Antimicrobial Activity of Essential Oil of Baccharis dracunculifolia DC (Asteraceae) Aerial Parts at Flowering Period

Baccharis dracunculifolia DC (Asteraceae) is a Brazilian native bush tree, and its leaf essential oil has been reported to possess some biological activities, but the antimicrobial activity of its aerial part essential oil at the flowering period is unknown or little studied, mainly against agents that cause foodborne diseases. Thus, this study aimed to determine the chemical composition and evaluate the antimicrobial activity of the essential oil of B. dracunculifolia aerial part at flowering period. This essential oil was obtained by hydro distillation and its chemical composition was determined by gas chromatography coupled with mass spectrometry (GC–MS). The minimum inhibitory concentration, minimum bactericidal concentration, and minimum fungicidal concentration of the essential oil were evaluated against eight bacteria and eight fungi using 96-well microtiter plates. The essential oil yield was 1.8 ± 0.07%, and spathulenol (27%) and trans-nerolidol (23%), both oxygenated sesquiterpenes, were the major compounds found among 30 chemical constituents identified. The essential oil presented bacteriostatic and bactericidal activities, mainly against Staphylococcus aureus, Bacillus cereus and Pseudomonas aeruginosa, and also fungistatic and fungicidal activities. However, its antibacterial activity was more effective than the antifungal one by using the essential oil at lower concentrations. Essential oil of B. dracunculifolia may be a potential alternative for food applications in order to reduce synthetic chemicals in a more sustainable food industry.


INTRODUCTION
Baccharis dracunculifolia DC (Asteraceae), known as "alecrim-do-campo" (rosemary-of-the-field) and "vassourinha" (small broom) in Brazil (Barroso, 1976), a perennial woody bush tree that can reach up to around 2 m of height with alternate spike lanceolate leaves (Borges and Forzza, 2008), 1 to 1.5-mm long achene (Barroso, 1976), dioicous with male and female inflorescences in separate plants (Park et al., 2004), is native to biomes of Brazilian Cerrado, Atlantic Rainforest and Pampas. It is reported as the main source of propolis and bee honey production, followed by Araucaria angustifolia and Eucalyptus citriodora in Brazil (Sforcin et al., 2012).
Despite of some reports on the antimicrobial activity of B. dracunculifolia, mainly related to its antifungal activity, the antimicrobial activity spectrum of B. dracunculifolia essential oil in the flowering phenophase and against agents causing foodborne diseases is unknown or little studied. Food spoilage is a serious widely neglected problem, mainly by mycotoxins in grain storage due to poor harvesting practices, inappropriate drying, handling, packaging, storage, and transport conditions (Bhat et al., 2010). The utilization of synthetic chemicals has increased microorganism resistance, mostly against antibiotics, and the search for alternative agents to control microorganisms is therefore necessary (Rossolini et al., 2014;Ventola, 2015). An alternative to reduce synthetic chemicals is the search for antimicrobials from medicinal plants (Anyanwu and Okoye, 2017) to extend shelf life and combat foodborne pathogens. Thus, the aim of this study was to determine the chemical composition and evaluate the antimicrobial activity of B. dracunculifolia aerial part essential oil for potential applications to preserve food by non-synthetic compounds.

Essential Oil Extraction
Baccharis dracunculifolia aerial parts were air dried under the shaded for 1 week. The essential oil was extracted from 200 g of dried aerial parts in 2 L of distilled water by hydrodistillation in a modified Clevenger apparatus for 2 h (Miranda et al., 2016) and stored at 4 • C (Pereira et al., 2016).
The essential oil yield (%) was calculated by mass (g) of essential oil per mass (g) of the dried aerial parts of the plant. The essential oil absolute density was determined in graduated capillaries (5.0 µL) and calculated by mass (g) per volume (mL) at 20 • C. The refraction index was determined using an Abbe refractometer (RL3 model) which was calibrated with distilled water (refraction index of 1.3330) at 20 • C (Brasil, 1988).

Chemical Characterization
The essential oil chemical identification was carried out by GC-MS (Agilent 19091J-433) equipped with an HP-5MS capillary column (30 m × 0.25 mm × 0.25 µm), with initial temperature from 40 • C (2 min) to 230 • C (3 • C/min), and kept at this temperature for 20 min. Helium was utilized as the carrier gas at the linear speed of 1 mL/min up to 300 • C, and pressure release of 56 kPa. The injector temperature was 250 • C; the injection volume was 1 µL; the injection occurred in split mode (20:1). The temperatures of the transfer line, ion source, and quadrupole were 285, 230, and 150 • C, respectively. Mass spectrometry was obtained with a scan range of 40 to 550 m/z with solvent delay of 3 min, compounds were identified based on comparison of their retention indices obtained by a homologous series of n-alkane standard (C7-C28), and electron ionization mass spectra were compared with the Wiley 275 library spectra (Adams, 2012).

Antibacterial Activity
The antibacterial activity of B. dracunculifolia essential oil was tested against eight bacterium species: Gram-positive  Tsukatani et al., 2012) utilizing 96-well microtiter plates to determine the minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC). The bacterial suspensions were adjusted with sterile saline solution until the concentration of 1.0 × 10 5 CFU/mL. The inoculum was prepared daily and stored at 4 • C until its utilization. The inoculum was cultivated in solid medium to verify the absence of contaminations, and for validation. B. dracunculifolia essential oil was dissolved in a 5% dimethyl sulfoxide solution (Merck KGaA, Germany) containing 0.1% of polysorbate-80 (1 mg/mL) and added to Luria-Bertani (100 µL) medium with bacterial inoculum (1.0 × 10 4 CFU/well) to reach the desired concentrations. The microplates were incubated in a rotary agitator (160 rpm), for 24 h, at 37 • C. The lowest concentration without visible of the microbial biomass growth under the optical microscope were defined as the concentrations that completely inhibited bacterial growth.
Minimum bactericidal concentration was determined by 2 µL serial sub cultivation in microtiter plates containing 100 µL of broth per well and incubation during 24 h. The lowest concentration without visible microbial biomass growth under optical microscope was defined as MBC, indicating the death of 99.5% of the original inoculum. The optical density for each well was measured in a 655 nm wavelength with a Microplate Manager 4.0 (Bio-Rad Laboratories) and compared to a blank one (broth medium with diluted essential oil) and positive control. Streptomycin (Sigma P7794) and ampicillin (Panfarma, Belgrade, Serbia) were utilized as positive controls (1 mg/mL in sterile saline solution). A solution of 5% dimethyl sulfoxide was utilized as negative control.

Antifungal Activity
The antifungal activity of B. dracunculifolia essential oil was tested against eight fungi: Aspergillus fumigatus The fungi were kept in malt extract agar, and the cultures were stored at 4 • C and subcultivated once a month (Booth, 1971). A modified microdilution technique was utilized to investigate the antifungal activity (Hänel and Raether, 1988;Espinel-Ingroff, 2001). The fungal spores were washed with sterile saline solution at 0.85% containing polysorbate-80 (0.1%). The spore suspension was adjusted with sterile saline solution to a concentration of 1.0 × 10 5 in a final volume of 100 µL per well. The inoculums were stored at 4 • C for posterior utilization. The inoculum dilutions were cultivated in malt extract agar to verify the absence of contamination and validate each inoculum.
Minimum inhibitory concentration was determined by serial dilution technique using 96-well microtiter plates. The essential oil was dissolved in 5% dimethyl sulfoxide solution (Merck KGaA, Germany), containing 0.1% of polysorbate-80 (1 mg/mL), and added to a malt extract cultivation medium with inoculum. The microplates were incubated in a rotary agitator (160 rpm) for 72 h at 28 • C. The lowest concentrations without visible microbial biomass growth under optical microscope were defined as the concentrations that completely inhibited fungal growth.
The minimum fungicidal concentration (MFC) was determined by a 2 µL serial sub cultivation of the tested compound dissolved in a cultivation medium, and inoculated during 72 h in microtiter plates containing 100 µL of broth per well and with incubation for 72 h at 28 • C. The lowest concentration without visible biomass concentration was defined as MFC indicating the death of 99.5% of the original. The commercial fungicides bifonazole (Srbolek, Belgrade, Serbia) and ketoconazole (Zorkapharma, Šabac, Serbia) were used as positive controls (1-3500 µg/mL).

Statistical Analysis
The antimicrobial tests were carried out in duplicate and replicated three times. The results were expressed in values
Trans-nerolidol (3,7,11-trimethyl-1,6,10-dodecatrien-3-ol) is the second major compound of our study, and the mass spectrum was obtained by GC-MS (Figure 2). It is also known as peruviol, and is a naturally occurring sesquiterpene alcohol found in essential oil of several plants with floral odor. This compound has four different isomeric forms, which consist of two enantiomers and two geometric isomers: cis-and trans-nerolidol (Chan et al., 2016).
MIC values of the essential oil for bacteria varied from 0.50 to 12.65 mg/mL and the controls streptomycin and ampicillin ranged from 0.04 to 0.75 mg/mL ( Table 2). MBC values of the essential oil varied from 1.50 to 16.87 mg/mL and for the controls streptomycin and ampicillin from 0.10 to 1.20 mg/mL ( Table 2). S. aureus, B. cereus and P. aeruginosa were the most susceptible species to essential oil with MIC of 0.5, 1.1, and 1.05 mg/mL, respectively, and MBC of 2.1, 1.5, and 2.1 mg/mL, respectively. The most resistant species were L. monocytogenes and S. enterica with MIC and MBC values of 12.65 and 16.87 mg/mL, respectively, for both. These values varied from 32-to 63-fold higher than MIC and from 22-to 56-fold higher than MBC of controls.
Minimum inhibitory concentration values of the essential oil against fungi ranged from 8.43 to 16.87 mg/mL, and the controls   bifonazole from 0.10 to 0.20 mg/mL and ketoconazole from 0.20 to 2.5 mg/mL ( Table 3). MFC values of the essential oil varied from 16.87 to 33.60 mg/mL and the controls bifonazole from 0.20 to 0.25 mg/mL and ketoconazole from 0.30 to 3.5 mg/mL (  Table 3). MIC and MFC values for the essential oil varied for MIC from 3-to 169-fold higher and MFC from 5-to 168-fold higher than the controls.

DISCUSSION
The physical and chemical characteristics and the yield of B. dracunculifolia essential oil obtained in our study were different from the ones reported by Fabiane et al. (2008). For these authors, B. dracunculifolia leaf essential oil had yield of 1.5%, density of 91 g/mL, and refraction index of 1.4593. The differences may be related to the utilized plant part in the extraction of the essential oil. In our study, aerial parts were used and harvested during the flowering period when there is a natural increase in the production of the plant essential oil to attract pollinators. In the report by Fabiane et al. (2008) only leaves harvested during the vegetative phase of the plant were utilized. The plant phenophase is a determining factor in the yield, chemical composition, physical and chemical characteristics of the essential oil (de . Different major compounds of B. dracunculifolia essential oil have been reported (Table 4). Frizzo et al. (2008) identified seventy components in B. dracunculifolia aerial part essential oil from Brazil, Uruguay and Bolivia. The identified components accounted for 91.4 to 97.2% of the total essential oil compositions. For these authors, B. dracunculifolia essential oils from different origins were quite different in quantitative and qualitative composition. The Brazilian essential oils were characterized by high contents of (E)-nerolidol, whereas the Uruguayan essential oils showed a predominance of viridiflorol with absence of (E)nerolidol, and the Bolivian essential oils had higher contents of γ-cadinene, δ-cadinene, t-cadinol, and α-cadinol. Spathulenol was also identified in most B. dracunculifolia essential oils ( Table 4).
The differences found in B. dracunculifolia accesses make the importance of the essential oil chemical characterization in a bioassay evident. The chemical composition of the plant essential oil depends on the species, climate conditions, soil type, harvesting seasons, age of leaves, geographic region, and utilized extraction process. Seasonal variation is one of the main factors that affect the composition of essential oils (Chan et al., 2016). A study conducted by de Sousa et al. (2009) showed that the concentration of trans-nerolidol in B. dracunculifolia leaves was fivefold higher in March, 2005 (136.53 mg/100 g of plant) than that in July, 2004 (25.03 mg/100 g of plant).
Our studies showed that B. dracunculifolia essential oil presents bactericidal activity, mainly against S. aureus, B. cereus and P. aeruginosa, important pathogenic agents. S. aureus is a pathogenic gram-positive bacterium with high infection and mortality rates (Kong et al., 2016). P. aeruginosa is a pathogenic Gram-negative agent that is resistant to several medications and cause infections in the bloodstream (Piao and Yuan, 2017). B. cereus is a Gram-positive bacterium that causes gastrointestinal infections and that may result in osteomyelitis, meningitis, and pneumonia (Malik-Tabassum et al., 2017), inflammation and eyesight loss (Callegan et al., 2017).
Minimum inhibitory concentration and MBC values found in our study were higher for the essential oils than for the controls. However, Van Vuuren (2008), in a review of over 500 publications detailing the antimicrobial activity of plants, proposed that essential oils with MIC of 2 mg/mL or lower or natural products with MIC values below 1 mg/mL could be considered noteworthy; however, in extracts, they must have MIC values below 8 mg/mL to be considered. Therefore, B. dracunculifolia essential oil of our study with MIC lower than 2 mg/mL (S. aureus, B. cereus, and P. aeruginosa) is considered noteworthy to control microorganisms.
The main compounds of B. dracunculifolia essential oil found in our study were oxygenated sesquiterpenes such as spathulenol and nerolidol that have high hydrophobicity, thereby allowing easier penetration across the plasma membrane and interaction with intracellular proteins and/or intra-organelle sites (Chan et al., 2016). The surface polarity of spathulenol and nerolidol is 20.2 Å 2 and these compounds have one hydrogen bond donor and one acceptor (National Center for Biotechnology Information [NCBI]. PubChem Compound Database, 2018). Pajouhesh and Lenz (2005) reported that drugs with a PSA of 60 Å 2 or less are completely absorbed by the cell, whereas those with at least 140 Å 2 are not. Therefore, spathulenol as well as nerolidol are able to interact and pass through the cellular membrane.
For sesquiterpenes such as spathulenol and nerolidol, the antimicrobial action via cell membrane-disrupting mechanism is described and, hence, results in the leakage of K + ions from bacterial cells (Inoue et al., 2004). According to Togashi et al. (2007) terpene alcohols with carbon chains of C10 to C12, as nerolidol, exhibit a strong antibacterial activity against S. aureus. As reported by Chan et al. (2016) besides causing membrane disruption, nerolidol can be responsible for the downregulation of α-hemolysin gene hla expression in S. aureus determined via quantitative real-time PCR analyses (Lee et al., 2014). Chan et al. (2016) suggested that nerolidol is a therapeutic option to the development of drug combinations for antibacterial treatment, particularly against S. aureus and for multi-drug resistant bacteria.
The essential oil presented fungistatic and fungicidal action against Aspergillus, Penicillium, and Trichoderma genera. Aspergillus spp. are related to allergic reactions, respiratory problems and pulmonary infections (Yu et al., 2016). A. ochraceus is related to ochratoxin A production with pathogenic effects in animals and possible human carcinogen (Malir et al., 2016). T. viride causes adverse effects to health including respiratory problems (Larsen et al., 1996). P. verrucosum commonly grows on stored cereals (Bui-Klimke and Wu, 2015) and may produce ochratoxin A (Kuiper-Goodman and Scott, 1989) and citrinin (nephrotoxic) (Larsen et al., 2001). Thus, the identification of alternative compounds to control this fungus is relevant to manage damages caused by these fungi for food preservation, mainly with the increase in resistance to conventional chemical products. Koul et al. (2008) warned that the uses of synthetic fungicidesin the post-harvest treatments of vegetables -could develop resistant fungal strains and indicated bio-based essential oils as fungicides. Sivakumar and Bautista-Baños (2014) recommend the use of essential oils to preserve food and Burt (2004) recommend concentrations from 0.1 to 6%. The fungicidal concentrations of B. dracunculifolia essential oil reported in our study range from 16.87 to 33.60 mg/mL (equivalent to 1.6 to 3.3%, m/v, respectively). Thus, the values found in our study are within the concentration of essential oils used for food preservation (Sivakumar and Bautista-Baños, 2014), which makes the essential oil from our study an alternative to develop applications in food preservation.
The utilization of essential oils to preserve food and as flavoring agents is considered to be Generally Recognized as Safe (GRAS) (Burt, 2004;Sivakumar and Bautista-Baños, 2014). In addition, the essential oil compounds such as carvacrol, carvone, cinnamaldehyde, citral, p-cymene, eugenol, limonene, menthol, and thymol are indicated with no risk to the consumer's health. Also, there is no restriction for spathulenol and nerolidol, the major compounds of B. dracunculifolia essential oil described in our study.
One of the challenges to use essential oils to preserve food is the high cost that can be up to six times higher than chemical fungicides (Kouassi et al., 2012). However, Burt (2004) reported that the increase in the demand of essential oils can result in bioengineering of their synthesis in plants. Also, large scale production could reduce production cost. According to the European Pharmacopoeia, to develop the applications of essential oils, a minimum of 2 mL/kg of plants at flowering shoots is required (Nemeth and Bernath, 2008). Our study presents high yield (17 mL/kg dry plant) of B. dracunculifolia essential oil, increasing its potential utilization to control microorganisms.

CONCLUSION
The aerial parts of B. dracunculifolia present high essential oil yield (17 mL/kg dry plant) with 30 components, and the major ones are spathulenol and nerolidol, followed by heptacosane, β-pinene, bicyclogermacrene, (E)-caryophyllene and germacrene D. The essential oil presents greater action against bacteria than against fungi, mainly against S. aureus, B. cereus, and P. aeruginosa. Essential oil of B. dracunculifolia may be a potential alternative to food applications in order to reduce synthetic chemicals in a more sustainable food industry.

AUTHOR CONTRIBUTIONS
GL, NC, and ZG conceived of the presented study. GL and ZG developed the theory and verified the analytical methods. LC obtained the essential oil and produced the fractions. JGo determined the chemical composition of the oil and fraction. JGl and MS investigated antimicrobial activities and with NC interpreted the results in the discussion section about antimicrobial activity. JGl, MS, and JGo contributed to the interpretation of the results. LC wrote the manuscript with support of GL, NC, and ZG. The final version of the manuscript was discussed, rewritten, and approved by all authors before submitting to the Journal.
-finance code 001-, Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) for the financial support and the fellowship and the Ministry of Education, Science and Technological Development of Serbia (Grant No. OI 173032).