Comparative study on the antibaCterial aCtivity of volatiles from sage ( Salvia officinaliS l . )

Antibacterial activity of volatiles from sage against Gram-positive and Gram-negative bacteria from the ATCC collection was screened with the disk diffusion test. The essential oil and its fractions showed a significant antibacterial effect against S. aureus and B. subtilis. The minimum inhibitory concentrations were 1.25-2.5 μL/mL for S. aureus and 0.15-2.5 μL/mL for B. subtilis. The effect on S. aureus was bactericidal, while initial bactericidal effect on B. subtilis was impaired by the presence of a resistant fraction of the population, probably endospores. The results obtained with wild type and permeable strains of E. coli and S. typhimurium indicate that transport through the cell wall limits the antibacterial effect of sage volatiles.


INTRODUCTION
Plant extracts and essential oils, as well as their constituents, are used in the food, cosmetics, and pharmaceutical industries (S t a m m a t i et al. 1999).Many essential oils and their ingredients have been shown to possess diverse biological activities, including antibacterial, antifungal, and antiviral effects (O p l a c h en o v a and O b r e s h k o v a , 2003; M a r i n k o v i ć et al. 2002; S a t t a r et al. 1995).Nowadays, with the alarming incidence of antibiotic resistance in bacteria, there is a need for effective alternatives.It is therefore of interest to reexamine the way in which essential oils delay or inhibit the growth of pathogenic or food contaminating bacteria and apply them to actual practice.
A number of Lamiaceae species are aromatic plants used in traditional medicine and as culinary herbs worldwide.In traditional medicine sage was used for many ailments, including inflammation of the mouth and throat (B a r i č e v i ć et al. 2001).Traditional use of sage has been justified in a number of studies (B a r i č e v i ć et al. 2000; C a p a s s o et al. 2004; R e n et al. 2004; Đ a r m a t i et al. 1993, 1994).Moreover, antimutagenic and cancer preventive activities of sage have been reported (C r a i g , 1999; S i m i ć et al. 2000; K n e ž e v i ć -V u k č e v i ć et al. 2001).
In the present work, we compared antimicrobial properties of the essential oil of sage (Salvia officinalis L.), its fractions, and major monoterpenes against S. aureus, S. epidermidis, P. aeruginosa, E. coli, B. subtilis, and S. faecalis strains from the ATCC collection, as well as against laboratory strains of E. coli and S. typhimurium with increased permeability of their cell walls.We used the disc-diffusion test for pre-screening of the antibacterial potential of agents, the broth macrodilution method to determine the minimum inhibitory concentration (MIC), and the timekill assay to examine the kinetics of antimicrobial activity.The fractions F1-F5 were prepared by vacuum rectification and analyzed by GC-FID and GC-MS (B r k i ć et al., 1999).The results of chemical analysis show that EO contains 44 terpenoids.Fractions F1-F4 contain mainly monoterpenes, while fraction F5 is with high content of sesquiterpenes (Table 1).The α+β thujone (94.48/3.50)used was from Extrasynthese, D,L-camphor was from ICN, and 1,8-cineole was kindly provided by the «Dr.Josif Pančić" Institute for Medicinal Plant Research.

Bacteria and media
The following bacterial strains were used: Staphylococcus aureus ATCC25923, Staphylococcus epidermidis ATCC12228, Pseudomonas aeruginosa ATCC27853, Escherichia coli ATCC25922, Bacillus subtilis ATCC10707, Streptococcus faecalis ATCC29212, and Escherichia coli K12 strains SY252 and IB112 from our laboratory collection; and Salmonella typhimurium TA100 and TA102 (M a r o n and A m e s , 1983).The IB112 and Salmonella strains are with increased permeability, due to lpcA and rfa mutations, respectively.
Bacteria were cultivated at 37 o C in Luria broth (LB) (yeast extract 5 g, bacto-tryptone 10 g, NaCl 5 g, distilled water 1 L) or M u e l l e r H i n t o n broth (MHB) from Oxoid.Luria agar (LA, LB plus 15 g agar) and M u e ll e r H i n t o n agar (MHA) from Oxoid were used as solid media.The essential oils, its fractions, and monoterpenes were dissolved in ethanol (1/10) and applied in different concentrations.Ethanol was used as a negative control and antibiotics (chloramphenicol, streptomycin, and gentamycin) as positive controls.

Antibacterial activity
The disk-diffusion assay was applied to determine the growth inhibition of bacteria by sage extracts (H i n d l e r , 1995).Overnight bacterial cultures (100 μL) were spread onto MHA.Sage extracts were applied to 10 mm disks (W h a t m a n paper No.1).After 24 h of incubation at 37 o C, the diameter of growth inhibition zones was measured.

MIC determination
The broth dilution test was performed in test tubes.In two-fold serial dilutions of EO or its fractions, a standardized suspension (M c F a r l a n d turbidity standard) of test bacteria (100 μL) was added to obtain a final concentration of 5x10 5 CFU/mL.A growth control tube and sterility control tube were used in each test.After overnight incubation at 37 o C, the MIC was determined visually as the lowest concentration that inhibits growth, evidenced by the absence of turbidity (H i n d l e r , 1995).Differences of more than two steps of dilutions were considered significant (O p l a c h e n o v a and O b r e s h k o v a , 2003).

Time-kill assay
Exponential cultures of test bacteria (100 μL) are inoculated into several tubes of LB containing MIC concentration of sage extract for S. aureus and double MIC for B. subtilis.A growth control tube was used in each test.Tubes were incubated at 37 o C for 24 h.At regular time intervals, samples were taken, diluted to obtain a countable number of colonies, and plated onto LA plates.

Growth curves
Overnight culture of SY252 was diluted 20-fold in LB with or without sage EO and incubated for 5 h at 37 o C with aeration.At regular time intervals, samples were taken and optical density at 600 nm was measured using a Shimadzu spectrophotometer.

RESULTS AND DISCUSSION
In a preliminary experiment, we screened the effect of essential oil (EO) of sage (S. officinalis L.) against S. aureus ATCC25923, S. epidermidis ATCC12228, P. with different content of mono-and sesquiterpenes.The test was performed with the most sensitive bacteria in the pre-screening test: S. aureus, B. subtilis, and S. faecalis.All tested concentrations of EO and fractions showed antibacterial activity (Table 2).The largest zones of growth inhibition appeared with the highest tested concentration (30 μL/disk).Streptococcus faecalis displayed lower sensitivity to EO and fractions than S. aureus and B. subtilis; it was even resistant to F5.  2001).To test this hypothesis, we compared the antibacterial effect of EO and fractions in wild type E. coli K12 strain SY252 and its permeable counterpart IB112.The IB112 strain was more sensitive to EO and fractions than SY252 in the disk-diffusion test.Similar results were obtained when major sage monoterpenes (thujone, cineole, and camphor) were tested, although the differences in size of the growth inhibition zones of IB112 and SY252 were less pronounced (Table 2).The permeable S. typhimurium strains TA100 and TA102 showed sensitivity to EO, its fractions, and monoterpenes similar to that of IB112 (data not shown).The obtained results indicate that transport of sage volatiles through the cell wall of Gram-negative bacteria is the major process limiting their antibacterial effect.
As evident from Table 2, the strongest antibacterial effect compared to EO was detected with the F2 and F3 fractions.The chemical structure of F2 and F3 is similar, and they both contain high concentration of α-thujone (Table 1).Toxicity of thujone has been demonstrated in mice, and the concentration with thujone was in correlation to its toxicity (F a r h a t et al. 2001).Our results demonstrate that, among major sage monoterpenes, thujone is toxic to all tested bacteria (Table 2).However, the antibacterial activity of EO and fractions is not correlated with their content of thujone, cineole, or camphor, indicating that their antibacterial effect probably involves some type of synergism between many constituents.The F5 fraction showed lower antibacterial effect in all tested bacteria.This fraction F5 contains mainly sesquiterpenoids, which probably enter bacterial cells in distinct quantities.3 show differential sensitivity of S. aureus and B. subtilis.The concentration of EO and fractions required to inhibit bacterial growth were higher for S. aureus than for B. subtilis.Moreover, the MIC values for S. aureus were similar for all tested fractions, while MIC for B. subtilis varied between 0.15 μl/ml and 2.50 μl/ml, depending on the fraction applied.Lower sensitivity of S. aureus compared to B. subtilis to sage EO was also reported by other authors (C a r v a l h o et al. 1999).
To date, there has been no standard method for studying the susceptibility of microorganisms to essential oils (O p l a c h e n o v a and O b r e s h k o v a , 2003).Our results show significant differences between antibacterial activities of some fractions obtained with the disk-diffusion test and the MIC assay, probably caused by differences in water solubility and diffusion rates of the compounds, as well as by physiology of the tested bacteria.Although the MIC test is considered more accurate for quantitative evaluation of antimicrobial activity, it does not represent an absolute value either.The «true» MIC is somewhere between the lowest test concentration which inhibits the bacterial growth and the next lower test concentration.Also, inhibitory concentrations of plant extracts are higher when the incubation time is extended for 5 or 7 days (O p l a c h e n o v a and O b r e s h k o v a , 2003).
Bactericidal activity of antimicrobial agents can also be assessed by performing an in vitro time-kill assay.Table 4 presents data on the kinetics of survival of S. aureus and B. subtilis in the presence of EO and fractions (MIC for S. aureus and double MIC for B. subtilis).The EO, F2, and F3 exhibited a strong bactericidal effect on S. aureus, and within 4 hours the bacterial population was completely inactivated.Fraction F1, F4, and F5 gradually reduced the counts of S. aureus during 24 h of incubation, indicating that longer incubation or increased concentrations are needed for complete loss of viability.In contrast, EO and all tested fractions rapidly reduced the counts of B. subtilis, but after the initial reduction a constant fraction of the bacterial population survived; with EO and the F3 fraction, it even recovered after 24 h of incubation.The survival of B. subtilis is probably due to the presence of endospores, which are resistant to conditions to which vegetative cells are intolerant.Considering our results and the finding that EO of sage at a concentration of 0.35 μL/mL could not prevent the germination of B. subtilis INRA L2104 spores, but extended the lag phase of the culture by 60% (Va l e r o and S a l m e r o n , 2003), we can speculate that further incubation with EO and fractions would probably result in complete recovery of the B. subtilis population.
We also examined the effect of sage EO on the initial growth parameters of E. coli SY252 by monitoring optical density of the culture.There was no increase in the optical density of a culture containing 1 μL/mL of EO during 5 h of incubation, time sufficient for the control to enter the stationary phase (data not shown).
Obtained with different experimental methods, the presented data show a significant antibacterial effect of sage EO and its fractions.The activities demonstrated against B. subtilis, S. aureus, and E. coli, the traditional use of sage as a culinary herb, and the recently reported antigenotoxic potential of sage extracts in mice (V u j o š e v i ć and B l a g o j e v i ć , 2004) indicate that sage oil or its fractions can be considered for application in controlling food contaminations.
Salvia officinalis L. plants were cultivated in Pančevo by the «Dr.Josif Pančić» Institute for Medicinal Plant Research.Essential oil (EO) was prepared by steam distillation of plants cut at ground level in a 2 m 3 steam distiller (Hromil) for 2 hours at the pressure 3-4 bars and temperature of 135-145 o C according to Ph. Jug.IV.
aeruginosa ATCC27853, E. coli ATCC25922, B. subtilis ATCC1070 and S. faecalis ATCC29212 in disk-diffusion assay.The results showed antibacterial activity of EO (2-20 μL/disc) against all tested bacteria (data not shown).Moreover, Gram-positive bacteria were more sensitive than Gram-negative bacteria to the killing effect of EO, confirming results already reported (P a l o m b o and S e m p l e , 2001; K u d i at al. 1999; M a r i n o et al. 2001).The disk-diffusion test was further used to compare antibacterial activity of EO and its fractions (F1-F5)

A
common feature of plant volatiles is their hydrophobic nature, and the cell membrane has been proposed as the primary target of their antimicrobial action.Plant volatiles appear to accumulate in the cell membrane causing the leakage of ions, enzymes, and metabolites (I n o u e et al., 2004).It has been suggested that high resistance to plant extracts in Gram-negative bacteria is due to the outer membrane of their cell wall, acting as a barrier to many environmental substances including antibiotics (P a l o m b o and S e m p l e , 2001; K u d i at al. 1999; M a r i n o et al. Taking into account data obtained in the disk-diffusion test, we decided to determine MIC of EO and fractions for S. aureus and B. subtilis.Both species are important food-borne pathogens (P a l o m b o and S e mp l e , 2001).In addition, multidrug resistant S. aureus strains are often isolated from human clinical specimens (O p l a c h e n o v a and O b r e s h k o v a , 2003).The results reported in Table

Table 1 .
Composition of essential oil of sage and its fractions

Table 3 .
MIC values of EO and fractions F1-F5

Table 4 .
Results of the time kill assay in S. aureus and B. subtilis * MIC of EO and F1-F5.** double MIC of EO and F1-F5