lable at ScienceDirect

Toxicon 125 (2017) 74e83

Contents lists avai

Toxicon

journal homepage: www.elsevier .com/locate/ toxicon

Inhibition of local effects induced by Bothrops erythromelas snakevenom: Assessment of the effectiveness of Brazilian polyvalentbothropic antivenom and aqueous leaf extract of Jatropha gossypiifolia

Juliana F�elix-Silva a, Jacyra A.S. Gomes a, Jacinthia B. Xavier-Santos a, Júlia G.R. Passos a,Arn�obio A. Silva-Junior a, Denise V. Tambourgi b, Matheus F. Fernandes-Pedrosa a, *

a Laborat�orio de Tecnologia & Biotecnologia, Faculdade de Farm�acia, Universidade Federal do Rio Grande do Norte, Natal, RN, Brazilb Laborat�orio de Imunoquímica, Instituto Butantan, S~ao Paulo, SP, Brazil

a r t i c l e i n f o

Article history:Received 12 September 2016Received in revised form31 October 2016Accepted 23 November 2016Available online 25 November 2016

Keywords:Bothrops erythromelasCaatinga's lanceheadJatropha gossypiifoliaAntiophidic activityBothropic antivenom

Abbreviations: AE, aqueous leaf extract of Jatrophaof variance; AUC0e2h, area under the time-course curantivenom; BeV, Bothrops erythromelas venom; BjV, BoBrazilian Genetic Heritage Management Council; CIOOrganizations of Medical Sciences; CONCEA, NationaAnimal Experimentation of Brazil; DAMP, damage-aECM, extracellular matrix; ELISA, enzyme-linked immhuman embryonic kidney 293 cells; IL, interleukin;phosphate buffered saline; PLA2, phospholipase A2; Selectrophoresis with sodium dodecil sulphate; SEM, stBrazilian Biodiversity Authorization and Informationmetalloproteinase; TNF-a, tumor necrosis factor alph* Corresponding author. Av. Gal. Cordeiro de Farias

570, Natal, RN, Brazil.E-mail address: mpedrosa@ufrnet.br (M.F. Fernand

http://dx.doi.org/10.1016/j.toxicon.2016.11.2600041-0101/© 2016 Elsevier Ltd. All rights reserved.

a b s t r a c t

Bothrops erythromelas is a snake of medical importance responsible for most of the venomous incidentsin Northeastern Brazil. However, this species is not included in the pool of venoms that are used in theBrazilian polyvalent bothropic antivenom (BAv) production. Furthermore, it is well known that anti-venom therapy has limited efficacy against venom-induced local effects, making the search for com-plementary alternatives to treat snakebites an important task. Jatropha gossypiifolia is a medicinal plantwidely indicated in folk medicine as an antidote for snakebites, whose effectiveness against Bothropsjararaca venom (BjV) has been previously demonstrated in mice. In this context, this study assessed theeffectiveness of the aqueous extract (AE) of this plant and of the BAv against local effects induced byB. erythromelas venom (BeV). Inhibition of BeV-induced edematogenic and hemorrhagic local effects wasassayed in mice in pre-treatment (treatment prior to BeV injection) and post-treatment (treatment post-envenomation) protocols. Inhibition of proteolytic, phospholipase A2 (PLA2) and hyaluronidase enzy-matic activities of BeV were evaluated in vitro. BAv cross-reactivity and estimation of antibody titersagainst BeV and BjV were assessed by Ouchterlony double diffusion test. The results show that in pre-treatment protocol AE and BAv presented very similar effects (about 70% of inhibition for edemato-genic and 40% for hemorrhagic activities). However, BAv poorly inhibited edema and hemorrhage inpost-envenomation protocol, whilst, in contrast, AE was significantly active even when used after BeVinjection. AE was able to inhibit all the tested enzymatic activities of BeV, while BAv was active onlyagainst hyaluronidase activity, which could justify the low effectiveness of BAv against BeV-induced localeffects in vivo. Ouchterlony's test showed positive cross-reactivity against BeV, but the antibody titerswere slightly higher against BjV. Together, these data indicate that despite the presence of immuno-logical cross-reactivity, Brazilian polyvalent bothropic antivenom presented low inhibitory potentialagainst biological and enzymatic effects of BeV, illustrating the need for new strategies in the productionof antivenom with broad neutralizing potential in the treatment of Bothrops spp. envenomationthroughout the country. Together, the results highlight the antiophidic potential of J. gossypiifolia,

gossypiifolia; ANOVA, analysisves after 2 h; BAv, bothropicthrops jararaca venom; CGEN,MS, Council of Internationall Council for the Control ofssociated molecular pattern;unosorbent assay; HEK-293,MPO, myeloperoxidase; PBS,DS-PAGE, polyacrylamide gelandard error of mean; SISBIO,System; SVMP, snake venoma., s/n, Petr�opolis, CEP 59012-

es-Pedrosa).

J. F�elix-Silva et al. / Toxicon 125 (2017) 74e83 75

suggesting that it can be considered a potential adjuvant in the treatment of bothropic envenomationlocal effects.

© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Snake envenoming is a serious public health problem in manyregions around the world. In 2013e2015, the Bothrops snakescaused about 87% of snakebites that occurred in Brazil (Brasil,2016). Bothropic envenomation is characterized by immediateand intense local effects (edema, hemorrhage and necrosis, whichcan often lead to permanent disabilities), cardiovascular alterations(especially systemic bleeding and hypovolemic shock), coagulop-athy and renal alterations (which could evolve into acute kidneyinjury) (Brasil, 2010; Guti�errez et al., 2009a; Guti�errez andLomonte, 1989). The specific treatment for bothropic envenom-ation is the intravenous administration of bothropic antivenom(BAv). This antivenom is produced in Brazil by hyperimmunizationof horses with a pool of venoms from Bothrops jararaca, Bothropsjararacussu, Bothrops moojeni, Bothrops alternatus and Bothropsneuwiedi snakes, with the aim of neutralizing all the Bothropsvenoms distributed in this country (Brasil, 2010). However, someprevious studies have shown pre-clinical evidence that the anti-venom used in Brazil may not fully neutralize the toxic activitiesinduced by all bothropic venoms, suggesting that other venomsshould be included in the immunization pool for the preparation ofa universal bothropic antivenom (Muniz et al., 2000; Queiroz et al.,2008). Many factors including phylogeny, sex, geographic origin,season, age and prey preference may influence the venomcomposition, which could affect the neutralizing capacity of anti-venom (Chippaux et al., 1991). Bothrops erythromelas (Amaral,1923), commonly known as “Caatinga lancehead”, “jararaca daseca” or “jararaca malha de cascavel”, is a small terrestrial venomoussnake responsible for most of the snakebites in Northeastern Brazil(Lira-da-Silva et al., 2009). Therefore, since B. erythromelas is notincluded in the pool of venoms that are used in the Brazilianpolyvalent bothropic antivenom production, this calls into questionthe efficacy of BAv.

Snake venom toxins such as metalloproteinases (SVMPs),phospholipases A2 (PLA2) and hyaluronidases are mostly respon-sible for the local tissue damage frequently observed in bothropicenvenomation. The action of these toxins are, in general, poorlyneutralized by antibodies present in antivenom, so it is well knownthat antivenom therapy has limited efficacy against venom-induced local effects, making the search for complementary alter-natives to treat snakebites an important task (Guti�errez et al.,2006). In addition to poor inhibition of local damage, there is arisk of development of immunological reactions, the high cost ofproduction and very difficult access in some regions (Guti�errezet al., 2011; Silva et al., 2015). This inability to treat local effects,as well the increased time between accident and treatment, are themain reasons for the temporary or permanent disability observedin many victims, which can lead to serious negative social, eco-nomic, and health impacts, given that most victims live in ruralareas (Guti�errez et al., 2013).

The use of medicinal plants against snakebites is a historicalpractice carried out throughout human history, with this knowl-edge being transferred down among the rural communities fromgeneration to generation (Butt et al., 2015). Nowadays, these herbalantidotes used in traditional folk medicine have gained muchattention by toxinologists worldwide as a tool for designing potent

inhibitors against snake venom toxins (Sulochana et al., 2015). Thepossible advantages of a potential application of medicinal plants asantiophidic agents are that, in general, they would be cheap, easilyavailable, stable at room temperature and could be able toneutralize a broad spectrum of toxins, including those related to thelocal tissue damage (Gomes et al., 2010; Santhosh et al., 2013).

Jatropha gossypiifolia is a medicinal plant popularly known inBrazil as “pinh~ao-roxo” or worldwide as “bellyache-bush”, and itbelongs to the Euphorbiaceae family. It is largely used in folkmedicine for various purposes, especially as antiophidic, anti-inflammatory, anti-hemorrhagic, hemostatic and healing,amongst others (F�elix-Silva et al., 2014a). The effectiveness of theaqueous extract of this plant against enzymatic and biological ef-fects of Bothrops jararaca snake venom has been previouslydemonstrated, including significant inhibitory potential against thelocal tissue damage induced by this venom in mice (F�elix-Silvaet al., 2014d).

In this context, this study was carried out aiming two maingoals: (a) to evaluatewhether the polyspecific bothropic antivenom(BAv) manufactured in Brazil is effective against local effectsinduced by B. erythromelas snake venom (BeV), and (b) to assess theefficacy of the aqueous leaf extract of J. gossypiifolia (AE) againstthese effects.

2. Material and methods

2.1. Chemicals and reagents

Azocasein, hexadecyltrimethylammonium bromide, hyaluronicacid and o-dianisidine were purchased from Sigma-Aldrich (St.Louis, MO, USA). Hydrogen peroxide was purchased from Merck(Darmstadt, HE, Germany). Agarosewas purchased fromUnisciencedo Brasil (S~ao Paulo, SP, Brazil). Coomassie Brilliant Blue R-250 waspurchased from Vetec Química Fina Ltda. (Duque de Caxias, RJ,Brazil). Sodium thiopental was purchased from Crist�alia (Itapira, SP,Brazil). Drabkin's reagent (120mMpotassium ferricyanide,150mMpotassium cyanide and 0.1% non-ionic detergent in 200 mM po-tassium dihydrogen phosphate) and hemoglobin standard solutionwere purchased from Bioclin (Belo Horizonte, MG, Brazil). All otherreagents and solvents usedwere of analytical grade. Thewater usedwas purified by reverse osmosis. Phosphate buffered saline (PBS)was utilized and contained the following constituents: 137 mMNaCl, 3 mM KCl, 1.5 mM KH2PO4, 10 mM Na2HPO4, pH 7.4.

2.2. Snake venom

Venoms from Bothrops erythromelas (BeV) and Bothrops jararaca(BjV) snakes were a kind gift from Kathleen Fernandes Grego,Instituto Butantan, SP, Brazil. The venomswere obtained bymanualextraction from adult specimens and then lyophilized and keptat �20 �C until used. Venom solutions were prepared with PBS attime of use. The amount of venom was expressed by protein con-tent, determined by Bradford protein assay using albumin asstandard (Bradford, 1976). The scientific use of the venoms wasapproved by the Brazilian Genetic Heritage Management Council(CGEN) (Process number 010844/2013-9).

J. F�elix-Silva et al. / Toxicon 125 (2017) 74e8376

2.3. Plant material

2.3.1. Plant collection and identificationLeaves of Jatropha gossypiifolia L. (Euphorbiaceae) plant were

collected in April 2012 in Entrocamento's road, a village near Car-naubais, a municipality of Rio Grande do Norte State, Brazil (5.27�S,36.8�W). The botanical identification of the material was per-formed by Msc. Alan de Araújo Roque and a voucher specimen wasdeposited at the Herbarium in the Centro de Biociencias of Uni-versidade Federal do Rio Grande do Norte (UFRN 12561).

The leaves were dried at room temperature, triturated andstored in hermetically sealed bottles protected from light and hu-midity until use for extract preparation.

The collection of the plant material was conducted underauthorization of Brazilian Biodiversity Authorization and Informa-tion System (SISBIO) (Process number 35017) and Brazilian GeneticHeritage Management Council (CGEN) (Process number 010844/2013-9).

2.3.2. Aqueous extract (AE) preparationAqueous extract (AE) from J. gossypiifolia was prepared by

decoction, by extracting dried leaves with purified water to theratio 1:10 (w/v) for 15 min at 100 �C. The method of extraction waschosen based on previous literature that indicates that this is themain form of utilization in folk medicine (tea decoction) and onprevious research by our group (F�elix-Silva et al., 2014a, 2014c,2014d). A yield of 13.57% (relative to dry plant) was obtained. Theaqueous extract obtained after vacuum filtration was freeze-driedand dissolved in PBS at the time of use, at adequate concentra-tions for the biological assays.

2.4. Bothropic antivenom (BAv)

The bothropic antivenom (BAv) used was produced by theInstituto Butantan, S~ao Paulo, SP, Brazil, from plasma of horses thathad been immunized with a mixture of the following venoms:Bothrops jararaca (50%), Bothrops neuwiedi (12.5%), Bothrops alter-natus (12.5%), Bothrops moojeni (12.5%) and Bothrops jararacussu(12.5%). The batch number was 135101A, and according to the labelof the product, each millilitre of BAv is able to neutralize 5 mg ofB. jararaca venom (reference venom) in lethality tests in mice.

2.5. In vivo experiments

2.5.1. AnimalsAll the procedures involving animals were performed in

agreement with the recommendations of the Brazilian NationalCouncil for the Control of Animal Experimentation (CONCEA) andthe International Guiding Principles for Biomedical ResearchInvolving Animals of the Council of International Organizations ofMedical Sciences (CIOMS). The experimental protocols wereapproved by the Ethics Committee on Animal Use from Uni-versidade Federal do Rio Grande do Norte (protocol no. 004/2013).

Swiss albino mice (about 30 g, 6e8 weeks old), from both sexes,supplied by the animal facility of Centro de Ciencias da Saúde fromUniversidade Federal do Rio Grande Norte were used. The micewere matched by sex and age in all procedures. The animals werehoused in standard polypropylene cages (30 � 19 � 13 cm) andmaintained under controlled temperature (22 ± 2 �C) in a 12 hlight/dark cycle. The mice were fed with a standard laboratoryextruded food (Presence®, purchased from Agroline, CampoGrande, MS, Brazil) and water ad libitum. At the end of the exper-iments, the animals were euthanized by sodium thiopental over-dose (100 mg/kg) by intraperitoneal (i.p.) route.

2.5.2. Treatment protocols and doses employedDoses of AE and BAv were administered following two treat-

ment protocols: pre-treatment (AE or BAv used 60 min beforevenom was injected) and post-treatment (AE or BAv administered1 min after venomwas injected). In both protocols, AE (400 mg/kg)or BAv (100 mL/mice) were administered by intraperitoneal route.AE doses used were based on dose-effect studies (result notshown), fromwhere we chose the most effective one. Dose and viaof administration of BAv were chosen based on pilot assays, pre-vious studies found in literature and the actual label of the product,which indicates that 1 mL of BAv neutralizes 5 mg of B. jararacavenom (reference venom), which corresponds to a proportion of1:0.2 BjV: BAv (w/v). Considering the challenge of doses employedin this work, an excess of BAv in relation to the label of the productwas used, to avoid possible false-negative results due to low dosageof BAv (Supplementary Table 1). Groups where envenomed animalswere treated with PBS were used as control (venom control). Inaddition, as negative control, animals received only PBS instead ofvenom and AE or BAv (health control).

2.5.3. Edematogenic activityThe edematogenic activity of BeV was evaluated using the paw

edema model, as previously described in the literature with a fewmodifications (F�elix-Silva et al., 2014d). BeV (1 mg/50 mL of PBS) wasinjected subcutaneously into the right hind paw. Control animalsreceived equal volume of PBS (normal control). The BeV dose waschosen based on pilot assays, where the selected dose inducedsignificant paw edema without producing paw hemorrhage. Theindividual right hind paw thickness was measured immediatelybefore venom injection (basal value) and at selected time intervalsafter edema induction (0.25, 0.5, 1, 1.5 and 2 h), using a digitalcaliper (Digimess, S~ao Paulo, SP, Brazil). Edema was expressed asthe percentage difference between the thickness of the paw after(at respective time points) and before (basal values) venom injec-tion, as mean ± standard error of mean, with n ¼ 5 animals pergroup. In addition, the area under the time-course curve after 2 h(AUC0e2h) was calculated using trapezoidal rule.

After 2 h of BeV injection, the animals were euthanized and theirright hind paws were collected for quantification of myeloperox-idase (MPO) enzyme activity, as a biochemical marker of neutrophilmigration to the inflammation site, as previously described(Bradley et al., 1982). Briefly, paw skin tissues were weighed,chopped and homogenated in 0.5% hexadecyltrimethylammoniumbromide buffer (1 mL of buffer for each 50 mg of tissue). Then,samples were sonicated in an ice bath for 30 s, submitted to threefreeze-thaw cycles and finally sonicated for 30 s once more, forMPO enzyme extraction. The supernatant obtained after centrifu-gation at 10,000 g for 10 min at 4 �C was used for MPO activitydetermination, by mixing 20 mL of each supernatant with 200 mL of50 mM potassium phosphate pH 6.0 containing 0.0005% hydrogenperoxide and 0.167 mg/mL o-dianisidine. The measurement of theactivity was carried out at 460 nm using a microplate reader(Epoch-Biotek, Winooski, VT, USA), through kinetic reading at1 min intervals, during 3 min. One unit of MPO is defined as theequivalent to the consumption of 1 mmol of hydrogen peroxide perminute, considering that 1 mmol of hydrogen peroxide gives achange in absorbance of 1.13 � 10�2 per minute (Posadas et al.,2004). Samples of each animal were analysed in triplicate and themean of these determinations was used to express the result asmean ± standard error of mean, with n ¼ 5 animals per group.

2.5.4. Hemorrhagic activityThe hemorrhagic activity of BeV was evaluated using the local

skin hemorrhage model, as previously described in the literature,with a fewmodifications (Domingos et al., 2015; Roodt et al., 2000).

J. F�elix-Silva et al. / Toxicon 125 (2017) 74e83 77

BeV (25 mg/100 mL of PBS) was injected subcutaneously in thecentral region of the abdomen of mice. Control animals receivedequal volume of PBS (normal control). After 3 h, the animals weresacrificed and the inner surface of the skin was exposed. Thehemorrhagic skin was excised out and weighed on an analyticalbalance. The hemorrhage was expressed as the mass of hemor-rhagic halo formed after subcutaneous injection of venom, ingrams, as mean ± standard error of mean, with n ¼ 5 animals pergroup.

Thereafter, the hemorrhagic skin was fragmented and homog-enized with 3 mL of Drabkin's reagent for hemoglobin extraction.After incubation for 48 h at 8 �C, the samples were firstly centri-fuged at 392 g for 10 min, at room temperature. Then, 1 mL of thissupernatant was centrifuged at 16,000 g for 30 min, at 4 �C. Finally,the clear supernatant obtained was read at 540 nm in a microplatereader (Epoch-Biotek, Winooski, VT, USA), using Drabkin's reagentas blank. The hemoglobin content in excised hemorrhagic halo wasdetermined through a standard curve using hemoglobin. Thehemorrhage was expressed as the hemoglobin content extractedfrom excised hemorrhagic halo formed after subcutaneous injec-tion of venom, in mg/mL. Samples of each animal were analysed intriplicate and the mean of these determinations was used to ex-press the result as mean ± standard error of mean, with n ¼ 5animals per group.

2.6. In vitro enzymatic assays

2.6.1. AE and BAv pre-incubationFor inhibition studies, a fixed amount of BeV was pre-incubated

for 30 min at 37 �C with varying amounts of AE or BAv. Then, thevenom þ antivenom mixtures were submitted to enzymatic assayas described below. For phospholipase A2 (PLA2) and proteolyticactivities, the 1:25, 1:50 and 1:100 BeV: AE (w/w) ratios wereemployed, while in hyaluronidase activity, the ratios used were 1:1,1:2.5 and 1:5 (w/w). For BAv inhibitory evaluation, in all enzymatictests performed, the ratios BeV: BAv (w/v) 1:0.25, 1:0.5 and 1:1were used (Supplementary Table 1). As in the in vivo studies,considering the challenge doses employed in this work, there is anexcess of BAv to avoid possible false-negative results due to lowdosage of BAv, as explained in Supplementary Table 1. Tubes whereAE or BAv were substituted for PBS (ratio 1:0, BeV: antivenom, w/v)were used as control (venom control).

2.6.2. Proteolytic activity upon azocaseinProteolytic activity of BeV was determined colorimetrically us-

ing azocasein as substrate, as previously described in the literature,with some adaptations (Perea~nez et al., 2013). For the assay,15 mg ofBeV were pre-incubated for 30 min at 37 �C with different con-centrations of AE or BAv in PBS, at a final volume of 100 mL. Then, ineach tube, 100 mL of azocasein (10 mg/mL in 50 mM Tris-HCl pH 7.4buffer containing 200 mM NaCl and 5 mM CaCl2) was added andincubated for 90 min at 37 �C. The enzymatic reaction was stoppedby adding 100 mL of 5% trichloroacetic acid. After 30min standing atroom temperature, the tubes were centrifuged at 15,000 g for10 min at 22 �C. The supernatant (100 mL) was removed and mixedwith equal volume of 0.5 M NaOH in a 96-well microplate. After10 min at room temperature, the samples were read at 440 nm in amicroplate reader (Epoch-BioTek, Winooski, VT, USA). Blanks foreach concentration of AE or BAv were prepared the same way,except by adding the substrate only after trichloroacetic acidaddition. The proteolytic activity was calculated as the percentageof degradation products of azocasein formed, taking into accountthe absorbance increase compared to the control in which onlysubstrate was incubated (absence of venom, AE and/or BAv). Theresults were expressed as percentage of proteolytic activity in

relation to control in which only BeV was incubated (absence of AEor BAv, considered 100% of proteolytic activity), as mean ± standarderror of mean, with n ¼ 3.

2.6.3. Phospholipase A2 (PLA2) activityPLA2 activity of BeV was determined turbidimetrically in 96-

well microplates using an egg yolk suspension as substrate, aspreviously described in the literature, with some adaptations (Liuet al., 2015; Marinetti, 1965). First, egg yolk was mixed with PBSin a 1:3 (v/v, egg yolk: PBS) ratio, and then centrifuged at 400 g for2 min at room temperature to obtain in the supernatant the stocksuspension of egg yolk. Before the assay, the substrate concentra-tion was determined as that which gives an absorbance of 0.6 at925 nm, whichwas a 5% dilution of the stock suspension of egg yolkin 50mM Tris-HCl pH 7.4 buffer containing 200mMNaCl and 5mMCaCl2. For the assay, 100 mL of this suspension was mixed with100 mL of a solution containing 2.5 mg of BeV pre-incubated for30 min at 37 �C with different concentrations of AE or BAv in PBS.Blanks for each concentration of AE or BAv were prepared similarly,except for replacing the substrate with an equal volume of assaybuffer. After 30 min at 37 �C, the absorbance was read at 925 nm,using a microplate reader (Epoch-BioTek, Winooski, VT, USA). ThePLA2 activity was calculated as the percentage of remaining sub-strate, taking into account the turbidimetric decrease compared tothe control in which only substrate was incubated (absence ofvenom, AE and/or BAv). The results were expressed as percentageof PLA2 activity in relation to control in which only BeV was incu-bated (absence of AE or BAv, considered 100% of PLA2 activity), asmean ± standard error of mean, with n ¼ 3.

2.6.4. Hyaluronidase activityHyaluronidase activity of BeV was determined turbidimetrically

in 96-well microplates using hyaluronic acid as substrate, as pre-viously described in the literature, with some adaptations (Paix~ao-Cavalcante et al., 2015). The assay mixture contained acetate buffer(0.2 M sodium acetate pH 6.0 containing 0.15 M NaCl), 10 mg ofhialuronic acid (0.5 mg/mL in acetate buffer) and 3 mg of BeV pre-incubated for 30 min at 37 �C with different concentrations of AEor BAv in PBS, in a final volume of 100 mL. Blanks for each con-centration of AE or BAv were prepared similarly, except forreplacing the substrate with an equal volume of assay buffer. Thismixture was incubated for 60 min at 37 �C and then the enzymaticreaction was stopped by adding 200 mL of 2.5% hexadecyl-trimethylammonium bromide buffer in 2% sodium hydroxide. After10 min at room temperature, the absorbance was read at 925 nm,using a microplate reader (Epoch-BioTek, Winooski, VT, USA). Thehyaluronidase activity was calculated as the percentage ofremaining substrate, taking into account the turbidimetric decreasecompared to the control in which only substrate was incubated(absence of venom, AE and/or BAv). The results were expressed aspercentage of hyaluronidase activity in relation to control in whichonly BeV was incubated (absence of AE or BAv, considered 100% ofhyaluronidase activity), as mean ± standard error of mean, withn ¼ 3.

2.7. Ouchterlony double immunodifusion test

2.7.1. Comparison of BeV and BjV antigensBAv cross-reactivity was investigated by double immunodifu-

sion test, as previously described (Ouchterlony, 1953). B. jararacavenom (BjV), which is the main antigen present in the pool ofimmunization for BAv production (comprises 50% of antigenicmixture) was used for comparison. Briefly,1% agarosewas preparedin PBS, heated, poured into a clean glass Petri dish and allowed tosolidify, producing a 3 mm thick gel. Using a template, three wells

Fig. 1. Inhibition of edematogenic activity of Bothrops erythromelas venom (BeV) byaqueous extract (AE) of Jatropha gossypiifolia and bothropic antivenom (BAv). (A) Time-course of paw edema. **P < 0.01 and ***P < 0.001, when compared to the BeV controlgroup in two-way ANOVA followed by Bonferroni's test. (B) Area under the time-course curves after 2 h (AUC0e2h) values. ***P < 0.001, when compared to the BeVcontrol group in one-way ANOVA followed by Tukey's test. Data showed asmean ± SEM (n ¼ 5/group).

J. F�elix-Silva et al. / Toxicon 125 (2017) 74e8378

(5mm)were punched 10mmapart, forming an equilateral triangle.In the central well, 15 mL of BAv was applied, while in the peripheralwells 15 mL of solutions of BeV and BjV (1 mg/mL, each one) wereused. The plates were then incubated at 37 �C in a humidifiedchamber overnight, so as antigen-antibody complexes formed byimmunodifusion. The positive results were taken as the presence ofprecipitin lines between BAv and each venom. The pattern of linesformed was classified by pattern of identity, partial identity or non-identity, to evaluate the similarity between BeV and BjV and theirpossible cross-reactivity.

2.7.2. Antibody titers estimationThe titers of antibodies against BeV and BjV were determined

using serial dilutions of BAv and a double immunodifusion test aspreviously described, with a few adaptations (Lanari et al., 2014).Agarose gels were prepared as described above. Using anothertemplate, six wells (5 mm) were punched 10 mm apart, forming aregular hexagon. In a seventh well, located in the centre of thehexagon, 15 mL of BeV or BjV (1 mg/mL) was applied, while in thesixth peripheral well 15 mL of serial dilutions of BAv (1/1 to 1/32)were placed. The plates were then incubated at 37 �C in a humid-ified chamber, overnight. To increase technique sensitivity, afterdeveloping the precipitin bands, the gels were removed from thePetri dishes to be washed 3 times for 15 min with 0.1 M NaCl so-lution, then dried at 37 �C for 60 min, and stained with 0.3% Coo-massie Brilliant Blue R-250 (in 40% methanol and 10% acetic acidsolution) for 5 min. After washing in destaining solution (35%methanol and 10% acetic acid), gels were dried in air and photodocumented. The antibody titers against BeV and BjV venoms weredefined as the reciprocal of the highest dilution of antivenom givinga positive precipitin band.

2.8. Data analysis

All results are presented as mean ± standard error of mean(SEM). One-way ANOVA followed by Tukey's test or Two-wayANOVA followed by Bonferroni's test were performed usingGraphPad Prism version 5.00 (San Diego, CA, USA). P values lessthan 0.05 were considered significant.

3. Results

3.1. Inhibition of edematogenic activity by AE and BAv

The edematogenic activity of BeV was almost completelyinhibited (about 90% of inhibition) by AE 2 h after venom injectionin mice paws (P < 0.001), when this extract was administeredbefore venom injection (Fig. 1, Supplementary Table 2). This resultwas very similar to that of BAv, also administered before BeV.However, in the post-treatment protocol only AE was effective,inhibiting about 67% of BeV edema at 2 h (P < 0.001), while BAv didnot reach statistical significance after 2 h of edema observation(P > 0.05). As observed in Fig.1A and Supplementary Table 2, in pre-treatment protocol, AE reached a significant inhibitory percentage(66.3%) as quickly as 15 min (P < 0.001), while in post-treatment, asexpected due to pharmacokinetic reasons, a lower inhibition per-centage was observed (42.8%), although also statistically significant(P < 0.001). Also as shown in Fig. 1A and Supplementary Table 2, itcould be observed that BAv reached significant (P < 0.01) inhibitorypercentage (40.2%) at 15 min, when administered before venom.However, when the post-treatment protocol was used, no inhibi-tory activity was observed; only at the end of the period of obser-vation (about 1.5e2 h after envenomation) could be observed asmall extent of inhibition, but this did not reach statistical signifi-cance (P > 0.05).

Considering the total edema effect, calculated by the area underthe time-course curves after 2 h (AUC0e2h) values (Fig. 1B), it couldbe observed that while in pre-treatment protocol both AE and BAvwere active, when treatment was given after envenomation only AEwas active (P < 0.001).

A similar behaviour was also observed regarding the myelo-peroxidase (MPO) extracted from the mice paws, i.e. both AE(P < 0.001) and BAv (P < 0.01) were active in pre-treatment pro-tocol, inhibiting about 70 and 42%, respectively (Table 1); but onlyAE presented statistically significant anti-inflammatory activitywhen dosed after envenomation (P< 0.001, about 63% of inhibition)(Fig. 2). However, while pre-treatment with BAv diminished thepaw edema, this did not occur with MPO activity, as it did not reachstatistical significance (P > 0.05), indicating that this inflammatoryparameter was not really influenced by BAv treatment.

3.2. Inhibition of local hemorrhagic activity by AE and BAv

The hemorrhagic activity of BeV was partially inhibited by AEand BAv (shown in Fig. 3 and Table 1). In general, AE, in both preand post-treatment protocols, showed similar results in relation toBAv when used prior to envenomation. As observed for edemato-genic activity, BAv when administered after BeV injection did notproduce a statistically significant inhibitory effect (P > 0.05).

Two parameters were used to evaluate BeV-induced skin hem-orrhage: hemorrhagic halo weight and hemoglobin content fromthese haloes. Regarding hemorrhagic halo weight (Fig. 3A), AE inboth protocols presented similar inhibitory percentages (about41%), with different significance levels (P < 0.01 and P < 0.001 for

Table 1Summary of inhibition percentages of aqueous extract of Jatropha gossypiifolia (AE) and bothropic antivenom (BAv) against edematogenic and hemorrhagic local effectsinduced by Bothrops erythromelas venom (BeV) in vivo.

Treatment Protocol Edematogenic activity Hemorrhagic activity

AUC0e2h MPO activity Halo weight Hemoglobin

AE Pre-treatment 72.9 ± 3.2*** 69.2 ± 12.5*** 40.8 ± 5.1** 30.8 ± 5.5*Post-treatment 50.2 ± 6.2*** 63.0 ± 8.6*** 42.7 ± 10.1*** 41.9 ± 11.2**

BAv Pre-treatment 73.0 ± 2.1*** 41.6 ± 1.6*** 48.1 ± 4.4*** 52.3 ± 8.3***Post-treatment 17.4 ± 8.9 30.4 ± 6.9 26.2 ± 8.0 24.9 ± 2.6

AUC0e2h: area under time-course after 2 h. MPO:myeloperoxidase. Data showmean ± SEM (n¼ 5/group). *P < 0.05, **P < 0.01 and ***P < 0.001, when compared to BeV controlgroup in one-way ANOVA followed by Tukey's test. Inhibition percentage calculated as follows: [1 e (%Activitytest e %ActivityPBS control mean) ÷ (%ActivityBeV control mean e %ActivityPBS control mean)] � 100.

Fig. 2. Effect of aqueous extract (AE) of Jatropha gossypiifolia and bothropic antivenom(BAv) on myeloperoxidase (MPO) activity in mice paws injected with Bothrops eryth-romelas venom (BeV). ***P < 0.001, when compared to the BeV control group in one-way ANOVA followed by Tukey's test. Data showed as mean ± SEM (n ¼ 5/group).

Fig. 3. Inhibition of hemorrhagic activity of Bothrops erythromelas venom (BeV) byaqueous extract (AE) of Jatropha gossypiifolia and bothropic antivenom (BAv). (A)Weight of hemorrhagic haloes formed after subcutaneous injection of BeV in mice skinafter 3 h. (B) Hemoglobin content in hemorrhagic haloes obtained in (A). *P < 0.05,**P < 0.01 and ***P < 0.001, when compared to the BeV control group in one-wayANOVA followed by Tukey's test. Data showed as mean ± SEM (n ¼ 5/group).

J. F�elix-Silva et al. / Toxicon 125 (2017) 74e83 79

pre- and post-treatment protocols, respectively). BAv, in pre-treatment protocol, inhibited almost 50% of hemorrhagic haloweight, thus being a little bit more active than AE, although thisdifference was not statistically significant (P > 0.05). However, inpost-treatment protocol, BAv inhibited only about 26% of hemor-rhagic halo weight, and did not reach statistical significance(P > 0.05).

When evaluating the extracted hemoglobin content from thehemorrhagic haloes (Fig. 3B), similar results were observed(Table 1). AE was active in both treatment protocols, however,interestingly, AE was a little bit more active when administeredafter envenomation (P < 0.01, 41.9% of inhibition), thanwhen it wasused before venom application (P < 0.05, 30.8% of inhibition);however, this difference between treatment protocols was notstatistically significant (P > 0.05). The same as what occurred inhemorrhagic halo weight, BAv in pre-treatment protocol decreasedthe hemoglobin content in 52.3% (P < 0.001), but did not reachstatistical significance in the post-treatment protocol (P > 0.05),although it did achieve an inhibitory potential of almost 25% of thehemoglobin content (Table 1).

3.3. Inhibition of enzymatic activities by AE and BAv

AEwas able to inhibit significantly the proteolytic (Fig. 4A), PLA2(Fig. 4B) and hyaluronidase (Fig. 4C) activities of BeV. For PLA2 andhyaluronidase activities, the inhibition percentage reached 100%.Proteolytic activity was not totally inhibited but reached up to 83%of inhibition.

BAv, on the other hand, was able to inhibit completely only thehyaluronidase activity of BeV (Fig. 5C), while PLA2 activity wasinhibited only partially (up to about 16% of inhibition) (Fig. 5B) andthere was no inhibition of proteolytic activity (Fig. 5A), even usingmuch higher BeV: BAv ratios than necessary, according to thoseindicated on the BAv product label (Supplementary Table 1).

3.4. BAv cross-reactivity and antibody titers against BeV and BjV

Cross-reactivity between BAv and BeV and BjV was measured byagarose gel immunodifusion, in which precipitin bands wereobserved against both Bothrops venoms. A complete identitypatternwas observed for BeV and BjV, which indicates the presenceof common antigenic determinants in these venoms. When deter-mining the antibody titers of BAv against each venom, themaximum dilution of antivenom in which immunoprecipitationlines were observed were 1:16 for BeV and 1:32 for BjV, suggestinga slightly higher antibody titer against BjV. In addition, some minorprecipitin bands between practically all dilutions of BAv against BjVwere observed, while against BeV this was only observed in lowerdilutions of BAv.

Fig. 4. Inhibition of enzymatic activities of Bothrops erythromelas venom (BeV) by aqueous extract (AE) of Jatropha gossypiifolia. (A) Proteolytic activity. (B) Phospholipase A2 (PLA2)activity. (C) Hyaluronidase activity. Data shown as mean ± SEM (n ¼ 3).

Fig. 5. Inhibition of enzymatic activities of Bothrops erythromelas venom (BeV) by bothropic antivenom (BAv). (A) Proteolytic activity. (B) Phospholipase A2 (PLA2) activity. (C)Hyaluronidase activity. Data shown as mean ± SEM (n ¼ 3).

J. F�elix-Silva et al. / Toxicon 125 (2017) 74e8380

4. Discussion

B. erythromelas venom (BeV) is not included in the antigenicpool for production of bothropic antivenom (BAv) distributed inBrazil, which puts into question the efficacy of the national anti-venom against this venomous species of high incidence in North-eastern Brazil. In this context, and taking into account previousstudies in mice showing the potentiality of J. gossypiifolia aqueousleaf extract as an antiophidic agent (F�elix-Silva et al., 2014d), thisstudy aimed to evaluate the inhibitory potential of the aqueousextract (AE) of this vegetal species against local effects induced byBeV, as well as to assess whether the polyspecific bothropic anti-venom (BAv) manufactured in Brazil is effective against theseeffects.

In the present work, two treatment protocols were used forin vivo experiments. The pre-treatment protocol (treatment beforevenom injection) is relevant since it ensure that there arebioavailable compounds in circulation at the time of envenoming,thus avoiding the obtainment of false-negative results due the veryrapid onset of local effects induced by snake venom. On the otherhand, the post-treatment protocol (treatment after envenomation)aimed to simulate the practical clinical use of the products as

curative remedies for snake envenoming, since it mimics moreclosely the real practical use of antivenom and show the efficacy ofthe compounds when the poisoning is already more or lessestablished (since until the products are absorbed, the venom hasalready begun to produce its toxic effects).

A summary of inhibition percentages of AE and BAv against BeV-induced local effects in these treatment protocols is shown inTable 1. In vivo, AE was significantly active in both treatment pro-tocols, while BAv was active only in the pre-treatment protocol.These results were corroborated with the in vitro enzymatic assays,where BAv was poorly active against proteolytic and PLA2 activities,being active only against hyaluronidase activity of BeV (Fig. 5),while AE was able to inhibit all the enzymatic activities tested,reaching 100% of inhibition in most of the cases (Fig. 4).

Bothropic envenomation is characterized by the rapid devel-opment of an inflammatory process at the site of venom injection.The pathophysiology of edema formation is multifactorial,involving direct action of venom components on the microvascu-lature, increasing of the permeability of capillaries and venules, andthe effect of endogenous mediators released by venom compo-nents, such as histamine, prostaglandins, kinins and activatedproducts from the complement system (Guti�errez and Lomonte,

J. F�elix-Silva et al. / Toxicon 125 (2017) 74e83 81

1989; Guti�errez et al., 2009b). This response often causes severeedema, ischemia and neural compression, which may result in acompartmental syndrome that can lead to permanent loss of tissueor amputation of the affected limb due to necrosis (Teixeira et al.,2009).

In general, it was observed that AE and BAv presented higherinhibitory effect when used before venom than when used after.This could be expected since in pre-treatment both AE and BAvhave more time to be absorbed into circulation and thus bedistributed. In post-treatment protocol, on the other hand, only AEwas significantly active. A possible justification for this poor actionof BAv is the fact that antivenom neutralizes toxins, but cannotneutralize the action of the subsequently released endogenous in-flammatory mediators. In fact, an early study shows that local ef-fects induced by some Bothrops species are poorly neutralized bycommercial antivenoms when they were administered right beforeenvenomation by intravenous route (Picolo et al., 2002). Anotherstudy, evaluating the efficacy of bothropic antivenom on microcir-culatory effects induced by B. jararaca snake venom, showed thatthe antivenom therapy has low effectiveness against these effectsdue to impaired and delayed venom and antivenom interaction atthe site of injury due local events caused by the venom (Battellinoet al., 2003). In addition, it is important to emphasize that addi-tional experiments administering BAv by intravenous route(instead of intraperitoneal route) showed similar results, excludingthus the influence of the via of administration in the low efficacy ofBAv against BeV (data not shown).

Based on the inhibitory action of AE against PLA2 activity of BeV(Fig. 4), the action via inhibition of this class of toxins may behighlighted, since prior studies showed a significant positive cor-relation of edematogenic activity of some Bothrops venoms withPLA2 activity in vitro (Souza et al., 2015). Another possibility may bean inhibitory action upon SVMPs, since the extract also presentedanti-hemorrhagic effect (Fig. 3) and inhibited proteolytic activity ofBeV (Fig. 4). SVMPs play a relevant role in the prominent local in-flammatory response, since they induce edema, activate endoge-nous matrix metalloproteinases and are capable of releasing TNF-afrom its membrane-bound precursor (Guti�errez and Rucavado,2000). In addition, the role of the inhibition of hyaluronidase ac-tivity in the antiedematogenic activity of AE must be pointed out.Hyaluronidases, popularly known as “spreading factors”, generallyacts upon hyaluronic acid from the extracellular matrix (ECM) ofsoft connective tissue, weakening the structural integrity of ECM bydegrading hyaluronic acid and facilitating the easy diffusion of thetoxins (Kemparaju and Girish, 2006). The degraded products ofhyaluronic acid are described as damage-associated molecularpatterns (DAMPs) and are immunostimulatory and pro-inflammatory in nature, triggering several inflammatory path-ways and increasing the expression of pro-inflammatory cytokinessuch as TNF-a, IL-1b and IL-6, aggravating inflammation (Sunithaet al., 2015). BAv, in the present work, also inhibited completelythe hyaluronidase activity from BeV, but this inhibition was insuf-ficient to inhibit completely the local edematogenic effect of thisvenom, since it involves the participation of several toxins besideshyaluronidases, as discussed above.

Besides its action upon snake venom toxins, the anti-edematogenic activity of AEmay be attributed to the indirect actionof the extract upon endogenous inflammatory mediators releasedby venom toxins. Indeed, some studies have shown that anti-inflammatory and/or antioxidant drugs could inhibit the inflam-matory response against Bothrops venoms (Araújo et al., 2007;Sunitha et al., 2015). Previous studies with AE have shown signifi-cant anti-inflammatory and antioxidant activities in this plant(F�elix-Silva et al., 2014b, 2014c). In this context, the use of anti-ophidic plants as adjuvants in antivenom therapy is interesting due

to their potential ability to neutralize a broad spectrum of toxinsand greatly inhibit local tissue damage due to an indirect actiontowards endogenous mediators.

Hemorrhage is one of the most characteristic effects induced byBothrops venoms. The local hemorrhage is associated with the ac-tion of hemorrhagic SVMPs, also called hemorrhagins, which causeproteolysis of basal lamina components of microvessels, with lossof vascular wall integrity, leading to blood extravasation to skin(Guti�errez et al., 2016; Guti�errez and Lomonte, 1989). As observedin Fig. 3, AE presented very similar results in both treatment pro-tocols. BAv, as occurred in the paw edema model, was active onlywhen used before venom. The inhibitory action of AE could beexplained by its significant inhibitory potential against proteolyticactivity of BeV, which suggests an inhibitory action against SVMPs(Fig. 4). BAv, by the other hand, was not able to inhibit the in vitroproteolytic activity of BeV upon azocasein, which could justify thelow effectiveness of this antivenom against hemorrhagic activity ofBeV. However, owing to the known cross-reactivity of viperid snakevenom proteinases, this lack of neutralization was surprising. So,additional experiments using a different substrate (fibrinogen)were performed and visualized by polyacrylamide gel electropho-resis (SDS-PAGE), according to previously described (F�elix-Silvaet al., 2014d). In these experiments, it was observed some extentof neutralization of proteolytic activity only at higher levels of BAv,which indicates that BAv can neutralize, at least partially, someproteinases from BeV (data not shown).

The general low potency of BAv against BeV observed in thiswork may be related with the fact that BeV is not included in theantigenic mixture for production of this antivenom in Brazil.Despite the broad cross reactivity of the epitopes that are sharedbetween the snake venom toxins, the existence of particular com-ponents in each venom may result in an antivenom with poorneutralizing activity coverage. So, to assess how much BAv canrecognize immunologically BeV, the Ouchterlony double diffusiontest was employed.

According to our experiments, precipitin lines were observedagainst both BeV and BjV, thus indicating immunological recogni-tion of both venoms. In addition, a slightly higher antibody titeragainst BjV was suggested, since precipitin lines were still observedin a slightly higher antivenom dilution when compared to BeV.Owing the low effectiveness of BAv against BeV, this result mayindicate the presence of highly immunogenic components in BeV,which are toxicologically non-relevant for the toxic venom action,which could justify the presence of immunological recognitionwith little neutralizing potential. Interestingly, Muniz et al. (2000),while analyzing the effectiveness of BAv against some AmazonianBothrops species, observed that the immunological reactivityagainst BjV was higher than against other Bothrops venoms andthat this antivenom, in general, failed to inhibit the lethal, hemor-rhagic and PLA2 activities of Bothrops venoms from the Amazonianrain forest (Muniz et al., 2000).

Using different experimental protocols, some other works havealso shown the low efficacy of BAv against BeV. Bezerra (2000)showed that BAv was less active than a monoespecific BeV-antisera against several enzymatical and biological activities ofBeV, and presented a poor capacity to form immunocomplexes,indicating the low cross-reactivity between B. erythromelas and theother Bothrops species used for BAv production. In another work, itwas observed that BAv was less active against BeV when comparedto BjV (Castro Junior, 2008). Queiroz et al. (2008) showed by ELISAthat BAv presented only intermediate antibody titers for BeV, whencompared to the venoms used for BAv production and, interest-ingly, observed that BAv was able to neutralize significantly onlythe hyaluronidase activity of BeV (about 65% of inhibition), whileproteolytic and PLA2 activities were not.

J. F�elix-Silva et al. / Toxicon 125 (2017) 74e8382

Regarding the J. gossypiifolia plant, in previous studies con-ducted by our group, chromatographic analysis of AE showed thepresence of alkaloids, terpenes, steroids, phenolic compounds,flavonoids, tannins, amines and sugars (F�elix-Silva et al., 2014c).From these, flavonoids may be highlighted, since they seem to bethe leading compounds of the aqueous leaf extract of the plant(F�elix-Silva et al., 2014c). Studies are being carried out in our labaiming to isolate and evaluate the bioactivity of these phytocon-stituents. Regarding safety, previous results from our group showedthat AE did not present hemolytic nor cytotoxic action against HEK-293 cells, suggesting the low toxicity of this plant extract (F�elix-Silva et al., 2014c).

In conclusion, the data presented in this work indicate thatdespite the presence of some extent of immunological cross-reactivity, Brazilian polyvalent bothropic antivenom presentedpoor inhibitory potential against biological and enzymatic effectsinduced by B. erythromelas snake venom, illustrating the need fornew strategies in the production of antivenoms with broadneutralizing potential in the treatment of bothropic envenomationthroughout the country. Together, the results highlight the anti-ophidic potential of J. gossypiifolia aqueous leaf extract, suggestingthat it can be considered as a potential future adjuvant in thetreatment of bothropic envenomation local effects.

Conflict of interest

The authors have no conflict of interest to disclose.

Acknowledgements

The authors thank CAPES (23038000814/2011-83) for financialsupport. M. F. Fernandes-Pedrosa and D. V. Tambourgi are CNPqfellowship-honored researchers. J. F�elix-Silva and J. A. S. Gomesgive thanks to CAPES for the PhD Scholarship. The authors givethanks to Andrew Alastair Cumming for editing this manuscript forthe English revision.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.toxicon.2016.11.260.

References

Amaral, A., 1923. New genera and species of snakes. Proc. N. Engl. Zool. Club 8,85e105.

Araújo, S.D., De Souza, A., Nunes, F.P.B., Gonçalves, L.R.C., 2007. Effect of dexa-methasone associated with serum therapy on treatment of Bothrops jararacavenom-induced paw edema in mice. Inflamm. Res. 56, 409e413.

Battellino, C., Piazza, R., Moura da Silva, A.M., Cury, Y., Farsky, S.H.P., 2003. Assess-ment of efficacy of bothropic antivenom therapy on microcirculatory effectsinduced by Bothrops jararaca snake venom. Toxicon 41, 583e593.

Bezerra, M.M.P., 2000. Comparaç~ao da capacidade neutralizante dos antisorosbotr�opicos comercial e monoespecífico frente a peçonha de B. erythromelas.Departamento de Biofísica. Universidade Federal de Pernambuco, Recife,p. 80.

Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of micro-gram quantities of protein utilizing the principle of protein-dye binding. Anal.Biochem. 72, 248e254.

Bradley, P.P., Priebat, D.A., Christensen, R.D., Rothstein, G., 1982. Measurement ofcutaneous inflammation: estimation of neutrophil content with an enzymemarker. J. Investig. Dermatol. 78, 206e209.

Brasil, 2010. Doenças infecciosas e parasit�arias: guia de bolso, eighth ed. Minist�erioda Saúde, Departamento de Vigilancia Epidemiol�ogica, Brasília.

Brasil, 2016. Acidente por animais peçonhentos - notificaç~oes registradas no Sis-tema de Informaç~ao de Agravos de Notificaç~ao (SINAN) - SINAN NET, DATASUS(Tecnologia da Informaç~ao a Serviço do SUS. Minist�erio da Saúde, Secretaria deVigilancia Sanit�aria).

Butt, M.A., Ahmad, M., Fatima, A., Sultana, S., Zafar, M., Yaseen, G., Ashraf, M.A.,Shinwari, Z.K., Kayani, S., 2015. Ethnomedicinal uses of plants for the treatmentof snake and scorpion bite in Northern Pakistan. J. Ethnopharmacol. 168,164e181.

Castro Junior, N.C., 2008. Comparaç~ao do potencial neutralizante dos soros anti-botr�opico comercial e experimental frente �as atividades biol�ogicas dos venenosde Bothrops jararaca e Bothrops erythromelas. Programa de P�os-graduaç~aoInterunidades em Biotecnologia/USP/Instituto Butantan/IPT. Universidade deS~ao Paulo, S~ao Paulo, p. 80.

Chippaux, J.P., Williams, V., White, J., 1991. Snake venom variability: methods ofstudy, results and interpretation. Toxicon 29, 1279e1303.

Domingos, T.F., Vallim, M.A., Cavalcanti, D.N., Sanchez, E.F., Teixeira, V.L., Fuly, A.L.,2015. Effect of diterpenes isolated of the marine alga Canistrocarpus cervicornisagainst some toxic effects of the venom of the Bothrops jararaca snake. Mole-cules 20, 3515e3526.

F�elix-Silva, J., Giordani, R.B., Silva Jr., A.A., Zucolotto, S.M., Fernandes-Pedrosa, M.F.,2014a. Jatropha gossypiifolia L. (Euphorbiaceae): a review of traditional uses,phytochemistry, pharmacology, and toxicology of this medicinal plant. Evid.Based Complement. Altern. Med. 2014, 1e32.

F�elix-Silva, J., Gomes, J.A.S., Barbosa, L.M.Q., Pinheiro, I.T.M.G., Soares, L.A.L., Silva-Júnior, A.A., Zucolotto, S.M., Fernandes-Pedrosa, M.F., 2014b. Systemic and localanti-inflammatory activity of aqueous leaf extract from Jatropha gossypiifolia L.(Euphorbiaceae). Int. J. Pharm. Pharm. Sci. 6, 142e145.

F�elix-Silva, J., Souza, T., Camara, R.B.G., Cabral, B., Silva Jr., A.A., Rebecchi, I.M.M.,Zucolotto, S.M., Rocha, H.A.O., Fernandes-Pedrosa, M.F., 2014c. In vitro antico-agulant and antioxidant activities of Jatropha gossypiifolia L. (Euphorbiaceae)leaves aiming therapeutical applications. BMC Complement. Altern. Med. 14,405.

F�elix-Silva, J., Souza, T., Menezes, Y.A.S., Cabral, B., Camara, R.B.G., Silva-Junior, A.A.,Rocha, H.A.O., Rebecchi, I.M.M., Zucolotto, S.M., Fernandes-Pedrosa, M.F., 2014d.Aqueous leaf extract of Jatropha gossypiifolia L. (Euphorbiaceae) inhibits enzy-matic and biological actions of Bothrops jararaca snake venom. PLoS One 9,e104952.

Gomes, A., Das, R., Sarkhel, S., Mishra, R., Mukherjee, S., Bhattacharya, S., Gomes, A.,2010. Herbs and herbal constituents active against snake bite. Indian J. Exp. Biol.48, 865e878.

Guti�errez, J.M., Escalante, T., Rucavado, A., 2009a. Experimental pathophysiology ofsystemic alterations induced by Bothrops asper snake venom. Toxicon 54,976e987.

Guti�errez, J.M., Escalante, T., Rucavado, A., Herrera, C., 2016. Hemorrhage caused bysnake venom metalloproteinases: a journey of discovery and understanding.Toxins 8, 93.

Guti�errez, J.M., Le�on, G., Burnouf, T., 2011. Antivenoms for the treatment ofsnakebite envenomings: the road ahead. Biologicals 39, 129e142.

Guti�errez, J.M., Lomonte, B., 1989. Local tissue damage induced by Bothrops snakevenoms. A review. Mem. Inst. Butantan 51, 211e223.

Guti�errez, J.M., Rucavado, A., 2000. Snake venom metalloproteinases: their role inthe pathogenesis of local tissue damage. Biochimie 82, 841e850.

Guti�errez, J.M., Rucavado, A., Chaves, F., Díaz, C., Escalante, T., 2009b. Experimentalpathology of local tissue damage induced by Bothrops asper snake venom.Toxicon 54, 958e975.

Guti�errez, J.M., Theakston, D.G., Warrell, D.A., 2006. Confronting the neglectedproblem of snake bite envenoming: the need for a global partnership. PLoSMed. 3, 727e731.

Guti�errez, J.M., Warrell, D.A., Williams, D.J., Jensen, S., Brown, N., Calvete, J.J.,Harrison, R.A., 2013. The need for full integration of snakebite envenomingwithin a global strategy to combat the neglected tropical diseases: the wayforward. PLoS Negl. Trop. Dis. 7, e2162.

Kemparaju, K., Girish, K.S., 2006. Snake venom hyaluronidase: a therapeutic target.Cell Biochem. Funct. 24, 7e12.

Lanari, L.C., Olvera, A., Costa de Oliveira, V., Laskowicz, R.D., Boyer, L., Lago, N.R.,Alejandro, A., Roodt, A.R., 2014. Intraspecific differences in the immuno-chemical reactivity and neutralization of venom from Argentinean Bothrops(Rhinocerophis) alternatus by specific experimental antivenoms. Toxicon 85,31e45.

Lira-da-Silva, R.M., Mise, Y.F., Casais-e-Silva, L.L., Ulloa, J., Hamdan, B., Brazil, T.K.,2009. Serpentes de importancia m�edica do Nordeste do Brasil. Gaz. M�ed. Bahia79, 7e20.

Liu, Y., Staerk, D., Nielsen, M.N., Nyberg, N., J€ager, A.K., 2015. High-resolution hy-aluronidase inhibition profiling combined with HPLC-HRMS-SPE-NMR foridentification of anti-necrosis constituents in Chinese plants used to treatsnakebite. Phytochemistry 119, 62e69.

Marinetti, G.V., 1965. The action of phospholipase A on lipoproteins. Biochim.Biophys. Acta 98, 554e565.

Muniz, E.G., Maria, W.S., Estev~ao-Costa, M.I., Buhrnheim, P., Ch�avez-Ol�ortegui, C.,2000. Neutralizing potency of horse antibothropic Brazilian antivenom againstBothrops snake venoms from the Amazonian rain forest. Toxicon 38,1859e1863.

Ouchterlony, €O., 1953. Antigen-antibody reactions in gels. IV. Types of reactionsin coordinated systems of diffusion. Acta Pathol. Microbiol. Scand. 32,231e240.

Paix~ao-Cavalcante, D., Kuniyoshi, A.K., Portaro, F.C., Silva, W.D., Tambourgi, D.V.,

J. F�elix-Silva et al. / Toxicon 125 (2017) 74e83 83

2015. African adders: partial characterization of snake venoms from three Bitisspecies of medical importance and their neutralization by experimental equineantivenoms. PLoS Negl. Trop. Dis. 9, e0003419.

Perea~nez, J.A., Pati~no, A.C., Rey-Suarez, P., Nú~nez, V., Henao Casta~neda, I.C.,Rucavado, A., 2013. Glycolic acid inhibits enzymatic, hemorrhagic and edema-inducing activities of BaP1, a PeI metalloproteinase from Bothrops aspersnake venom: insights from docking and molecular modeling. Toxicon 71,41e48.

Picolo, G., Chacur, M., Guti�errez, J.M., Teixeira, C.F.P., Cury, Y., 2002. Evaluation ofantivenoms in the neutralization of hyperalgesia and edema induced byBothrops jararaca and Bothrops asper snake venoms. Braz. J. Med. Biol. Res. 35,1221e1228.

Posadas, I., Bucci, M., Roviezzo, F., Rossi, A., Parente, L., Sautebin, L., Cirino, G., 2004.Carrageenan-induced mouse paw oedema is biphasic, age-weight dependentand displays differential nitric oxide cyclooxygenase-2 expression. Br. J. Phar-macol. 142, 331e338.

Queiroz, G.P., Pessoa, L.A., Portaro, F.C.V., Furtado, M.F.D., Tambourgi, D.V., 2008.Interspecific variation in venom composition and toxicity of Brazilian snakesfrom Bothrops genus. Toxicon 52, 842e851.

Roodt, A.R., Dolab, J.A., Dokmetjian, J.C., Litwin, S., Segre, L., Vidal, J.C., 2000.

A comparison of different methods to assess the hemorrhagic activity ofBothrops venoms. Toxicon 38, 865e873.

Santhosh, M.S., Hemshekhar, M., Sunitha, K., Thushara, R.M., Jnaneshwari, S.,Kemparaju, K., Girish, K.S., 2013. Snake venom induced local toxicities: plantsecondarymetabolites as an auxiliary therapy.Mini Rev.Med. Chem.13,106e123.

Silva, H.A., Ryan, N.M., Silva, H.J., 2015. Adverse reactions to snake antivenom, andtheir prevention and treatment. Br. J. Clin. Pharmacol. 81, 446e452.

Souza, L.L., Stransky, S., Guerra-Duarte, C., Flor-Sa, A., Schneider, F.S.,Kalapothakis, E., Chavez-Olortegui, C., 2015. Determination of toxic activities inBothrops spp. snake venoms using animal-free approaches: correlation betweenin vitro versus in vivo assays. Toxicol. Sci. 147, 458e465.

Sulochana, A.K., Raveendran, D., Krishnamma, A.P., Oommen, O.V., 2015. Ethno-medicinal plants used for snake envenomation by folk traditional practitionersfrom Kallar forest region of South Western Ghats, Kerala, India. J. Intercult.Ethnopharmacol. 4, 47e51.

Sunitha, K., Hemshekhar, M., Thushara, R.M., Santhosh, M.S., Sundaram, M.S.,Kemparaju, K., Girish, K.S., 2015. Inflammation and oxidative stress in viperbite: an insight within and beyond. Toxicon 98, 89e97.

Teixeira, C., Cury, Y., Moreira, V., Picolo, G., Chaves, F., 2009. Inflammation inducedby Bothrops asper venom. Toxicon 54, 988e997.