The Culture of Pediococcus pentosaceus T 1 Inhibits Listeria Proliferation in Salmon Fillets and Controls Maturation of Kimchi

Lactic acid bacteria (LAB), highly benefi cial microorganisms for humans, have been used for a long time in fermented products such as fermented milk, sausages and kimchi (1,2). They are usually Gram-positive, catalase-negative, and non-spore-forming bacteria (1). They are classifi ed into various genera including Lactobacillus, Leuconostoc, Streptococcus, Lactococcus and Pediococcus. The properties of LAB are numerous: the enhancement of food preservation and fl avour by their metabolites, antimicrobial eff ect against harmful bacteria, and supply of nutrients. The biological eff ects of LAB on human health have been studied in various research areas (1). These effects include activation of immunity, anticancer activity, reduction of cholesterol level and liver protection (3–6). A recent study has shown the suppressive eff ect on allergy such as atopic dermatitis by lactic acid bacteria via cell line and animal studies (7). ISSN 1330-9862 original scientifi c paper


Introduction
Lactic acid bacteria (LAB), highly benefi cial microorganisms for humans, have been used for a long time in fermented products such as fermented milk, sausages and kimchi (1,2).They are usually Gram-positive, catalase-negative, and non-spore-forming bacteria (1).They are classifi ed into various genera including Lactobacillus, Leuconostoc, Streptococcus, Lactococcus and Pediococcus.The properties of LAB are numerous: the enhancement of food preservation and fl avour by their metabolites, antimicrobial eff ect against harmful bacteria, and supply of nutrients.The biological eff ects of LAB on human health have been studied in various research areas (1).These effects include activation of immunity, anticancer activity, reduction of cholesterol level and liver protection (3)(4)(5)(6).A recent study has shown the suppressive eff ect on allergy such as atopic dermatitis by lactic acid bacteria via cell line and animal studies (7).
Kimchi is a traditional fermented vegetable dish in Korea, which has centuries long historical records of consumption (1).Its fermentation is a spontaneous process that is initiated by various microorganisms originally present in the raw materials for kimchi production (1).The microorganisms in kimchi include approx.200 species of bacteria and several yeasts, which are involved in a series of fermentation stages.Various LAB species, among which Pediococcus spp., have been isolated from kimchi, and their diff erent technological characteristics have been studied (8,9).This strain is known to be used in the American-style fermented meat and vegetables as a main starter culture (10).It produces bacteriocin called pediocin, which usually possesses antilisterial activity.Recent studies have shown that pediocins or pediococcus cultures inhibit Listeria monocytogenes in fermented sausages or salami (11,12).L. monocytogenes, a major human pathogen, is a bacterium causing listeriosis, a serious bacterial disease (13).Elderly people, newborns, and pregnant women, who have weakened immune systems, are susceptible to this disease, which is accompanied by sepsis and meningitis with high mortality rate (14).
Safe preservation of food is one of the critical issues in food industry.Traditionally, control of temperature such as by heating or refrigeration has usually been used for food preservation.However, these treatments can have high cost and cause the change of the components of food products, which results in the loss of food nutrients and changes in fl avour, recognized as unnatural by consumers.In addition, the occurrence of psychrophilic pathogens does not guarantee safety in food preservation based on low temperature.Synthetic preservatives are used as an alternative way for food preservation, but they can be unfavourable for human health (15).Therefore, in recent years, biopreservation using biomaterials has received attention as a way of food preservation, with a trend demanding fresh and natural products.LAB are one of the good sources for biopreservation.
One of the benefi cial properties of LAB is the production of antimicrobial substances like bacteriocin (13), which are used for biopreservation.Bacteriocin such as nisin is admi ed as a GRAS (Generally Recognized As Safe) (16), and many European countries use nisin as a food preserva tive in commercial food products including canned food, mayonese and cheese (17).
In our previous study, we isolated a strain from kimchi and identifi ed it by ribosomal DNA sequence analysis as the antilisterial strain Pediococcus pentosaceus T1 (18).In this study, we examine the antilisterial eff ect of P. pentosaceus T1 culture in fi sh products like salmon fi llet, and its eff ect on maturation and quality of kimchi.

Isolation and identifi cation of the LAB that produce antibacterial agents
Commercial kimchi was obtained from a store (Seoul, Korea), the samples were unwrapped, transferred to stomacher fi lter bags and mixed with sterile phosphate buff er (0.625 mM, pH=7.2).The samples were homogenized with a BagMixer ® 400 VW (Interscience, Saint--Nom-la-Bretèche, France) at 300 × g for 5 min, then serially diluted and plated onto de Man, Rogosa and Sharpe (MRS) agar (BD Difco, Detroit, MI, USA), followed by incubation under anaerobic conditions using the GasPak TM system (GENbox anaerobic indicator, bioMérieux S.A, Marcy l'Etoile, France) at 37-42 °C for 48 h.Colonies were Gram stained and tested for catalase.Gram-positive and catalase-negative bacilli or coccobacilli were selected (18).For identifi cation of LAB that produce antibacterial agents, rDNA PCR analysis was performed.Genomic DNA was extracted using DNeasy tissue kit (Qiagen, Hiden, Germany), and PCR reaction for the amplifi cation of 16S rDNA was performed using 20 pmol of universal bacterial primers: 27F (50-AGAGTTTGATCCTGGCTCA-30) and 1492R (50-GGTTACCTTGTTACGACTT-30) (19), and template DNA, 100 mM dNTP, 1 U of Taq DNA polymerase (Roche, Mannheim, Germany).A er thermocycling amplifi cation (18), agarose gel electrophoresis was performed to confi rm PCR products.16S rDNA from the gel was collected, purifi ed using Solgent gel and PCR purifi cation system (Solgent, Daejeon, South Korea), and then compared with 16S rDNA sequences of other strains using the BLAST programs in the National Center for Biotechnology Information database (Rockville Pike, Bethesda, MD, USA) and the EzTaxon server v. 2.1 (20) by Solgent.Phylogenetic analyses of the 16S rRNA gene sequences were conducted using Molecular Evolutionary Genetics Analysis (MEGA) so ware, v. 5 (21).

Antilisterial activity of LAB from kimchi
Antilisterial activity of the isolated LAB strains was tested using an agar well diff usion method, as described by de Carvalho et al. (22).One hundred and twenty fi ve LAB were cultured overnight by inoculating 10 5 CFU/mL in tryptic soy broth (TSB; BD Difco).The agar well diff usion assay was performed by spreading Listeria monocytogenes cultures on tryptic soy agar (TSA) plates (BD Difco).Wells of 6.5 mm in diameter were punched in these plates, fi lled with 50 mL of cell-free culture supernatants of LAB and incubated at 35 °C for 24 h.Antilisterial activities were measured by examining the diameters of the inhibition zones around the wells.The inhibitory activities corresponding to the diameters of the inhibition zones were expressed in mm.

Culture conditions and preparation of crude supernatant
The composition of the culture medium was as follows (in %): sucrose 1.5 and fructose 1.5 (carbon source), soya peptone and yeast extract 1.5 (nitrogen source), K 2 H-PO 4 0.1, sodium acetate 0.1, tryptophan 0.05, cysteine 0.05, MgSO 4 0.01, and MnSO 4 0.005.A 5-litre laboratory scale fermentor (FMT ST-D, Fermentech, Cheongju, South Korea) was used for the growth of LAB under anaerobic conditions at 35 °C, with stirring at 100 × g for 20 h.The fermented culture was centrifuged at 8000 × g for 30 min, and the supernatant was autoclaved at 100 °C for 15 min to inactivate proteases.Organic acids in the culture were removed by ultrafi ltration (molecular mass cut-off <3 kDa).The fi lter sludge was lyophilized for the study.

Listeria cultivation, salmon medium preparation, and antilisterial determination
Listeria monocytogenes KCCM 40307 was inoculated on TSA (BD Difco) corresponding to the cell number of 10 8 cells per mL.This Listeria solution was diluted to 10 5 cells per mL in 200 mL of TSB (BD Difco) containing Pediococcus pentosaceus T1 culture at mass per volume ratios of 1, 2, 3 and 4 %, and incubated at 35 °C.The Listeria culture was harvested at 6, 9, 12, 15 and 18 h to count viable cell numbers on Listeria selective medium, an Oxford Medium Base (BD Difco) containing antimicrobial supplement (BD Difco).For antimicrobial activity of nisin (Sigma-Aldrich, St. Louis, MO, USA) and P. pentosaceus T1 culture, raw salmon (10 g) was ground under aseptic conditions, and added to TSB and phosphate buff er (0.625 mM, pH=7.2) (10 mL) to make a salmon-based medium.A volume of 100 µL of Listeria culture was added to the salmon-based medium followed by the addition of nisin and P. pentosaceus T1 culture with serial dilutions (20, 10, 5, 2.5, 1.25 and 0.625 mg/mL).The culture was incubated at 35 °C for 24 h, and spread onto Listeria selective medium.Minimal inhibitory concentration (MIC) was set where viable Listeria was not observed on the plate.

Antilisterial activity in raw salmon fi llet
Frozen salmon was thawed, and sliced into fi llets (200 g).Three fi llets were used to examine the antilisterial activity of each P. pentosaceus T1 culture or sodium hypochlorite (ACL-60G, Namkang, Bucheon, South Korea) treatment.The fi llets were inoculated with Listeria culture (10 6 CFU/mL), and then rested for 2 h at room temperature.A erwards, they were dipped in sodium hypochlorite (0.2 mg/mL) or the P. pentosaceus T1 culture solution (6 g per 100 mL) for 10 min, or sprayed with sodium hypochlorite or the culture solution.The fi llets were incubated in the refrigerator at 4 °C for 24 h.Listeria cells were taken from the fi llets by grinding them and diluting with phosphate buff er (0.625 mM, pH=7.2),followed by spreading on the Listeria selective medium for counting the Listeria cells.

Antimicrobial activity on LAB from kimchi
Antimicrobial activity of the isolated P. pentosaceus T1 on LAB from kimchi was tested using an agar well diff usion method, as described by Jang et al. (18).Indicator strains, including 16 LAB strains, were cultured overnight by inoculating 10 5 CFU/mL in MRS medium (BD Difco).Sixteen strains of LAB were obtained from Korean Collection of Type Cultures (KCTC, Daejeon, Korea).The agar well diff usion assay was performed by spreading the LAB cultures on MRS agar plates (BD Difco).Wells of 6.5 mm in diameter were punched in these plates, fi lled with 50 µL of cell-free supernatants of P. pentosaceus T1, from which organic acid was removed, and incubated at 35 °C for 24 h.Antimicrobial activity was examined by measuring the diameters of inhibition zones around the wells.When the diameters of the clear zones were wider than 6.5 mm, the LAB were considered to be inhibited by P. pentosaceus T1.The inhibitory activity corresponding to the diameters of the inhibition zones was expressed in mm.

Preparation of kimchi
Kimchi was prepared in batches up to 500 kg at a kimchi factory (Our Home Co. Ltd, Seongnam, South Korea) using their production line.Chinese cabbage (Brassica campestris L. ssp.pekinensis Rupr.) was soaked in a solution of refi ned salt (80 g/L, Hanju Co, Ulsan, Korea) for 2 to 4 h, and washed 3 times with tap water.The washed Chinese cabbages were le to drain any excess water in a wicker container at 5 to 10 °C for 2 to 4 h.The salted Chinese cabbages were then mixed with the other kimchi ingredients including red pepper powder, radish, garlic, ginger, onion, sugar and fermented fi sh sauce.The fi nal salt mass fraction of the kimchi was adjusted to 1.9 to 2.1 % using refi ned salt.The fi ltered culture of Pediococcus pentosacesus T1 was inoculated into kimchi preparation (1 %, 10 g/kg).As a control, the same kimchi recipe was used without the fi ltered culture of Pediococcus pentosacesus T1.The prepared kimchi was vacuum packed into 500-gram retort packages with polyethylene resin and incubated at 10 °C in a refrigerator (Daehan Science, Seoul, Korea) for 105 days.

Chemical analysis of kimchi
Ripened kimchi (500 g) was macerated using a hand blender (Hanil, Seoul, South Korea) for 2 min.The kimchi juice was centrifuged at 5000 ×g for 5 min, and the pH of the supernatant was tested with a pH meter (Me ler Toledo, Virofl ay, France).The supernatant was then titrated with 0.1 M NaOH to pH=8.3 to determine the total titratable acidity (TTA), which was expressed as: where V is the volume of 0.1 M NaOH (mL), f is the factor of 0.1 M NaOH solution, and k is the constant of organic acid equivalent to 1 mL of 0.1 M NaOH solution (in the case of lactic acid k=0.009).

Microbial analysis of kimchi
For the microbial analysis, kimchi samples were randomly selected and blended for 2 min.The juice samples were fi ltered with a sterile sieve (pore size: 0.15 mm, Chung Gye Sang Gong SA, Seoul, South Korea) and the aliquots of each fi ltrate were serially diluted with 0.1 % peptone water and spread onto plate count agar (PCA; Merck, Darmstadt, Germany) for total microbial counts.The plates were counted a er 2 to 3 days of incubation at 37 °C.Among the serially diluted plates, those with 30 to 300 CFU/mL were used for enumeration of the total microbial population in the kimchi samples.

Statistical analysis
Statistical analysis was performed using the SPSS-PC v. 11.0 so ware (SPSS, Chicago, IL, USA).Data were sub-jected to ANOVA, and the mean values were separated using Duncan's multiple-range test, with signifi cance at p<0.05.For the signifi cance of the diff erences between the given samples and control group, Student's t-test was used (p<0.05).

Isolation of antibacterial LAB from kimchi
One hundred and twenty fi ve strains of LAB were isolated based on Gram-positive staining and catalase reaction.Antilisterial activity of the LAB was examined using well diff usion assay.Twenty LAB strains (SN2, SN3, SN5, SN8, SN10, SN12, SN14, SN19, SN22, SN43, SN46, SN47, SN48, SN52, SN54, SN59, SN65, SN66, SN68 and SN101) were shown to have an inhibitory eff ect on L. monocytogenes.Among them, the strain SN43 demonstrated the strongest antilisterial activity by showing the inhibition zone of over 6.5 mm in agar well diff usion assay (Table 1).We had identifi ed SN43 strain as Pediococcus pentosaceus T1 via rDNA PCR analysis in the previous study (18).Therefore, we used the culture of P. pentosaceus T1 for subsequent experiments.

Eff ect of P. pentosaceus T1 culture on L. monocytogenes proliferation
The inhibitory eff ect of the P. pentosaceus T1 culture against Listeria at diff erent doses was determined.Listeria monocytogenes growth greatly increased until 12 h in the control group, reaching 1.3•10 9 CFU/mL, a er which it gradually decreased to 5.8•10 7 CFU/mL (Fig. 1).In contrast, the treatment with the culture signifi cantly inhibited cell proliferation of Listeria at all tested concentrations.A er 6 h, the number of Listeria monocytogenes cells in all samples treated with the culture was 10 4 to 10 5 CFU/mL.The culture containing 1 % P. pentosaceus T1 showed a small increase of the number of Listeria monocytogenes cells a er 6 h (Fig. 1).However, the number of Listeria monocytogenes cells at the other mass per volume ratios of the culture (2, 3 and 4 %) continually decreased a er 6 h.In particular, 4 % of the culture caused a dramatic decrease a er 18 h, with less than 10 2 CFU/mL of Listeria cells.Thus, the treatment with P. pentosaceus T1 culture showed an eff ective inhibitory eff ect on Listeria monocytogenes in a dose-dependent manner.Our results indicate that the substances produced by P. pentosaceus T1 have an ability to inhibit Listeria monocytogenes.
Previously, we had isolated P. pentosaceus T1 as an antilisterial LAB from kimchi (18).In general, Pediococcus spp.have been known to exhibit antilisterial activity like Lactobacillus spp.(23).Recent studies have reported antilisterial activity of Pediococcus spp.from various sources (23,24).Pediococcus spp.were also known to inhibit other  pathogens such as Escherichia coli and Staphylococcus aureus (25).A major antimicrobial substance from Pediococcus spp.has been found to be a bacteriocin called pediocin, which is classifi ed into Class II (24).Its molecular size is less than 5 kDa containing 36-48 residues (25).However, our series of analyses including chromatography showed that P. pentosaceus T1-derived antilisterial material was proteinous substance with a molecular size of 23 kDa (18).In addition, liquid chromatography-mass spectrometry showed that P. pentosaceus T1-derived antimicrobial substance contained LysM domain (18), which is known to hydrolyze peptidoglycan, a cell wall component (26).Therefore, the active substance from our sample could be considered as a novel antilisterial substance, which is different from pediocin.LAB have been known to produce organic acids to inhibit other microbes, and these acids are possibly major antilisterial substances found in this work.However, we removed organic acids from the culture by ultrafi ltration (cut off < 3 kDa) to exclude this potential from our investigation.Lactic and acetic acids, at concentrations of 19.9 and 2.6 g/L respectively, before ultrafi ltration were completely removed a er ultrafi ltration from the culture.

Antilisterial eff ects of P. pentosaceus T1 culture and nisin in raw salmon medium
We compared the antilisterial eff ect of our sample (P.pentosaceus T1 culture) with that of nisin, a known bacteriocin with antilisterial activity.For this experiment, the raw salmon medium was inoculated with Listeria monocytogenes culture.Our culture and nisin were tested on Listeria in salmon in various doses to get minimum inhibitory concentrations (MIC).The number of Listeria cells in raw salmon medium signifi cantly decreased with both treatments in a dose-dependent manner.Based on our data, inhibitory eff ect of the culture on Listeria proliferation was shown to be stronger than that of nisin (Fig. 2), showing a more decreased number of Listeria cells at most of the treatment concentrations.MIC value of the culture was 20 mg/mL, while that of nisin was over 20 mg/mL.This result shows that P. pentosaceus T1 produces stronger antilisterial substances than nisin.Considering that our culture sample was not completely purifi ed, it is believed that real antilisterial activity of P. pentosaceus T1-derived active substance is underestimated.However, this result might be diff erent under diff erent conditions because optimal conditions for antimicrobial activity of nisin can be diff erent from those of our samples.Generally, nisin is known to be more active in acidic pH, which is related to its cell membrane permeation (27)(28)(29).In contrast, another study reported that nisin was rather less sensitive to foodborne pathogens such as Staphylococcus aureus and Listeria monocytogenes in the acidic pH (pH=4.5-5)(30).Our test of antilisterial activity of nisin and P. pentosaceus T1 culture was performed under the optimal growth conditions for L. monoctogenes (pH=7.2 and 35 °C) to count viable Listeria cells in the selective medium.Since our previous study had shown that the culture had broader spectrum of pH and temperature for maximal antilisterial activity (31), it could be more favourable as an antilisterial agent in food products.However, the application of antimicrobial materials on food is more complex due to various factors such as salt concentration and temperature, which aff ect the antimicrobial activity (30).Therefore, further detailed analysis of antilisterial eff ects of both samples will be performed in the next study.

Antilisterial eff ect of P. pentosaceus T1 culture on salmon fi llets inoculated with Listeria monocytogenes
Fish is highly susceptible to contamination with food pathogens like L. monocytogenes, causing a serious food--derived infection globally (32).We tested the antilisterial eff ect of our culture sample on a fi sh product contaminated with Listeria.Listeria-inoculated salmon fi llets were dipped and sprayed with our samples and sodium hypochlorite.A er incubation for 24 h at 4 °C, signifi cant decreases in the bacterial cell numbers were observed in sample-treated groups (Fig. 3).Treatment with sodium hy pochlorite, a disinfectant normally used in fi sh product processing, also showed a signifi cant reduction of Listeria cells.Interestingly, our culture caused a dramatic decrease in the number of Listeria cells a er the treatment (Fig. 3).The culture showed a much stronger inhibitory eff ect on Listeria growth compared with sodium hypochlorite (0.2 mg/mL, ACL-60G), which served as a positive control.However, we cannot directly compare the inhibitory effects of P. pentosaceus T1 culture and ACL-60G disinfectant because the concentrations of the two samples were diff erent in the treatments, where 0.2 mg/mL of ACL-60G is maximum allowed criterion in food processing.Nevertheless, P. pentosaceus T1 culture (6 g per 100 mL) clearly has an inhibitory eff ect on Listeria growth in salmon product.Therefore, P. pentosaceus T1 culture could be used as an inhibitor of Listeria contamination in fi sh products.This result is correlated with the data derived from the experiment performed in raw salmon medium (Fig. 2).
Our data suggest an applicable potential of P. pentosaceus T1 in raw fi sh product processing in food industry.

Eff ects of P. pentosaceus T1 culture on acidity and pH changes in kimchi during fermentation
To examine the eff ect of P. pentosaceus T1 culture on the maturation of kimchi, we determined the changes in the pH values and titratable acidity of kimchi preparations during fermentation at 10 °C for 105 days (Fig. 4).When the kimchi was prepared (0 day fermentation), the total acidity was 0.40-0.42%, and pH values were 5.4-5.6.The acidity of the fermented kimchi without the culture treatment (control) increased faster than that of the culture-treated kimchi, reaching 1.21 % (pH=4.02)within 21 days (Fig. 4), a er which it remained stable until the 105th day (1.28 %, pH=3.89).The acidity of the kimchi treated with the culture increased very slowly during fermentation, and reached up to 0.66 % at pH=4.66 in 105 days (Fig. 4).Fast increase of acidity during fermentation in control kimchi resulted in rapid decrease of pH value, while the culture treatment deferred these changes.According to our data, normal kimchi acidity reached 1 % at around day 15, but the culture-treated kimchi never reached that acidity during entire fermentation (Fig. 4a).
Sensory quality of kimchi depends on the duration of maturation, which means that its optimal quality is maintained only for a certain period of time.Today, kimchi is commercially produced and sold via distribution networks.The supply of fresh and tasty kimchi is one of the important challenges for producers.During fermentation of kimchi, prolonged maturation allows proliferation of other putrefying or spoilage bacteria as well as deterioration of quality, which increases the acidity above 1 % (33).Thus, acidity is used as a direct indicator of prolonged maturation (33).Usually, the acidity of kimchi in early fermentation stage is known to be in the range of 0.4-0.6 %, a er which it increases over 1 % at around day 30 (34).The acidity of kimchi in our work reached 1 % in 15 days, earlier than in the study by Shin et al. (34), suggesting faster maturation of the kimchi in our study.This result could be due to the diff erences in compositions of ingredients, which can aff ect LAB proliferation.Several studies have shown that optimal acidity is 0.5-0.75%, and kimchi with an acidity level over 1 % is recognized as unacceptable for consumption (34,35).Our results show that the treatment with the culture suppresses the increase of kimchi acidity during fermentation, maintaining it at 0.6 %.Normally fermented kimchi has an optimal acidity for maturity only for a short period of time (within 20 days), while culture-treated kimchi maintains optimal acidity for a longer time (105 days) (Fig. 4).Our data indicate that culture treatment could play an important role in controlling the acidity of kimchi with optimal maturity, which is desirable for distribution and storage of the product, and meets the commercial demands.Our fi ndings suggest that the shelf life of kimchi could be properly extended using the P. pentosaceus T1 as a starter culture under optimal maturation conditions.

Eff ect of P. pentosaceus T1 culture on total viable cell number in kimchi during fermentation
Total viable cell number in two kimchi sample groups (control and culture-treated group) was determined during fermentation (Fig. 5).The initial total cell number was around 1.0-1.6•10 6CFU/mL in control and culture-treated kimchi.Similar numbers of cells in the two samples show a clear diff erence a er 3 days of fermentation; cell number of control kimchi signifi cantly increased to 2.1•10 7 CFU/mL, while the cell number of culture-treated kimchi decreased to 5.4•10 5 CFU/mL.The increase of the cell number in control kimchi is correlated with the increase of acidity during early fermentation period.In control kimchi, cells continued to increase up to 2.7•10 8 CFU/mL and gradually decreased to 10 7 CFU/mL in 105 days, but the culture-treated kimchi had a stable cell number in the range of 2 to 5•10 5 CFU/mL during the investigated fermentation period (Fig. 5).This result shows that the culture treatment inhibited cell proliferation during fer mentation.The suppression of cell number occurred at an early stage of fermentation of samples treated with P. pentosaceus T1 culture, indicating that the culture could inhibit the growth of other LAB, which is responsible for overmaturation of kimchi, by producing antimicrobial substances including LysM domain.According to our results, P. pentosaceus T1 culture treatment could control the number of total cells, which can aff ect kimchi maturity or fermentation quality.

Growth inhibition of indicator LAB by P. pentosaceus T1 culture
The above data show that P. pentosaceus T1 culture suppressed the total cell viability in kimchi during fermentation.We then determined which LAB were inhibited by P. pentosaceus T1.Sixteen strains of LAB including Leuconostoc spp., Lactobacillus spp.and Weisella spp.were tested against P. pentosaceus T1.Well diff usion assay showed that the wells treated with P. pentosaceus T1 had signifi cant inhibition zones, over 12 mm, only against Leuoconostoc mesenteroides and Lactobacillus sakei (Fig. 6), suggesting that P. pentosaceus T1 has a strong antimicrobial activity on these two bacteria (Table 2).P. pentosaceus T1 also inhibited the other tested strains, although the inhibitory eff ects on them weakened compared to those on Leu.mesenteroides and L. sakei (Table 2).Leu.mesenteroides and L. sakei are the major LAB in kimchi, which are responsible for its maturation during fermentation (1).Leu.mesenteroides is known to be a predominant strain in the early or middle stages of kimchi fermentation, and L. sakei is one of the predominant strains in the late stage of fermentation (36,37).Moreover, the prolonged predominance of L. sakei can result in an excessively acidic taste and so texture of kimchi (38).Thus, our P. pentosaceus T1 is thought to control the change of microfl ora during fermentation of kimchi by inhibiting various LAB including Leu. mesenteroids and L. sakei.Our result indicated that the inhibition of LAB by P. pentosaceus T1 could prevent prolonged maturation, maintaining proper fermentation level which could be induced by other non-inhibited LAB.In addition, it would be interesting to analyze the changes of overall LAB microfl ora in the presence or absence of P. pentosaceus T1 during kimchi fermentation in the next study.

Sensory evaluation of kimchi
We showed that our P. pentosaceus T1 culture could positively aff ect the kimchi quality by controlling the acidity and bacterial cell number.However, if the treatment with P. pentosaceus T1 culture negatively aff ected kimchi sensory properties, its benefi cial eff ect such as antilisterial activity would be less meaningful.Therefore, we performed sensory evaluation of kimchi samples treated with P. pentosaceus T1 and the control.Sensory characteristics of kimchi include a proper combination of sour, sweet and salty tastes along with freshness, fi zzy mouthfeel, and crunchy texture.Sensory properties of kimchi are shown in Table 3. Kimchi treated with P. pentosaceus T1 received a higher score on most of the items in sensory evaluation including overall acceptability compared with the control.Specifi cally, sourness, off -fl avour  and fi zzy mouthfeel were greatly improved when P. pentosaceus T1 culture was added to the kimchi as a starter.
Colour diff erences between the starter and nonstarter kimchi preparations were not signifi cant, but the kimchi treated with P. pentosaceus T1 had a brighter appearance than the control.Interestingly, the culture-treated kimchi had a characteristic fl avour.This result showed that P. pentosaceus T1 delayed maturation stage, maintaining optimal kimchi quality.Accordingly, our data indicate that the use of P. pentosaceus T1 as a starter in kimchi preparation could prolong optimal conditions of kimchi fermentation, maintaining optimal sensory properties during storage or distribution.

Conclusion
In this study, we compared the antilisterial eff ect of Pediococcus pentosaceus T1 isolated from kimchi using salmon with a commercial bacteriocin (nisin) and a disinfectant (sodium hypochlorite).P. pentosaceus T1 was evaluated as a competitive antilisterial agent that showed a stronger inhibitory eff ect than nisin and the disinfectant.Current study also showed benefi cial eff ects of P. pentosaceus T1 used as a starter culture on kimchi quality.P. pentosaceus T1 eff ectively controlled maturation of kimchi by suppressing lactic acid bacteria such as Leu.mesenteroides and L. sakei, which are responsible for kimchi maturation.Moreover, P. pentosaceus T1 culture improved organoleptic quality of kimchi, as shown by sensory evaluation.We suggest that P. pentosaceus T1 be used as an antilisterial agent in fi sh products as well as a starter to control the fermentation of kimchi.

Fig. 2 .
Fig. 2. Antilisterial eff ects of P. pentosaceus T1 culture (T1) and nisin, which were added to L. monocytogenes-inoculated salmon medium at serially diluted concentrations to test the antilisterial eff ects for 24 h at 35 °C.Data are expressed as mean values± S.D. (N=3)

Fig. 5 .Fig. 6 .
Fig. 5. Eff ect of P. pentosaceus T1 culture (T1, 6 g per 100 mL) on total viable cell number in kimchi during fermentation.Kimchi samples were blended to prepare the juice.The juice samples were fi ltered, diluted, and spread onto plate count agar.The plate agar was counted a er 2 to 3 days of incubation at 25 °C.Data are expressed as mean values±S.D. (N=3)

Table 1 .
Antilisterial activity of LAB from kimchi

Table 3 .
Sensory evaluation of kimchi Signifi cant diff erences between the values of the same tested property (p<0.05,independent samples t-test).Results are expressed as mean values of scores on a 9-point hedonic scale *