A COMPARATIVE STUDY OF THE ANTIFUNGAL ACTIVITY OF ESSENTIAL OILS OF Varronia curassavica Jacq. OBTAINED BY DIFFERENT DISTILLATION METHODS ESTUDO COMPARATIVO DA ATIVIDADE ANTIFÚNGICA DOS ÓLEOS ESSENCIAIS DE Varronia curassavica Jacq. OBTIDOS POR DIFERENTES

This work aimed to compare the antifungal activity of the essential oil of Varronia curassavica obtained by hydrodistillation and microwave against the fungus Colletotrichum musae and verify the alterations caused by these extraction methods on the leaf surface. This study used four essential oil samples obtained by different methods, two by hydrodistillation [HD1 (1.0 L of water and 100 min.) and HD2 (2.0 L of water and 140 min.)] and two by microwave [MI1 (500W, 20 min, without water) and MI2 (700W, 40 min, with 50 mL of water added to fresh leaves)]. Essential oils concentrations of 0.05, 0.1, 0.5, 1.0, and 3.0% (v / v) were tested in PDA medium. The mycelial growth of C. musae was evaluated by measuring the diameter, every 24 hours up to 144 hours after the beginning of the incubation. Untreated leaves and leaves treated with HD1 and MI1 were prepared for observation in a scanning electron microscope (SEM) LEO EVO 40. The most abundant compounds detected in the essential oil samples analyzed by gas chromatography were: shyobunol, germacrene D-4-ol, E-caryophyllene, bicyclogermacrene, and α-cadinol. Up to 72 hours after the beginning of the incubation, C. musae presented no mycelial growth, even at the lowest essential oil concentration. Conversely, mycelial growth was detected in the control (PDA + DMSO) from 24 hours after incubation. At 144 hours after incubation, regardless of the concentration, the essential oil samples obtained by HD provided lower mycelial growth of C. musae (1.49 cm) when compared with samples obtained by MI (1.80 cm). This difference possibly occurred due to the reduction to less than half of the germacrene D-4-ol content in the samples obtained by MI. The four essential oil samples tested inhibited the mycelial growth and thus presented a inhibitory effect on C. musae. The SEM revealed more drastic changes on the surface of the leaf treated with MI than on those treated with HD. The essential oil of V. curassavica, mainly when obtained by hydrodistillation, has the potential for use in the control of C. musae.


INTRODUCTION
Commonly known as "erva-baleeira," Varronia curassavica (Jacq.) is a medicinal species originated in Brazil, used both in the folk medicine and in the pharmaceutical industry for the treatment of inflammation. Several studies have stated that the anti-inflammatory property of "erva-baleeira" is attributed to the presence of the sesquiterpene αhumulene in the essential oil contained in the leaves MEDEIROS et al., 2007;ROGÉRIO et al., 2009;PARISSOTTO et al., 2012).
Essential oils are plant metabolites, mostly consisting of a mixture of compounds, mainly mono and sesquiterpenes. However, within each species, the chemical composition may vary according to several factors, such as plant genetic characteristics, climatic conditions, soil characteristics, biotic interferences, besides the methods used at plant tissue processing and essential oil extraction (GOBBO-NETO;LOPES, 2007;VERMA et al., 2010). Among these factors, the extraction method determines the essential oil quality (TONGNUANCHAN; BENJAKUL, 2014).
If the extraction is improperly performed, the procedure may alter the chemical composition of the essential oil and alter its natural characteristics, such as color, aroma, taste, and viscosity. All these changes may result in the loss of the biological activity of this metabolite.
Although hydrodistillation is the most widely used method for the distillation of essential oils in the laboratory, new techniques have been proposed aiming mainly to reduce the extraction 1952 A comparative study… NIZIO, D. A. C. et al. time and the water and energy consumption. In this sense, microwave-assisted essential oil extraction has aroused the interest of the scientific community (SHAH; GARG, 2014) for allowing a high and faster reproducibility, simplified handling, and reduction in water and energy consumption. Nevertheless, the most appropriate extraction conditions for each species must be established. In a study carried out with V. curassavica, the authors reported that both methods, hydrodistillation and microwave, were efficient in the essential oil extraction; however, the essential oil chemical composition differed depending on the extraction method and condition (NIZIO et al., 2018b). Besides the anti-inflammatory properties of essential oil of "erva-baleeira," other activities have been reported, such as antibacterial (MATIAS et al., 2010;PINHO et al., 2012;RODRIGUES et al., 2012) and antiprotozoal (NIZIO et al., 2018a). Studies have also reported its efficiency against phytopathogenic fungi (SILVA et al., 2012;HOYOS et al., 2012;SILVA et al., 2014;NIZIO et al., 2015).
Among the most critical fungal species in Brazil, Colletotrichum musae (Berk & Curt.) von Arx. stands out for causing anthracnose, the primary disease responsible for the post-harvest deterioration of banana fruits during the transportation, storage, and commercialization stages (MORAES; ZAMBOLIM; LIMA, 2006;PESSOA et al., 2007), causing severe losses to farmers and traders (THANGAMANI et al., 2011). Despite the significant contribution of synthetic fungicides to reducing losses caused by fungal diseases, some characteristics, such as the persistence and toxicity of these products, have led to several environmental problems, affecting the human and animal health. Moreover, fungicide-resistant fungal races have emerged due to the indiscriminate use of synthetic fungicides (VERWEIJ et al., 2009;ARENDRUP et al., 2010;TATEISHI et al., 2014). Therefore, several plant species have been studied, and the essential oil of many of them have presented antifungal activity, such as Corymbia citriodora, Cymbopogon nardus (AGUIAR et al., 2014), Myrcia ovata , Myrcia lundiana ALVES et al., 2018), Lippia alba (PEIXOTO et al., 2018), Curcuma longa (HU et al., 2017), and Mentha piperita (RACHITHA et al., 2017). These metabolites are considered as a safer alternative for the control of these pathogens since they are biodegradable and have low toxicity and low risk of developing resistance in microorganisms (RUSSO et al., 2013). Also, several essential oils have been classified as "Generally Recognized as Safe (GRAS)" by the FDA.
Thus, this work aimed to compare the antifungal activity of the essential oil of "ervabaleeira" obtained by hydrodistillation and microwave against the fungus Colletotrichum musae and verify the changes that these methods cause on the leaf surface after the essential oil extraction.

MATERIAL AND METHODS
This study used four essential oil samples obtained from the V. curassavica accession VCUR-201, from the Active Germplasm Bank of Medicinal and Aromatic Plants of the Federal University of Sergipe (SISGEN Register number A8CCB3B). Two samples were obtained by hydrodistillation (HD) in a modified Clevenger apparatus under different conditions: 1.0 L of water and 100 minutes of extraction (HD1); and 2.0 L of water and 140 minutes of extraction (HD2). In both situations, 100 g of fresh leaves and 3.0 L flasks were used. Two other samples were obtained by microwave (MI) (NEOS, Milestone, Italy), using 50 g of fresh leaves, under the following conditions: 500W power level, 20 minutes of extraction, without adding water to the leaves (MI1); and 700W power level, 40 minutes of extraction, with 50 mL of water added to fresh leaves (MI2) (NIZIO et al., 2018b).
The analysis of the essential oil chemical composition was performed in a GC-MS/FID mass spectrometer (QP2010 Ultra, Shimadzu Corporation, Kyoto, Japan), coupled with an autosampler AOC-20i (Shimadzu), according to Nizio et al. (2018b). Table 1 shows the chemical composition of the essential oils used in this study. For the evaluation of the antifungal activity, a pure culture of the fungus Colletotrichum musae was obtained from the mycology collection of the Phytopathology Laboratory of the Federal University of Sergipe-UFS. The experimental design was completely randomized with three replications. Four essential oil samples were tested, two of them obtained by hydrodistillation, HD1 and HD2, and two obtained by microwave, MI1 and MI2. Essential oil concentrations of 0.05, 0.1, 0.5, 1.0, and 3.0% (v/v) were tested.
Essential oils were solubilized in 1% DMSO and homogenized in PDA culture medium (Potato Dextrose Agar, HIMEDIA). Solutions were then poured into 9.0 cm-diameter Petri dishes. A 7 mmdiameter disk containing mycelia of the fungus culture was superimposed in the center of the Petri dishes, which were then sealed, identified, and incubated in a B.O.D. chamber at 22 ± 3 ºC, with a 12-hour photoperiod. Evaluations were performed by measuring the mycelial diameter (mean of two diametrically opposite measures), using a pachymeter, every 24 hours up to 144 hours after the beginning of the incubation. Petri dishes without essential oil, containing the BDA medium and the DMSO solvent, were used as the control.
The scanning electron microscopy (SEM) was used to observe the changes in the leaves subject to the different methods of essential oil extraction. Analyses were performed at the Laboratory of Electron Microscopy and Ultra Structural Analysis, in the Department of Phytopathology of the Federal University of Lavras (UFLA), Lavras, MG. Leaves without any treatment and those treated with HD1 (1.0L of water and 100 minutes of distillation) and MI1 (500W, 20 minutes, without water) were used. Samples were fixed and prepared according to Alves et al. (2008), mounted on aluminum brackets (stubs), coated with gold (Balzers SCD 050 evaporator), and observed in a scanning electron microscope LEO EVO 40.
For the mycelial growth data, graphs were obtained from the mean diameter (mean ± standard error of the mean) of the mycelial growth for each concentration used, in function of the incubation time, using the Graph Pad Prism® software. The analysis of variance (ANOVA) was performed from the data of the 144-hour evaluation, considering a factorial scheme, with four essential oil samples and five concentrations, by the Sisvar software. Means were compared by the Tukey's test at 5% probability. Afterward, a histogram was generated using the Graph Pad Prism® software.

RESULTS AND DISCUSSION
Up to 72 hours after the beginning of the incubation, C. musae presented no mycelial growth, even at the lower concentrations. Conversely, mycelial growth was detected in the control after 24 hours of incubation (Figure 1). The four essential oil samples tested inhibited the mycelial growth and thus presented a inhibitory effect on C. musae. After 144 hours of incubation, the fungus of the control treatment reached the maximum mycelial growth and occupied the entire Petri dish (9.0 cm diameter). The analysis of variance performed for the incubation time of 144 hours revealed the absence of significant interaction between the essential oil samples and the concentrations tested. Regardless of the concentration used, samples HD1 and HD2 showed significantly lower mycelial growth (1.48 and 1.51 cm) when compared with samples MI1 (1.75 cm) and MI2 (1.84 cm) (Figure 2A). Regardless of the essential oil sample used, the concentration of 3.0% provided lower mycelial growth of C. musae (1.08 cm) when compared with the other concentrations ( Figure 2B). Although the essential oil samples tested had no fungicidal effect, the maximum mycelial growth observed after 144 hours was 2.70 cm for the treatment MI1 at the lowest concentration tested (0.05%), evidencing the existence of an inhibitory effect when compared with the control (Figure 1).  The differences in the chemical composition of the essential oil samples obtained by the two distinct extraction methods might have affected the inhibitory activity against C. musae. Among all the compounds, the highest difference was observed for germacrene D-4-ol, which was found at higher levels in essential oil samples obtained by hydrodistillation, with 11.40 and 10.36% for HD1 and HD2, respectively (Table 1). Also, the compound farnesyl acetate (2Z, 6E) was detected only in the essential oil obtained by the microwave extraction. The mycelial growth inhibition provided by the essential oil of V. curassavica can be attributed to the compounds present at higher levels, such as shyobunol, bicyclogermacrene, and germacrene D-4-ol. However, the presence of other compounds at lower levels can both potentiate the antifungal effect of the essential oil and reduce its toxicity.
Among the major compounds identified in the essential oil of "erva-baleeira" plants used in the present study, only germacrene-D-4-ol has been reported as presenting antifungal activity. The essential oil extracted from Zanthoxylum fagara fruits, containing germacrene D-4-ol as the major compound, exhibited strong activity against Colletotrichum acutatum (PRIETO et al., 2011). In Pinus nigra, the antimicrobial activity of essential oils against fungi and bacteria was attributed to the synergistic effect between some compounds and germacrene D-4-ol owing to the interaction of this compound with FtsZ binding pocket, which is an important enzyme related to cell division (ŠARAC et al., 2014). Other toxic effects of the essential oils on microorganisms may be related to the damage caused by these metabolites to the cell walls, making them permeable (OLIVEIRA et al., 2011) and causing extravasation of the cellular content either by lipids oxidation, induced by some of the essential oils compounds (MONTANARI et al., 2012) or by the inhibition of the ergosterol synthesis (KEDIA et al., 2014). The essential oil of V. curassavica caused degenerative changes in Oidium eucalypti at a concentration of 0.25% (v/v). The primary impairments observed were the lysis of the hyphal walls and the shrinking of conidia and conidiophores (SILVA et al., 2014).
When working with in natura foods, such as freshly harvested fruits, the most important thing is to inhibit the development of the disease and not necessarily to entirely eliminate the pathogen. A product that can inhibit the fungus growth, causing this process to occur more slowly, may be sufficient to preserve the fruit at the post-harvest, during the transportation, until reaching the final consumer, without presenting deterioration sings due to the disease. The absence of mycelial growth provided by the essential oil of "erva-baleeira" until three days after the beginning of the incubation shows the potential of this metabolite to control this pathogen in banana, favoring the fruit quality over its shelf life.
SEM was used to visualize the morphological changes in the leaves of "ervabaleeira" after the different extraction methods. Figure 3 shows the micrographs of untreated leaves (A), leaves treated with hydrodistillation extraction (B), and leaves treated with microwave extraction (C). In the untreated leaf, the globular glandular trichomes are intact and rounded, with no apparent damage. In the leaf subject to HD extraction, the glandular trichomes are wilted and deformed.
However, in the leaf treated with MI extraction, the trichomes are entirely disfigured. The most drastic changes observed on the surface of the leaves treated with MI when compared with those treated with HD are due to the heating process of the former. In the microwave extraction, the heat transfers from the center to the external part of the sample by the absorption of the irradiation emitted by the equipment. This absorption leads to a rapid increase in the temperature and consequently an increase in the pressure inside the gland. This phenomenon exceeds the gland's expansion capacity and causes a faster rupture and release of the essential oil than when using HD extraction (LI et al., 2012;QI et al., 2014;GAVAHIAN et al., 2015), where the heat transfers from the outside to the inside and depends on the thermal conductivity of the equipment and the sample. In the HD method, the essential oil must permeate the cuticle to be extracted (SAHRAOUI et al., 2008). Materials subject to the microwave extraction were more damaged than those subject to the HD extraction, as in the case of fruits of Schisandra chinensis (MA et al., 2012) and leaves of Dryopteris fragrans (LI et al., 2012), Cajanus cajan (QI et al., 2014), and Rosmarinus officinalis (FARHAT et al., 2017).
The present results showed that the essential oil of V. curassavica obtained by hydrodistillation provided greater mycelial growth inhibition of C. musae when compared with that obtained by microwave extraction, possibly due to the alterations that the extraction methods caused to the chemical composition of this metabolite. Nevertheless, a considerably high mycelial growth inhibition was detected even at the lowest essential oil concentration obtained by the microwave extraction, compared with the control. The present study also revealed that the microwave-assisted essential oil extraction causes more drastic changes to the leaf surface, especially in glandular trichomes, when compared with hydrodistillation extraction. The essential oil of V. curassavica presents the potential of use to control the fungus C. musae. Nevertheless, in vivo studies must be performed by applying the essential oil to banana fruits.

ACKNOWLEDGMENTS
The authors thank CNPq, FAPITEC/SE, CAPES, FINEP, and RENORBIO for their financial support for this work.