Studies have shown that lipophilic substances can often induce hemolysis because they are capable of destabilizing the lipid bilayers present in cell membranes, causing lysis of erythrocytes and increasing plasma hemoglobin levels. This effect can result in several complications: hemolytic anemia, multiple organ failure and even death  . However, many natural compounds have properties that cause the reduction in the membrane fluidity of erythrocytes, reducing hemolytic processes that lead to lower blood viscosity. This property has a potential for pharmaceutical use.
Hemolysis is a process of destruction of red blood cells (erythrocytes) in which the rupture of the plasma membrane occurs, resulting in the release of hemoglobin and causing serious damage to vital organs such as liver, kidneys and heart. Hemolysis is caused not only by chemical compounds such as penicillin, methyldopa, some types of antibiotics and anti-inflammatory agents, but also by natural compounds such as animal venoms. Several plant extracts with hemolytic activity have been described, some of which are cytotoxic or genotoxic, making it necessary to perform pharmacological and toxicological analyses of essential oils and plant extracts  .
The phospholipids constituting the membranes can be degraded by numerous substances, including natural compounds, and they can interact with different compounds, resulting in destabilization of the membranes and alteration in the flow of liquids and ions through the membranes. These substances might also act by inhibiting the action of phospholipases from various sources, both animal and human, thereby exerting anti-inflammatory activity and interfering in pro- cesses such as blood coagulation and platelet aggregation. The phospholipids constituting the membranes can be degraded by numerous substances, including natural compounds, and they can interact with different compounds, resulting in destabilization of the membranes and alteration in the flow of liquids and ions through the membranes. These substances might also act by inhibiting the action of phospholipases from various sources, both animal and human, exerting anti-inflammatory activities and interfering in processes such as blood coagulation and platelet aggregation. These processes are closely related to the action of eicosanoids generated from arachidonic acid, one of the products of the breakdown of phospholipids  .
Ophidic poisoning has been of great concern for public health, especially in tropical and neotropical countries, both because of the incidence and the action of venoms on living organisms  . However, snake venoms are mixtures of substances, mainly proteins (for example, phospholipases A2 and proteases), that have various biological activities such as enzymatic, myotoxic, cardiotoxic and cytotoxic activities  .
Phospholipases A2 (PLA2) and proteases (metalloproteases and serinoproteases), present in venoms of snakes of the Bothrops genus, can act directly on eryth- rocytes, myocytes, blood coagulation cascade factors, epithelial cells and vascular endothelium, causing severe physiological disorganization and resulting in coagulation or intravascular hemolysis or predisposing the organism to the development of diseases  . Venoms have been widely used as laboratory tools for physiological studies and for the characterization of several compounds, mainly of vegetal origin, as exemplified by plant extracts, plant drugs and essential oils. The aim of this study was to evaluate the enzymatic inhibition of phospholipases A2 and cytotoxicity to human erythrocytes using the essential oils from M. piperita, C. citratus, R. officinalis, P. boldus and F. vulgare.
2. Material and Methods
2.1. Extraction, Identification and Quantification of Essential Oils
The essential oils from M. piperita, C. citratus, R. officinalis, P. boldus and F. vulgare were extracted in the Laboratory of Essential Oils of the Department of Chemistry of the Federal University of Lavras by hydrodistillation over a 2-h period using a modified Clevenger apparatus. They were identified by gas chromatography coupled to a mass spectrometer (CG-MS) and quantified by gas chromatography coupled to a flame ionization detector (FID) according to the procedure previously described by Rezende et al. (2017)  .
2.2. Hemolytic Activity: Cytotoxicity to Human Erythrocytes
The analyses involving human blood were approved by the Human Research Ethics Committee (COEP) of the Federal University of Lavras, with registration number 48793115.0.0000.5148. The erythrocyte suspension was prepared using 10 mL of blood collected in tubes containing sodium citrate, which were centrifuged for 10 min at 4˚C and 2500 g. After centrifugation, the plasma was removed, and the erythrocytes were suspended in phosphate-buffered saline (PBS) (pH = 7.2 - 7.4) and centrifuged again under the same conditions. This procedure was repeated three times to obtain a packed red blood cell pellet.
The analysis of the hemolytic activity of the essential oils was accomplished using the methodology of Price, Wilkinson and Gentry (1982)  , with modifications. The medium was prepared with 1% agar in PBS and 0.01 M calcium chloride, 0.1 g sodium azide, and 1% blood erythrocytes. Cavities 3 mm in diameter were made in the medium after solidification in Petri dishes for the application of 0.6, 1.2 and 1.8 μL aliquots of the essential oils, and the plates remained in a cell culture chamber at 37˚C for 16 hours.
The essential oils (0.6, 1.2 and 1.8 μL aliquots) were incubated with the Bothrops jararacussu venom in a water bath at 37˚C for 30 minutes to evaluate a possible inhibitory action of the oils on the hemolysis induced by the venom. The formation of a translucent halo around the cavity in the gel is indicative of activity, and this halo was measured in millimeters for the quantification of hemolytic activity.
2.3. Activity of Phospholipase A2
The phospolipase activity was determined in a solid medium in accordance with the method described by Gutiérrez et al. (1988)  . A gel similar to that described for the determination of hemolytic activity was prepared except that 1% lecithin from egg yolk was substituted for the erythrocytes. After solidification of the medium, cavities 3 mm in diameter were prepared for appliction of the samples.
The essential oils (0.6, 1.2 and 1.8 μL) were incubated with B. jararacusu venom in a water bath at 37˚C for 30 minutes and then applied to the plates, which were maintained for 16 hours at 37˚C in a cell culture chamber. The formation of a translucent halo around the hole in the gel was indicative of activity, and this halo was measured in millimeters for the quantification of phospholipase activity.
2.4. Statistical Analysis
For the cytotoxic and phospholipase activities, the test was performed by comparing averages one at a time (each volume was separately compared to the control). The data were submitted to analysis of variance, and the means were compared by the Scott-Knott test at the 5% probability level. The statistical program used was SISVAR  .
3. Results and Discussion
According to Rezende et al. (7), the main constituents of the M. piperita, were carvone (84.34%) and limonene (10.97%). The main constituents of the essential oil from C. citratus were geraniale (47.74%), neral (35.43%) and myrcene (8.46%). The main constituents in the essential oil from R. officinalis were 1,8- cineole (62.26%), camphor (17.34%) and α-pinene (9.07%). The main components of the essential oil from P. boldus were α-terpinyl formate (61.99%), p-cymene (15.45%), 1,8-cineole (10.59%), ascaridol (2.73%), and terpinen-4-ol (2.03%), and the main component of the essential oil from F. vulgare was methyl chavicol, also known as estragole (89.48%), followed by limonene (6.15%) and fenchone (3.80%).
In Figure 1, the classes of constituents present in the essential oil from M. piperita can be observed. The oxygenated monoterpene class is predominant (86%). The essential oil from M. piperita (mint) contained carvone and limonene as the main components.
The essential oil from C. citratus contained approximately 91% monoterpenes, as can be observed in Figure 2. The essential oil from C. citratus (lemongrass) contained geranial, neral and myrcene as the main constituents.
The classes to which the constituents of the essential oil of R. officinalis belong, being 87% oxygenated monoterpenes, are presented in Figure 3. The classes of constituents found in the essential oil of P. boldus are shown in Figure 4. The essential oil of F. vulgare was the only one that presented a component belonging to the class of phenylpropanoides in its composition (Figure 5). This component represented 90% of the oil.
Figure 1. Classification of the constituents of the essential oil from Mentha piperita.
Figure 2. Classification of the constituents of the essential oil from Cymbopogon citratus.
Figure 3. Classification of the constituents of the essential oil from Rosmarinus officinalis.
Figure 4. Classification of the constituents of the essential oil from Peumus boldus.
Figure 5. Classification of the constituents of the essential oil from Foeniculum vulgare.
3.1. Hemolytic Activity: Cytotoxicity to Human Erythrocytes
With the exception of the essential oil from R. officinalis, all the oils evaluated
induced hemolysis (Figure 6); halos between 9 and 15 mm in diameter were observed for the oils from C. citratus and P. boldus. The 15 mm halo, referring to the 1.8 μL volume of oil from P. boldus, does not differ significantly from the control containing only venom.
Pre-incubation of the essential oil from P. boldus with B. jararacussu venom resulted in potentiation of hemolytic activity by approximately 30% for all the volumes of oil evaluated (Figure 7). An inhibition of approximately 45% occurred with the 0.6-μL aliquot of the essential oil from M. piperita, whereas the 1.2- and 1.8-μL volumes potentiated the action of the hemolytic enzymes present in the venom, represented mainly by proteases and phospholipases A2.
Contrary to the oil from R. officinalis, the lowest volume (0.6 μL) of M. piperita oil evaluated potentiated the lithotripsy induced by venom (20%) and inhibited this activity (35%) when the highest volumes (1.2- and 1.8-μL) were tested. For the essential oil from F. vulgare, 100% inhibition was observed with the 0.6- and 1.2-μL volumes. However, the essential oil from C. citratus did not significantly alter the hemolytic activity induced by the venom (Table 1). These differentiated activities for each essential oil can be explained by the difference in their compositions.
The different performances of the essential oils evaluated for the hemolytic activity induced by B. jararacussu venom are presented in Table 1. The data suggest the presence of specific interactions between the constituents of some oils and the hemolytic toxins present in the venom because potentiation was not observed for C. citratus oil (hemolytic in volumes of 0.6-, 1.2- and 1.8-μL) and F. vulgare (hemolytic with the 1.8 μL volume). Potentiation would be expected if there was a sum of the effect of the oils with that of the venom. In addition, significant potentiation was observed with the 0.6-μL volume of R. officinalis oil and significant inhibition with the 1.2- and 1.8-μL aliquots. These results differ from those observed with the oil from M. piperita, which caused inhibition at the lowest volume evaluated and potentiation at higher volumes.
Figure 6. Evaluation of hemolytic activity against human erythrocytes induced by essential oils from Mentha piperita, Cymbopogon citratus, Rosmarinus officinalis, Peumus boldus and Foeniculum vulgare alone and by the Bothrops jararacussu venom. *Differ from the control containing only venom by the Scott-Knott test at 5% of significance.
Table 1. Quantitative data of the effect of the essential oils of Mentha piperita, Cymbopogon citratus, Rosmarinus officinalis, Peumus boldus and Foeniculum vulgare on the hemolytic activity induced by Bothrops jararacussu venom on human erythrocytes.
*0% value represents the absence of an effect in tests in which the oils did not inhibit or potentiate the action of the venom. **Differ from the positive control (activity of the snake venom considered as 100%) at 5% of significance.
Figure 7. Effect of the essential oils from Mentha piperita, Cymbopogon citratus, Rosmarinus officinalis, Peumus boldus and Foeniculum vulgare on the hemolytic activity induced by Bothrops jararacussu venom (10 μg) in human erythrocytes after incubation of the oils with the venom at 37˚C for 30 minutes. The values obtained for the pure venom were considered to represent 100% of activity. *Differs from the positive control by the Scott-Knott test at 5% significance.
The total or partial inhibition of hemolytic activity may be related mainly to the action of phospholipases A2, which may represent the presence or absence of interactions between the molecules present in the toxins and essential oil constituents, as well as the possibility that the inhibitory action might also result from antioxidant mechanisms  . These mechanisms are closely related to the number of molecules (enzymes and active plant compounds) present in the reaction environment and justify the different actions (inhibitory, potentiating or no effect) observed for the various oil volumes analyzed.
3.2. Phospholipase A2 Activity
According to Figure 8, the action of the phospholipases A2 present in the venom was potentiated only by the 1.8-μL aliquots of the essential oils from M. piperita and F. vulgare and the 0.6-μL volume of the C. citratus oil. The smallest volume evaluated (0.6-μL) of the essential oil from R. officinalis caused a 10% inhibition of the phospholipase activity, whereas the essential oil from P. boldus did not alter the activity induced by the venom.
Some studies have described the action of plant compounds on the different classes of enzymes present in snake venoms. Silva et al. (2017)  evaluated the inhibitory potential of essential oils from Mentha viridis (L). L. and Mentha pulegium L. on phospholipase A2 present in snake venoms and observed that both essential oils were able to inhibit the degradation of phospholipids induced by Bothrops venoms. The essential oils also presented hemolytic activity, the activity of the oil from Mentha viridis (L). L. being observed only at the highest concentrations (14.6 and 29 μL∙mL−1).
Figure 8. Effect of the essential oils from Mentha piperita, Cymbopogon citratus, Rosmarinus officinalis, Peumus boldus and Foeniculum vulgare on the phospholipase activity induced by Bothrops jararacussu venom (10 μg) after incubation of the venom with the oils at 37˚C for 30 minutes. The values obtained for the pure venom were considered to represent 100% of activity. *Differs from the positive control by the Scott-Knott test at 5% significance.
Miranda et al. (2016)  investigated the inhibitory properties of the essential oils from Baccharis dracunculifolia, Conyza bonariensis, Tithonia diversifolia and Ambrosia polystachya by means of the coagulation and fibrinogenolytic activities induced by Bothrops and Lachesis snake venoms. They observed that the essential oils exhibited some therapeutic properties because they inhibited the coagulation and fibrinogenolysis induced by the poisons. The authors suggested that the topical use of the oils, in general, does not require specific pharmaceutical preparations and can be applied directly after the extraction. Many oils with antimicrobial, anti-inflammatory and curative properties are described in the literature, and these actions are of great value in the treatment of snake poisonings.
Miranda et al. (2014)  studied the effect of the essential oil from Hedychium coronarium on the fibrinogenolytic and coagulant activities induced by Bothrops and Lachesis venoms. They observed significant inhibition of the coagulation induced by both venoms, suggesting their possible use as a complementary alternative to serum therapy because the essential oils do not require specific formulations and their topical use can be performed immediately upon extraction.
Yamaguchi and Veiga-Junior (2013)  evaluated the hemolytic capacity of essential oils obtained from Endlicheria citriodora branches and leaves and observed no damage to the membrane. They also reported that both oils were composed basically of methyl geranate (monoterpenoid ester), which corresponds to 95.15% and 93.75% of the oils from the branches and leaves, respectively.
Phospholipases A2 in Bothrops snake venom induce the hydrolysis of membrane phospholipids and can generate arachidonic acid, which is a precursor of prostaglandins, thromboxanes, leukotrienes and other bioactive lipids that act mainly in inflammatory processes and in the blood coagulation cascade, thereby altering hemostasis  . The different venom toxins can be inhibited by several molecules, including chelating agents such as heparin, plasma factors of animal origin and plant extracts  .
Studies by Silva et al. (2017)  using Bothrops venom showed that both the essential oil from Mentha pulegium and that from Mentha viridis had an inhibitory effect on phospholipases A2 on the order of 4.1% at the concentration of 14.6 μL∙mL−1.
In the year 2016, Oliveira et al.  evaluated the possible interactions between vitamins and enzymes present in Bothrops atrox and Crotalus durissus terrificus venoms in vitro. Inhibition assays for proteolysis, hemolysis, coagulation, and hemagglutination were performed using different proportions of vitamins to inhibit the minimum effective dose of each venom. The authors observed that the vitamins were responsible for the 100% reduction in the cleavage of azocasein by C.d.t. poison, induced thrombolysis by the B. atrox venom, and also observed the induction of fibrinogenolysis by both poisons.
Oliveira et al. (2016)  observed possible interactions between the vitamins and the active site of the enzymes. These interactions may occur in the hydrophobic regions present in the enzymes and vitamins, as well as in the inhibitions exerted by the antioxidant mechanism.
According to reports by Borges et al. (2000)  , the aqueous extract of Casearia sylvestris inhibited the hemorrhagic activity caused by the venom of several snakes of the Bothrops genus. An aqueous extract of Mandevilla velutina was an effective inhibitor of phospholipase A2 and inhibited some of its toxic effects, such as hemorrhage  .
Carvalho et al. (2013)  reported the importance of plant species in treating snakebites, especially in places that do not have access to serotherapic treatment. Phospholipases A2, being among the main constituents of Bothrops snake venoms, can be inhibited by components of these plants, such as phenolic compounds, flavonoids, alkaloids, steroids, terpenoids (mono-, di- and triterpenes), and polyphenols (vegetable tannins).
In the present work, approximately 10% inhibition of phospholipases occurred in the presence of the oil from R. officinalis (rosemary). This oil is composed of terpenes, alcohols and ethers. This result agrees with the work reported by Mors, Nascimento and Pereira (2000)  that demonstrated the inhibitory action of several pentacyclic triterpenes, such as oleanolic acid, lupeol, ursolic acid, taraxerol, taraxasterol, α, β-amirina and friedeline, on snake venoms. Considering that phospholipase activity is only exerted by PLA2s and that hemolytic activity is exerted by both PLA2s and proteases, the observed results point to the presence of specific protease inhibitors in the evaluated oils, especially in the oils from M. piperita, R. officinalis and F. vulgare, which induced significant inhibition of the hemolytic activity exerted by the B. jararacussu venom.
The essential oil of R. officinalis inhibited almost 40% of the proteases and approximately 10% of the phospholipases A2, followed by the essential oils of M. piperita and F. vulgare that were able to inhibit protease activity, but did not inhibit of phospholipases A2. The essential oil of C. citratus induces hemolysis and potentiated the activity of phospholipases A2. The essential oil of P. boldus was induced hemolysis and bridged the action of proteases, but showed no effect on phospholipase activity. The results suggest that the essential oils studied can be used as phytotherapics in inflammation processes because they are able to inhibit the action of phospholipases A2, these enzymes being part of the inflammation cascade.
The authors express their gratitude to the Fundação de Amparo a Pesquisa do Estado de Minas Gerais (FAPEMIG), the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and the Conselho Nacional de Desenvolvi- mento Científico e Tecnológico (CNPq), for financial support and a PVNS fel- lowship.
 Ajdzanovic, V., Spasojevic, I., Pantelic, J., Sosic-Jurjevic, B., Filipovic, B., Milosevic, V. and Severs, W. (2012) Vitex agnus-castus L. Essential Oil Increases Human Erythrocyte Membrane Fluidity. Journal of Medical Biochemistry, 31, 222-227.
 Carvalho, B.M.A., Santos, J.D.L., Xavier, B.M., Almeida, J.R., Resende, L.M., Martins, W., Marcussi, S., Marangoni, S., Stábeli, R.G., Calderon, L.A., Soares, A.M., Da Silva, S.L. and Marchi-Salvador, D.P. (2013) Snake Venom PLA2s Inhibitors Isolated from Brazilian Plants: Synthetic and Natural Molecules. BioMed Research International, 2013, Article ID: 153045.
 Jiang, X., Xu, J. and Yang, Q. (2009) Soluble Expression, Purification, and Characterization of Gloydius shedaoensis Venom Gloshedobin in Escherichia coli by Using Fusion Partners. Applied Microbiology and Biotechnology, 85, 635-642.
 Rezende, D.A.C.S., Souza, R.V., Magalhães, M.L., Caetano, A.R.S., Carvalho, M.S.S., Souza, E.C., Guimarães, L.G.L., Nelson, D.L., Batista, L.R. and Cardoso, M.G. (2017) Characterization of the Biological Potential of the Essential Oils from Five Species of Medicinal Plants. American Journal of Plant Sciences, 8, 154-170.
 Oliveira, C.H.M., Simão, A.A., Trento, M.V.C., César, P.H.S. and Marcussi, S. (2016) Inhibition of Proteases and Phospholipases A2 from Bothrops atrox and Crotalus durissus terrificus Snake Venoms by Ascorbic Acid, Vitamin E, and B-Complex Vitamins. Anais da Academia Brasileira de Ciências, 88, 2005-2016.
 Silva, L.F., Cardoso, M.G., Preté, P.S.C., Teixeira, M.L., Nelson, D, L., Magalhães, M.L., Ferreira, V.R.F., Souza, R.V., Soares, L.I. and Marcussi, S. (2017) Essential Oils from Mentha viridis (L). L. and Mentha pulegium L.: Cytogenotoxic Effects on Human Cells. American Journal of Plant Sciences, 8, 1423-1437.
 Miranda, C.A.S.F., Cardoso, M.G., Marcussi, S. and Teixeira, M.L. (2016) Clotting and Fibrinogenolysis Inhibition by Essential Oils from Species of the Asteraceae Family. Brazilian Archives of Biology and Technology, 59, e16150775.
 Miranda, C.A.S.F., Cardoso, M.G., Mansanares, M.E., Gomes, M.S. and Marcussi, S. (2014) Preliminary Assessment of Hedychium coronarium Essential Oil on Fibrinogenolytic and Coagulant Activity Induced by Bothrops and Lachesis Snake Venoms. Journal of Venomous Animals and Toxins including Tropical Diseases, 20, 39-46.
 Yamaguchi, K.K.L. and Veiga-Junior, V.F. (2013) Atividades biológicas dos óleos essenciais de Endlicheria citriodora, uma lauraceae rica em geranato de metila. Química Nova, 36, 826-830.
 Trento, E.P., Garcia, O.S., Rucavado, A., França, S.C., Batalini, C., Arantesd, E.C., Giglio, J.R. and Soares, A.M. (2001) Inhibitory Properties of the Anti-Bothropic Complex from Didelphis albiventris Serum on Toxic and Pharmacological Actions of Metalloproteases and Myotoxins from Bothrops asper Venom. Biochemical Pharmacology, 62, 1521-1529.
 Borges, M.H., Soares, A.M., Rodrigues, V.M. et al. (2000) Effects of Aqueous Extract of Casearia sylvestris (Flacourtiaceae) on Actions of Snake and Bee Venoms and on Activity of Phospholipases A2. Comparative Biochemistry and Physiology B, 127, 21-30.
 Biondo, R., Pereira, A.M.S., Marcussi, S., Pereira, P.S., França, S.C. and Soares, A.M. (2003) Inhibition of Enzymatic and Pharmacological Activities of Some Snake Venoms and Toxins by Mandevilla velutina (Apocynaceae) Aqueous Extract. Biochimie, 85, 1017-1025.