Proteomics and immunocharacterization of Asian mountain pit viper (Ovophis monticola) venom

The venomic profile of Asian mountain pit viper Ovophis monticola is clarified in the present study. Using mass spectrometry-based proteomics, 247 different proteins were identified in crude venom of O. monticola found in Thailand. The most abundant proteins were snake venom metalloproteases (SVMP) (36.8%), snake venom serine proteases (SVSP) (31.1%), and phospholipases A2 (PLA2) (12.1%). Less abundant proteins included L-amino acid oxidase (LAAO) (5.7%), venom nerve growth factor (3.6%), nucleic acid degrading enzymes (3.2%), C-type lectins (CTL) (1.6%), cysteine-rich secretory proteins (CRISP) (1.2%) and disintegrin (1.2%). The immunoreactivity of this viper’s venom to a monovalent antivenom against green pit viper Trimeresurus albolabris, or to a polyvalent antivenom against hemotoxic venom was investigated by indirect ELISA and two-dimensional (2D) immunoblotting. Polyvalent antivenom showed substantially greater reactivity levels than monovalent antivenom. A titer for the monovalent antivenom was over 1:1.28x107 dilution while that of polyvalent antivenom was 1:5.12x107. Of a total of 89 spots comprising 173 proteins, 40 spots of predominantly SVMP, SVSP and PLA2 were specific antigens for antivenoms. The 49 unrecognized spots containing 72 proteins were characterized as non-reactive proteins, and included certain types of CTLs and CRISPs. These neglected venom constituents could limit the effectiveness of antivenom-based therapy currently available for victims of pit viper envenomation.


Introduction
Envenomation from snakebites affects over 2.7 million people in tropical and subtropical countries each year, leading to more than 130,000 deaths among victims [1]. Severe injuries and complications from bites also can lead to permanent disabilities and long-term health a1111111111 a1111111111 a1111111111 a1111111111 a1111111111

Snakes, venom and antivenoms
All O. monticola pit vipers (Fig 1) were captured in the wild and transferred to Snake Farm, Queen Saovabha Memorial Institute (QSMI) before being quarantined. All procedures were performed following the safety protocol for working with venomous snakes (No. SN 001/2016). Routine snake care and the venom collection was conducted according to the specific protocol. All protocols were approved by the Ethic Committee of the Queen Saovabha Memorial Institute Animal Care and Use (No. QSMI-ACUC-02-2018) in accordance with the guideline of the National Research Council of Thailand. Information about individual snakes used in this study is shown in Table 1.

O. monticola venom preparation and one-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE)
Crude venom of O. monticola was mixed with lysis buffer (containing of 1% Triton X-100 (Merck, Germany), 1% sodium dodecyl sulfate (SDS) (Merck, Germany), and 1% NaCl (Merck, Germany). The venom was estimated for protein concentration by Quick Start™ Bradford Protein Assay (Bio-Rad, USA). A 30 μg sample of O. monticola venom was separated by 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (Bio-Rad, USA) and stained by Coomassie R-250 solution (Bio-Rad, USA) as previously described [13]. The whole lane of venom was excised into 10 pieces and further subjected to in-gel digestion.

Two-dimensional polyacrylamide gel electrophoresis (2DE)
A 100 μg protein was mixed with IPG sample buffer containing 8 M urea, 2% (w/v) 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS), 15 mM dithiothreitol (DTT), and 0.5% IPG sample buffer [14]. Afterwards, the protein solution was rehydrated overnight into a non-linear immobilized pH gradient (IPG) strip (pH 3-10; Amersham Bioscience, USA). Isoelectric focusing (pI) was done using an Ettan IPGphorII instrument (Amersham Bioscience, USA) with the following settings: 30 V for 14 h, 200 V for 1 h, 500 V for 1 h, 1000 V for 1 h, 3500 V for 1 h, and 8000 V for 18 h. The IPG strips were equilibrated with DTT for 15 min and with iodoacetamine for 15 min. After incubation, the strips were placed onto a 12% SDS-PAGE gel. All three 2DE gels were stained with silver stain and the immunoreactive spots in these gels were excised and pooled for mass spectrometric analysis. Other two 2DE gels were used for immunoblotting.

In-gel digestion
A mixture of 50% acetonitrile (ACN) in 50 mM ammonium bicarbonate was used for de-staining the blue color from gel slides [13]. Venom proteins were reduced by 4mM DTT and incubated at 60˚C for 15 min. The reduced proteins were further alkylated by 250 mM iodoacetamine (IAA) (Sigma-Aldrich, USA) and incubated at room temperature for 30 min in dark. The gel pieces were dehydrated by removing all solution and adding 100% ACN (Thermo Scientific, USA). For tryptic digestion, trypsin (Sigma-Aldrich, USA, T6567) in 50 mM ammonium bicarbonate (Sigma-Aldrich, USA) was added to rehydrate the gels, which were then incubated overnight at 37˚C. Peptide extraction was performed by adding 100% ACN and incubating for 15 min. The resulting solution was transferred into a new microcentrifuge tube and dried using a centrifugal concentrator (TOMY, Japan). The peptide mixtures were stored at -20˚C prior to mass spectrometric analysis.  [8]. Key characters are body coloration and pattern: predominantly tan or reddish-grey with irregular short, black-edged crossbars or blotches along the vertebral ridge, including smaller irregular dark blotches on both sides of the body along the edges of the dorsal scales (Fig 1). b Voucher no. was attached to each preserved snake after it died.

Mass spectrometric analysis
Venom peptides were dissolved in 0.1% formic acid (Sigma-Aldrich, USA) and subjected to an Ultimate1 3000 Nano-LC systems (Thermo Scientific, USA). The peptides were eluted and infused to a microTOF-Q II (Bruker, Germany). The acquisition was operated by HyStar™ version 3.2 (Bruker, Germany), and the resulting data were processed and converted to mascot generics format (.mgf) files using Compass DataAnalysis™ software version 3.4 (Bruker, Germany). A database search was performed using Mascot Daemon software (Matrix Science, USA) against the NCBI snake database with the following parameters: one missed cleavage site, variable modifications of carbamidomethyl (C) and oxidation (M), 0.8 Da for MS peptide tolerance and 0.8 Da for MS/MS tolerance. The significance threshold was set at 95%. Three biological replications were performed for protein identification.

Indirect enzyme-linked immunosorbent assay (ELISA)
Immunoreactivity of protein antigens in O. monticola venom to monovalent and polyvalent antivenom was assessed by indirect enzyme-linked immunosorbent assay (ELISA) modified from Gawtham and colleagues [15].

Immunoblot analysis
The separated polypeptide spots from 2DE gels were transferred to nitrocellulose membrane for 90 min at 18 V on a Trans-blot semi-dry Transfer CellTM (Biorad) in semi-dry transfer buffer (48 mM Tris and 2.93 g glycine) pH 9.2 containing 20% methanol. The membranes were blocked using 5% (w/v) non-fat milk in PBS for 2 h at room temperature. The membranes were rinsed twice with PBS-T buffer pH 7.4 (8 mM sodium phosphate, 2 mM potassium phosphate, 140 mM NaCl, 2.7 mM KCl and 0.5% v/v Tween) for 30 s each. The blotted membranes were incubated with either monovalent or polyvalent antivenom (1:1000 in 0.2% BSA-PBS). After washing the membrane three times with PBS-T, 50 μL of horseradish peroxidase-conjugated goat anti-horse-IgG (Abcam, Cambridge, UK) in PBS-T (1:2000) was added, and the mixture was incubated for 1 h at ambient temperature under constant agitation. Membranes were washed three times with PBS-T buffer and one time with PBS. Immunogen spots were visualized by detection of peroxidase activity using Ultra TMB-Blotting Solution (Ther-moFisher Scientific, UK).

Statistical analysis
Quantitative data are presented as mean ± SEM. Statistical significance between groups was analyzed using standard t-tests or two-way ANOVA followed by the Bonferroni test. Significant p-values are indicated within the figure panels. Error bars indicate SEM.

Proteomic analysis of O. monticola venom
Detectable proteins in venom of O. monticola were between 10-95 kDa (Fig 2A). Intense protein bands at 10, 15, 50 and 72 kDa and faint bands at 26, 28, 30, 34 and 95 kDa were recorded. There were 247 proteins found in O. monticola venom (a list of all proteins is shown in S1 Table). A classification of constituent proteins based on their biological properties is presented in Fig 2B.  Table 2.

Immunoreactivity of protein antigens in O. monticola venom to monovalent and polyvalent antivenoms by indirect ELISA
Since there is no homospecific antivenom to Ovophis spp. venoms currently available, all pit viper envenoming victims are recommended to receive either monovalent antivenom (raised against T. albolabris venom) or hematotoxic polyvalent antivenom (produced against venoms of C. rhodostoma, D. siamensis and T. albolabris) to alleviate symptoms [16]. Indirect ELISA was used to determine the cross-reactivity of these antivenoms to O. monticola venom. Hematotoxic polyvalent antivenom exhibited a significantly greater level of immunoreactivity than the monovalent antivenom by 30-50% (up to the dilution 1: 1.6x10 6 ), P<0.001 (Fig 3). A titer for the monovalent antivenom was over 1:1.28x10 7 dilution, while that of hematotoxic polyvalent antivenom was 1:5.12x10 7 (Fig 3).

O. monticola venom protein analysis by two-dimensional electrophoresis (2DE)
In order to explore the protein antigens present in O. monticola venom, crude venom was subjected to 2DE gel electrophoresis. There were 89 spots detected, with pI values ranging from 3 to 10 and MW from 10 to 95 kDa. Within the particular MW regions of 10-15, 30-34, 50 and 72-90 lies the greatest abundance of protein spots ( Fig 4A). Using MALDI-TOF/TOF-MS/MS, all protein spots in 2DE gels were identified, and are listed in S2 Table. There were 461 different sequences, which correspond to 173 peptide accession identities.

Immunoreactive proteins in O. monticola venom by immunoblot analysis
The immunoblot analysis was performed with either monovalent or polyvalent antivenom to characterize specific protein antigens within the O. monticola venom. Twenty-six immunoreactive spots were detected with monovalent antivenom, with pI values ranging from 3 to 6, and MW ranging from 17 to 95 kDa. Most of these spots were observed at MW 50 to 95 kDa and pI between 3 to 5 ( Fig 4B). When probed with hematotoxic polyvalent antivenom, 40 immunoreactive spots were recorded with a broader range of pI values from 3 to 8, and MW ranging from 17 to 95 kDa. A high number of the spots were detected within a MW range of 40-55 kDa ( Fig 4C). Comparing the immunoreactive spots obtained from polyvalent antivenom with all protein spots visualized by silver staining (Fig 4A), 49 spots (numbered 24-25, 31-32, 38-40, 42-45, 48-51, 54-73 and 76-89) were not immunologically recognized. These non-reactive spots were grouped according to their MW and pI values into three clusters. Cluster 1 appeared in the MW range from 26

Identification of immunoreactive and non-reactive peptides in O. monticola venom by LC-MS/MS
LC-MS/MS analysis revealed a total of 202 distinct sequences in O. monticola venom identified within 101 protein types that were immunologically reactive with a polyvenom. All

PLOS ONE
Ovophis monticola venomic profiles and immunoreaction with antivenom immunoreactive proteins recognized by monovalent and polyvalent antivenom are listed in Table 3. However, from all 173 protein types appearing by silver staining, 72 proteins were left unrecognized by both antivenoms, and are shown in Table 4. Interestingly, the majority of these neglected peptides are well-known members of CTLs (e.g., C-type lectin, snaclec and galactose-binding lectin) and different CRISPs including okinavin, catrin and piscivorin.

Discussion
The proportion seemed to be different. Among these three Ovophis spp., the most abundant proteins were of SVSP (35-53%), followed by PLA 2 (19-25%) and LAAO (5-17%). SVMP (11-19%) was detected at a lower percentage than in our venom [10]. Such venom variation in venom composition could would not only be attributed by speciation but also other factors including prey diversity reflecting different ecological habitats [11], snake sex, [17] and age [18,19]. In addition, the variation in quantity of identified toxin types within snake venom might be resulted from the different quantitative approaches and accompanying calculation methods as well as proteomic database availability [20]. The 2DE indicated that O. monticola venom contained more acidic than basic protein spots. This finding was confirmed by our list of all identified peptides obtained from LC-MS/MS showing that the majority possessed pI values lower than 7. Our spot pattern also confirmed previous 2DE analysis of Trimeresurus sumatranus (another Viperid) venom, in which more proteins were identified in the acidic range than in Elapid venom [21]. The overall acidic properties of 4 main protein groups of vipers greatly contribute to the hemorrhagic and coagulopathic effect on victims [22,23].
Our results revealed SVMPs as representing more than one-third of the entire O. monticola venom. The greatest amount found in O. monticola was sp|O93517.1|VM3S4_GLOBR Zinc metalloproteinase/disintegrin or disintegrin-like salmosin-4, first identified within Korean Agkistrodon halys brevicaudus snake venom [24]. SVMPs potentially inhibit platelet aggregation and integrin-dependent cell adhesion via interrupting glycoprotein IIb-IIIa/fibrinogen interaction and fibrinogenolysis [25,26]. Additionally, SVMPs interact with the various types of cellular matrix and exerts the most haemorrhagic effect on hosts [27]. SVSPs were found to be second-most abundant in O. monticola venom. The majority are thrombin-like enzymes including sp|Q9PRP4.1|VSPF_LACMR thrombin-like enzyme LMR-47 and sp|Q9PRP4.1| VSPFLACMR thrombin-like enzyme calobin-2. Known as fibrinogen-clotting enzymes, they are common, and found in large amounts in the venoms of the genera Agkistrodon, Bothrops, Lachesis and Trimeresurus [28]. Thrombin-like enzymes demonstrate strong hydrolytic activity, primarily against triad residues of His57, Asp102 and Ser195 of fibrinogen [29]. Resembling thrombin, they act on blood plasma by forming friable and translucent clots which later     are degraded due to non-functional cross-linked structures, leading to coagulopathy and hypofibrinogenemia [30,31]. However, in terms of enzyme proportion, our findings contrast with previous reports on the transcriptomic analysis of O. okinavensis venom glands, which mainly contained SVSP (93.1%) and relatively little SVMP (4.2%) [32]. It remains unclear whether the microenvironment within the venom gland might preferentially activate or interfere with the functioning of newly synthesized enzymes. Compared with SVMP and SVSP, we detected smaller amounts of PLA 2 and LAAO in O. monticola venom. PLA 2 hydrolyzes phospholipids at the sn-2 position, generating fatty acids and lysophospholipids [33]. Group II PLA 2 is expressed exclusively in the venoms of the Viperidae [34]. We found both acidic and basic PLA 2 subtypes in O. monticola venom. It is noteworthy that the acidic PLA 2 (sp|P81478.1|PA2A2TRIGA and sp|P82896.1|PA2A5TRIST were present in high quantities partially contributing to more acidic properties of the venom. These were previously found to trigger oedema [35]. PLA 2 elicits inflammatory responses through the overproduction of pro-inflammatory cytokines (such as TNFα, IL-1β and IL-6) largely by immunocompetent cells (monocytes, neutrophils and mast cells) [36]. LAAOs can act in concert with PLA 2 in local inflammatory reactions. LAAOs from Calloselasma rhodosthoma venom were shown to induce superoxide anion and hydrogen peroxide production by human neutrophils [37]. The most abundant LAAO in O. monticola venom is sp|P0C2D5.2|OXLA_PROFL, also known as Okinawa habu apoxin protein-1. This protein was first characterized from the venom of Protobothrops flavoridis to induce apoptosis in glioma cells [38]. The roles of both PLA 2 and LAAOs in snake venoms are multi-faceted. Their catalytic as well as cytotoxic properties have been extensively investigated for pharmaceutical potential against cancers and other diseases [39].
Due to the unavailability of homospecific antivenom to Ovophis spp. venoms, all pit viper envenoming victims usually receive either monovalent antivenom (raised against T. albolabris venom) or hematotoxic polyvalent antivenom (produced against venoms of C. rhodostoma, D. siamensis and T. albolabris) [16]. The latter gave considerably higher immunoreactive levels (30-50%) to O. monticola venom proteins than the former. Relatively greater levels of reactivity of the polyvalent antivenom was previously reported with the venoms of C. rhodostoma, Hypnale hypnale and Trimeresurus hageni, and even Trimeresurus albolabris when compared with those of monovalent antivenom [40]. With the combination of 2DE immunoblotting and LC-MS/MS analyses, we found that hematotoxic polyvalent antivenom reacted with a wider range of proteins and peptides accounting for 58% of the entire range of proteins and covering all major enzymatic groups. Nonetheless, we were able to observe that an array of LAAOs and SVSPs did not react with a monovalent antivenom specific only to T. albolabris venom. This finding suggests the shared antigenic epitopes particularly from Ovophis, Calloselasma and Trimeresurus venoms used to generate antivenom. In this context, proteomic analysis of Malayan pit viper C. rhodostoma venom revealed a similar SVMP dominance (41.17%), with other major constituents of snaclec (26.3%) and SVSP (14.9%) [41]. A study of phylogenetic relationships based on geographic distribution and mitochondrial and nuclear gene sequences also demonstrated that O. monticola is less distantly separated from C. rhodostoma than from T. albolabris [42]. Thus, the antivenom against immunogenic epitopes from C. rhodostoma venom should be further investigated for the possible adjunctive treatment of O. monticola bite victims.
Our current study revealed that 72 proteins (42% of venom proteins) were left unrecognized by both antivenoms. The majority of immunologically non-reactive proteins have low molecular mass. They include a number of SVSPs, PLA 2 and certain SVMP. The poor immunogenicity of these low molecular venomic proteins has been obviously reported, although some possess high toxicity [43,44]. In addition, an array of non-enzymatic CTLs such as galactose-binding lectins, snaclec, alboaggregin A, and CRISPs such as triflin, okinavin, catrin, and piscivotin were found unrecognizable by antivenom. This reflects the difference in antigenic abundance between O. monticola venom and those venoms employed to generate horse immunoglobulins, as the proportions of CTLs and CRISPs in our O. monticola venom were only 1.6% and 1.2%, respectively. Nonetheless, their biological impacts on host cells and tissues should not be neglected. In the context of CTLs, alboaggregin A was shown to bind strongly with platelet glycoproteins IB and VI, and hence, activated platelet aggregation [45]. In addition, evidence of enhanced platelet activation and thrombotic microangiopathy-like symptoms has been documented with other related snaclecs [46]. In terms of CRISPs, the unique pdb| 1WVR|A chain A triflin, as well as sp|P0DL18.1|CRVP_OVOOK okinavin from the related hime habu O. okinavensis were not reactive with the antivenoms. They have been previously described to have a calcium channel-impairing effect, leading to aberrations in muscle contraction [47,48]. Considering the pathophysiological effects, our findings address the suite of protein targets which could be additional antigens for future antivenomic design. Furthermore, in order to alleviate the symptoms of mountain pit viper envenomation, these would facilitate the development of specific drug schemes allowing patients to recover more quickly. The proteomic profile of O. monticola venom not only provides insight into the venomic phenotypes reflecting the evolutionary path among Viperid snakes, but also accelerates the discovery of novel candidates for medical and pharmaceutical use.
Supporting information S1 Raw image. (PDF) S1 Table. List of all proteins found in Ovophis monticola venom. (XLSX) S2 Table. List of proteins in 89 spots in Ovophis monticola venom. Crude venom was subjected to 2DE gel. Proteins were separated in the first dimension in the pH range 3-10. (DOCX)