Antioxidant Potential, Lipid Peroxidation Inhibition and Antimicrobial Activities of Satureja montana L. subsp. kitaibelii Extracts

The antioxidant activity of different Satureja montana L. subsp. kitaibelii extracts was tested by measuring their ability to scavenge reactive hydroxyl radical during the Fenton reaction, using ESR spectroscopy. Also, the influence of these extracts on lipid peroxyl radicals obtained during lipid peroxidation of: (I) sunflower oil (37 ºC, 3h) induced by 4,4′-azobis(4-cyanovaleric acid) (ACVA) and (II) liposomes induced by 2,2′-azobis(2- amidino-propane)dihydrochloride (AAPH) was studied. n-Butanol extract had the best antioxidant activity (100% at 0.5 mg/mL in Fenton reaction system; 89.21% at 5 mg/mL in system I; 83.38% at 5 mg/mL in system II). The antioxidant activities of the extracts significantly correlated with total phenolic content. The antimicrobial activity of Satureja montana L. subsp. kitaibelii extracts was investigated. Petroleum ether, chloroform and ethyl acetate extracts expressed a wide range of inhibiting activity against both grampositive and gram-negative bacteria.


Introduction
Phenolic compounds, important constituents in many plants, have received considerable attention as potentially protective factors against cancer and heart diseases because of their antioxidant potency and their ubiquity in a wide range of commonly consumed foods of plants origin [1][2][3][4]. The presence of phenolic compounds in plants is important for normal growth development and defence against infection and injury. Also, phenolics may have an important effect on the oxidative stability and microbial safety in injured plants. Phenolic compounds possess one or more aromatic ring-bearing a hydroxyl substituents, including their functional derivatives such as esters, methoxy compounds and glycosides. Approximately 8000 naturally occurring compounds are known which include simple phenolic, phenolic acids (benzoic and cinnamic acid derivatives), lignans, lignins, coumarins, styrylpyrones, flavonoids, isoflavonoids, stilbenes, flavonolignans and tannins [5,6]. It is generally accepted that phenolic compounds behave as antioxidants as a result of the reactivity of the phenolic moiety. There are several mechanisms of antioxidant activity but it is believed that radical scavenging via hydrogen atom donation is the predominant mode. Other established antioxidant mechanisms involve radical complexing of prooxidant metals as well as quenching through electron donation and singlet oxygen quenching. Many studies have reported that phenolic compounds possess other biological activities such as antiinflammatory, antiulcer, antispasmodic, antisecretory, antiviral, antidiarrhoeal, antitumor, etc. [7].
Herbal and especially herbal extracts are very attractive not only in modern phytotherapy but also for food industry as spices and additives [8]. Winter savory (Satureja montana L., Lamiaceae) is native to the Mediterranean region, but is now found all over Europe, Russia and Turkey. In the Mediterranean region, it has been used as a culinary herb since antiquity. The leaves, flowers, and stems are used for herbal tea and, in traditional medicine homeopathic, to treat various ailments (with bactericidal, carminative, digestive, expectorant, fungicidal, laxative, antidiuretic, etc. activities) [9]. This plant contains various biologically active constituents such as essential oil [10,11], triterpenes [12] and phenolic compounds [13,14]. Essential oil as well as ethanolic extract of winter savory possesses antioxidant [15] and antimicrobial properties (antibacterial and antifungal) [16,17].
The aim of the present study was to examine the antioxidative activities of the different winter savory (Satureja montana L. subsp. kitaibelii) extracts by electron spin resonance (ESR) spectroscopy using three different reaction systems: the Fenton model system, the lipid system consisting of sunflower oil and the 4,4'-azobis(4-cyanovaleric acid) (ACVA; system I), and the liposomal model system with 2,2'-azobis(2-amidino-propane)dihydrochloride (AAPH; system II). Also, the relationship between phenolic contents and antioxidant activities of winter savory extracts was evaluated.

Results and Discussion
Phenolic compounds are very important plant constituents because they exhibit antioxidant activity by inactivating free radicals or preventing decomposition of hydroperoxides into free radicals. In our previous work, the total phenolic content in winter savory was determined by the Folin Ciocalteu method (expressed as a chlorogenic acid equivalent per g dry weight of plant), while phenolic compounds were identified and quantified by the HPLC method [14]. The results obtained by the Folin Ciocalteu method showed that the quantity of phenolic compounds in ethyl acetate (969.43 µg/g) and n-butanol (1358.14 µg/g) extracts were significantly higher (p<0.001) than in other investigated extracts. HPLC analysis proved that the ethyl acetate and n-butanol extracts contained hydroxybenzoic acid derivatives (protocatechuic, syringic and vanillic acids), hydroxycinnamic acid derivatives (caffeic, p-coumaric and ferulic acids), and flavanols ((±)-catechin and (-)-epicatechin) (Figure 1).   According to HPLC analysis, total content of phenolic acids in ethyl acetate and n-butanol extracts was 47.59 and 96.70 µg/g, respectively, while flavanols were present in ethyl acetate and n-butanol extracts in the extent of 638.76 and 438.90 µg/g, respectively. The chromatograms of petroleum ether, chloroform and water extracts showed that these extracts were poor in phenolic compounds. This fact is in accordance with the determined low value of total phenolic content in these three extracts (0 µg/g for petroleum extract, 8.36 µg/g for chloroform extract and 96.36 µg/g for water extract).
Since the main mechanism of antioxidant action in foods is radical scavenging, many methods have been developed in which the antioxidant activity is evaluated by the scavenging of synthetic radicals in polar organic solvents at room temperature. The antioxidant activity of extracts was tested by measuring their ability to scavenge the reactive hydroxyl radical during the Fenton reaction, using ESR spectroscopy. Also, the influence of these extracts on lipid peroxyl radicals obtained during lipid peroxidation of sunflower oil induced by ACVA and in the liposomal model system was studied.
Hydroxyl radicals generated by the Fenton reaction (Fe 2+ /H 2 O 2 ) in the presence of the spin trapping agent 5,5-dimethyl-1-pyroline-N-oxide (DMPO) generated a 1:2:2:1 quartet of lines with hyperfine coupling parameters (a N =a H =14.9 G) (Figure 2a). The intensity of the ESR signal, corresponding to the concentration of formed the DMPO-OH spin adducts, was changed in a dose-depend manner in the presence of different amounts of Satureja montana L. subsp. kitaibelii extracts. This activity could be due to a direct scavenging effect and/or to the inhibition of hydroxyl generation. The second mechanism occurs by ferrous ion chelation. The antioxidant activity (AA OH ) of the different concentrations of investigated extracts on the hydroxyl radical is shown in Figure 3. The nbutanol extract showed a very high AA OH , and in the investigated range of concentration (0.1-0.5 mg/mL) it changed from 42.67 to 100%. The abilities of all samples to scavenge hydroxyl radicals decreased in the order: n-butanol>ethyl acetate> water>chloroform. The petroleum ether extract did not exert any AA OH .

Concentration (mg/mL) n-Butanol Ethylacetate Water Chloroform Petrolether
The azo-initiators, such as ACVA and AAPH, used in our study, generate the alkyl or/and peroxyl radical by their thermal cleavage [18]. These radicals induce lipid peroxidation and the formation of lipid peroxyl radicals. The ability of different extracts to scavenge lipid peroxyl radicals was investigated by the ESR -spin trapping method using α-phenyl-N-tert-butylnitrone (PBN) as the spin trap.
Typical ESR spectra of the PBN-peroxyl radical (PBN-OOL) spin adduct were detected during the peroxidation of sunflower oil induced by ACVA (Figure 2b

Concentration (mg/mL) n-Butanol Ethylacetate Water Chloroform Petrolether
The antioxidant activity of ethyl acetate, n-butanol and water extracts increased dose-dependently at mass concentrations ranging from 1 to 5 mg/mL. The highest antioxidant activity on the peroxyl radical (AA LOO,I =89.21%) was observed for the n-butanol extract of Satureja montana L. subsp. kitaibelii at 5 mg/mL (Figure 4), while the same concentration of ethyl acetate extract exhibited somewhat lower activity (AA LOO,I =80.47%). Butylated hydroxyanisole (BHA) eliminated 57.69% of the peroxyl radicals at 5 mg/mL (data not shown). Petroleum ether and chloroform extracts did not show any antioxidant activity.
The high lipid content of cell membranes make them one of the main targets of the reactive oxygen species that can be generated in vivo by both enzymatically and non enzymatically. The incubation of liposomes with hydrophilic AAPH induced lipid peroxidation, as measured by the accumulation of the very reactive lipid peroxyl radical and resulting PBN-OOL spin adduct exhibits a characteristic signal (Figure 2c). The influence of different concentrations of the extracts of Satureja montana L. subsp. kitaibelii on peroxyl radical formation during the AAPH-induced lipid peroxidation of liposomes is presented in Figure 5.

Concentration (mg/mL) n-Butanol Ethylacetate Water Chloroform Petrolether
It is evident that the antioxidant activity (AA LOO,II ) on peroxyl radicals during the AAPH-induced lipid peroxidation of liposomes depends on the type and concentration of the investigated extracts. Only at higher concentrations of applied ethyl acetate, n-butanol and water extracts, the scavenging effects on peroxyl radicals in the investigated system were similar to the results of the scavenging effects on hydroxyl radicals. Generally, the antioxidant activity of ethyl acetate, n-butanol and water extracts increased dose-dependently at mass concentrations ranging from 1 to 5 mg/mL. The highest antioxidant activity on the peroxyl radical (AA LOO,II =83.38%) was observed for the n-butanol extract of Satureja montana L. subsp. kitaibelii at 5 mg/mL ( Figure 5), while the same concentration of ethyl acetate extract exhibited somewhat lower activity (AA LOO,II =72.35%). BHA eliminated 57.69% of peroxyl radicals at 5 mg/mL (data not shown). Petroleum ether and chloroform extracts did not show any antioxidant activity.
The relationship between the total content of phenolic compounds obtained by the Folin-Ciocalteu method and the HPLC method were correlated with antioxidant activities (Figure 6). Total Phenol content (µg/g)

AA (%)
Peroxyl radical assay I Peroxyl radical assay II Hydroxyl radical assay The phenolic content, obtained by the Folin-Ciocalteu method and the HPLC method, and antioxidant activities are very closely correlated for the all cases. The correlation coefficient between total phenolic content, determined by the Folin Ciocalteu method, and AA was very good (in the case of hydroxyl radicals r 2 =0.87; in the case of peroxyl radicals in system I and system II, r 2 =0.84 and r 2 =0.9, respectively). Lower ones were obtained between the total phenolic content obtained by HPLC and AA (in the case of hydroxyl radicals r 2 =0.62; in the case of peroxyl radicals in system I and system II, r 2 =0.79 and r 2 =0.64, respectively).
The relationship between structure and activity can be proposed from this series of the phenolics. Generally, phenolic compounds with o-dihydroxyl group in aromatic ring possess stronger antioxidant activity than monophenolics [19]. For example, caffeic acid reacted better with oxygen centered radicals than p-coumaric acid. Also, methoxy substitution of the hydroxyl group in the ortho position of the diphenolics, as in ferulic or vanillic acid, resulted in to a decrease in the free radical scavenging reaction [20]. According to literature data, (+)-catechin and (-)-epicatechin possess marked free radical scavenging activity [21]. In both compounds the presence of o-dihydroxyl group in B-ring is fundamental for their scavenging properties, but m-hydroxyl groups at C 5 and C 7 in ring A also contribute to scavenging activity.
The values of antimicrobial activity of petroleum ether, chloroform, ethyl acetate, n-butanol and water extracts of Satureja montana L. subsp. kitaibelii and standards are shown in Tables 1 and 2.  The maximum antibacterial activity (MIC=10 mg/mL) was shown by the ethyl acetate extract against Pseudomonas aeruginosa and Staphylococcus aureus, and by the chloroform extract against Pseudomonas aeruginosa and Bacillus cereus. The petroleum ether extract remarkably inhibited the growth of Staphylococcus aureus; Sarcina lutea and Bacillus cereus, while n-butanol were weakly active against Staphylococcus aureus and Bacillus cereus. On the other hand, a very low antimicrobial activity (MIC ≥100 mg/mL) of all extracts was found against the tested yeasts.
Phenolic compounds are known to be synthesized by plants in response to microbial infection. It is therefore logical that they have been found in vitro to be effective antimicrobial substances against a wide array of microorganisms [22]. The structure-activity relationships of the antimicrobial activity of phenolic compounds are contradictory. For example, Chabot et al. [23] reported that less polar compounds, i.e. flavonoids lacking hydroxyl groups on their B ring, are more active against microorganisms than those with OH groups. On the other hand, Mori et al. [24] found that a free 3',4',5'-trihydroxy B-ring and a free 3-OH were necessary for antibacterial activity. Based on this, it can be concluded that the antimicrobial activity of investigated extracts depend not only on phenolic compounds. The presence of different secondary metabolites also contributed to the antimicrobial activity. An additional research on chemical composition of each extract is required for comprehensive assessment of individual compounds exhibiting antimicrobial activity.

Conclusion
The results of ESR analysis indicate that ethyl acetate, n-butanol and water extracts have significant antioxidant activities against hydroxyl and peroxyl (obtained in two different systems) radicals, which is in correlation with the contents of total phenolic compounds. The following order of antioxidant activities (AA OH , AA LOO,I and AA LOO,II ) has been established: n-butanol > ethyl acetate > water. Satureja montana L. subsp. kitaibelii extracts were also observed to possess antibacterial activity depending on the solvent used and microorganisms tested. All of these results show that the Satureja montana L. subsp. kitaibelii extracts can be used accessible sources of natural antioxidants, as a possible food supplements and as a phytochemicals.

Preparation of Extracts
The dried aerial parts (moisture content 3.72%) of winter savory were ground in a grinder (DLFU Bühler Miag Laboratory Disc Mill, Germany) with a 2 mm in diameter mesh. This material (100 g) was macerated with 70 % methanol in water (2 x 2500 mL) at room temperature for 2 x 24 h. The obtained extract was concentrated under reduced pressure to remove methanol, and then extracted sequentially with petroleum ether (2 x 200 mL), chloroform (2 x 200 mL), ethyl acetate (2 x 200 mL) and n-butanol (2 x 200 mL). The solvent extractions were performed in separation funnels, shaking for 10 min. The petroleum ether, chloroform, ethyl acetate, n-butanol and remaining water extract were evaporated to dryness under reduced pressure at 40°C with a water bath.

Oxidation of Sunflower Oil Induced by Azo-initiator ACVA
Oxidation was investigated in a solution containing 1 mL of sunflower oil, 0.0018g ACVA, 0.0274 g of α-phenyl-N-tert-butylnitrone (PBN) and 380 µL of various concentrations of methanolic solutions of the investigated extracts (final concentrations were in range 1-5 mg/mL). The solutions were incubated at 37 o C for 3 h under aerobic conditions.

Oxidation of the PC Liposomes Induced by Azo-initiator AAPH
Liposome suspension was prepared by adding 1 g of commercial liposomes to 10 mL water and incubated with stirring at 37 o C for 1h. Oxidation was investigated in the solution containing 1 mL of liposome suspension, 0.0019 g AAPH, 0.0274 g of PBN and 380 µL of various concentrations of methanolic solutions of the investigated extracts (final concentrations were in the range 1-5 mg/mL). The solutions were incubated at 37 o C for 3 h under aerobic conditions. Lipid oxidation in both systems was measured by following the formation of peroxyl radicals with ESR spin trapping method. The ESR spectra of the formed PBN-peroxyl radical spin adducts were recorded with the following spectrometer settings: field modulation 100 kHz, modulation amplitude 1.021 G, receiver gain 5⋅10 5  Magnetic field scanning determinations were calibrated using Fremy's salt. Splitting constants were calculated from computer-generated second derivatives of the spectra, after optimizing signal-to-noise ratios, and were verified by computer simulations.

Antimicrobial activity test
For these investigations the disc diffusion method and microbroth dilution methods were applied. From the primary isolation medium 2-3 colonies of investigated bacteria were taken by flamed loop, suspended in Mueller-Hinton or Sabouraud broth, and they were incubated at 37 o C. The suspension for inoculation was prepared from the broth cultures. The number of cells in 1 mL of suspension for inoculation measured by the McFarland nephelometer was 1x10 7 cfu/mL. A volume of 1 mL of this suspension was homogenized with 9 mL of melted (45 o C) Mueller-Hinton or Sabouraud dextrose agar and poured into Petri dishes.
For screening, sterile 6 mm discs (HiMedia, Mumbai, India) were impregnated with 10 µL of 100 mg/mL of Satureja montana L. subsp. kitaibelii extracts. After incubation for 24-48 hours in the thermostat, inhibition zone diameters, ZI, (including disc) were measured and expressed in mm. The presence of the inhibition zone indicates the activity of tested extracts against bacteria or yeasts. The minimum inhibitory concentration (MIC) was reported as the lowest concentration of the extracts capable of inhibiting the growth of the bacterium tested. The MIC was determined by the broth macrodilution test [25] using an inoculum of 1x10 7 cfu/mL. Final concentrations of the investigated extracts were 6.25, 12.5, 25, 50, 75 and 100 mg/mL. Both tests were done in three replications.
In parallel with the antibacterial investigation of Satureja montana L. subsp. kitaibelii extracts, pure solvent was tested too, and it did not exhibit any antibacterial activity (data are not shown). Bacteria were obtained from the stock cultures of the Microbiology Laboratory, Faculty of Technology, University of Novi Sad.

Statistical analysis
All measurements were carried out in triplicate, and presented as mean ± SME or ± SD. Regression analyses and significance of differences were carried out using a SPSS Statistical Software package (SPSS for Windows, 8.0, 1997, SPSS Inc., Chicago, IL, USA).