Aqueous Leaf Extract of Jatropha mollissima (Pohl) Bail Decreases Local Effects Induced by Bothropic Venom

Snakebites are a serious worldwide public health problem. In Brazil, about 90% of accidents are attributed to snakes from the Bothrops genus. The specific treatment consists of antivenom serum therapy, which has some limitations such as inability to neutralize local effects, difficult access in some regions, risk of immunological reactions, and high cost. Thus, the search for alternative therapies to treat snakebites is relevant. Jatropha mollissima (Euphorbiaceae) is a medicinal plant popularly used in folk medicine as an antiophidic remedy. Therefore, this study aims to evaluate the effect of the aqueous leaf extract from J. mollissima on local effects induced by Bothrops venoms. High Performance Liquid Chromatography with Diode Array Detection analysis and Mass Spectrometry analysis of aqueous leaf extract confirmed the presence of the flavonoids isoschaftoside, schaftoside, isoorientin, orientin, vitexin, and isovitexin. This extract, at 50–200 mg/kg doses administered by intraperitoneal route, showed significant inhibitory potential against local effects induced by Bothrops erythromelas and Bothrops jararaca snake venoms. Local skin hemorrhage, local edema, leukocyte migration, and myotoxicity were significantly inhibited by the extract. These results demonstrate that J. mollissima extract possesses inhibitory potential, especially against bothropic venoms, suggesting its potential as an adjuvant in treatment of snakebites.


Introduction
Snakebites represent a serious worldwide public health and social problem because of their high frequency, morbimortality, and sequelae left in the victims. Moreover, accidents caused by snakes are considered a neglected disease mainly in Africa, Latin America, Asia, and Oceania [1,2]. Data indicate that, worldwide, more than 5 million people suffer snakebites every year, resulting in 25,000 to 125,000 deaths and leaving approximately 400,000 people with permanent disabilities [1]. In Brazil, an estimated number of 25,000 snakebites occur per year [3]. Most of these accidents are caused by snake species of the Bothrops genus, which is responsible for about 90% of the cases in Latin America [3,4]. The main representative species of the Bothrops genus in Brazil are Bothrops jararaca (South and Southeast), Bothrops erythromelas (Northeastern), Bothrops atrox (North), and Bothrops moojeni (Center-West) [3]. The Bothrops snakes have high complexity and variation in the protein composition of their venom. This variation is due to factors such as diet, age, seasonal variation, sexual dimorphism, and geographical origin, which occurs within the species, interfamily,

Plant Material.
Leaves from Jatropha mollissima were collected in the city of "Rafael Godeiro," 6 ∘ 04 40 S; 7 ∘ 42 54 W, RN, Brazil, in January 2014. The collection of the plant material was conducted under authorization of the Brazilian Authorization and Biodiversity Information System (SISBIO) (Process number 35017) and the Brazilian Access Authorization and Dispatch Component of the Genetic Patrimony (CGEN) (Process 010844/2013-9). The botanical identification of the material was performed by Dr. Jomar Gomes Jardim and a voucher specimen was deposited at the Herbarium of the Bioscience of the Federal University of Rio Grande do Norte, Brazil (UFRN 16879). After identification and confirmation of the plant species, the leaves were dried at room temperature, triturated with an industrial blender, and stored in hermetically sealed bottles until used for aqueous extract preparation.

Snake Venom.
Lyophilized B. erythromelas and B. jararaca snake venoms were used in this work. B. erythromelas was generously supplied by the Instituto Butantan, SP, Brazil. B. jararaca was purchased from Sigma-Aldrich (St. Louis, MO, USA) (product number V5625). The scientific use of the material was approved by the Brazilian Access Authorization and Dispatch Component of Genetic Patrimony (CGEN) (Process 010844/2013-9). The venom was weighed and dissolved with phosphate buffer saline (PBS) and the protein content quantified by the Bradford method [26].

Animals.
Male and female Swiss albino mice (30-35 g), 6-8 weeks of age, used in this study were maintained under standard environmental conditions with free access to water and food. On the day of the experiment, the animals were placed in the experimental room for at least one hour prior to tests, for acclimation. All animals were euthanized by sodium thiopental overdose associated with 2% lidocaine by intraperitoneal (i.p.) route, at the end of the experiments. The experimental protocols using animals were performed in agreement with the National Council for the Control of Animal Experimentation of Brazil (CONCEA) and the International Guiding Principles for Biomedical Research Involving Animals of the Council of International Organizations of Medical Sciences (CIOMS). The animal experiments were approved by the Ethics Committee on Animal Use from the UFRN (protocol number 053/2014). The total number of animals used was 230.

Preparation of the Aqueous Extract from the Leaves of J.
mollissima. Dried leaves were submitted to decoction (10% w/v, plant : water) for 15 min at a temperature of around 100 ∘ C to obtain the aqueous leaf extract of J. mollissima (yield: 12.5% relative to dry plant). The aqueous extract obtained after vacuum filtration was freeze-dried and dissolved in PBS at adequate concentrations for the biological assays.  The mass spectrometer source parameters were set as follows: capillary voltage at 3.0 kV and end plate offset at 500 V. Nitrogen (N 2 ) was used as nebulizing (60 psi) and drying gas (10 L min −1 , 320 ∘ C). Full-scan MS and MS/MS spectra were obtained by scanning m/z from 50 to 1300. The electrospray ionization (ESI) source was operated in the positive and negative ionization mode. The data were acquired using amplitudes of 0.7 V (MS2) and 1.0 V (MS3). The data were processed through Bruker Compass Data Analysis 4.1 software (Bremen, Germany).

Inhibition of the Local Hemorrhagic
Activity. The hemorrhagic activity of B. erythromelas and B. jararaca venoms was induced using the in vivo model of local hemorrhage, as previously described in the literature with few modifications [27]. Groups of 5 animals were treated with different doses of the extract (50-200 mg/kg, i.p.). After 30 min, the animals received a subcutaneous (s.c.) injection of 25 g of both venoms (in 100 L of PBS) in the dorsal region. 3 h later, the animals were sacrificed and had the inner surface of the skin exposed. After photo documentation of the produced hemorrhagic halos, the hemorrhagic skin was removed and weighed. The group in which animals received s.c. injection of venom and i.p. treatment of PBS was used as control (venom control). Another group that received s.c. injection and i.p. treatment of PBS was used as negative control (PBS control).

Inhibition of the Edematogenic Activity.
The edematogenic activity of B. erythromelas and B. jararaca venoms was induced using the in vivo model of paw edema as previously described in the literature with few modifications [28]. Groups of 5 animals were treated with different doses of J. mollissima extract (50, 100, and 200 mg/kg, i.p.) or dexamethasone (2 mg/kg, i.p.). After 30 min, the animals received an intraplantar (i.pl.) injection of 1 g or 0.5 g of B. erythromelas or B. jararaca venoms, respectively, in 50 L of PBS, in the right hind paw. The individual right hind paw thickness was measured immediately before injection of the venoms (basal values) and at different time intervals (30,60,90, and 120 minutes) after injection of the venoms using a digital caliper (Digimess, São Paulo, SP, Brazil). A group of animals that received i.pl. injection of venoms and i.p. treatment of PBS was used as control (venom control). Another group that received i.pl. injection and i.p. treatment of PBS was used as negative control (PBS control).

Inhibition of the Cell Migration into Peritoneal Cavity.
The ability of the B. erythromelas and B. jararaca venoms to induce migration of leukocytes into the peritoneal cavity was evaluated as described in the literature with few modifications [29]. Groups of 5 animals were treated with different doses of J. mollissima extract (50, 100, and 200 mg/kg i.p.) or dexamethasone (2 mg/kg i.p.). After 30 min, the animals received by i.p. route an injection of 5 g or 2.5 g of B. erythromelas or B. jararaca venoms, respectively, in 500 L of PBS. After 6 hours (B. erythromelas venom) or 4 hours (B. jararaca venom), the animals were sacrificed and the peritoneal exudates were collected through abdominal laparotomy for total leukocyte count. To facilitate the collection, all animals received an injection of 3 mL of heparinized PBS (5 IU/mL) and had the abdomen massaged to release the adhered cells. The lavage fluid was centrifuged (at 392 g for 5 min) and the cell button resuspended in 500 L of PBS. The samples were then diluted 1 : 20 in Türk solution (acetic acid and crystal violet 1%) for total leukocyte count in Neubauer's chamber. Cytospin preparations were stained with Leishman's stain for the differential cell counts. A group of animals that received i.p. injection of venoms and i.p. treatment of PBS was used as control (venom control). Another group that received i.p. injection and i.p. treatment of PBS was used as negative control (PBS control).

Inhibition of the Myotoxic Activity.
The myotoxic activity of B. erythromelas and B. jararaca venoms was induced using the serum creatine kinase (CK) level, as previously described in the literature with few modifications [28]. Groups of 5 animals were treated with different doses of J. mollissima extract (50-200 mg/kg, i.p.) or dexamethasone (2 mg/kg, i.p.). After 30 min, all the animals received an intramuscular (i.m.) injection of 25 g of both venoms (in 50 L of PBS) in the right thigh. 3 h later, the animals were anesthetized with sodium thiopental and the blood was collected. The blood samples were incubated for 10 min at 37 ∘ C and, then, centrifuged at 10,000 g for 10 min to obtain the serum. The serum CK activity was determined using a commercial kit according to the manufacturer's protocol adapted for reading in the microplate reader (Epoch-Biotek, Winooski, VT, USA). A group in which animals received i.m. injection of venom and i.p. treatment of PBS was used as control (venom control). Another group that received s.c. injection and i.p. treatment of PBS was used as negative control (PBS control).
2.11. Statistical Analysis. All results were presented as mean ± standard error of mean (SEM). One-way ANOVA with Tukey's posttest and regression analysis were performed using GraphPad Prism version 5.00 (San Diego, CA, USA). p values less than 0.05 were considered significant.

High Performance Liquid Chromatography (HPLC-DAD)
Profile. The chromatographic fingerprint obtained by HPLC-DAD of aqueous extract of J. mollissima is depicted ( Figure 1). It is possible to observe that J. mollissima exhibits at least six major peaks (1-6). Among them, most have UV spectra similar to glycosylated flavonoid derivatives from apigenin (267 nm II band and 336 nm I band) and luteolin (253 and 267 nm II band and 349 nm I band) [30]. Glycosylated derivatives from these two flavonoids, orientin and isoorientin (derived from luteolin) and vitexin and isovitexin (derived from apigenin), have similar absorption II bands for these aglycones, differing mainly by the maximum absorption of the I band. Thereafter, by the analysis of standards and coinjection of extract + standard, it was possible to observe the increase in peak area of each standard analyzed.     [31]. Applying the systematic analyses carried out by Ferreres and coworkers [31,32], it was possible to identify the compounds as schaftoside and isoschaftoside flavonoid, respectively. According to such report, preferential fragmentation is of the sugar moiety at the 6-C rather than the 8-C position.  (Table 1). Considering the preferential fragmentation at the C-6 position, chromatographic signal 1 corresponded to the isoschaftoside flavonoid and chromatographic signal 2 corresponded to the schaftoside flavonoid. Furthermore, peaks 3, 4, 5, and 6 were confirmed as isoorientin, orientin, vitexin, and isovitexin as suggested by the coinjection performed with these standards (see Section 3.1.2). All the signals present in Table 1 are in full agreement with the previous published data. Figure 2 shows flavonoids identified by HPLC-DAD-MS/MS for the species J. mollissima.   Figure 3(a). The maximum inhibition by extract was 44% at a dose of 200 mg/kg for 3 hours. This result can be seen with the decrease in hemorrhagic halo weight and with the visual decreased halo diameter, in each dose tested. On the other hand, for B. erythromelas, it was observed that the hemorrhage caused by this venom was reduced by the extract, particularly at doses of 50 and 100 mg/kg, but this decrease was not statistically significant ( > 0.05) (Figure 3(b)).

Inhibition of the Edematogenic Activity. B. erythromelas
and B. jararaca venoms showed a marked edematogenic effect for 120 min after intraplantar injection, compared to PBS ( < 0.05) (Figure 4). The B. jararaca venom proved to be more potent in causing edema in relation to the B. erythromelas venom, since a lower dose of B. jararaca (0.5 g/paw) produced an effect similar to that of the B. erythromelas venom (1.0 g/paw). A lower dose of the B. jararaca venom was used since this venom is very hemorrhagic and we intended to evaluate edema dissociated from hemorrhage. The treatment with dexamethasone significantly reduced the edema induced by both venoms. It could be observed that the J. mollissima extract (50-200 mg/kg), administered half an hour before the injection of the B. erythromelas (Figure 4(a)) and the B. jararaca (Figure 4(b)) venoms, inhibited the edematogenic activity after 120 min of the venom injection ( < 0.05  All the results show that the animals treated with the J. mollissima extract or dexamethasone half an hour before the intraperitoneal injection of the venoms significantly inhibited the migration of these cells into the peritoneal cavity compared to the control group (which received only the venoms) ( < 0.05). For B. erythromelas, the maximum inhibition by extract was 80% at a dose of 100 mg/kg while for B. jararaca the maximum inhibition by extract was 80.18% at a dose of 200 mg/kg. Maximum inhibition for dexamethasone was 87.27% for B. erythromelas and 66.97% for B. jararaca.

Inhibition of the Myotoxic Activity.
The intramuscular injection of both venoms induced a significant increase in the CK in serum after 3 h, compared to PBS ( < 0.05) (Figures 7(a) and 7(b)). Treatment with dexamethasone reduced serum CK induced by these venoms ( < 0.05). In the same way, treatment with extract (50-200 mg/kg) showed significant reduction in the serum CK levels induced by the B. erythromelas (Figure 7(a)) and the B. jararaca (Figure 7(b)) venoms ( < 0.05). For B. erythromelas, the maximum inhibition by extract was 81.70% at a dose of 200 mg/kg while for B. jararaca the maximum inhibition by extract was 72.73% at a dose of 100 mg/kg. Maximum inhibition for dexamethasone was 92.66% for B. erythromelas and 96.10% for B. jararaca. Table 2 summarizes the maximum percentage of inhibition of the J. mollissima extract in the hemorrhagic, edematogenic, inflammatory, and myotoxic activities induced by B. erythromelas and B. jararaca venoms.

Discussion
Currently, the only available specific treatment for snakebites is the antivenom serum therapy, which has some limitations, such as reduced effectiveness against local effects, risk of immunological reactions, high cost, and difficult access in some regions [11,12]. Even though the antivenom causes the inhibition of systemic effects, the neutralization of the local tissue damage is much more difficult [4,28]. Given these limitations, it is important to find alternative treatments and/or complementary therapies. In this context, the use of medicinal plants could be highlighted, since many of them could be able to neutralize a broad spectrum of toxins (including the local tissue damage) [28,33]. Indeed, several PBS Ber Dexa * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * medicinal plants are rich sources of natural inhibitors and have pharmacologically active components. They are also able to be stable at room temperature, having easy access and low cost [34].
In this work, a phytochemical study of J. mollissima was conducted for better comprehension of the chemical compounds presented in the aqueous extract of leaves of this species. The TLC analysis identified the presence of flavonoids that could be suggested as major compounds, judging by the number, size, and intensity of spots when revealed with Reagent A Natural (specific spray reagent for this class of compounds). By the HPLC-DAD and HPLC-DAD-MS/MS analysis, the presence of six flavonoids (isoschaftoside, schaftoside, isoorientin, orientin, vitexin, and isovitexin) was confirmed. These flavonoids have been described in the literature for this species and some others of the Jatropha genus [16,[35][36][37]. These results are interesting, since several studies show that flavonoids have significant inhibitory activities against some snake venom enzymatic toxins such as PLA 2 and hyaluronidases [38,39]. These compounds could inhibit snake venom toxins directly, acting as enzyme inhibitors and chemical inactivators, or indirectly as immunomodulators, interacting with biological targets [38]. Additionally, these phenolic compounds are capable of chelating metal ions, which are essential for the activity of toxins such as SVMPs and PLA 2 [34,38]. Among the flavonoids detected in J. mollissima aqueous leaf extract, the presence of glycosylated derivatives from luteolin could  be highlighted, as these compounds have been shown to possess antiophidic properties [40,41]. Therefore, a plausible hypothesis is that the major compounds of the J. mollissima extract could be responsible for the inhibitory properties experimentally observed in this work. To confirm such assumption, the isolation of these compounds is currently underway in our group. Local hemorrhage is one of the main symptoms of bothropic envenomation [42]. The hemorrhagic SVMPs (hemorrhagins) are the main compounds responsible for this effect, causing proteolysis of basal lamina components of the microvasculature leading to a rupture of the blood vessels with the appearance of fissures, which will result in leakage of blood to the exterior [43]. Additionally, SVMPs are responsible for hydrolyzing proteins in the cell membrane, such as integrins, cadherins, collagen type IV, laminin, and fibronectin [42]. Indeed, SVMPs are the key enzymes that contribute to the toxicity of Bothrops genus, as it has been estimated that they comprise at least 30% of the total protein content [7]. SVMPs are zinc-dependent enzymes with molecular masses that range from 20 to 100 kDa and play different functions in snake envenoming such as proteolytic degradation, alterations in blood coagulation, proinflammatory activity, inhibition of platelet aggregation, and myotoxic and myonecrotic activity. The SVMPs are divided into three classes and several subclasses: PI SVMPs (no hemorrhagic activity), PII SVMPs (disintegrin domain, with proteolytic activity), and PIII SVMPs (disintegrin domain and cysteinerich domain, being the most hemorrhagic) [7,44,45]. To date, several SVMPs, from PI, PII, and PIII classes, have been purified and/or characterized in B. jararaca venom [7,44], while in B. erythromelas venom, only a very few SVMPs have already been characterized, being PI and PIII SVMPs [46]. This study revealed that the J. mollissima extract was able to inhibit the hemorrhage caused by the B. jararaca venom. In fact, through the external appearance of the hemorrhagic halos, a decrease in the hemorrhage according to the increase of the extract dose was observed (Figure 3(a)). So, this result could indicate an inhibitory action upon the SVMP action. On the other hand, the J. mollissima extract was also able to inhibit the hemorrhage caused by the B. erythromelas venoms, but this decrease was not significant (Figure 3(b)). This may be due to the different composition of SVMPs in both venoms and the possible selectivity of the extract against SVMPs from the B. jararaca venom Edema is one of the first effects caused by the bothropic envenoming. Various toxins may be responsible for edematogenic activity produced by bothropic venoms including Asp 49 or Lys 49 PLA 2 and hemorrhagic or nonhemorrhagic SVMPs [47,48]. This activity is the result of combined action of various toxins found in Bothrops venoms, acting rapidly in the connective and muscle tissue, inducing the release of various endogenous inflammatory mediators. This is the reason for the decrease in the efficacy of conventional antivenom serum therapy against these local inflammatory reactions [4,49]. In fact, this therapy is able to neutralize the toxins but cannot neutralize the effects produced by the endogenous inflammatory mediators [49,50]. Moreover, the injection of the B. jararaca venom in mice paws induces edema, which is mainly mediated by the metabolites of the arachidonic acid and the involvement, in a low level, of histamine, serotonin, and platelet-activating factor [51,52]. SVMPs also have an important role in the inflammatory response by degrading the extracellular matrix, an effect that can affect wound healing and tissue regeneration [48]. The results obtained in this study revealed that J. mollissima extract efficiently inhibited the edematogenic activity produced by B. erythromelas (Figure 4(a)) and B. jararaca (Figure 4(b)) venoms after 120 minutes. This result was similar to that produced by dexamethasone, which is a steroidal anti-inflammatory drug widely used in medical practices. Dexamethasone inhibits the PLA 2 and, consequently, there is a decrease in the production of the products derived from the arachidonic acid, which is generated by the cyclooxygenase and the lipoxygenase route [49]. Previous studies show that dexamethasone decreased the acute inflammatory response induced by the Bothrops moojeni in mice because of its ability to decrease the formation of eicosanoids in the presence of the venom [49,50]. So, in this context, two explanations could be addressed for the inhibitory effect presented by the J. mollissima extract: it could be directly inhibiting the toxins involved in the inflammatory effect produced by the venoms and/or it could be acting as a potent anti-inflammatory agent.
After the formation of the edema, the next local reaction is the recruitment of leukocytes, which selectively migrate to the site of the inflammation [53]. In the acute inflammatory response, there is a predominant accumulation of neutrophils. These cells represent the first line of defense in the body and have a phagocytic capacity for the removal of the aggressor agent. In the later stages of the inflammatory response, mononuclear cells are observed [54]. Previous studies have shown that the metalloproteinases, present in the B. asper venom, when injected into the peritoneal cavity of mice, induced an increase in the IL-1 levels, followed by an increased expression of adhesion molecules. These metalloproteases were also responsible for the activation of the complement system, resulting in an increase in the cell migration [55,56]. Bothrops venom also possesses PLA 2 toxins, which are important for the induction of the leukocyte migration, most likely by inducing more potently the release of proinflammatory mediators [57,58]. In this study, an increase in the leukocytes was observed in the peritoneal cavity after 6 hours and 4 hours induced by B. erythromelas and B. jararaca venoms, respectively. Similar to the dexamethasone, J. mollissima extract efficiently inhibited the number of total leukocytes (Figures 5(a) and 5(b)) and the number of mononuclear and polymorphonuclear cells that migrated into the peritoneal cavity induced by the B. erythromelas and B. jararaca venoms at all tested doses (Figures 6(a), 6(b), 6(c), and 6(d)). Therefore, it can be suggested that the J. mollissima extract is inhibiting the Asp 49 and/or the Lys 49 PLA 2 or that the J. mollissima extract has a potent inhibitory effect against the endogenous chemical mediators released by the action of the toxins. Another possibility may be an inhibitory action upon SVMPs, since the J. mollissima extract also presented an antihemorrhagic effect ( Figure 3).
Damage to the muscle tissue (myonecrosis) is a serious local effect of the bothropic envenomation, since it can lead to permanent loss of tissue, disability, and even amputation [59,60]. The myotoxicity may be due to a direct action of the Asp 49 or the Lys 49 PLA 2 , which directly injures skeletal muscle cells, affecting the integrity of their plasmatic membrane. The PLA 2 damages the sarcolemma, resulting in a loss of calcium permeability and, consequently, causing rupture of this membrane, leading to a rapid release of the cytosolic markers such as lactate dehydrogenase (LDH) and creatine kinase (CK) [55,61]. SVMPs can lead to myotoxicity by an indirect action, due to the ischemia caused by the vascular disorders resulting from hemorrhagic action, which can lead to muscle necrosis and the consequent release of CK [4,61]. The inflammatory reaction induced by the snake venoms contributes to further development of muscle damage [62,63]. Patrão-Neto et al. (2013) [49] demonstrated that dexamethasone decreased the late myotoxicity triggered by bothropic venom, since this compound has significant antiinflammatory properties. In this present work, J. mollissima extract was able to decrease CK levels in animals injected with B. erythromelas and B. jararaca venoms, as shown in Figures  7(a) and 7(b), respectively. This result was similar to the antiinflammatory drug dexamethasone, which possibly inhibits the inflammatory effects produced by Bothrops venoms. This result suggests that the extract possesses antimyotoxic action, possibly by inhibiting inflammation induced by the venom, since inflammation is an important finding in the local muscle damage. Furthermore, a possible inhibition of the direct action of PLA 2 could be suggested in addition to this antiinflammatory effect. Additionally, the action of the extract on the indirect myotoxic action of SVMPs could be supported since the extract also presented an antihemorrhagic effect.

Conclusions
Overall, these results demonstrate the potential of the J. mollissima extract in the treatment of the local effects produced by bothropic venoms. It could be concluded that probably the aqueous extract from the leaves of J. mollissima has substances that can inhibit or inactivate the toxins presented in the B. erythromelas and the B. jararaca venoms, as well as acting indirectly upon endogenous mediators. Therefore, the set of results provides scientific evidence of the potentiality of the J. mollissima extract. Other experimental models should be tested by our research group in order to suggest the usefulness of this plant as a future potential adjuvant in the treatment of local effects due to snakebites, along with antivenom therapy.