Chemical Composition, Enantiomeric Distribution, and Sensory Evaluation of the Essential Oils Distilled from the Ecuadorian Species Myrcianthes myrsinoides (Kunth) Grifo and Myrcia mollis (Kunth) DC. (Myrtaceae)

The essential oils of Myrcianthes myrsinoides and Myrcia mollis, belonging to the Myrtaceae family, were obtained by steam distillation. They were analyzed by gas chromatography-mass spectrometry (GC-MS), gas chromatography-flame ionization detector (GC-FID), enantioselective gas chromatography, and gas chromatography-olfactometry (GC-O). A total of 58 compounds for Myrcianthes myrsinoides essential oil (EO) and 22 compounds for Myrcia mollis EO were identified and quantified by GC-MS with apolar and polar columns (including undetermined components). Major compounds (>5.0%) were limonene (5.3%–5.2%), 1,8-cineole (10.4%–11.6%), (Z)-caryophyllene (16.6%–16.8%), trans-calamenene (15.9%–14.6%), and spathulenol (6.2%–6.5%). The enantiomeric excess of eight chiral constituents was determined, being (+)-limonene and (+)-germacrene D enantiomerically pure. Eight components were identified as determinant in the aromatic profile: α-pinene, β-pinene, (+)-limonene, γ-terpinene, terpinolene, linalool, β-elemene and spathulenol. For M. mollis, the major compounds (>5.0%) were α-pinene (29.2%–27.7%), β-pinene (31.3%–30.0%), myrcene (5.0%–5.2%), 1,8-cineole (8.5%–8.7%), and linalool (7.7%–8.2%). The enantiomeric excess of five chiral constituents was determined, with (S)-α-pinene and (+)-germacrene D enantiomerically pure. The metabolites β-pinene, 1,8-cineole, γ-terpinene, terpinolene, linalool, and (E)-β-caryophyllene were mainly responsible for the aroma of the EO. Finally, the M. myrsinoides essential oil has an inhibitory activity for cholinesterase enzymes (IC50 of 78.6 μg/mL and 18.4 μg/mL vs. acethylcholinesterase (AChE) and butyrylcholinesterase (BChE) respectively). This activity is of interest to treat Alzheimer’s disease.

The undetermined compounds with a molecular weight of 204 or 220 are most probably hydrocarbon or oxygenated sesquiterpenoids. Their amount accounted for 0.2%-2.2% in the M. myrosinoides essential EO and 0.4%-3.2% in the M. mollis EO.

Enantioselective Analysis
The distribution of the enantiomeric pairs in both species' essential oil was determined with two capillary columns coated with a chiral selector: diethyl terbutylsilyl-β-cyclodextrin and diacetyl terbutylsilyl-β-cyclodextrin [38,39].
In M. myrsinoides EO, two chiral constituents, (+)-limonene and (+)-germacrene D, are baseline separable one another only with the first chiral selector, where they resulted to be enantiomerically pure. The enantiomeric distribution and enantiomeric excess were calculated, and the results are shown in Table 3.
The two above chiral columns were used to measure the enantiomeric distribution and enantiomeric excess of five enantiomeric compounds in the M. mollis essential oil. α-Pinene, germacrene D, and α-thujene were only separated by the diethyl terbutylsilyl-β-cyclodextrin column. The results are reported in Table 4.

Sensory Evaluation
The olfactive active compounds were estimated in each of the two investigated essential oils by GC-O. Table 5 reports their linear retention indices and the corresponding sensory description. In addition, an aromagram was constructed and flavor dilution factors (FD) measured by aroma extract dilution analysis (AEDA) [40], which can be visualized in Figures 1 and 2.

Biological Activity
The inhibitory activity of the two investigated EOs were then tested on two cholinesterase enzymes: AChE and BChE. Only M. myrsinoides presented an inhibitory activity for the investigated enzymes, with an IC 50 of 78.6 µg/mL for acetylcholinesterase and of 18.4 µg/mL for butyrylcholinesterase.
The IC 50 value for the M. mollis EO was > 50 µg/mL for both enzymes. The IC 50 of donepezil, the positive control, was 0.04 µg/mL for AChE and 3.6 µg/mL for BChE.
GC-MS-FID analysis led to the quantification of more than 90% of the EO components for both species. Quantitation was based on the determination of the relative response factor (RRF) according to the combustion enthalpy [42]. This method assumes that the RRF depends on the elemental composition of the molecules, so that compounds with the same molecular formula and number of aromatic rings have the same RRF [43]. The present study constitutes one of the first applications of this method with external calibration, using isopropyl caproate as a calibration standard and n-nonane as internal standard.
The sensory description of each odorant was determined by two trained panelists and resulting descriptors were consistent with the data reported in the literature. These results derived from the evaluation of two panelists are indeed not exhaustive, from a statistical point of view, to describe correctly the aromatic EO profile. However, they can be considered a contribution useful to justify the similar aroma for two chemically different products. A distinctive woody note was described for α-pinene and β-pinene [16], an herbal note for γ-terpinene, a plastic predominant odor for terpinolene, and a floral note for linalool. With M. myrsinoides EO, limonene was perceived with a flavor dilution factor (FD) of 16, while in the M. mollis EO, β-pinene and 1,8-cineole presented a FD of 8. FD were determined by aroma extract dilution analysis (AEDA), which allows a reliable qualitative and quantitative odor analysis of each analyte eluting from the chromatographic column to be obtained [15]. In this study, we can observe that the chemical analysis, showing a quite different composition, should deny the previous sensory statement on the similar aroma for the two oils. However, the GC-O evaluation indicates that M. myrsinoides and M. mollis share four main odorants (α-pinene, γ-terpinene, terpinolene, and linalool). This fact fully justifies the similar odor perception and demonstrates that GC-O is actually the main technique to deeply investigate the aromatic profile of an odorous mixture.
The biological roles of AChE and BChE are different. The former plays a fundamental role in the human nervous system, when hydrolyzing acetylcholine, a neurotransmitter of cholinergic synapses, allowing the restoration of the functionality of the postsynaptic terminals [26]. The latter does not have a well-defined physiological function but is known to act as an endogenous suppressant of anticholinergic compounds by hydrolysis of hydrophobic and hydrophilic esters [45]. The results obtained with the biological activity tests for the M. myrsinoides EO indicated an average maximum inhibitory concentration of 78.6 µg/mL anti-AChE and 18.4 µg/mL anti-BChE, i.e., values close to those reported for other essential oils rich in pinene isomers [46]. Furthermore, compared to the positive control, M. myrsinoides essential oil was almost inactive for AChE and 4.8 less active for BChE. However, the much higher activity against BChE than AChE is consistent with a similar case reported by the authors in a recent publication [25], where the difference was hypothetically attributed to a possible selective inhibitory mechanism. On the other hand, the lack of activity for M. mollis EO, which presents a more important monoterpene composition, could be related to the intriguing possibility that the opposite enantiomeric excess of some chiral constituents, such as α-pinene and limonene, is responsible for the different activity.

Materials and Methods
The chemical, enantioselective, and GC-O analyses were carried out with an Agilent Technologies GC-MS system consisting of a 6890N gas chromatograph with a 7683 autoinjector combined with a 5973 INERT mass spectrometric detector (Santa Clara, CA, USA) and a flame ionization detector (FID). The instrument was also equipped with a Gerstel ODP 3 sniffing port (Gerstel GmbH Co., Mülheim an der Ruhr, Germany).
The mass spectrometer detector operated in SCAN mode (40-350 m/z), with an electron ionization source at 70 eV.
The qualitative and quantitative analyses were run with both an apolar and a polar capillary column. The apolar column was 5% phenyl methylpolysiloxane stationary phase (DB-5ms from Agilent Technologies, 30 m long, 0.25 mm internal diameter, and 0.25 µm film thickness); the polar column was a polyethylene glycol stationary phase (HP-INNOWax from Agilent Technologies, 30 m × 0.25 mm × 0.25 µm).
The GC-O analyses were run with the above DB-5ms column at the exit of which a flow splitting of 50% between detector (FID) and sniffing port was applied. Carrier gas for all analysis was helium GC purity (Indura, Guayaquil, Ecuador).
The enzyme inhibition tests were carried out with a Varioskan Flash detection system, purchased from Thermo Scientific (Waltham, MA, USA).
All solvents, alkanes and internal standard were analytical grade (purity > 99%), from Sigma-Aldrich (Saint Louis, MO, USA). The calibration standard was isopropyl caproate, obtained in the authors' laboratory by synthesis and purified to 98.8% (GC-FID purity).

Distillation of the Essential Oil
The essential oils of both plants were obtained by steam distillation for 4 hours, using a Clevenger-type stainless steel apparatus. The fresh M. myrsinoides plant material (leaves) was distilled in four repetitions, two with 2.2 kg and two with 0.6 kg; the yield was 0.3% ± 0.01% (w/w). The fresh M. mollis leaves were also distilled in four repetitions of 1.9 kg each, obtaining a yield of 0.2% ± 0.02% (w/w). The EOs were immediately dried on anhydrous sodium sulfate and stored in the dark at −4 • C.

Chemical Analyses
The analytical samples were prepared by diluting an exactly weighed amount of essential oil (corresponding to 10 µL) with 1 mL of internal standard solution, previously prepared by diluting 0.7 mg of n-nonane to a total volume of 10 mL with cyclohexane. Such a preparation was repeated for each EO of the two species. These samples were directly used for the qualitative, quantitative, and enantioselective analyses.
GC-O analyses: two concentrated EO samples were prepared by diluting 30 µL of M. myrsinoides EO in 500 µL of cyclohexane and 10 µL of M. mollis EO in 1 mL of the same solvent.
Qualitative analysis: the GC-MS analyses of M. myrsinoides EO for both DB-5ms and HP INNOWax columns were carried out under the following conditions: temperature program from 60 • C (5 min) to 180 • C at 3 • C/min then to 250 • C (5 min) at 15 • C/min. The injector operated in split mode, with a ratio of 40:1; injection volume of 1 µL, and temperature of 250 • C; helium flow rate: 1 ml/min. The GC-MS qualitative analyses of M. mollis EO for both DB-5ms and HP INNOWax columns were carried out under the following conditions: temperature program from 60 • C (5 min) to 165 • C at 3 • C/min then to 250 • C (5 min) at 15 • C/min. The injector operated in split mode, with a ratio of 50:1; injection volume of 1 µL and temperature of 250 • C; helium flow rate: 1 ml/min. Additionally, a mixture of n-alkanes (C 9 -C 25 ) was injected under the same conditions to determine the linear retention indices (LRIs).
Quantitative analysis: all samples of M. mollis and M. myrsinoides EOs, prepared as previously described, were analyzed by GC-FID with both columns, under the same instrumental conditions described for qualitative analyses. Four calibration curves were obtained, injecting six dilutions of isopropyl caproate (calibration standard) and n-nonane (internal standard) for each EO in both columns. The dilutions were obtained by diluting 0.8 mg, 1.8 mg, 4.1 mg, 8.3 mg, 16.9 mg, and 34.5 mg of isopropyl caproate and an exactly weighed amount of 7.6-7.8 mg of n-nonane to 10 mL with cyclohexane. All calibration curves achieved a R 2 > 0.999.

Enantioselective Analyses
The enantioselective analyses were carried out by GC-MS under the following temperature program for both EOs: from 60 • C (5 min) to 220 • C (5 min) at 2 • C/min. The injector operated in split mode, with a ratio of 40:1; injection volume of 1 µL, and temperature 220 • C. A mixture of n-alkanes (C 9 -C 25 ) was injected under the same conditions as for conventional analysis to determine LRIs.
The identification of each enantiomer was achieved by injecting a series of enantiomerically pure standards, available from one of the authors (C.B.).

AEDA Analysis
The GC-O analyses were performed according to the AEDA method for both species, by injecting 2 µL of each sample, prepared as described in Section 4.4, in a sequence of increasing dilutions until no odor was perceived by all the panelists. The dilutions were obtained by operating on the split values of 5:1, 10:1, 20:1, 40:1, and 80:1 for both species, and up to 160:1 for M. myrsinoides EO.
The following temperature program was applied: 40 • C (1 min) to 280 • C (10 min) at 12 • C/min. Helium flowrate: 2 mL/min. Two operators ran the olfactory analysis, perceiving the odors without visualizing the chromatogram and describing the perceived aroma of each analyte eluting at the sniffing port. The adopted acceptance criteria implied that a perception, to be accepted, need to be detected by at least one panelist in two following dilutions or by both panelists in a single dilution. The results allowed the construction of an aromagram over the chromatogram, based on the LRI and the FD of each odor described.

Biological Activity
The activities of cholinesterase (ChE) were evaluated by following a colorimetric protocol adapted from the literature [46][47][48]. The catalyst efficiently hydrolyzes acetylthiocholine (ACh), the sulfur analogue of the natural substrate of these enzymes. After hydrolysis, this substrate analogue produces acetate ion and thiocoline. Thiocoline reacts with the highly reactive 5,5'-dithiobis-(2-nitrobenzoic acid) ion (DTNB) to give a yellow color, which can quantitatively be monitored by measuring its spectrophotometric absorption at 412 nm. The inhibition assay volume contained 200 µL of phosphate buffered saline (pH 7.4), DNTB (1.5 mM) and test sample in DMSO (1% v/v). The assay was carried out on Electrophorus electricus acetylcholinesterase and equine serum butyrylcholinesterase that were both dissolved in PBS pH 7.4 and used at 25 mU/mL. After 10 minutes of preincubation, the substrate acetylthiocholine iodide (1.5 mM) was added to start the reaction. Multiple 96-well microliter sites were read in the detection system during 30 min of incubation at 30 • C. All measurements were carried out in triplicate. When possible, the IC 50 values were calculated using the GNUPLOT package online (www.ic50.tk, www.gnuplot.info). Donepezil was used as reference ChE inhibitor and showed an IC 50 = 100 nM for AChE and 8500 nM for BChE. With this assay, the possibility of false positive inhibition results previously described for high concentration (> 100 µg/mL) of amine or aldehyde compounds cannot be excluded [25].

Conclusions
In this study, the chemical, enantiomeric, and sensory profiles of two EOs were investigated for the first time. The EO of M. myrsinoides was established to consist mainly of sesquiterpenoids, while in the M. mollis EO, monoterpenoids prevail. Despite the different chemical compositions, the similarity in the AEDA evaluation could justify the similarity of the two aromas.
Furthermore, the EO of M. myrsinoides was determined to be a weak selective inhibitor of BChE, with an inhibitory activity hypothetically attributable to a chirality dependent mechanism of monoterpenes (IC 50 of 78.6 µg/mL and 18.4 µg/mL vs. AChE and BChE, respectively).