Chemical Composition of the Essential Oil of the Endemic Species Micromeria frivaldszkyana (Degen) Velen.

Micromeria frivaldszkyana is an endemic species found only in Bulgaria. Its essential oil (EO) composition is unknown. This study assessed the EO yield and composition of M. frivaldszkyana as a function of the location and of drying prior to the EO extraction. M. frivaldszkyana was sampled from two natural habitats, Uzana and Shipka in the Balkan Mountains; the EO was extracted via hydrodistillation and analyzed on GC/MS. The plants from the two locations had distinct EO composition. The EO content (in dried material) was 0.18% (Uzana) and 0.26% (Shipka). Monoterpene ketones were the major group of the EO constituents. Also, hydrocarbons predominated in the EO from Shipka, and alcohols predominated in the EO from Uzana. The EO from Uzana had a greater concentration of menthone (56% vs. 17% from Shipka) and neomenthol (7.8% vs. 2.4%). Conversely, the EO from Shipka had greater concentrations of pulegone (50% vs. 20% from Uzana), limonene (10.1% vs. 2.6%), and germacrene D (3.4% vs. 1.1%). Drying prior to the EO extraction altered the concentration of some constituents. This is the first report of M. frivaldszkyana EO yield and composition. The EO showed some similarities with the chemical profile of other Micromeria species, but overall, it has an unique chemical profile and may have distinctive applications.


Introduction
Micromeria frivaldszkyana (Degen) Velen. is an endemic plant species found only in Bulgaria. The genus Micromeria Benth. belongs to the family Lamiaceae, subfamily Nepetoideae. The number of species in the Micromeria genus varies in different taxonomic schemes [1]. According to Harley et al. [2], the genus comprises about 70 species, but according to another taxonomic study [3], the genus Micromeria includes 54 species, 32 subspecies, and 13 varieties. Plants of the genus are widely distributed from the Himalayan region to the Macaronesian Archipelago (Madeira, Cape Verde, and Canary Islands) and from the Mediterranean to South Africa and Madagascar [4]. Genetic

Comparison of the Chemical Families of EO Constituents as a Function of the Collection Site (Uzana and Shipka) and Drying Prior to the EO Extraction
Overall, ketones were the major group of the EO chemical constituents, and there were no differences in their concentrations due to the location or drying of the material prior to oil extraction (Table 1). Monoterpene ketones were by far the largest subgroup, with cyclic and sesquiterpene ketones being negligible. While hydrocarbons constituted the second major EO chemical group in plants from Shipka, alcohols were the second major EO chemical group in plants from Uzana. There were significant differences between the plants from Uzana and Shipka with respect to the overall content of monoterpene and sesquiterpene hydrocarbons. Drying of the plant material prior to EO extraction also resulted in a significant shift in the chemical groups of the EO constituents; hydrocarbons and alcohols were greater in the EO from the dried material than in the EO from the fresh material (Table 1). .85 a * Means followed by the same letter within a row are not significantly different at the 5% level of significance. * A coelution between Piperitone oxide <cis-> and Piperitone on SLB-5ms column. * B coelution between 1,5-di-epi-beta-bourbonene and Elemene <β-> on SLB-5ms column. b the sum of the not identified components.

Essential Oil (EO) Composition from Dried vs. Fresh Material Collected at Mount Shipka
Drying had a significant effect on the EO composition of plant material collected at Mount Shipka. Drying of the plant material prior to oil extraction significantly affected the concentration of two major EO constituents; drying increased the concentration of limonene (10.1% vs. 6.9%) in the EO from the fresh material and the concentration of neomenthol (2.43% vs. 0.93%) but decreased the concentration of pulegone (50.5% vs. 61.2%) in the EO from the fresh material. In addition, the concentrations of Hex-(2E)-enal, sabinene, and trans-isopulegone were greater in the EO from the fresh material (Table 1).

Discussion
The EO yield of dried M. frivaldszkyana in this study (0.18% from Uzana and 0.26% from Shipka) was comparable to the one in previous reports. Overall, the EO yield in other Micromeria species was found to vary significantly, from around 0.05 up to 4% [15]. For example, the EO yields in M. cristata ssp. phrygia collected from three sites in Turkey were 0.03-0.08% [15], the oil content of M. cristata and M. juliana collected in Serbia and Montenegro was 0.1% [10], and the EO content of M. fruticosa from Israel was 0.5 to 0.72% [13].
Pulegone, menthone, and limonene are monterpenes, found as constituents in the EO of a wide range of plant species [1]. Pulegone is constituent of the EO of species from the mint family such as Nepeta cataria and Mentha piperita [20], and in Mentha pulegium, it can constitute up to 83% of the total oil [21]. Pulegone has shown insecticidal properties and is also utilized widely as flavor and fragrance agents in perfumery, cosmetics, and aromatherapy. In 2018, the United States Food and Drug Administration (FDA) [22] reconsidered the safety and toxicity of substances and withdrew six flavoring substances (including synthetic pulegone) from the GRAS (Generally Recognized As Safe) list [22]. However, this recent ruling of 2018 did not affect natural (derived from plants) pulegone. Therefore, we anticipate greater commercial demand for the sourcing of natural pulegone. The endemic plant M. frivaldszkyana, a subject of this study, would have a potential as a new source for natural pulegone, if introduced into culture.
Menthone is a major EO constituent of Mentha piperita and other mints and geraniums and is utilized widely in perfumery and cosmetics, pharmaceutical products, e-cigarettes, and other products due to its easily identifiable soothing minty scent. Menthone is a predominant EO constituent in peppermint young growing leaves [20]. At maturation, and at senescence, however, menthone is reduced to menthol and isomenthol [20,23]. On the other hand, pulegone is reduced by pulegone reductase to produce menthone and isomenthone in peppermint [20], demonstrating the close biochemical and physiological relationships between these compounds in peppermint oil glands as a function of environmental conditions, phenological phase, or the age of the individual plant leaves, and day length [23]. Further research is needed to reveal the biosynthetic processes in M. frivaldszkyana. One may speculate that the identified chemotypes in Micromeria are mostly a function of growth stage, plant part, and the environment, unless there is a side-by-side comparison of these chemotypes. Limonene, the most common terpene in nature, is constituent of citrus peel EO, and it is used extensively as a flavoring agent in the food industry and various cleaning products due to its fresh citrus aroma [24]. In addition, limonene is used as a precursor for the commercial production of carvone.
There are no previous reports on the EO composition of the endemic species M. frivaldszkyana. However, there are reports on other species from the same family [8,9,12,13,25].
In a study of Micromeria species from Bulgaria and Macedonia, the authors reported pulegone (35.8%), piperitenone (18.6), and trans-p-methane-3-one (15.8) as the main EO constituents of M. dalmatica [9]. Furthermore, caryophyllene oxide (0-14.3%), α-bisabolol (0-38.5%), geracrone (0-18.5%), and β-atlantol (0-9.9%) were the main constituents of the EO from M. cristata collected from three locations in Bulgaria and Macedonia. In addition, caryophyllene oxide (11.2%), spatulenol (5.6%), and trans-2-cren-4-ol (3.8%) were the main EO constituents of M. juliana from Macedonia [9]. In the same study, monoterpenes were the major group of EO constituents in the EO of M. dalmatica from Bulgaria and M. cristata from one of the locations in Bulgaria, while sesquitepenes were the major group in the M. cristata from the other two locations in Bulgaria and Macedonia and in M. juliana from Macedonia. In a study with M. graeca (L.) from two locations in Greece, the authors [14] reported caryophyllene oxide (17.0%) and epi-α-bisabolol (12.8%) as major constituents from one of the locations and linalool (18.1%) and β-chamigrene (12.5%) as the main EO constituents from the other location. trans-Verbenol was also relatively high in the EO of M. graeca from both locations. However, pulegone and menthone were not identified in the EO of M. graeca, whereas germacrene D constituted 0.7% and 7.5% in the EO from location 1 and 2, respectively, and limonene was 1% of the EO from one of the locations [14].
In a study on M. fruticosa in Israel, Dudai et al. [13] reported EO content (yield) variations from 0.5 to 0.72% and (+)-pulegone from 65 to 78% of the total oil as a function of day/night temperature regime and day length. Interestingly, the pulegone concentration in the leaves varied from 0% in the low base leaf pair to 71% in the tip leaf pairs, demonstrating dramatic changes within a plant and individual branch as a function of leaf age, development stage, and location. Also, the concentration of isomenthol varied from 0% in young leaves to more than 60% in older leaves, suggesting significant changes in monoterpene synthesis and accumulation during the development stages [13]. The EO content of M. fruticosa reported from Israel [13] was a bit higher than the one of M. frivaldszkyana in this study.
Kremer et al. [8] conducted a study on related species M. kerneri and M. juliana in several locations in the Balkans, including Croatia, Bosnia and Herzegovina, Montenegro, Republic of Macedonia, and Northern Greece, locations that are geographically similar and relatively close to the collection sites of M. frivaldszkyana in this study. The two Micromeria species from the above locations did not contain pulegone, menthone, or neomenthol, the major EO constituents of M. frivaldszkyana in this study. The limonene concentration in M. kerneri and M. juliana varied from 0% to 5.4%, while germacrene varied from 1.5% to 4.9% in the EO of the two species.
This literature overview of the EO composition of related species, some of which were collected in the same region, confirmed our hypothesis that the EO composition of M. frivaldszkyana is unique and different from the EO profile of other Micromeria species in the region. Therefore, the unique EO may have novel applications. M. frivaldszkyana, an endemic plant, is found only on dry rocky outcrops in the Balkan Mountains of Bulgaria; it grows on a limited amount of poor soil onto the rocks and apparently has a great ecological adaptability to environments with limited nutrients and water supply. In view of its unique chemical composition, EO yield, and ecological plasticity, the plant may have a potential as a new crop and as a commercial source for high-menthone and high-pulegone EO for the flavor and fragrance industries and possibly for the food, beverage, and pharmaceutical industries. Further research is needed to evaluate the bioactivity of the plant biomass and its EO.

Essential Oil (EO) Extraction of the M. frivaldszkyana Biomass Samples
Subsamples from the whole above ground plant parts of Micromeria frivaldszkyana (stems, leaves, and inflorescences) were submitted to hydrodistillation for EO extraction. The EO was extracted in 2 L hydro-distillation units (Laborbio Ltd., Sofia, Bulgaria, laborbio.com); each distillation was done in two replicates. The oil was measured by volume, transferred in 2 mL vials and placed in a freezer. Afterwards, the oil samples were separated from the remaining water, measured on an analytical scale, and kept in a freezer until they were analyzed. Six samples of Micromeria (2 locations × 2 replicates extracted from dried material and one sample from Mount Shipka (Balkan Mountains, Bulgaria) in two replicates was extracted fresh). Samples were stored at −3 • C. Before analysis, the samples were defrosted at room temperature. (Table S1 and Figures S1-S14)

Samples, Sample Preparation, and Gas-Chromatography/MS
All samples (10 µL) were dissolved in 990 µL of n-hexane and injected on the GC-MS and GC-FID systems.

GC-MS Analysis
The essential oil (EO) analyses were carried out on a GCMS-QP2020 (Shimadzu, Milan, Italy) equipped with a split-splitless injector, an AOC-20i autosampler, and a quadrupole MS detector. MS parameters were as follows: mass range 40-550 amu, scan speed 3333 amu/s, ion source temperature 220 • C, and interface temperature 250 • C. A SLB-5ms 30 m × 0.25 mm id × 0.25 µm film thickness column (Merck Life Science (Merck KGaA, Darmstadt, Germany)) (silphenylene polymer virtually equivalent in polarity to poly (5% diphenyl/95% dimethyl siloxane phase)) was used for the characterization of the volatile fraction. The column operated under a programmed temperature of 50 • C to 280 • C (5 min) at 3.0 • C/min. Injection volumes and mode were 1.0 µL; the split ratio was 10:1. Helium was used as a gas carrier at a constant linear velocity of 30 cm/s. The GCMS solution software (version 4.41 Shimadzu, Milan, Italy) was used for data collection and handling. A homologous series of n-alkanes (C7-C30 Saturated Alkanes, Merck Life Science, Darmstadt, Germany) standard solution has been used for Linear Retention Indices (LRIs) calculation that supported the identification of analytes on the SLB-5ms column. The peaks assignment was carried out based on a double filter, namely the MS similarity spectra (over 85%) and a LRIs ± 5 compared to the values reported in the spectral library. For mass spectral identification, Shimadzu FFNSC 3.01 was mainly used.

GC-FID Analysis
Quantitative analyses were carried out on a GC-2010 Plus (Shimadzu, Milan, Italy) equipped with a split-splitless injector (280 • C), an AOC-20i autosampler, and a flame ionization detector (FID). A SLB-5 ms 30 m × 0.25 mm id × 0.25 µm film thickness column (Merck Life Science) operated under the programmed temperature of 50 • C to 280 • C (5 min) at 3.0 • C/min. Injection volume and mode were 1.0 µL; the split ratio was 10:1. Helium was used as the carrier at a constant linear velocity of 30 cm/s and a pressure of 99.5 KPa. The FID temperature was set at 280 • C (sampling rate 60 ms), and gas flows were 40 mL/min for hydrogen, 30 mL/min for make up (nitrogen), and 400 mL/min for air. Data were processed through the LabSolution software (version 5.92 Shimadzu, Milan, Italy). Quantification was performed using the GC-FID data. Each sample was analyzed for three consecutive runs for a major precision of data.

Statistical Analysis
All data was analyzed with the one-way analysis of variance (ANOVA) to reveal significant differences between the means of each of the essential oil constituents from the two locations and between dried and fresh material using JMP Pro program (SAS Institute, Cary, North Carolina, United States). When F < 0.05, the means were compared using Tukey HSD and letter groupings were generated by the system.