Chemical Composition and Antimicrobial Activity of the Essential Oils of Three Closely Related Hypericum Species Growing Wild on the Island of Crete, Greece

The volatile compositions of three closely related Hypericum species growing wild on the island of Crete were studied, all belonging to the section Coridium. Hydro-distillation in a modified Clevenger-type apparatus was performed according to the Hellenic Pharmacopoeia in order to obtain the essential oils, which were analyzed by GC-MS. Identification of the compounds was carried out by comparison of MS spectra and retention indices with literature data, as well as by co-chromatography with authentic samples. In total, 123 different compounds were identified and the main compounds were by order of their abundance as follows: H. empetrifolium: α-pinene, germacrene D, β-pinene, E-caryophyllene; H. amblycalyx: β-elemene, β-selinene, α-pinene, E-caryophyllene, α-selinene; H. jovis: trans-calamenene, α-selinene, β-elemene. The chemical results revealed the differences and similarities (qualitative and quantitative) between the studied oils, supporting the hypothesis that essential oils from Hypericum spp. do not serve as chemotaxonomic markers. Moreover, the essential oils were subjected to antimicrobial screening. According to the given results, the essential oils possessed better antifungal and anticandidal activities than antibacterial activities.


Introduction
The genus Hypericum L. of the family Hypericaceae consists of more than 460 species divided in 36 sections [1] and has a long-term use in traditional medicine. It has been described by Hippocrates [2], Dioscorides [3], and later on in the Medieval era by Nikolaos Myrepsos [4,5]. H. perforatum (St. John's Wort) has been used as an ancient folk cure for diseases including wounds, kidney and lung ailments, and depression, while in modern times, its monograph was included in 2009 in the European Pharmacopoeia [6]. Hypericum plants are also used in foodstuff, after the evaluation of the concentration of hypericin and xanthone derivatives (Council of Europe, 2000) [7]. Externally, Hypericum is a valuable healing and anti-inflammatory remedy for burns, wounds, sores, bruises, and other skin disorders [6].

Hydro-Distillation of Essential Oils
In order to obtain the EOs, the air-dried plant material from each sample was finely comminuted and separately subjected to hydro-distillation for 3 h. About 45 g from each plant in 500 mL of distilled water were used and the distillation was performed in a modified Clevenger-type apparatus with a water-cooled oil receiver to reduce artifacts produced during hydro-distillation by overheating, according to the Hellenic Pharmacopoeia [16]. The EO yield (% v/w; Table 1) was calculated based on dry weight. GC (Gas Chromatography) grade n-pentane was used for the collection of the EOs, with the addition of anhydrous sodium sulphate to reduce any moisture. The EOs were subsequently analyzed by GC-MS (Gas chromatography-mass spectrometry analysis) and finally stored at −20 °C.

Gas Chromatography-Mass Spectrometry Analysis
GC-MS analyses were carried out using a Hewlett-Packard 7820A-5977B MSD system operating in EI mode (70 eV), equipped with a HP-5MS fused silica capillary column (30 m × 0.25 mm; film thickness 0.25 µm), and a split-splitless injector. The temperature program was from 60 °C at the time of the injection, raised to 300 °C at a rate of 3 °C/min and subsequently held at 300 °C for 10 min. Helium was used as a carrier gas at a flow rate of 2.0 mL/min. The injected volume of the samples was 2 µL. The analyses were carried out twice, resulting in reproducible results.

Identification of Compounds
Retention indices for all compounds were determined according to the Van der Dool approach [17], with reference to a homologous series of n-alkanes from C9 to C25. The identification of the chemical components was based on comparison of both relative retention times and mass spectra with those reported in the NIST/NBS and Wiley libraries, as well as those described by Adams [18] and other literature data. In many cases, the essential oils were subjected to co-chromatography with authentic compounds (Fluka, Sigma). Component relative percentages were calculated based on GC peak areas without using correction factors. Optical rotation values of the Eos

Hydro-Distillation of Essential Oils
In order to obtain the EOs, the air-dried plant material from each sample was finely comminuted and separately subjected to hydro-distillation for 3 h. About 45 g from each plant in 500 mL of distilled water were used and the distillation was performed in a modified Clevenger-type apparatus with a water-cooled oil receiver to reduce artifacts produced during hydro-distillation by over-heating, according to the Hellenic Pharmacopoeia [16]. The EO yield (% v/w; Table 1) was calculated based on dry weight. GC (Gas Chromatography) grade n-pentane was used for the collection of the EOs, with the addition of anhydrous sodium sulphate to reduce any moisture. The EOs were subsequently analyzed by GC-MS (Gas chromatography-mass spectrometry analysis) and finally stored at −20 • C.

Gas Chromatography-Mass Spectrometry Analysis
GC-MS analyses were carried out using a Hewlett-Packard 7820A-5977B MSD system operating in EI mode (70 eV), equipped with a HP-5MS fused silica capillary column (30 m × 0.25 mm; film thickness 0.25 µm), and a split-splitless injector. The temperature program was from 60 • C at the time of the injection, raised to 300 • C at a rate of 3 • C/min and subsequently held at 300 • C for 10 min. Helium was used as a carrier gas at a flow rate of 2.0 mL/min. The injected volume of the samples was 2 µL. The analyses were carried out twice, resulting in reproducible results.

Identification of Compounds
Retention indices for all compounds were determined according to the Van der Dool approach [17], with reference to a homologous series of n-alkanes from C 9 to C 25 . The identification of the chemical components was based on comparison of both relative retention times and mass spectra with those reported in the NIST/NBS and Wiley libraries, as well as those described by Adams [18] and other literature data. In many cases, the essential oils were subjected to co-chromatography with authentic compounds (Fluka, Sigma). Component relative percentages were calculated based on GC peak areas without using correction factors. Optical rotation values of the Eos [α] 20 D were determined at 20 • C at 589 nm in cyclohexane (cHex), Polarimeter: Perkin Elmer 341.

Microbial Cultures
The Gram-positive bacteria Bacillus cereus (human isolate), Micrococcus flavus (ATCC 10240), Staphylococcus aureus (ATCC 11632), and the Gram-negative bacteria Escherichia coli (ATCC 25922), Pseudomonas aeruginosa (ATCC 27853) and Salmonella typhimurium (ATCC 13311) were used in order to determine the potential antimicrobial activity of EOs of the three Hypericum species. For the determination of the antifungal activity, three yeast species Candida albicans (clinical isolate), Candida tropicalis (ATCC 750) and Candida krusei (clinical isolate) and four micromycetes Aspergillus fumigatus (ATCC 9197), Aspergillus niger (ATCC 6275), Penicillium funiculosum (ATCC 36839), Penicillium verrucosum var. cyclopium (food isolate) were tested for their susceptibility. The bacterial strains were cultured on solid Tryptic Soy agar (TSA), while micromycetes were cultured on solid malt agar (MA) and yeast were sustained on Sabouraud dextrose agar (SDA) medium. The cultures were sub-cultured once a month and stored at 4 • C for further utilization. All the tested microorganisms are deposited at the Mycological Laboratory, Department of Plant Physiology, Institute for Biological Research "Siniša Stankovic", University of Belgrade.

In Vitro Antibacterial and Antifungal Assays
The minimum inhibitory and bactericidal/fungicidal concentrations (MICs, MBCs and MFCs) were determined using 96-well microdilution plates with a flat bottom [19,20]. The oils were dissolved in 5% dimethylsulfoxide (DMSO) solution that contained 0.10% Tween 80 (v/v) and added appropriate medium with bacterial/fungal inoculum.
For antibacterial assay, bacterial suspensions were adjusted with sterile saline to a concentration of 1.0 × 10 5 CFU mL −1 . Afterward, the bacterial inoculum was added to this mixture in order to achieve the appropriate concentrations. The microplates were incubated for 24 h at 37 • C. The MIC/MBC values for bacteria were detected following the addition of 40 µL of p-iodonitrotetrazolium violet (INT) 0.2 mg/mL (Sigma I8377) and incubation at 37 • C for 30 min [21]. The lowest concentration with no visible growth was defined as the MBC, indicating 99.5% killing of the original inoculum. Streptomycin (Sigma-Aldrich S6501, St Louis, MO, USA) and ampicillin (Sigma-Aldrich A9393, Germany) were used as positive controls for bacteria (1 mg mL −1 in sterile physiological saline). Sterilized distilled water containing 0.1% Tween 80 and 5% DMSO was used as negative control.
The assay for antifungal activity was performed in the following manner: fungal spores/yeast cells were washed from the surface of agar plates with sterile 0.85% saline containing 0.1% Tween 80 (v/v). The spore suspension was adjusted with sterile saline to a concentration of approximately 1.0 × 10 5 in a final volume of 100 µL per well. MIC/MFC determination was performed by a serial dilution technique using 96-well plates. Microplates were incubated for 24 h at 37 • C for yeasts and for other fungi 72 h at 28 • C. Commercially available antifungal agent-ketoconazole (Zorkapharma, Šabac, Serbia) was used as a positive control. Sterilized distilled water containing 0.1% Tween 80 and 5% DMSO was used as negative control.

Antimicrobial Activity
The antibacterial and antifungal activities of the three Hypericum EOs are presented as MIC and MBC/MFC in Tables 3 and 4.

Antimicrobial Activity
The antibacterial and antifungal activities of the three Hypericum EOs are presented as MIC and MBC/MFC in Tables 3 and 4.

Discussion
The three under investigation species contained EOs in low amounts, consistent with previous studies reporting Hypericum as an EO-poor genus [9]. Despite this fact, many studies have been conducted on Hypericum spp. regarding the volatile constituents [9][10][11]; in many cases, the plant material has been collected in Greece [22][23][24][25][26]. Regarding chemotaxonomy, some controversial hypotheses exist; some researchers accept volatile constituents of the genus Hypericum as possible chemotaxonomic markers [27,28] and make use of statistical analysis to divide Hypericum spp. into taxonomic sections [24,29,30], while Crockett [9] mentioned the variability of the EOs from this genus concerning a broader distribution, taxonomic rank or phenological state. Our results are in accordance with the latter author, as both differences and similarities (qualitative and quantitative) were observed between the studied oils. The investigated Hypericum species are closely related taxa, as all belong to the section Coridium; HA and HJ are very close to HE, referred as recent derivatives of the latter and they are considered as the eastern and western member of a species pair, respectively [15]. The Hypericum taxa of the section Coridium (six species in total) are distributed in disjunct geographic areas, except from the derivative taxa of HE in Crete, while the rest of them are totally absent from Greece [15]. HA and HJ are distributed in restricted areas of Crete only, thus being narrow endemic, and this is the first report of the chemical composition of their EOs, although studies have been conducted concerning their content in phloroglucinols derivatives [31,32]. Regarding HE, its populations occupy a wider geographical range: Crete and other Greek Islands, Central Greece, Albania, W. Turkey, Cyrenaica [15,33]. Intraspecific-variability has been observed, in comparison to some previous studies using plant material collected from Mount Parnitha, Attica [21] and Kos Island, Aegean Sea [26]. In those studies, the plant material has also been subjected to hydro-distillation using a Clevenger apparatus [24,26], allowing the comparison of the chemical composition of the EOs. The main compound of the EO from HE was α-pinene (19.0%), a result that is consistent with previous studies for the dominant compound. However, both Petrakis et al. [24] and Fanouriou et al. [26] reported similarly higher percentages (approx. 35%), while the latter study also describes a high amount of iswarane (30.5%), a compound totally absent in the analysis of Petrakis et al. [21], which was revealed in small amounts (2.0%) in our study. The same compound was also identified in HA (2.3%). Generally, the most abundant group of compounds was found to be sesquiterpene hydrocarbons; however, some frequently reported constituents to other Hypericum spp., such as germacrene-D and β-caryophyllene [9,10], were either absent or revealed only in small amounts regarding HA and HJ. Such differences have also been found in the composition of H. perforatum growing wild in Greece [11]. It is noteworthy that HJ was found to be deficient in monoterpene hydrocarbons (0.1%), while this group of compounds was present in HE (36.9%) and HA (15.1%). HA and HJ were also found to be deficient in aliphatic hydrocarbons, like n-nonane, n-decane and n-undecane, which are common compounds in other Hypericum spp. [9,10]; e.g., H. caprifolium and H. myrianthum are characterized by the production of n-nonane and n-undecane, 55% and 20%, respectively [34]. Similar results have been reported for H. coris, the type species of the section Coridium [35]. Finally, H. ericoides, also belonging to the section Coridium, has been studied regarding its EO; however, these studies report quite diverse results, which could be explained by the different extraction methods that have been applied: isolation of the EO from the hexane extract of the plant [36], hydro-distillation [37], and headspace solid phase microextraction [38].
Regarding their anti-microbial potential, the obtained EOs were further subjected to the evaluation of antimicrobial activity in a variety of bacteria and fungi. The selected bacteria were some clinical and food isolates. More specifically, most food-borne illness is caused by infection of Staphylococcus aureus, Bacillus cereus, Listeria monocytogenes, Salmonella typhimurium and Escherichia coli, while S. aureus, Pseudomonas aeruginosa, E. coli give rise to some hospital-acquired infections. The tested fungi of Aspergillus and Penicillium genera are plant, animal and human pathogens, food contaminators and producers of potentially carcinogenic mycotoxins. Penicillium species can produce nephrotoxic, carcinogenic, teratogenic and immunotoxic metabolites, such as ochratoxin A, and contaminate cereals and cereal-based products [39]. Thus, in European countries, P. verrucosum, which produces this mycotoxin, is a major cause of cereal contamination [40]. Therefore, the presence of toxigenic fungi in cereals and food poses a potential risk to human and animal health. Moreover, Aspergillus fumigatus can cause invasive aspergillosis in immunocompromised patients [41]. Regarding the selected Candida species, they are responsible for the fungal infections called candidiasis. C. albicans is the cause for oral and genital infections, while C. krusei is the pathogen for fungemia (bloodstream infection).
According to given results of the antibacterial tests, EOs showed good activity against tested strains, but in variable degree. Minimal inhibitory concentrations ranged between 0.0025-0.030 mg/mL, and minimal bactericidal concentration was 0.005-0.060 mg/mL. All EOs showed higher antibacterial activity in comparison to commercial drugs Ampicillin and Streptomycin. The Gram-negative E. coli and P. aeruginosa were the most sensitive bacteria, since they were inhibited by all tested oils (MIC values in the range of 0.015-0.010 mg/mL). In contrast, the tested oils did not exhibit antibacterial activity against S. typhimurium.
The examined oils were more effective compared to ketoconazole (MIC 0.200-1.500 mg/mL and MFC 0.500-2.000 mg/mL), a commercially available fungicide used as positive control. The most sensitive fungi were the yeasts C. tropicalis and C. krusei with MIC 0.001-0.010 mg/mL and MFC 0.002-0.030 mg/mL. No remarkable activity was detected against A. niger and C. albicans.
Generally, among all tested essential oils, antimicrobial potential could be presented as following: HA > HJ > HE. The Hypericum EOs possessed better antifungal and anticandidal activities than antibacterial activities, with MIC values ranging between 0.001 mg/mL to 0.030 mg/mL, whereas fungicidal activates were within the 0.002 mg/mL−0.060 mg/mL range. Their inhibition and bactericidal/fungicidal concentrations (<100 mg/mL) indicate a good potency level as antibiotics [42], and support their use in the traditional medicine against diseases caused by different species of microorganisms. This is in accordance with previous reports on the antimicrobial activities of EOs from the genus Hypericum, and could be explained due to the presence of α-pinene, β-pinene and (E)-caryophyllene, which are known for their antimicrobial effects [11].