Essential Oil Content and Chemical Composition in 14 Selected Species From a Stretch of Restinga in Southern Brazil

The restinga is an Atlantic Forest ecosystem characterized by tree, shrub, and herb species that are rich sources of essential oils. In this study, we aim to quantify the essential oil content and determine the chemical constituents of fresh leaves of 14 plant species in a restinga stretch in southern Brazil. Essential oils were obtained by hydrodistillation in a Clevenger-type apparatus and analyzed by gas chromatography coupled to mass spectrometry. Campomanesia reitziana, Cortaderia selloana, and Sophora tomentosa had no essential oils. Total essential oil content ranged from 0.01% (Mikania involucrata) to 1.56% (Varronia curassavica). In total, 60 chemical constituents were identified, representing between 46.2% and 96.5% of the chemical composition of the essential oils. Limonene was the common constituent in all species in which the essential oils were present. The major constituents were ar-curcumene (15.1%) and cis-chrysanthenol (14.2%) in Ambrosia elatior; benzyl benzoate (43.5%) and benzyl salicylate (23.7%) in Aniba firmula; caryophyllene oxide (35.7%) and spathulenol (10.6%) in Austroeupatorium inulaefolium; spathulenol (19.8%) and caryophyllene oxide (14.0%) in Baccharis spicata; caryophyllene oxide (16.3%) in Eugenia astringens; curzerene (30.0%), limonene (13.0%), and germacrone (11.9%) in Eugenia uniflora; caryophyllene oxide (17.1%) and ledol (11.3%) in Lantana camara; caryophyllene oxide (27.7%) and limonene (12.7%) in M. involucrata; 1,8-cineole (19.8%) in Psidium cattleianum; limonene (10.2%) in Schinus terebinthifolius, and allo-aromadendrene (15.2%) in V. curassavica. We expect that our results can assist in selecting species of potential interest for herbal, phytotherapeutic, and cosmetic products.


Introduction
The restinga is an ecosystem type that originated from Quaternary marine deposits and is part of the Atlantic Forest biome. Restingas are characterized by dunes and sandy coastal plains, with vegetation growing in open and/or inaccessible places such as lagoons, lakes, and marshes. These communities include a mosaic of plants with physiognomic and xeromorphic variations that respond to the numerous constraints imposed by nutrient-poor sandy soils, drought, salinity, solar radiation, constant winds, and high air and soil temperatures (Reinert et al., 1997). The unique character of the restinga comes from a plant community with high ecological plasticity. Many restinga species colonize, grow, and survive in inhospitable situations despite their origin in forest environments.
The ecological balance of species in the restinga is largely maintained through the propagation of specific plants, including the abundant aromatic herbs, shrubs, and trees. The botanical families of Asteraceae, Fabaceae, Myrtaceae, and Poaceae are the most representative of this habitat (Melo Junior & Boeger, 2015). Other common families include Anacardiaceae, Boraginaceae, Lauraceae, and Verbenaceae. Species of this ecosystem are characterized by adaptations to its adverse conditions. Plants use various strategies to deal with their difficult environmental conditions (Amorim & Melo Júnior, 2017). These include changes in secondary metabolism, resulting in the production of a wide variety of compounds, including essential oils. For quantification, the essential oils were injected into a gas chromatograph (Agilent 7890) equipped with a flame ionization detector operated at 280 °C. Hydrogen was used as a support gas at a flow rate of 1.5 mL/min, using the same column and conditions described above. The quantification of each constituent was estimated by electronic integration of the flame ionization detector with the corresponding peak area, which was determined using the average of three injections.

Identification of the Chemical Constituents of the Essential Oil
Identification of the chemical constituents of the essential oil was performed by comparing Kovats indices (KIs) obtained from a correlation of the homologous series of alkanes (C 8 -C 26 ) and matching their mass spectra with those of libraries, and comparing KIs from the literature (Adams, 2007).

Statistical Analyses
Essential oil content data were tested for homogeneity using Bartlett's test. An analysis of variance (ANOVA) was performed using ASSISTAT® software, version 7.7 (Silva & Azevedo, 2016), and a Tukey test was used to determine significance at the p > 0.05 level.

Results and Discussion
The essential oil was obtained through the hydrodistillation process from 11 of the 14 species sampled. Although these species are recorded in other coastal regions of the Atlantic Forest biome (Silva et al., 2021), to our knowledge, there is no information about the chemical compounds of essential oil found in these populations. Sophora tomentosa, Campomanesia reitziana, and Cortaderia selloana did not have essential oil in their leaves (Table 1). Though the plants exhibited presumed aromatic potential at the time of collection, these may be attributed to the presence of other compounds. Many water-soluble substances have odors that can be confused with the presence of essential oils, such as free amino acids, soluble carbohydrates, and aliphatic oxygenated compounds (Eisenreich et al., 1997). Note. * Content expressed as % of essential oil extracted from fresh leaves by hydrodistillation. ** Means followed by the same letter are not significantly different from each other according to the Tukey test at the 5% probability level. *** No essential oil present in their leaves.
The highest essential oil content was observed in Varronia curassavica (1.56%), while essential oil content ranged between 0.01% and 1.04% in the remaining plants (Table 1). Although phytochemical studies have been carried out for the selected species, comparisons of essential oil content are not an easy task due to their heterogeneous profiles. For example, the essential oil content of Lantana camara reported in the literature ranges jas.ccsenet.org Journal of Agricultural Science Vol. 13, No. 11; from 0.004% (Zhu et al., 2013) to 0.09% (Sousa et al., 2010). These differences may be attributed to several reasons, including the duration and method of extraction, population genetics of each species (Nizio et al., 2015), the plant part used (Cole et al., 2014), collection time (Sousa et al., 2010), exposure to sunlight (Feijó et al., 2014), seasonality, temperature, and precipitation (Matias et al., 2016).
A total of 60 chemical constituents were identified in the essential oils extracted, comprising between 46.2% and 96.5% of their chemical compositions (Table 2). Of these constituents, 7.5-18.7% were from the hydrocarbon monoterpene class, 0.3-27.2% oxygenated monoterpenes, 2.5-32.1% hydrocarbon sesquiterpenes, 3.9-69.8% oxygenated sesquiterpenes, 1.4% phenylpropanoids, and 1.8-68.9% were esters. Limonene was the only common constituent in all the species analyzed, with a concentration ranging between 4.9% and 13.0% (Table 2). This similarity may be associated with the role of limonene as a precursor of monoterpene biosynthesis (Trombin- Souza et al., 2017;de Souza et al., 2021).  Limonene was only the most abundant constituent in Schinus terebinthifolius, accounting for 10.2% of the essential oil. The presence of 9-epi-(E)-caryophyllene (10.1%) and p-cymen-7-ol (22.5%) have been reported in fresh leaves of the species , as well as germacrene D (23.8%), bicyclogermacrene (15.0%) (Santana et al., 2012), and δ-3-carene (68.78%) (Uliana et al., 2016). However, the concentrations of these constituents measured in this study were lower or absent (Table 2). Quantitative and qualitative variations in the species' essential oil may be related to the metabolic plasticity of S. terebinthifolius. The production of secondary metabolites is likely influenced by the peculiarities of each environment, including abiotic and edaphic conditions, as well as herbivores, pollinators, and seed dispersers. Furthermore, alterations in essential oil biosynthesis may also reflect a possible deviation in metabolic pathways to help the plant survive in particular environments.
The most abundant constituents in the Asteraceae species were spathulenol (19.8%), caryophyllene oxide (14.0%), α-cadinol (12.0%), and epi-α-muurolol (10.4%) in Baccharis spicata; caryophyllene oxide (35.7%) and spathulenol (10.6%) in Austroeupatorium inulaefolium; caryophyllene oxide (27.7%) and limonene (12.7%) in Mikania involucrate, and ar-curcumene (15.1%) and cis-chrysanthenol (14.2%) in Ambrosia elatior ( Table 2). The chemical profiles of these oils indicated a predominance of sesquiterpenes (3.8-69.8%) over monoterpenes (2.8-18.4%). These findings can be interpreted as a competition between two pathways for the same precursor. It is known that the concentrations of monoterpenes and sesquiterpenes are negatively correlated (Ghaffari et al., 2011). Thus, the highest flux of isopentenyl diphosphate (IPP) among the species studied tended to be in the cytosol (the site of sesquiterpene biosynthesis) in the restinga environment. Higher proportions of sesquiterpenes may also indicate the stressful conditions that plants undergo in this ecosystem since high temperatures, strong winds, and solar radiation contribute to the volatilization of smaller molecules such as monoterpenes. In contrast jas.ccsenet.org Vol. 13, No. 11; to our results, the essential oils of other Asteraceae species collected in non-coastal areas of the Atlantic Forest had roughly equal proportions of monoterpenes and sesquiterpenes . This suggests that site-specific characteristics (i.e., environmental differences) are determining factors in terpene variation.
In Aniba firmula, the main constituents were benzyl benzoate (43.5%) and benzyl salicylate (23.7%; Table 2). The essential oils of Brazilian species of Lauraceae are generally divided into groups of chemotypes based on their main constituents. Aniba firmula belongs to the benzoate group. Species in this family can also belong to the linalool and allylbenzene chemotypes, depending on the principal constituents, which remain consistent across each species (Morais et al., 1972). Similarly, Aniba firmula exhibited low variation in the main constituents of its essential oil and lower sensitivity to environmental characteristics. These findings are interesting because they reveal that the restinga conditions did not result in significant changes in the essential oil composition.
The main constituents found in species of Myrtaceae were caryophyllene oxide (16.3%) in Eugenia astringens; limonene (13.0%), curzerene (30.0%), and germacrone (11.9%) in Eugenia uniflora, and 1,8-cineol (19.8%) in Psidium cattleianum ( Table 2). The chemical constituents of Myrtaceae essential oils belong predominantly to the hydrocarbons (14.5-18.7%), oxygenated monoterpenes (0-27.2%), and oxygenated sesquiterpenes (22.5-58.4%). This finding contrasts with earlier results for Myrtaceae plants in the Atlantic Forest, which showed that sesquiterpenes generally predominated Albuquerque et al., 2012). In the restinga, an increase in hydrocarbon and oxygenated monoterpenes has been observed Defaveri et al., 2011). Although monoterpenes volatilize easily under conditions of high temperature and solar intensity (Arruda and Victório, 2011), the abundance of these compounds in species of this family can be explained by their thick and wax-covered leaves, especially in plants from the restinga (Donato and Morretes, 2007). Thus, the functional traits of the leaves indicate the existence of mechanisms to reflect incident light and protect against the loss of water and volatile substances.
This study reports the chemical diversity present in the essential oils of plant species collected in the restinga ecosystem of southern Brazil. Although E. uniflora and V. curassavica are commercially exploited, in this work we report that these species have a high content of the substance of economic interest such as curzurene (30.0%) and α-humulene (2.4%), which may represent a potential commercial. Likewise, the selection of matrices with economic value can be subsidized with sustainable use practices of the species, since they are distributed in Biome highly threatened by anthropogenic disturbance (de . This information is critical when selecting species with economic potential for phytotherapeutic products, as well as for the phytosanitary and cosmetic industries.

Conclusions
In conclusion, our study reports the yield and chemical composition of essential oils from 14 species distributed on the coast of Santa Catarina, Brazil. The EO content ranges from 0.01% (M. involucrata) to 1.56% (V. curassavica). The major constituents are ar-curcumene (15.1%) and cis-chrysanthenol (14.2%) in A.elatior;