World J Microbiol Biotechnol (2008) 24:699–707 DOI 10.1007/s11274-007-9528-y ORIGINAL PAPER Characterization of the three selected probiotic strains for the application in food industry Blazenka Kos Æ Jagoda Šušković Æ Jasna Beganović Æ Krešimir Gjuračić Æ Jadranka Frece Æ Carlo Iannaccone Æ Francesco Canganella Received: 15 May 2007 / Accepted: 31 July 2007 / Published online: 20 August 2007
Springer Science+Business Media B.V. 2007 Abstract Previously selected bacterial probiotic strains Enterococcus faecium L3, Lactobacillus plantarum L4 and Lactobacillus acidophilus M92 have shown their potential as functional starter cultures in silage, white cabbage and milk fermentation. Therefore, the phenotypic and genotypic characteristics important for their application in food industry were investigated. Pulsed-field gel electrophoresis (PFGE) of NotI digested genomic DNA, in combination with physiological traits determined by API tests, made a useful tool for identification of these probiotic strains and differentiation among them. Lyophilized probiotic cells remained viable during 75 days of storage at 20, +4 and +15C, while fresh concentrated cells remained viable only at 20C with addition of glycerol as cryoprotectant. After the lyophilization with addition of skim milk as lyoprotectant, the viability of L. acidophilus M92, L. plantarum L4 and E. faecium L3 was reduced by only 0.37, 0.44 and 0.50 log, respectively. Furthermore, probiotic strains L. acidophilus M92, L. plantarum L4, and E. faecium L3, demonstrated anti-Salmonella activity, and L. acidophilus M92 having also antilisterial activity demonstrated by in vitro competition test. Overnight cultures and cell-free supernatants of the three probiotic strains exerted also an antagonistic effect against the Gram-positive and GramB. Kos (&)
J. Šušković
J. Beganović
J. Frece Faculty of Food Technology and Biotechnology, University of Zagreb, 6 Pierotti St., 10 000 Zagreb, Croatia e-mail: bkos@pbf.hr K. Gjuračić GlaxoSmithKline Research Centre Zagreb Ltd Zagreb, 29 Baruna Filipovića St., 10 000 Zagreb, Croatia C. Iannaccone
F. Canganella Department of Agrobiology and Agrochemistry, University of Tuscia, Via S. Camillo de Lellis, 01100 Viterbo, Italy negative test microorganisms examined, demonstrated by the agar-well diffusion test. The inhibition of Listeria monocytogenes, Salmonella typhimurium, Yersinia enterocolitica, and Acinetobacter calcoaceticus obtained, achieved by the neutralized, 5-fold concentrated supernatant ofL. plantarum L4, may be the result of its bacteriocinogenic activity. On the basis of these results, the application of the three examined probiotic strains may become a point of great importance in respect of food safety. Keywords Antagonistic effect
Enterococcus faecium
Lactobacillus acidophilus
Lactobacillus plantarum
Lyophilization
Probiotics
Pulsed-field gel electrophoresis (PFGE) Introduction Food fermentation has been shown to have not only preservative effects and the capability of aiding the modification of physico-chemical properties of various foods, but also the capability to provide significant impact on the nutritional quality and functional performance of the raw material (Knorr 1998). This offers a possibility to explore the use of probiotics as functional starter cultures for the manufacture of fermented foods. Functional starter cultures are defined as starters that possess at least one inherent, functional property, aimed at improving the quality of the end product (De Vuyst 2000). According to the definition of the World Health Organization, probiotics are living microorganisms which, when administered in adequate amounts, confer a health benefit on the host (Gilliland et al. 2001). The increasing application of probiotic cultures in food products underscores the need to 123 700 properly identify and distinguish these beneficial bacteria among the originally presented microbial population. Moreover, certain probiotic activities are strain-specific and thus, identification of probiotics to the strain level is necessary. Current strain-specific techniques used for probiotics, comprise multiple DNA-based methods such as pulsed-field gel electrophoresis (PFGE), random amplified polymorphic DNA (RAPD) PCR, ribotyping and proteinbased methods such as SDS-PAGE (Yeung et al. 2004; Klaenhammer et al. 2005; Zoetendal and Mackie 2005). Characterization of bacteria below the strain level could be used to estimate the microbial diversity present in natural populations, and to determine the technological contribution of individual strains or biotypes during the manufacture and ripening processes. One of the primary benefits associated with probiotic bacterial cultures, is the exclusion of pathogenic bacteria in the small and large intestine. Furthermore, inactivation of undesirable microorganisms during fermentation is an essential part of food preservation. On the other hand, fermentation which is a process to improve the digestibility, quality, safety and physico-chemical properties of the raw material, can be counterproductive to the viability of microorganisms (Knorr 1998; Brul 2005). Technological challenges include the necessity to obtain high productivity and viability of starter cultures, probiotic strains and functional starter cultures. Several aspects, including general, functional and technological characteristics, have to be taken into consideration while selecting probiotic strains (Sanders and Huis in’t Veld 1999; Šušković et al. 2001). Based on in vitro selection criteria, three potential probiotic strains were selected in advance: Lactobacillus acidophilus M92, Lactobacillus plantarum L4 and Enterococcus faecium L3. These three strains have been shown to have the ability to survive conditions mimicking those in the gastrointestinal tract. Because of their bile resistance and cholesterol assimilation in the presence of bile, it is postulated that these strains might help in lowering serum cholesterol in vivo (Šušković 1996; Kos et al. 2000; Šušković et al. 2000). The probiotic strains examined stimulated humoral immune response, and have the ability to survive and adhere in the mouse intestinal tract, and operate as effective probiotics that positively influence the intestinal microflora of the host (Kos et al. 2003; Frece et al. 2005a). The results obtained on the aggregation and adhesion of L. acidophilus M92, suggested that these processes are mediated by proteinaceous components (S-layer) of the cell surface (Frece et al. 2005b). Furthermore, L. plantarum L4 and E. faecium L3 strains were successfully applied as starter cultures for silage fermentation (Runjić-Perić 1996), while L. plantarum L4 successfully fulfilled the role of the starter culture in the process of white cabbage fermentation 123 World J Microbiol Biotechnol (2008) 24:699–707 (Beganović et al. 2005). Especially, L. acidophilus M92 has a great potential as probiotic strain for fermented milk products, because of the protective role of S-layer proteins, exhibited during the transit through the gastrointestinal tract (Frece et al. 2005a). The main objectives of this study were phenotypic and genotypic characterization, the estimation of viability exhibited during culture production and storage, and the estimation of antagonism to pathogens, carried out by the three selected probiotic strains. Materials and methods Bacterial strains and growth conditions The three selected probiotic strains, Lactobacillus acidophilus M92, Lactobacillus plantarum L4, and Enterococcus faecium L3 were from the culture collection of the Department of Biochemical Engineering, Laboratory of Antibiotic, Enzyme, Probiotic and Starter Cultures Production, University of Zagreb. Lactobacillus acidophilus ATCC 4356 and Leuconostoc mesenteroides LMG 7954 strains were also used. All of these strains were stored at 70C in the DeMann Ragosa Sharpe (MRS) broth (Difco) with 30% (v/v) glycerol. Before the experimental use, these cultures were sub-cultured twice in the MRS broth. The three selected probiotic strains were examined for their antagonistic activities against the following test microorganisms: Escherichia coli DH5a, Salmonella enterica serovar typhimurium, Acintetobacter calcoaceticus, Listeria monocytogenes, Yersinia enterocolitica, Gardnerella vaginalis, Bacillus cereus, Pseudomonas sp. S12+ and Vibrio anguillarum. All test microorganisms were from the Department of Agrobiology and Agrochemistry, Faculty of Agricultural Science, University of Tuscia, Viterbo. Test microorganisms were cultured aerobically at 37C on the PCA broth and agar (Difco). Carbohydrate fermentation The ability of Lactobacillus acidophilus M92, Lactobacillus plantarum L4, and Enterococcus faecium L3, to ferment various carbohydrates, was determined using the API 50 CH (for lactobacilli) and API 20 Strep (for Enterococcus), applied according to the manufacturer’s instructions (API systems, BioMérieux). Pulsed-field gel electrophoresis (PFGE) Intact genomic DNA was isolated and digested in agarose plugs as described by Yeung et al. (2004), with only minor World J Microbiol Biotechnol (2008) 24:699–707 modifications. Purified DNA was digested with five restriction enzymes (SmaI, SfiI, SalI, ApaI, NotI), all produced by Fermentas GmbH. NotI was found to be the best enzyme, i.e., it yielded reproducible and more informative digestion patterns with complete DNA digestion into a suitable number of fragments, hence it was used for further optimization of the electrophoretic conditions. The chromosomal DNA digests were separated by PFGE using a CHEF DRIII apparatus (Bio-Rad). Electrophoresis was performed through 1% w/v agarose gels at 14C and 6 V/cm in 0.5· Tris–Borate–EDTA buffer (TBE). Two pulse range times were applied; 0.5–25 s for 12 h and 25–50 s for 6 h. Viability of probiotic strains during preparation and storage The cultures were cultivated overnight in MRS broth. The cells were collected by centrifugation (10,000 rev/min/ 20 min), washed and resuspended in skim milk (10% w/v) for lyophilization or in phosphate buffer saline (PBS) containing 50% (v/v) glycerol. Cells were frozen at 20C overnight, then lyophilized in bench top freeze-dryer (B. Braun Biotech International, model Christ Alpha 1-4). Lyophilized cells and cells suspended in PBS containing 50% (v/v) glycerol were stored for 75 days at 20, +4 and +15C. Viable cells recovery was determined by colony formation on MRS agar using the standard pour-plating method. Antibacterial activity of probiotic strains Probiotic strains were tested for their antibacterial activity by in vitro competition test and agar-well diffusion test. In vitro competition test were performed in Erlenmeyer flasks containing 200 ml of PTG broth. The broth media were inoculated with probiotic strain and test microorganism in cell ratio 1:1, 2:1, and 3:1 log units. The test microorganism and the probiotic strain were added in the media at the same time, or, alternatively, the test microorganism was added 2 h after the probiotic strain. The flasks were incubated aerobically for 24 h at 37C, and the number of viable cells alone or associated within the broth, were determined by the standard pour-plating method using selective media (MRS for lactic acid bacteria, Brilliant green agar for Salmonella), or the temperature selective for growth of the particular cells (e.g., 7C for the growth of Listeria). Using the agar-well diffusion test described by Tagg et al. (1976), antibacterial activity of the three probiotic strains was further examined. Overnight cultures and cellfree supernatants of L. acidophilus M92, L. plantarum L4, 701 and E. faecium L3, were used for the examination of their antibacterial activity. The cells were removed by centrifugation (10,000 rev/min at 4C for 20 min) and the supernatants were filtered through a 0.22 lm Millipore filter. The cell-free supernatant was concentrated in a 50ml Amicon cell (Amicon, Beverly, USA) equipped with a selective (10,000 Da) membrane. In order to avoid any pH effect, the 5-fold concentrated supernatants were tested for their inhibitory activity at the pH adjusted to 6.5 and using 1 M NaOH solution. Furthermore, catalase was added at the final concentration of 1 mg/ml, to provide against the possible presence of hydrogen peroxide. Briefly, the PCA medium was seeded with the overnight cultures of the test microorganisms at the final concentration of about 106 c.f.u./ml. Wells (9 mm) were cut in the solidified agar using a sterile metal cork borer, and filled with 120 ll of the overnight culture, cell-free supernatant and neutralized 5-fold concentrated supernatant. The plates were kept at 4C for 2 h to allow the diffusion on the assay material, and then incubated at 37C for 18 h. The diameters of the clear inhibition zones were then measured. Results Biochemical and molecular characterization of the three selected probiotic strains The phenotypic characterization of the probiotic bacteria (to ferment various carbon sources) has been done using the API tests (API 50 CH and API 20 Strep). As a result, the carbohydrate fermentation patterns of L. acidophilus M92, L. plantraum L4, and E. faecium L3, were observed and their identification was confirmed in comparison with the type species from the database of the API systems (BioMérieux) with 78.6, 99.9 and 93.9% similarity, respectively (Tables 1 and 2). Prior to the evaluation of PFGE patterns, the selection of the restriction enzymes was performed. In order to determine the restriction enzyme which provides suitable fragment patterns, five restriction enzymes were tested (SmaI, SfiI, SalI, ApaI, NotI). It was found that NotI could generate clear and easy-to-interpret PFGE patterns of all examined strains (Fig. 1). The PFGE pattern of the strain L. acidophilus M92 was similar to those observed for the reference strain L. acidophilus ATCC 4356. All other strains produced distinctly different patterns (Fig. 1). Viability of probiotic strains during preparation and storage The effect of lyophilization on the viability of the three probiotic strains is shown in Table 3. Although 123 702 World J Microbiol Biotechnol (2008) 24:699–707 Table 1 Fermentation patterns of Lactobacillus acidophilus M92, Lactobacillus plantarum L4 and Enterococcus faecium L3 on API 50 CH Carbohydrates M92 L4 L3 Carbohydrates M92 L4 L3 Control Arbutin + + ± Glycerol Esculin ± + + Erthritol Salicin ± + ± D-arabinose Cellobiose + + + L-arabinose + + Maltose + + + Ribose + + Lactose ± + + D-xylose Melibiose L-xylose Saccharose Adonitol Trehalose b-methyl-xyloside Inulin ± + + + + Melezitose + + + + D-raffinose + D-fructose + + + Amidon D-mannose + + + Glycogen Rhamnose Xylitol b-gentiobiose ± + ± Dulcitol D-turanose ± + + Inositol D-lyxose ± + + D-tagatose Sorbitol + D-fucose a Methyl-D-mannoside + L-fucose a Methyl-D-glucoside + + ± L-arabitol Amygdalin + + ± Gluconate 2-ketogluconate Test Voges-Proskauer + Fermentation of: Reactions Ribose + Esculin hydrolysis + Arabinose + Pyrrolidonyl arylamidase ± Mannitol + a-galactosidase Sorbitol b-glucuronidase Lactose + + Trehalose + Leucine aminopeptidase + Raffinose Arginine dihydrolase + Alkaline phosphatase Inulin Starch + 5-keto-gluconate Reactions Hippurate hydrolysis + D-arabitol N-acetylglucosamine Test + Glycogen , negative reaction lyophilization had a deleterious effect on the viability of probiotic microorganisms, they had shown high a survival rate in the presence of skim milk as lyoprotectant. The viability of L. acidophilus M92, L. plantarum L4 and E. faecium L3, was reduced after lyophilization by only 0.37, 0.44 and 0.50 log, respectively (Table 3). Furthermore, the viable cell counts of the three probiotic strains 123 + + Table 2 Reactions of Enterococcus faecium L3 in API 20 Strep +, Positive reaction; + D-glucose Mannitol b-galactosidase ± + Galactose L-sorbose , Negative reaction, the color did not change; +, positive reaction, the color changed to yellow in 48 h; ±, the color ranged between green and yellow + remained constant during 75 days of storage at 20, +4 and +15C (data not shown). When the experiment was performed with concentrated fresh cells instead of the lyophilized ones, the recovery of viable cells during 75 days of storage with glycerol as cryoprotectant, was revealed to be the highest at 20C, which goes for all of the three tested strains (Fig. 2). The number of E. faecium L3 cells remained constant at +4 and +15C during 15 days of storage, but L. plantarum L4 cells remained viable in the same period only when stored at +4C. The greatest survival at +4C had been shown by E. faecium L3, with the viable cell counts of about 7 log units after 75 days of storage (Fig. 2). At the end of the storage period, the viable cell counts of L. plantarum L4 and E. faecium L3, steadily approximated 104 cells/ml at +15C (Fig. 2). Antagonistic activity of probiotic strains Two different methods were used for the examination of antibacterial activity of the three probiotic strains. Antagonistic activity of the probiotic strains was first tested by in vitro competition test, using PTG broth inoculated with World J Microbiol Biotechnol (2008) 24:699–707 703 (a) 10 l o g CF U/ ml 8 6 4 2 0 0 10 20 30 40 50 60 70 time (days) (b) 10 Fig. 1 Electrophoretic karyotype of NotI-digested genomic DNA of the lactic acid bacteria examined. Lane M, Lambda-DNA PFGE molecular mass markers; lane 1, Lactobacillus acidophilus ATCC 4356; lane 2, Lactobacillus acidophilus M92; lane 3, Lactobacillus planatarum L4; lane 4, Enterococcus feacium L3; lane 5, Leuconostoc mesenteroides LMG 7954 l og CFU/ ml 8 6 4 2 0 Table 3 Survival of potential probiotic strains after lyophilization with skim milk as lyoprotector 0 10 20 30 40 50 60 70 50 60 70 time (days) (c) 10 8 l og CFU/ml both the probiotic strain and the test microorganism (Salmonella typhimurium or Listeria monocytogenes). When the probiotic strain and the test microorganism were inoculated into the growth medium at the same time (with the ratio of the cell counts 3:1 log units, respectively), the inhibition of S. typhimurium occurred after 12 h of incubation with L. plantarum L4 (Fig. 3A). However, by inoculating the test microorganism 2 h after the probiotic strain, S. typhimurium was inhibited almost immediately (Fig. 3B). Similar results were obtained by E. faecium L3 and L. acidophilus M92 against S. typhimurium (data not shown). With the inoculation of the lower initial number of probiotic bacteria L. acidophilus M92 (with the ratio of cell 6 4 2 0 0 10 20 30 40 time (days) Microorganisms Lactobacillus acidophilus M92 Viable count (log10) c.f.u./ml of probiotic culturesa Before lyophilization After lyophilization 8.87 ± 0.23 8.50 ± 0.17 Lactobacillus plantarum L4 8.71 ± 0.31 8.27 ± 0.28 Enterococcus faecium L3 8.89 ± 0.29 8.39 ± 0.20 a Measurements are expressed as means ± standard error of the three replicates of the each individual strain Fig. 2 Survival of the concentrated cultures of Lactobacillus acidophilus M92 (a), Lactobacillus plantarum L4 (b), and Enterococcus faecium L3 (c), stored with glycerol as protector at 20C (j), +4C (h) and +15C (x) counts 2:1 log units instead of 3:1 log units), the inhibition started after 2 h of incubation, and S. typhimurium was completely inhibited after 10 h of incubation with the probiotic strain L. acidophilus M92 (Fig. 4). Furthermore, 123 704 World J Microbiol Biotechnol (2008) 24:699–707 10 9 (a) 8 7 log CF U/ml log CFU/ml 8 6 4 2 6 5 4 3 2 1 0 0 2 4 6 8 10 12 14 16 18 20 22 0 24 0 2 4 6 8 10 time (h) 14 16 18 20 22 24 time (h) Fig. 4 Inhibition of Salmonella typhimurium when grown in association with Lactobacillus acidophilus M92 in PTG broth at 37C. Ratio of cell numbers Lactobacillus:Salmonella was 2:1 log units. j, log N of S. typhimurium associated; h, log N of S. typhimurium alone; x, log N of L. acidophilus M92 alone and associated 10 (b) 8 log CF U/m l 12 6 4 2 0 2 4 6 8 10 12 14 16 18 20 22 24 time (h) Fig. 3 Inhibition of Salmonella typhimurium when grown in association with Lactobacillus plantarum L4 in PTG broth at 37C. S. typhimurium was inoculated at the same time (a) and 2 h after probiotic strain L. plantarum L4 (b). Ratio of cell numbers Lactobacillus:Salmonella was 3:1 log units. j, log N of S. typhimurium associated; h, log N of S. typhimurium alone; x, log N of L. plantarum L4 alone and associated log CFU/ml 0 11 10 9 8 7 6 5 4 3 2 1 0 0 2 4 6 8 10 12 14 16 18 20 22 24 time (h) after inoculation of the same initial number of the probiotic strain and the test microorganism, during the first 12 h of incubation with L. acidophilus M92, the number of viable L. monocytogenes decreased for 2 log-values in comparison to the control, and was completely inhibited after 24 h of incubation (Fig. 5). When the overnight cultures and cell-free supernatants were examined by the agar-gel diffusion test, the probiotic strains exerted antibacterial activity against different Gram-positive and Gram-negative test microorganisms (Table 4). Furthermore, positive results were obtained with the usage of the neutralized, 5-fold concentrated supernatant of L. plantarum L4 against Listeria monocytogenes, Salmonella typhimurium, Yersinia enterocolitica, and Acinetobacter calcoaceticus. Discussion Lactic acid bacteria have a worldwide industrial use as starter cultures in the manufacturing of different fermented products. Their confirmed probiotic properties are of great 123 Fig. 5 Inhibition of Listeria monocytogenes when grown in association with Lactobacillus acidophilus M92 in PTG broth at 37C. Ratio of cell numbers Lactobacillus: Listeria was 1:1 log units. j, log N of L. monocytogenes associated; h, log N of L. monocytogenes alone; x, log N of L. acidophilus M92 alone and associated importance for the application of the three selected strains, as functional starter cultures for milk (L. acidophilus M92) and vegetable (L. plantarum L4 and E. faecium L3) fermentations. These strains fulfill in vitro and in vivo selection criteria, such as survival, competition, adhesion and colonization in the gastrointestinal tract, as well as immune-modulating capability (Šušković et al. 2000; Kos et al. 2003; Frece et al. 2005b). However, increasing application of probiotics in food products underscores the need to properly identify these beneficial bacteria and distinguish them from among other microorganisms present in the products. Using strain-specific patterns to verify phenotypic and genotypic characteristics of the applied probiotic bacteria represents a necessary approach as to provide quality control of probiotic products. The carbohydrate fermentation analysis performed by the API system was first carried out as the primary method for phenotypic World J Microbiol Biotechnol (2008) 24:699–707 705 Table 4 Antibacterial activity of the overnight culture, cell-free supernatant and neutralized 5-fold concentrated supernatant of Lactobacillus acidophilus M92, Lactobacillus plantarum L4 and Enterococcus faecium L3, obtained by agar-well diffusion test Test microorganisms Zones of inhibition (mm) obtained by Overnight culture Cell-free supernatant Neutralized concentrated supernatant M92 L4 L3 M92 L4 L3 M92 L4 L3 Escherichia coli DH5a 25 18 18 17 19 15 – – – Salmonella typhimurium 29 27 22 18 20 15 – 12 – Acintetobacter calcoaceticus 15 22 14 15 17 13 – 10 – Listeria monocytogenes 21 22 15 15 14 10 – 11 – Yersinia enterocolitica 24 28 22 17 28 12 – 10 – Gardnerella vaginalis 25 22 19 14 14 12 – – – Bacillus cereus 27 27 20 19 17 14 – – – Pseudomonas sp. S12+ 22 23 17 14 15 11 – – – Vibrio anguillarum 26 22 15 15 13 12 – – – characterization of the three probiotic strains. High similarity of the carbohydrate utilization profile between the probiotic strains examined and the corresponding typestrains originating from the API system database (bioMérieux), was demonstrated. The results obtained could be useful in the application of the three strains as functional starter cultures for different fermented products. However, the carbohydrate fermentation pattern can be affected by experimental conditions such as incubation time and temperature, and limited in terms of its discriminating ability and accuracy. In contrast, molecular methods such as the PFGE, offer the advantage of having a good level of taxonomic resolution at species or subspecies level, so being able to distinguish added probiotic cultures among microbial population originally present in the functional food products (Yeung et al. 2004). PFGE patterns of NotIdigested genomic DNA of L. acidophilus M92, L. plantarum L4 and E. faecium L3 have shown that this method can be successfully applied for the identification of these strains and the differentiation among them. A highly reproducible method of characterizing and distinguishing closely-related strains is the PFGE performed by infrequently cutting endonucleases (Pepe et al. 2004). Namely, based on the DNA–DNA hybridizations, six different species were identified within the L. acidophilus-group, and two of them (L. acidophilus and L. johnsonii) are mainly used in fermented dairy products and show nearly identical phenotypical properties (Fujisawa et al. 1992; Reuter et al. 2002). Genetic differentiation of strains of several lactic acid bacterial species has been successfully performed by PFGE, and has revealed that bacterial isolates of the same species can exhibit closely related, but not identical patterns (Yeung et al. 2004). This is confirmed by the similar, but not identical restriction fragments obtained by the PFGE as regards L. acidophilus M92 and type strain L. acidophilus ATCC 4356. A remarkable genetic restriction polymorphism was obtained by the PFGE after the digestion of the genomic DNA of the 30 L. plantarum strains by NotI (Pepe et al. 2004). The PFGE profiles have shown the large number of restriction fragments ranging between 9 and 150 kb in size, similar to those obtained with the examined probiotic strain L. plantarum L4. Furthermore, NotI digestion pattern of the probiotic strain E. faecium L3, was similar to those obtained for E. faecium ATCC 19436. It yielded five DNA bands in the range of 10–1,000 kb (Oana et al. 2002). Very clear, reproducible restriction patterns of the three probiotic strains examined were obtained, allowing monitoring of these strains during food processing. The viability and activity of probiotic bacteria during preparation and storage, is also very important for their industrial application. They can be added to probiotic products as fresh or lyophilized cells. High population levels, between 106 and 108 microbial cells per ml, should be present in probiotic products (Sanders and Huis in’t Veld 1999). However, during preparation for industrial application, bacteria are subjected to stress conditions such as freezing, drying and concentration stress, which diminish cell viability. The frequency of cell death is usually correlated with cell membrane damage. Cryoprotectants (for freezing) and lyoprotectants (for lyophilization) are usually used for membrane stabilization. These are small molecules with osmotic behavior, or polymers, which promote the formation of amorphous or ‘‘glassy’’ solids and reduce ice formation that can be cell-damaging (Conrad et al. 2000; Capela et al. 2006). For protection of L. acidophilus M92, L. plantarum L4 and E. faecium L3 cells, glycerol as cryoprotectant and skim milk as lyoprotectant have been used. All of the three strains examined had shown the ability to withstand the stresses associated with lyophilization, and to retain their viability during a 75 day-storage period in skim milk at 20, +4 and +15C. 123 706 Higher survival rates of L. acidophilus M92, exhibited during lyophilization, may be attributed to its specific cell surface composition. Namely, L. acidophilus M92 possesses surface layer (S-layer) proteins protective for this strain during its transit through the intestinal tract of mice, which also mediate adhesion (Frece et al. 2005b). There exists an increasing amount of evidence that S-layercarrying bacteria may use S-layer protein genes for the adaptation to different stress factors, such as drastic changes in environmental conditions (Jakava-Viljanen et al. 2002). Furthermore, fresh concentrated cells of the three probiotic strains, can be stored at 20C for 75 days without any loss of viability. Additionally, E. faecium L3 cells retained their high viability (107 cells/ml) even at +4C for 75 days of storage. Another functional, strain-specific property of probiotic strains is their antagonism to pathogens. The antimicrobial activity of probiotic strains in the variety of food products, may contribute to an improvement in the quality of fermented foods, achieved through the control of spoilage and pathogenic bacteria, extending shelf-life and improving sensory quality (Wei et al. 2006; Siripatrawan and Harte 2007). Two common pathogens, Salmonella typhimurium and Listeria monocytogenes, were used for the investigation antagonistic activity of the probiotic bacteria examined, carried out via the in vitro competition test. All of the three strains have shown strong antibacterial activity against S. typhimurium. Additionally, L. acidophilus M92 exerted anti-listerial bactericidal activity which could be of great technological importance, since L. monocytogenes is able to survive milk fermentation conditions as well as cheese manufacturing, within the frame of which L. acidophilus M92 can be applied as probiotic strain. Based on the suggestion that high competitive power is an essential property of a protective culture (Wei et al. 2006), the obtained results indicated that both early and late contaminations with the examined pathogens could be combated by application of probiotic strains L. acidophilus M92, L. plantarum L4, or E. faecium L3. On the occasion of the agar-well diffusion test, overnight cultures of these three strains and their cell-free supernatants also exhibited antibacterial activity against a wide range of test microorganisms. By means of neutralization of concentrated cellfree supernatants, the antimicrobial activity of lactic acid was eliminated. The inhibition of some pathogens obtained attained by L. plantarum L4 supernatants, may be the result of its bacteriocinogenic activity. Furthermore, the antibacterial activity of the supernatants against L. monocytogenes was eliminated or reduced by their neutralization. As L. monocytogenes tolerates low pH environments, its inhibition attained by overnight cultures and supernatants could be the result of bacteriocinogenic activity of the examined probiotic strains. However, the 123 World J Microbiol Biotechnol (2008) 24:699–707 activity of the produced bacteriocins could be affected by the change in pH value of the medium, which could explain the reduction of L. monocytogenes inhibition. Namely, bacteriocins are ribosomally synthesized antimicrobial peptides produced by one bacterium, active against other bacteria, either in the same species (narrow spectrum) or across genera (broad spectrum). They are produced by food-grade bacteria, are usually heat-stable, and can inhibit a number of primary pathogenic and spoilage organisms that cause problems in minimally processed foodstuffs. Although bacteriocins characterized by a wide spectrum of activity, are usually those most sought after, other factors including pH optima, solubility and stability are as important and are major considerations in choosing bacteriocins. However, up to now, only nisin and pediocin PA1/AcH have found widespread use in the food domain. Bacteriocins can have implications on the development of desirable flora in fermented food, and can be produced in situ by bacterial cultures that substitute for all, or part of the starter cultures (Cotter et al. 2005). A wide range of the inhibitory spectrum of bacteriocins produced by L. plantarum strains have been reported included anti-listerial bactericidal activity (Messi et al. 2001; Maldonado et al. 2003; Bernbom et al. 2006). Interestingly, bactericidal activity of the concentrated neutralized supernatant of L. plantarum L4 was observed with Gram-negative strains, and was best-achieved against Y. enterocolitica. Activity against Gram-negative strains exhibited by Gram-positive bactreriocin producers, has rarely been reported (Messi et al. 2001; Elgado et al. 2004). Further characterization of antibacterial substances produced by L. plantarum L4 will be performed. Acknowledgments This work was partially supported by the FEMS (the Federation of European Microbiological Societies) 3-month Fellowship hosted by the Department of Agrobiology and Agrochemistry, University of Tuscia, Viterbo, Italy. The authors are also grateful for financial support provided by the Ministry of Science, Education and Sport of the Republic of Croatia, Projects: 0058009 and HITRA TP-5822. References Beganović J, Kos B, Gjuračić K, Šušković J, Frece J (2005) Genotypic and phenotypic diversity of lactic acid bacteria isolated from controlled sauerkraut fermentation. In: Book of abstracts of the second congress of Croatian geneticists. Itg Digitalni tisak, Zagreb, Croatia, p 62 Bernbom N, Licht TR, Saadbye P, Vogensen FK, Nørrung B (2006) Lactobacillus plantarum inhibits growth of Listeria monocytogenes in an in vitro continuous flow gut model, but promotes invasion of L. monocytogenes in the gut of gnotobiotic rats. Int J Food Microbiol 108:10–14 Brul S (2005) The role of fermentation in food preservation and food quality. In: Nout RMJ, de Vos WM, Zwietering MH (eds) Food fermentation. Wageningen Academic Publishers, Wageningen, pp 69–78. ISBN 9076998833 World J Microbiol Biotechnol (2008) 24:699–707 Capela P, Hay TKC, Shah NP (2006) Effect of cryoprotectants, prebiotics and microencapsulation on survival of probiotic organisms in yoghurt and freeze-dried yoghurt. Food Res Int 39:203–211 Conrad PB, Miller DP, Cielenski PR, de Pablo JJ (2000) Stabilization and preservation of Lactobacillus acidophilus in saccharide matrices. Cryobiology 41:17–24 Cotter PD, Colin H, Ross RP (2005) Bacteriocins: developing innate immunity for food. Nature Rev Microbiol 3:777–788 De Vuyst L (2000) Technology aspects related to the application of functional starter cultures. Food Technol Biotechnol 38:105–112 Elgado FB, Guerra MARV, Macayan RA, Mendoza HA, Lirazan MB (2004) Spectrum of bacteriocin activity of Lacotobacillus plantarum BS and fingerprinting by RAPD-PCR. Int J Food Microbiol 95:11–18 Frece J, Kos B, Beganović J, Vuković S, Šušković J (2005a) In vivo testing of functional properties of three selected probiotic strains. World J Microbiol Biotechnol 21:1401–1408 Frece J, Kos B, Svetec IK, Zgaga Z, Mrša V, Šušković J (2005b) Importance of S-layer proteins in probiotic activity of Lactobacillus acidophilus M92. J Appl Microbiol 98:285–292 Fujisawa T, Benno Y, Yaeshima Y, Mitsouka T (1992) Taxonomic study of the Lactobacillus acidophilus group, with recognition of Lactobacillus gallinarum sp. nov. and Lactobacillus johnsonii sp. nov. and synonymy of Lactobacillus acidophilus group A3 (Johnson et al. 1980) with type strain of Lactobacillus amylovorus (Nakamura 1981). Int J Syst Bacteriol 42:487–491 Gilliland SE, Morelli L, Reid G (2001) Health and nutritional properties of probiotics in food including powder milk with live lactic acid bacteria. Paper presented at the Joint FAO/WHO expert consultation on evaluation of health and nutritional properties of probiotics in food including powder milk with live lactic acid bacteria, Cordoba, Argentina Jakava-Viljanen M, Avall-Jaaskelainen S, Messner P, Sleyter B, Palva A (2002) Isolation of three new surface layer protein genes (slp) from Lactobacillus brevis ATCC 14869 and characterization of the change in their expression under aerated and anaerobic conditions. J Bacteriol 184:6786–6795 Klaenhammer TR, Barrangou R, Buck BL, Azcarate-Peril MA, Altermann E (2005) Genomic features of lactic acid bacteria effecting bioprocessing and health. In: de Vos WM, Kuipers OP, Kleerebezem M (eds) Proceedings of the 8th symposium on lactic acid bacteria: genetics, metabolism and applications. Elsevier Science B.V., Egmond Aan Zee, The Netherlands, pp 393–410. FEMS Microbiology Reviews 29 Knorr D (1998) Technology aspects related to microorganisms in functional foods. Trends Food Sci Technol 9:295–306 Kos B, Šušković J, Goreta J, Matošić S (2000) Effect of protectors on the viability of Lactobacillus acidophilus M92 in simulated gastrointestinal conditions. Food Technol Biotechnol 36:121– 127 Kos B, Šušković J, Vuković S, Šimpraga M, Frece J, Matošić S (2003) Adhesion and aggregation ability of probiotic strain Lactobacillus acidophilus M92. J Appl Microbiol 94:981–987 Maldonado A, Ruiz-Barba JL, Jiménez-Dı́az R (2003) Purification and genetic characterization of plantaricin NC8, a novel 707 coculture-inducible two-peptide bacteriocin from Lactobacillus plantarum NC8. Appl Environ Microbiol 69:383–389 Messi P, Bondi M, Sabia C, Battini R, Manicardi G (2001) Detection and preliminary characterization of a bacteriocin (plantaricin 35d) produced by a Lactobacillus plantarum strain. Int J Food Microbiol 64:193–198 Oana K, Okimujra Y, Kawakami Y, Hayashida N, Shimosaka M, Okazaki M, Hayashi T, Ohnishi M (2002) Physical and genetic map of Enterococcus faecium ATCC 19434 and demonstration of intra- and interspecific genomic diversity in enterococci. FEMS Microbiol Lett 207:133–139 Pepe O, Blaiotta G, Anastasio M, Moschetti G, Ercolini D, Villani F (2004) Technological and molecular diversity of Lactobacillus plantarum strains isolated from naturally fermented sourdoughs. Syst Appl Microbiol 27:443–453 Reuter G, Klein G, Goldberg M (2002) Identification of probiotic cultures in food samples. Food Res Int 35:117–124 Runjić-Perić V (1996) Cultivation of conjoint lactic acid bacteria for ensiling starter cultures. Ph.D. Thesis, Faculty of Food Technology and Biotechnology, University of Zagreb, Croatia Sanders ME, Huis in’t Veld JHJ (1999) Bringing a probioticcontaining functional food to the market: microbiological, product regulatory and labelling issues. In: Konings WN, Kuipers OP, Huis in’t Veld JHJ (eds) Proceedings of the 6th symposium on lactic acid bacteria: genetics, metabolism and applications. Kluwer Academic Publishers, Antonie van Leeuwenhoek, Veldhoven, The Netherlands, pp 293–316. ISBN 0792359534 Siripatrawan U, Harte BR (2007) Solid phase microextraction/gas chromatography/mass spectrometry integrated with chemometrics for detection of Salmonella typhimurium contamination in a packaged fresh vegetable. Anal Chim Acta 581:63–70 Šušković, J (1996) The growth and probiotic effect of chosen lactic acid bacteria. Ph.D. Thesis, Faculty of Food Technology and Biotechnology, University of Zagreb, Croatia Šušković J, Kos B, Matošić S, Besendorfer V (2000) The effect of bile salts on survival and morphology of potential probiotic strain Lactobacillus acidophilus M92. World J Microbiol Biotechnol 16:673–678 Šušković J, Kos B, Goreta J, Matošić S (2001) Role of lactic acid bacteria and bifidobacteria in synbiotic effect. Food Technol Biotechnol 39:227–235 Tagg JR, Adjoin AS, Watchmaker LW (1976) Bacteriocins of Gram positive bacteria. Microbiol Rev 40:722–756 Wei H, Wolf G, Hammes WP (2006) Indigenous microorganisms from iceberg lettuce with adherence and antagonistic potential for use as protective culture. Innovative Food Sci Emerg Technol 7:294–301 Yeung PSM, Kits CL, Cano R, Tong PS, Sanders ME (2004) Application of genotypic and phenotypic analyses to commercial probiotic strain identity and relatedness. J Appl Microbiol 97:1095–1104 Zoetendal EG, Mackie RI (2005) Molecular methods in microbial ecology. In: Tannock GW (ed) Probiotics & prebiotics: scientific aspects. Caister Academic Press, Wymondham, UK, pp 1–24. ISBN 1-904455-01-8 123