Chemometric Screening of Fourteen Essential Oils for Their Composition and Biological Properties

Essential oils (EOs) obtained from aromatic plants are widely used worldwide, especially in cosmetic and food products due to their aroma and biological properties and health benefits. Some EOs have significant antimicrobial and antioxidant activities, and thus could effectively increase the shelf lives of foodstuff and beverages. In this study, fourteen essential oils (clove, eucalyptus, fennel, lavender, oregano, palmarosa, pepper, star anise, tea tree, turmeric, Chinese yin yang, Japanese yin yang, and ylang ylang) from different medicinal plant families were screened by gas-chromatography–mass spectrometry (GC–MS) for their different chemical profiles and bioassays were performed to assess their antifungal and antioxidant activities. The results obtained were assessed by principal component analysis (PCA). PCA distinguished six groups characterized by different terpene chemotypes. Amongst the EOs studied, the clove EO showed the strongest antioxidant activity characterized by an EC50 of 0.36 µL/mL. The oregano EO had the greatest antiyeast activity characterized by a minimal inhibitory concentration of 10 µL/mL. In conclusion, clove and oregano EOs are strong antifungal and antioxidant agents, respectively, with great potential in the food industry to avoid spoilage and to increase shelf life.


Introduction
Aromatic plants (APs) have been used since antiquity in folk medicine and as food preservatives due to their many biologically active components including phenolic compounds (e.g., phenolic acids, flavonoids, coumarins, lignans, stilbenes, and tannins), nitrogen-containing compounds (alkaloids, amines, and betalains) and terpenoids (including carotenoids). Essential oils (EOs) are the volatile fractions extracted from aromatic plants. EOs are highly complex natural mixtures of low-molecular mass volatile compounds such as terpenes and aldehydes that are responsible for the typical aroma of any APs [1][2][3][4][5][6]. EOs are produced by the different parts of plants, including buds, flowers, leaves, stems, twigs, seeds, fruits, roots, wood and bark, and stored in secretory cells, cavities, canals, epidermic cells and glandular trichomes [1,2,[4][5][6]. Nearby 3000 EOs originated from 2000 plant species and are distributed around 60 botanical genera which are produced worldwide. In nature, EOs play important biological and ecological roles since they are directly involved in plant  In this work, chemotyping of EOs was performed by studying the chemical variability between the samples and transforming the original correlated variables into new, reduced, artificial, and orthogonal ones, called principal components (F1 and F2 in Figure 1). The score-plot shows the relationship between different EOs (the observations) and the loading plot shows how strongly each characteristic influences a principal component (other variables). Graphically, EOs close to each other in the score-plots have similar chemotype. In this regard, the first two principal components (F1 and F2) had the highest share of the total variance (33.6%), due to the relatively high number of variables analyzed (177 compounds). In fact, twelve principal components would explain 100% of the total variation. It is, however, worth noting that the number of principal components depends on the number of variables, as the percent of total variance increases with lower number of variables. Generally, by increasing the number of variables, the proportion of total variance decreases.
PCA has been shown to be very useful when combined with targeted and untargeted analytical fingerprinting techniques and applied to herbal extracts and EOs [43].
In this work, chemotyping of EOs was performed by studying the chemical variability between the samples and transforming the original correlated variables into new, reduced, artificial, and orthogonal ones, called principal components (F1 and F2 in Figure 1). The score-plot shows the relationship between different EOs (the observations) and the loading plot shows how strongly each characteristic influences a principal component (other variables). Graphically, EOs close to each other in the score-plots have similar chemotype. In this regard, the first two principal components (F1 and F2) had the highest share of the total variance (33.6%), due to the relatively high number of variables analyzed (177 compounds). In fact, twelve principal components would explain 100% of the total variation. It is, however, worth noting that the number of principal components depends on the number of variables, as the percent of total variance increases with lower number of variables. Generally, by increasing the number of variables, the proportion of total variance decreases.
The first principal component F1, which accounted for 17.5% of the total variance, was positively correlated with eucalyptol (0.85), the major compound found in EOs of eucalyptus and Chinese yin yang, and with D-limonene (0.70), the main compound found in EOs of bitter orange and pepper. The F2, explaining 16.1% of the total variance, was positively correlated with carvacrol (0.9), the most abundant compound found in oregano EOs. Based on the PCA analysis, the 14 EOs studied were grouped into six different clusters, which allowed to distinguish six chemotypes as shown in Figure 1. The first group (I) includes eucalyptus and Chinese yin yang, both characterized by a high content of eucalyptol, whereas the second one (II) was made up of bitter orange and pepper, which contains high percentage of D-limonene. Cluster The first principal component F1, which accounted for 17.5% of the total variance, was positively correlated with eucalyptol (0.85), the major compound found in EOs of eucalyptus and Chinese yin yang, and with d-limonene (0.70), the main compound found in EOs of bitter orange and pepper. The F2, explaining 16.1% of the total variance, was positively correlated with carvacrol (0.9), the most abundant compound found in oregano EOs.
Based on the PCA analysis, the 14 EOs studied were grouped into six different clusters, which allowed to distinguish six chemotypes as shown in Figure 1. The first group (I) includes eucalyptus and Chinese yin yang, both characterized by a high content of eucalyptol, whereas the second one (II) was made up of bitter orange and pepper, which contains high percentage of D-limonene. Cluster III groups together star anise and fennel, based on the high quantity of anethole. Group IV was clustered by the presence of sesquiterpenes and was composed by cloves, palmarosa, and ylang ylang. Furthermore, it was observed that the EOs of turmeric, lavender, and oregano (Group V) were clustered because of the presence of eucalyptol and o-cymene. In the same way, the similarity between tea tree and Japanese yin yang can be observed (Group VI) due to the high relative percentage of the components eucalyptol and d-limonene.

Antiyeast Activity
Problems with chemical preservatives and the growing demand of consumers for natural food additives have turned the attention to plant-derived natural antimicrobials such as EOs. In fact, EOs could represent an alternative to synthetic preservatives against spoilage due to yeasts and molds. The antiyeast effect of the 14 EOs was tested by the solid medium diffusion method. Table 2 shows the mean diameters of inhibition halos of each EO obtained on the food-spoiling yeast S. cerevisiae. From the results obtained, yeast was resistant to star anise, turmeric, and ylang ylang, as the inhibition halo was completely absent. On the other hand, tea tree, fennel, lavender, pepper and Chinese yin yang showed inhibition halos ranging from 4 to 9 mm. Oregano was the best performing inhibitor of S. cerevisiae growth among all the EOs analyzed by displaying a 35-mm inhibition halo ( Table 2). The solid medium diffusion technique was proven to be a useful screening method in order to obtain preliminary data about the antiyeast effect of the EOs. The bitter orange, clove, eucalyptus, oregano, palmarosa, and Japanese yin yang EOs showed growth inhibition halos with diameters equal to or greater than 10 mm. Their minimum inhibitory concentration (MIC) values (Table 3) were showed a more than 90% reduction in the measured absorbance [44]. Amongst the 14 EOs tested, oregano showed the highest antiyeast activity, followed by clove, palmarosa, bitter orange, eucalyptus and Japanese yin yang. Thus, the oregano and clove EOs could be considered as potential antimicrobial agents to be used in the food bioconservation industry.
Oregano was the most active EO against the growth of S. cerevisiae, displaying an MIC of 10 µL/mL (Table 3); this result was in accordance with previous data [6,44]. Antiyeast activity of oregano EO could be associated with the high relative amount of carvacrol, as evidenced by its terpene composition (Table 1). In fact, carvacrol could be absorbed by the double phospholipidic layer of yeasts and could increase the fluidity and permeability of the membrane. Yeast cells, in the presence of carvacrol, have been shown to change the composition of their membrane's fatty acids as an adaptation mechanism to maintain the correct structure and function of the membrane [6,45].

Antioxidant Activity
Oxidation is one of the major causes of food degradation, which can occur along the entire food chain. In particular, oxidation is a process that causes unwanted quality changes, organoleptic variation, as well as affect the safety and nutritional value of foodstuffs. It occurs mainly through discoloration, odor generation and off-flavor, or through the formation of potentially toxic substances [2,12]. For this reason, the protection of food from oxidative deterioration is an important goal in food technology. EOs are regarded as GRAS thank to their chemical composition, biological effects and toxicity and are extensively exploited as natural antioxidants to be used in the food sector in contrast to chemical preservatives with known negative effects on human health [2,12].
DPPH is a stable free radical widely used to test the free-radical scavenging ability of various EOs. Clove, fennel, lavender, oregano, palmarosa, pepper, star anise, tea tree, turmeric, Chinese yin yang, and ylang ylang EOs were able to inhibit 50% of the radical scavenging activity of DPPH, as can be seen in Table 4. On the contrary, bitter orange, eucalyptus, and Japanese yin yang revealed no antioxidant activity. The lowest EC 50 values were found in clove (0.36 ± 0.02 µL/mL) and Chinese yin yang (5.35 ± 0.13 µL/mL), thus they are classified as very strong antioxidants, accordingly to Scherer and Godoy [51] and Cautela et al. [52].
Clove EO was renowned as one of the strongest antioxidants as previously reported by Teixera et al. [22], Misharina and Samusenko [24], Jirovetz et al. [34], and Wei and Shibamoto [53]. The high antioxidant activity of clove could be linked to the presence of eugenol as the EO's main constituent, revealed by the GC-MS analysis reported in Table 1. This compound is a phenylpropanoid derived from guaiacol and is known to possess antioxidant activity [34,53,54]. Clove EO, due to its high antioxidant activity, could be used as an antioxidant agent in order to prevent natural oxidation and deterioration of food and thus for increasing the shelf life.
Additionally, oregano and ylang ylang exhibited a strong antioxidant activity and their EC 50 values were 11.58 ± 0.22 and 12.71 ± 0.17 µL/mL, respectively. Oregano EO was known to possess antioxidant activity due to the presence of carvacrol, as reported in the literature [54], and our study was in accordance with this as this monoterpene phenol resulted to be very abundant (about 70%), as depicted in Table 1. Ylang ylang EO was characterized as having caryophyllene as its main component and so the strong antioxidant activity of EO could be related to presence of this sesquiterpenes, as previously described [57].
Only turmeric displayed a moderate antioxidant activity with 24.99 ± 0.44 µL/mL, while the remaining EOs (fennel, lavender, palmarosa, pepper, star anise, and tea tree) revealed poor antioxidant activity, as reported in Table 4. Overall, clove exhibited the highest antioxidant activity amongst the EOs studied, followed by Chinese yin yang, oregano, and ylang ylang.

Essential Oils
EOs of 13 plants were purchased from different companies, as reported in Table 5. EOs (clove, eucalyptus, fennel, lavender, oregano, palmarosa, star anise, tea tree, turmeric, Chinese yin yang, Japanese yin yang, and ylang ylang) were mostly obtained by steam distillation. Instead, bitter orange was extracted by means of cold pressing.
Moreover, pepper EO from Piper niger was obtained by hydrodistillation in Clevenger-type apparatus according to the European Pharmacopeia method 2005.2812 58 [58]. Briefly, 0.25 kg of pepper leaves were placed in a spherical bottom flask with a volume of 1 L. For optimization of the process, the volume of EO recovered by the Clevenger system was monitored at regulated intervals until the maximum yield was obtained.

GC-MS Analysis
Diluted EO samples (1:100 v/v in heptane) were analyzed by gas-chromatography-mass spectrometry (GC-MS) using the Trace 1300 GC coupled to the TSQ DUO triple quadrupole mass spectrometer (Thermo Scientific, Walthan, MA, USA) equipped with an electron impact ion source. Samples were injected without derivatization into the DB-5 column, 30-m length, 0.25-mm internal diameter, 0.25-µm film (Thermo Scientific, Walthan, MA, USA) using the 1:10 split mode. The following parameters were used: ionization energy of 70 eV, mass range between 50 and 550 m/z and interface temperature of 250 • C. The GC oven temperature was as follows: initial oven temperature of 70 • C and an isotherm for 1 min; subsequently, at 24 • C/min to 180 • C and an isotherm for 2 min, and then reached 280 • C at 50 • C/min where the isotherm was kept for a further 2 min. The carrier gas was helium (He, purity 99.999%) with a constant flow of 1.2 mL/min.
The acquisition data and the control of the instrument were performed through the Chromeleon Chromatography Data System software, CDS (Thermo Scientific, Walthan, MA, USA). The identification of the GC peaks corresponding to the components of the EOs was based on a direct comparison of the retention times and mass spectral data with those of standard compounds, computer matching with the National Institute of Standards and Technology (NIST) library.
Results were presented as a relative percentage of normalized peak area abundances, without the use of correction factors. The percentage data shown were mean values of two injections.

Antiyeast Activity on Saccharomyces cerevisiae
Saccharomyces cerevisiae obtained from the strain collection of the Institute of Research on Terrestrial Ecosystems (IRET) of the National Research Council (CNR) was used to evaluate the antiyeast activity of EOs. Stock culture was maintained at 4 • C on Malt Extract Agar (malt extract 30 g/L; peptone 5 g/L; agar 15 g/L). Inocula were obtained from overnight cultures on MEA plates at 28 • C. S. cerevisiae was grown in malt extract (ME 30 g/L; peptone 5 g/L) for 24 h in an orbital shaking incubator at 120 rpm at 28 • C [59]. The growth was monitored both by measuring the absorbance at 600 nm and by counting on plates (CFU/mL).

Solid Medium Diffusion Method
A solid medium diffusion procedure using wells in dishes was used to determine the antiyeast activity of all EOs [22,44]. For this, 1 mL of S. cerevisiae suspension with a concentration of 10 6 CFU/mL was uniformly spread on a sterile MEA petri dish (diameter 9 cm). After inoculum absorption by agar, wells were made using sterile glass tubes (diameter 6 mm) which were filled with 10 µL of each EO. The disc radius was not included. Negative controls were prepared with only 10 µL of DMSO. Petri dishes were incubated at 28 • C for 48 h; antiyeast activity was evaluated by measuring the diameter of the growth inhibition halos and was expressed in millimeters. All determinations were carried out in triplicate.

Minimum Inhibitory Concentration
EOs showing growth inhibition, as clear white zone with diameters equal to or greater than 10 mm with the solid medium diffusion method, were considered to determine the minimum inhibitory concentration (MIC) using a liquid medium. The MIC was defined as the lowest concentration of an EO that resulted in a reduction of >90% in the measured absorbance [15,22,44,60].
The microplate bioassay was used to study the antiyeast activity of EOs. The 24-well plates were prepared by dispensing into each well 1.8 mL of Malt Extract broth and 0.2 mL of yeast inoculum with a final concentration of 10 6 CFU/mL. An aliquot (20 µL) of each EO (with concentrations ranging from 0.1 to 1000 µL/mL) was transferred into a well. Negative controls were prepared adding only 20 µL of DMSO to ME medium. The microplates were sealed and incubated on a plate shaker (100 rpm) at 28 • C for 48 h. The yeast growth was evaluated by measuring the absorbance at 600 nm.

Antioxidant Activity
The radical scavenging activity (RSA) of EOs was evaluated by 2,2 -diphenyl-1-picrylhydrazyl (DPPH) assay according to the procedure of Blois [61]. Briefly, 150 µL of each EOs (with a concentration ranging from 0.1 to 1000 µL/mL) were mixed with 1.35 mL of 60-µM DPPH methanolic solution. The absorbance reduction at 517 nm of the DPPH was determined continuously for 60 min. The RSA was calculated as a percentage of DPPH discoloration, using the following equation: where A S is the absorbance of the solution when the EO was added and A DPPH is the absorbance of the DPPH solution, as reported [62]. The extract concentration (EC) necessary to achieve 50% radical DPPH inhibition (EC 50 ) was obtained by plotting the RSA percentage as a function of EOs' concentrations and was expressed as microliters per milliliter (µL/mL). Moreover, in order to classify the EOs, the antioxidant activity index (AAI) was determined as follows: AAI = DPPH concentration in reaction mixture/EC 50 (2) EOs were categorized as showing poor antioxidant activity (AAI < 0.5), moderate (0.5 < AAI < 1.0), strong (1.0 < AAI < 2.0) and very strong (AAI > 2.0), as reported by Scherer and Godoy [51] and Cautela et al. [52].

Statistical Analysis
Samples were analyzed in triplicate and all results were expressed as mean ± standard deviation (SD). Means, SD, calibration curves and linear regression analyses (R 2 ) were determined using Microsoft Excel 2013 (Microsoft Corporation, Redmond, WA, USA). The multivariate analysis was performed applying Principal Component Analysis (PCA) through XLSTAT Statistical Software using Microsoft Excel 2013.

Conclusions
From analyzing the chemical composition of the 14 EOs, bitter orange was found to consist mainly of monoterpenes while tea tree, eucalyptus, lavender, palmarosa, Chinese yin yang, and Japanese yin yang were characterized by high amounts of monoterpenoids and oxygenated monoterpenes. Sesquiterpenes, on the other hand, were abundant in ylang ylang, while star anise, fennel, clove, and oregano were rich in phenylpropanoids. Turmeric and pepper EOs exhibited a more heterogeneous chemoprofile and could not be associated with a prevalent terpenic class.
Overall, the 14 EOs studied were classified into six different chemotypes according to their chemical compositions and their relative abundances. The exploratory PCA technique allowed us to visualize, by reducing the dimension of the original data, and provide phytochemical relationships among all the EOs studied. Summarizing our results, amongst the 14 EOs studied, clove showed the highest antioxidant activity with an EC 50 of 0.36 µL/mL, followed by Chinese yin yang, oregano and ylang ylang. Moreover, oregano had the greatest antiyeast properties, inhibiting the growth of S. cerevisiae with an MIC of 10 µL/L, followed by clove, palmarosa, bitter orange, eucalyptus and Japanese yin yang EOs. Therefore, considering the EOs studied here, clove for its high antioxidant activity and oregano for its great antiyeast activity may have potential uses in the food and beverage sectors to increase shelf life and avoid deterioration.