat and prey between the Southern Chinese N. kaouthia and N. atra drove the development of similar α-neurotoxin profile between the two species in the region. Meanwhile, although the biochemical profile and toxic activities of Indian N. kaouthia venom 18

have been reported [52, 53], the composition of its 3FTx and intra-species variation remain to be further investigated. It is interesting to note that NK-V venom contains weak neurotoxins (WTXs) as its major type of neurotoxin, and the content of the WTX (19.3%) is markedly higher than that in NK-T and NK-M venoms ( 9%). The NK-V venom also contains relatively more variants of WTX. The most highly expressed isoform, weak neurotoxin 6, is similar to that cloned from N. sputatrix (UniProtKB: O42256) [54] (Table 2). It has been reported that the WTXs isolated from N. kaouthia (geographical origin unspecified) venom were antagonists of human and rat nicotinic receptors [55, 56], but less lethal (LD 50 ~5-80 µg/g) as compared to the typical α-ntx (LD 50 ~0.1 µg/g). On the other hand, muscarinic toxin-like proteins (MTLP), a 3FTx subtype of minor abundance were detected in all three venoms investigated, with a relatively higher abundance in NK-V venom. MTLP and WNTX both act on the cholinergic receptor system, where the MTLPs selectively antagonize distinct subtypes of muscarinic receptors (machrs) [57], while WTXs (as unconventional neurotoxins) possess low affinity toward both muscular-type and neuronal-type of nicotinic receptors (nachrs) [54, 55]. It has also been demonstrated that i.v. injection of WTX induced a dose-dependent decrease in blood pressure and heart rate in rodents [58]. The observation could be related to autonomic disturbances induced by the interaction of WTXs with muscarinic cholinoceptors [54, 55, 59], although the specific effect has not been well reported in N. kaouthia envenomation. Our results also reveal major differences in the CTX content of the three N. kaouthia venoms. For NK-M and NK-V venoms, CTXs constitute approximately 45% of the total venom proteins; but for NK-T, CTX content was only 27.6% (Figure 2; Table 2). CTX has a wide range of pharmacological activities [19, 25, 40]; its main pathogenic action is cytotoxic and cytolytic leading to tissue destruction [60-62]. It is however usually 10-time less lethal (LD 50 ~ 1.0-2.5 µg/g) [63, 64] compared to most cobra α-ntx (LD 50 ~ 0.1 µg/g) [63]. The major CTX subtypes present in NK-M venom (protein families cytotoxin 1, 2, and 2A) are found different from those in the venoms of NK-T and NK-V. For both NK-T and NK-M venoms, the major CTX subtypes are comprised of cytotoxin 3 and cytotoxin 5a (Table 2). The differences in the 3FTx composition of the three N. kaouthia venoms are reflected in the differences of their lethality. The Thai N. kaouthia venom, with the highest amount of the highly lethal LNTX, has a much lower LD 50 (0.2 µg/g) as compared to the 19

Malaysian and Vietnamese species (LD 50 =0.9 µg/g for both venoms) [19, 21] (Table 3). This is further compared and discussed in sections regarding lethality and neutralization study. 3.3 Other toxin components of the N. kaouthia venoms Phospholipase A 2 (PLA 2 ) is the second most abundant toxin family in all three N. kaouthia venoms examined (17.4-23.5%), followed by CRISP, SVMP, LAAO, CVF, KUN, NP, PDE, 5 NUC, vespryn, CTL and NGF. The two protein families CRISP and SVMP constitute 4.3% and 3.3% of the NK-M venom proteins but their abundances are lower in the NK-T and NK- V venoms. The rest of toxin families generally constitute less than 1% of total venom proteins (Figure2). The low abundances of these toxins suggest that these toxins play a minor role in the toxic actions of the venom, and their functions are probably ancillary. Secretory PLA 2 is a protein family rapidly expanding throughout the evolution of venomous snakes. With a typical molecular mass of 13-15 kda, PLA 2 is present in virtually all snake venoms, usually with multiple isoforms and varied pharmacological activities. The PLA 2 in all three N. kaouthia venoms belong to Group IA, and constitute a relatively high content in the venom (12-23%), in agreement with high PLA 2 activity reported previously for the Thai and Indian N. kaouthia venom [19, 52]. It should be noted that all PLA 2 isoforms of N. kaouthia venoms identified in the current study are acidic subtypes, which have been reported to be less toxic compared to basic PLA 2 [65, 66]. Previous studies indicated that acidic PLA 2 from Indian N. kaouthia venom was non-toxic to mice up to a dose of 10 µg/g. The enzyme, however, exhibited cytotoxic and anticoagulant activity [67, 68]. In addition, cobra acidic PLA 2 may interact synergistically with cytotoxins or cardiotoxins, hence potentiating the cytolytic effect of the venom that leads to local tissue necrosis. Besides, PLA 2 is also known to enhance the lethal effect of cobra venoms [63, 69]. Other than 3FTxs and PLA 2, CRISP and SVMP are the two protein families that account for >2% of venom mass in NK-M and NK-T venoms. CRISP is a single chain polypeptide widely distributed in numerous animal tissues and reptilian venoms [70]. It exhibits a wide range of biological activities, including blockage of various ion channels and induction of hypothermia in prey animals. However, most reptilian venom CRISPs have no identifiable function and no known acute toxic effects [70]. The CRISPs identified in the N. kaouthia venoms were homologous to natrin isolated from N. atra (Table 2), an antagonist at 20

the high-conductance calcium-activated potassium (BKca) channel and ryanodine (RyR1) receptors [71-74]. SVMP, a high molecular mass enzymatic toxin, is usually the dominant toxins in viperid venoms. It is involved in destruction of basal membrane, causing haemorrhage and presumably also plays a role in prey digestion [75-78]. SVMPs identified in the three N. kaouthia venoms appear to be above 40 kda (Figure 1) and are homologous to SVMP P-III subtype. The low expression level and the lack of coagulopathic feature in cobra envenomed patients suggest that the biological role of SVMPs in N. kaouthia venoms is likely for prey digestion purpose, though the enzymes may contribute to local tissuedamaging effect. LAAO constitute only 0.5-1.1% in all the three N. kaouthia venoms, a level consistent with the low level of activity reported in Thai N. kaouthia venom [19]. The expression level of LAAO in snake venom can be highly variable. Usually it is a minor component [79, 80], but in certain species such as Ophiophagus hannah, Calloselasma rhodostoma and Hypnale hypnale, the venom LAAO content can exceed 10% of total venom protein [81]. The enzyme has been isolated from Naja kaouthia venom and shown to exhibit platelet aggregation activity [82, 83]. In view of its low content, LAAO is unlikely to play a major role in the pathophysiological action of N. kaouthia venom. The biological significance of LAAO in the venom may be more of antimicrobial. All three N. kaouthia venoms contained a small amount (0.7-1.1%) of cobra venom factor (CVF). CVF is a complement-activating protein in cobra venom with structural and functional homology to complement C3. It is able to initiate the alternative pathway by forming with factor B an extremely stable C3/C5 convertase, generating excessive anaphylatoxins and depleting complements in the serum of many vertebrates [84]. Although CVF is unlikely to play a direct role in the lethality of the venom, the protein can cause the release of anaphylatoxins such as C3a and C5a that promote local pro-inflammatory response through vasodilatation and the chemotaxis and activation of leukocytes. These increase the vascular permeability and blood flow at the bite site, thus enhancing the absorption and spread of toxins, a strategy needed for fast distribution of toxins to the target organs. On the other hand, Thaicobrin, a vespryn class of toxin was found in minor amount in all three venoms (0.2-0.7 µg/g). There is a lack of published literature on its functional characterization. The protein, however, exhibits high sequence homology to Ohanin isolated from king cobra (Ophiophagus hannah) venom, hence Thaicobrin possibly possesses similar 21

pharmacological activities as Ohanin, i.e. inducing hyperalgesia and hypolocomotion in prey animals, which were believed to be complementary actions in predation [85, 86]. Our proteomic results also revealed the presence of protein families KUN, NP, PDE, 5 NUC and CTL, previously not reported in N. kaouthia venom, although PDE and 5 -NUC have been detected enzymatically [19, 63]. KUN was detected in the venoms of NK-M and NK-V, and appeared similar to serine protease inhibitor isolated from N. naja venom which strongly inhibits trypsin [87]. Natriuretic peptide (NP) was detected in NK-T venom. This is a minor toxin component reported to induce rapid relaxation of phenylephrine-precontracted rat aortic strips, increase cgmp formation and hypotensive effect [88]. Besides, both PDE and 5 NUC were detected in minute amounts (0.2-0.4%) in the three N. kaouthia venoms. These enzymes function to liberate nucleosides that may help in prey immobilization. PDE has been shown to associate with a drop in arterial pressure and locomotor depression in animals injected with phosphodiesterase isolated from vipers venom [89]. Together, they exert hypotensive effect [90-92], which may act to slow down the prey. The current study also revealed the presence of CTL in two of the N. kaouthia venoms (NK-M and NK-T). CTL is a toxin that targets a wide range of plasma components and cells particularly platelets, causing either platelet aggregation activation or inhibition [93]. CTL is reported mainly in viperid venoms, its toxic role in N. kaouthia envenomation is unlikely to be significant in view of the lack of report on platelet dysfunction or thrombocytopenia in envenomation. Another minor ancillary toxin, NGF, is present in all three venoms examined with a content of 0.2-1.3%. NGF has been shown to be cytotoxic and capable of inducing apoptosis [94]. Cobra NGF is potentially useful in the development of neurotrophic drug, as it has been shown to be a natural antagonist of human trka-receptor of a lower potency but of similar efficacy compared with mammalian NGFs [95]. Again, it is unlikely this very minor component plays important role in the lethal action of the cobra venom. 3.4 Comparison with the venom proteomes of other cobras Our findings indicate relatively more protein families (a total of 16 families) in Naja kaouthia venom than that reported for the venoms of several other cobras: Chinese Naja atra (3 protein families) [96], African spitters N. nigricollis, N. katiensis, N. pallida, N. nubiae, N. mossambica (6) [20], Pakistani N. naja (6) [97], Moroccan N. haje (10) [26] and Malaysian N. 22

sumatrana (10) [25]. The discrepancies could be due to differences in the resolution of proteomic techniques or the database used; nevertheless, the dominating presence of 3FTx (particularly for neurotoxins and cytotoxins/cardiotoxins) and PLA 2 has been consistent in all the venom proteomes of African and Asian cobras studied thus far. Although not all the studies reported the relative abundance of toxins, a distinct pattern of 3FTx distribution is worth noted: venoms of the African spitters, the Moroccan N. haje, and the Malaysian N. sumatrana possess high content of 3FTx (60-73% of venom protein), consistent with that found in N. kaouthia proteomes in this study. The consistent dominance of 3FTx in cobra venoms however is not represented in some genera of terrestrial elapids, for instance, the Asiatic banded krait Bungarus fasciatus and Malayan krait B. candidus [27], and certain species of the New World coral snake such as Micrurus nigrocinctus [98], and New Guinean small-eyed snake, Micropechis ikaheka [99] that contain high abundance of PLA 2, comparable to or exceeding the 3FTx. Within the 3FTx family, the Thai N. kaouthia venom reported here appears to contain the highest amount of alpha-neurotoxins (> 40%) compared to the venoms of African spitters (0.4-15%), the Morrocon cobra (13%), the Malaysian black spitter (15%), and the Pakistani common cobra (minor component of 3FTx by qualitative description). This is consistent with clinical reports that a high percentage of patients envenomed by N. kaouthia in Thailand experienced neurotoxic symptoms [100]. These congeneric cobras as well as the N. kaouthia from Vietnam and Malaysia, nonetheless, share a higher content of cytotoxins / cardiotoxins (33-73%) than the N. kaouthia from Thailand (28%). This is particularly apparent in the venom proteomes of African spitting cobras characterized by approximately 70% of CTX, supporting that the venoms are cytotoxic in principle and less neurotoxic [20]. A comparison on the existing data of intravenous LD 50 values of these cobra venoms revealed that the African spitters have LD 50 ranging from 0.6-1.1 μg/g mouse, comparable to N. sumatrana (0.5 μg/g), as well as N. kaouthia from Malaysia and Vietnam (0.9 μg/g), but less potent compared to that from Thailand (0.2 μg/g). A recent enzymatic study of Southeast Asian cobra venoms [19] demonstrated the activities of hyaluronidase, alkaline phosphomonoesterase and acetylcholinesterase, which presence were not detected in this current proteomic study. This could be due to exceptionally low amount of these enzymes in the venom, complicated by the lack of specific sequence information in the database. In proteomic studies, PLA 2 nonetheless appeared to be substantial in all Naja venoms (15-30%) except the Moroccan cobra (4%). A previous 23

enzymatic study coupled with cation-exchange chromatography [19] has provided valuable PLA 2 information for comparison and correlation in our current study. The Southeast Asian spitting cobras (N. sumatrana, N. siamensis, N. sputatrix) contain substantial amount of basic PLA 2, which was also reported for the African spitting cobras N. nigricollis [101], however not detected in the non-spitting N. kaouthia venom. The chromatographic result is in agreement with our proteomic finding, where only acidic PLA 2 isoforms were detected at the abundance of 12-24% (for NK-T, NK-M and NK-T). The presence of basic PLA 2 along with the high content of CTX in the venoms of spitting cobras from African and Malaysia perhaps indicates synergistic interactions that contribute to highly potent cytotoxicity, in line with the pathogenic effect of venom ophthalmia [102]. Proteomic comparison of N. kaouthia venom with the venoms of other Southeast Asian cobras in the region remains to be investigated, especially on the case of N. siamensis as its separation with N. kaouthia was not distinct prior to the major revision of cobra taxonomy [103], hence raising questions on the validity of some earlier reports using Thai cobra venom in general. 3.5 Comparison of median lethal doses (LD 50 ) of the three N. kaouthia venoms The marked variations in the composition of 3FTx among NK-T, NK-M and NK-V venoms were reflected in the differences in their median lethal doses (LD 50 ). Venoms were administered via two different routes into mice: intravenously to ensure total systemic access of the challenge dose, and subcutaneously to mimic real envenomation mode. The median lethal dose via s.c. injection is not significantly different from that administered via i.v. injection (Table 3) considering the small differences of LD 50 with overlapping 95% C.I.. This implies that the bioavailability of the subcutaneously injected principal lethal toxins is close to 100%. This finding is consistent with the result of the pharmacokinetic study on N. sumatrana venom and toxins [69], where the high bioavailability is attributable to effective absorption of 3FTxs of small molecular size. Table 3 The medium lethal dose (LD 50 ) of N. kaouthia venoms Malaysia, Thailand and Vietnam administrated by intravenous (i.v.) and subcutaneous (subcut.) routes. Values of 95% C.I. were in parentheses. Venom i.v. LD 50 Subcut. LD 50 Naja kaouthia (Malaysia) 0.90 (0.59-1.36) 1.00 (0.88-1.14) Naja kaouthia (Thailand) 0.18 (0.12-0.27) 0.20 (0.16-0.25) Naja kaouthia (Vietnam) 0.90 (0.59-1.36) 1.11 (0.73-1.69) 24

Table 3 showed that the venom of Thai N. kaouthia has the lowest LD 50, approximately five-fold more lethal than that for Malaysian and Vietnamese samples (LD 50 ~0.9 µg/g each). The markedly higher lethality of Thai N. kaouthia venom is related to the high abundance (33.3%) of LNTX in the venom. Venoms of the Malaysian and Vietnamese N. kaouthia, on the other hand, have a much higher content of CTX and both venoms contain a lesser amount of the highly lethal α-neurotoxins. Interestingly, the Vietnamese sample is unique with a higher content of the non-conventional weak neurotoxins (WTX). Kaouthiotoxin (KTX), a WTX from N. kaouthia venom has been found to non-covalently interact with PLA 2 [104], forming a synergistic complex that potentiates the cytotoxic action of PLA 2 and causes more intense membrane damage [104]. Therefore, it is hypothesized that with high WTX (19.3%) and PLA 2 content (17.4%), along with highly abundant CTX (44.9%), NK-V venom may exert more potent cytotoxic effect compared to NK-M and NK-T venoms. The findings encourage future studies on comparative cytotoxicity of the venoms and CTX. 3.6 Neutralization of N. kaouthia venoms by two antivenoms In view of the substantial variations in the proteomes and LD 50 values of N. kaouthia venoms from the three different geographical locations, we further investigated if all three venoms can be effectively neutralized by the commonly available antivenoms in the region, namely N. kaouthia monovalent antivenom (NKMAV) and Neuro polyvalent antivenom (NPAV). Both the antivenoms are produced by QSMI, Bangkok, using the venom of Thai N. kaouthia as immunogen (in the case of the polyvalent antivenom, other venoms are included in the immunogen mixture as stated above). Firstly, antivenom and venom were mixed and incubated ex vivo for immunocomplexation, followed by intravenous injection of the supernatant (if the centrifugation step was included) or the whole mixture (if the centrifugation step was omitted) to assess efficacy of neutralization (this is referred to as the preincubation method). Whether the centrifugation step was added or omitted, our study using NPAV against the three venoms found that the antivenom neutralization potencies remain similar for both protocols (data not shown), implying that the immunocpmplexes were inactive, or eliminated (phagocytosed) effectively in vivo. The preincubation method has the advantage of ensuring total systemic access of venom-antivenom mixture under controlled titration. The results indicated that both the antivenoms produced for Thai species 25

were able to effectively neutralize the lethality of N. kaouthia venoms examined, regardless of their origins of location (Table 4). Using NPAV, Although monovalent antivenom has been suggested to perform better than polyvalent antivenom [105], both NKMAV and NPAV appear comparable in efficacy in this study, consistent with earlier reports that polyvalent antivenom is not necessary inferior in term of efficacy to that of monovalent antivenom [6, 106]. Our results indicated that despite the variation in the proteomes (in particular the content of major lethal toxins), the venom proteins of the N. kaouthia possess sufficiently similar immunogenicity to be neutralized effectively by the antivenoms. The findings were consistent with an in vivo protection assay (challenge-rescue experimental model) where independent administration of venom (s.c.) and antivenom (i.v.) were carried out in mice. The NPAV was selected for this model as it has been shown to perform slightly more superior to the monovalent antivenom from pre-incubation assay and previous studies [21]. Moreover, it is the preferred choice of antivenom in most neurotoxic snakebite cases in Malaysia and Thailand where the diagnosis of biting species cannot be ascertained. Our findings showed that NPAV is effective for all three N. kaouthia venoms when administered independently into mice which were experimentally envenomed (Table 5). The findings nonetheless revealed that NPAV was able to neutralize the venom of Malaysian N. kaouthia with a higher potency (approximately 70% in excess) than to the venom of Thai species. This perhaps indirectly reflects that antivenom treatment for N. kaouthia envenomation in Malaysia can be optimised by tapering the dose accordingly, thus lowering the risk of hypersensitivity and treatment cost of antivenom. Nevertheless, the change in dose according to locality should only be made if the monocled cobras within a population do not exhibit remarkable interindividual venom variation. This relies on a wide-scale study on potential inter-individual venom variation to be explored in the future. Table 4 - Neutralization of lethality of N. kaouthia venoms from different locations by N. kaouthia monovalent antivenom (NKMAV) and Neuro polyvalent antivenom (NPAV): Experiments with preincubation of venom and antivenom prior to injection. In Challenged NKMAV NPAV Venom i.v. LD 50 dose ED 50 (µl) ER 50 (mg/ml) P ( mg/ml) ED 50 (µl) ER 50 (mg/ml) P ( mg/ml) Naja kaouthia (Malaysia) 0.90 (0.59-1.36) 5 LD 50 78.29 1.38 (0.90-2.08) 1.10 70.68 1.43 (0.94-2.16) 1.14 Naja kaouthia (Thailand) 0.18 (0.12-0.27) 5 LD 50 18.75 1.15 (0.77-1.73) 0.92 17.67 1.17 (0.78-1.76) 0.94 Naja kaouthia (Vietnam) 0.90 (0.59-1.36) 5 LD 50 120.86 0.87 (0.57-1.32) 0.70 89.89 1.10 (0.72-1.66) 0.88 26

Table 5 - Neutralization of lethality of N. kaouthia venoms from different locations by Neuro polyvalent antivenom (NPAV): Experiments with independent administration of venom and antivenom (challenge-rescue experiment). Species Subcut. LD 50 Challenged dose NPAV ED 50 (µl) ER 50 (mg/ml) P ( mg/ml) Naja kaouthia (Malaysia) 1.00 (0.88-1.14) 5 LD 50 53.16 2.06 (1.82-2.36) 1.66 Naja kaouthia (Thailand) 0.20 (0.16-0.25) 5 LD 50 19.57 1.20 (0.96-1.50) 0.96 Naja kaouthia (Vietnam) 1.11 (0.73-1.69) 5 LD 50 78.29 1.42 (0.93-2.16) 1.13 4 Conclusion Our proteomic findings revealed as many as 16 different protein families (13 of which are toxins) in the venoms of N. kaouthia from three Southeast Asian locations (Malaysia, Thailand, Vietnam). The 3FTxs constitute the major arsenal (~60-80%) of the venoms, supplemented mainly by PLA 2 (~20%). Most of the other protein families comprise a small proportion (<5%) of the venom proteins. The 3FTx profiles of N. kaouthia venoms vary remarkably according to the geographical locations. The variations in these major toxins indicate that the venoms may have evolved under different levels of selection pressure leading to substantial molecular diversification. The effects of ecological niche on the toxin specialization of this species remain to be further studied. On the other hand, the variations in 3FTx profile correlated with the levels of lethality of the three venoms tested in mice. Regardless of the proteomic variations (in particular the 3FTx profile), both the commercially available Thai Naja kaouthia monovalent antivenom and Neuro polyvalent antivenom are effective in neutralizing the three N. kaouthia venoms in both the pre-incubation and the challenge-rescue experimental models. From the clinical standpoint, the regional antivenoms indicated for Thai N. kaouthia envenomation would be useful for bites by the same species in Malaysia and Vietnam, presumably due to conserved immunological determinants despite the variability of their 3FTx compositions. The higher efficacy of the Thai antivenoms in neutralizing Malaysian N. kaouthia venom may be indicative of dose reduction; this however requires further study to validate that the N. kaouthia population in Malaysia does not exhibit remarkable inter-individual venom variations. 27

Acknowledgements: The authors express high gratitude to Prof. Kavi Ratanabanangkoon from the Chulalongborn Research Institute, Bangkok, for his generous gift of venoms. This work was supported by UM High Impact Research Grant UM-MOE UM.C/625/1/HIR/MOE/E00040-20001 and Fundamental Research Grant FP028-2014A from Ministry of Education, Malaysia, RG521-13HTM, RG282-14AFR and PG043-2013B.from University of Malaya. References: [1] Gutierrez JM, Theakston RD, Warrell DA. Confronting the neglected problem of snake bite envenoming: the need for a global partnership. PLoS Med. 2006;3:e150. [2] Harrison RA, Hargreaves A, Wagstaff SC, Faragher B, Lalloo DG. Snake envenoming: a disease of poverty. PLoS Negl Trop Dis. 2009;3:e569. [3] Williams DJ, Gutierrez JM, Calvete JJ, Wuster W, Ratanabanangkoon K, Paiva O, et al. Ending the drought: new strategies for improving the flow of affordable, effective antivenoms in Asia and Africa. J Proteomics. 2011;74:1735-67. [4] Gutierrez JM, Warrell DA, Williams DJ, Jensen S, Brown N, Calvete JJ, et al. The need for full integration of snakebite envenoming within a global strategy to combat the neglected tropical diseases: the way forward. PLoS Negl Trop Dis. 2013;7:e2162. [5] Keyler DE, Gawarammana I, Gutierrez JM, Sellahewa KH, McWhorter K, Malleappah R. Antivenom for snakebite envenoming in Sri Lanka: the need for geographically specific antivenom and improved efficacy. Toxicon. 2013;69:90-7. [6] Chippaux JP. [Guidelines for the production, control and regulation of snake antivenom immunoglobulins]. Biol Aujourdhui. 2010;204:87-91. [7] Kasturiratne A, Wickremasinghe AR, de Silva N, Gunawardena NK, Pathmeswaran A, Premaratna R, et al. The global burden of snakebite: a literature analysis and modelling based on regional estimates of envenoming and deaths. PLoS Med. 2008;5:e218. [8] Jamaiah I, Rohela M, Ng TK, Ch'ng KB, Teh YS, Nurulhuda AL, et al. Retrospective prevalence of snakebites from Hospital Kuala Lumpur (HKL) (1999-2003). Southeast Asian J Trop Med Public Health. 2006;37:200-5. [9] (WHO) WHO. Guidelines for the prevention and clinical management of snakebite in Africa. Brazzaville: WHO Regional Office for Africa. 2010. [10] Chew KS, Khor HW, Ahmad R, Rahman NH. A five-year retrospective review of snakebite patients admitted to a tertiary university hospital in Malaysia. Int J Emerg Med. 2011;4:41. [11] Mohapatra B, Warrell DA, Suraweera W, Bhatia P, Dhingra N, Jotkar RM, et al. Snakebite mortality in India: a nationally representative mortality survey. PLoS Negl Trop Dis. 2011;5:e1018. [12] Stiles BG. Acetylcholine receptor binding characteristics of snake and cone snail venom postsynaptic neurotoxins: further studies with a non-radioactive assay. Toxicon. 1993;31:825-34. [13] Mackessy SP. The field of reptile toxinology. Snakes, lizards, and their venoms. In S. P. Mackessy (Ed), Handbook of venoms and toxins of reptiles. (pp. 3-23). Boca Raton, FL: Taylor and Francis Group, CRC Press. 2009. [14] Barlow A, Pook CE, Harrison RA, Wuster W. Coevolution of diet and prey-specific venom activity supports the role of selection in snake venom evolution. Proc Biol Sci. 2009;276:2443-9. [15] Gibbs HL, Mackessy SP. Functional basis of a molecular adaptation: prey-specific toxic effects of venom from Sistrurus rattlesnakes. Toxicon. 2009;53:672-9. [16] Glenn JL, Straight RC, Wolfe MC, Hardy DL. Geographical variation in Crotalus scutulatus scutulatus (Mojave rattlesnake) venom properties. Toxicon. 1983;21:119-30. 28

[17] Warrell DA. Unscrupulous marketing of snake bite antivenoms in Africa and Papua New Guinea: choosing the right product--'what's in a name?'. Trans R Soc Trop Med Hyg. 2008;102:397-9. [18] Gutierrez JM, Burnouf T, Harrison RA, Calvete JJ, Kuch U, Warrell DA, et al. A multicomponent strategy to improve the availability of antivenom for treating snakebite envenoming. Bull World Health Organ. 2014;92:526-32. [19] Yap MK, Tan NH, Fung SY. Biochemical and toxinological characterization of Naja sumatrana (Equatorial spitting cobra) venom. J Venom Anim Toxins incl Trop Dis. 2011;17:451-9. [20] Petras D, Sanz L, Segura A, Herrera M, Villalta M, Solano D, et al. Snake venomics of African spitting cobras: toxin composition and assessment of congeneric cross-reactivity of the pan-african EchiTAb-Plus-ICP antivenom by antivenomics and neutralization approaches. J Proteome Res. 2011;10:1266-80. [21] Leong PK, Sim SM, Fung SY, Sumana K, Sitprija V, Tan NH. Cross neutralization of Afro- Asian cobra and Asian krait venoms by a Thai polyvalent snake antivenom (Neuro Polyvalent Snake Antivenom). PLoS Negl Trop Dis. 2012;6:e1672. [22] Leong PK, Tan CH, Sim SM, Fung SY, Sumana K, Sitprija V, et al. Cross neutralization of common Southeast Asian viperid venoms by a Thai polyvalent snake antivenom (Hemato Polyvalent Snake Antivenom). Acta Trop. 2014;132:7-14. [23] Wei JF, Lu QM, Jin Y, Li DS, Xiong YL, Wang WY. Alpha-neurotoxins of Naja atra and Naja kaouthia snakes in different regions. Sheng Wu Hua Xue Yu Sheng Wu Wu Li Xue Bao (Shanghai). 2003;35:683-8. [24] Kulkeaw K, Chaicumpa W, Sakolvaree Y, Tongtawe P, Tapchaisri P. Proteome and immunome of the venom of the Thai cobra, Naja kaouthia. Toxicon. 2007;49:1026-41. [25] Yap MK, Fung SY, Tan KY, Tan NH. Proteomic characterization of venom of the medically important Southeast Asian Naja sumatrana (Equatorial spitting cobra). Acta Trop. 2014;133:15-25. [26] Malih I, Ahmad rusmili MR, Tee TY, Saile R, Ghalim N, Othman I. Proteomic analysis of Moroccan cobra Naja haje legionis venom using tandem mass spectrometry. J Proteomics. 2014;96:240-52. [27] Rusmili MR, Yee TT, Mustafa MR, Hodgson WC, Othman I. Proteomic characterization and comparison of Malaysian Bungarus candidus and Bungarus fasciatus venoms. J Proteomics. 2014;110C:129-44. [28] Howard-Jones N. A CIOMS ethical code for animal experimentation. WHO Chron. 1985;39:51-6. [29] Correa-Netto C, Junqueira-de-Azevedo Ide L, Silva DA, Ho PL, Leitao-de-Araujo M, Alves ML, et al. Snake venomics and venom gland transcriptomic analysis of Brazilian coral snakes, Micrurus altirostris and M. corallinus. J Proteomics. 2011;74:1795-809. [30] Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680-5. [31] Finney DJ. Probit Analysis. 2nd ed. Cambridge University Press. England. 1952. [32] Ramos-Cerrillo B, de Roodt AR, Chippaux JP, Olguin L, Casasola A, Guzman G, et al. Characterization of a new polyvalent antivenom (Antivipmyn Africa) against African vipers and elapids. Toxicon. 2008;52:881-8. [33] Morais V, Ifran S, Berasain P, Massaldi H. Antivenoms: potency or median effective dose, which to use? Journal of Venomous Animals and Toxins including Tropical Diseases. 2010;16:191-3. [34] Calvete JJ, Fasoli E, Sanz L, Boschetti E, Righetti PG. Exploring the venom proteome of the western diamondback rattlesnake, Crotalus atrox, via snake venomics and combinatorial peptide ligand library approaches. J Proteome Res. 2009;8:3055-67. [35] Fry BG, Wuster W. Assembling an arsenal: origin and evolution of the snake venom proteome inferred from phylogenetic analysis of toxin sequences. Mol Biol Evol. 2004;21:870-83. [36] Vejayan J, Khoon TL, Ibrahim H. Comparative analysis of the venom proteome of four important Malaysian snake species. J Venom Anim Toxins Incl Trop Dis. 2014;20:6. [37] Tan NH, Tan CS. A comparative study of cobra (Naja) venom enzymes. Comp Biochem Physiol B. 1988;90:745-50. [38] Menez A. Functional architectures of animal toxins: a clue to drug design? Toxicon. 1998;36:1557-72. [39] Tsetlin V. Snake venom alpha-neurotoxins and other 'three-finger' proteins. Eur J Biochem. 1999;264:281-6. 29

[40] Hegde RP, Rajagopalan N, Doley R, Kini RM. Snake venom three-finger toxins. In: Mackessy SP, (Ed.) Handbooks of venoms and toxins of reptiles. (pp. 287-301). Boca Raton, FL: CRC Press. 2009. [41] Kini RM, Doley R. Structure, function and evolution of three-finger toxins: mini proteins with multiple targets. Toxicon. 2010;56:855-67. [42] Chanda C, Sarkar A, Sistla S, Chakrabarty D. Anti-platelet activity of a three-finger toxin (3FTx) from Indian monocled cobra (Naja kaouthia) venom. Biochem Biophys Res Commun. 2013;441:550-4. [43] Endo T, Tamiya N. Structure-function relationship of postsynaptic neurotoxins from snake venoms. In: Harvey, A.L. (Ed.) Snake Toxins. (pp. 165-222). Pergamon Press, New York. 1991. [44] 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 alpha7 nicotinic receptor. J Biol Chem. 1997;272:24279-86. [45] Servent D, Menez A. Snake neurotoxins that interact with nicotinic acetylcholine receptor. In: Massaro, E.00000J., (Ed.) Handbook of Neurotoxicology: Vol I. (pp385-425). Humana Press, Totowa, New Jersey. 2001. [46] Wongtongkam N, Wilde H, Sitthi-Amorn C, Ratanabanangkoon K. A study of Thai cobra (Naja kaouthia) bites in Thailand. Mil Med. 2005;170:336-41. [47] Barber CM, Isbister GK, Hodgson WC. Alpha neurotoxins. Toxicon. 2013;66:47-58. [48] Karlsson E, Eaker D. Isolation of the principal neurotoxins of Naja naja subspecies from the Asian mainland. Toxicon. 1972;10:217-25. [49] Chiou SH, Lin WW, Chang WP. Sequence characterization of venom toxins from Thailand cobra. Int J Pept Protein Res. 1989;34:148-52. [50] Cheng Y, Meng Q, Wang W, Wang J. Structure-function relationship of three neurotoxins from the venom of Naja kaouthia: a comparison between the NMR-derived structure of NT2 with its homologues, NT1 and NT3. Biochim Biophys Acta. 2002;1594:353-63. [51] Meng QX, Wang WY, Lu QM, Jin Y, Wei JF, Zhu SW, et al. A novel short neurotoxin, cobrotoxin c, from monocellate cobra (Naja kaouthia) venom: isolation and purification, primary and secondary structure determination, and tertiary structure modeling. Comp Biochem Physiol C Toxicol Pharmacol. 2002;132:113-21. [52] Mukherjee AK, Maity CR. Biochemical composition, lethality and pathophysiology of venom from two cobras-- Naja naja and N. kaouthia. Comp Biochem Physiol B Biochem Mol Biol. 2002;131:125-32. [53] Das T, Bhattacharya S, Biswas A, Gupta SD, Gomes A, Gomes A. Inhibition of leukemic U937 cell growth by induction of apoptosis, cell cycle arrest and suppression of VEGF, MMP-2 and MMP- 9 activities by cytotoxin protein NN-32 purified from Indian spectacled cobra (Naja naja) venom. Toxicon. 2013;65:1-4. [54] Poh SL, Mourier G, Thai R, Armugam A, Molgo J, Servent D, et al. A synthetic weak neurotoxin binds with low affinity to Torpedo and chicken alpha7 nicotinic acetylcholine receptors. Eur J Biochem. 2002;269:4247-56. [55] Utkin YN, Kukhtina VV, Kryukova EV, Chiodini F, Bertrand D, Methfessel C, et al. "Weak toxin" from Naja kaouthia is a nontoxic antagonist of alpha 7 and muscle-type nicotinic acetylcholine receptors. J Biol Chem. 2001;276:15810-5. [56] Utkin YN, Kukhtina VV, Maslennikov IV, Eletsky AV, Starkov VG, Weise C, et al. First tryptophan-containing weak neurotoxin from cobra venom. Toxicon. 2001;39:921-7. [57] Kukhtina VV, Weise C, Muranova TA, Starkov VG, Franke P, Hucho F, et al. Muscarinic toxinlike proteins from cobra venom. Eur J Biochem. 2000;267:6784-9. [58] Ogay AY, Rzhevsky DI, Murashev AN, Tsetlin VI, Utkin YN. Weak neurotoxin from Naja kaouthia cobra venom affects haemodynamic regulation by acting on acetylcholine receptors. Toxicon. 2005;45:93-9. [59] Mordvintsev DY, Rodionov DI, Makarova MV, Kamensky AA, Levitskaya NG, Ogay AY, et al. Behavioural effects in mice and intoxication symptomatology of weak neurotoxin from cobra Naja kaouthia. Basic Clin Pharmacol Toxicol. 2007;100:273-8. 30

[60] Feofanov AV, Sharonov GV, Astapova MV, Rodionov DI, Utkin YN, Arseniev AS. Cancer cell injury by cytotoxins from cobra venom is mediated through lysosomal damage. Biochem J. 2005;390:11-8. [61] Osipov AV, Kasheverov IE, Makarova YV, Starkov VG, Vorontsova OV, Ziganshin R, et al. Naturally occurring disulfide-bound dimers of three-fingered toxins: a paradigm for biological activity diversification. J Biol Chem. 2008;283:14571-80. [62] Konshina AG, Boldyrev IA, Utkin YN, Omel'kov AV, Efremov RG. Snake cytotoxins bind to membranes via interactions with phosphatidylserine head groups of lipids. PLoS One. 2011;6:e19064. [63] Tan NH. The Biochemistry of venoms of some venomous snakes of Malaysia. Tropical Biomedicine. 1991;8. [64] Debnath A, Saha A, Gomes A, Biswas S, Chakrabarti P, Giri B, et al. A lethal cardiotoxiccytotoxic protein from the Indian monocellate cobra (Naja kaouthia) venom. Toxicon. 2010;56:569-79. [65] Joubert FJ, Taljaard N. Purification, some properties and amino-acid sequences of two phospholipases A (CM-II and CM-III) from Naja naja kaouthia venom. Eur J Biochem. 1980;112:493-9. [66] Karlsson E. Chemistry of protein toxins in snake venoms. In: Chen-Yuan, Lee. (Ed.) Snake Venoms. Handbook of Experimental Pharmacology Springer-Verlag, Berlin-Heidelberg-New York. 1979;52. [67] Mukherjee AK. Correlation between the phospholipids domains of the target cell membrane and the extent of Naja kaouthia PLA(2)-induced membrane damage: evidence of distinct catalytic and cytotoxic sites in PLA(2) molecules. Biochim Biophys Acta. 2007;1770:187-95. [68] Mukherjee AK, Kalita B, Thakur R. Two Acidic, Anticoagulant PLA2 Isoenzymes Purified from the Venom of Monocled Cobra Naja kaouthia Exhibit Different Potency to Inhibit Thrombin and Factor Xa via Phospholipids Independent, Non-Enzymatic Mechanism. PLoS One. 2014;9:e101334. [69] Yap MK, Tan NH, Sim SM, Fung SY, Tan CH. Pharmacokinetics of Naja sumatrana (equatorial spitting cobra) venom and its major toxins in experimentally envenomed rabbits. PLoS Negl Trop Dis. 2014;8:e2890. [70] Heyborne WH, Mackessy SP. Cysteine-rich secretory proteins in reptile venoms. In: Mackessy, S.P. (Ed.) Handbook of venoms and toxins of reptiles (pp. 325-336). Boca Raton, FL: Taylor and Francis Group, CRC Press. 2009. [71] Chang LS, Liou JC, Lin SR, Cheng YC. Purification and characterization of Taiwan cobra venom proteins with weak toxicity. Toxicon. 2005;45:21-5. [72] Wang J, Shen B, Guo M, Lou X, Duan Y, Cheng XP, et al. Blocking effect and crystal structure of natrin toxin, a cysteine-rich secretory protein from Naja atra venom that targets the BKCa channel. Biochemistry. 2005;44:10145-52. [73] Wang F, Li H, Liu MN, Song H, Han HM, Wang QL, et al. Structural and functional analysis of natrin, a venom protein that targets various ion channels. Biochem Biophys Res Commun. 2006;351:443-8. [74] Zhou Q, Wang QL, Meng X, Shu Y, Jiang T, Wagenknecht T, et al. Structural and functional characterization of ryanodine receptor-natrin toxin interaction. Biophys J. 2008;95:4289-99. [75] Markland FS, Jr., Swenson S. Snake venom metalloproteinases. Toxicon. 2013;62:3-18. [76] Fox JW, Serrano SM. Insights into and speculations about snake venom metalloproteinase (SVMP) synthesis, folding and disulfide bond formation and their contribution to venom complexity. FEBS J. 2008;275:3016-30. [77] Fox JW, Serrano SM. Timeline of key events in snake venom metalloproteinase research. J Proteomics. 2009;72:200-9. [78] Fox JW, Serrano SM. Structural considerations of the snake venom metalloproteinases, key members of the M12 reprolysin family of metalloproteinases. Toxicon. 2005;45:969-85. [79] Tan NH. L-Amino acid oxidases and lactate dehydrogenases. In G. S. Bailey (Ed.). Enzymes from snake venom (pp. 579-598). Fort Collins, CO: Alaken. 1998. [80] Mackessy SP. Biochemistry and pharmacology of colubrid snake venoms. Toxin Reviews. 2002;21:43-83. [81] Tan CH, Tan NH, Sim SM, Fung SY, Gnanathasan CA. Proteomic investigation of Sri Lankan hump-nosed pit viper (Hypnale hypnale) venom. Toxicon. 2015;93:164-70. 31

[82] Tan NH, Swaminathan S. Purification and properties of the L-amino acid oxidase from monocellate cobra (Naja naja kaouthia) venom. Int J Biochem. 1992;24:967-73. [83] Sakurai Y, Takatsuka H, Yoshioka A, Matsui T, Suzuki M, Titani K, et al. Inhibition of human platelet aggregation by L-amino acid oxidase purified from Naja naja kaouthia venom. Toxicon. 2001;39:1827-33. [84] Vogel CW, Muller-Eberhard HJ. The cobra venom factor-dependent C3 convertase of human complement. A kinetic and thermodynamic analysis of a protease acting on its natural high molecular weight substrate. J Biol Chem. 1982;257:8292-9. [85] Pung YF, Wong PT, Kumar PP, Hodgson WC, Kini RM. Ohanin, a novel protein from king cobra venom, induces hypolocomotion and hyperalgesia in mice. J Biol Chem. 2005;280:13137-47. [86] Pung YF, Kumar SV, Rajagopalan N, Fry BG, Kumar PP, Kini RM. Ohanin, a novel protein from king cobra venom: its cdna and genomic organization. Gene. 2006;371:246-56. [87] Shafqat J, Beg OU, Yin SJ, Zaidi ZH, Jornvall H. Primary structure and functional properties of cobra (Naja naja naja) venom Kunitz-type trypsin inhibitor. Eur J Biochem. 1990;194:337-41. [88] Zhang Y, Wu J, Yu G, Chen Z, Zhou X, Zhu S, et al. A novel natriuretic peptide from the cobra venom. Toxicon. 2011;57:134-40. [89] Russell FE, Buess FW, Woo MY. Zootoxicological properties of venom phosphodiesterase. Toxicon. 1963;1:99-108. [90] Dhananjaya BL, D'Souza CJ. The pharmacological role of nucleotidases in snake venoms. Cell Biochem Funct. 2010;28:171-7. [91] Aird SD. The role of purine and pyrimidine nucleosides in snake venoms. In S. P. Mackessy (Ed.). Handbook of venoms and toxins of reptiles (pp. 393-419). Boca Raton, FL: Taylor and Francis Group, CRC Press. 2009. [92] Aird SD. Ophidian envenomation strategies and the role of purines. Toxicon. 2002;40:335-93. [93] Du XY, Clemetson KJ. Reptile C-Type Lectins. In S. P. Mackessy (Ed.). Handbook of venoms and toxins of reptiles (pp. 359-375). Boca Raton, FL: Taylor and Francis Group, CRC Press. 2009. [94] Lavin MF, Earl S, Birrell G, St. Pierre L, Guddat LW, J. Jd, et al. Snake venom nerve growth factors. In S. P. Mackessy (Ed.). Handbook of venoms and toxins of reptiles (pp. 377-391). Boca Raton, FL: Taylor and Francis Group, CRC Press. 2009. [95] Katzir I, Shani J, Goshen G, Sela J, Ninary E, Dogonovski AM, et al. Characterization of nerve growth factors (NGFs) from snake venoms by use of a novel, quantitative bioassay utilizing pheochromocytoma (PC12) cells overexpressing human trka receptors. Toxicon. 2003;42:481-90. [96] Li S, Wang J, Zhang X, Ren Y, Wang N, Zhao K, et al. Proteomic characterization of two snake venoms: Naja naja atra and Agkistrodon halys. Biochem J. 2004;384:119-27. [97] Ali SA, Yang DC, Jackson TN, Undheim EA, Koludarov I, Wood K, et al. Venom proteomic characterization and relative antivenom neutralization of two medically important Pakistani elapid snakes (Bungarus sindanus and Naja naja). J Proteomics. 2013;89:15-23. [98] Fernandez J, Alape-Giron A, Angulo Y, Sanz L, Gutierrez JM, Calvete JJ, et al. Venomic and antivenomic analyses of the Central American coral snake, Micrurus nigrocinctus (Elapidae). J Proteome Res. 2011;10:1816-27. [99] Paiva O, Pla D, Wright CE, Beutler M, Sanz L, Gutiérrez JM, et al. Combined venom gland cdna sequencing and venomics of the New Guinea small-eyed snake, Micropechis ikaheka. Journal of Proteomics. 2014;110:209-29. [100] Viravan C, Veeravat U, Warrell MJ, Theakston RD, Warrell DA. ELISA confirmation of acute and past envenoming by the monocellate Thai cobra (Naja kaouthia). Am J Trop Med Hyg. 1986;35:173-81. [101] Fletcher JE, Yang CC, Rosenberg P. Basic phospholipase A2 from Naja nigricollis snake venom: phospholipid hydrolysis and effects on electrical and contractile activity of the rat heart. Toxicol Appl Pharmacol. 1982;66:39-54. [102] Chu ER, Weinstein SA, White J, Warrell DA. Venom ophthalmia caused by venoms of spitting elapid and other snakes: Report of ten cases with review of epidemiology, clinical features, pathophysiology and management. Toxicon. 2010;56:259-72. [103] Wuster W, Thorpe RS. Asiatic cobras: systematics and snakebite. Experientia. 1991;47:205-9. [104] Mukherjee AK. Phospholipase A2-interacting weak neurotoxins from venom of monocled cobra Naja kaouthia display cell-specific cytotoxicity. Toxicon. 2008;51:1538-43. 32

[105] Ahmed SM, Ahmed M, Nadeem A, Mahajan J, Choudhary A, Pal J. Emergency treatment of a snake bite: Pearls from literature. J Emerg Trauma Shock. 2008;1:97-105. [106] Raweerith R, Ratanabanangkoon K. Immunochemical and biochemical comparisons of equine monovalent and polyvalent snake antivenoms. Toxicon. 2005;45:369-75. 33

Captions / legends for figures Figure 1 Figure 2 RP-HPLC liquid chromatogram obtained from C 18 reverse-phase fractionation of N. kaouthia venoms (3 mg) sourced from three different geographical locations and SDS-PAGE profiles of the individual fractions under reducing conditions. A broad range protein ladder (10 to 260 kda) was used for calibration. A) N. kaouthia from Malaysia. B) N. kaouthia from Thailand. C) N. kaouthia from Vietnam. Pie charts showing the relative abundance of venom protein families identified by mass spectrometry following reverse-phase HPLC and SDS-PAGE of N. kaouthia venoms. A) N. kaouthia from Malaysia. B) N. kaouthia from Thailand. C) N. kaouthia from Vietnam. Abbreviations: 3FTx; three-finger toxin, LNTX; long-chain neurotoxin, SNTX; short-chain neurotoxin, CTX; cytotoxin/cardiotoxin, MTLP; muscarinic toxinlike protein, WTX; weak neurotoxin/toxin, PLA 2 ; phospholipase A 2, CRISP; cysteine-rich secretory protein, SVMP; snake venom metalloproteinase, LAAO; L-amino acid oxidase, CVF; cobra venom factor, KUN; Kunitz-type protease inhibitor, NP; natriuretic peptide, PDE; phosphodiesterase, 5 NUC; 5 nucleotidase, CTL; c-type lectin, and NGF; nerve growth factor. Left: 3FTx subtypes; Right: Others protein families in the venoms. 34

Fig 1 35

Fig 2 36

Graphical abstract 37