Antibacterial Activity of Indian Borage ( Plectranthus amboinicus Benth ) Leaf Extracts in Food Systems and Against Natural Microflora in Chicken Meat

Indian borage (Plectranthus amboinicus Benth), also known as country borage, is a medicinal plant used widely in Indian medicine. It is a succulent, aromatic, perennial herb belonging to the family Lamiaceae (1), and decoction of its leaves is used for several medicinal purposes (2). The antimutagenic, antitumorigenic and antigenotoxic eff ects of Indian borage leaves are well documented (3), and recently its antimicrobial properties have also been reported (4–6).


Introduction
Indian borage (Plectranthus amboinicus Benth), also known as country borage, is a medicinal plant used widely in Indian medicine.It is a succulent, aromatic, perennial herb belonging to the family Lamiaceae (1), and decoction of its leaves is used for several medicinal purposes (2).The antimutagenic, antitumorigenic and antigenotoxic eff ects of Indian borage leaves are well documented (3), and recently its antimicrobial properties have also been reported (4)(5)(6).
Increase in the consumer demand for safe and non--synthetic alternatives to chemical preservatives has directed research interest in exploring extracts of various plant materials and several essential oils are being investigated for their biological potency to delay or inhibit the growth of pathogenic and/or toxin-producing microor-ganisms in food (7,8).Several plant extracts have been found to be useful as antimicrobials against spoilage and pathogenic microorganisms.However, their activity decreases considerably when added to a food system (8).Pomegranate peel extracts have been reported to increase the storage life of chicken meat (9), while grape seed and pine bark extracts were successfully used to control the growth of artifi cially inoculated bacteria in raw ground beef (10).Extracts from cinnamon bark, oregano, clove, pomegranate peel and grape seeds were also reported to decrease various microbial contaminants in raw pork during storage at ambient temperature (11).
Various antimicrobial mechanisms of natural preservatives have been hypothesized.Damage to cytoplasmic membrane and membrane proteins, leakage of intracellular contents, coagulation of cytoplasm, degradation of the ISSN 1330-9862 original scientifi c paper doi: 10.17113/ft b.54.01.16.3973 cell wall, alteration of genetic material and depletion of proton motif force are proposed mechanisms of growth inhibition of microorganisms (7,8,12).Dillenia indica extracts were reported to cause disintegration of cell wall and leakage of genetic material (13).Increase in membrane permeability and leakage of bacterial cell cytoplasm by cinnamaldehyde treatment was documented, wherein cinnamaldehyde interacts with cell wall enzymes (14,15).Phenolic compounds were reported to be the major constituents of various plant extracts, and bacterial cell wall lysis aft er treatment with phenolics was the main cause of cell death (16).
Although there are numerous studies on antibacterial properties of Indian borage leaf, they mainly focus on methanol extracts.In our earlier study, we found that extracts obtained by sequential acetone and ethyl acetate extraction had higher activity than the methanol extract, and both these extractions individually also had potential antibacterial activity in laboratory conditions (4,17).Therefore, we have used acetone and ethyl acetate extracts of Indian borage leaves to study their antibacterial effi cacy in cabbage, papaya and chicken meat, in order to evaluate their antibacterial potency under realistic conditions and establish their usefulness as food preservatives.We have also studied the eff ect of these extracts on bacterial cell wall and leakage of cell constituents to fi nd their possible antibacterial mechanism.

Plant material and chemicals
The leaves of Plectranthus amboinicus were collected from the campus of CSIR-Central Food Technological Research Institute (CFTRI), Mysore, India.The plant material was identified at Fruit and Vegetable Technology Department, CFTRI, Mysore, and a specimen voucher was deposited in the Fruit and Vegetable Technology Depart ment herbarium (FVT DH No. LGMGHCC-PA-8/ 2011).All solvents and chemicals used were of analytical reagent grade (Qualigens Fine Chemicals Pvt. Ltd., Mumbai, India), and microbiological media and sterilized plates were from HiMedia (Mumbai, India).

Plant extract preparation
The fresh leaves of P. amboinicus were washed with running tap water and dried at 55 °C in a hot air oven (Industrial and Laboratory Tools Corp., Chennai, India).The dried leaves were powdered using a mixer grinder (Johnson, Mumbai, India) and extracted with either ethyl acetate or acetone (leaf per solvent ratio of 1:4) to obtain fractions with high antibacterial activity.Flasks were placed in a shaker (60 rpm) for 24 h at ambient temperature.Aft er 24 h, the extract was fi ltered with Whatman No. 1 fi lter paper and the fi ltrate was concentrated in a fume hood.The chemical profi ling of these extracts was undertaken in our earlier study (17).Total phenolics in the acetone extract (85.15 mg of gallic acid equivalents per g of extract) were higher than in the ethyl acetate extract (67.83 mg/g), and the major phenolics present in acetone and ethyl acetate extract were quercetin (0.193 and 0.142 %), rutin (0.028 and 0.021 %), coumaric acid (0.014 and 0.010 %), caff eic acid (0.007 and 0.005 %) and gallic acid (0.003 and 0.001 %), respectively.

Microorganisms and minimum inhibitory concentration determination
Bacterial strains, namely Bacillus cereus F 4810 (Public Health Laboratory, London, UK), Staphylococcus aureus FRI 722 (Public Health Laboratory, Amsterdam, The Netherlands), and Escherichia coli MTCC 108 and Yersinia enterocolitica MTCC 859 (both from Microbial Type Culture Collection, Institute of Microbial Technology, Chandigarh, India) were grown overnight in brain heart infusion (BHI) broth at 37 °C.The minimum inhibitory concentration (MIC) of the extracts was determined by agar dilution method (in vitro) in our earlier study (17), and the values were 0.063, 0.075, 0.063 and 0.113 % of ethyl acetate extract and 0.038, 0.075, 0.063, and 0.100 % of acetone extract against B. cereus, S. aureus, E. coli and Y. enterocolitica, respectively.

Preparation of inoculum
Overnight cultures of B. cereus, Y. enterocolitica, E. coli and S. aureus were harvested by centrifugation at 2111×g for 10 min.The pellet was washed with 2 mL of phosphate buff ered saline (PBS; pH=7.4) and was again centrifuged.Then it was resuspended in 1 mL of PBS.An aliquot of 30 μL of the cell suspension was taken and made up to 3 mL using PBS.The absorbance of the suspension was read at 620 nm using UV-Vis spectrophotometer (Shimadzu, Tokyo, Japan) and adjusted to approx.0.3 (10 8 CFU/mL) with PBS in order to maintain uniformity in the diff erent sets of experiments.

Application in cabbage food model
Cabbage food model was prepared by adding fi nely shredded cabbage to the sterile deionized water (1:2, by mass per volume) and ground in a mixer for 2-3 min (18).The suspension was fi ltered using muslin cloth and pH was adjusted to 7.2 using dilute NaOH/oxalic acid.The juice (50 mL) was dispensed in 250-mL conical fl asks and autoclaved at 121 °C for 20 min.Aft er cooling, Indian borage extracts were added at diff erent concentrations (MIC or 2×MIC), and the juice was inoculated with each bacterium separately to obtain a cell density of 10 5 CFU/mL.Simultaneously, a control without extracts was used and samples were incubated at 37 °C for 30 days.The juice was withdrawn at regular intervals and appropriate dilution was pour plated onto nutrient agar plates.The colonies developed aft er incubation at 37 °C for 24 h were counted and the count was expressed in log CFU/mL.

Application in papaya food model
Aft er washing with tap water, the exterior surface of the ripe papaya fruit was cott on scrubbed with an alcoholic solution of iodine (2 %, by mass per volume) and allowed to air dry inside a laminar airfl ow cabinet.The fruit pulp without seeds (pH=5.3)was aseptically removed with sterilized spoon, transferred to a sterilized fl ask and homogenized using the spoon.Pulp (50 g) was transferred to sterilized fl asks and pasteurized at 80 °C for 1 min in order to eliminate the background microfl ora (19).Aft er cooling, the Indian borage extracts were added at diff erent concentrations (MIC or 5×MIC) and the mixture was inoculated with each bacterium separately to obtain cell density of 10 5 CFU/g.Simultaneously, a control (without extracts) was also used and all the samples were incubated at 37 °C for 7 days.The pulp was withdrawn at regular intervals and plated using appropriate dilution onto nutrient agar plates.The colonies developed aft er incubation at 37 °C for 24 h were counted and the count was expressed in log CFU/g.

Application in chicken meat system
Chicken meat system was prepared as described by Shan et al. (11).Fresh chicken meat purchased from a local market in Mysore, India was packed in presterilized polyethylene covers and transported immediately in an ice bucket to laboratory.Excess fat and accessory muscles on the surface were trimmed with a sanitized knife under aseptic conditions.The meat was cut into pieces of uniform size and placed in a sterile conical fl ask.Plant extracts (acetone and ethyl acetate) were added separately at diff erent concentrations (5×MIC or 10×MIC), mixed thoroughly and incubated for 7 days at 37 °C.Chicken samples without plant extracts were used as control.Sampling was done at regular intervals and natural microfl ora of chicken meat samples were counted by pour plating of appropriate serial dilutions onto nutrient agar.The count was expressed in log CFU/g.

Eff ect of Indian borage leaf extracts on nucleic acid leakage from bacterial cells
The eff ect of acetone and ethyl acetate extracts on nucleic acid leakage was estimated as described earlier (20).MIC of the extracts was added to 50 μL of the cell suspension (10 6 CFU), the volume was made up to 1 mL using PBS and incubated at 37 °C.At diff erent time intervals (0 (immediately aft er the addition of extracts), 15, 30 and 60 min), 50 μL of mixture were added to 1.95 mL of PBS and absorbance was measured at 260 nm against blank (PBS) in a UV-Vis spectrophotometer.For control, only cell suspension (50 μL) was added to 1.95 mL of PBS and read as above.The leakage of nuclear material to the incubating medium was measured by the change of A 260 nm at each incubation period.

Eff ect of borage extracts on bacterial cell wall
Scanning electron microscopy (SEM) was used to visualize the eff ect of extracts on bacterial cell wall (21).Overnight cultures were centrifuged at 4138×g and 4 °C for 10 min in a refrigerated centrifuge (Plasto Craft , Mumbai, India), washed twice with 0.1 M phosphate buff er (pH=6.5)and volume was made up to 0.5 mL with the same buff er.Aft er the addition of the MIC of the extracts, fi nal volume was made up to 1 mL using phosphate buffer.The above cell suspension was incubated for 1 h and the cells were harvested at 3040×g and 4 °C for 10 min.The pellet was incubated in 1 % glutaraldehyde at 0 °C overnight and the cells were harvested at 3040×g and 4 °C for 10 min.The cells were dehydrated in ethanol gradient (10-100 %) and coated with thin layer of gold using Polaron SEM coating system (Quorum Technologies Ltd, Lewes, UK).The cells were observed with a LEO 435 VP scanning electron microscope (Leo Electron Microscopy Ltd., Cambridge, UK) and images were exported in TIFF format.

Statistical analysis
All the experiments to evaluate antimicrobial effi cacy of extracts in food systems and leakage of absorbing material measured at 260 nm were done in triplicate.Statistical analysis of the results was done in Microsoft Excel (Microsoft Corp., Redmond, WA, USA).The signifi cance of the results was based on the diff erence of the mean value of the treated samples and the control using t-test at α=0.05.

Antibacterial eff ect of Indian borage leaf extracts in food models
The antibacterial eff ect of Indian borage leaf extracts was tested against E. coli, S. aureus, B. cereus and Y. enterocolitica by artifi cially inoculating them (approx.5 log CFU) in cab bage and papaya models (Figs. 1 and 2).At the same time, the extracts were used in chicken meat to study the eff ect against natural microfl ora (Fig. 3).In cabbage food model, 2×MIC treatment of extracts resulted in lower viable counts of all bacteria than in control (Fig. 1).However, no defi nite trend was observed, as MIC and 2×MIC of both extracts signifi cantly (p<0.05)decreased the counts of B. cereus, whereas only ethyl acetate extract at 2×MIC lowered the counts of E. coli throughout storage.2×MIC of both extracts caused a signifi cant (p<0.05)decrease of the count of S. aureus aft er 7 days, and 2×MIC of ethyl acetate extract decreased the count of Y. enterocolitica aft er 7 and 14 days (Fig. 1).Almost similar trend of reduction in bacterial population was observed in papaya model (Fig. 2) albeit at a higher concentration.Both extracts at 5×MIC caused a signifi cant (p<0.05)decrease in the count of Y. enterocolitica and S. aureus in papaya model throughout the storage, whereas 5×MIC of both extracts caused a signifi cant decrease of bacterial count of B. cereus up to 3 days and of E. coli up to 2 days.In case of S. aureus, 5×MIC of acetone extract was able to reduce signifi cantly the counts throughout storage, whereas ethyl acetate extract reduced the counts only up to 5 days at a similar concentration.Both extracts at 5×MIC reduced signifi cantly (p< 0.05) the counts of Y. enterocolitica throughout storage (Fig. 2).There was no signifi cant diff erence in bacterial population in MIC-and 2×MIC-treated papaya samples (data not presented for 2×MIC).In chicken meat, the count of natural microfl ora was reduced signifi cantly (p<0.05) by 5×MIC and 10×MIC of both extracts during storage (Fig. 3).No defi nite trend was observed of the reduction of bacterial counts by 5×MIC and 10×MIC of both extracts during storage.However, 5×MIC of ethyl acetate and 10×MIC of both extracts reduced the growth to approx. 1 log CFU/g aft er 5 days.

Infl uence of borage extracts on nucleic acid leakage from bacterial cells
In the present study, the leakage of 260-nm-absorbing material from bacterial cells (Fig. 4) was observed aft er treatment with MIC of ethyl acetate and acetone extracts of Indian borage leaves.The increase in A 260 nm was 60-95 % when using ethyl acetate extract and 40-80 % when using acetone extract in comparison with control.The observed leakage was signifi cantly (p<0.05)higher in the treated cells of B. cereus, Y. enterocolitica (except for 60 min with acetone extracts) and S. aureus (except for 30 and 60 min with both extracts) than that in control.However, that of the treated E. coli cells was statistically (p<0.05)similar to the control.Both extracts caused statistically (p<0.05)similar leakage of 260-nm-absorbing material from the tested bacteria at all time intervals, with the exception of Y. enterocolitica, where ethyl acetate extract caused statistically (p<0.05)higher leakage than acetone extract aft er 15 min of incubation (Fig. 4).

Degradation of bacterial cell wall by borage extracts
Treatment with extracts caused degradation of bacterial cell wall, as can be observed on the scanning electron micrographs of both extracts (Fig. 5).MIC of acetone extract was able to degrade the cell wall of B. cereus and Y. enterocolitica, but not many changes were observed in the

Discussion
In all food models (Figs.1-3), there was a marked increase in bacterial population in control samples, whereas the bacterial numbers decreased aft er treatment with ex-tracts, with certain exceptions.However, the eff ect of the extract concentration depended on the complexity of a food system.Lower inhibitory activity of the antimicrobial extracts in food models than in in vitro systems as well as diff erences in inhibitory activity of antimicrobial compounds in diff erent food models have been reported (22).Penteado and Leitão (19) also observed that Listeria monocytogenes growth was not inhibited even at low temperature storage, although a decrease in the count was observed at lower pH.It is well known that the potency of antimicrobial compounds in food systems is reduced when compared to in vitro conditions, as the presence of fats, carbohydrates, proteins, salt and pH strongly infl uences the eff ectiveness of these agents (23,24).Although chitosan, a natural preservative, exerts bactericidal eff ect against bacteria in a model system, the microfl ora of apple juice was not inactivated by it, even at higher concentrations (25).Studies on the eff ect of essential oils in fruit and vegetable products have recorded the eff ect of foodstuff s on microbial resistance (26) and att ributed it to the greater availability of nutrients in foods compared to laboratory media, which may enable bacteria to repair damaged cells faster (27).
The physical structure of a food may also limit the antibacterial effi cacy of the extracts.In the present study a higher resistance of microorganisms in papaya pulp (Fig. 2) than in the cabbage model (Fig. 1) was observed.Gutierrez et al. (22) evaluated the effi cacy of essential oils of lemon balm, marjoram, oregano and thyme on food models based on lett uce, milk and meat.They reported that food composition and structure had a signifi cant eff ect on   the growth of microorganisms.We also observed that, as the complexity of a system increased, the extent of control of microbial population decreased and higher concentration of extract was required for the control of similar bacterial populations (in cabbage 2×MIC, papaya 5×MIC, and chicken 10×MIC; Figs.1-3).A similar trend of reduction of bacterial count in chicken by sea buckthorn leaf extract was reported by Dhanze et al. (28), where the meat treated with 3 % extract was preserved for 7 days as compared to the one treated with 1 % extract (3 days) or untreated meat, which was spoiled aft er 1 day of storage.
Earlier studies have indicated loss of 260-nm-absorbing material as one of the killing mechanisms and suggested that increase in A 260 nm of incubation medium may be due to the death of the organism as a consequence of the loss of cytoplasm macromolecules (29,30).In the present study the leakage of 260-nm-absorbing material was also observed and most of the leakage occurred during initial period, followed by a slight increase with prolonging the incubation period (Fig. 4).Plant phenolics are known to interact with DNA (31,32); and Hammer et al. (33) observed signifi cant loss of 260-nm-absorbing material in Candida albicans treated with tea tree oil.The leakage of 260-nm-absorbing material from bacterial cells treated with Indian borage extracts observed in the present study may probably be due to the alteration in cell membrane caused by the presence of phenolics (17).Stojković et al. (30) also observed that protocatechuic acid, the main phenolic compound in Veronica montana water extract, inactivates L. monocytogenes by causing permeability changes in cell membrane.
The antibacterial activity of various compounds such as phenols, fl avonoids, terpenoids, coumarin and alkaloids present in natural preservatives is due to several mechanisms, including cell wall disintegration and degradation of genetic material (12).In the present study, alteration in cell structure was observed in bacteria treated with Indian borage leaf extracts (Fig. 5).Previously, Carson et al. (20) reported that the content of some cells of S. aureus was depleted aft er treatment with tea tree oil.In our study, both extracts caused rupture of cells in B. cereus and Y. enterocolitica, but ethyl acetate extract caused more structural alterations in S. aureus and E. coli.These fi ndings are similar to the earlier reported observations of decrease in the size of Listeria cells by essential oil (34), and structural alterations induced by nitrofuran (31).

Conclusions
This study showed that Indian borage leaf extracts have good antibacterial activity in diff erent food systems as they reduced the counts of artifi cially inoculated bacteria in food models (papaya and cabbage) and natural microfl ora in chicken meat.These extracts probably cause leakage of cell constituents and degradation of bacterial cell wall, thereby decreasing the population of bacteria.However, the concentrations of extracts required for reducing the bacterial counts were high in food models (2to 10-fold of MIC observed in in vitro system), therefore, the eff ect of extract addition on sensory and textural properties of foods needs a deeper study.

Fig. 4 .Fig. 3 .
Fig. 4. Eff ect of Indian borage leaf acetone and ethyl acetate extracts at MIC on the leakage of 260-nm-absorbing material from diff erent bacteria: a) B. cereus, b) E. coli, c) S. aureus and d) Y. enterocolitica.MIC=minimum inhibitory concentration