Proteomic Investigations of Two Pakistani Naja Snake Venoms Species Unravel the Venom Complexity, Posttranslational Modifications, and Presence of Extracellular Vesicles

Latest advancement of omics technologies allows in-depth characterization of venom compositions. In the present work we present a proteomic study of two snake venoms of the genus Naja i.e., Naja naja (black cobra) and Naja oxiana (brown cobra) of Pakistani origin. The present study has shown that these snake venoms consist of a highly diversified proteome. Furthermore, the data also revealed variation among closely related species. High throughput mass spectrometric analysis of the venom proteome allowed to identify for the N. naja venom 34 protein families and for the N. oxiana 24 protein families. The comparative evaluation of the two venoms showed that N. naja consists of a more complex venom proteome than N. oxiana venom. Analysis also showed N-terminal acetylation (N-ace) of a few proteins in both venoms. To the best of our knowledge, this is the first study revealing this posttranslational modification in snake venom. N-ace can shed light on the mechanism of regulation of venom proteins inside the venom gland. Furthermore, our data showed the presence of other body proteins, e.g., ankyrin repeats, leucine repeats, zinc finger, cobra serum albumin, transferrin, insulin, deoxyribonuclease-2-alpha, and other regulatory proteins in these venoms. Interestingly, our data identified Ras-GTpase type of proteins, which indicate the presence of extracellular vesicles in the venom. The data can support the production of distinct and specific anti-venoms and also allow a better understanding of the envenomation and mechanism of distribution of toxins. Data are available via ProteomeXchange with identifier PXD018726.


Introduction
Pakistan has a particular geographical location and hosts an array of habitats such a, mountains, glaciers, coastal areas, swamps, plane areas, fresh water, and sandy areas [1]. The country is located between two zoogeographical regions (Palearctic and Oriental) and hosts a diverse venomous fauna. Nine habitat zones are recognized according to the distribution of snakes in Pakistan [2]. Seventy-two snake species are known to Pakistan, among which 14 marine and 12 terrestrial are venomous [1]. According to ITIS (Integrated Taxonomic Information System) database there are 29 snake species belonging to the genus Naja [3]. Among these two are found in Pakistan, i.e., Naja naja and Naja oxiana [2]. Both of these snakes are non-spitting cobras [4]. These snakes are shy of humans. However, upon assessing threat they lift the anterior part of their body, display a hood, and if provoked, hiss loudly and sway their hood to frighten their adversary. These snakes attack very furiously, chewing the bitten part. They usually feed on rodents, birds, frogs, lizards, and snakes. They are found in rocky, stony foothills, forests and around the villages [2]. N. naja (black cobra) is known to have variable color and pattern. However, in Pakistan juveniles and young adults tend to be grey with hood marks, but the adult specimens are usually uniformly black. In addition, the throat pattern is obscured in adult snakes, due to pigmentation [4,5]. N. naja is distributed in North West Pakistan, south and desert areas, except most of Baluchistan. N. oxiana (brown cobra) occurs sympatrically in the Northern half of Pakistan with N. naja. Adult N. oxiana is normally brown in color [5,6]. These snakes and their geographical distribution are shown in Figure 1. Only a few reliable data exist reporting the frequency of morbidity because of snakebites in developing countries. However, it is predicted that snakebite is responsible for a substantial amount of morbidity and mortality in remote areas [7]. The hidden toll of suffering continues to affect the families of the deceased, and patients who survived with crippling deformity [8]. World Health Organization (WHO), included snakebite in its list of "Neglected Tropical disease" in 2007 [9]. Recently, WHO also venom by gel filtration chromatography followed by gel electrophoresis, worked best in their hands. The same group reported the proteomic study of South Indian N. naja venom, recently [47].
In this work, they separated the crude venom components by 1D gel electrophoresis. The mass spectrometric analysis of the tryptic peptide was performed on QTOF. The results of this study derive a comparison of common and unique toxins in N. naja venom obtained from all the three different Indian regions. Our results revealed remarkable differences in the relative abundance of the venom components, as compared to the previous studies. In addition, our investigations unveiled new venom components, not reported before in these venoms. The variation in the results could be different geographical of the snakes from which we collected the venom samples. Further, our workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. Bold text in the first column indicates protein families exclusively identified in N. naja venom. Blue coloured text indicates protein family identified only in N. oxiana. Check mark (  ) in the second column, indicates that this work is the first report of the identification of the corresponding protein families in N.naja venom. The dash sign indicate that the protein family was not identified in the venom. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.

CNF
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. Bold text in the first column indicates protein families exclusively identified in N. naja venom. Blue coloured text indicates protein family identified only in N. oxiana. Check mark (  ) in the second column, indicates that this work is the first report of the identification of the corresponding protein families in N.naja venom. The dash sign indicate that the protein family was not identified in the venom. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.

Results
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1.  workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.

GP
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. workflow did not involve any pre fractionation of the venom. Pre fractionation by gel or liquid chromatography might lead to the loss of some low abundant venom compo used a modern version of the orbitrap mass spectrometer in this work which is equipment.
To the best of our knowledge, this is the first report on the proteomic study of Naj The abbreviations used for proteins and peptides are given in Table 1. ) in the second column, indicates that this work is the first report of the identification of the corresponding protein families in N.naja venom. The dash sign indicate that the protein family was not identified in the venom.

Results
The venom proteome of N. naja (NN) and N. oxiana (NO) snakes was investigated by mass spectrometric analysis, using a shotgun proteomic approach. We were able to provide an extensive Toxins 2020, 12, 669 5 of 41 overview of various protein families present in both venoms, based on data base searches and BLAST analysis of the de novo sequenced tandem mass spectra. A total of 735 peptides from NN and 254 peptides from NO were sequenced (Supplementary Table S1 and S2). Subsequently 365 proteins in NN venom (Table 2) and 140 proteins were identified in NO venom ( Table 3). The sequences of the protein fragments are listed in Supplementary Tables S1 and S2. The results obtained allowed us to cluster the venom protein content into 34 protein families for N. naja and in 24 protein families for the N. oxiana venom. Figure 2A illustrates the preparation for MS acquisition and Figure 2B represents the strategy applied for data base searches. In the present work, we performed data base search against Serpents, King cobra utilizing Uniprot data base. The venom of Ophiophagus hannah has been well studied and genomic and proteomic data are available in the database [28,48,49]. A recent study showed similarity between the genome of Indian cobra and King cobra [50]. This group analyzed 139 N. naja venom gland toxin genes to identify orthologs in the King cobra. It was determined that 96 genes matched while 43 did not. It was suggested that, although some genes are likely to be unique to Indian cobra, the majority were not annotated in King cobra genome. The possible reason could be its highly fragment assembly. Based on this similarity, we searched our data against King cobra database also. Further, in the data base complete proteome of only King cobra is available. The details of our search against Serpent database are presented in Supplementary              The bold text indicates the proteins identified to have N-terminal acetylation.
Toxins 2020, 12, 669     The bold text indicates the proteins identified to have N-terminal acetylation.

of 41
A comparative summary of the protein families of the two venoms is presented in Table 1. Figure 3, presents a comparison of the relative abundance of different venom protein families as pie charts. From the pie charts, it can be observed that there are significant differences in the proteome of two snake venoms. In the venom of N. naja, three-finger toxins (3FTx) are more abundant, while in N. oxiana venom, both 3FTXs and snake venom metalloproteinase (SVMPs) are almost equally abundant. In NO, snake venom serine proteases (SVSPs) and phospholipase A 2 (PLA 2 s) are much more abundant than in NN. There are other subtle variations in the relative abundance of protein families between the two venoms. For example, Cysteine-rich Secretory Protein (CRISP) family is much more abundant in NN as compared to NO. Further, NN venom contains 11 protein families, which could not be found in NO venom, listed in Table 1 and highlighted in red color. Whereas NO venom contains serpins, which are absent in NN venom. Figure 3 shows that NN venom is much more versatile and contains a number of different proteins (Table 1). Data base searches revealed that our data provide a deeper insight of the NN and NO venom proteomes. There are several protein families, which have not been reported earlier in NN venom, including western and eastern Indian N. naja. In Table 1, the protein families discovered and reported for the first time in terms of our investigations are shown with check mark ( workflow did not involve any pre fractionation of the venom. Pre fractionation by gel electrophoresis or liquid chromatography might lead to the loss of some low abundant venom components. Also, we used a modern version of the orbitrap mass spectrometer in this work which is very sensitive equipment. To the best of our knowledge, this is the first report on the proteomic study of Naja oxiana venom. The abbreviations used for proteins and peptides are given in Table 1. Bold text in the first column indicates protein families exclusively identified in N. naja venom. Blue coloured text indicates protein family identified only in N. oxiana. Check mark (  ) in the second column, indicates that this work is the first report of the identification of the corresponding protein families in N.naja venom. The dash sign indicate that the protein family was not identified in the venom.

Results
). Interestingly previous studies reported PLA 2 as the second most abundant protein family found in N. naja venom, and that SVMP was present in relatively low abundance [8,[42][43][44][45][46]. In contrast, our data showed SVMP as the second most abundant protein family in N. naja. The venom proteome of N. oxiana displays that, both 3FTXs and SVMP are equally abundant like that of king cobra (Ophiophagus hannah) [51], as illustrated in Figure 3A. In the present work, three types of posttranslational modification were also observed, i.e., N-terminus pyro-glutamate, methionine oxidation and N-terminal acetylation (N-ace). Pyro-glutamate posttranslational modifications of the venom proteins has been described before and are known to confer stability to the proteins and peptides [52][53][54][55]. However, modification of methionine residues and pyro-glutamate cannot be excluded during sample preparation. Therefore, keeping in view this possibility we have not discussed the observed methionine and pyro-glutamate modifications. The current study is the first description of N-terminal acetylation of venom proteins. In N. naja venom we were able to identify three peptide fragments (Muscarinic toxin-like protein 3, Acid phospholipase A 2 and weak neurotoxin 7) containing N-ace modification. Whereas in N. oxiana one peptide (Muscarinic toxin-like protein 3) was identified with N-ace. These sequences have been highlighted with green colour in Supplementary Tables S1 and S2. In the present work, we have identified a number of proteins like cobra serum albumin, leucine repeat, zinc finger containing protein, venom lectin protein, Ras-like protein. The presence of Ras-like protein demonstrates the presence of extracellular vesicles in the venom of Naja naja. The comparison of our proteomic data with that of N. naja snake both from western and eastern India, reveals that such proteins were not identified in Indian N. naja, Further in Pakistani Naja naja snake we not could identify cholinesterase, butyrylcholinesterase, hyalurinidase and snaclec proteins which were previously reported in Indian N. naja venom [45,46].
Below we briefly describe and discuss the different venom protein families identified.

Three-Finger Toxins
The detailed proteomic investigations of the, NN and NO snake venoms identified two main types of three-finger toxins, i.e., neurotoxins and cytotoxins. The venom of NN consists of an overall higher abundance and a greater diversity of 3FTXs, as compared to NO (Tables 1 and 2, Figure 3). Neurotoxins are predominant in both venoms, as compared to cytotoxins, Figure 4. Among the neurotoxins, long, muscarinic, weak, 3FTxs precursor and aminergic toxin families are present in both venoms. In case of NO, a rather low amount of long neurotoxin is present, represented by one neurotoxin, cobratoxin. Whereas, long neurotoxins constitute a major proportion of neurotoxins found in the NN venom. Figure 4 indicates that in NO venom, muscarinic toxins are present in relatively higher amounts as compared to NN venom. It is interesting to note that NN venom contains an aminergic neurotoxin with homology to Dendroaspis angusticeps venom toxin AdTx1. This toxin is known to impair G-protein-coupled receptors [56,57]. Previous studies have shown that 3FTXs make up approximately 56-84% of venom proteome in various species of Naja [58]. However, our results of Pakistani Naja venom samples show a much lower percentage of 3FTXS as compared to other investigations, which is 21% in NN and 16% in NO of the total venom proteins. In contrast to Pakistani N. naja venom proteome, the eastern Indian N. naja venom comprises of 61% 3FTXs and western Indian N. naja contains 68% 3FTXs. Interestingly eastern Indian N. naja lacks short neurotoxins, which are present in both western Indian and Pakistani N. naja snake venom [45,46]. Investigating 3FTXs are not only of interest to characterize their toxicity, but also of great significance for structural studies, as well as for biotechnological, biomedical and evolutionary studies [59][60][61][62][63]. Already, 3FTXs have proven to be an efficient tool to analyze various receptor types, and to study diseases like Parkinson's disease, myasthenia gravis and cancer [64][65][66][67][68][69][70][71]. The aminergic toxins from mamba venom served as good candidates for protein resurrection methodology [72].

Phospholipase A 2
Both Naja snake venom contain PLA 2 . The percentage abundance of PLA 2 enzymes (12.6%) is higher in NO as compared to NN. PLA 2 s make up 6% of the venom of NN (Table 1). A recent study reported the comparative enzymatic activity of PLA 2 s in ten different Naja species, with highest activity in N. siamensis and lowest in N. nivea [73]. The venom proteome study of Indian N. naja venom carried out by A. K. Mukherjee research group reported that Indian N. naja contains 20-27% PLA 2 s [45,46]. This is much higher than the amount of PLA2s present in Pakistani N. naja. A proteomic study of N. kaouthia venom reported PLA 2 s as one of the most abundant venom proteins [74]. While another study on the venom proteome of N. annulifera did not detect PLA 2 s [75]. In the venom of N. philippinensis PLA 2 s made up 22.88% of venom proteome [76]. Another study showed distinct distribution of PLA2s in Afro-Asian cobra venom. Asian spitting cobras showed highest PLA 2 activity. Asian non-spitting and African spitting cobras showed moderate activity and low activity was shown by African non-spiting cobras [77]. Table 3 shows that both venom comprise of acidic and basic PLA 2 s. However, acidic PLA 2 s are more abundant in the two venoms. Two fragments of phospholipases from NO bear homology to neutral PLA 2 s paradoxin-like beta chain from Oxyuranus microlepidotus. This protein was found to be one of the most potent presynaptic neurotoxins [78]. Eleven peptide fragments bearing homology to acidic phospholipase in the venom of Naja sputatrix were identified (Table 3). In the Naja naja peptide fragments having homology to acidic PLA 2 s from the venom of other Naja species were determined ( Table 2). Six peptide fragments showed homology to acidic PLA 2 from the venom of Pseudonaja textilis. A previous study reported this molecule to have moderate enzymatic activity and procoagulant property and was found to be non-lethal [79]. In the NN venom two peptide fragments matching Basic phospholipase A2 from Bungarus candidus venom and one matching with basic PLA 2 with sea krait was identified. While in NO venom only one peptide fragment having homology to a basic PLA 2 from Bungarus candidus was found. The activity and specificity of basic phospholipases from Agkistrodon h. blomhiffii and Pakistani N. naja was studied on intact human erythrocytes. Although belonging to two different snake families, similar response was reported for these molecules, from both venoms. Basic PLA 2 induced the hydrolysis of membrane phospholipids and total cell hemolysis [80]. Despite the fact that acidic PLA 2 s are found abundantly in the snake venom, their role is poorly understood [81]. In spite of having high catalytic activity as compared to basic PLA 2 s, they do not induce toxicity [82]. Studies have suggested acidic PLA 2 s to participate in prey digestion [83]. Other studies have suggested that acidic PLA 2 s work synergistically with other venom toxins, as PLA 2 s, metalloprotease and cytotoxins [84][85][86].
PLA 2 is ubiquitously found in nature [87,88]. In mammals, they are known to play important and vital role in many life processes [89][90][91]. On the other hand, snake venom PLA 2 s are toxic and interfere with a number of physiological processes, upon envenomation [87]. Phospholipase A 2 are also known to be responsible for the hepatic injury, inflammation and anticoagulation in a victim [26].

Snake Venom Metalloproteinase
The present study shows that N. naja and N. oxiana snake venom contain significant amounts of metalloproteinases, which are the second most abundant protein family. Proteomics study of other Naja species shows the presence of SVMPs in varying amounts ranging from as low as 0.9% to 16% [74,[92][93][94][95][96][97][98][99][100]. Previous proteomic studies reported a lower abundance of SVMP in Pakistani N. naja venom [42][43][44]. Three SVMPs bearing relatively higher homology with snake venom metalloproteinase from N. atra were determined in each of the two venoms. Twenty fragments of SVMPs were detected in N. naja venom, which are homologous to K-like SVMPs from N. atra. 13 Peptide fragments were found to match with SVMPs from N. kaouthia (Table 2). The data shows that in case of N. oxiana venom higher number of peptide fragments match with SVMPs from N. atra venom ( Table 3). The eastern Indian N. naja contains only 6% SVMPs in contrast to Pakistani N. naja, which contains 10% of SVMPS. It is interesting to note that western Indian N. naja contains 16% SVMPs as determined by A. K. Mukherjee and his research group [45,46].
SVMPs are found in all advanced snakes and make up the major component of the venom of Crotalid and Viperid snakes [101][102][103][104]. SVMPs are structurally versatile and act on different hemostatic targets of prey upon envenomation [105]. These toxins provoke many systemic changes, such as hemorrhage, acute renal failure, coagulopathy, and/or platelet aggregation inhibition [106]. The SVMPs identified in terms of our investigations, in both of the venoms, belong to subfamily P-III. The P-III SVMPs possess gelatinolytic and hemorrhagic activities [105]. A previous study reported the hemorrhagic response of Pakistani N. naja venom in chicken egg [107]. The determination of a higher amount of SVMPs in both NO and NN venom indicates that there is potential for hemorrhage as a response of NO and NN snakebite envenomation.

L-Amino Acid Oxidase
Snake venom L-Amino acid oxidase (LAAOs) belong to the Flavin monoamine oxidase family and are dependent on FAD group for their activity. These proteins are present in both venoms studied and constitute approximately 4-5% of the venom proteome (Table 1). Peptide fragments bearing sequence similarity to LAAO from different snake venoms were detected and summarized (Tables 2  and 3). In contrast to our results, studies of western Indian N. naja venom report only 0.31% LAAO. However, Indian N. naja venom contains 3% LAAO, which is similar to that of Pakistani N. naja [45,46]. In terms of our investigations we identified that LAAO from both, NN and NO venom, have homology with LAAO from N. atra venom. LAAO is known to be prevalent in many snake venoms [108] but its role in snake venom envenomation pathology is not clear. A recent study reported that LAAO might contribute to severe tissue disruption. This study suggested that LAAO might elicit its toxicity by catalyzing the intracellular substrates [108]. Different biological activities of the isolated LAAO have been reported like, edema, cytotoxic, antibacterial, antiparasitic, and/or platelet aggregation effects [109,110]. Also some investigations reported antitumor effects of LAAO [111]. LAAO is a glycoprotein and glycosylation is also considered to play a significant role in the toxicity of LAAO, and cause cell death by interacting with the cell surface [112,113].

Cobra Venom Factor
Cobra venom factor (CVF) belong to the venom complement C 3 homologue family. CVF constitutes approximately 9% of the total proteins identified in both venoms. The identified CVF peptides bear homology mainly to the CVFs from N. kaouthia. Fragments matching to CVF alpha chain and gamma chain were also analyzed. Peptide fragments showing sequence similarity to CVF proteins from other Elapidae and Colubridae have also been identified (Tables 2 and 3). Proteomic study of Indian N. naja venom showed that it contains only 0.03-1.7% CVF [45,46], which is significantly less compared to our results obtained for the Pakistani N. naja. A venom proteome study of Naja philippinensis showed that it contains less than 4% [76]. The venom of N. ashei contains only 0.12% CVF [99]. Cobra venom factor is a complement activating protein and is functionally and structurally similar to complement component C3b. It is a glycoprotein and herein glycosylation contributes in the immunogenicity of CVF [114,115]. In vivo studies have shown that CVF causes an acute inflammatory injury in the lungs [116]. CVF serves as a gold standard molecule for the evaluation of drugs for trials, to control diseases involving the complement system [117]. A recent study reported CVF as a promising candidate for the treatment of IRI-induced hepatic injury [118]. Our data reveals that CVF is one of the abundant proteins in the venom of Pakistani Naja naja and Naja oxiana (Figure 3). Therefore, these venom can be a good source of isolating CVF for use in biomedical research.

Cysteine-Rich Secretory Protein
Cysteine-rich secretory proteins (CRISPs) have been identified in many animal venoms. These proteins have two domains, a pathogenesis related domain at the N-terminal region and a cysteine rich domain at the C-terminus. Based on sequence homology the CRISP family is classified into four classes, and snake venom CRISPs belong to the group III [119]. CRISPs were found in much higher abundance in N. naja (7%) as compared to N. oxiana (2.8%) and peptide fragments showing similarities to CRISPs from different snake venoms were found in both venoms. However, highest similarity was found with the cysteine-rich venom protein natrin-1(NA-CRVP1) from N. atra. Investigations indicated that NA-CRVP1 could act as inflammatory modulator that could perturb the wound-healing process of a bitten victim by regulating the expression of adhesion molecules in endothelial cells. This study also showed that natrin contains a zinc-binding domain at the N-terminus and elicits its proinflammatory activity in a zinc and heparan-sulfate dependent manner [119]. Natrin has also been reported as a potassium channel blocker and in this context can weakly block muscle contraction [120][121][122][123]. In our study six peptide fragments matching CRISP from N. haje. A study reported this CRISP was found to be non-toxic when administered to crickets [124]. The venom proteome of N. haje contain 10% CRISP [92]. Different species of Naja contain varying amounts of CRISP, from as low as 0.2% to 10% of the total venom proteome. The Indian N. naja contains 1.14-3% CRISPs [45,46].

Snake Venom Serine Proteinase
Snake Venom Serine Proteinase (SVSPs) belong to the peptidase S1 family. N. oxiana venom proteome shows relatively higher abundance of serine proteinases (4%) as compared to N. naja venom, which contains only 2% (Table 1; Figure 3). Both of the venoms contain peptide fragments, which bear homology to tissue-type plasminogen activators from Ophiophagus hannah and the thrombin-like enzyme TLP from Indian N. naja. In addition to this, peptide fragments having sequence similarity to SVSP have also been identified (Tables 2 and 3). SVSPs have been identified in only few Naja species venom. In western Indian N. naja the SVSPs contributed only 0.69% to venom proteome [46]. N. philippinensis venom proteome consists of 0.35% SVSPs [76]. Previous studies showed that SVSPs are absent in Eastern Indian N.naja venom, while a small percentage (0.03%) was reported for the western Indian snake [45,46].
SVSPs have high specificity towards their substrates. Based on their biological roles, they have also been classified as activators of the fibrinolytic system, procoagulant, anticoagulant and platelet-aggregating enzymes [125]. A few SVSPs, like ancrod and batroxobin have already applications in the treatment of cardiovascular problems, while reptilase serves today as a diagnostic probe for dysfibrinogenemia [126].

Snake Venom Nerve Growth Factor
Snake venom Nerve Growth Factor (NGF) were identified in both venoms but were relatively more abundant in the venom of N. oxiana (4%) as compared to N. naja (2%), Table 3. In both venoms, peptides sharing homology with Ovophis okinavensis, N. sputatrix, and Bitis gabonica NGF were identified (Tables 1  and 2). In N. naja seven peptides bearing homology with Pseudechis australis were also identified. Further, additional peptide fragments of NGF were also identified in terms of our investigations (Tables 2 and 3). A previous proteomic study also showed Pakistani N. naja venom to contain 2% NGF [42]. N. naja snake venom from east India contained 3.1% and 1% in western India sample. In the same study N. kaouthia from eastern India was shown to contain 1% NGF [45,46]. N. philippinensis contain only 0.06% NGF [76]. Proteomic analysis of other Naja species venom have also shown them to contain NGF but their relative abundance was not calculated [93,100]. Moroccan cobra venom contains 5% NGF of total venom proteome [92].
Till now not much is known about the contribution and function of NGFs in envenomation. Various bioassays have shown that NGFs have neurotropic activity. Snake venom NGFs have been suggested as a pharmacological tool to study the structure function relationship of human trkA receptor [127]. Studies show that NGFs assert venom toxicity indirectly, either by acting as a carrier of other neurotoxins, which do not have specific recognition sites, like phospholipase or by inducing plasma extravasation at the site of snakebite. NGF is known to coexist with CVF in cobra snake venom, and might be responsible for enhancing the toxic mechanism of CVF in an indirect manner [128]. In 1986, two scientists were awarded a Nobel Prize for their pioneering work, which allowed to explain cell growth regulation. And in context of this investigations Cohen serendipitously discovered NGF from snake venom of Agkistrodon piscivorus [129].

Snake Venom Phosphodiesterase
A lower abundance of snake venom phosphodieterases (PDEs) was determined in both venoms, although relatively higher in N. oxiana, i.e., 3.1%. N. naja venom contains only 1.1% of PDEs. Peptides fragments matching with PDEs from the venom of N. atra, Ovophis okinavensis, and Borikenophis portoricensis were identified in both venom. A recent study determined PDE activity in the venom of ten different species of Naja. All the species showed PDE activity with minor variation [73]. The Indian N. naja venom was reported to contain less than 1 % PDEs, which is similar to Pakistani N. naja.
PDEs are ubiquitously present in snake venom but their activity is higher in Viperidae venom as compared to Elapidae family [130]. In recent years, there has been considerable interest in snake venom PDEs due to their potential applications as pharmacological tool and drug lead. The endonuclease activity of PDEs rendered their use in sequencing of polynucleotides and oligonucleotides [130]. Phosphoribosylation, a protein modification, can also be processed by PDEs [131]. Recent innovative approaches, have utilized snake venom PDEs to digest genomic DNA into single nucleosides to study modifications of DNA [132][133][134].

Minor Protein Components (Relative Abundance ≤2%)
A large number of low abundant proteins were found in both venoms, particularly in N. naja (Tables 2 and 3). Ras-like proteins, identified in the venom of N. naja were of particular interest, as they indicate the presence of extra0cellular vesicles in the venom. Snake venom extracellular vesicles (SVEVs) have previously been isolated from the venom of Agkistrodon contortrix contortrix, Crotalus atrox, Crotalus viridis, and Crotalus cerberus oreganus. The size distribution of SVEVs was found to be between approximately 50-500 nm. Proteomic investigations revealed that SVEVs could be assigned to eight different protein classes, such as SVMP, SVSP, and disintegrins [135].
Exosome-like vesicles have also been reported in the venom of Gloydius blomhoffii blomhoffii [136]. In this context extracellular vesicles (EV) are known to carry a diverse cargo of molecules as proteins, DNA, RNA, and/or lipids [137]. Further, EVs are involved in cell-to-cell communication, immune response and apoptotic rescue [138,139] and participate in the maintenance of normal as well as pathophysiological conditions, like cancer [140][141][142]. The proteomic study of extracellular vesicles released from cancer cells have shown the presence of Ras proteins functioning as biomarkers for extracellular vesicles [137,[143][144][145]. Studies have shown that Ras proteins are involved in the regulation and assembly of extracellular vesicles cargo [145][146][147][148]. Therefore, the identification of Ras-like proteins indicates the presence of extracellular vesicles in the venom of N. naja. However direct experimental work needs to be done to confirm the presence of such vesicles in the venom. SVEVs in the venom may be involved in another mechanism to secrete membrane proteins like aminopeptidase A and G coupled receptors. SVEVs may also offer an additional route for the envenomation process, thereby facilitating toxins to translocate inside the prey cells.
In the present work, a number of proteins have been identified for the first time in the proteome of these venoms, like G-protein coupled receptors, zinc finger proteins, ankyrin repeat, leucine repeat, Ubiquitin carboxyl-terminal hydrolase and a number of other protein. It can be assumed that these proteins have also a function in the venom. Ankyrin repeats and zinc finger proteins were also identified recently, in the venom of King cobra, Naja annulifera and Micrurus pyrrhocryptus [51,100,149]. A rather old publication reported cobra serum albumin in the venom of cobra snakes [150]. Our data also revealed the presence of cobra serum albumin in the venom of N. naja. It is possible that upon envenomation cobra serum albumin is responsible or supporting the transportation of other venom proteins in the prey serum. Previous studies have reported Cobra blood serum albumin as an antitoxic protein, having the potential to sequester endogenous toxins [151,152]. Cobra serum albumin was also reported in the venom proteome of N. sumatrana [93]. Further, we identified glutathione peroxidase in both venoms. A recent proteomic study also reported the presence of glutathione peroxidase in the venom of Micrurus pyrrhocryptus and N. annulifera [100,149]. It can be speculated that this protein might be involved in protecting the venom gland from oxidative damage. Phospholipase A 2 inhibitors, bearing similarity to PLA 2 inhibitor isolated from the serum of Elaphe quadrivirgata and Naja kaouthia snakes, were identified in the venom of N. naja. This inhibitor was shown to inhibit the enzymatic activity of all till now known PLA 2 enzymes [153,154]. Phospholipase B was also identified in both venom. Only 0.1% constituted the venom proteome of NN while that of NO contained 1.6% of the total venom proteome. Studies have shown that PLBs make up approximately 0.34% of venom components, and in Viperidae venom it varies between 0.23% to 2.5% [155]. Insulin and Transferrin proteins were also identified in the venom of N. naja. Transferrin is a metal binding proteins. Transferrin was also reported before in the venom of P. australis, utilizing two dimensional gel electrophoresis [156]. Snake venom VEGF bearing similarities to that isolated also in Bitis arietans venom, identified in N. naja venom as well. Studies have shown that different variants of snake venom VEGF induce angiogenesis and vascular permeability through different mechanisms [157,158]. Snake venom VEGF are potential candidates for therapeutic angiogenesis [159]. A low abundance of Kunitz type serine protease inhibitors (KSPI) was identified in the venom of both snakes. Snake venom KSPI have the potential to selective inhibit distinct serine proteases [35]. Some of the snake venom KSPI have evolved as potassium channel blockers [160]. BF9 a snake venom KSPI, which act as potassium channel blockers and retain the serine protease inhibitory activity. This bifunctional KSPI was reported in the venom of Bungarus fasciatus [161]. Interestingly another type of serine protease inhibitor, i.e., serpin, was identified in the venom of N. oxiana. Proteins belonging to Ohanin/Vespryn family were found in both venoms. They are small proteins with an average mass of approx. 12 kDa, and are neurotoxic in nature [162]. Further, we could identify 5 -nucleotidase in both venoms. This family of protein is found in different snake venoms [163]. These enzymes play a major role in the release of adenosine upon envenomation, which facilitates prey immobilization. In vivo studies have shown that 5-nucleotidases act synergistically with other venom components like phospholipases, disintegrins to exert a pronounced anticoagulant effect [164]. Aminopeptidase was identified in both N. naja and N. oxiana venoms. Aminopeptidase A activity has been found in the venoms of snakes belonging to Elapidae and Viperidae families. Till now not much is known about the contribution of this enzymes within envenomation pathology [165]. Cystatin, having similarity to cystatin from the venom of N. kaouthia, was identified for the first time in the venom of N. naja in the present study. Cystatin is a cysteine protease inhibitor [166]. Natriuretic peptides were identified in both N. naja and N. oxiana venom. These peptides are known to induce hypotension upon envenomation [167,168]. Cathelicidin was identified in the venom of N. naja, and previous studies have shown it to be potent antimicrobial peptide [169].

Posttranslational Modifications
In terms of our investigations, we were able to identify N-terminal acetylation (N-ace) for the first time in the snake venom. This posttranslational modification is known to carry and support out different functions in the cell. A most analyzed function of N-ace is the regulation of protein half-life, by labelling proteins for polyubiquitation and thus degradation by the proteasome [170,171]. N-ace modification plays a role in protein folding and protein complex formation [172,173]. Furthermore, studies have shown that N-ace modification mediates the interaction of proteins with membrane and subcellular localization [173]. A probable role of this modification in snake venom proteins could be to stabilize them against proteolytic cleavage, and to assist in distinct protein-protein interactions upon envenomation. In both venoms a peptide fragment bearing homology to muscarinic toxin like-protein 3, from the venom of Naja kaouthia was found to be N-terminal acetylated. Whereas in Naja naja two other peptide fragments were identified to be N-terminal acetylated. One bearing homology to phospholipase A 2 and other to a weak neurotoxin 7 (Supplementary Table S1 and S2).

Conclusions
Using the MS shotgun approach we could provide a holistic view of the venom profile of the two Pakistani cobra snakes N. naja and N. oxiana. Our data shows for the first time the venom proteome of N. oxiana. The comparative evaluation of the venom proteome of the two snakes reveals differences, as well as similarities in their venom composition. Both snake venoms contain different types of three-finger toxins in their venom, although they exit sympatrically. There are a few toxin families, which were only found in the venom of N. naja, like cystatin, VEGF, TGF, BPP, and Cathelicidin. Therefore, we can suggest, that venom samples from both species should be utilized for the production of effective antivenoms. Also, applying state-of-the-art mass spectrometric tools allowed to identify a number of proteins not known before to be in these venoms, like Ras-GTPase, Ankyrin repeats, leucine repeat, G-protein coupled receptor, zinc finger protein, holiday junction protein, and endonuclease. In this context, identification of Ras-like proteins provided a clue about the presence of extracellular vesicles. These vesicles might function as an additional carrier to transport venom components in the prey upon envenomation. Further, our data highlight N-terminal acetylation of venom proteins for the first time and the results delineate the unique complexity of snake venoms, and open routes for further research to understand function of these molecules upon envenomation.

Venom Collection
Venom was milked manually from adult snake species of N. naja (black cobra/Indian cobra/ Spectacled cobra) and N. oxiana (brown cobra/Caspian cobra/Central Asian cobra). For the proteomic studies of each species the venom was collected from three adult healthy snakes and pooled. The sex of the snakes was not determined. N. naja snakes were captured from the rural surroundings of Mianwali district, while N. oxiana snakes were caught from Lahore, Punjab province, Pakistan. The venom was freeze dried and kept at −20 • C till further use.

Sample Preparation for LC-MS/MS
For LC-MS/MS analysis the lyophilized crude venom from N. naja (black cobra) and N. oxiana (brown cobra) was dissolved in 10 mM Triethylamonium bicarbonate (TEAB, Thermo Fisher Scientific), 1% v/w Sodium deoxycholate (SDC, Sigma) buffer. Protein concentration was determined using a bicinchoninic acid protein assay (Pierce™ BCA Protein Assay Kit, Thermo Fisher Scientific) and 20 µg of venom protein was tryptically digested. In brief, cysteine residues were reduced for 30 min, in the presence of 10 mM dithiothreitol (DTT, Sigma) at 60 • C and alkylated for 30 min, with 20 mM iodoacetamide (IAA, Sigma) at 37 • C in the dark. Thereafter, sequencing grade trypsin (Promega) was added in a protease/protein ratio of 1:100 at 37 • C to hydrolyze venom proteins overnight. Enzymatic activity was quenched by addition of 1% v/v formic acid (FA, Fluka) and the SDC was precipitated by centrifugation at 16,000× g for 5 min. The peptide containing supernatant was vacuum dried and reconstituted in 0.1% FA for LC-MS/MS analysis.
MS/MS data acquisition was performed in data dependent mode for the 15 most abundant precursor ions. Precursor ions with charge stages between 2+ and 5+ were selected for fragmentation if they exceeded an intensity threshold of 100,000. For MS/MS spectra acquisition the set AGC-target was 100,000 with a maximal ion injection time of 50 ms. Precursor ions were fragmented at a normalized collision energy (NCE) of 25 and the fragment ions were measured with a resolution of 17,500 at 200 m/z. To avoid redundant precursor sampling a dynamic exclusion was applied for 20 s.

Data Analysis
For protein identification, the generated raw data were processed using the Proteome Discoverer™ Software 2.0.0.802. Database search was performed with the SEQUEST algorithm against an Ophiophagus hannah (txid:8665, King cobra) protein database (UniProt), since it represents the closest sequence database to the analyzed samples. Carbamidomethylation of cysteine was used as fixed modification. Furthermore, oxidation of methionine, conversion of glutamine to pyro-glutamic acid at peptide N-termini, loss of N-terminal methionine and the acetylation of protein N-termini were considered as variable modifications. Precursor and fragment ion tolerance were set at 10 ppm and 0.02 Da, respectively. Peptide-spectra matches with a maximum delta Cn of 0.05 were used by Percolator for FDR estimation.
Unidentified spectra were exported to a new mgf file and de novo sequencing was performed with Novor [174] via DenovoGUI 1.16.2 [175]. Modifications and allowed mass tolerances were identical to the database search approach. Hits with a Novor score above 80 were considered for a protein BLAST approach. Protein BLAST for the sequenced peptides was conducted with the NCBI BLAST p algorithm (2.9.0+) with default settings against non-redundant protein sequences (nr) narrowed down to serpents (taxid: 8570). Alignments were chosen according to the max Score, the query coverage and if the homologous proteins were related to venom activity. With this information, a venom specific peptide database was created to support database searching for further analyses. Similarly, the data search was also performed against Serpents protein data base from UniProt.
The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository [176] with the dataset identifier PXD018726 and 10.6019/PXD018726. Venom components were classified according to protein families and their relative abundances calculated as, reported previously [51]. Briefly, the proteins analyzed were sorted into different groups of protein families. The relative abundance of each family was calculated as percent of total number of venom proteins detected by the mass spectrometer. The mathematical relationship below was used to calculate the relative abundance of each protein group. Pie chart ( Figure 3) and Table 1, presents the percentage relative abundance of proteins.
Number of proteins (protein family) Total venom proteins detected using LC − MS/MS × 100 (1) Supplementary Materials: The following are available online at http://www.mdpi.com/2072-6651/12/11/669/s1 Table S1: De novo peptide sequencing Naja naja venom. Table S2: De novo peptide sequencing Naja oxiana venom. Table S3: Proteomic data of Naja naja venom searched against Serpents Uniprot protein data base. Table S4: Proteomic data of Naja naja venom searched against King cobra Uniprot protein data base. Table S5: Proteomic data of Naja oxiana venom searched against Serpents Uniprot protein data base. Table S6: Proteomic data of Naja oxiana venom searched against Serpents Uniprot protein data base.