Phytochemical Study of the Ecuadorian Species Lepechinia mutica (Benth.) Epling and High Antifungal Activity of Carnosol against Pyricularia oryzae

The plant Lepechinia mutica (Benth.) Epling (family Lamiaceae) is endemic to Ecuador. In the present study, we report some major non-volatile secondary metabolites from the leaves and the chemistry of the essential oil distilled from the flowers. The main identified compounds were carnosol, viridiflorol, ursolic acid, oleanolic acid, chrysothol, and 5-hydroxy-4′,7-dimethoxy flavone. Their structures were determined by X-ray diffraction and NMR and MS techniques. The essential oil showed a chemical composition similar to that distilled from the leaves, but with some qualitative and quantitative differences regarding several minor compounds. The main constituents (>4%) were: δ-3-carene (24.23%), eudesm-7(11)-en-4-ol (13.02%), thujopsan-2-α-ol (11.90%), β-pinene (7.96%), valerianol (5.19%), and co-eluting limonene and β-phellandrene (4.47%). The volatile fraction was also submitted to enantioselective analysis on a β-cyclodextrin column, obtaining the separation and identification of the enantiomers for α-thujene, β-pinene, sabinene, α-phellandrene, limonene and β-phellandrene. Furthermore, the anti-fungal activity of non-volatile secondary metabolites was tested in vitro, with carnosol resulting in being very active against the “blast disease” caused by the fungus Pyricularia oryzae.


Introduction
The family Lamiaceae comprises approximately 224 genera and more than 5600 species distributed all over the world. One of these genera is Lepechinia Willd., consisting of 43 species growing from Northern California, in Western USA, to Central Argentina, in South America [1][2][3][4]. The flora of Ecuador contains 21 genera of Lamiaceae, for a total number of 135 species, 33 of which are endemic. In particular, the genus Lepechinia Willd. includes nine species, among which four are endemic to Ecuador [2,5]. Lepechinia species are perennial herbs to shrubs, rarely gynodioecious or dioecious, often
The structure of carnosol (1) was established by single crystal X-ray diffraction analysis. The crystallographic data were the same as those reported in the literature [19]: orthorhombic crystal; P21212; a = 15.762 (1), b = 13.757 (1), c = 7.7747 (7) Å, Z = 4, V = 1688.2 Å3, and R = 0.031. The structure elucidation was also supported by the comparison of the 1 H NMR, 13 C NMR, and ESI-MS data with those in the literature [20,21]. The 13 C NMR and DEPT spectra indicated the presence of four methyls, four methylenes, two methines, one oxymethine, six aromatic carbons, two quaternary carbons and one ester carbon [22,23]. The aromatic ring must be penta-substituted and one substituent was an isopropyl group, according to the 1 H NMR spectrum [20].
The pentacyclic triterpenoid acids ursolic acid (3) and the isomeric oleanolic acid (4) are widely occurring in plants in both free and glycosidic forms (saponins) [28]. They were identified by comparison of the 1 H NMR, 13 C NMR, and ESI-MS data with the literature [29][30][31][32]. 5-Hydroxy-4 ,7-dimethoxy flavone (5) was identified by comparison of the 1 H NMR, 13 C NMR, and ESI-MS data with those reported in the literature [33][34][35]. The 1 H NMR spectrum clearly indicated that compound 5 was a flavone containing a free phenolic OH group and two methoxy groups. The one-proton singlet at δ 13.60 in the 1 H NMR spectrum, exchangeable with D 2 O and assignable to a phenolic proton bridged to the carbonyl group, suggested a free phenolic OH group in position C-5 [35].

Chemical Analysis of the Volatile Fraction from Flowers
The flowers of L. mutica provided an essential oil in a yield of 0.317 ± 0.096% (w/w), referred to fresh plant material. The relative density was d 20 = 0.8526 ± 0.0127, and the refractive index n 20 = 1.4906 ± 0.0015. The specific rotation was [α] D 20 = −6.0 (c 0.08 in CH 2 Cl 2 ). The qualitative and percent compositions are reported in Table 1. About 93.9% of the volatile components have been identified and most of them quantified (see Sections 4.5 and 4.6, respectively). All constituents with an abundance <0.1% were considered as trace. The significant difference between calculated and reference linear retention indices (LRIs) [40] is within the experimental error. It can be due to different analysis conditions, carried out with a more recent instrumentation.

Enantiomeric Analysis of the Volatile Fraction from the Flowers
The results of the GC-MS analysis on a β-cyclodextrin enantioselective column are reported in Table 2, including both the enantiomeric distribution and the enantiomeric excess (ee) for each enantiomeric couple. The chromatogram of the enantioselective analysis is shown in Figure 2; only the enantiomeric couples that underwent separation have been reported. It is worthy of note that most of the detected enantiomers were baseline separated. Table 2. Enantiomeric analysis of the essential oil from flowers of Lepechinia mutica on a 30% 2,3-diethyl-6-tert-butyldimethylsilyl-β-CDX column.

Enantiomers
Retention

Biological Activity
Antifungal activities were determined for carnosol (1) and the results are shown in Table 3. Carnosol (1) showed a potent antifungal activity against the dermatophyte Mircrosporum canis (CBS 136538) and against Pyricularia oryzae (LM 120). The minimum inhibitory concentration (MIC) and minimum fungicidal concentration (MFC) values (mg/mL) determined for carnosol against P. oryzae are very close to those of the reference well-known pesticide flutriafol (0.01 mg/mL). Although this result does not validate a property of the plant extract, it constitutes a new biological property described for compound 1. These results clearly show that the activity of carnosol (1) is similar to that of flutriafol on Pyricularia oryzae but not competitive to that of intraconazol on Mircrosporum canis. More experiments are under way in this respect.

Non-Volatile Fraction
Over 71 years ago, a bitter principle was isolated from sage, Salvia carnosa Dougl. The natural substance, having the formula C 20 H 26 O 4 , was named carnosol and noted to be a diphenolic ester, containing a hydrophenanthrene moiety, presumably identical to a compound also obtained from S. officinalis L. Indeed, the correct structure was eventually deduced from chemical studies mainly performed on a sample isolated from Rosmarinus officinalis. It was also showed that carnosol was identical with pikrosalvin (picrosalvin) isolated from S. officinalis L. [23,39]. Other sources of carnosol are S. triloba L., Sphacele eremophila Boiss., S. pachyphyla, S. clevelandii, and Lepechinia hastate [39,42]. Carnosol was also isolated, as the main diterpene constituent, from the aerial parts of Sphacele chamaedryoides, and it was determined to be more cytotoxic against gastric tumour cells AGS than normal fibroblasts [43]. Moreover, it showed a significant activity against the gram (+) bacteria Staphylococcus aureus, Bacillus subtilis, Escherichia coli, and Candida albicans [44].
Carnosol showed strong inhibitory effect on the EBV-EA induction and exhibited a remarkable effect on mouse skin tumor promotion in an in vivo two-stage carcinogenesis model. Carnosol has also been shown to be partly responsible for the antioxidant and anti-tumorogenesis properties of rosemary (Rosmarinus officinalis L.). Recently, carnosol has been proved to be efficacious in the chemoprevention and/or chemotherapy of hormone-responsive cancers [45][46][47].
The cytotoxicity activity of carnosol (1) has been determined in vitro against A-549 (human lung carcinoma), HeLa (human carcinoma of the cervix), HEp-2 (human carcinoma of the larynx), and MCF-7 (human breast adenocarcinoma) cell lines, and against a Vero (African green monkey kidney) cell line [46].
Oleanolic acid (4) was isolated from more than 120 plant species, as it results from a partial survey about its occurrence in plants used in folk medicine, and its biological activity. As a typical example, both oleanolic acid (4) and ursolic acid (3) are present in Sambucus chinesis Lindl. This plant is used in folk medicine to treat inflammatory disorders and acute hepatitis. Compounds 3 and 4 were identified as the metabolite responsible for this activity [29]. Ursolic acid is chemically and pharmacologically similar to oleanolic acid. The multiple uses of oleanolic or ursolic acid containing plants in folk medicines are due to anti-inflammatory, hepatoprotective, analgesic, cardiotonic, sedative and tonic effects, among others [29].
Ursolic acid (3) was identified as an active hepatoprotective component in the preparation of Sambucus chinesis Lindl., Solanum incanum L., Tripterospermum taiwanense, and Eucalyptus hybrid. In addition to its protective effects against CCl 4 -induced liver injury, ursolic acid (3) also shows protective properties against D-galactosamine-induced liver injury in rats, and preventive effects against acetaminophen-induced cholestasis [28].
Ursolic acid (3) and oleanolic acid (4) was also isolated from Lepechinia caulescens. The vascular activity of ursolic acid (3) depended on the presence of a functional endothelium and possible NO release [29].

Volatile Fraction
The essential oil distilled from flowers of L. mutica is composed of monoterpenoid and sesquiterpenoid compounds (55.6% and 44.4% respectively). The main constituents (>4%) were: δ-3-carene (24.2%), eudesm-7(11)-en-4-ol (13.0%), thujopsan-2-α-ol (11.9%), β-pinene (8.0%), valerianol (5.2%), and co-eluting monoterpenoids limonene and β-phellandrene (4.5%). Since major constituents of the essential oil distilled from flowers are the same as the one distilled from leaves, the two chromatograms result in being qualitatively quite similar. However, quantitative analysis demonstrates that the overall composition of the two volatile fractions is actually different. In Table 1, the quantitative chemical composition of the two essential oils, expressed as percent abundance, is reported. Aliphatic monoterpens and sesquiterpens are the families showing the greatest difference, passing respectively from 24.5% to 48.9% in flowers and from 9.9% to 36.0% in leaves. LC-MS spectra were obtained with a Thermo Scientific LTQ XL Linear ION Trap mass spectrometer, equipped with an ESI source. X-ray crystallographic analysis was performed at the "Centro Grandi Strumenti (CGS)" of the University of Pavia. NMR spectra were recorded on an NMR Varian 400 MHz spectrometer or, alternatively, on Bruker 300 MHz and 200 MHz instruments. Chemical shifts (δ are given in ppm and coupling constants (J) in Hz. GC-MS and GC-FID analyses were performed with an Agilent Technologies 6890N GC, coupled to a mass spectrometry detector model 5973 inert, and to a flame ionization detector. In qualitative and quantitative GC analyses, a J&W DB-5 capillary column (30 m × 0.25 mm × 0.15 µm), purchased from Agilent Technologies, was used. GC enantioselective analysis was performed with a cyclodextrin-based capillary column (30% 2,3-diethyl-6-tert-butyldimethylsilyl-β-CDX, 25 m × 0.25 mm × 0.25 µm), purchased from Mega (Legnano, Italy).

Plant Material
The collection of L. mutica, authorized by the Ministry of Environment of Ecuador (MAE), with authorization no. 001-IC-FLO-DBAP-VS-DRLZCH-MA, was performed in the canton Quilanga, Loja province, Ecuador, in September 2009. The botanical specimen was identified by Dr. Bolivar Merino at the herbarium of the "Universidad Nacional de Loja". A voucher specimen, with number PPN-la-005, was deposited at the herbarium of the "Universidad Técnica Particular de Loja". After collection, leaves and flowers were separated and submitted to different processes. After drying in the dark at 35 • C, the leaves were submitted to solvent extraction. Fresh leaves and flowers were both steam distilled to obtain the essential oils. A quite complete study of the essential oil from leaves has recently been published [17].

Extraction and Purification of Non-Volatile Metabolites
The extract was obtained from 250 g of dried leaves of L. mutica, using EtOAc as the extraction solvent. The process was carried out statically, at room temperature, for three consecutive times, one hour each. After solvent evaporation, 24.1 g of the EtOAc extract were obtained (yield: 9.6%). The extract underwent subsequent removal of chlorophylls by filtration through a C-18 SPE column; elution was done with MeOH-H 2 O, 9:1 v/v, followed by 100% acetone. After chlorophyll removal and solvent evaporation chlorophyll, a free EtOAc extract (13.8 g) was obtained. In addition, 1 g of this extract was fractionated by medium-pressure flash chromatography on a C-18 cartridge, containing 100 g of stationary phase, with an elution gradient from methanol/water (2:1) to 100% methanol. The eluent flow was set at 30 mL/min, giving 19 main fractions. Fraction #11 (511.5 mg) was subjected to normal phase (silica gel) column chromatography. Elution with a mixture of hexane/EtOAc (97:3) gave 38.5 mg of carnosol (1). Fraction #12 and fraction #13 (58.6 mg) were collected together and submitted to silica gel column chromatography. From the elution with a mixture of hexane/EtOAc (95:5), 10.0 mg of viridiflorol (2) was obtained. Fraction #18 (79.9 mg) was submitted to C-18 reversed phase column chromatography. Elution with a mixture of MeOH/H 2 O (95:5) afforded 14.2 mg of ursolic acid (3), and 23.5 mg of oleanolic acid (4). A second amount of the EtOAc chlorophyll-free extract (11 g) was fractionated by preparative column chromatography on silica gel in order to get larger amounts of metabolites from L. mutica. Elution with an increasing polarity gradient of a hexane/EtOAc mixture afforded 16 main fractions. Fraction #11 (4.9 g) was subjected to silica gel column chromatography. Elution with an increasing polarity gradient of a hexane/EtOAc mixture gave 11.9 mg of 5-hydroxy-4 ,7-dimethoxy flavone (5), and 127.7 mg of carnosol (1). Collected fractions #13 and #14 (2.4 g) were submitted to silica gel column chromatography. Elution with an increasing polarity gradient of a hexane/EtOAc mixture gave 11 main fractions (F38-F49). Fraction #44 (158.3 mg) was also submitted to silica gel column chromatography. Elution with an increasing polarity gradient of a dichloromethane/EtOAc mixture afforded 77.7 mg of isolated chrysothol (6).

Distillation of the Volatile Fraction
Fresh flowers of L. mutica (5 kg) were divided into four equal amounts and steam was distilled in four stainless steel Clevenger-type apparatus for 4 h. After distillation, each organic layer was separated from the aqueous phase, dried over anhydrous sodium sulfate and weighted. For each distilled fraction, yield, relative density, refractive index and specific rotation were measured. The flowers were handled and distilled in the same way described for leaves [17].

Qualitative Analysis of the Essential Oil
The qualitative analysis of the essential oil was performed by GC-MS, injecting 1 µL of each distilled fraction, 1% (v/v) diluted in cyclohexane. The injector was kept at 220 • C, operating in split mode with a split ratio of 40:1. The carrier gas (He) was set at a constant flow of 1 mL/min. The analysis was performed in thermal gradient conditions, with the following temperature program: 60 • C for 5 min, increased to 110 • C at a rate of 5 • C/min, then to 148 • C at a rate of 2 • C/min. and to 250 • C at a rate of 20 • C/min, and then held at 250 • C for 2.4 min. The MS was operated in a SCAN mode, with a scan rate of 2 scan/s within a mass range of 40-350 m/z at 70 eV. For each chromatographic peak, the corresponding linear retention index (LRI) was calculated, according to Van den Dool and Kratz [41], with reference to a mixture of a homologous series of n-alkanes, from nonane to heptadecane. The constituents of the essential oil were identified, by comparing their LRIs and EI-MS spectra with data present in literature [40]. The identification was considered as acceptable in a range of ±13 units of LRI values, according to the injection of some standard compounds belonging to the different terpenic families.

Quantitative Analysis of the Essential Oil
The quantitative analysis of the essential oil was performed by GC-FID, with the same instrumental configuration and method of the qualitative analysis. The constituents were quantified by external calibration, using n-nonane as the internal standard. According to the literature [54], the response factor in FID detection is almost constant and close to value 1 within isomeric compounds, allowing the use of a single standard to quantify a family of analytes, all characterized by the same molecular formula. The same considerations are reported in reference [55]. Therefore, an isomer was used to quantify isomeric metabolites; if an isomer was not available, a structurally closely related terpene was selected. Hence, the following terpenoids (purity > 98%) were used as calibration standards: limonene for aliphatic monoterpene hydrocarbons (R 2 = 0.9962), p-cymene for aromatic monoterpene hydrocarbons (R 2 = 0.9986), linalool for monoterpene alcohols (R 2 = 0.9956), carvone for monoterpene ketones (R 2 = 0.9958), cedrene for sesquiterpene hydrocarbons (R 2 = 0.9998), and nerolidol for sesquiterpene alcohols (R 2 = 0.9997). All calibration curves were built on six points. Quantitative results were reported as the main values and standard deviations of three replicates for each distillation.

Enantioselective Analysis of the Essential Oil
The enantioselective analysis was performed by GC-MS, under the same conditions reported above except for the oven temperature that was set according to this program: initial temperature hold at 60 • C for 2 min, then increased to 220 • C at a rate of 2 • C/min, and then held at 220 • C for 2 min. The elution order of the separated enantiomers was determined by injection of enantiomerically pure standards, available in the laboratory of one of the authors (C.B.). LRIs were calculated as in qualitative analysis, according to Van den Dool and Kratz (see Section 4.5).

Antifungal Activity
Antifungal activities were expressed as the minimum inhibitory concentration (MIC) and minimum fungicidal concentration (MFC). MIC and MFC values were measured using a stereoscope and were determined for two strains of Microsporum canis (CBS 136538, from CBS-KNAW culture collection, NL), a human dermatophyte fungus, and a Pyricularia oryzae (LM120 strain, isolated by S.T. and A.M.P. from leaf lesions of Italian rice varieties), a plant pathogenic fungus. The compounds were dissolved in dimethylsulfoxide (DMSO), 1% v/v, in the liquid medium culture Sabouraud. The tested concentrations (expressed as mg/mL) were: 0.1000, 0.0500, 0.0250, 0.0125, and 0.0060.
The activities of the reference pesticides are assumed as criteria to evaluate the carnosol (1) antifungal properties.

Conclusions
In this work, the results of the first chemical investigation on both non-volatile and volatile secondary metabolites from the leaves and flowers, respectively, of L. mutica are reported. Carnosol (1), ursolic acid (3), oleanolic acid (4), viridoflorol (2), chrysothol (6), and 5-hydroxy-4 ,7-dimethoxy flavone (5) have been isolated from the leaves. Furthermore, the chemical analysis and the enantiomeric recognition of the essential oil distilled from the flowers of L. mutica are described. The qualitative compositions of the essential oils from the flowers and leaves of L. mutica have also been compared. δ-3-Carene, eudesm-7(11)-en-4-ol, thujopsan-2-α-ol, β-pinene, valerianol, limonene, and β-phellandrene were the major components of both oils, while at least 16 minor compounds can be considered qualitatively different in the two volatile fractions.
The known phenolic diterpene lactone carnosol (1) was the main secondary metabolite isolated from L. mutica. It showed a powerful antifungal activity against the plant pathogenic fungus P. oryzae and an important antifungal activity against the dermatophyte fungus Microsporum canis.