Combined venomics, antivenomics and venom gland transcriptome analysis of the monocoled cobra (Naja kaouthia) from China

Journal of Proteomics 159 (2017) 19–31

Contents lists available at ScienceDirect

Journal of Proteomics journal homepage: www.elsevier.com/locate/jprot

Combined venomics, antivenomics and venom gland transcriptome analysis of the monocoled cobra (Naja kaouthia) from China Ning Xu a, Hong-Yan Zhao a, Yin Yin a, Shan-Shan Shen a, Lin-Lin Shan a, Chuan-Xi Chen a, Yan-Xia Zhang a, Jian-Fang Gao a,⁎, Xiang Ji b,⁎ a b

Hangzhou Key Laboratory for Animal Adaptation and Evolution, College of Life and Environmental Sciences, Hangzhou Normal University, Hangzhou 310036, Zhejiang, China Jiangsu Key Laboratory for Biodiversity and Biotechnology, College of Life Sciences, Nanjing Normal University, Nanjing 210023, Jiangsu, China

a r t i c l e

i n f o

Article history: Received 7 December 2016 Received in revised form 23 February 2017 Accepted 28 February 2017 Available online 3 March 2017 Keywords: Venomics Antivenomics Transcriptome Naja kaouthia Toxicological activity Enzymatic activity

a b s t r a c t We conducted an omics-analysis of the venom of Naja kaouthia from China. Proteomics analysis revealed six protein families [three-finger toxins (3-FTx), phospholipase A2 (PLA2), nerve growth factor, snake venom metalloproteinase (SVMP), cysteine-rich secretory protein and ohanin], and venom-gland transcriptomics analysis revealed 28 protein families from 79 unigenes. 3-FTx (56.5% in proteome/82.0% in transcriptome) and PLA2 (26.9%/13.6%) were identified as the most abundant families in venom proteome and venom-gland transcriptome. Furthermore, N. kaouthia venom expressed strong lethality (i.p. LD50: 0.79 μg/g) and myotoxicity (CK: 5939 U/l) in mice, and showed notable activity in PLA2 but weak activity in SVMP, L-amino acid oxidase or 5′ nucleotidase. Antivenomic assessment revealed that several venom components (nearly 17.5% of total venom) from N. kaouthia could not be thoroughly immunocaptured by commercial Naja atra antivenom. ELISA analysis revealed that there was no difference in the cross-reaction between N. kaouthia and N. atra venoms against the N. atra antivenom. The use of commercial N. atra antivenom in treatment of snakebites caused by N. kaouthia is reasonable, but design of novel antivenom with the attention on enhancing the immune response of non-immunocaptured components should be encouraged. Biological significance: The venomics, antivenomics and venom-gland transcriptome of the monocoled cobra (Naja kaouthia) from China have been elucidated. Quantitative and qualitative differences are evident when venom proteomic and venom-gland transcriptomic profiles are compared. Two protein families (3-FTx and PLA2) are found to be the predominated components in N. kaouthia venom, and considered as the major players in functional role of venom. Other protein families with relatively low abundance appear to be minor in the functional significance. Antivenomics and ELISA evaluation reveal that the N. kaouthia venom can be effectively immunorecognized by commercial N. atra antivenom, but still a small number of venom components could not be thoroughly immunocaptured. The findings indicate that exploring the precise composition of snake venom should be executed by an integrated omics-approach, and elucidating the venom composition is helpful in understanding composition-function relationships and will facilitate the clinical application of antivenoms. © 2017 Elsevier B.V. All rights reserved.

1. Introduction The evolution of venoms in advanced snakes allows them to transit from a mechanical (constriction) to a chemical (venoms) means of predation, and play an important role in the diversification of the snakes involved [1]. The variability in snake venoms is caused by recruitment of new toxin genes or explosive diversification of existing toxin genes, which occurred before and across the evolutionary process of the advanced snakes [2,3]. The rapidly evolved toxins are speculated to be heavily driven by natural selection [4,5]. Generally, the complexity in functional role is highly correlated with the variability in venom ⁎ Corresponding authors. E-mail addresses: [email protected] (J.-F. Gao), [email protected] (X. Ji).

http://dx.doi.org/10.1016/j.jprot.2017.02.018 1874-3919/© 2017 Elsevier B.V. All rights reserved.

composition and abundance [6–9], and thus elucidating the venomics and venom gland transcriptomics of snakes would be helpful for clinical management of snakebites, filtration of medical components and preparation of antivenoms. There are some 60 species of venomous snakes in China, and the envenomation burden caused by snakebites is heavy in the country [10, 11]. Here, we studied the monocled cobra (Naja kaouthia), a clinically important cobra that is widely distributed in Northeast India, Bangladesh, Malaysia, Indochina Peninsula, Nepal and Southwest China (Yunnan, Sichuan and Guangxi Provinces) [12]. Envenomation by N. kaouthia is capable of inducing several very serious symptoms including ptosis, dysphagia and increased salivation, even leading to coma and death from respiratory paralysis [13–15]. One study conducted recently on the proteomic profile and potential role of N. kaouthia venom identified

20

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

12 protein families from crude venom, among which 3-FTx and PLA2 were the predominant components and accounted for 77.5% and 13.5% of the total venom, respectively [16]. Similar proteomic profiles have also been identified in N. kaouthia venoms from Malaysia, Thailand and Vietnam, although there are apparent geographical variations in venom composition and abundance that may correlate with the differences in venom lethality [17]. However, as the venom gland transcriptomic profile of N. kaouthia has not yet been studied, we still cannot have a deeper understanding of venom proteomic profile. As is well known, antivenoms are most efficient for treatment of snakebites caused by venomous snakes, and it has been recommended that a pan-regional polyvalent, regional monovalent or monoclonal antivenom should be designed and prepared [18–22]. However, the conflict between high investment in development and relatively low medical market demand of antivenom leads to a low diversity in commercial antivenoms, and victims envenomed by a snake sometimes have to be injected with a antivenom not raised against that snake but a phylogenetically related or congeneric species. For example, snakebites caused by N. kaouthia in China have been treated with a monovalent antivenom against the Chinese cobra Naja atra. Actually, in China the snakebites caused by N. kaouthia and N. atra are not strictly discriminated, because these two cobras have an overlapping range of distribution, are morphologically similar to each other, and even induce similar clinical symptoms. To our knowledge, there is no commercially available antivenom specifically raised against N. kaouthia in China, and there is no adequate proof of preclinical assessment for clinical use of commercial N. atra antivenom in treatment of envenomation caused by N. kaouthia. Here, we used a combined proteomic strategy and Illumina sequencing technology to resolve and compare the venom proteomic and venom gland transcriptomic profiles of N. kaouthia from China. To understand the potential role of venom on envenomation and assess the efficacy of commercially available N. atra antivenom on clinical treatment of envenomation caused by N. kaouthia in China, we also evaluated the toxicological and enzymatic activities of crude venom, and conducted second-generation antivenomics based on N. kaouthia venom and N. atra antivenom.

TFA in water and acetonitrile (ACN, solution B) at the following steps: isocratically (10% B) for 5 min, followed by 10–25% B for 15 min, 25– 45% B for 80 min, and 45–60% B for 20 min. Protein detection was performed at 215 nm. Fractions were collected manually, and concentrated in a Centrifugal Concentrator (CentriVap®, Labconco, USA). The proteins were dissolved in loading buffer, and applied to 18% or 12% SDSPAGE for separation. The gels were stained in 0.2% Coomassie brilliant blue (CBB) R-250, and scanned by UMAx2100 densitometer (Umax Technologies, China). Protein bands in the CBB-stained gels were excised, and subjected to automated reduction with DTT and alkylation with IAA, then digested by trypsin gold (Promega). The tryptic peptides were dried in centrifugal concentrator, and re-dissolved in 1.5 μl of 30% ACN and 0.1% TFA. The solutions were spotted onto a sample holder, air-dried to 0.5 μl, mixed with an equal volume of 5 mg/ml α-cyano-4-hydroxycinnamic acid (ABI) in 50% ACN and 0.1% TFA, dried completely, and subjected to MALDI-TOF/TOF-MS (Autoflex speed™, Bruker Dalton, Germany) according to the manual instructions. While for LC-MS/MS analysis, the tryptic peptides were dissolved in 2% ACN and 0.1% FA, then loaded on a C18 reverse phase column (100 μm × 10 cm, 3 μm resin, Michrom Bioresources, CA) and subjected to nESI-MS/MS (LTQ-Orbitrap, Thermo Electron, Germany) according to the manual instructions. The MS spectra were interpreted by FlexAnalysis or Xcalibur software, and the assignment of amino acid sequence similarity was performed against a non-redundant protein sequence database in NCBI (strict to the taxa Serpentes) using Mascot Search Engine 2.3.02. The mass tolerance was set at 0.6 Da. Carbamidomethyl (C) was set as fixed modification, and Acetyl (N-term) and Oxidation (M) was set as variable modification. We followed Calvete et al. [24] to estimate the relative abundance of protein family. The relative abundance of fractions was calculated by peak area measurement using Empower software (Waters, USA). When the fractions present one protein band in SDS-PAGE, the relative abundance was directly obtained from the peak area measurement; while the fractions present more than one protein band, the relative abundance of each band was estimated by densitometry using Tan4100 software (Tanon Science & Technology, China).

2. Materials and methods

2.3. Venom gland cDNA synthesis and sequencing

2.1. Venom and antivenom

Four days after venom milking, the aforementioned two snakes were anesthetized with sodium pentobarbital (s.c. 30 mg/kg). Venom glands of both sides were removed from each snake, rinsed with RNase-free water, and pooled. Total RNA of venom gland from each snake was extracted using TRIzol (Life Technologies, USA) following the manufacturer's protocol. RNA was purified and dissolved in 100 μl THE RNA storage solution (Ambion, Inc., USA). RNA degradation and contamination was assessed using 1% agarose gel electrophoresis. RNA purity and concentration was measured using the Nanophotometer (Implen, USA) and Qubit 2.0 fluorometer (Life Technologies, USA), and RNA integrity was evaluated using the Agilent 2100 system (Agilent Technologies, USA). The RNA for sequencing was pooled equally from the above two RNA samples. RNAseq libraries were constructed with TruSeq™ RNA Sample Preparation Kit (Illumina, USA) according to the manufacturer's instructions. Briefly, mRNA was purified and enriched using magnetic bead with oligo (dT). The mRNA was treated with fragmentation buffer, and used as a template to synthesize first-strand cDNA with reverse transcriptase and random hexamer primers. Second-strand cDNA was synthesized using dNTPs, DNA polymerase I and RNase H. Doublestranded cDNA was then purified with AMPure XP system (Beckman Coulter, USA), and underwent the processes of end pair, ligation of poly (A) tail and sequencing adapters. The adaptor ligated fragments were selected for PCR-amplification and purified using AMPure XP system (Beckman Coulter, USA) to create the final cDNA libraries. Deep sequencing with the Illumina HiSeq™ 2500 (Illumina, USA) platform was

We collected two adult N. kaouthia in 2014 in Baise, Guangxi, South China (Fig. 1), and brought them to our laboratory in Hangzhou for venom collection. The venom was milked by snake biting on parafilmwrapped jars. Fresh venom was centrifuged to remove impurities for 15 min at 10000g 4 °C, lyophilized, pooled equally and stored at − 80 °C until use. Commercial monovalent N. atra antivenom (batch 20,070,601, expiry date: 06/2010; 1000 IU/vail) purchased from Shanghai Serum Biological Technology Co., Ltd. (Shanghai, China) was prepared from the plasma of horses, and consisted of purified F(ab′)2 fragments. The antivenom was lyophilized and stored at − 80 °C, and re-dissolved when used. Protein concentrations of venom fractions were determined according to Bradford [23]. Concentration of antivenom was determined spectrophotometrically using 1 cm light pathlength cuvette based on an extinction coefficient (ε) of 1.4 for 1 mg/ml protein at 280 nm. 2.2. Isolation and characterization of venom proteins Crude venom (5 mg) was dissolved in 0.1% trifluoroacetic acid (TFA, solution A), centrifuged for 10 min at 10000g 4 °C, then the supernatant was collected and loaded on a Kromasil 300 C18 reverse phase column (250 × 4.6 mm, 5 μm particle, AkzoNobel, Sweden) using a Waters E600 HPLC system (Waters, USA). The flow-rate was set to 1 ml/min, and the venom proteins were separated with a mobile phrase system of 0.1%

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

21

Fig. 1. Characterization of the venom proteins of Naja kaouthia from China. Snake images were taken by Jian-Fang Gao in laboratory. Five milligrams of total venom were applied to a C18 column, and separated as described on Materials and methods. Fractions were collected manually and submitted to molecular determination by SDS-PAGE under reduced conditions. Protein bands were excised, tryptic digested and analyzed by MALDI TOF/TOF or nESI-MS/MS for their assignment to known protein families. The results are shown in Table 1.

performed by Novogene Bioinformatics Technology Co., Ltd., Beijing, China (www.novogene.cn) according to the manufacturer's protocol. 2.4. Transcriptome assembly, annotation and quantification Raw reads generated from Illumina HiSeq™ 2500 were cleaned by removing the adapter sequence. Low quality sequences (Q ≤ 20) and reads of which over 10% of the sequences were unknown (‘N’s) were also eliminated, and the remaining reads were assembled into contigs using Trinity with three independent software modules: Inchworm, Chrysalis and Buttersly [25]. In brief, Inchworm assembled all reads to a k-mer dictionary (default k = 25), and resulting in a collection of linear contigs. Chrysalis pooled contigs if they share at least on k – 1-mer and if the reads span the junction between contigs, and then built de Bruijn graphs from each pool. Butterfly reconciled the de Bruijn graphs with the original reads and paired-end reads, then reconstructed the full-length transcripts for spliced isoforms and teased the transcripts for paralogous genes. The transcript with longest sequence in each gene locus was defined as unigene after removing any redundancy, and used as reference sequence for downstream analyses. Gene annotation was conducted using BlastX with an E-value threshold of 10−5 in three databases, namely: NCBI non-redundant protein sequences database (NR), NCBI nucleotide sequences database

(NT) and Swiss-Prot protein database. To evaluate the abundance of unigenes, all of the original sequencing reads (clean reads) were matched with the reference unigenes using RSEM (RNA-Seq by Expectation Maximization) software with default parameters [26]. The number of reads aligned with a given unigene was defined as readcount, which was transferred to FPKM (expected number of fragments per kilobase of transcript sequence per millions base pairs sequenced) [27] for estimation of abundance. 2.5. Antivenomics analysis A second-generation antivenomics analysis [28] was used to evaluate the ability of commercial N. atra antivenom to recognize the N. kaouthia venom. NHS-activated Sepharose 4 fast flow medium (1 ml) (GE Healthcare) was packed in a column, and washed with 15 CV (column volumes) of 1 mM ice cold HCl, followed by 3 CV of coupling buffer (0.2 M NaHCO3, 0.5 M NaCl, pH 8.3). Then, the matrix was incubated overnight at 4 °C with 50 mg N. atra F(ab′)2 antivenom fragments in 0.5 CV of coupling buffer, the concentration of coupled fragments was estimated measuring the uncoupled fragments by quantitative band densitometry of SDS-PAGE. Non-reacting groups were blocked with 0.5 CV of 0.1 M Tris-HCl, pH 8.0 for 4 h at room temperature using an orbital shaker. After then, the matrix was alternately washed with 3 CV of

22

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

low (0.1 M acetate buffer, 0.5 M NaCl, pH 4.0) and high (0.1 M Tris-HCl, pH 8.5) pH buffer, and repeated six times, and equilibrated with 3 CV of binding buffer (20 mM PBS). For immunoaffinity analysis, 500 μg of N. kaouthia venom was dissolved in 0.5 ml binding buffer and loaded on the column, incubated overnight at 4 °C on the orbital shaker. The antivenom: venom ratio was corresponding to about 8-fold molar per “10 kDa of toxins (the average molecular mass assumed)”. After eluting and collecting the non-retained fractions with binding buffer, the retained fractions were eluted with 3 CV of 0.1 M glycine-HCl, pH 2.0, and neutralized with 1 M Tris-HCl, pH 9.0 immediately. Both nonretained and retained fractions were concentrated, and analyzed by RP-HPLC according to the procedure described above. 2.6. Enzyme-linked immunoassays

and 0.81 U/ml horseradish peroxidase), mixed and incubated at 37 °C for 1 h. The reaction was ended by 50 μl 2 M H2SO4, the absorbance was recorded at 490 nm. H2O2 was used as the standard, and the activity was expressed as nmol of H2O2 degraded/min/mg venom. 2.7.5. 5′-Nucleotidase activity Venom (1.6 μg) was added into 90 μl substrate system (50 mM TrisHCl, PH 7.4, containing 10 mM MgCl2, 50 mM NaCl, 10 mM KCl and 10 mM 5′-AMP), and incubated at 37 °C for 30 min. The reaction was ended by 0.42% ammonium molybdate in 10% L-ascorbic acid and 1 M sulphuric acid at room temperature for 30 min. The absorbance was recorded at 660 nm. KH2PO4 was used as the standard, and the activity was expressed as nmol of inorganic phosphate released/min/mg venom.

Micro-ELISA plate (96 wells) was coated with 100 μl N. kaouthia/N. atra venom (2 μg/ml in 0.1 M Na2CO3-NaHCO3, pH 9.6) per well overnight at 4 °C. After washing three times with PBST (0.05% Tween-20 in 10 mM PBS, pH 7.4), the plate was blocked with 150 μl 2% fat-free milk powder in PBST at 37 °C for 1 h. After washing three times, 100 μl suitably diluted horse serum/commercial N. atra antivenom (initial concentration 4 μg/μl) in 1% BSA in PBST was added into each well and incubated at 37 °C for 1 h. The plate was washed again, and incubated with series diluted HRP-labelled anti-horse IgG (Sigma) at 37 °C for 1 h. Finally, the plate was washed with PBST to rinse out the unbound secondary antibodies. Then, 100 μl substrate solution (0.5 mg/ml OPD and 0.006% H2O2 in 0.15 M citrate buffer, pH 5.0) was added into each well, and incubated at room temperature for 20 min. Aliquots (50 μl) of 2.5 M sulphuric acid was added into the well to stop the reaction, the absorbance was recorded at 490 nm with SpectraMax 384 microplate reader (Molecular Devices, Inc., Canada).

2.7.6. Phospholipase A2 activity Aliquots (5 μl) of venoms with various amounts were added into 195 μl substrate system (0.1 M NaCl, 10 mM CaCl2, 7 mM Triton X100, 0.35% soybean lecithin and 98.8 mM phenol red, pH 7.6) and gently mixed, the absorbance was continuously recorded at 558 nm for 2.5 min at room temperature. The activity was expressed as a change in absorbance of 0.3 unit/min/μg venom.

2.7. Enzymatic and toxicological activity

3.1. Venom proteomics

2.7.1. Lethality Groups of 4–8 male ICR mice (18–22 g) from Laboratory Animal Center of Hangzhou Normal University were intraperitoneally injected with various doses of crude N. kaouthia venom, which was dissolved in 100 μl saline, while the control were only injected with saline of the same volume. Deaths were recorded during 24 h after injection, and the median lethal dose (LD50) was calculated using the SpearmanKarber method.

Unveiling the whole venomics is an important way to understand the potential function of snake venoms [7,9,29–32]. In this study, proteomic characterization of N. kaouthia venom was performed by a proteomic strategy combining RP-HPLC, SDS-PAGE, in-gel digestion, MALDITOF/TOF MS and LC-MS analysis. Sixteen chromatographic fractions (peaks) were resolved in the crude N. kaouthia venom by RP-HPLC, and 50 protein bands from SDS-PAGE were identified by MS analysis (Fig.1 and Table 1). The results indicated that the N. kaouthia venom from China is comprised of six protein families, including three-finger toxin (3-FTx, 56.5% of total venom), phospholipase A2 (PLA2, 26.9%), ohanin (vespryn, 9.2%), cysteine-rich secretory protein (CRISP, 5.4%), snake venom metalloproteinase (SVMP, 1.1%) and nerve growth factor (NGF, 1.0%) (Fig. 3 and Table 1). In summary, chromographic peaks 1– 7 and 10–13 mainly contained high abundance of 3-FTx, while peak 6 contained relatively low abundance of NGF, peaks 8–11 and peaks12– 13 respectively contained PLA2 and ohanin, and peaks15–16 mainly contained SVMP and CRISP. However, compared with previous publications [16,17], the proteomic profile of N. kaouthia venom appeared to be underestimated (only six protein families were identified) in this study. This might be caused by the neglect of the components with relatively low abundance during the procedure of separation, and this neglect would make the comparison with other studies relatively more difficult or less convincing.

2.7.2. Myotoxicity Three male ICR mice (18–22 g) were injected intramuscularly in the right thigh muscle with 4 μg crude N. kaouthia venom in 20 μl saline, while the control were only injected with the saline of the same volume. After 3 h, the mice were bled, and the plasma creatine kinase (CK) activity was measured using the CK assaying kit (Batch 20,151,017; Nanjing Jiancheng Bioengineering Institute, China). The activity was defined in U/L. 2.7.3. Proteolytic activity Venom (40 μg in 0.1 ml ddH2O) was added into 0.5 ml substrate system (0.2 M Tris–HCl, pH 8.5, containing 2% casein from bovine milk), mixed and incubated at 37 °C for 2 h. And 0.6 ml 0.44 M TCA was added into the mixture and incubated at 37 °C for 30 min to end the reaction. After then, the mixture was centrifuged at 12000g, 4 °C for 15 min. Aliquots (0.8 ml) of supernatant were collected, then mixed with 2.0 ml 0.4 M Na2CO3 and 0.4 ml folin reagent, the absorbance was recorded at 660 nm. L-Tyrosine was used as the standard, and the activity was expressed as nmol of L-Tyrosine released/min/mg venom. 2.7.4. L-amino acid oxidase activity Venom (2 μg) was added into 90 μl substrate system (50 mM Tris– HCl, pH 8.0, containing 0.25 mM L-Leucine, 2 mM o-phenylenediamine

2.7.7. Statistical analyses The LD50 was calculated using the Trimmed Spearman-Karber Program 1.5, and expressed as 95% confidence limits. Descriptive statistics of other toxicological and enzymatic activities were calculated using Statistica 8.0 (StatSoft Inc., USA), and expressed as mean ± standard error (SE). 3. Results and discussion

3.2. Venom-gland transcriptomics Exploring the transcriptomic profile of venom gland is essential for interpreting the functional role of snake venom, and it was carried out by next generation sequencing. In our investigation, a total of 59,626,766 pairs of clean reads were filtered from 65,736,528 pairs of raw reads collected from Illumina sequencing, and assembled to 96,138 transcripts by Trinity. A total of 81,450 unigenes were then

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

23

Table 1 Assignment of the reverse-phase chromatographic fractions from Naja kaouthia venom to protein families by MALDI-TOF/TOF and nESI-MS/MS of selected peptide ions from in-gel digested protein bands separated by SDS-PAGE. Peak

%

MW (kDa)

1

4.18 3.90

14 11

2

0.43 0.21

13 10

3

1.33

11

4

2.71

11

5

0.06

28

0.08

23

0.39

13

0.19

11

0.23

9

0.21

26

0.13

22

0.63

14

1.39

11

7

13.56

11

8 9 10

12.79 7.26 0.08

14 13 14

0.15

10

6

Peptide Ion m/z

z

726.8 443.7 658.8 864.9 576.9 591.3 658.3 696.3 1121.5 567.3 620.8 591.3 658.3 675.8 696.3 774.4 1121.5 790.7 591.3 696.3 1121.5 591.3 658.3 696.3 774.4 1121.5 439.2 498.3 613.8 660.3 807.3 1094.5 782.0 815.7 1264.0 591.3 675.8 696.3 774.4 1121.5 790.7 567.3 620.8 647.3 708.3 711.4 564.3 647.3 708.3 711.4 824.4 1131.0 793.4 868.4 647.3 708.3 658.3 696.3 1121.5 407.7 1289.5 835.3 835.3 579.3 594.8 603.3 805.8 849.4 913.8 1178.6 371.2 714.9

2 2 2 2 3 2 2 2 2 2 2 2 2 2 2 2 2 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 2 2 2 2 2 2 3 2 2 2 2 2 2 2 2 2 2 2 3 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

MS/MS-derived sequence

Protein family/species/accession

NGXEXNCCTTDR WWSDHR VBPGVNXNCCR XECHNBBSSBAPTTB XECHNBBSSBAPTTB CFXTPDXTSB TWCDAFCSXR VDXGCAATCPTVB TGVDXBCCSTDNCNPFPTR XTCXXCPEB EXVECCSTDB CFXTPDXTSB TWCDAFCSXR DCPNGHVCYTB VDXGCAATCPTVB RVDXGCAATCPTVB TGVDXBCCSTDNCNPFPTR TGVDXBCCSTDNCNPFPTRB CFXTPDXTSB VDXGCAATCPTVB TGVDXBCCSTDNCNPFPTR CFXTPDXTSB TWCDAFCSXR VDXGCAATCPTVB RVDXGCAATCPTVB TGVDXBCCSTDNCNPFPTR XTCVBEB WHMXVPGR DVXECCSTDB GCAATCPXAENR DVXECCSTDBCNX SXFGVTTEDCPDGBNXCFB SXFGVTTEDCPDGBNXCFBR EBSXFGVTTEDCPDGBNXCFB GCAATCPXAENRDVXECCSTDB CFXTPDXTSB DCPNGHVCYTB VDXGCAATCPTVB RVDXGCAATCPTVB TGVDXBCCSTDNCNPFPTR TGVDXBCCSTDNCNPFPTRB XTCXXCPEB EXVECCSTDB XETACVCVXTB CBNPNPEPSGCR XETACVCVXTBB NPNPEPSGCR XETACVCVXTB CBNPNPEPSGCR XETACVCVXTBB GNTVTVMENVNXDNB GXDSSHWNSYCTETDTFXB TTATDXBGNTVTVMENVNXDNB EDHPVHNXGEHSVCDSVSAWVTB XETACVCVXTB CBNPNPEPSGCR TWCDAFCSXR VDXGCAATCPTVB TGVDXBCCSTDNCNPFPTR RPXSWR GCADTCPVGBPYEMXECCSTDB GDNDACAAAVCDCDR GDNDACAAAVCDCDR XSGCWPYFB GGSGTPVDDXDR NMXBCTVPSR GGNNACAAAVCDCDR TYSYECSBGTXTCB CCBVHDNCYNEAEB XAAXCFAGAPYNNNNYNXDXB XXPXASB MFMMSDXTXPVB

3-FTx (SNX); Naja kaouthia; P60771 3-FTx (SNX); N. kaouthia; P59276

3-FTx (SNX); N. kaouthia; P59276 3-FTx (LNX); N. kaouthia; P01391

3-FTx (WNX); Naja oxiana; P85520 3-FTx (LNX); N. kaouthia; P01391

3-FTx (LNX); N. kaouthia; P01391

3-FTx (LNX); N. kaouthia; P01391

3-FTx (Muscarinic); N. kaouthia; P82463

3-FTx (LNX); N. kaouthia; P01391

3-FTx (SNX); N. kaouthia; P59276 VNGF; N. kaouthia; P61899

VNGF; N. kaouthia; P61899

VNGF; N. kaouthia; P61899 3-FTx (LNX); N. kaouthia; P01391

3-FTx (WNX); N. kaouthia; P82935 PLA2; N. kaouthia; P00596 PLA2; N. kaouthia; P00596 PLA2; N. kaouthia; P00597

3-FTx (CTX); N. kaouthia; P60305 (continued on next page)

24

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

Table 1 (continued) Peak

%

MW (kDa)

11

6.63 10.51

17 13

11,12

2.53

19

12

1.05

29

3.09

17

7.19

13

1.85 3.23

29 23

3.23

16

7.41

13

0.11

71

0.04

57

0.03

46

0.02 0.03

40 33

0.04

31

0.09

25

13

14

Peptide Ion

MS/MS-derived sequence

m/z

z

784.9 835.3 350.2 409.3 444.7 474.7 544.2 552.8 656.3 477.9 767.9 871.4 948.4 1087.4 1361.5 748.4 906.0 371.7 457.2 748.4 906.0 757.4 440.3 474.7 478.7 544.2 552.8 656.3 449.9 747.4 801.9 871.4 906.0 393.7 516.2 648.3 371.7 457.2 748.4 906.0 371.2 572.3 678.4 714.9 792.9 398.2 516.3 543.3 544.3 622.8 686.8 749.4 945.9 1087.5 1497.8 1646.8 1087.5 1497.8 1087.5 532.3 629.7 633.8 785.3 1015.5 532.3 602.3 629.7 633.8 785.3 1007.5 516.2 648.3 771.4 925.4 1052.0

2 2 2 2 2 2 2 2 2 3 2 2 1 1 1 2 2 2 2 2 2 3 2 2 2 2 2 2 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

MFMMSDXTXPVBR GDNDACAAAVCDCDR TVPVBR XXPXAYB MFMVSNB GCXDVCPB YVCCNTDR RGCXDVCPB TCPAGBNXCYB XXPXAYBTCPAGB NXCYBMFMVSNB NSXXVBYVCCNTDR GCXDVCPB YVCCNTDR YVCCNTDRCN FDGSPCVXGSPGFR TVENVGVSBVAPDNPER XVPEER SPPGNWBB FDGSPCVXGSPGFR TVENVGVSBVAPDNPER ADVTFDSNTAFESXVVSPDBB XVPXFYB GCXDVCPB MFMVATPB YVCCNTDR RGCXDVCPB TCPAGBNXCYB MFMVATPBVPVB XVPXFYBTCPAGB NXCYBMFMVATPB NSXXVBYVCCNTDR TVENVGVSBVAPDNPER SVRPTAR CAASCFCR BNACBTEWMB XVPEER SPPGNWBB FDGSPCVXGSPGFR TVENVGVSBVAPDNPER XXPXASB CNBXXPXASB XXPXASBTCPAGB MFMMSDXTXPVB MFMMSDXTXPVBR SFAEWR XPCAABDEB TGCXVPVSPR EHBEYXXR CGTXYCTEXB CGTXYCTEXBB ERPBCXXNBPSR XBPHABCDSEECCEB EHBEYXXR ERPBCXXNBPSR TRVYEMVNYXNTB EHBEYXXR ERPBCXXNBPSR EHBEYXXR VSPTASNMXB SNCPASCFCR BSSCBDDWXB MEWYPEAASNAER VXEGXBCGESXYMSSNAR VSPTASNMXB RVSPTASNMXB SNCPASCFCR BSSCBDDWXB MEWYPEAASNAER VXEGXBCGESXYMSSNAR CAASCFCR BNACBTEWMB HHNVFSNCBSXAB NMXBMEWNSNAABNAB XGCGENXFMSSBPYAWSR

Protein family/species/accession

PLA2; N. kaouthia; P00596 3-FTx (CTX); N. kaouthia; P01445

3-FTx (CTX); Naja naja; P24780

Ohanin/vespryn; N. kaouthia; P82885 Ohanin/vespryn; N. kaouthia; P82885

3-FTx (CTX); Naja atra; Q98959

Ohanin/vespryn; N. kaouthia; P82885 CRISP; N. kaouthia; P84808

ohanin/vespryn; N. kaouthia; P82885

3-FTx (CTX); N. kaouthia; P60305

SVMP-III; N. atra; D5LMJ3

SVMP-III; N. atra; ADF43026

SVMP-III; N. atra; D5LMJ3 SVMP-III; N. atra; D5LMJ3 CRISP; N. atra; Q7T1K6

CRISP; N. atra; Q7T1K6

CRISP; N. kaouthia; P84808

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

25

Table 1 (continued) Peak

15

16

%

MW (kDa)

0.12 0.06

19 14

0.22

204

0.44

70

0.22

30

0.56

26

0.05

21

0.30

18

0.05

16

0.30

14

0.42

53

0.28

26

Peptide Ion m/z

z

835.3 371.2 474.7 544.2 552.8 656.3 714.9 784.9 871.4 1168.5 1195.6 1258.5 1324.6 1451.8 389.2 398.2 516.3 543.3 544.3 549.3 622.8 686.8 499.9 630.9 532.3 629.7 633.8 785.3 1007.0 532.3 532.3 602.3 629.7 633.8 785.3 892.9 1015.0 981.1 1168.5 1195.6 1258.5 1324.6 1451.8 2029.9 524.3 532.3 602.3 629.7 785.3 892.9 1007.0 1168.5 1195.6 1258.5 1324.6 1451.8 532.3 629.7 633.8 892.9 376.2 413.8 480.7 550.3 637.8 846.9 1168.5 1195.7 1258.5 1324.6 1451.8 2029.9

2 2 2 2 2 2 2 2 2 1 1 1 1 1 2 2 2 2 2 2 2 2 3 3 2 2 2 2 2 2 2 2 2 2 2 2 2 3 1 1 1 1 1 1 2 2 2 2 2 2 2 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1

MS/MS-derived sequence

Protein family/species/accession

GDNDACAAAVCDCDR XXPXASB GCXDVCPB YVCCNTDR RGCXDVCPB TCPAGBNXCYB MFMMSDXTXPVB MFMMSDXTXPVBR NSXXVBYVCCNTDR NVDFNSESTR EXVDXHNSXR SNCPASCFCR NVDFNSESTRR BBEXVDXHNSXR MVEPGTB SFAEWR XPCAABDEB TGCXVPVSPR EHBEYXXR FBGAETECR CGTXYCTEXB CGTXYCTEXBB ERPBCXXNBPSR XBPHABCDSEECCEB VSPTASNMXB SNCPASCFCR BSSCBDDWXB MEWYPEAASNAER VXEGXBCGESXYMSSNAR VSPTASNMXB XTNCDSXXB RVSPTASNMXB SNCPASCFCR BSSCBDDWXB MEWYPEAASNAER WANTCSXNHSPDNXR VXEGXBCGESXYMSSNAR XGPPCGDCPSACDNGXCTNPCTXYNB NVDFNSESTR EXVDXHNSXR SNCPASCFCR NVDFNSESTRR BBEXVDXHNSXR VXEGXBCGESXYMSSNAR VSPTASNMXB XTNCDSXXB RVSPTASNMXB SNCPASCFCR MEWYPEAASNAER WANTCSXNHSPDNXR VXEGXBCGESXYMSSNAR NVDFNSESTR EXVDXHNSXR SNCPASCFCR NVDFNSESTRR BBEXVDXHNSXR VSPTASNMXB SNCPASCFCR BSSCBDDWXB WANTCSXNHSPDNXR YXBAEB BTVXXPR TNTPEBDR DYBEYXXR YYNYDBPAXB TNTPEBDRYXBAEB NVDFNSESTR EXVDXHNSXR SNCPASCFCR NVDFNSESTRR BBEXVDXHNSXR VXEGXBCGESXYMSSNAR

PLA2; N. kaouthia; P00596 3-FTx (CTX); N. kaouthia; P60305

CRISP; N. atra; 1XX5_B

SVMP-III; N. atra; D5LMJ3

CRISP; N. atra; Q7T1K6

CRISP; N. atra; Q7T1K6

CRISP; N. atra; 1XX5_B

CRISP; N. kaouthia; P84805

CRISP; N. atra; 1XX5_B

CRISP; N. atra; Q7T1K6

SVMP-III; N. kaouthia; P82942

CRISP; N. atra; 1XX5_B

Abbreviations: 3-FTx, three-finger toxin; NGF, nerve growth factor; PLA2, phospholipase A2; CRISP, cysteine-rich secretory protein; SVMP, snake venom metalloproteinase; WNX, weak neurotoxin; CTX, cytotoxin; LNX, long chain α-neurotoxin; SNX, short chain α-neurotoxin. X: Leu/Ile; B: Lys/Gln. Methionine oxidation is underlined.

26

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

screened from these transcripts (N50/N90 = 1354/261), among which 70,053 unigenes passed the quality filter (FPKM N 1) for further bioinformatic analysis. BLAST alignment between unigenes and sequences in NCBI NR/NT and Swiss-Port databases (strict to serpentes) yielded 25,819 valid annotations, including 25,740 non-toxin coding unigenes and 79 toxin coding unigenes, whereas the remaining 44,234 unigenes could not be annotated. Toxin unigenes were found to be expressed at remarkably higher redundancies (7029.42 FPKM/unigene) compared to non-toxin unigenes (13.33 FPKM/unigene). Bioinformatic analyses indicated that the 79 toxin-coding unigenes (with partial and complete CDS) from venom gland transcriptome belonged to 28 protein families (Figs. 2, 3 and Table S1). The protein components were dominated by 3-FTx (82.0%), followed by PLA2 (13.6%), SVMP (1.1%), ohanin (0.9%) and C-type lectin (CTL, 0.5%). The remaining 23 protein families were only expressed in low abundances with a total FPKM of 1.8%, including antimicrobial peptide/protein (AMP), NGF, venom factor (VF), L-amino acid oxidase (LAAO), kunitztype proteinase inhibitor (KUN), 5′ nucleotidase (5′ NT), phosphodiesterase (PDE), phospholipase B (PLB), hyaluronidase (HA), acetylcholinesterase (AchE), aminopeptidase A/N (AP), cysteine-rich with EGF-like domain (CREGF), dipeptidylpeptidase IV (DPP IV), vascular endothelial growth factor (VEGF), CRISP, venom prothrombin activator (VPA), acid phosphomonoesterase (acid PME), glutaminyl-peptide cyclotransferases (QC), natriuretic peptide (NP), snake venom serine proteinase (SVSP), PLA2 inhibitor, waprin and cysteine-type inhibitor (cystatin). 3.3. Proteomic versus transcriptomic profiles Although the proteomic profile of N. kaouthia venom confirmed that the predominant components (83.4% being 3-FTx and PLA2) were consistent with the transcriptomic profile (95.6% being 3-FTx and PLA2), there were still significant quantitative and qualitative differences between venom proteome and venom-gland transcriptome in most protein families (Fig. 3). 3-FTxs are commonly found in venoms of snakes of the family Elapidae, and recognized as the largest group of non-enzymatic toxins

with a wide variety of functions [10,33–35]. As the most abundant and molecularly diversified toxin family in the venom proteome, 3-FTxs mainly contained cytotoxin (CTX, 27.9%), followed by weak α-neurotoxin (WNX, 14.9%), short chain α-neurotoxin (SNX, 8.7%), long chain α-neurotoxin (LNX, 4.6%) and muscarinic toxin (0.4%) (Fig. 3 and Table 1). This high abundance and diversity of 3-FTxs suggest that they may play a key role in the physiological symptoms of envenomations. While in the venom-gland transcriptome, it seems that 3-FTxs exhibit much higher abundance than in the venom proteome, and the 3-FTx subtypes are also different from those in venom proteome. The 3-FTxs encoded by 11 cDNA unigenes could be further classified into four different subtypes, namely WNX (70.2%), LNX (9.2%), muscarinic (2.4%), CTX (0.1%), SNX (0.02%) and several 3-FTx precursors (0.1%) (Fig. 3, Tables S1 and S2). Two of these unigenes with complete CDS sequences could encode WNX and 3-FTx-like precursor, while nine unigenes were only assembled into cDNA with partial CDS sequence. Similar to most Naja spp. venoms [6,17,36–38], PLA2 was expressed as the second most abundant component in N. kaouthia venom (Fig. 3 and Table 1). Actually, the N. kaouthia venom from China (26.9% in this study) even presented higher abundance in PLA2 than that from Malaysia (23.5%), Thailand (12.2–13.5%) and Vietnam (17.4%) [16,17]. PLA2 encoded by three unigenes (13.6%) was also the second most abundant toxin family in the N. kaouthia venom-gland transcriptome, although it showed less abundance than in venom proteome (Fig. 3 and Table S2). One unigene with relatively high abundance (13.6%) can encode the complete amino acid sequence of PLA2, and is identical to the acidic PLA2 [P00596]. The other two unigenes with very low abundance (0.01%) were only assembled into cDNA with partial CDS sequence. Particularly, ohanin (9.2%) was expressed in a relatively high abundance in N. kaouthia venom from China, and was significantly more abundant than that from Malaysia (0.3%), Thailand (0.6–0.7%) and Vietnam (0.2%) [16,17]. On the other hand, ohanin was only identified to account for 0.9% of the total venom-gland transcriptome. Only one fulllength protein sequence of ohanin (vespryn) could be assembled from N. kaouthia venom-gland transcriptome, which exhibited 92% similarity

Fig. 2. Expression levels of toxin transcripts identified in the venom gland transcriptome of N. kaouthia. 3-FTx, three-finger toxin; PLA2, phospholipase A2; SVMP, snake venom metalloproteinase; CTL, C-type lectin; AMP, antimicrobial peptide; NGF, nerve growth factor; VF, venom factor; LAAO, L-amino acid oxidase; KUN, kunitz-type inhibitor; 5′ NT, 5′ nucleotidase; PDE, phosphodiesterase; PLB, phospholipase B; HA, hyaluronidase; AchE, acetylcholinesterase; AP, aminopeptidase; CREGF, cysteine-rich with EGF-like domain; DPP IV, dipeptidylpeptidase IV; VEGF, vascular endothelial growth factor; CRISP, cysteine-rich secretory protein; VPA, venom prothrombin activator; Acid PME, acid phosphomonoesterase; QC, glutaminyl-peptide cyclotransferases; NP, natriuretic peptide; SVSP, snake venom serine proteinase.

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

27

Fig. 3. Comparison of venom proteomic and venom gland transcriptomic profiles of N. kaouthia. A: venom proteomic profile, the details are listed in Table 1; B: venom gland transcriptomic profile. 3-FTx, three-finger toxin; PLA2, phospholipase A2; SVMP, snake venom metalloproteinase; CTL, C-type lectin; AMP, antimicrobial peptide; NGF, nerve growth factor; VF, venom factor; LAAO, L-amino acid oxidase; KUN, kunitz-type inhibitor; 5′ NT, 5′ nucleotidase; PDE, phosphodiesterase; PLB, phospholipase B; HA, hyaluronidase; AchE, acetylcholinesterase; AP, aminopeptidase; CRISP, cysteine-rich secretory protein; WNX, weak neurotoxin; CTX, cytotoxin; LNX, long chain α-neurotoxin; SNX, short chain α-neurotoxin.

to the venom-gland ohanin [P83234] from Ophiophagus hannah (Fig. 3 and Table S2). CRISP was another protein family that exhibited large divergence in the abundance between venom proteome (5.4%) and venom-gland transcriptome (b0.01%) of N. kaouthia (Fig. 3 and Table S1). Moreover, the abundance of CRISP in venom proteome was also significantly higher than the reported values of 0.8–4.3% [16,17]. Here, the CRISP unigene was only assembled into cDNA with partial CDS sequence. Our result showed that the abundance of SVMP (1.1% proteomics and 1.1% transcriptomics) and NGF (1.0% and 0.3%) appeared not to diverge largely between the venom proteome and venom-gland transcriptome (Fig. 3). Ten unigenes were identified to encode the SVMPs, and only one unigene (accounting for 1.0% of total FPKM) could encode the complete amino acid sequence of a SVMP, which exhibited 94% similarity to the zinc metalloproteinase-disintegrin-like kaouthiagin-like [D3TTC1] from N. atra venom and 99% identity to the kaouthiagin [P82942] from N. kaouthia venom, therefore it was considered a novel sequence (Fig. 3, Tables S1 and S2). The remaining unigenes with low abundance (0.1%) were only assembled into cDNA with partial CDS sequence. The NGF was only encoded by one unigene with complete CDS sequence, and exhibited 98% identity to the reported component ([Q5YF89]) from Naja sputatrix. Although expressed in a low amount (0.5%) of the total venomgland transcriptome in N. kaouthia, CTL was encoded by eight cDNA unigenes, and appeared to be a molecularly diversified protein family (Fig. 2, Tables S1 and S2). Six novel proteins could be encoded by the unigenes with complete CDS sequences. Three of these proteins exhibited over 80% identity to the reported CTLs, and three exhibited b80% identity to the reported CTLs (Table S2). There were two unigenes that could only be assembled into cDNA with partial CDS sequences.

The remaining 44 cDNA unigenes were identified to encode another 21 protein families of minor expression (Figs. 2 and 3), and 17 of these unigenes with complete CDS sequences could encode novel proteins, which have not been cloned and expressed or separated with complete sequences before (Table S2). Additionally, compared to the previous studies [16,17,39,40], there were 15 protein families identified as novel components from N. kaouthia venom-gland transcriptome in this study, including AchE, acid PME, AP, AMP, PLB, CREGF, DPP IV, VEGF, QC, HA, PLA2 inhibitor, VPA, waprin, SVSP and NP (Table S2). Generally, the quantitative and qualitative divergence of venom composition between proteomic and transcriptomic analyses is common among snakes [8,41–50]. Such a “low degree of correspondence” between proteome and transcriptome is always speculated to be caused by translational and post-translational modifications [8,51–54], divergences in sampling time/source of venom and venom gland, inconsistent approaches for relative abundance estimation of proteins and mRNA [6], and even the inevitable overlook of the components of minor expression in proteomic analysis [42]. Therefore, an integrated analysis based on proteomic and transcriptomic approaches is needed to explain the venom profile, and this is especially true for the cases where the proteome may be underestimated. 3.4. Toxicological and enzymatic activity of N. kaouthia venom To understand the correlation between composition and functional role of N. kaouthia venom, toxicological and enzymatic activity of crude venom was determined in this investigation. Conventionally, αneurotoxins of 3-FTx in elapid snakes venoms are thought to act postsynaptically by binding to neuromuscular nicotinic AChRs, which eventually leads to paralysis, respiratory failure and death in the victims [35, 55]. Recently, α-neurotoxins have been demonstrated to be the most

28

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

relevant lethal components of the N. kaouthia venom [16,56]. Our result indicated that the estimated i.p. LD50 of the N. kaouthia venom (defined as China population) in mice was 0.79 μg/g (Table 2). While compared to the venoms from other populations, it seemed that the toxicity of N. kaouthia venom from China was slightly higher than that from Malaysia (i.v. 0.90 μg/g and s.c. 1.00 μg/g) and Vietnam (i.v. 0.90 μg/g and s.c. 1.11 μg/g), but was lower than that from Tailand (i.v. 0.18–0.24 μg/g and s.c. 0.20 μg/g) [16,17]. It has been demonstrated that the geographical divergence in lethality of N. kaouthia venom is mainly related to variation in the abundance of α-neurotoxin [17]. Available data on N. katouthia show that the abundance of α-neurotoxin is highest in venom from Thailand (49.9–53.2%) and lowest in venom from Malaysia (17.2%), with venom from China (28.2%) and Vietnam (28.5%) in between. However, a more reasonable and meaningful comparsion of the lethal toxicities of N. kaouthia venoms from different populations in mice depends on how injection of venom is done. Our adminsteration routes for LD50 testing (i.p. injection) differed from those (i.v. and s.c. injections) in previous studies [16,17], so it is possible that geographical variation in lethal toxicities observed in N. kaouthia might be caused by the difference in the injection routes. In addition to α-neurotoxin, cytotoxin is another important subtype of 3-FTx superfamily that is mainly responsible for the cardiotoxic and cytotoxic activities, and also for the hemolytic and membrane-damaging activities [35,57]. Similar to N. atra venom [58], the myotoxicity induced by N. kaouthia venom was declared to stem from the cytotoxin [13]. Here, the determined myotoxicity of N. kaouthia venom (4 μg) was 5939 ± 1020 U/l (Table 2), and appeared more active than N. atra venom (3263 ± 505 U/l) [6], although the N. kaouthia (27.9%) expressed much less abundant cytotoxin in total venom than N. atra (65.3%). One reasonable explanation is that the myotoxicity could be enhanced by the synergistic interaction between cobra acidic PLA2 and cytotoxins [17,59], while the acidic PLA2 expressed in N. kaouthia venom (26.9%) was more abundant than that in N. atra venom (12.2%). The synergistic interaction of cobra acidic PLA2 and cytotoxins has also been recognized to induce local tissue necrosis [16,17,60]. However, except a slight hemolysis, there was no significant swelling and tissue necrosis in the right thigh of the mice injected with the N. kaouthia venom, possibly due to the relatively low dose of injection (4 μg). Similar to the N. atra venom [6], the soybean lecithin was largely degraded by 0.4 μg N. kaouthia venom (5.11 ± 0.04 U/min/μg), and the activity of venom at 0.8 μg decreased significantly (Fig. 6). N. kaouthia exhibited higher activity in degrading soybean lecithin than N. atra venom (3.33 ± 0.29 U/min/μg), and it is likely attributed to the divergence in the amount of PLA2 expressed in both venoms (26.9% in N. kaouthia vs 12.2% in N. atra). Elapid venoms always express much lower abundance of SVMP than viperid venoms, and hence possess much weaker proteolytic activity. Similar to N. kaouthia venom from Thailand and N. atra venom [6,39], the N. kaouthia venom in this study only degraded casein with a weak activity (1.48 ± 0.05 nM/mg/min). Previous studies have revealed that the NGF plays a multifaceted role in N. kaouthia venom by preventing the death of cells in serum-free medium, inhibiting the growth of Ehrlich tumor cells and blocking the metalloproteinase-mediated platelet collagen receptor [61–63]. Additionally, L-amino acid

oxidase (17.9 ± 0.4 nM/mg/min) and 5′ nucleotidase (139.2 ± 6.5 nM/mg/min) activities were also detected in the N. kaouthia venom, although these two components were not detected in our venom sample. This may be due to the low amounts of these components in venom, and the low abundance of these components probably means the minor contribution to the integrative toxicity of the whole N. kaouthia venom. 3.5. Antivenomics and ELISA evaluation of commercial antivenom against venom Assessing the preclinical efficacy of commercial antivenom has always been necessary before its clinical application, and it has become more convenient since the first- and second-generation antivenomic approaches have been developed [19,28,64,65]. To evaluate the efficacy of commercial N. atra antivenom against the N. kaouthia venom, the second-generation antivenomic analysis was employed in this investigation. The results indicated that most venom components of N. kaouthia could be recognized by N. atra antivenom, in a ratio corresponding to 8-fold molar (antivenom) per “10 kDa of toxins” (venom), while there are still a number of venom components that could not be immunocaptured by N. atra antivenom (Fig. 4). These non-captured

Table 2 Toxicological and enzymatical activities of N. kaouthia venom. Activity

Animal/substrates

Descriptive results

Lethality (LD50, μg/g) Myotoxicity (CK, U/l)a Proteolytic activity (nM/min/mg)a LAAO activity (nM/min/mg)a

ICR mice ICR mice Bovine milk casein L-Leu

0.79 (0.73–0.86) 5939 ± 1020 1.48 ± 0.05 17.9 ± 0.4

5′-NT activity (nM/min/mg)a

AMP

139.2 ± 6.5

LD50: dose of venom that induces death in 50% of injected mice. Values in parentheses are 95% confidence limits. a Data are expressed as mean ± SE (n = 3).

Fig. 4. Immunocapture efficacy of commercial Naja atra antivenom towards N. kaouthia venom by RP-HPLC. Panels A–C show the chromatographic profiles of whole components, the retained components from affinity column, and non-retained components of the N. kaouthia venom, respectively.

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

components accounted for about 17.5% of the total venom based on calculation of chromatographic peak area, and mainly belonged to the 3FTx, PLA2, ohanin and CRISP families. The same proportion (17.5%) was calculated in N. atra based on the antivenomics analysis from our previous study [6], indicating that the commercial N. atra antivenom expressed similar capacity to immunorecognize N. kaouthia and N. atra venoms. The similar capacity was verified by ELISA analysis, which was conventionally employed for evaluating the preclinical efficacy of commercial antivenoms. We found that both N. kaouthia and N. atra venoms could cross-react with the N. atra antivenom in almost the same way, with the cross-reactivity showing a progressive increase as the concentration of antivenom increased (Fig. 5). Therefore, it was indirectly demonstrated that the use of commercial N. atra antivenom in the treatment of victims bitten by N. kaouthia in the past is reasonable. However, as the finding was only based on immunological assay, further interpretation of the cross-reactivity between commercial N. atra antivenom and Chinese N. kaouthia venom can be validated by functional analysis either using animals or suitable clinical study. Empirically, the phylogenetical relationship between venomous snakes is often considered for selection of substitutable antivenoms, but the injection dose of antivenoms is not self-evident. Therefore, it is strongly suggested that the preclinical efficacy of commercial antivenom should be assessed, especially for the antivenom that might be applied for the potential treatment of envenomation caused by heterologous snakes. On the other hand, the commercial antivenom always exhibits weak cross-reactivity in neutralizing the venom components with low molecular mass due to the relatively weak immunogenicity of these components [6,31,43,66–73], and it was also verified in application of the commercial N. atra antivenom against N. kaouthia venom in this study. To improve the efficacy of commercial antivenom against N. kaouthia venom, attention should be paid to enhancing the immune response of venom components with low molecular mass. Alternatively, a polyvalent antivenom raised against the pooled N. kaouthia and N. atra venoms may be designed. 4. Conclusion The findings of this study indicate that N. kaouthia venom is comprised of six protein families in proteomic analysis, including 3-FTx, PLA2, ohanin, CRISP, SVMP and NGF. On the other hand, 28 protein families identified in the venom-gland transcriptome of N. kaouthia were encoded by 79 cDNA unigenes, and dominated by 3-FTx, followed by PLA2, SVMP, ohanin and CTL, while the remaining 23 protein families were only expressed as minor proteins with a total FPKM of 1.8%. Compared to the proteomic profiles of N. kaouthia venom previously reported, 15 novel protein families were identified in the venom-gland

29

5.0

4.0

3.0

2.0

1.0

0.0 0.05 0.10

0.20

0.40

0.80

Fig. 6. Phospholipase A2 activity of N. kaouthia venom determined using soybean lecithin. Data are expressed as mean ± SE (n = 3).

transcriptome. The toxicity of N. kaouthia venom from China was slightly higher than N. kaouthia from Malaysia and Vietnam, but lower than that from Thailand while compared to the previous studies [16,17]. The N. kaouthia venom also exhibited relatively strong myotoxicity and PLA2 activity, but exhibited weak activity of SVMP, LAAO and 5′ NT. It probably means that the low abundance of venom components plays a minor contribution to the integrative function of the whole N. kaouthia venom. Antivenomic assessment revealed that a small number of venom components from N. kaouthia could not be thoroughly immunocaptured by the commercial N. atra antivenom. ELISA analysis revealed that there is no difference in the cross-reaction between N. kaouthia and N. atra venoms against the N. atra antivenom. Therefore, the utilization of commercial N. atra antivenom in treatment of snakebites caused by N. kaouthia is reasonable, but design of an antivenom against the N. kaouthia venom with the attention on enhancing the immune response of non-immunocaptured components should be encouraged in the future. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jprot.2017.02.018.

Conflict of interest We declare that we have no conflicts of interest concerning the manuscript entitled “Combined venomics, antivenomics and venom gland transcriptome analysis of the monocoled cobra (Naja kaouthia) from China”.

3.0

OD (490nm)

N. atra venom N. kaouthia venom N. atra venom N. kaouthia venom

2.0

1.0

Acknowledgements The protocol of animal studies complied with current laws on animal welfare and research in China, and was approved by the Animal Research Ethics Committee of Hangzhou Normal University (AREC201306-018). This work was supported by grants from the Natural Science Foundation of China (31101635 and 31471995), Zhejiang Provincial Foundation of Science (LY14C030007). References

0.0

1

2

4

8

16

32

64

128

Antivenom dilution (× 1000) Fig. 5. Cross-reaction between N. kaouthia/N. atra venom and commercial N. atra antivenom determined by ELISA. Normal horse serum (symbols with no color filling) was used as negative control. Data are expressed as mean ± SE (n = 3).

[1] A. Alape-Girón, L. Sanz, J. Escolano, M. Flores-Díaz, M. Madrigal, M. Sasa, et al., Snake venomics of the lancehead pitviper Bothrops asper: geographic, individual, and ontogenetic variations, J. Proteome Res. 7 (2008) 3556–3571. [2] B.G. Fry, From genome to “venome”: molecular origin and evolution of the snake venom proteome inferred from phylogenetic analysis of toxin sequences and related body proteins, Genome Res. 15 (2005) 403–420. [3] B.G. Fry, H. Scheib, L. van der Weerd, B. Young, J. McNaughtan, S.F.R. Ramjan, et al., Evolution of an arsenal: structural and functional diversification of the venom system in the advanced snakes (Caenophidia), Mol. Cell. Proteomics 7 (2008) 215–246.

30

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

[4] S.D. Aird, S. Aggarwal, A. Villar-Briones, M.M.-Y. Tin, K. Terada, A.S. Mikheyev, Snake venoms are integrated systems, but abundant venom proteins evolve more rapidly, BMC Genomics 16 (2015) 1–20. [5] D.R. Rokyta, K.P. Wray, A.R. Lemmon, E.M. Lemmon, S.B. Caudle, A high-throughput venom-gland transcriptome for the eastern diamondback rattlesnake (Crotalus adamanteus) and evidence for pervasive positive selection across toxin classes, Toxicon 57 (2011) 657–671. [6] L.-L. Shan, J.-F. Gao, Y.-X. Zhang, S.-S. Shen, Y. He, J. Wang, et al., Proteomic characterization and comparison of venoms from two elapid snakes (Bungarus multicinctus and Naja atra) from China, J. Proteomics 138 (2016) 83–94. [7] B. Lomonte, J. Fernandez, L. Sanz, Y. Angulo, M. Sasa, J.M. Gutierrez, et al., Venomous snakes of Costa Rica: biological and medical implications of their venom proteomic profiles analyzed through the strategy of snake venomics, J. Proteomics 105 (2014) 323–339. [8] N.R. Casewell, S.C. Wagstaff, W. Wüster, D.A.N. Cook, F.M.S. Bolton, S.I. King, et al., Medically important differences in snake venom composition are dictated by distinct postgenomic mechanisms, Proc. Natl. Acad. Sci. U. S. A. 111 (2014) 9205–9210. [9] J.J. Calvete, Snake venomics: from the inventory of toxins to biology, Toxicon 75 (2013) 44–62. [10] G.-P. Qin, China Poisonous Snake Research, Guangxi Science and Technology Press, Nanning, China, 1998. [11] A. Kasturiratne, A.R. Wickremasinghe, N. de Silva, N.K. Gunawardena, A. Pathmeswaran, R. Premaratna, et al., The global burden of snakebite: a literature analysis and modelling based on regional estimates of envenoming and deaths, PLoS Med. 5 (2008), e218. [12] E.-M. Zhao, Snakes of China, Anhui Science and Technology Publishing House, Hefei, China, 2006. [13] M. Reali, F.G. Serafim, M.A. da Cruz-Höfling, M.D. Fontana, Neurotoxic and myotoxic actions of Naja naja kaouthia venom on skeletal muscle in vitro, Toxicon 41 (2003) 657–665. [14] P.G. Sells, R.D.G. Theakston, D.A. Warrell, Development of α-neurotoxin antibodies in patients envenomed by the monocellate Thai cobra (Naja kaouthia), Toxicon 32 (1994) 1667–1671. [15] N. Wongtongkam, H. Wilde, C. Sitthi-Amorn, K. Ratanabanangkoon, A study of Thai cobra (Naja kaouthia) bites in Thailand, Mil. Med. 170 (2005) 336–341. [16] A.H. Laustsen, J.M. Gutiérrez, B. Lohse, A.R. Rasmussen, J. Fernández, C. Milbo, et al., Snake venomics of monocled cobra (Naja kaouthia) and investigation of human IgG response against venom toxins, Toxicon 99 (2015) 23–35. [17] K.Y. Tan, C.H. Tan, S.Y. Fung, N.H. Tan, Venomics, lethality and neutralization of Naja kaouthia (monocled cobra) venoms from three different geographical regions of Southeast Asia, J. Proteomics 120 (2015) 105–125. [18] J.M. Gutiérrez, D. Williams, H.W. Fan, D.A. Warrell, Snakebite envenoming from a global perspective: towards an integrated approach, Toxicon 56 (2010) 1223–1235. [19] D.J. Williams, J.-M. Gutiérrez, J.J. Calvete, W. Wüster, K. Ratanabanangkoon, O. Paiva, et al., Ending the drought: new strategies for improving the flow of affordable, effective antivenoms in Asia and Africa, J. Proteomics 74 (2011) 1735–1767. [20] R. Shashidharamurthy, K. Kemparaju, Region-specific neutralization of Indian cobra (Naja naja) venom by polyclonal antibody raised against the eastern regional venom: a comparative study of the venoms from three different geographical distributions, Int. Immunopharmacol. 7 (2007) 61–69. [21] R.A. Harrison, D.A. Cook, C. Renjifo, N.R. Casewell, R.B. Currier, S.C. Wagstaff, Research strategies to improve snakebite treatment: challenges and progress, J. Proteomics 74 (2011) 1768–1780. [22] K. Ratanabanangkoon, K.Y. Tan, S. Eursakun, C.H. Tan, P. Simsiriwong, T. Pamornsakda, et al., A simple and novel strategy for the production of a pan-specific antiserum against elapid snakes of Asia, PLoS Negl. Trop. Dis. 10 (2016), e0004565. [23] M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248–254. [24] J.J. Calvete, P. Juárez, L. Sanz, Snake venomics. Strategy and applications, J. Mass Spectrom. 42 (2007) 1405–1414. [25] M.G. Grabherr, B.J. Haas, M. Yassour, J.Z. Levin, D.A. Thompson, I. Amit, et al., Fulllength transcriptome assembly from RNA-Seq data without a reference genome, Nat. Biotechnol. 29 (2011) 644–652. [26] B. Li, C.N. Dewey, RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome, BMC bioinformatics 12 (2011) 323. [27] C. Trapnell, B.A. Williams, G. Pertea, A. Mortazavi, G. Kwan, M.J. Van Baren, et al., Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation, Nat. Biotechnol. 28 (2010) 511–515. [28] D. Pla, J.M. Gutiérrez, J.J. Calvete, Second generation snake antivenomics: comparing immunoaffinity and immunodepletion protocols, Toxicon 60 (2012) 688–699. [29] J.J. Calvete, L. Sanz, Y. Angulo, B. Lomonte, J.M. Gutiérrez, Venoms, venomics, antivenomics, FEBS Lett. 583 (2009) 1736–1743. [30] D. Georgieva, M. Öhler, J. Seifert, M.V. Bergen, R.K. Arni, N. Genov, et al., Snake venomic of Crotalus durissus terrificus—correlation with pharmacological activities, J. Proteome Res. 9 (2010) 2302–2316. [31] J.-F. Gao, J. Wang, Y. He, Y.-F. Qu, L.-H. Lin, X.-M. Ma, et al., Proteomic and biochemical analyses of short-tailed pit viper (Gloydius brevicaudus) venom: agerelated variation and composition–activity correlation, J. Proteomics 105 (2014) 307–322. [32] N.H. Tan, S.Y. Fung, K.Y. Tan, M.K.K. Yap, C.A. Gnanathasan, C.H. Tan, Functional venomics of the Sri Lankan Russell's viper (Daboia russelii) and its toxinological correlations, J. Proteomics 128 (2015) 403–423. [33] R.M. Kini, R. Doley, Structure, function and evolution of three-finger toxins: mini proteins with multiple targets, Toxicon 56 (2010) 855–867.

[34] A. Roy, X. Zhou, M.Z. Chong, D. D'Hoedt, C.S. Foo, N. Rajagopalan, et al., Structural and functional characterization of a novel homodimeric three-finger neurotoxin from the venom of Ophiophagus hannah (king cobra), J. Biol. Chem. 285 (2010) 8302–8315. [35] B.G. Fry, W. Wüster, R.M. Kini, V. Brusic, A. Khan, D. Venkataraman, et al., Molecular evolution and phylogeny of elapid snake venom three-finger toxins, J. Mol. Evol. 57 (2003) 110–129. [36] L.P. Lauridsen, A.H. Laustsen, B. Lomonte, J.M. Gutiérrez, Exploring the venom of the forest cobra snake: toxicovenomics and antivenom profiling of Naja melanoleuca, J. Proteomics 150 (2017) 98–108. [37] K. Sintiprungrat, K. Watcharatanyatip, W.D.S.T. Senevirathne, P. Chaisuriya, D. Chokchaichamnankit, C. Srisomsap, et al., A comparative study of venomics of Naja naja from India and Sri Lanka, clinical manifestations and antivenomics of an Indian polyspecific antivenom, J. Proteomics 132 (2016) 131–143. [38] D. Petras, L. Sanz, A. Segura, M. Herrera, M. Villalta, D. Solano, 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. 10 (2011) 1266–1280. [39] C.M. Modahl, A.K. Mukherjee, S.P. Mackessy, An analysis of venom ontogeny and prey-specific toxicity in the monocled cobra (Naja kaouthia), Toxicon 119 (2016) 8–20. [40] K. Kulkeaw, W. Chaicumpa, Y. Sakolvaree, P. Tongtawe, P. Tapchaisri, Proteome and immunome of the venom of the Thai cobra, Naja kaouthia, Toxicon 49 (2007) 1026–1041. [41] L. Sanz, J. Escolano, M. Ferretti, M.J. Biscoglio, E. Rivera, E.J. Crescenti, et al., Snake venomics of the South and Central American bushmasters. Comparison of the toxin composition of Lachesis muta gathered from proteomic versus transcriptomic analysis, J. Proteomics 71 (2008) 46–60. [42] C. Corrêa-Netto, I.L. Junqueira-de-Azevedo, D.A. Silva, P.L. Ho, M. Leitão-de-Araújo, M.L. Alves, et al., Snake venomics and venom gland transcriptomic analysis of Brazilian coral snakes, Micrurus altirostris and M. corallinus, J. Proteomics 74 (2011) 1795–1809. [43] L. Gonçalves-Machado, D. Pla, L. Sanz, R.J.B. Jorge, M. Leitão-De-Araújo, M.L.M. Alves, et al., Combined venomics, venom gland transcriptomics, bioactivities, and antivenomics of two Bothrops jararaca populations from geographic isolated regions within the Brazilian Atlantic rainforest, J. Proteomics 135 (2016) 73–89. [44] R.H. Valente, P.R. Guimarães, M. Junqueira, A.G.C. Neves-Ferreira, M.R. Soares, A. Chapeaurouge, et al., Bothrops insularis venomics: a proteomic analysis supported by transcriptomic-generated sequence data, J. Proteomics 72 (2009) 241–255. [45] S.C. Wagstaff, L. Sanz, P. Juárez, R.A. Harrison, J.J. Calvete, Combined snake venomics and venom gland transcriptomic analysis of the ocellated carpet viper, Echis ocellatus, J. Proteomics 71 (2009) 609–623. [46] J.J. Calvete, C. Marcinkiewicz, L. Sanz, Snake venomics of Bitis gabonica gabonica. Protein family composition, subunit organization of venom toxins, and characterization of dimeric disintegrins bitisgabonin-1 and bitisgabonin-2, J. Proteome Res. 6 (2007) 326–336. [47] R.S. Rodrigues, J. Boldrini-França, F.P.P. Fonseca, P. de la Torre, F. Henrique-Silva, L. Sanz, et al., Combined snake venomics and venom gland transcriptomic analysis of Bothropoides pauloensis, J. Proteomics 75 (2012) 2707–2720. [48] M.J. Margres, J.J. McGivern, K.P. Wray, M. Seavy, K. Calvin, D.R. Rokyta, Linking the transcriptome and proteome to characterize the venom of the eastern diamondback rattlesnake (Crotalus adamanteus), J. Proteomics 96 (2014) 145–158. [49] O. Paiva, D. Pla, C.E. Wright, M. Beutler, L. Sanz, J.M. Gutiérrez, et al., Combined venom gland cDNA sequencing and venomics of the New Guinea small-eyed snake, Micropechis ikaheka, J. Proteomics 110 (2014) 209–229. [50] C.H. Tan, K.Y. Tan, S.Y. Fung, N.H. Tan, Venom-gland transcriptome and venom proteome of the Malaysian king cobra (Ophiophagus hannah), BMC Genomics 16 (2015) 687. [51] J.-F. Gao, Y.-F. Qu, X.-Q. Zhang, Y. He, X. Ji, Neonate-to-adult transition of snake venomics in the short-tailed pit viper, Gloydius brevicaudus, J. Proteomics 84 (2013) 148–157. [52] A. Zelanis, S.M.T. Serrano, V.N. Reinhold, N-glycome profiling of Bothrops jararaca newborn and adult venoms, J. Proteomics 75 (2012) 774–782. [53] L.T. García, L.T. Parreiras e Silva, O.H.P. Ramos, A.K. Carmona, P.A. Bersanetti, H.S. Selistre-de-Araujo, The effect of post-translational modifications on the hemorrhagic activity of snake venom metalloproteinases, Comp. Biochem. Physiol. C 138 (2004) 23–32. [54] S. Aird, Y. Watanabe, A. Villar-Briones, M. Roy, K. Terada, A. Mikheyev, Quantitative high-throughput profiling of snake venom gland transcriptomes and proteomes (Ovophis okinavensis and Protobothrops flavoviridis), BMC Genomics 14 (2013) 790. [55] Y.N. Utkin, V.V. Kukhtina, E.V. Kryukova, F. Chiodini, D. Bertrand, C. Methfessel, et al., “Weak toxin” from Naja kaouthia is a nontoxic antagonist of α7 and muscle-type nicotinic acetylcholine receptors, J. Biol. Chem. 276 (2001) 15810–15815. [56] K. Tan, C. Tan, S. Fung, N. Tan, Neutralization of the principal toxins from the venoms of Thai Naja kaouthia and Malaysian Hydrophis schistosus: insights into toxin-specific neutralization by two different antivenoms, Toxins 8 (2016) 86. [57] A.V. Feofanov, G.V. Sharonov, M.A. Dubinnyi, M.V. Astapova, I.A. Kudelina, P.V. Dubovskii, et al., Comparative study of structure and activity of cytotoxins from venom of the cobras Naja oxiana, Naja kaouthia, and Naja haje, Biochemistry (Moscow) 69 (2004) 1148–1157. [58] L. Xue, G. Liu, Q. Wang, K. Li, Acute and long-term toxicities of Chinese cobra (Naja naja atra) venom cytotoxin in experimental animals, J. Fujian Med. Univ. 30 (1996) 109–113. [59] A.K. Mukherjee, Correlation between the phospholipids domains of the target cell membrane and the extent of Naja kaouthia PLA2-induced membrane damage:

N. Xu et al. / Journal of Proteomics 159 (2017) 19–31

[60] [61]

[62]

[63]

[64]

[65]

[66]

[67]

evidence of distinct catalytic and cytotoxic sites in PLA2 molecules, BBA-Gen. Subjects 1770 (2007) 187–195. C.-M. Teng, Y.-P. Kuo, L.-G. Lee, C. Ouyang, Characterization of the anticoagulants from Taiwan cobra (Naja naja atra) snake venom, Toxicon 25 (1987) 201–210. V.V. Kukhtina, V.I. Tsetlin, Y.N. Utkin, L.S. Inozemtseva, I.A. Grivennikov, Two forms of nerve growth factor from cobra venom prevent the death of PC12 cells in serum-free medium, J. Nat. Toxins 10 (2001) 9–16. L.C. Wijeyewickrema, E.E. Gardiner, E.L. Gladigau, M.C. Berndt, R.K. Andrews, Nerve growth factor inhibits metalloproteinase-disintegrins and blocks ectodomain shedding of platelet glycoprotein VI, J. Biol. Chem. 285 (2010) 11793–11799. A.V. Osipov, T.I. Terpinskaya, E.V. Kryukova, V.S. Ulaschik, L.V. Paulovets, E.A. Petrova, et al., Nerve growth factor from cobra venom inhibits the growth of Ehrlich tumor in mice, Toxins 6 (2014) 784–795. B. Lomonte, J. Escolano, J. Fernández, L. Sanz, Y. Angulo, J.M. Gutiérrez, et al., Snake venomics and antivenomics of the arboreal neotropical pitvipers Bothriechis lateralis and Bothriechis schlegelii, J. Proteome Res. 7 (2008) 2445–2457. J.M. Gutiérrez, G. Solano, D. Pla, M. Herrera, Á. Segura, M. Villalta, et al., Assessing the preclinical efficacy of antivenoms: from the lethality neutralization assay to antivenomics, Toxicon 69 (2013) 168–179. M. Villalta, D. Pla, S.L. Yang, L. Sanz, A. Segura, M. Vargas, et al., Snake venomics and antivenomics of Protobothrops mucrosquamatus and Viridovipera stejnegeri from Taiwan: keys to understand the variable immune response in horses, J. Proteomics 75 (2012) 5628–5645. L. Fahmi, B. Makran, D. Pla, L. Sanz, N. Oukkache, M. Lkhider, et al., Venomics and antivenomics profiles of North African Cerastes cerastes and C. vipera populations reveals a potentially important therapeutic weakness, J. Proteomics 75 (2012) 2442–2453.

31

[68] D. Pla, L. Sanz, P. Molina-Sánchez, V. Zorita, M. Madrigal, M. Flores-Díaz, et al., Snake venomics of Lachesis muta rhombeata and genus-wide antivenomics assessment of the paraspecific immunoreactivity of two antivenoms evidence the high compositional and immunological conservation across Lachesis, J. Proteomics 89 (2013) 112–123. [69] J.M. Gutierrez, B. Lomonte, L. Sanz, J.J. Calvete, D. Pla, Immunological profile of antivenoms: preclinical analysis of the efficacy of a polyspecific antivenom through antivenomics and neutralization assays, J. Proteomics 105 (2014) 340–350. [70] R.J. Jorge, H.S. Monteiro, L. Goncalves-Machado, M.C. Guarnieri, R.M. Ximenes, D.M. Borges-Nojosa, et al., Venomics and antivenomics of Bothrops erythromelas from five geographic populations within the Caatinga ecoregion of northeastern Brazil, J. Proteomics 114 (2015) 93–114. [71] A.J. Saviola, D. Pla, L. Sanz, T.A. Castoe, J.J. Calvete, S.P. Mackessy, Comparative venomics of the prairie rattlesnake (Crotalus viridis viridis) from Colorado: identification of a novel pattern of ontogenetic changes in venom composition and assessment of the immunoreactivity of the commercial antivenom CroFab®, J. Proteomics 121 (2015) 28–43. [72] K.Y. Wong, C.H. Tan, N.H. Tan, Venom and purified toxins of the spectacled cobra (Naja naja) from Pakistan: insights into toxicity and antivenom neutralization, Am.J.Trop. Med. Hyg. 94 (2016) 1392–1399. [73] P.K. Leong, S.Y. Fung, C.H. Tan, S.M. Sim, N.H. Tan, Immunological cross-reactivity and neutralization of the principal toxins of Naja sumatrana and related cobra venoms by a Thai polyvalent antivenom (neuro polyvalent snake antivenom), Acta Trop. 149 (2015) 86–93.