Study of the effect of aqueous garlic extract ( Allium sativum ) on some Salmonella serovars isolates

There has been a consistent increase in the search for alternative and efficient compound for food conservation, aiming a partial or total replacement of antimicrobial chemical additives. Garlic offers a promising alternative for food safety and bioconservation. Filter sterilized, aqueous garlic extract was tested for ability to inhibit the growth of some isolated Salmonella serovars. The fresh aqueous garlic extract (A.G.E., 57.1% (w/v), containing 324 μg/ml allicin) inhibited the growth and killed most of the tested Salmonella serovars. The effect of bacteriostatic concentration of A.G.E. on the growth of the different tested serovars, revealed a pattern of inhibition characterized by: (i) a transitory inhibition phase whose duration was proportional to A.G.E concentration (ii) a resumed growth phase which showed a lower rate of growth than in uninhibited controls, and (iii) an entry into stationary phase at a lower culture density. The minimal inhibitory concentration and minimum bactericidal concentrations were very close, garlic MIC range 10-12.5 mg/ml; MBC range 13-15 mg/ml. Garlic extract could be stored at 4°C because no detectable loss of antibacterial activity at this temperature over several days was observed. However, excessive warming, or longer periods at higher temperatures should be avoided. Among enzymatic activities followed with the API-ZYM system, significant changes during the inhibition phase were detected. These biochemical changes could represent an adaptative response towards the garlic stress.


Introduction
The increasingly high numbers of bacteria that are developing resistance to classical antibiotics (Levy, 1997;Cohen, 1992;Walsh, 2000;Kiessling et al., 2002;Cole et al., 2002;Teuber, 1999) drive much of the current interest on plant antimicrobial molecules in hope that they may provide useful leads into antiinfective drug candidates.Several antimicrobial agents were isolated from plant including secondary metabolites as essential oil and Terpenoides, amongst which can be cited xanthones, benzophenones, coumarins and flavonoids (Nkengfack et al., 2002;Ouahouo et al., 2004;Komgeum et al., 2005) Garlic (Allium sativum) has traditional dietary and medicinal applications as an anti-infective agent (Ross et al., 2001).Garlic is a common food spice widely distributed and used in all parts of the world as a spice and herbal medicine for the prevention and treatment of a variety of diseases, ranging from infections to heart diseases (Rivlin, 2001).Garlic is thought to have various pharmacologic properties and medical applications.It is mainly consumed as a condiment in various prepared food (Amagase et al., 2001).
The main antimicrobial constituent of garlic has been identified as the oxygenated sulphur compound, thio-2propene-1-sulfinic acid S-allyl ester, which is usually referred to as allicin.Allicin is produced catalytically when garlic cloves are crushed and the enzyme allinase (alliin lyase E.C. 4.4.1.4) of the bundle sheath cells mixes with its substrate, alliin, which is released from mesophyll cells (Miron et al., 2000;Curtis et al., 2004).
Allicin is one of the principals of freshly crushed garlic homogenates, is a volatile molecule that is poorly miscible in aqueous solutions liquid, responsible for the pungent smell of garlic and is chemically an unstable and highly reactive molecule.Allicin is a short lived molecules, this rather unstable compound has been suggested by Lawson and coworkers to transform rapidly into secondary products (In vivo) such as allylmercaptan and others (Koch and Lawson 1996;Lawson and wang, 1993).
Moreover, since massive generation of allicin could also be toxic for the plant tissues and enzymes, its very limited production and short-lived activity, which is confined to the area where the microbial attack takes place, minimizes any potential cell-damage to the plant (Ankri and Mirelman, 1999).
Allicin undergoes thiol-disulphide exchange reactions and can react with free thiol groups in proteins (Ankri and Mirelman, 1999;Miron et al., 2002).SHcontaining enzymes so far shown to be inhibited by allicin include: succinic dehydrogenase, urease, papain, xanthine oxidase, choline oxidase hexokinase, cholinesterase, glyoxylase, triose phosphate dehydrogenase, alcohol dehydrogenase, and cysteine proteases (Ankri et al., 1997;Wills, 1956).Additionnally, Focke et al. (1990) provided evidence for specific inhibition of acetyl-CoA synthetase (E.C.6.2.1.1)by allicin which occurred by a non-covalent, reversible binding of allicin to the enzyme.In contrast to others reports of enzymes inhibition by allicin this effect could not be competed out by thiol reagents such as dithioterythritol (Focke et al., 1990).Allicin's reactivity with enzymes, and its radical-trapping properties and ready membrane permeability, are regarded as the basis of its biological activity (Miron et al., 2000;Rabinkov et al., 1998).
The anti-microbial effects is due to the chemical reaction of the allicine with the thio groups of several enzymes such as ARN Polymerase, by delaying and inhibiting DNA, RNA and protein synthesis (Ankri andMirelman, 1999, Feldberg et al., 1988) The search for new antibacterial agents associated with specific plant families should be continued, and recent focus has shifted to determining the antimicrobial activity of plant extracts used in folk medicine (Rios and Recio, 2005).The screening of plant extracts and plant products for antimicrobial activity has shown that higher plants are a potential source of novel antibiotic substitutes (Rios and Recio, 2005).
Although much has been reported on the medicinal properties of garlic and allicin (Ali et al., 2000;Ankri and Mirelman, 1999;Singh et al., 1998), not much is known about its proteinaceous constituents (Terras et al., 1993;Van Damme et al., 1993).Lixin and Ng (2005) has isolated an antifungal protein from garlic, designated alliumin, with a molecular mass of 13 kDa.Alliumin presents antifungal activity against Mycospharella arachidicola, inhibitory activity to the bacterium Pseudomonas fluorescences and exerted antiproliferative activity toward leukemia L1210 cells (Lixin and Ng, 2005).
Various garlic preparations have been shown to exhibit a wide spectrum of antibacterial activity against Gramnegative and Gram-positive bacteria including such species as Esherichia, Salmonella, Streptococcus, Staphylococcus, Klebsiella, Proteus, and Helicobacter pylori (Ankri and Mirelman, 1999;Small et al., 1947).Even acid-fast bacteria such us Mycobacterium tuberculosis are sensitive to garlic (Uchida et al., 1975).
Salmonella serovars was chosen as the Gram-negative model organism, as it is one of the major human pathogens and food poisoning cases (Reed, 1993).It's well demonstrated the Salmonella can disseminate and survive in various environmental niches for a long periods of time.They are pervasive in nature and many contaminate animals, vegetables, water and especially food during its production and distribution (Davies and Wray, 1996).
The aim of this study was first to investigate if the antimicrobial activity exhibited by the fresh aqueous garlic extract against six serovars of Salmonella, isolated from Tunisian foods, coproculture and wastewater, is just due to the allicin or to the presence of others bioactive molecules.Second, to study the garlic cell growth effects and third to evaluate the bioresponses of treated bacteria, particularly biochemical changes induced by bacteriostatic aqueous garlic extract concentrations.

Bacteria and growth conditions
Salmonella serovars included in this study were recovered from different samples, purchased from different region in Tunisia (Table 1).
A total of six Salmonella serovars, isolated from Tunisian fast foods, coproculture and wastewater were used in this study.These were Salmonella cerro, Salmonella enteritidis, Salmonella lindenburg, Salmonella montevideo, Salmonella hadar and Salmonella nikolaifleet which are also listed in Table 1.Long-term storage of Salmonella samples was at -20°C in sterile glycerol (15%).A preculture was prepared by transfer from this culture to fresh sterile liquid medium (Pronadisa, Hispanlab) and cultivated for 18 h at 37°C with shaking.

Inhibition studies
Overnight cultures of the six serovars of Salmonella were grown in liquid medium and diluted into 50 ml of fresh sterile broth medium.Aqueous garlic extract was added at various concentrations directly to the flask and turbidity was monitored by measuring the optical density at 600 nm of the medium.

Preparation of aqueous garlic extracts (A.G.E.)
Fresh garlic cloves (70 g) were blended in 35 ml sterile distilled water, centrifuged at 5000 rpm and sterilised by filtration (0.45µm).By subtracting the weight of the insoluble material from the weight of the original cloves, the final concentration of the garlic extract in solution was determined to be 57.1% (w/v).Aliquots were stored at -20°C until required.The concentration of allicin in each preparation was determined spectrophotometrically by reaction with the thiol, 4-mercaptopyridine (Miron et al., 2002).Briefly, varying quantities of garlic extract were incubated with 4mercaptopyridine (10 -4 mM) in 50mM phosphate buffer, 2 mM EDTA pH 7.2 which results in the formation of a mixed disulphide, 4-allylmercaptothiopyridine and the consequent shift in absorbance at 324 nm was monitored after 30 min incubation period at room temperature.

Minimal inhibitory concentration (MIC) and Minimal bactericidal concentration (MBC) determination by broth dilution
For MIC and MBC determinations, serial twofold dilutions of aqueous garlic extract in 5 ml of broth inoculated with 50 µl of fresh precultures (inoculum, ∼10 8 CFU/ml).The tubes were incubated at 37°C overnight with shaking and the highest dilution in which there were no growth was recorded as the MIC.For MBC testing, aliquots (20 µl) of broth from tubes containing no growth were plated onto solid medium and again incubated overnight at 37°C.The highest dilution in which there were no survivors was recorded as the MBC.In the above method, controls for each organism were performed using the sterile liquid medium without aqueous garlic extract.All MICs and MBCs were confirmed by triplicate assays.

Stability and heat inactivation of garlic extract
To assess the effects of storage on the antimicrobial activity in garlic extract an aliquot was kept at room temperature while another sample was stored at 4°C.Afterwards, the residual antibacterial activity was measured.
To study heat-inactivation, 1 ml aliquots of extract in 1.5 ml Eppendorf tubes were exposed to various temperatures for 10 min in a heating block, cooled to room temperature and 20 µl plated in the standard assay to assess the effect on antimicrobial activity.

Effect of garlic extract on some enzyme activity
The changes of different cellular enzymatic activities of treated or untreated bacteria, collected from the inhibition and the exponential growth phase culture, was evaluated using the API-ZYM system (BioMérieux).The cells were harvested, washed twice with cold 5. 10 -2 mol L -1 phosphate buffer (pH 7) and resuspended in 0.85% (w/v) sterile NaCl solution.
Each API-ZYM gallery was inoculated with two drops of adequate suspension of cells and incubated for 4 h at 28°C.The galleries were then activated by adding one drop of ZYM A and ZYM B reagents and after 5 min, values ranging from 0 to 5 in relation to the color developed in each enzymatic reaction, were visually assigned by means of the color chart supplied with the system.

Statistical analysis
Data were analyzed by ANOVA using a SAS system procedure (SAS Institute, Cary, NC, USA).A multiple comparison test (least significant difference) was used to test for significant differences between the treatment means (P< 0.05).

Salmonella growth
The aqueous garlic extract (57.1% (w/v), containing 324 µg/ml allicin) inhibited the growth and killed most of the tested salmonella serovars.In order to study the effect of the A.G.E., different concentrations were tested.No effect on the different cell growth curves was observed when we added an A.G.E. to a final concentration less than 11 mg/ml (mean 89.1 µg/ml of allicin).So, we had studied the effect of garlic extract added to a diluted final concentration, which range from 11 mg/ml to 13 mg/ml.
Compared with the control cell suspensions without garlic extract, we observed a modification of the classical cell growth curves of the different salmonella serovars.Figure 1 shows the basic phenomena observed in the inhibition of cell growth by aqueous garlic extract.
The addition of A.G.E. to these final concentrations at cell density of 0.05 induces the apparition of an inhibition phase in all the Salmonella growth curves.After this inhibition phase, cell growth resumed but at rate inferior to the control cell suspensions.In addition, cultures exposed to A.G.E.entered stationary phase at a cell density substantially lower than that of the control culture.This phenomenon was observed in all cases of A.G.E.treated Salmonella.
The effect of increasing the A.G.E.concentration on bacterial growth is indicated in Figure 1.As the aqueous garlic extract concentration was increased from 11 to 13 mg/ml (means 89.1 to 97.2 µg/ml of allicin), the duration of the inhibition phase increased, the rate of the growth after inhibition decreased, and the cell density at which stationary phase was entered also decreased.The duration of the inhibition phase was proportional to the aqueous garlic extract concentration and variable according to the different Salmonella serovars tested.
The garlic extract added at these final concentrations had a bacteriostatic effect on Salmonella cerro, Salmonella lindenburg, Salmonella montevideo and Salmonella hadar, but the final concentration 13 mg/ml had a bactericide effect on Salmonella enteritidis and Salmonella nikolaifleet.
Examination of the effect of A.G.E.concentration on the different Salmonella serovars cell growth revealed that (i) the duration of inhibition varied from the different Salmonella serovars and it was clearly proportional to the amount of A.G.E.applied.A longer growth inhabitation phase was observed using increased A.G.E.concentration (Table 2).(ii) The resumed growth rate, expressed as a percentage of uninhibited growth rate, was proportional to A.G.E.concentration and it was different on the Salmonella serovars tested.
The lowest resumed growth rate was observed in the case of Salmonella enteritidis using 13 mg/mL of A.G.E.Our results demonstrate that the different Salmonella serovars did not present the same duration of inhibition and resumed growth rate.It seems that the sensibility to A.G.E.depend on Salmonella serovars.
In order to study the proportionality between the amount of active substance and diameter of inhibition zone, garlic extract (20µl) or dilutions were applied directly to the surface of the plate or pipetted onto a stack of six 5 mm diameter filter-paper discs (we used six 5mm diameter filter-paper to be able to applied the 20µl garlic extract volume) cut with a hole-punch.The size of the inhibition halo was clearly proportional to the amount of A.G.E.applied, and it showed a linear relationship when plotted against the log of the diameter of the inhibition zone (Figure 2b).This positive correlation was observed using the different Salmonella serovars.

Determination of MIC and MBC
Table 3 shows the MIC and MBC values obtained, expressed in terms of the garlic extract concentration and the deduced allicin concentration estimated spectrophotometrically.The MIC values shown were determined by broth dilution.MBC values were mostly higher than MIC values, although they were occasionally identical.
MICs range from 10 mg/ml garlic concentration (estimated allicin concentration 8.1 µg/ml) to 12.5 mg/ml garlic concentration (estimated allicin concentration 10.1 µg/ml allicin).Salmonella lindenburg present the lowest MIC value overall, whereas, Salmonella nikolaifleet had the highest MIC of all Salmonella tested serovars.The MBC values range from 13 mg/ml garlic concentration (means 10.5 µg/ml of allicin) to 15 mg/ml garlic concentration (mean 12.1 µg/ml allicin).Salmonella nikolaifleet present the lowest MBC value, whereas, Salmonella montevideo had the highest MBC.

Stability of fresh aqueous garlic extract
Figure 3 shows the stability of A.G.E. on storage at 4°C and room temperature (22°C) changes with the storage time.
While the extract stored at 4°C was found to be relatively stable and even after 10 days still retained over 90% of its original inhibitory potential, the solution stored at 22°C lost the antibacterial activity slowly over a period of 6 to 10 days.
In order to study the effect of heat treatment, several A.G.E.samples were incubated for 10min at different temperature (20, 40, 60, 80 and 100 °C).Afterwards, the residual antibacterial activity was measured.We had found that a decrease in the A.G. E. inhibition strength is correlated with the decrease in the allicin content in the extract stored both at 4 and at 22°C (data not shown).
Figure 4 shows that the antibacterial activity was relatively stable at higher temperatures less than 80°C but it lost at 100°C.This result suggests that the antibacterial activity against the different salmonella serovars is due to allicin and other compounds such as the antimicrobial peptides, which could be present in the A.G.E and stable at high temperature (80°C).

Biochemical changes of salmonella serovars
In order to assess biochemical changes of Salmonella, induced by A.G.E.during the cell growth, several bacterial enzymatic activities were analyzed during the inhibition (treated bacteria) and the exponential phase cells culture (treated and untreated bacteria).Using API ZYM plates, 19 enzyme activities were tested, the results are summarised in the figure 5. Compared with the untreated cells, we observed that some enzyme activities were maintained at low level, others were already induced and others inhibited.
Salmonella cerro, S. enteritidis, S. montevideo and S. hadar α-glucosidase activity was total inhibited but in the case of S. lindenburg and S. nikolaifleet we observed that it was maintained at a low level.
During the exponential phase cell culture, treated bacteria present different enzymatic activities, compared to the untreated cells.We observed that the αmannosidase activity was already induced in all stressed cells (Figure 5), whereas, the α-glucosidase activity was inhibited in S. lindenburg and S. montevideo.These results showed that the different Salmonella serovars did not present the same biochemical changes, also during the inhibition and the exponential phase cell culture.

Discussion
With the emergence of antibioticresistant bacteria, it is reasonable to explore new sources of natural compounds with antimicrobial activity.Edible plants have been proven to be harmless and are economical.
In this study, a marked inhibitory effect of aqueous garlic extract against Salmonella serovars, which has been also described against other bacteria was observed and confirmed.The phenomenon of inhibition is characterized by: i) an inhibition phase observed during the lag phase which appears longer than that observed in the cell growth control, the same result was described by Feldberg et al. (1988) andO' Gara et al. (2000) ii) a resumed secondary logarithmic growth rate less than the original rate of growth.These observations confirm those described by Feldberg et al. (1988).
The duration of inhibition was proportional to the final aqueous garlic extract concentration, there was a simple linear relationship between inhibition activity and A.G.E.amount (mean allicin).A differential effect of garlic extract on Salmonella serovars was also reflected in the cell growth curves performed in this study.
After this inhibition phase, cells were able to overcome the inhibition and resume growth, which suggests that they were able to metabolize the allicin to noninhibitory compound.Although aqueous garlic extract at these concentrations is bacteriostatic rather than bacteriocidal, the observation that the resumed growth rate is substantially lower than the unihibitedculture growth rate suggests that the inhibited cells are not totally able to recover from aqueous garlic extract inhibition.The lower secondary growth rate could indicate either the presence of some unrepaired lesion in the cells or the depletion of limiting nutrients from the medium during the inhibition phase.Cellini et al. (1996) demonstrated that aqueous garlic extract effectively inhibited sixteen clinical isolates and three reference stains of Helicobacter pylori.Chowdhury et al. (1991) also investigated the ability of garlic to inhibit antibiotic-resistant stains of bacteria.The in vitro MIC of garlic extract was 5µl/ml, and was affective against Shigilla dysenteriae, Sh. flexneri, Sh. sonnei and Escherichia coli.Multiple resistant stains of bacteria were used by Singh and Shukla (1984) to investigate garlic's antibiotic potential.They found that garlic was more effective than any of the test antibiotics (penicillin, ampicillin, doxycycline, streptomycin and cephalexin) against clinical strains of Staphylococcus, Escherichia, Proteus, Pseudomonas and klebsiella bacteria.
The mechanism of bacterial inhibition by allicin was investigates by Feldberg et al. (1988).They reported a typical cycle of inhibition: initially, there was a lag time of approximately 15 minutes between addition of allicin and onset of inhibition; then there was a "transitory inhibition phase" whose duration was directly proportional to the allicin concentration, and inversely proportional to the culture density; this was followed by a resumed growth phase reached a stationary phase at a lower culture density than the uninhibited controls.Feldberg et al. (1988) reported that the in vitro mechanism of Salmonella typhimirium inhibition growth by allicin at bacteriostatic concentrations (0.2 to 0.5 mM) was found to be due to a delayed and partial inhibit of DNA and protein synthesis and an immediate inhibition of RNA synthesis, suggesting that this is the primary target of allicin action.
The mechanism responsible for all these activities is believed to be allicin and its chemical reaction with thiol groups of various enzymes et al., 1998).It is of interest that cooked garlic and commercially available garlic tablets lost the antibacterial effect found in raw garlic.
This antibacterial activity diminution observed during room temperature A.G.E.conservation, seems to be due to the allicin instability and its transformation with time into more stable components: polysulfides and thiosulphonates (Block, 1992).Garlic extract could be stored at 4°C as our result showed that there was no detectable loss of activity at this temperature over several days (Figure 3).However, excessive warming or longer periods at higher temperatures should be avoided (Figure 4).
This heat stability would be a very useful characteristic if the antimicrobial peptides were to be used as a food preservative, because many foodprocessing procedures involve a heating step.It seems that garlic extract contains antibacterial peptides or proteins, which exhibit an antibacterial activity stable at high temperature less than 80°C.These antimicrobial peptides have a broad antimicrobial spectrum, including Grampositive and Gram-negative bacteria, and they are produced by plants to protect themselves from pathogen invasion (Zasloff, 2002).
In 2005 Lixin and Ng had isolated alliumin, a novel protein with antifungal and antiproliferative activities from the multiple-cloved garlic bulbs.This protein presents a molecular mass of 13 kDa and an N-terminal sequence similar to a partial sequence of glucanase and demonstrating antifungal activity against Mycosphaerelle arachidicola.Alliumin was inhibitory to bacterium Pseudomonas fluorescens and exerted antiproliferative activity toward leukemia L1210 cells.However, it was devoid of ribonuclease activity, protease activity, mitogenic activity toward mouse splenocytes, and antiproliferative activity toward hepatoma Hep G2 cells.It is noteworthy that the antifungal action of alliumin is preserved after exposure to 100°C for 1 h.
Several antimicrobial agents were isolated from plant including secondary metabolites as essential oil and Terpenoides, amongst which can be cited xanthones, benzophenones, coumarins and flavonoids (Nkengfack et al., 2002;Komguen et al., 2005) Recently, there has been increasing interest in discovering new natural antimicrobials.This has been also true in food microbiology.Plant products with antimicrobial properties have notably obtained emphasis for possible application in food production in order to prevent bacterial and fungal growth (Fattouch et al., 2007;Lanciotti et al., 2004).Much focus on determining the antimicrobial activity of plant extracts is found in folk medicine (Rios and Recio, 2005).
The antimicrobial peptides appear to offer benefits such as selective inhibition and low risk of development of resistance, as well as being bactericidal in some cases (Pereira, 2006).In recent years, many antimicrobial peptides in both constitutive and inducible forms have been discovered in animals, insects and plants, and these have been recognized as important components of the innate defense system (Reddy et al., 2004;Boman, 2000).Since plants apparently do not have an active immune response, in many cases defense against microbial invaders likely depends on a variety of antimicrobial compounds.
Peptides offer genetically efficient means of providing this.Many plants produce small cysteine-rich antimicrobial peptides as an innate defense against pathogens (Dangl and Jones, 2001).
Based on their amino acid sequence homology, these have mostly been classified as α-defensins (Broekaert et al., 1995), thionins (Florack and Stiekema, 1994), lipid transfer proteins, cyclotides, snakins and hevein-like peptides.These classes have recently been centralized in a database named PhytAMP (Hammami et al., 2009).These antimicrobial peptides have been found to be excellent candidates for the development of a new generation of antimicrobial agents (Hammami et al., 2009).
Our results demonstrated that garlic has, by far, the highest antibacterial activity among the common vegetables and fruits.Furthermore, it is active against a large spectrum of bacteria.Garlic has long been known to have antibacterial activity from in vitro and in vivo (Uchida et al., 1975, Chowdhury et al., 1991).It is effective against Helicobacter pylori (Sivam, 2001), a bacteria associated with peptic ulcer disease and gastric cancer.It has synergic effect with omeprazole against Helicobacter pylori (Jonkers et al., 1999) and with streptomycin or chloromphenicol against Mycobacterium tuberculosis (Gupta and Viswawnathan, 1955).It has been shown to prevent the formation of Staphylococcus enterotoxin (Gonzalez-Fondos et al., 1994) which causes food poisoning.Garlic also has antifungal (Yoshida et al., 1987;Yamada and Azuma, 1977), antiviral and antiparasitic properties (Tsai et al., 1985).
Garlic seems to be safe agents that have the potential for broader applications to take advantage of it antibacterial activities.It is suggested that garlic, as a natural herb, could be used to extend the shelf-life of meat products, providing the consumer with food containing natural additives, which might be seen more healthful than those of synthetic origin.
To sum up, the reported results shed new light on the behavior of pathogenic Gram-negative bacteria under A.G.E.stress.This antibacterial agent triggers several biological and biochemical changes enabling Salmonella to escape the deleterious effects of Garlic.
Moreover, the natural resistance of such cells to garlic compounds is unknown.Work is in progress in our laboratory to determine the mechanisms of garlic toxicity and to identify molecules involved in these adaptive responses.Focke, M., A. Feld and K. Lichtenthaler. 1990 H.Belguith et al.

Figure 2 .
Figure 2. (a) Relationship curve between allicin amount and aqueous garlic extract volume.Reaction was done at room temperature in phosphate buffer, pH 7.2.(b) Regression plot of the log of the diameter of the inhibition zone against the volume of aqueous garlic extract.S. enteritidis-seeded Petri plate was used.

Figure 3 .
Figure 3. Antibacterial activity evolution of aqueous garlic extract during storage in the dark at room temperature ( ) or at 4°C ( ).

Figure 4 .
Figure 4. Effect of the heat treatment on the A.G.E.antibacterial activity against different Salmonella serovars.The A.G.E was treated for 10 min at various temperatures.