Chemical Composition and Biological Activities of Essential Oil from Salvia sclarea Plants Regenerated in vitro

The essential oils obtained by hydrodistillation of dried aerial parts of Salvia sclarea L. plants, regenerated in vitro and reproduced from seeds, were analyzed by GC and GC-MS. The oils from in vitro and in vivo plants were compared in respect to their chemical composition as well as antimicrobial and cytotoxic activities. The chemical profiles of both oils were very similar, although the yield of essential oil from in vitro plants was lower (0.1%, v/w) than the oil yield isolated from in vivo S. sclarea plants (0.2%, v/w). Both oils showed antimicrobial and cytotoxic activity. The oil from in vitro regenerated plants of S. sclarea exhibited stronger cytotoxic action against NALM-6 cell lines in comparison with the essential oil from in vivo plants.


Introduction
The genus Salvia consists of many species, which have wide applications in folk medicine and also many commercial uses, especially in the production of essential oils and flavoring agents. Salvia sclarea L. (clary sage) is an example of such species. This plant, occurring in the Mediterranean basin and Iran [1], is one of the most important aromatic plants cultivated worldwide as a source of essential oils. The essential oils or extracts of the aerial part of the S. sclarea plant have a broad spectrum of effects: analgesic, antiinflammatory [2], antioxidant, antifungal [3], and antibacterial [4][5]. Apart from the various medicinal uses, essential oils of clary sage are widely applied in the food and cosmetic industries.
In vitro propagation techniques offer a powerful tool for mass multiplication of plants with a high level of secondary metabolites [6]. In the present study, we report on the yield and chemical composition of the essential oil isolated from S. sclarea plants obtained from shoot tip culture. For comparison, the essential oil from clary sage plants propagated from seeds was also isolated and analyzed. The essential oils isolated from in vitro-and in vivo-derived plants of S. sclarea were tested for their cytotoxic and antimicrobial activity. All the plants grew in the field under the same conditions and were collected at the same stage of development. Up to now, we have found no studies on essential oils produced by micropropagated plants of S. sclarea. In the literature, there have also been no attempts to investigate the chemical composition of essential oil of Salvia sclarea cultivated under Polish climatic conditions, although several studies on the composition of essential oils of the plant species cultivated in many other countries have been published [7][8][9].

Results and Discussion
The hydrodistillation of the aerial parts of S. sclarea in vitro and in vivo plants gave yellow oils in 0.1% and 0.2 % (v/w; ml 100 g -1 dry weight) yield, respectively. The chemical composition of the essential oils was analyzed by GC and GC-MS. Thus, eighty-two constituents were identified, representing about 96% of the total oils. The compounds, together with their retention indices and relative percentage concentrations are presented in Table 1, according to the elution order on the CP Sil 5 CB column. It was found that the chemical profiles of both oils were similar. The oils comprised 21 oxygenated monoterpenes (74% and 75% in the oils from in vitro and in vivo plants, respectively), 13 monoterpene hydrocarbons (8.8%; 7.8%), 19 sesquiterpene hydrocarbons (6.6%; 4.3%), 18 oxygenated sesquiterpenes (5.8%; 6.4%), and 7 oxygenated diterpenes (1.1%; 2.2%) ( Table 1). The principal components of both oils were identified as linalool (42.3% in the oil from in vitro plants and 38.6% in the oil from in vivo plants), α-terpineol (13.4%; 14.3%), geraniol (6.3%; 7.7%), its acetate derivative (5.4%; 5.8%), and myrcene (3.3%; 3.4%). Linalool is one of the most useful monoterpene alcohols for the perfumery industry as well as for synthesis route to vitamin E [10]. Furthermore, this compound has exhibited antinociceptive [11], anticonvulsant [12] and sedative activities [13].
A major difference between the oils from in vitro and in vivo S. sclarea plants was observed only in the content of germacrene D; in the oil from micropropagated plants the amount of the sesquiterpene was almost four times higher than that found in the oil from in vivo plants. Similarity in the chemical composition of essential oils from in vitro and in vivo plants has been reported by some other authors [14][15]. For example, Fortunato and Avato (2008) [14] have reported that a chemical profile of the essential oil from Origanum vulgare L. ssp. hirtum in vitro plants was comparable to that of the control mother plants, with carvacrol as the main compound. On the other hand, comparative studies on the essential oils from in vitro and in vivo plants of Salvia przewalskii showed numerous differences between the two oil profiles [16]. We have found that the samples of oils studied by us were different from samples of oils obtained from S. sclarea plants grown for example in Greece, Spain, or Yugoslavia [7][8]. According to Souleles and Argyriadou (1997) [7], in the oil isolated from S. sclarea grown in Greece, 72 constituents were identified. Linalool was the main constituent of this oil, but its content (17.2%) was over twice lower than that detected in essential oils isolated in the present study. In contrast, linalyl acetate was found in the essential oil from Greek clary sage in a higher amount (14.3%) [7] as compared to the samples of oils analyzed in the present work (1.1-2.6%). In the essential oil from aerial parts of S. sclarea growing in Yugoslavia [8], the percentage content of geraniol (1.2%) was significantly lower than that in in vitro and in vivo clary sage plants examined in our work (6.3%; 7.7%). The sample of oil from plants growing in Spain showed an increased content of germacrene D (7.6%) [7] in comparison with oils analyzed in our laboratory ( Table 1). The changes in composition of oils could be related to climate or soil conditions. The oils from in vitro and in vivo plants of S. sclarea were evaluated for their cytotoxicity and antimicrobial activity (Tables 2 and 3). Among the bacteria tested, Escherichia coli strain was the most sensitive microorganism to clary sage essential oils (Table 2). However, the significant activity was also noted against S. aureus and S. epidermidis strains (MIC values, respectively 10.0 and 5.0 mg mL -1 ). The antimicrobial activity of the essential oils obtained from studied S. sclarea, propagated in vitro or in vivo, may well be due to the presence of synergy, antagonism or additive effects of the tested major components of the oils, which possess various potency of activity. Further studies are needed to explain which essential oils component, out of 82 main constituents demonstrated, are the most important one and is responsible for the bactericidal effects observed in this study.
As shown in Table 3 both oil samples tested exhibited cytotoxic activity against the NALM-6 and HL-60 cell lines, although the promyelocytic leukemia HL-60 cells were more sensitive to the essential oils (IC 50 = 6.5 µg/mL). The essential oil isolated from in vitro regenerated plants of S. sclarea had over 2-fold stronger cytotoxic activity against the NALM-6 cell line (IC 50 = 8.1 µg/mL) than the volatile oil obtained from in vivo clary sage plants (IC 50 = 20.1 µg/mL). This difference was statistically significant with p value less than 0.01, as calculated by Student's t-test. Higher cytotoxic activity could be explained with a higher content of germacrene D in the oil isolated from in vitro plants (Table 1). Antiproliferative activity of the sesquiterpene against the MDA-MB-231, MCF7, Hs 578T, PC-3 and Hep-G2 cell lines has been reported by Setzer et al. (2006) [17]. Germacrene D has exhibited a 7-fold stronger cytotoxic activity against the Hs 578T cell line in comparison with αpinene and limonene and a 6-fold stronger action against Hs 578T and Hep-G2 than 1,8-cineole, linalool, 4-terpineol and α-terpineol [17]. RI -Relative retention index on CP Sil 5 CB column; V -essential oil from in vitro plants; S -essential oil from in vivo plants; t -trace (percentage value less than 0.01%).

Conclusions
Our studies have shown that the yield of the oil isolated from in vitro plants of S. sclarea at the flowering stage was slightly lower to that of in vivo plants. However, the chemical profile and relative amounts of compounds were very similar in both oils. These results strongly suggest that using in vitro propagated plants of S. sclarea as a source of essential oil is possible. It is important in view of the biological activity and commercial application of the essential oil.

Plant material
According to the reported procedure [18], shoots of Salvia sclarea were multiplied from shoot tips on solid (0.7% agar) Murashige and Skoog (MS) medium [19] supplemented with indole-3-acetic acid (IAA) (0.1 mg L -1 ), 6-benzylaminopurine (BAP) (1.0 mg L -1 ) and sucrose (3%). For rooting, axillary shoots were transferred on agar MS medium without growth regulators. Cultures were maintained in the growth room at 26 ± 2 ºC and 70% humidity with photoperiod of 16 h of light/8 h of darkness. Illumination was supplied by cool white fluorescent lamps with a light intensity of 40µmol m -2 s -1 . Four-week-old rooted shoots were transferred to the pots with the sterile mixture of soil, sand and peat

Isolation of essential oil
The essential oil was obtained by hydrodistillation of the air-dried aerial parts of in vitro and in vivo plant materials (shoots and flowers) (ca 140 g dry weight) using a Clevenger-type apparatus for 5 h.

GC and GC-MS analyses
The GC analysis of the essential oil samples was performed using a Carlo-Erba Vega 6000 apparatus equipped with an FID detector and a capillary column CP Sil 5 CB (30 m × 0.32 mm i.d., film thickness 0.25 µm, Quadrex Corporation, New Haven). The oven temperature was programmed as follows: 60 -300 ºC at 4 ºC/min and the final temperature was held for 10 min; injector temperature 320 ºC; detector temperature 310 ºC; carrier gas N 2 at a flow rate of 1.5 ml min -1 . Peaks were measured by electronic integration. The percentage composition of the essential oil samples was computed from the peak areas using the normalization method. The GC-MS analysis was carried out with a Fison MD 800 mass spectrometer (ion source 200 ºC, EI 70 eV) connected to a GC 8000 gas chromatograph; helium served as the carrier gas with a flow rate of 0.8 mL min -1 . A CP Sil 5 CB column was used at the same parameters as described for GC. The identification of compounds was based on the comparison of their retention indices and mass spectra with those in commercial libraries NIST 98.1 and MassFinder 3.1. The results are presented in Table 1.

Biological activity of essential oils
For antibacterial and antifungal activity testing, three Gram-positive bacteria: Staphylococcus aureus ATCC 29213, Staphylococcus epidermidis ATCC 12228, Enterococcus faecalis ATCC 29212; two Gram-negative bacteria: Escherichia coli NCTC 8196, Pseudomonas aeruginosa NCTC 6749 and yeast Candida albicans ATCC 10231 were used. The susceptibility of microorganisms to essential oils was determined by the standard CLSI (Clinical Laboratory Standards Institute) microdilution method. Sterile stock solutions of each oil at the concentration of 160 mg mL -1 were prepared in dimethyl sulphoxide (DMSO). The oils concentration range used was 10.0-0.15 mg mL -1 , prepared for bacteria in Mueller-Hinton broth (Difco), and for yeast in RPMI-1640 medium supplemented with L-glutamine and NaHCO 3 (Biomed, Poland). Overnight grown bacterial suspensions were standarized to 10 6 CFU mL -1 (E. faecalis -10 7 CFU mL -1 ); the yeast suspension was standarized to 10 5 CFU mL -1 . One hundred µL of each suspension was added to the wells containing essential oils serial dilutions (in triplicate). The oils solubilizer was used as a growth control (no antimicrobial activity was noted). The culture medium was used as a blank control. The following standard antibacterial/antifungal agents were used: ampicillin (Serva), ofloxacin (Sigma), itraconazole (Jansen-Cilag Int., N.V.). After incubation at 37 0 C, 24 h (bacteria) and at 28 0 C, 48 h (yeast), turbidometric (OD 600 ) measures were carried out using the multifunction counter Victor2 (Wallac, Finland) and the minimal inhibitory concentrations (MICs) were specified. The results are shown in Table 2.