An overview of Bothrops erythromelas venom

This review discusses studies on the venom of Bothrops erythromelas published over the past 36 years. During this period, many contributions have been made to understand the venomous snake, its venom, and its experimental and clinical effects better. The following chronological overview is based on 29 articles that were published between 1979 and 2015, with emphasis on diverse areas. The complexity of this task demands an integration of multidisciplinary research tools to study toxinology. This science is in need of renewed conceptual and experimental platforms aimed at obtaining a profound understanding of the highly complex pathophysiology of snakebite envenoming and toxins isolated from snakes. Key-words: Bothrops erythromelas. Snake. Venom.


INTRODUCTION
Venomous and non-venomous snakes are distributed worldwide, especially in tropical and subtropical areas (1) .Envenoming is neglected in many countries, which makes it a public health concern (2) (3) (4) .Snake venoms are fascinating models for drug design (1) and their antidotes are still under development.As a result, research in this area has been in existence since the 19 th century (5) .Three hundred and eighty-six species of snakes have been identifi ed in Brazil, of which 62 are venomous, 32 belong to the family Elapidae, and 30 belong to the family Viperidae (6) .Bothrops erythromelas, which belongs to the family Viperidae (Figure 1), is responsible for most of the snakebite accidents in Northeast Brazil (7) .
The species is found in the Caatinga ecoregion, an exclusive Brazilian biome that covers an area of about 850,000km 2 and includes part of the Brazilian States of Piauí, Ceará, Rio Grande do Norte, Paraíba, Pernambuco, Alagoas, Sergipe, Bahia, Maranhão, and Minas Gerais (7) .

OVERVIEW ON BOTHROPS ERYTHROMELAS VENOM
To date, 29 studies have been published on Bothrops erythromelas.The fi rst study on this snake dates back to 1979.Nahas et al. (8) , compared the coagulating activity of 26 different Bothrops snake venoms.Using a specifi c clotting system, they showed that B. erythromelas venom presents no thrombin-like activity, because it was not able to directly clot fi brinogen.Furthermore, it is speculated that the absence of the thrombinlike activity is due to a fi brinogenolytic effect of the venom.
In 1990, Moura-da-Silva et al. (9) evaluated the differences in the distribution of myotoxins from different Bothrops species.Briefl y, antigens with high myotoxic activities were isolated from Bothrops jararacussu venom and their cross reactivities were analyzed by western blotting and enzymelinked immunosorbent assay using Bothrops antivenoms.Bothrops jararacussu, Bothrops moojeni, Bothrops neuwiedi, and Bothrops pradoi antivenoms were found to be active against the isolated myotoxins; however, B. erythromelas antivenom showed no activity against the antigens.Moreover, in vivo myotoxicity studies confi rmed the absence of related myotoxins in B. erythromelas venom.
In 1991, two related studies were conducted.The fi rst one compared 9 snake venoms from adult B. erythromelas females and their offspring (10) ; this study found that caseinolytic and fi brinolytic activities were lower in venoms from the newborn snakes than in those from their mothers.Although variations were observed in the amidolytic activities of the venoms from most Bothrops snakes and their offspring, the venom of B. erythromelas showed no amidolytic activity.In a comparative analysis of the coagulating activities of Bothrops venoms, B. erythromelas venom presented the highest levels of factor X (FX) and prothrombin activators without showing a thrombinlike activity.It was emphasized that the degree of procoagulation in the newborn snakes' venoms is related to the ability to activate prothrombin and FX.However, this procoagulating activity seems to decrease as the newborn snakes grow.Moreover, these researchers demonstrated that the venom of an adult B. erythromelas has a higher protein content than that from a newborn snake does.
The second study isolated and compared myotoxins from different species of Bothrops by fast protein liquid chromatography and isoelectrofocalization (11) .No basic proteins with phospholipase and/or myotoxic activity in the B. erythromelas venom was detected in this study.This confi rmed the data obtained by Moura-da-Silva et al. (9) .
In 1992, Maruyama et al. (12) complemented Nahas et al.'s work (8) by investigating the enzymatic properties of factor II (FII) and FX activators from B. erythromelas venom.They demonstrated that both activators are inhibited by ethylenediaminetetraacetate and 1,10-phenanthroline.They also hypothesized that metalloproteinases with molecular weights of 70-90kDa were present in the venom.These researchers observed that FII activator activity of B. erythromelas was approximately 30 times greater than that of Oxyuranus scutellatus venom but similar to that of Daboia russelli venom.Furthermore, they found that FX activator activity was calciumdependent and that the venom of B. erythromelas contains two hemorrhagic factors and two fi brinolytic enzymes.
Flores et al. (13) were the fi rst to study the proinfl ammatory effects of B. erythromelas venom; they demonstrated that Bothrops erythromelas and Bothrops alternatus venoms induced the migration of neutrophils into the peritoneal cavities of rats.When injected into the peritoneal cavity of rats and into the air pouch, the venom induced migration of the cells to the injection site.This migratory response was thought to occur due to the phospholipase activity of the venom.Moreover, B. alternatus venom showed a phospholipase activity that was two times higher than that exhibited by B. erythromelas venom.Additionally, B. erythromelas venom induced neutrophil recruitment 2-3 times more than that induced B. alternatus venom.Moreover, treatment with dexamethasone or nordihydroguaiaretic acid, followed by stimulation with Bothrops venom, resulted in a signifi cant reduction in neutrophil migration.This suggests that leukotriene B 4 , which is a lipoxygenase metabolite of an arachidonic acid derivative, acts as a chemotactic mediator.Macrophages are the main source of leukotriene B 4 .They are also the predominantly resident cells in the rat air pouch; therefore, rats injected with thioglycolate and then with B. erythromelas or B. alternatus venom had more neutrophils in their peritoneal cavities than the control rats had.
The reproductive aspects of some viviparous species from Bahia (Brazil), including B. erythromelas, was investigated by Lira-da-Silva et al. (14) .The researchers observed that the gestation period of B. erythromelas was about 123 days and that parturition occurred preferably in the summer season.Furthermore, the female snakes produced an average of 11 hatchlings/gestation with 16.80-19.20cm.
In 1998, Vasconcelos et al. (15) studied the in vivo distribution of B. erythromelas venom.In that study, the venom was labeled with 131 I and administered subcutaneously to mice.The results showed a higher amount of venom in subcutaneous tissues than in the heart, bladder, brain or diaphragm.This indicates that B. erythromelas venom does not target most internal organs; hence, systemic effects of envenomation from B. erythromelas would be related to an indirect action of the venom.
In some regions in Brazil, heparin is used for treating bothropic accidents associated with antibothropic serum (ABS).In 2001, Boechat et al. (16) verifi ed the action of heparin (Liquemine, Roche, Brazil) on the main biological activities (hemorrhagic, in vitro and in vivo coagulating, edematogenic, phospholipasic, and lethal) of Bothrops atrox and Bothrops erythromelas venoms.This study verifi ed that heparin (3 and 6IU) was not effective in neutralizing the hemorrhagic and coagulating activities of both venoms.However, heparin, at dose of 6 IU, was capable of neutralizing the edematogenic activity of B. erythromelas venom.It also increased the effectiveness of ABS.Moreover, heparin neutralized the phospholipase A 2 (PLA 2 ) activity of B. erythromelas venom by 28%.Furthermore, heparin was more effective in neutralizing the venom's lethal activity when it was administered with ABS.
Camey et al. (17) compared the toxic effects of the venoms from fi ve Bothrops snakes (Bothrops alternatus, Bothrops moojeni, Bothrops neuwiedi, Bothrops jararacussu, and Bothrops jararaca) and a combination of these venoms (AgB).They investigated the ability of polyvalent ABS, produced by Ezequiel Dias Foundation, to recognize and neutralize toxic compounds in the venoms.The results showed that ABS inhibited the toxic effects of each of the fi ve venoms, as well as those of Bothrops erythromelas, Bothrops atrox, and Bothrops leucurus venoms.
Silva et al. (18) were the fi rst to conduct a study involving the molecular cloning, purifi cation, and characterization of a prothrombin activator of B. erythromelas.Their fi ndings complemented those obtained by Nahas et al. (8) , and they successfully purifi ed the prothrombin activator berythrativase, by single cation-exchange chromatography.They demonstrated that berythrativase makes up approximately 5% of the venom.Additionally, berythrativase presented as a single band protein with a molecular weight of 78kDa under reducing conditions in sodium dodecyl sulfate polyacrylamide gel electrophoresis.This study showed that prothrombin hydrolysis by berythrativase resulted in a fragment pattern similar to that generated by Group A prothrombin activators.The latter convert prothrombin into meizothrombin; however, this occurs independent of the prothrombinase complex but is typical of a metalloproteinase.Furthermore, the enzymatic activity of berythrativase was rapidly inhibited by chelators, such as ethylenediaminetetraacetic acid and o-phenanthroline.It was also observed that after prolonged incubation with berythrativase, the Aα-chain of human fi brinogen was slowly digested; however, no effects on the β-or γ-chains were observed.Additionally, the enzyme triggered procoagulating and proinfl ammatory responses.It also positively regulated the surface expressions of intracellular adhesion molecule-1 and E-selectin on human umbilical vein endothelial cells (HUVECs).Berythrativase is functionally similar to Group A prothrombin activators and its primary structure is related to that of hemorrhagic metalloproteinases from snake venoms.As a result, it does not show a hemorrhagic activity, which is characteristic of other snake venom metalloproteinases.
In 2004, Zamuner et al. (19) compared the myotoxic and neurotoxic effects of Bothrops venoms (B.erythromelas, B. jararaca, B. jararacussu, B. moojeni, and B. neuwiedi) and evaluated their neutralization using a commercial antivenom produced by the Institute Vital Brazil (Rio de Janeiro, RJ, Brazil).Although all the venoms were myotoxic when they were assessed by their release of creatine kinase, the B. erythromelas venom seemed less active than the other venoms.This was in agreement with the data obtained by Moura-da-Silva et al. (9) and Moura-da-Silva et al. (11) .However, Zamuner et al. (19) indicated that the B. erythromelas venom was more lethal than the others were.In vitro neurotoxicity studies showed that the venoms were neurotoxic to chick nerve-muscle preparations.A commercial antivenom was used to neutralize the myotoxic effect of the venoms; however, its neuroprotective effect was variable, indicating that the neurotoxic venom component(s) differ among the venoms.Junqueira-de-Azevedo et al. (20) have investigated the B. erythromelas snake venom vascular endothelial growth factor (svVEGF), which is an angiogenic protein.The protein seems to be responsible for many of the features of Viperidae envenomation, such as hypotension and increase in vascular permeability, which results in spreading of the venom.In the work by Junqueira-de-Azevedo et al. (20) , western blot assays were conducted using mice antisera against svVEGF isolated from Bothrops insularis.The results indicated the presence of svVEGFs in B. erythromelas venom.The complete sequence of B. erythromelas svVEGF cDNA was found to contain 1213 nucleotides.The deduced protein showed an open reading frame of 146 amino acid residues, with an initiation codon (ATG) at position 205 and a stop codon (TGA) at position 643, which are similar to the respective codons in the B. insularis svVEGF sequence.
In 2004, the chromatographic behavior of an acidic PLA 2 isolated from B. erythromelas snake venom by size-exclusion chromatography was studied by Aird (21), who observed that the PLA 2 interacted hydrophobically with the matrix resin, which was constituted of agarose and dextran, thereby retaining the protein on the matrix.Additionally, it was highlighted that different buffers at various pHs, as well as organic solvents, such as acetonitrile (30%), could be used to improve chromatographic resolution.
Schattner et al. (22) complemented the results reported by Silva et al. (18) by evaluating the effects of two P-III snake venom metalloproteinases (SVMPs), berythrativase and jararhagin, isolated from B. erythromelas and B. jararaca, respectively, on HUVECs.Both SVMPs were capable of stimulating the releases of interleukin-8 (IL-8), nitric oxide, and prostacyclin (PGI 2 ) by the HUVECs.Berythrativase also increased the expression of decay-accelerating factor; however, it did not affect the viability of the HUVECs even when it was used at high concentrations.The former effect may be the reason for the hemorrhagic activity of berythrativase.
Grazziotin and Echeverrigaray (23) performed a random amplifi ed polymorphic DNA analysis to study the genetic relationships among 11 Bothrops species and found that B. erythromelas showed genetic relations with B. moojeni.
Some animals have a natural resistance to the effects of snake venoms, which in many cases can be explained by the presence of neutralizing factors in the animal's serum (24) (25) .Moreover, resistance to the effects of snake venom has been studied in the serum of Didelphis marsupialis, a very common opossum in South America (26) (27) (28) (29) .In 2005, Martins et al. (30) investigated the action of an antibothropic factor isolated from D. marsupialis serum on the renal effects of B. erythromelas venom without systemic interference.Their study showed that B. erythromelas venom reduced renal perfusion pressure (PP) and renal vascular resistance (RVR).Additionally, the venom decreased glomerular fi ltration rate (GFR) at 60 minutes and then increased it at 120 minutes after perfusion.Furthermore, the urinary fl ow (UF) was increased signifi cantly, whereas the tubular transport percentages of sodium (%TNa + ) and potassium (%K + ) decreased.They observed that the isolated antibothropic factor at 10μg/mL blocked the effects of the venom on PP, RVR, %TNa + , and %TK + .However, it was not effective in reversing the effects of the venom on UF and GFR.At higher concentrations (30μg/mL), the antibothropic factor was able to reverse all the renal effects induced by the B. erythromelas venom.Finally, they concluded that the B. erythromelas venom altered all the renal functional parameters that were evaluated.Additionally, the antibothropic factor inhibited all the renal effects induced by the venom on isolated kidneys from Wistar rats.
In 2006, Pereira et al. (31) studied berythrativase and jararhagin to complement the studies conducted by Schattner et al. (22) and Silva et al. (18) .Pereira et al. (31) studied the differences in the biological effects of the two SVMPs on different hemostatic properties.Next, they characterized the biological effects of the proteins and compared their effects on HUVECs.They evaluated the release and modulation of coagulation and fi brinolytic factors by the cells, as well as the expressions of their related genes.The results showed that berythrativase and jararhagin induced the release of von Willebrand factor but did not modulate its gene expression level.Additionally, berythrativase increased the expression of the tissue factor in the HUVECs.This study concluded that each SVMP acts in a specifi c manner.Specifi cally, jararhagin has a preferential local action, while berythrativase is a systemic procoagulating protein that acts on the surfaces of HUVECs.
De Albuquerque Modesto et al. (32) isolated a novel acidic phospholipase A 2 with an aspartate at position 49 (PLA 2 Asp-49) from B. erythromelas venom, named BE-I-PLA 2 , characterized it, and described its complete sequencing.BE-I-PLA 2 has a molecular weight of 13,649.57Daand a cDNA sequence of 457 base pairs (bp).They incubated BE-I-PLA 2 with platelet-rich plasma and showed that the former has a potent inhibitory effect on aggregation induced by arachidonic acid and collagen but not that by adenosine diphosphate.Moreover, BE-I-PLA 2 showed no binding to/interference with principal platelet receptors.BE-I-PLA 2 was shown to stimulate endothelial cells to release PGI 2 but not nitric oxide, which suggests that BE-I-PLA 2 has a potential antiplatelet activity in vivo.
These data are interesting because they highlight that berythrativase, a P-III snake venom metalloproteinases (SVMPs) from B. erythromelas, can stimulate endothelial cells to release nitric oxide and PGI 2 (22)   .Therefore, the effects of both toxins contribute to the effects observed after B. erythromelas bites.
Moura da Silva et al. (33) published another study on metalloproteinases.In that study, the effects of jararhagin, which is a hemorrhagic P-III SVMP, and berythrativase, which is a procoagulant and non-hemorrhagic P-III SVMP, were compared.The results showed that both SVMPs inhibited collagen-induced platelet aggregation.Moreover, the monoclonal antibody MAJar 3, which neutralizes the hemorrhagic effect of Bothrops venoms and inhibits the binding of jararhagin to collagen, did not react with berythrativase.Furthermore, the jararhagincollagen complex retained the catalytic activity of the toxin, as was evidenced by the hydrolysis of fi brin.Moreover, the three-dimensional structures of the metalloproteinases were studied to clarify why the two SVMPs exhibited different effects.The authors pointed out that the subdomain disintegrin located in front of the catalytic domain found in jararhagin is able to mediate the binding of the latter to collagen and react with the monoclonal antibody.The study therefore revealed a novel function of the disintegrin domain during hemorrhage.
Furthermore, this study describes that a bradykinin-potentiating peptide, QGGWPRPGPEIPP, is common to the seven Bothrops venoms.QGGWPRPGPEIPP is a specific marker because it is not present in the venom of Crotalus durissus terrifi cus (rattlesnake).The PCA of the peptides showed that the venom of B. erythromelas is phylogenetically close to those of B. jararaca and B. insularis.Additionally, its peptide profi le was similar to that of B. jararaca.B. alternatus is phylogenetically different from the other Bothrops species examined in the study, with the order of decreasing proximity being B. erythromelas, B. jararaca/B.insularis, B. jararacussu, B. moojeni, and B. leucurus.This relationship is the same as the one observed in the PCAs of the peptides in the various venoms, which showed that B. erythromelas venom is the closest to B. alternatus venom but the most different from B. leucurus venom.These researchers concluded that fi ngerprinting using direct infusion nano-ESI-MS in positive ion mode, followed by chemometric analysis provides a rapid means to analyze low-molar-mass peptides in snake venom samples.This provides a reliable mass fi ngerprinting spectra for venom classifi cation and quality control.
Rocha et al. (35) continued the studies conducted by Vasconcelos et al. (15) by investigating the plasma pharmacokinetics of B. erythromelas venom labeled with 131 I in the presence and absence of an antivenom in mice.In the presence of the antivenom, the percentage radioactivity of B. erythromelas in plasma was higher and its elimination half-life was longer than the respective values were in the absence of the antivenom.Thus, the data indicated a redistribution of the venom from the tissues to the vascular compartment associated with the treatment of envenomed mice with anti-venom 15 min after venom injection.The researchers concluded that the pharmacokinetics of B. erythromelas venom in the presence of an antivenom follows a modifi ed profi le and is likely the result of redistribution of the venom from the peripheral compartment to the central compartment.This may contribute to the understanding and optimization of treatment against envenoming in humans.
Estevão-Costa et al. (36)  Studies on the clinical and epidemiological profi les of snakebites caused by Bothrops and Bothropoides snakes, including Bothrops erythromelas, which are responsible for most of the snakebite accidents in Northeast Brazil (7) were reported only after 2010.Oliveira et al. (37) evaluated victims of snakebites who were admitted to the Paraiba Information Centre for Toxicological Assistance and found that, the annual incidence of snakebite accidents was 5.5 for every 100,000 inhabitants.They also found that the lethality rate following such accidents was 0.2%.The accidents were considered mild and the patients were examined about 6 hours after the bite.Most of the patients were rural male workers aged between 11 and 20 years old.The study also showed that the feet and toes were the most affected body parts.
Another research in this area was conducted by Luna et al. (38) , who collected blood samples from patients bitten by B. erythromelas before and after the patients were administered an antivenom.Humoral response [immunoglobulin M (IgM) production] was analyzed to ascertain the effectiveness of the treatment.In all the blood samples, a protein with a molecular weight of 38kDa was found before and after serum therapy.The results suggested that this protein could be used as a marker to indicate envenomation by B. erythromelas.
A study performed by Luna et al. (39) showed that, Bothrops erythromelas and Crotalus durissus cascavella venoms induced high interferon-gamma and IL-6 production in mouse splenocytes.Nitric oxide was signifi cantly produced in the splenocytes only by B. erythromelas venom, which also induced a higher rate of cell death than C. durissus cascavella venom did.The results of the study showed that B. erythromelas and C. durissus cascavella venoms induce distinct in vitro responses through cytokine and nitric oxide productions.This study complemented the data obtained by Flores et al. (13) , which fi rst showed the proinfl ammatory effects of B. erythromelas venom.
In a study by Machado et al. (40) , 140 venom samples were collected from 93 different localities and used to investigate specie boundaries.The study aimed to hypothesize the phylogenetic relationships among the venoms based on 1,122bp of cyt b and ND4 from mitochondrial DNA.Additionally, they investigated the patterns and processes involved in the evolutionary history of the group of venoms.The results indicated that B. neuwiedi venom is highly monophyletic.However, Bothrops diporus, Bothrops lutzi, Bothrops erythromelas, Bothrops mattogrossensis, Bothrops neuwiedi, Bothrops marmoratus, and Bothrops pauloensis were polyphyletic based on their morphologies.
Two recent studies by Jorge et al. (41) applied a venomics approach to defi ne the proteome and geographic variability of venoms of adult B. erythromelas from fi ve geographic regions.They found that the fi ve venoms exhibited highly conserved venom proteomes.The overall toxin profi le of a snake venom explains the local and systemic effects observed in envenomation.The five venoms showed qualitatively and quantitatively overlapping antivenomic profi les against the antivenoms [antibothropic-crotalic-laquetic, produced by the Instituto Clodomiro Picado (San José, Costa Rica); antibothropic, produced by Instituto Vital Brazil (Niterói, Brazil)] generated using different bothropic venoms in immunization mixtures.The large immunoreactive epitope conservation across the Bothrops genus offers promise for the generation of broad-spectrum bothropic antivenoms.
Another study by Santoro et al. (42) investigated the thrombinlike activity of venoms from hybrids born in captivity from the mating of a female B. erythromelas and a male B. neuwiedi, which are two species whose venoms are known to display ontogenetic variations.They found that the features of venoms from hybrid snakes are genetically controlled during ontogenetic development.Despite the presence of thrombin-like enzyme genes in hybrid snakes, they are silenced during the fi rst six months of life.

CONCLUSIONS
This overview discusses articles on the venom of Bothrops erythromelas that have been published over the past 36 years.During this period (1979-2015), many contributions have enabled a better understanding of the venomous snake, its venom, and its experimental and clinical effects.However, because B. erythromelas is the cause of most snakebite accidents in Northeast Brazil, the number of available publications is not suffi cient to elucidate the venom's variations and mechanism of action.Further studies must be conducted to improve the knowledge on the venom's components, which can be used in basic science and clinical applications to enhance the effectiveness of treatments.Moreover, the introduction of new methods in proteomics and genomics can lead to the discovery of new compounds.These can serve as research tools or templates for the development of drugs.The application of these sensitive and comprehensive methods allows studying any venom and/or its components possible.As such, the complexities of these tasks demand the integration of multidisciplinary research tools to study toxinology.
studied phospholipase A 2 inhibitors (PLIs) present in snake serum.Three different structural classes of PLIs (α, β, and γ) were noted from the study.The γ class members are potent inhibitors of PLA 2 and are from the venoms of Viperidae snakes.They further documented the γPLIs in the venoms of six Bothrops snakes (B.erythromelas, B. neuwiedi, B. leucurus, B. jararacussu, B. moojeni, B. alternatus, and B. jararaca).The mature proteins possessed 181 amino acid residues following a 19-residue signal peptide, similar to the γPLIs in the venom of Crotalus durissus terrifi cus.Two of the deduced proteins from B. erythromelas and B. neuwiedi venoms were considered as exceptions.They showed consistent insertions of 4-amino acid residues in their structures.However, further studies should be conducted on γPLIs because this class of proteins may be useful in the development of selective inhibitors of secretory PLA 2 from several sources.