Snake venomics of Bothrops punctatus, a semiarboreal pitviper species from Antioquia, Colombia

Introduction

The Chocoan forest lancehead, Bothrops punctatus, known in Colombia as ‘rabo de chucha’, is a large semi-arboreal pitviper, ranging from 1.0 to 1.5 m in length. (Campbell & Lamar, 2004) described its distribution from the Pacific foothills and coastal plain of eastern Panamá through western Colombia to northwestern Ecuador, with an altitudinal range between 135 and 2300 m. In Colombia (Daza, Quintana & Otero, 2005) reported the occurence of B. punctatus in the Cauca and Magdalena river basins of Antioquia to eastern Chocó. Although Bothrops species are clearly predominant in the epidemiology of snakebite accidents occuring in Colombia (Otero, 1994; Paredes, 2012), published reports of proven envenomings caused by B. punctatus appear to be rare. The protein composition of the venom of this species has not been investigated, although at least two reports characterized its toxicological properties, in comparative studies of snake venoms from Colombia (Otero et al., 1992) and Ecuador (Kuch et al., 1996), respectively. The lethal potency of this venom to mice was highest among the different Bothrops venoms analyzed in these two studies, being only second to that of Crotalus durissus terrificus venom (Otero et al., 1992; Kuch et al., 1996). Due to the lack of knowledge on the venom composition of B. punctata, this work aimed at characterizing its proteomic profile using the ‘snake venomics’ analytical strategy (Calvete, Juárez & Sanz , 2007; Calvete , 2011), in combination with the assessment of its enzymatic or toxic activities in vitro.

Methods

Statistical analyses

The significance of differences between means was assessed by ANOVA, followed by Dunnett’s test, when several experimental groups were compared with the control group, or by Student’s t-test, when two groups were compared. Differences were considered significant if  p < 0.05.

Results and Discussion

B. punctatus has been included in the ‘red list’, a report categorizing conservation status, as a threatened species (Carrillo et al., 2005). Very scarce information on its venom is available in the literature. In comparative studies of snake venoms from Colombia (Otero et al., 1992) and Ecuador (Kuch et al., 1996), respectively, this venom was found to induce local effects such as hemorrhage, edema, and myonecrosis, as well as systemic alterations such as defibrination, in similarity to venoms from other Bothrops species. Developments in proteomic techniques have brought new possibilities to examine the detailed toxin composition of snake venoms, increasing knowledge on their evolution, toxicological properties, and correlation with clinical features of envenomings (Calvete, Juárez & Sanz , 2007; Calvete , 2013; Fox & Serrano, 2008; Valente et al., 2009; Ohler et al., 2010; Georgieva et al., 2010). Therefore, the venom of B. punctatus was analyzed for the first time using proteomic tools, to gain a deeper understanding on its protein composition and relationships to toxic and enzymatic actions.

RP-HPLC of the crude venom resulted in the separation of 30 fractions (Fig. 1C), which were further subjected to SDS-PAGE (Fig. 1B), in-gel digestion of the excised bands, and MALDI-TOF-TOF analysis of the resulting peptides. The amino acid sequences obtained allowed the unambiguous assignment of 29 out of the 37 components analyzed, to known protein families of snake venoms (Table 1). Protein family relative abundances were estimated by integration of the chromatographic areas, combined with gel densitometric scanning. Results showed that the predominant proteins in this secretion are metalloproteinases (41.5%; SVMP), followed by C-type lectin/lectin-like proteins (16.7%; CTL), bradykinin-potentiating peptide-like peptides (10.7%; PEP), phospholipases A₂ of both the D49 (8.0%) and K49 (1.3%) subtypes (for a combined 9.3%; PLA₂), serine proteinases (5.4%; SP), disintegrins (38%; DIS), L-amino acid oxidases (3.1%; LAO), vascular endothelial growth factor (1.7%; VEGF), and cysteine-rich secretory proteins (1.2%; CRISP), as summarized in Fig. 2 and Table 1. An estimated 6.6% of the proteins remained unidentified, and owing to the scarcity of the venom, their assignment could not be further pursued.

A recent phylogenetic analysis of the genus Bothrops (sensu lato) by (Fenwick et al., 2009) grouped B. punctatus within the same clade as Bothrops atrox and Bothrops asper. Since the proteomic profile of the venoms of the latter two species has been reported (Núñez et al., 2009; Alape-Girón et al., 2008), a comparison of their venom compositions, together with those of two other pitviper species distributed in Colombia, Bothrops ayerbei (Mora-Obando et al., 2014) and Bothriechis schlegelii (Lomonte et al., 2008), was compiled (Table 2). Venoms from these five species have been analyzed by the same methodological strategy, therefore allowing reliable comparisons. The composition of B. punctatus venom resembles that of the other Bothrops species listed in Table 2 only in terms of their high content of metalloproteinases (41.5–53.7%), but overall, its composition departs from the relative protein abundances observed in any of the other four pitvipers. The high proportion of CTL proteins in B. punctatus is of note, doubling the abundance observed in B. atrox, and close to that of B. ayerbei, while in contrast such proteins are expressed only in trace amounts in B. asper, and have not been detected in B. schlegelii (Table 2). Further, B. punctatus venom presents a modest amount of VEGF (1.7%), which has not been found in any of the venoms listed in Table 2. Similar to the venom of the arboreal snake B. schlegelii, but also with the terrestrial species B. ayerbei, the venom of B. punctatus presents a high content of BPP-like peptides, strikingly differing from B. asper and B. atrox venoms in this regard. The possible trophic relevance of these vasoactive peptides among viperids remains elusive, and no clear correlations with prey types or habitats have been disclosed thus far. BPPs are oligopeptides of 5–14 amino acid residues, rich in proline residues and often presenting a pyroglutamate residue, which display bradykinin-potentiating activity. Their pharmacological effect is related to the inhibition of angiotensin I-converting enzyme (ACE) (Ianzer et al., 2007). Peak 4 of the HPLC separation of B. punctatus venom components (Fig. 1C) was identified as a BPP (Table 1), and its inhibitory activity on ACE was confirmed showing a half-maximal inhibition of this enzyme at 0.9 mg/mL (Fig. 3A). Interest in snake venom BPPs stems from their potential in the development of hypotensive drugs, as exemplified by Captopril®;. Overall, the comparison of B. punctatus venom with those of other pitvipers distributed in Colombia (Table 2) highlights the remarkable divergence of compositional profiles that have arisen through the evolution and diversification of snakes (Casewell et al., 2013).

The protein composition of B. punctatus venom correlates with the enzymatic activities assayed, as well as with those described in earlier studies (Otero et al., 1992; Kuch et al., 1996). L-amino acid oxidase (Fig. 3C), proteolytic (Fig. 4A), and PLA₂ (Fig. 4B) activities of this venom were corroborated. Interestingly, its proteolytic activity was higher than that of B. asper venom (Fig. 4A), and this might be related to the stronger hemorrhagic potency that was reported for B. punctatus venom in comparison to B. asper venom (Otero et al., 1992). Hemorrhage induced by viperid venoms is mainly dependent on the proteolytic action of SVMPs upon the microvasculature and its extracellular matrix support (Bjarnason & Fox, 1994; Gutiérrez et al., 2005), and this effect can be enhanced by venom components affecting haemostasis, such as procoagulant SPs with thrombin-like activity, or some CTL components and disintegrins that potently interfere with platelets, among others (Gutiérrez, Escalante & Rucavado, 2009; Calvete et al., 2005). Considering that the proportion of SVMPs is lower in B. punctatus than in B. asper venom (Table 2), the higher hemorrhagic action reported for the former (Otero et al., 1992) suggests that its abundant CTL components (16.7%) might include toxins that affect platelets, a hypothesis that deserves future investigation. On the other hand, the PLA₂ activity of B. punctatus venom was lower than that of B. asper (Fig. 4B), in agreement with their corresponding relative contents of these enzymes (Table 2). However, a major contrast was evidenced in the cytotoxic activity of these two venoms upon myogenic cells in culture, B. punctatus being essentially devoid of this effect, while B. asper causing overt cytolysis and LDH release under identical conditions (Fig. 4C). Since cytolysis of myogenic cells, an in vitro correlate for in vivo myotoxicity (Lomonte et al., 1999), has been shown to be mediated mainly by basic PLA₂s in the case of viperid venoms (Gutiérrez & Lomonte, 1995; Lomonte & Rangel, 2012), this finding anticipates that the catalytically active (D49) PLA₂s present in B. punctatus venom are likely to belong to the acidic type of these enzymes, which despite frequently having higher enzymatic activity than their basic counterparts, usually display very low, or even no toxicity (Fernández et al., 2010; Van der Laat et al., 2013). In contrast, the venom of B. asper is rich in basic D49 and K49 PLA₂s/PLA₂ homologues with strong cytolytic and myotoxic effects (Angulo & Lomonte, 2005, 2009) that would explain the present findings. Although at least one PLA₂ component of B. punctatus venom was shown to belong to the K49 type of catalytically-inactive, basic PLA₂ homologues (fraction 23–25a; Table 1), its low abundance (1.3%) in the venom would be in agreement with the observed lack of cytotoxicity (Fig. 4C).

In summary, the general compositional profile of B. punctatus venom was obtained through the analytical strategy known as ‘snake venomics’. The present data add to the growing body of knowledge on the remarkable diversity of compositional strategies in snake venom ‘cocktails’, in spite of the reduced number of gene families that encode their proteins/toxins (Casewell et al., 2013; Calvete , 2013). Due to the key adaptive role of venoms, this knowledge, in combination with toxicological, ecological, and natural history information, could lead to a deeper understanding of the evolutionary trends and selective advantages conferred by particular venom compositions in the divergence of snakes. In addition, compositional data may offer a more comprehensive basis to foresee the features of envenomings by this pitviper species, largely unreported in the literature.