Chemical Composition and Antibacterial Activity of Essential Oils from Ferula L. Species against Methicillin-Resistant Staphylococcus aureus

Essential oils (EOs) were obtained by hydrodistillation of various parts of Ferula ovina (Boiss.) Boiss., Ferula iliensis Krasn. ex. Korovin, and Ferula akitschkensis B. Fedtsch. ex Koso-Pol., collected in the flowering/budding and fruiting stages. Eight samples of EOs isolated from F. ovina and four samples from F. akitsckensis were analyzed by gas chromatography–mass spectrometry (GC-MS). The major constituents of F. ovina EOs were α-pinene (6.9–47.8%), β-pinene (1.5–7.1%), sabinene (0.1–20.5%), β-phellandrene (0–6.5%), trans-verbenol (0.9–7.4%), eremophilene (3.1–12%), and 6Z-2,5,5,10-tetramethyl-undeca-2,6,9-trien-8-one (0–13.7%). The major constituents of F. akitsckensis EOs were α-pinene (0–46.2%), β-pinene (0–47.9%), sabinene (0–28.3%), eremophilene (0–10.6), β-caryophyllene (0–7.5%), himachalen-7-ol (0–28.2%), and an himachalol derivative (0–8.3%). Samples of EOs from F. ovina, F. iliensis, and F. akitsckensis were evaluated for antibacterial activity against methicillin-resistant Staphylococcus aureus (MRSA) pulse-field gel electrophoresis type USA300 (LAC). EOs from F. ovina exhibited the highest antibacterial activity compared to samples from other Ferula spp., with the most potent EOs being isolated from roots at the flowering and fruiting stages and stems at the fruiting stage (IC50 values of 19.1, 20.9, and 22.9 µg/mL, respectively). Although EOs demonstrated concentration-dependent inhibition of MRSA growth, analysis of the major constituents (α-pinene, β-pinene, and sabinene) showed that they had low activity, suggesting that other components were likely responsible for the observed bioactivity of the unfractionated EOs. Indeed, correlation of the GC-MS data with antibacterial activity suggested that the putative components responsible for antibacterial activity were, either individually or in combination, eremophilene and trans-verbenol. Overall, these results suggest that the EOs from F. ovina could have potential for use as alternative remedies for the treatment of infectious diseases caused by MRSA.


Introduction
Methicillin-resistant Staphylococcus aureus (MRSA) is one of the main causative agents of skin and soft tissue infections. Infections caused by MRSA have limited treatment options since these strains are resistant to the entire class of β-lactam antibiotics. Vancomycin still remains the treatment of choice for serious MRSA infections [1]; however, vancomycin must be administered intravenously, which makes administration outside of hospital or clinical settings challenging. Additionally, S. aureus strains that have vancomycin intermediate resistance are prevalent and although rare, vancomycin resistant S. aureus strains have also been isolated [2]. Thus, there is an increased interest in finding alternative methods of treatment, including natural compounds such as essential oils (EOs), that are effective against bacterial infections [3,4]. The antimicrobial properties of EOs have been reported in several studies (reviewed in [5][6][7]), and combination of antibiotics with EOs targeting multidrug resistant bacteria could lead to new choices to overcome the problem of bacterial resistance [8,9]. Thus, EOs offer promise as an alternative treatment option. Ferula spp. are a good source of biologically active compounds, such as sesquiterpenes, terpenoid coumarins, and sulfur containing compounds [10][11][12][13][14][15][16]. The genus Ferula (Apiaceae) comprises ∼185 species distributed throughout Central Asia, the Mediterranean, and northern Africa, and many species of Ferula L. have been used in traditional medicine [13,17,18]. For example, one of the plant species in our study, Ferula iliensis Krasn. ex. Korovin, is a native plant of Kazakhstan that is widely used by the local population as an anti-inflammatory treatment [19]. The main constituents of most reported EOs from Ferula spp. exhibiting antimicrobial activity are monoterpenes, oxygenated monoterpenes, sesquiterpenes, and oxygenated sesquiterpenes [20][21][22][23]. Monoterpenes and sesquiterpenes are a frequently occurring group of compounds in EOs and have a broad spectrum of pharmacological properties, including antimicrobial activity [24][25][26][27][28]. Previously, we reported the chemical composition and immunomodulatory activity of EOs isolated from Ferula iliensis and Ferula akitschkensis [12,19]. Likewise, several sulfur compounds, including sec-butyl disulfide derivatives, were found in EOs and/or ole-gum resins obtained from various Ferula spp. [15,[29][30][31][32]. The fruit oil of Ferula latisecta contains a high amount of polysulfide compounds, of which (Z)-1-propenyl sec-butyl disulfide (65.2%) and (E)-1-propenyl sec-butyl disulfide (6.8%) are the major constituents [33]. Monoterpene hydrocarbons dominated (85.7%) over all other compound groups in EOs from F. akitschkensis [12]. Ferula ovina (Boiss.) Boiss. is a fodder plant in Kazakhstan, and there is no information on the use of this plant in traditional medicine. However, F. ovina is a flavoring agent used as an ingredient in Iranian spices and condiments [34]. Aqueous extracts of F. ovina possess anti-spasmodic, anticholinergic, and smooth muscle relaxant activities [35], and antibacterial activity of F. ovina EOs against S. aureus was demonstrated by Syed et al. [36]. Radulovic et al. reported that bornyl 4-methoxybenzoate was one of the constituents of EOs from F. ovina, and it was shown that this compound induces hyperalgesia in mice [34].
In the present study, the chemical composition of EOs isolated from several samples of F. ovina and F. akitschkensis was evaluated. Antibacterial activity of EOs obtained from various parts of F. ovina, F. iliensis, and F. akitschkensis against MRSA was also assessed. Finally, three main constituents of the EOs (α-pinene, β-pinene, and sabinene) were evaluated for antibacterial activity.  Table 1). The chemical composition of two additional EOs isolated from umbels with seeds and stems at the fruiting stage of F. akitschkensis were reported previously [12]. Hydrodistillation of the umbels with seeds and stems produced 0.7 and 0.02% EOs, respectively [12]. The chemical composition of all EOs from F. iliensis were reported recently, and yields of their EOs varied from 0.4 to 1.1% [19].
Antibacterial activity of the most active samples (FOEO Rfl , FOEO Rfr , and α/β-pinenes) against MRSA was also evaluated by enumerating the number of colony-forming units (CFU). Following a 1-h incubation of bacteria with the selected EOs, the bacteria were plated on solid media and incubated overnight. FOEO Rfl and FOEO Rfr significantly inhibited growth of MRSA, even at the lowest concentrations tested (6.25 µg/mL), and only a few bacterial colonies were observed at the highest tested concentrations (100 µg/mL) ( Figure 1A). However, the individual constituents (±)-α-pinene and (−)-β-pinene demonstrated much weaker activity, even at the highest concentrations tested ( Figure 1B).
To date, more than 70 species of Ferula have been chemically investigated [56][57][58]; however, there are only a few reports on the biological activity of EOs isolated from Ferula spp. In some studies, the bacteriostatic properties of EOs from Ferula spp. were associated with a high content of α-pinene and β-pinene or polysulfides [56]. EOs from F. assa-foetida contained sulfur compounds and had antimicrobial activity against S. aureus, Staphylococcus epidermidis, Bacillus subtilis, Escherichia coli, Pseudomonas aeruginosa, and Klebsiella pneumoniae [57], while EOs from F. latisecta were active against S. aureus and Candida albicans [33]. However, disulfides exhibited much lower antimicrobial activity than other sulfur containing compounds [58]. In the present studies, EOs from F. iliensis, which also mainly contain sulfur compounds, did not demonstrate a high level of antibacterial activity against MRSA. Likewise, Iranshahi et al. reported that EOs from the fruits of F. latisecta, which have a high content of polysulfides (mainly sec-butyl-(Z)-propenyl disulfide), exhibited only moderate antibacterial activity against S. aureus (ATCC 6538p) [33]. Although there are several reports on the antibacterial activity of EOs against S. aureus (e.g., see [59][60][61]  Although there are several reports on the antibacterial activity of EOs against S. aureus (e.g., see [59][60][61]), many of these studies involved high EO concentrations and only a few studies evaluated the effects of EOs at concentrations below 50 µg/mL. For example, Yamani et al. reported that EOs from Ocimum tenuiflorum at 2.25-2.5 µg/mL had bacteriostatic activity against two S. aureus strains, including MRSA [62]. The main volatile constituents of O. tenuiflorum EOs are monoterpenes and sesquiterpenes [62]. Likewise, EOs of Aloysia polystachya at 3.64, 7.28, and 29.13 µg/mL inhibited S. aureus ATCC 25923, S. aureus ATCC 29213, and MRSA, respectively [63]. The main compounds in A. polystachya EOs are carvone (78.9%) and limonene (14.2%) [63]. Here we found that EOs from F. ovina exhibited antibacterial activity against MRSA, with FOEO Rfl , FOEO Rfr , and FOEO Sfr at concentrations of 19-22 µg/mL (Table 2). Thus, this is the first study showing effective antibacterial activity of EOs from F. ovina against a clinically-relevant MRSA strain (USA300).
Studies on the antimicrobial activity of monoterpenes showed that only the (+)-enantiomers of α-pinene and β-pinene had antibacterial activity against C. albicans, Cryptococcus neoformans, Rhizopus oryzae, and MRSA [27]. In our experiments, (±)-α-pinene and (−)-β-pinene demonstrated lower activity compared to unfractionated F. ovina EOs, and (±)-sabinene also had low activity. The highest percentage of the (+)-enantiomer of β-pinene was in FOEO Sfr . Although it could be suggested that this enantiomer was responsible for the antibacterial activity of unfractionated F. ovina EOs, some active EO samples (FOEO Rfl , FOEO Rfr , and FOEO Sfr ) had lower levels of β-pinene (1.9 and 1.5%, respectively) ( Table 1), which is not consistent with this conclusion. Additionally, α-pinene is present at high levels in F. akitschkensis EOs, yet these EOs had no antibacterial activity [12]. Thus, it is unlikely that (±)-α-pinene and (−)-β-pinene contribute significantly to the overall antibacterial activity observed.
To account for inactive EOs from F. akitschkensis, we also plotted the reciprocal values of antibacterial activity (1/IC 50 ), where inactive samples were assigned a value of zero, and obtained a good linear correlation for trans-verbenol and eremophilene (Table 3 and Figure 2A,B). Antibacterial activity also correlated with the total quantity of sesquiterpenes present in the EO samples ( Figure 2C), supporting the finding for eremophilene, an eremophilane-type sesquiterpene [65]. Moreover, various EOs isolated from Verbenaceae spp., which have a high amount of sesquiterpenes, were highly active against S. aureus (reviewed in [6]). Although we did not find a correlation with total amount of oxygenated monoterpenes (Table 3), our finding of trans-verbenol supports previous studies showing that oxygenated terpenoids may have more antimicrobial activity than some other EO constituents [66]. For the remaining major constituents, including α/β-pinenes and other chemical classes, no significant correlation between antibacterial activity and their concentrations in the EOs was found (Table 3). This is also consistent with previous studies showing that the presence of α/β-pinenes does not correlate with antimicrobial/antifungal activities [67,68]. Overall, trans-verbenol and eremophilene seem to represent reasonable targets for further analysis to define the anti-MRSA activity of the active EOs.
pinenes does not correlate with antimicrobial/antifungal activities [67,68]. Overall, trans-verbenol and eremophilene seem to represent reasonable targets for further analysis to define the anti-MRSA activity of the active EOs.   Concentration of compound(s) in EO samples are expressed as relative %. a n.s., no correlation (p > 0.05).
Unfortunately, these compounds are not commercially available and will require isolation, which is difficult due to their low concentrations, or possibly synthesis. Therefore, further studies are clearly warranted and are the focus of our ongoing research.
In general, our analysis performed using two activity representations (LogIC 50 and 1/IC 50 ) suggests that anti-MRSA activity of the EOs could be attributed to the presence of eremophilene and/or trans-verbenol and/or their additive or synergistic effect with α/β-pinenes, sabinene, and other constituents. Thus, compounds present in the greatest proportions are not necessarily responsible for the largest share of the antibacterial activity, and involvement of less abundant constituents should be considered. For example, evaluation of the major compounds of Piper hispidinervum EOs showed that a low quantity of terpinolene increased the nematicidal effect of safrole when binary combinations of these compounds were tested [69]. However, the interactive effects of major active constituents of EOs from Glossogyne tenuifolia (linalool, 4-terpineol, α-terpineol, ρ-cymene) were additive instead of synergistic, as determined by checkerboard analysis with pathogenic bacteria, including S. aureus [70].
In conclusion, we report that EOs isolated from selected Ferula species have antibacterial activity against MRSA USA300, which is a relevant clinical strain. The most active EOs were isolated from F. ovina and were characterized by an abundance of monoterpene hydrocarbons, oxygenated monoterpenes, and sesquiterpenes. On the other hand, F. iliensis EOs had low antibacterial activity, suggesting that (E)-propenyl sec-butyl disulfide and (Z)-propenyl sec-butyl disulfide do not have significant activity against MRSA. Finally, F. akitsckensis EOs possessed weak or no antibacterial activity. Although EOs from F. ovina demonstrated concentration-dependent inhibition of MRSA growth, their major constituents (α-pinene, β-pinene, and sabinene) had low activity, suggesting that they were not responsible for the observed bioactivity of the unfractionated EOs. On the other hand, correlation of the GC-MS data with antibacterial activity suggested that the sesquiterpene hydrocarbon eremophilene and the oxygenated monoterpene trans-verbenol could be the constituents responsible for antibacterial activity. Further studies are clearly necessary to evaluate efficacy and elucidate the exact mechanisms by which EOs from F. ovina exhibit their antibacterial effects.

Isolation of EOs
EOs were obtained from air-dried plant material (30-60 g depending on plant parts) by hydrodistillation for 3 h using a Clevenger-type apparatus. For the hydrodistillation, the conditions accepted by the European Pharmacopoeia (European Directorate for the quality of Medicines, Council of Europe, Strasbourg, France, 2014) were applied. The yield of EOs was calculated on a dry weight basis. Solutions of the EOs were prepared in DMSO (10 mg/mL) for antibacterial evaluation and n-hexane (10% w/v) for gas chromatographic analysis.

Gas Chromatography-Mass Spectrometry (GC-MS) Analysis
Chemical composition of the EOs was determined as reported previously [11] using GC-FID and GC-MS. GC-MS analysis was performed with an Agilent 5975 GC-MSD system (Agilent Technologies, Santa Clara, CA, USA). An Innowax FSC column (60 m × 0.25 mm, 0.25 µm film thickness) was used with He as carrier gas (0.8 mL/min). GC oven temperature was kept at 60 • C for 10 min, increased to 220 • C at a rate of 4 • C/min, kept constant at 220 • C for 10 min, and then increased to 240 • C at a rate of 1 • C/min. The split ratio was adjusted to 40:1, and the injector temperature was 250 • C. MS were collected at 70 eV with a mass range from m/z 35 to 450. GC analysis was performed using an Agilent 6890N GC system. To obtain the same elution order as with GC-MS, simultaneous injection was performed using the same column and appropriate operational conditions. Flame ionization detector (FID) temperature was 300 • C. The EO components were identified by co-injection with standards (whenever possible), which were purchased from commercial sources or isolated from natural sources. In addition, compound identities were confirmed by comparison of their mass spectra with those in the Wiley GC-MS Library (Wiley, New York, NY, USA), MassFinder software 4.0 (Dr. Hochmuth Scientific Consulting, Hamburg, Germany), Adams Library, and NIST Library. Confirmation was also achieved using the in-house "Başer Library of Essential Oil Constituents database, obtained from chromatographic runs of pure compounds performed with the same equipment and conditions. A C8-C40 n-alkane standard solution (Fluka, Buchs, Switzerland) was used to spike the samples for the determination of relative retention indices (RRI). Relative percentage amounts of the separated compounds were calculated from FID chromatograms.

Chiral GC-MS Analysis
Chromatographic separation on a chiral column was performed for α-pinene, β-pinene, and sabinene. GC-MS analysis of the enantiomers in the oil was performed with an Agilent 7890 GC equipped with a FID and 5975 MSD with a triple-axis detector and an Agilent G 4513 autoinjector, integrated with a Gerstel CIS (Gerstel, Mülheim an der Ruhr, Germany; SEM Ltd., Istanbul, Turkey). Chiral separation was performed on a Lipodex G column (25 m × 0.25 mm × 0.125 µm film thickness; Macherey-Nagel, Düren, Germany) with He as the carrier gas (65 min at 5 mL/min, average velocity 77.985 cm/s). Injection quantity was 1 µL (10% in hexane). The temperature program for separation of α-pinene, β-pinene, and sabinene enantiomers was 50 min at 35 • C and then increased 40 • C/min to 200 • C for 10.875 min. Run time was 65 min. The split ratio was adjusted to 40:1, and the injector temperature was at 250 • C. FID temperature was 250 • C.

Bacterial Strain and Culture
MRSA pulse-field gel electrophoresis type USA300 cultures were grown in TSB containing 0.5% glucose. Overnight cultures of bacteria were diluted 1:200 in 20 mL TSB in a 125 mL flask and grown at 37 • C with shaking at 250 rpm. For all experiments, cultures were grown to mid-exponential growth phase (optical density at 600 nm [OD 600 ] = 1.5).

Bacterial Growth Inhibition Assays
For analysis of antibacterial activity in culture, bacteria (2.5 × 10 7 CFU/mL) were resuspended in TSB and incubated for 4 h at 37 • C with 5 different concentrations of EOs (6.25, 12.5, 25, 50, and 100 µg/mL) or with each of the constituents (α-pinene, β-pinene, and sabinene at 31.25, 62.5, 125, 250, and 500 µg/mL) in 96-well tissue culture plates. EOs or pure compounds diluted in DMSO were added to the wells (final concentration of DMSO was 1%). DMSO was used as a negative control. The growth suppression of bacteria was monitored as absorbance (λ = 600 nm) every 5 min for 4 h using a SpectraMax 190 microplate reader. Spectinomycin was used as positive control, and 50 µg/mL of this antibiotic completely inhibited bacteria growth.
For analysis of EO or constituent effects on bacterial survival, bacteria (2 × 10 5 ) were resuspended in TSB and added to 96-well tissue culture plates with different concentrations of compounds diluted in TSB. The plates were incubated for 1 h at 37 • C, and the samples were plated onto TSA in Petri dishes. At the indicated time points, samples were serially diluted (1:10) in water, and CFU were enumerated the next day, as reported previously [71].

Statistical Analyses
The inhibitory effect of EOs against MRSA USA300 (LAC) was determined by calculation of the inhibitory concentration values (IC 50 ) as the mean ± S.D. of three independent experiments. To calculate median IC 50 , curve fitting was performed by nonlinear regression analysis of the dose-response curves generated using Prism 7 (GraphPad Software, Inc., San Diego, CA, USA). One-way analysis of variance (ANOVA) was performed on the datasets, followed by Dunnett's test. For correlation analyses, the Spearman rank correlation coefficient (r) was calculated.