In Vitro Antibacterial Activity and in Silico Analysis of the Bioactivity of Major Compounds Obtained from the Essential Oil of Virola surinamensis Warb ( Myristicaceae )

day in two seasons (rainy and dry), as well as their antimicrobial activity against pathogenic bacterial strains of Pseudomonas aeruginosa, Escherichia coli, and Staphylococcus aureus . In addition, we investigated, using in silico tools, the antimicrobial activity of the major chemical compounds present in the essential oil of V. surinamensis . The samples collected at diﬀerent seasons and times showed a similar chemical proﬁle, characterized by the major constituents α -pinene ( > 33%) and β -pinene ( > 13%). The essential oil of V. surinamensis showed an interesting antibacterial activity, exhibiting low inhibitory concentrations against the tested bacterial species. The computational investigation indicated that limonene, myrcene, and β -pinene could be related to the antibacterial activity against the tested pathogenic bacterial strains. Our results shed light on the possible constituents of essential oil that could be related to its activity against bacterial species and might be useful for further experimental tests that aim to discover new potential antibacterial agents for food preservation.


Introduction
Virola is one of the five genera of Myristicaceae and this genus is native to the American continent, especially in neotropical regions. e largest number of species is found in the Amazon and some of the most important species include Virola surinamensis Warb. and Virola sebifera Aubl. [1]. V. surinamensis is an Amazon species, which habits the floodplain and "igapós." In the Amazon region, it is popularly known by its Portuguese names Ucuúba, Sucuúba, Súcuba, and Leite-de-mucuiba [2,3]. It is described by its medicinal properties in the treatment of infections and inflammations [4,5]. e antibacterial activity of the genus Virola sp. has been investigated in some studies. Cordeiro et al. 2019 [6] obtained the oil extract from the seeds of V. surinamensis using Soxhlet and supercritical fluid extraction. ey evaluated the antimicrobial activity against Candida albicans, Staphylococcus aureus, and Escherichia coli, highlighting that the oil extract obtained by supercritical fluid extraction, which was characterized by high levels of myristic acid and lauric acid, was sensitive for S. aureus. In another study, Velasco et al. 2005 [7] showed that the aqueous extract of V. sebifera Aubl. had activity against the infection caused by the methicillin-resistant strain of S. aureus. e essential oil of V. surinamensis was reported by its antimalarial activity, which was associated with the sesquiterpene nerolidol [5,8]; antitumoral activity in colon cancer cells, which was investigated using the essential oils from bark and leaves, both composed mostly by the sesquiterpenes aristolene, β-elemene, and byciclogermacrene [2]. To the best of our knowledge, there are no studies about the use of the essential oil of V. surinamensis as an antibacterial agent. Essential oils with antimicrobial properties may have applications in the food industry, acting as preservatives. Food safety is fundamental in the food industry and is related to food preservation. ere are a variety of synthetic substances that are used as food preservatives, such as nitrates, sorbates, and benzoates. However, natural products have a great appeal to the consumer due to the negative perception associated with synthetic preservatives.
erefore, the scientific community has been looking for substances from natural products that act as food preservatives [9,10]. Since ancient times, aromatic plants with antimicrobial properties have been recognized as food preservatives and traditional medicines. Essential oils are composed of bioactive compounds which are described by their antioxidant and antimicrobial activities. us, those natural products are natural alternatives to replace or reduce the use of synthetic preservatives and they can be potentially used for food preservation, as well as for food packing [11][12][13][14]. Different studies have shown the feasibility of using essential oils in their free form or encapsulated form as food preservatives in bakery products [15,16], vegetables [17,18], ice cream [19], meat products [20,21], and in fruit preservation [22][23][24].
In the present study, we describe the chemical composition of the essential oil of the V. surinamensis leaves collected at two different periods of the day (8 a.m. and 4 p.m.) in rainy and dry seasons, and their antimicrobial activity against the American Type Culture Collection (ATCC) bacterial strains of Pseudomonas aeruginosa ATCC 27253, Escherichia coli ATCC 25922, and Staphylococcus aureus ATCC 29213. In addition, we investigated using in silico tools the potential antibacterial activities of the major chemical constituents present in the essential oil of V. surinamensis. Our results could be useful for further experimental studies that aim to develop and discover new antibacterial agents for food conservation.

Plant Materials.
e leaves of V. surinamensis were collected in the municipality of Belém (Pará state, Brazil) in the rainy season (March) and dry season (September) at 8 a.m. and 4 p.m. After collection, the samples were dried in a forced convection oven at 35°C, grounded in a blade mill, and submitted to hydrodistillation. e botanical identification was performed by comparison with an authentic specimen of V. surinamensis (MG 180285) previously deposited in the Herbarium of the Museu Paraense Emílio Goeldi (Belém, Pará state, Brazil).

Essential Oil Extraction.
Leaves of V. surinamensis obtained from rainy and dry seasons were subjected to hydrodistillation using a modified Clevenger glass extractor for 3 h. We utilized 160 g of the plant material for each experiment. e yield (%) of the essential oil was expressed as the percentage of essential oil concerning the dry matter of the leaves [25][26][27].

Chemical Analysis.
e chemical composition was analyzed with gas-phase chromatography coupled to mass spectrometry (GC/MS) using a Shimadzu QP 2010 Plus System (Shimadzu Corporation, Japan), equipped with silica capillary column (DB-5MS, length of 30 m, inner diameter of 0.25 mm, and film thickness of 0.25 μm); carrier gas: helium, linear velocity: 36.5 cm·s −1 ; type of injection: splitless (solution of 2 μL of oil in 500 μL of hexane); injector temperature: 250°C, temperature range: 60°C to 250°C, the gradient of 3°C·min −1 ; electron impact mass spectrometry: 70 eV; ion source temperature and connection parts: 220°C. e quantification of each component was performed by peak-area normalization using a flame ionization detector (GC-FID, Shimadzu, QP 2010 system) under the same conditions as GC/MS, except for the carrier gas, which was hydrogen.
e components were identified based on the retention index (RI), which was calculated using the retention times of a homologous series of n-alkanes (C8-C40, Sigma-Aldrich, USA). e pattern of fragmentation observed in the spectra was compared with existing patterns of authentic samples in data system libraries and the literature [28]. ese microorganisms were provided by the Microbiology Laboratory of the Institute Ophir Loyola, Belém, Pará, Brazil. We used ATCC bacterial strains from the Ophir Loyola Hospital that are used in the quality control of the microbiology laboratories, so they were not originated from clinical samples.

Agar Diffusion Assay.
For the preparation of the bacterial inoculum, 3 to 4 colonies of each strain isolated in nutrient agar (Kasvi, Brazil) were added in 1 mL of 0.9% physiological solution and compared with 0.5 MacFarland standard to adjust the turbidity of bacterial suspension with approximate cell count density of 1.0 x 10 8 colony forming units·mL −1 (CFU·mL −1 ). e antibacterial activity of the essential oil was performed by the disc diffusion method in agar, according to the Clinical and Laboratory Standards Institute. Using a swab, the inoculum (1.0 x 10 8 CFU·mL −1 ) was sowed on the surface of the Mueller-Hinton agar in three different directions. Sterile filter paper discs of 6 mm in diameter were impregnated with 10 μL of essential oil of V. surinamensis, placed on the surface of the inoculated Mueller-Hinton agar, and incubated for 18 hours at 37°C. e antimicrobial activity was evaluated for each microorganism by measuring the diameter of the inhibition zone (in millimeters ± standard deviation). e tests were performed in triplicate. Bacterial colonies that showed inhibition zones smaller than 8 mm were considered insensitive to the essential oil of V. surinamensis.

Minimum Inhibitory Concentration (MIC) Assay.
e minimum inhibitory concentration (MIC) was determined by the microdilution method, using the Mueller-Hinton broth and sterile 96-well microplates. e V. surinamensis essential oil was solubilized in 2% Tween 80 and successive dilutions were performed to obtain different concentrations (350, 175, 87.5, 43.75, 21.87, 10.93, 5.46, 2.73, and 1.36 µg mL −1 ). en, each dilution was added to the wells in triplicate. After that, the suspension of microorganisms was added to each well. One of the wells received the 2% Tween solution, being the test blank, to verify if the solvent interfered with bacterial growth. For the positive control, the Mueller-Hinton broth and bacterial suspension were used; the negative control consisted of the culture medium and sterile saline solution. e plates were incubated at 36°C for 24 h and posteriorly an 0.5% (v/v) aqueous solution of 2,3,5-triphenyl tetrazolium chloride-TTC (Merck, Germany) was added to each well. en, the plates were incubated at 36°C for 3 hours. e analyses were performed by colorimetric changes in each well. e red color indicates metabolic activity and bacterial growth, so it indicates that the analyzed bacterial strain is not sensitive to the essential oil. No color change indicates the absence of bacterial growth and demonstrates bacterial sensitivity to the specified concentration of the essential oil, indicating that the lowest concentration of the essential oil was able to inhibit bacterial growth.

Statistical Analyses.
Principal component analysis (PCA) was used to investigate similarities in the chemical composition between the essential oil samples extracted at different times (8 a.m. and 4 p.m.) and different seasons (rainy and dry). e analysis was performed with the chemical constituents identified and quantified by CG/MS and CG-FID, respectively, according to. Hierarchical cluster analysis (HCA) was also performed using the Euclidean distance and complete connection [29,30]. Statistical analyses were performed in the Software Statistica ® 13.1 using a correlation of the matrix type (TIBCO Software Inc., USA).

Computational Prediction of the Antibacterial Activity.
To identify potential bacterial species that are targets of the natural compounds of the V. surinamensis essential oil, we compared the structure of the major constituents of the essential oil with the structure of bioactive compounds deposited in ChEMBL [31].
e ChEMBL is a chemostructural database that contains information regarding the bioactivity of compounds belonging to different chemical classes. Initially, the structures of the major compounds of the essential oil were obtained using the Marvin Sketch program [32], and the valency and stereochemical errors were corrected using the Structure Checker program [32]. en, to perform the structural search, we used the Tar-getHunter server [33], applying a cutoff 2D similarity equal to 0.7 for each compound structure. e molecular targets were only considered in the final results if they had a value of an inhibitory activity expressed by Ki, IC 50 , ED 50 , MIC, and EC 50 . Finally, a summary table was organized with information on each natural product from essential oils, their SMILES code, ChEMBL accession codes, matched compounds, predicted molecular targets, and references relating to bioactivity.

Yield and Chemical Composition of the Essential Oils.
Our analyses demonstrated that the collection time did not influence the yield of essential oil; however, higher yields were achieved in the dry season (September), which correspond to the period of lower incidence of rainfall (Table 1). In previous studies, Anunciação et al. (2020) [2] reported yields equivalent to 1.21 ± 0.10% (bark) and 1.78 ± 0.17% (leaves) of the V. surinamensis essential oil collected in the Amazon region (Amazonas state, Brazil). Several factors can influence the yield of essential oils, such as seasonal and circadian variations [34,35]. e Brazilian amazon climate is characterized only by dry and rainy seasons and this classification is directly related to the spatial and seasonal heterogeneity of rainfall in the Amazon region [36].
Differently from the results reported in the present study, Lopes et al. (1997) [37] reported that seasonal and circadian variations did not influence the yield of the V. surinamensis essential oil, which was approximately constant (0.5%); however, they noted a strong variation in the essential oil chemical composition. Information about the influence of seasonality and circadian rhythm on the yield of V. surinamensis, or its influence on species of the Virola genus remains scarce. However, several studies have investigated the possible correlations between essential oil yield and environmental factors along with seasonal and circadian variations. For example, Ribeiro et al. (2014) [38] evaluated the influence of the collection time and seasonality on the yield of Lippia origanoides Kunth (Verbenaceae) essential oil. As in the present study, for the seasonal study, they also found better yields during the dry period. Furthermore, they pointed out that these results are useful to program the plant collection according to time, month, and chemosystematics interest. In another study, Figueiredo et al. (2018) [39] studied the seasonal influence on the  [40]. erefore, the influence of seasonal and circadian factors is particular for each species and it can vary according to several environmental factors.
ere are still few studies reporting the chemical composition of the essential oil of V. surinamensis leaves. Lopes et al. (1997) [37] evaluated the influence of seasonality and circadian rhythm on the composition of the essential oil of V. surinamensis specimen collected in the metropolitan region of Belém (Amazon region, Pará state, Brazil) in February, June, and October, at 6 a.m., 12 a.m., 6 p.m., and 9 p.m. e authors did not identify the constituents β-maaliene and δ-selinene. According to their results, the chemical composition varied throughout the seasonal and circadian rhythms. Only limonene was among the major constituents of V. surinamensis essential oil, regardless of collection time and season. In February samples, in addition to limonene (11.87-19.85%), high levels of phenylpropanoid elemicin (17.65-26.51%) were also found. In June, limonene (22.92-26.67%) and elemicin (9.19-12.02%) remained among the major constituents, accompanied by α-pinene In another study, Anunciação et al. (2020) [2] evaluated the inhibition of essential oil samples obtained from bark and leaves of V. surinamensis against HCT116 carcinoma cells using in vitro and in vivo assays. ey reported as major compounds the β-elemene (9.61 ± 1.02%), bicyclogermacrene (8.1 ± 2.0%), germacrene D (7.44 ± 1.80%), α-cubebene (5.69 ± 0.67%), and α-copaene (5.02 ± 0.77%); they indicated that some of these major compounds of V. surinamensis essential oil can be an alternative in the development of anticancer compounds to treat the colon cancer.
Multivariate analysis was used to investigate possible similarities in the chemical composition (Table 1) between the samples of V. surinamensis essential oils related to the seasonal variation and circadian rhythm. By analyzing the PCA data (Figure 2 In qualitative terms, the samples collected at different seasons and times showed a similar chemical profile. Multivariate analyses showed that the variation in the collection time in the rainy season did not influence the chemical composition of essential oils. In addition, the samples collected during the rainy season provided essential oils with a chemical composition similar to the sample extracted from the V. surinamensis leaves at 4 p.m. in the dry season. Evaluating the chemical composition of the samples collected in the dry season, we observed that the variation in the collection time caused a quantitative change in the chemical profiles of the essential oil samples analyzed in this season (Figure 2(c)). V. surinamensis essential oil presented a chemical composition composed of forty-seven chemical constituents (Table 1). Possibly, the percentage variation of sesquiterpene hydrocarbons β-maaliene and δ-selinene were some of the factors that most influenced the formation of the groups, since the percentage of these chemical constituents was approximately twice higher in the essential oil of the sample collected at 8 a.m. in the dry season. Table 2 shows the inhibition analyses against the bacterial strains using the disk-diffusion and minimum inhibitory concentration (MIC) assays with the essential oil obtained from V. surinamensis leaves collected at 8 a.m and 4 p.m in March and September.

Disk-Diffusion Assay.
In the initial screening, the essential oil samples showed antibacterial activity against E. coli and S. aureus that are above the reference values (greater than 8 mm in diameter). However, the disk-diffusion tests performed with the P. aeruginosa (Gram-negative bacteria) strain showed inhibition zone values equal to or less than 8 mm. ese results were obtained independently of the collection sample time and the seasonal period, thus indicating that essential oil showed to be inactive against this bacterial species.
To the best of our knowledge, the literature did not report similar results of the V. surinamensis essential oil against the bacterial species investigated in our study. Analyzing the antimicrobial activities against E. coli, we obtained inhibition zones of 12 mm (rainy/8 a.m.) and 14 mm (dry/8 a.m.). e samples obtained in the rainy season (4 p.m.) and dry season (4 p.m.) showed inhibition zones of diameters equal to 13 mm. Different from our results, previous studies did not identify the formation of inhibition zones when testing the V. surinamensis extract from stem bark [47] and oil from seed [6] against E. coli. e V. surinamensis essential oil tested against the S. aureus (Gram-positive) strain showed inhibition zones equal to 12.4 mm for samples from the rainy season (8 am and 4 pm) and inhibition zones equal to 13.7 mm for the dry season (8 am and 4 pm). Interestingly, Costa et al. (2008) [47] reported inhibition zones greater than 8 mm when they evaluated the antibacterial activity of V. surinamensis extract from stem bark against S. aureus; however, Cordeiro et al. (2019) [6] verified that the oil extracted from the V. surinamensis seeds were inactive against S. aureus.

Minimum Inhibitory Concentration.
e essential oil of V. surinamensis showed an interesting antibacterial activity, exhibiting low inhibitory concentrations against the tested bacterial species.
e MIC values ranged from 2.73 µg·mL −1 to 87.5 µg·mL −1 for the essential oil samples tested ( Table 2). It is worth mentioning that the blank sample (2% tween 80) did not interfere with bacterial growth. e essential oil obtained from the rainy season showed MIC values equal to 10.93 µg·mL −1 (8 a.m./rainy) and 2.73 µg·mL −1 (4 p.m./rainy) against the S. aureus strain. Regarding the dry season, the essential oil samples obtained from this period showed MIC values equal to 87.5 µg·mL −1 (8 a.m./dry) and 21.87 µg·mL −1 (4 p.m./dry). us, we noticed that the circadian and seasonal effects did not show a clear relationship with the antibacterial activity of the essential oil samples against the S. aureus strain. e essential oil samples obtained in the dry season showed MIC values equal to 21.87 µg·mL −1 (8 a.m. and 4 p.m.) against the E. coli strain. Regarding the rainy season, the essential oil samples showed MIC values equal to 87.5 µg·mL −1 (8 a.m. and 4 p.m.). us, we noted a significant difference in the antibacterial activity of the essential oil samples due to the seasonality against the E.coli strain. We noted that in the HCA plot (Figure 2(c)), majority of chemical constituents showed quantitative differences related to the collection period and it could be corroborated by the variation of the antibacterial activity of the essential oil samples. Despite the chemical profile showing quantitative variability as a function of the seasonal and circadian study, the MIC value for essential oil samples of V. surinamensis against the P. aeruginosa was equal to 87.5 µg·mL −1 , independently of the season and time of collection. It is important to highlight that the P. aeruginosa strain is a Gram-negative bacteria that forms a biofilm that could impair the permeability of substances with antibacterial activity [48][49][50][51][52]. Some essential oils and their chemical constituents can destabilize the bacterial lipid layer, promoting cell membrane degeneration and bacterial death [53,54].
e chemical composition of all collected samples of V. surinamensis essential oils was characterized by a high percentage of α-pinene (>33%) and β-pinene (>13%), which probably influenced their antibacterial activities. Studies have demonstrated that these two compounds in the isolated form show antibacterial activity against different bacterial species [55][56][57]. Furthermore, essential oils with high content of α-pinene and β-pinene have been reported by antibacterial activity against strains of P. aeruginosa [58], E. coli [59], and S. aureus [60].
However, it is important to highlight that essential oils are composed of a high variety of substances, and thus, their chemical constituents act synergically in the organisms [61,62]. e antibacterial activity could be attributed to different substances that compose the   Journal of Food Quality 7 essential oil and not only to the majority of components [63,64].

Computational Identification of Molecular Targets.
Different in silico methods have been used to identify the molecular targets and the bioactivity of chemical compounds, thus offering a low-cost and effective approach to explore their potential applications and elucidate their molecular mechanisms of action [65][66][67][68][69]. In the present study, we performed a similarity search using the Target-Hunter server to identify the bioactivity of the major constituents present in the essential oil samples of V. surinamensis. TargetHunter uses a fingerprint-based algorithm named "Targets Associated with its Most Similar Counterparts" (TAMOSIC) to explore similar compounds in ChEMBL, a chemo-structural database that contains information on compound structures and their bioactivity [33]. e β-pinene exhibited a high structural similarity with (+)-aromadendrene (ChEMBL code: CHEMBL2269083, TargetHunter score: 0.71), which is the main chemical constituent of the volatile extract of Callicarpa japonica leaves which showed antibacterial activity against different bacterial species, such as E. coli, S. aureus, and Bacillus cereus [70]. Regarding myrcene, we identified a similarity with squalene that has demonstrated inhibition against Mycobacterium tuberculosis (TargetHunter score: 0.79) [71]. Similarly, limonene showed structural matches with two diterpenes compounds containing the dolabellane skeleton obtained from the extracts of the brown alga Dilophus spiralis.
Natural products have been widely investigated for the development of new bioinspired drugs [69,73,74] and computational methods have been used as complementary and useful predictive approaches to identify interesting molecular targets of small compounds, their biological activities, and their pharmacokinetic properties [67,[75][76][77][78][79]. However, it is important to highlight that our computational results are only indicative of the antibacterial activity of these major constituents of the essential oil; thus, posterior experimental studies must be performed to confirm the inhibitory activity of these compounds against these bacterial species.
Several essential oils obtained from aromatic plants, such as clove, rosemary, ginger, thyme, and basil, have been used in food matrices due to their antimicrobial activities [11,12,23]. It is important to emphasize that after proving the antimicrobial action of essential oil, further studies are needed to ensure that the essential oil can be an effective agent in food preservation. Our in silico analyzes could shed light on the main compounds that are involved with the antibacterial activity of essential oil, thus favoring the rational choice for their isolation or synthesis to be applied in food matrices.

Conclusions
e essential oil yield was not influenced by the circadian rhythm; however, the higher yields were achieved in the dry season (September). e monoterpene hydrocarbons α-pinene and β-pinene were the major compounds identified in the essential oil samples of V. surinamensis, regardless of seasonal or circadian variations. Concerning the chemical profile of the samples of essential oils, in qualitative terms, the chemical composition was very similar. e essential oil samples of V. surinamensis showed satisfactory antibacterial activity against S. aureus and E. coli, compared with P. aeruginosa. Our computational investigation indicated that some major chemical components from the essential oil, such as limonene, myrcene, and β-pinene could be related to the antibacterial activity against the tested pathogenic bacterial strains. e essential oil of V. surinamensis extracted from the leaves collected in the rainy season showed promising results against S. aureus, indicating that this essential oil could be a potential natural source of bioactive compounds with antibacterial activity and an alternative to control the growth of microorganisms in food matrices. However, additional experiments related to food safety must be carried out.

Data Availability
e data used to support the findings of this study are available upon request to the corresponding author.

Conflicts of Interest
e authors declare that they have no conflicts of interest.