Chemical Composition and Antioxidant Activity of Essential Oils from Leaves of Two Specimens of Eugenia florida DC.

Eugenia florida DC. belongs to the Myrtaceae family, which is present in almost all of Brazil. This species is popularly known as pitanga-preta or guamirim and is used in folk medicine to treat gastrointestinal problems. In this study, two specimens of Eugenia florida (Efl) were collected in different areas of the same region. Specimen A (EflA) was collected in an area of secondary forest (capoeira), while specimen B (EflB) was collected in a floodplain area. The essential oils (EOs) were extracted from both specimens of Eugenia florida by means of hydrodistillation. Gas chromatography coupled to mass spectrometry (GC/MS) was used to identify the volatile compounds present, and the antioxidant capacity of the EOs was determined by antioxidant capacity (AC-DPPH) and the Trolox equivalent antioxidant (TEAC) assay. For E. florida, limonene (11.98%), spathulenol (10.94%) and α-pinene (5.21%) were identified as the main compounds of the EO extracted from sample A, while sample B comprised selina-3,11-dien-6α-ol (12.03%), eremoligenol (11.0%) and γ-elemene (10.70%). This difference in chemical composition impacted the antioxidant activity of the EOs between the studied samples, especially in sample B of E. florida. This study is the first to report on the antioxidant activity of Eugenia florida DC. essential oils.


Introduction
In the secondary metabolism of plants, there is the production and accumulation of compounds of different chemical natures [1], and these chemical compounds, called secondary metabolites, are important for the plant's ability to defend itself against pathogens, predators and environmental stress [2]. The interactions that secondary metabolites exert in plants have awakened great interest in studies, mainly due to the antioxidant properties that some plants have [3].
The search for natural antioxidants, particularly originating from plants, has increased significantly, which may be related to the presence of some compounds of the secondary metabolism of plants that can exhibit oxidizing activity, and, in a way, contribute to the combat and inhibition of free radicals [4], which are involved in the physiopathogenesis of found in studies with other species of the genus, such as E. uniflora (0.22-1.68%) [26] and (0.8-3.1%) [27].

Chemical Composition of the Essential Oils
The essential oils of the specimens were obtained by the hydrodistillation technique. Ninety-six chemical components were identified (Table 1), with a total of 93.09% in the EflA specimen and 88.04% in the EflB specimen. The EflA specimen had hydrocarbon (24.17%) and oxygenated (7.72%) monoterpenes, hydrocarbon (18.79%) and oxygenated (41.64%) sesquiterpenes. The EflB specimen showed hydrocarbon (44.06%) and oxygenated (43.98%) sesquiterpenes. The essential oil from the EflA specimen had limonene (11.98%) as the major component, which was absent in the EflB specimen. This substance is reported in the literature as having antibacterial and antifungal activity against foodborne pathogens [29,30]. In the food industry, for example, it can be used as an inhibitor of yeast growth during the fermentation process [31], and, in addition, this component also presents anti-inflammatory, antioxidant and anticancer activities [32]. Another major component found in the essential oil of the EflA specimen was the oxygenated sesquiterpene spathulenol (10.94%), which was also observed in the EflB specimen (1.2%). It is important to highlight that this substance has antioxidant, anti-inflammatory and antimicrobial activities [33], and insecticidal [34] and antinociceptive [35] activities. The monoterpene α-pinene (5.21%) is present in the EflA oil, but this substance was not identified in the EflB specimen. This monoterpene exists in nature and has (−)-α-pinene and (+)-α-pinene structural enantiomers [36]. In addition, these compounds have demonstrated biological activities, such as being antimicrobial, and are cytotoxic against the cancer cells which cause ovarian cancer [37,38].
The sesquiterpenes caryophyllene oxide (5.0%) and (E)-caryophyllene (4.49%) were also identified in the essential oil of the EflA specimen. The (E)-caryophyllene, at a lower content (2.35%), was found in the essential oil of the EflB specimen. Studies report that caryophyllene oxide has insecticidal activity against the Aedes aegypt vector, an important vector of diseases such as dengue, zika and chikungunya [39]. This compound also presents gastroprotective potential [40] and antiviral potential [41], as well as potential activity against leishmania [42].
The sesquiterpenes selina-3,11-dien-6α-ol (12.03%), eremoligenol (11.0%) and γ-elemene (10.70%) were the main constituents of essential oil EflB. Eremoligenol and γ-elemene were also found in the EflA specimen, but in low concentrations, with contents of 0.71% and 0.16%, respectively. γ-elemene is a sesquiterpene that proves to be toxic to some pest crops, and may be an alternative for the development of new pesticides [43] and insecticides [34,44]. Another compound identified in the essential oils of the specimens was δ-cadinene, with levels of 4.42% for EflB and 3.13% for EflA. It is important to highlight this compound for its acaricide activity [45], antimicrobial activity and as a causative agent of respiratory tract infections, such as pneumonia and sinusitis [46]. The α-cadinol was also found in the essential oils of the specimens, with contents of 4.31% for EflA and of 3.98% for EflB. This oxygenated sesquiterpene has antifungal [47] and cytotoxicity activities against some cancer cell lines [48].
There are few reports of the chemical composition of E. florida essential oils in the literature, but this study showed that the chemical composition of the studied specimens differed from each other, in which the specimen (EflB) had a high content of both hydrocarbon and oxygenated sesquiterpenes compared to the specimen (EflA), while the specimen (EflA) showed a high content of monoterpene hydrocarbons. The chemical profile of the studied specimens was different from that found in E. florida essential oils by Apel et al. [23]. The variability in the chemical profile of essential oils can be explained by different aspects, such as extraction techniques, climatic and geographical factors, type of soil, light, and temperature [49].

Antioxidant Activity
To measure the antioxidant activity of Eugenia florida essential oils, preformed free radicals DPPH • and ABTS •+ were used. Table 2 shows the ability of essential oils, which have been extracted from the dried leaves of Eugenia florida EflA and EflB specimens, to scavenge free radicals. According to the results, the TEAC of EflA and EflB specimens were 0.456 mM and 0.652 mM, respectively. When compared to the 1 mM concentration of Trolox, EflA specimen showed 45% inhibition of the ABTS •+ radical, and EflB 65%. Both activities were below the standard. Additionally, EflA and EflB specimens presented CA-DPPH • of 1.72 mM and 2.14 mM, respectively. According to these results, the DPPH • radical inhibition capacity of the EflA and the EflB specimen was 72% and 114% higher than the Trolox standard (1 mM), respectively. Based on these data, we observed that the results of the measurement of antioxidant capacity using the ABTS •+ radical scavenging capacity test were different from those obtained with the DPPH • radical. According to other studies, there are differences between the results obtained by DPPH • and ABTS •+ , resulting from the difference in reaction mechanisms that each of these free radicals presents against the antioxidant molecules presented in the samples [54,55]. In analyses using ABTS •+ , electron transfer can occur, and different antioxidant compounds provide electrons to reduce the radical cation, and despite the antioxidant compounds' potential, these compounds have time to fully react, allowing a measurement of the total antioxidant capacity. As for the DPPH • radical, the inhibition is based on the transfer reaction of hydrogen atoms, which can occur between antioxidants and peroxyl radicals. In this method, nitrogen radicals are created instead of peroxyl radicals, which are more stable and less transient, favoring their reaction with antioxidant compounds, which can result in higher levels of antioxidant capacity.
In addition, it is important to emphasize the synergistic interactions present in the chemical constituents of these essential oils, which may also have contributed to the antioxidant activity presented in each of the chemical profiles [26].

Preparation and Characterization of the Botanical Material
The samples from the Eugenia florida leaves were dried in a greenhouse with air circulation at a temperature of 35 • C for 5 days, and then shredded in a knife mill (Tecnal, model TE-631/3, Brazil). Moisture content was analyzed using an infrared moisture detector (ID50; GEHAKA, São Paulo, Brazil), in the temperature range of 60 to 180 • C, with 1 • C increments and bidirectional RS-232C output.

Extraction of Essential Oils
The samples were subjected to hydrodistillation in modified Clevenger-type glass systems for 3 h, coupled to a refrigeration system to maintain the condensation water at around 12 • C. After the extraction, the oils were centrifuged for 5 min at 3000 rpm, dehydrated with anhydrous sodium sulfate and centrifuged again under the same conditions. Oil yield was calculated in mL/100 g. The oils were stored in amber glass ampoules, sealed with flame, and stored in a refrigerator at 5 • C.

Chemical Composition Analysis
The chemical compositions of the EOs of E. florida (A and B), were analyzed using a Shimadzu QP-2010 plus (Kyoto, Japan) a gas chromatography system equipped with an Rtx-5MS capillary column (30 m × 0.25 mm; 0.25 µm film thickness) (Restek Corporation, Bellefonte, PA, USA) coupled to a mass spectrometer (GC/MS) (Shimadzu, Kyoto, Japan). The program temperature was maintained at 60-240 • C at a rate of 3 • C/min, with an injector temperature of 250 • C, helium as the carrier gas (linear velocity of 32 cm/s, measured at 100 • C) and a splitless injection (1 µL of a 2:1000 hexane solution), using the same operating conditions as described in the literature [67,68]. Except for the carrier hydrogen gas, the components were quantified using gas chromatography (CG) on a Shimadzu QP-2010 system (Kyoto, Japan), equipped with a flame ionization detector (FID) (Kyoto, Japan), under the same operating conditions as before. The retention index for all volatile constituents was calculated using a homologous series of n-alkanes (C 8 -C 40 ) Sigma-Aldrich (San Luis, USA), according with Van den Dool and Kratz [69]. The components were identified by comparison (i) of the experimental mass spectra with those compiled in libraries (reference) and (ii) their retention indices to those found in the literature [28,70,71].
To measure the antioxidant capacity, 2.97 mL of the ABTS •+ solution was transferred to the cuvette, and the absorbance at 734 nm was determined using a Biospectro SP 22 spectrophotometer (São Paulo, Brazil). Then, 0.03 mL of the sample was added to the cuvette containing the ABTS •+ radical and, after 5 min, the second reading was performed. The synthetic antioxidant Trolox (6-hydroxy-2,5,7,8-tetramethylchromono-2-carboxylic acid; Sigma Aldrich; 23881-3; São Paulo, Brazil) was used as a standard solution for the calibration curve (y = 0.4162x − 0.0023, where y represents the value of absorbance and x, the value of concentration, expressed as mM; R 2 = 0.9789). The results were expressed as mM. The values found for the samples were compared to the Trolox standard (1 mM).

Antioxidant Capacity by Inhibition of Radical DPPH • (AC-DPPH • )
The test was carried out according to the method proposed by [74]. To measure the antioxidant capacity, initially, the absorbance of DPPH • solution (2,2-diphenyl-1picrylhydrazyl; Sigma-Aldrich; D9132; São Paulo, Brazil) 0.1 mM diluted in ethanol was determined. Subsequently, 0.6 mL of DPPH • solution, 0.35 mL of distilled water and 0.05 mL of the sample were mixed and placed in a water bath at 37 • C for 30 min. Thereafter, the absorbances were determined in a spectrophotometer Bioespectro SP 22 (São Paulo, Brazil) at 517 nm. The synthetic antioxidant Trolox (6-hydroxy-2,5,7,8-tetramethylchromono-2carboxylic acid; Sigma-Aldrich; 23881-3; São Paulo, Brazil) was used as a standard solution for the calibration curve (y = 0.1271x − 0.0023, where y represents the value of absorbance and x, the value of concentration, expressed as mM; R 2 = 0.9856). The results were expressed as mM. The values found for the samples were compared to the Trolox standard (1 mM).

Statistical Analysis
Results were expressed as the mean of three replicates ± standard deviation of percent inhibition. The activity of essential oils from E. florida leaves was analyzed by Student's T-Test, considering p < 0.05 as significant.

Conclusions
The essential oils from Eugenia florida specimens showed different chemical compositions, which may have been influenced by the type of ecosystem where the samples were obtained, i.e., specimen A presented the hydrocarbon monoterpenes and oxygenated monoterpenes classes, predominantly, while specimen B presented hydrocarbon sesquiterpenes and oxygenated sesquiterpenes as major classes. This difference may have affected the potential antioxidant activity of the samples, as specimen B showed superior antioxidant activity for both analyzed methods (TEAC and DPPH). The essential oils of Eugenia florida DC. may be a promising source of antioxidant compounds.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.