Essential oil analysis of eight Nepeta taxa in Iran

. Nepeta is one of the largest and important genera of Lamiaceae that is found in many parts of the world as wild plants. These aromatic plants produce essential oil for various pharmaceutical and industrial products. The essential oil composition in eight taxa of Nepeta was analyzed. One natural population from each taxon was selected, and their essential oils extracted using Clevenger apparatus. Moreover, GC and GC/MS analysis methods allowed to reveal the variability in essential oil composition and profile among the studied taxa. The oxygenated monoterpenes were dominant in the oils of N. meyeri , N. mirzayanii , N. racemosa , N. binaludensis , and N. glomerulosa. Phytol was the major compound in the essential oil of N. kotschyi var. persica and N. saccharata (11.56% and 27.04%, respectively). 4aα,7α,7aβ-nepetalactone (73.89%) and 4aα,7β,7aα-nepetalactone (83.92%) were the major constituents in essential oil of N. mirzayanii and N. meyeri , respectively. 1,8-cineol was the principal constituent in the oil of N. glomerulosa var. carmanica , N. binaludensis , N. pogonosperma and N. racemosa (23.34%, 43.49%, 53.94% and 70.89%, respectively). The studied taxa were classified into four distinct groups according to the UPGMA tree with high level of bootstrapping support. Each group was characterized by special trait(s) that could be used for identification of them. Therefore, four chemotypes were separated among the studied taxa: 1,8-cineol, 4aα,7α,7aβ-nepetalactone, 4aα,7β,7aα-nepetalactone, and carvacrol. It was also noticed that the composition of essential oil was highly varied compared to previous results.


Introduction
Healing with medicinal plants is as old as mankind itself (Newman et al., 2000;Paterson & Anderson, 2005), while the classification of these plants is still the most difficult research problem (Singh & Geetanjali, 2018). Over the years, taxonomists suggested many different approaches toward plant classifications such as morphological, anatomical (Talebi et al., 2015), chemotaxonomic (Singh, 2016), and DNA based methods, like genetic structure (Sheidai et al., 2014) and phylogeny (Molvary et al., 1999). Singh (2016) considered the chemotaxonomy was the most important toward plant classification based on specific class of second-ary metabolites and their biosynthetic pathways. Indeed, the chemotaxonomy really can resolve specific taxonomical problems.
Nepeta L. is a multiregional genus of Lamiaceae (Mint) family. The name Nepeta originated from the ancient Italian city Nephi (Sharma & Cannoo, 2013). It has 280 species over the world, extensively grown in many regions of Asia, Africa, North America, central and southern parts of Europe. Around 30 species are found in the plains and foothills of the Indian Himalayas (Pojarkova, 1954).
Nepeta species are mostly herbaceous perennials or annuals occurring mainly in Eurasia, while this is one of the largest genera of Lamiaceae in Southwest Asia. In Iran, Nepeta is the largest genus of the family, with about 60% endemic species (Jamzad, 2009). Rechinger (1982) identified 63 species from Iran, classified into 13 sections according to their morphological characteristics. Jamzad et al. (2003), in their molecular analysis based on sequences of nuclear ITS of 34 Iranian species of Nepeta recognized two clades, one of them subdivided into four sub-clades. The phylogenetic relationships among the species of these groups are congruent with the distribution of some floral traits, e.g., corolla shape, bract texture, and exine ornamentation of pollen. Recently, Jamzad (2012) listed 79 species of Nepeta in Iran and divided them into six taxonomic groups based on morphological traits and DNA sequences.
According to Sajjadi (2005), Nepeta species have been divided into two groups with regard to the major compounds of their essential oils: one with different isomers of nepetalactone and the second that contained non-nepetalactone isomers like 1,8-cineole, β-caryophyllene, caryophyllene oxide, β-farnesene, α-citral and β-citronellol as their major constituents. The essential oils obtained from various organs of the plant, such as roots, stems, leaves and inflorescence of Nepeta taxa, have been highly contaminated with terpenoid hydrocarbons and sesquiterpene hydrocarbons, including their oxygenated analogs (Gkinis et al., 2003).
The present study aims to identify the essential oil compositions of eight Nepeta taxa and to compare our results with previous investigations and species classification according to the chemical oil compositions.

Plant materials
Eight taxa of Nepeta were collected from different parts of Iran (Table 1). The plants were sampled from wild populations at the initial stage of blooming in 2015-2016 and were identified according to Flora Iranica (Rechinger, 1982) and Flora of Iran (Jamzad, 2012)

Isolation and identification of essential oil compositions
The aerial parts of the plant samples were air-dried at room temperature in a shady place for eight days. 50 g of dried samples were hydro-distilled for 3 h using a Clevenger-type apparatus to extract the essential oil according to the European Pharmacopoeia method (Anon., 2005). Essential oil yields were calculated using the equation: ROu= (M/B m ) × 100, where (M) is the mass of the extracted oil (g) and (B m ) is the initial plant biomass (g) (da Costa et al. 2014). The extracted essential oils were dried over anhydrous sodium sulfate and stored in sealed vials at 4°C before gas chromatography (GC) and gas chromatography-mass spectrometric (GC-MS) analysis. The GC analysis was performed on an Agilent 6890 N GC system equipped with a 5975 MSD and an FID, with HP-5MS column (30 m × 0.25 mm, 0.25 mm film thickness). Injection volume was 2µl, the temperature of the injector was 200°C with a 1:10 split ratio. Helium was the carrier gas, and the rate of gas flow was 1.0 ml/min (constant flow mode). The column temperature was linearly programmed in the range of 60-280°C at the rate of 3°C / min and held at 280°C for 5 min. The transfer line was heated at 250°C, and the ionization energy for the MS detector was 70eV. The linear retention indices (LRI) of essential oil compounds were calculated by linear interpolation relative to retention times of two series of n-alkanes (C8-C20 and C21-C40) under equal chromatographic conditions. The compounds were identified by comparing their linear retention indices (LRI, with those reported in the literature (Massada, 1976;Connoll & Hill, 1991;Adams, 2007) and also by comparing their mass spectra with Wiley Registry 9 th edition Library.

Statistical analyses
After the acquisition of the data by CG-MS and CG-FID, the identified volatile substances were grouped according to their presence, absence and intensities in each of the taxa studied. From these data, each volatile substance and its respective intensity were considered the variables for the analysis of Principal Coordinate Analysis (PCA) and Unweighted Paired Group Method with Arithmetic Mean (UPGMA). In order to perform it, data were standardized (Mean of 0 and standard deviation of 1) (Higgs, 1991;Podani, 2000).
The bootstrapping analysis was used to verify the correctness of the UPGMA tree. Bootstrapping calcu-lates a support value for each node of the tree according to the fraction of samples that support the nodes. Bootstrap provides a number on each node (0-100). The highest support value is 100, while values below 70 are usually considered weak. Values below 50 are not shown and these branches are collapsed and shown as a polytomy (Tribble, 2018). We have used MVSP 2.0 (1998) and PAST 3.14 software for statistical analyses.

Results
Based on the dry weight, the yield of essential oil from eight taxa of Nepeta picked at the flowering stage ranged from 0.1 to 0.9%. The analysis of the chemical profile of the essential oils showed 106 compounds, including oxygenated monoterpenes (5.2-96.4%), oxygenated sesquiterpenes (0-18.8%), monoterpene hydrocarbons (0-14.1%), sesquiterpene hydrocarbons (0-6.5%) and the other compounds (0-74.3%). All the identified compounds, along with their percentages, appear in Table 2, where they have been ordered according to their Linear Retention Index (LRI).
The studied taxa were clustered separately in the UPGMA tree (Figure 1). The tree had two branches, N. meyeri was placed in the small branch (with 100% bootstrap support), while the big branch was composed of two sub-branches. N. mirzayanii was placed separately (by 87% bootstrap support), and other species were clustered in two groups. One group was presented by N. binaludensis, N. racemosa with N. pogonosperma (bootstrapping value of 96%), and another group was composed of N. saccharata, N. kotschyi var. persica and N. glomerulosa var. carmanica (by 95% bootstrap support). In this group N. saccharata and N. kotschyi var. persica were clustered as a pair and N. glomerulosa var. carmanica was joined them with a relatively long distance.
Furthermore, the PCA plot produced similar results (Figure 2). The X axis (Axis 2) divided the taxa into two main groups. N. meyeri and N. mirzayanii were placed far from the other taxa. These species constituents the first and second clades in the UPGMA tree. The highest amount of 4aα,7β,7aα-nepetalactone identified N. meyeri; furthermore, this species had 3-octanone and (Z)-β-ocimene in its essential oil, while these compounds were not reported from the other studied species. N. mirzayanii that made the second clade in the UPGMA tree had the highest amount of 4aα,7α,7aβ-nepetalactone; furthermore, its oil contained some minor compounds such as 3-(Z)-hexenyl benzoate, hexahydrofarnesyl acetone, and 2-octanone which were absent in the other species. In the plot, there was a clear division of two large groups by the Y-axis (Axis 1). It was the cut factor to leave N. binaludensis, N. racemosa, N. glomerulosa var. carmanica and N. pogonosperma that grouped in the third cluster of the UPGMA tree. All species were identified according to the highest amounts of 1,8-cineol and also the presence of α-pinene and γ-terpinene, which were not detected in other taxa.  Furthermore, this axis separated N. saccharata and N. kotschyi that grouped in the fourth cluster of the UPGMA tree. Both taxa had the highest amounts of phytol and carvacrol among the studied taxa. Besides, some compounds such as (E)-β-Ionone, 1-nor-Bourbonan-1-one, widdrol, octadecane and dibutyl phthalate were found in N. saccharata and N. kotschyi, while these compounds were absent in the other studied taxa.
Therefore, we have divided the studied taxa into four primary groups. N. mirzayanii and N. meyeri were characterized by high amounts of 4aα,7α,7aβ-nepetalactone and 4aα,7β,7aα-nepetalactone and formed the first and second groups, respectively. The characteristic of N. binaludensis, N. racemosa, N. glomerulosa var. carmanica and N. pogonosperma was high concentrations of 1,8-cineol, and they formed the third group. N. saccharata and N. kotschyi var. persica were characterized by high percentages of phytol and carvacrol and formed the fourth group (Figure 3). Figure 3. PCA bi-plot of the studied taxa and essential oil compositions.

Discussion
Our results of essential oil composition suggested higher variability in comparison with previous studies of these taxa. Some authors (Nejad Ebrahimi et al., 2008;Saharkhiz & Mohammadi, 2011) showed some differences in the essential oil compositions even among the populations of the same species.
We have shown that phytol (15.6%), 4aα,7α,7aαnepetalactone (13.4%), carvacrol (9.9%), citronellol (8.4%) and geranyl acetate (4.8%) were the main essential oil compounds in N. kotschyi var. persica. Nevertheless, Hadi et al. (2016) investigated essential oil compositions of 18 populations of this species in two consecutive years. In the first and second years, the oil extraction was performed in summer and spring, respectively. Their results revealed annual variation in the amount of 4aα,7α,7aα-nepetalactone, the major compound of most populations. The maximum and minimum annual changes were reported for Khorasan and Taft1 populations by 17% and 1.3%, respectively.
The annual differences in essential oil compositions may relate to harvest time. Marques et al. (2012) have suggested the seasonal fluctuations in the essential oil compositions of aromatic plants. The seasonality has a direct influence on the vegetative development and foliar biomass, which are the key points for oil production.
Moreover, amounts of 4aα,7α,7aα-nepetalactone differed among the studied populations of each year.
In the first year, highest amount of this compound was registered in Khorasan population (85.9%), while Taft 5 population had its lowest amount (0.5%). In the second year, the maximum and minimum percentages of this compound were 82.4 % (in Yazd population) and 0.3% (in Taft 5 populations), respectively. Abdelmajeed et al. (2013) believed that different physical and chemical environmental factors (e.g. temperature, hydric and osmotic stress, relative humidity, photoperiod, nutrition, and soil properties) and genetic structure could affect the quality and quantity of the essential oils. It seems that the variations in essential oil compositions among the populations each year related to the mentioned above environmental factors.
We registered that two isomers of nepetalactone, 4aα,7β,7aα-and 4aα,7α,7aβ-nepetalactone, accounted for more than 90% of essential oil of N. meyeri, while 2-metoxy-para-cresol made about of 2.59%. However, Sefidkon & Shaabani (2004) analyzed essential oil of N. meyeri (Ghom population) and considered 4aα,7α,7aβnepetalactone (53.2%) and 1,8-cineole (29.3%) as the main oil constituents. In both studies, two isomers of nepetalactone were the major compounds of the oil, which constituent more than 53% of oil composition. The Ghom population was harvested in semi-dried habitat in the central part of Iran at the altitude of about 1000 m asl, while our studied population was collected from the slopes of Albourz mountains at an altitude of 2650 m asl. Not only these populations were harvested from different habitats, but also from different altitudes, more than 1600 m in some cases.
According to Kofidis & Bosabalidis (2008), altitude is one of the most important environmental factors. Several environmental factors such as wind exposure, partial CO 2 pressure, light intensity, and ozone concentration may also differ among different altitudes. Moreover, several investigations (Giuliani et al., 2013;Sadeghi et al., 2015;Talebi et al., 2019) have revealed that altitude significantly affected the chemical composition of essential oil in Lamiaceae taxa.
In this study phytol (31.2%), carvacrol (22.4%) and n-hexadecanoic acid (9.3%) were the main compounds in N. saccharata oil, but, Rustaiyan et al. (2013) reported that the major compounds of this species were neo-isomenthol (18.6%) and n-hexadecanoic acid (12.1%). The habitat characteristics widely differed between these populations. Our studied population belongs to the Atropatenian district (northwest of Iran) that is largely influenced by the Mediterranean climate. Its thermal climate is similar to that of the eastern Mediterranean;, its climate is continental. In addition, spring arrives in late March and its precipitation tends to be moderate in amount. The studied population by Rustaiyan et al. (2013) was harvested from central Iran district (Isfahan) with continental desert climate, whose climate pattern is similar to that of central Asia desert. This region is characterized by a distinct seasonal climate, with strong, long-lasting frost in winter and extremely hot summer, in combination with very low annual precipitation of less than 100 mm. Moreover, spring arrives as soon as early March (Zohary, 1973). Although, both populations were harvested in May, because of the environmental conditions, they were in different growth phases. Esfahan population was at the end of the annual growth, while Zanjan population reaches its maximum growth rate in Mey. These conditions highly influence essential oil compositions of these populations. Dudai et al. (2001) have suggested that when growth rates are in maximal, pulegone constituted up to 80% of the essential oil in Micromeria fruticose, in a period of growth-rest, pulegone levels dropped dramatically to a few percents, while isomenthol constituted up to 80% of the essential oil.
It seems that the enhancement of n-hexadecanoic acid percentage in Isfahan population might be related to environmental temperatures. Habitat of Isfahan population is warmer than Zanjan population. Shamloo et al. (2017) have reported that a significant increase in n-hexadecanoic acid amount was recorded at an increase in growth temperature for all wheat genotypes.
A relatively high amount of dibutyl phthalate (3.3%) was found in the essential oil composition of this species. Manayi et al. (2014) have suggested that phthalate are chiefly used as plasticizers and cause several human health and also environmental hazards. Due to the problems regarding waste disposal in developing countries such as Iran, phthalate derivatives can easily release from waste disposal to the water and soil, resulting in probable absorption and accumulation by medicinal and dietary plants. Therefore, it seems that a relatively high concentration of phthalate in N. saccharata and N. kotschyi var. persica comes from environmental pollutions.
Our studied population was harvested in June at an altitude of 2100 m asl. The samples of Rustaiyan et al. (2000) and Dabiri & Sefidkon (2003) studies were harvested in June at the altitude of 3100 m asl and in September at the altitude of more than 3000 m asl, respectively. Although our study and Rustaiyan et al. (2000) populations were selected in the same month of the year, a 1000 m difference exists between their altitudes. Moreover, these populations were harvested in different years (at least 15 years difference). Nepetalactones are oxygenated monoterpenes compounds, and several reports (Niinemets et al., 2004;Figueiredo et al., 2008;Lakusic et al., 2012) have revealed that biosynthesis of terpene is influenced by different factors, primarily the genotype of species, but also by several biotic and abiotic environmental factors. Therefore, it is not unusual for populations of the same species to have various essential oil compositions. Setzer (2016) has suggested there are eight possible stereoisomers of nepetalactone, of those six stereoisomers have been found in nature as we can see, the authors registered different isomers of nepetalactone as the main parts of essential oil of this species except for 1,8-cineol.
These populations were harvested from habitats with more than 900m variation in the altitude. According to Hadi et al. (2016), percentages of nepetalactone isomers have annually and geographically variations. Moreover, altitude has a strong effect on environmental conditions and the synthesis of secondary metabolites (Talebi et al., 2019). It is important to know that cresols are released via high traffic areas, automobile exhaust, and gas stations are likely to have high levels of cresols. Moreover, these compounds are also the product of combustion of wood, municipal solid waste, and coal. So, residents near coal and petroleum-fueled facilities, as well as residents near municipal waste incinerators, may have amplified exposure to cresols. Several studies revealed that cresols are highly corrosive and toxic compounds (Tisserand & Young, 2014). It seems that a high amount of 2-metoxy-para-cresol in oil of this plant comes from environmental pollutions.
The main compound of all studies was 1,8-cineol, but with various percentages. Tounekti et al. (2011) have reported that some environmental factors such as salinity stress affect some secondary plant metabolite such 1,8-cineol. It was found that soil salinity significantly increased the relative abundance of 1,8-cineol. Based on the au-thor´s field observations, the habitat of populations in the study of Mohammadpour et al. (2013) is more saline than other populations, which could lead to a higher percentage of 1,8-cineol. However, our findings of N. binaludensis oil composition were more similar to Rustaiyan & Nadji (1999) rather than Mohammadpour et al. (2013). Three main compounds that constitute more than 71% of the oil were the same with a nearly equal amount. The populations in our study, as well as in the study of Rustaiyan & Nadji (1999), were harvested in the same month (July), while Mohammadpour et al. (2013) harvested the plant samples in August. Moreover, according to Hadi et al. (2016), the percentages of nepetalactone isomers were strongly affected by harvest time, ecological, and genetic parameters. Therefore, the similarity of essential oil compositions of these populations seems reasonable.
We have recorded that 1,8-cineol (23.3%), isobornyl acetate (6%), geraniol (5.4%), terpinen-4-ol (5.27%), and borneol (4.29%) were the main compounds in the oil of Nepeta glomerulosa subsp. carmanica. Furthermore, Sajja-di & Ghassemi (1999) analyzed the essential oil composition of Isfahan population and reported α-pinene (18.3%), 1,8-cineole (13.9%), limonene (9.7%) and linalool (4.8%) as the major compounds. We have found more than 59% variation in 1,8-cineol amounts between these populations. Against, other main compounds of Isfahan populations were found in our population in very low amounts (less than 3%). Both populations were harvested in the same month of the year, while they were selected from habitats with more than 500 m difference in altitude. Meanwhile, Zohary (1973) has suggested that the soil salinity in Kerman province (the habitat sampled in our study) is higher than Isfahan province. It seems that the observed variations in the amount of 1,8-cineol could be mainly related to soil salinity. According to Tounekti et al. (2011), salinity stress has a strong effect on 1,8-cineol amount. Diefendorf et al. (2011) have stated that chemotaxonomy of plant species is the interpretation of proxy records based on n-alkanes and the compositions of their associated carbon isotope. We calculate bootstrap percentages for all groups in UPGMA tree. All bootstrapping values were more than 87% for all groups in the tree. Burleigh et al. (2006) have suggested that the nonparametric bootstrapping methods are useful for assessing confidence in a super-tree inference. Moreover, non-parametric bootstrapping may be an over-conservative estimator of node reliability (Hillis & Bull, 1993), and values below 70 are usually considered weak. Therefore, all identified groups in the UPGMA tree of essential oil data have reliable support. As the chemical composition of essential oil varied significantly in the studied taxa, and according to UPGMA and PCA analysis, we have clustered these taxa into four groups. We also identified four chemotypes among these species: N. racemosa, N. pogonosperma, N. binaludensis and N. glomerulosa var. carmanica (with a high concentration of 1,8-cineole), N. mirzayanii and N. meyeri (due to high percentages of 4aα,7α,7aβ-nepetalactone and 4aα,7β,7aα-nepetalactone, respectively). According to the high phytol and carvacrol amounts, the separate group has been formed by N. kotschyi var. persica and N. saccharata.

Conclusions
The obtained results suggested significant variations in essential oil compositions among eight Nepeta taxa in Iran, which could be the diagnostic characteristics for their identification. The UPGMA hierarchical clustering method formed the similar five groups as the PCA plot. Furthermore, the bootstrapping analysis showed strong support for these groups. Therefore, we divided these taxa into four chemotypes based on the essential oil compositions. Each of these chemotypes had a major compound with the highest percentage with the existence of some specific minor compounds. The chemotypes were 1,8-cineol; 4aα,7α,7aβ-nepetalactone; 4aα,7β,7aα-nepetalactone; and carvacrol. We also considered that the essential oil compositions of the studied species significantly differed with results from previous similar investigations.