Bromocresol purple (BCP) agar medium was used for the culture-based initial screening of LAB, whereas nutrient broth (NB) medium was used to cultivate pathogenic bacteria. Both media were purchased from Sigma-Aldrich (Sigma, St. Louis, MO, USA). The de Man, Rogosa, and Sharpe (MRS) agar medium used to isolate and culture LAB was purchased from Difco (USA). Other chemicals and reagents used were of high purity. Spectrophotometric measurements were made using an enzyme-linked immunosorbent assay (ELISA) instrument. A 18∼24 h grown culture of P. pentosaceus 4I1 was centrifuged followed by freeze-drying and desired test concentrations of its cell free supernatant (CFS) were prepared in double-distilled sterilized water.

The foodborne pathogens tested were Salmonella enterica ATCC-4731, Staphylococcus aureus KCTC-1621, Listeria monocytogenes KCTC-3569, Bacillus subtilis KCTC-3569, and Escherichia coli O157:H7. Test pathogens were cultured in NB medium and incubated at 37°C. For regular experiments, cultures were maintained on nutrient agar (NA) medium and stored at 4°C.

Zacco koreanus specimens were collected from a local river using catch per unit effort (CPUE) methods following taxonomic identification (Kim and Park, 2002) and isolation procedure of LAB (Cho et al., 2013). Initial screening was performed using the agar spot test on BCP agar medium (Trias et al., 2008). Selected LAB working cultures were then routinely grown in MRS broth and stored as stock cultures in MRS broth containing 15% glycerol (cryoprotective agent) in cryovials at -20°C. Ethical approval regarding “Animal Care and Use” was secured beforehand from the ethical committee of Daejeon National Science Museum Daejeon, South Korea with an approval # NSMD-MSIFP-KOR208. Experiments were performed in accordance with the relevant guidelines and methods.

Morphological, biochemical, and molecular characterizations of 4I1 were performed as we previously described (Bajpai et al., 2016a). In brief, 4I1 was identified by gram-staining and based on microscopic observations of colony shapes and cell morphologies (Holt et al., 1994). Further, 4I1 was characterized biochemically using API 50 CHL strips with API 50 CHL medium at the species level according to the manufacturer’s instructions (API 50 CHL, BioMerieux, France).

Partial 16S rRNA gene sequencing analysis was used to characterize 4I1 at the molecular level. Briefly, genomic DNA was isolated from 4I1 and then the 16S rRNA gene was amplified by PCR (Bajpai et al., 2016b). PCR reactions were carried out using a Biometra thermal cycler (M Biotech, Inc., Canada) with the following cycle parameters: an initial denaturation at 94°C for 2 min, followed by 35 cycles of denaturation at 94°C for 30 s, annealing at 52°C for 30 s, and elongation at 72°C for 1 min. The PCR products were sequenced and analyzed, and gene sequences obtained were compared in the National Center for Biotechnology Information (NCBI) for homology using BLAST and multiple-aligned with 16S rRNA gene sequences of different strains for similarity using the ClustalW program coupled with MEGA 5. The neighbor-joining method was used to construct the phylogenic tree using MEGA 5 software.

A detailed analysis of the chemical composition of the CFS of P. pentosaceus 4I1 was performed as described by Sjögren et al. (2003) using a gas chromatography–mass spectrometry (GC/MS) system (Jeol JMS 700 mass spectrometer, Agilent 6890N, Agilent Technologies, Santa Clara, CA, USA) equipped with a fused silica capillary column (30 m length × 0.25 mm ID × 0.25 μm film thickness). The GC–MS conditions used were as previously described (Bajpai et al., 2013). Relative proportions of the extract constituents were expressed as percentages by peak area normalization. Extract components were identified based on GC retention times and computer matching of mass spectra using the Wiley and National Institute of Standards and Technology Libraries for the GC–MS system used.

The standard agar well-diffusion method was used to determine the antibacterial efficacy of 4I1 CFS (Murray et al., 1995). To obtain the CFS of 4I1, supernatant from a 24 h grown culture of 4I1 was collected by centrifugation (8,000 × g; 10 min) and freeze-dried (lyophilized) (Saadatzadeh et al., 2013). Briefly, NA medium (20 mL) was poured into Petri-plates and allowed to solidify. Plates were then dried and 1 mL of standardized bacterial inoculum (10⁷ CFU/mL) was poured and uniformly spread onto agar surfaces, and then allowed to stand for 5 min. Wells were made in the agar by using a sterilized borer and 100 μL CFS of P. pentosaceus 4I1 was poured into each well against each of the tested pathogen. Negative controls were prepared using the same medium (sterilized distilled water or MRS medium) employed to dissolve the samples. Antibacterial activities were evaluated by measuring the diameters of zones of inhibition (including diameter of well: 6 mm) against the tested bacteria. All assays were performed in triplicate.

The MICs of CFS of P. pentosaceus 4I1 were determined using the twofold serial dilution method (Bajpai et al., 2013). Freeze-dried CFS of P. pentosaceus 4I1 (4 mg) was first dissolved in 1 mL distilled water as stock, and incorporated into NB medium to an initial concentration of 2,000 μg/mL, and then was serially diluted to 1,000, 500, 250, 125, 62.5, 31.25, 15.62, and 7.81 μg/mL concentrations of the CFS of 4I1. A 10 μL standardized bacterial suspension of each tested pathogen (10⁷ CFU/mL) was transferred to each tube. The treatment and control tubes which contained only bacterial suspensions were incubated at 37°C for 24 h. The lowest concentration of CFS, which did not show any visible growth of tested organisms after macroscopic evaluation, was determined as MIC, and was expressed in μg/mL. Further, the concentrations showing complete inhibition of visual growth of bacterial pathogens were identified, and 50 μL of each culture broth was transferred onto the agar plates and incubated at 37°C for 24 h. The complete absence of growth of bacterial colonies on the agar surface is the lowest concentration of the sample and was defined as MBC. Each assay in this experiment was replicated three times.

Freshly grown bacterial colonies of the selected pathogenic bacteria were inoculated in NB medium at 37°C for 24 h, and then bacterial cultures were serially diluted to 10⁷ CFU/mL (Shin et al., 2007). To determine the effect of CFS of P. pentosaceus 4I1 on cell viabilities, two selected foodborne pathogenic bacteria, S. aureus KCTC-1621 and E. coli O157:H7 were used. Briefly, each of the tubes containing bacterial suspension (10 μL; approximately 10⁷ CFU/mL) of S. aureus KCTC-1621 and E. coli O157:H7 was inoculated with 100 μL of CFS of 4I1 at its MIC in 890 μL NB broth at 37°C. Samples for viable cell counts were taken out at 0, 40, 80, 120, 160, and 200 min time intervals. Viable plate counts were monitored on NB agar as we previously described (Bajpai et al., 2013). Colonies were counted after incubation for 24 h at 37°C. The controls were inoculated without CFS of 4I1 for each pathogenic bacteria using the same experimental condition. Assay were performed in triplicate.

The effects of CFS of P. pentosaceus 4I1 on the efflux of potassium ion from S. aureus KCTC-1621 and E. coli O157:H7 were determined as we previously described (Bajpai et al., 2013). Concentration of free potassium ion in bacterial suspensions of S. aureus KCTC-1621 and E. coli O157:H7 was measured after exposing bacterial cells to CFS of P. pentosaceus 4I1 at their MICs in sterile peptone water (8.5 g NaCl + 1 g peptone in 1 L sterilized distilled water) for 0, 30, 60, 90, and 120 min. At each pre-established interval, the extracellular potassium concentration was measured by a photometric procedure using the Calcium/Potassium kit (Quantofix, GmbH, Wiesbaden, Germany). Similarly, control was also tested without adding CFS. Results were expressed as the amount of extracellular free potassium (mmol/L) in the growth media in each interval of incubation.

The release of 260-nm-absorbing materials from S. aureus KCTC-1621 and E. coli O157:H7 cells was monitored in aliquots of 2 mL of the bacterial inocula in sterile peptone water (0.1 g/100 mL). The reaction solution containing MIC of CFS of P. pentosaceus 4I1 was incubated at 37°C. At 0, 30, and 60 min time interval of treatment, cells were centrifuged at 3,500 × g, and the absorbance of the obtained supernatant was measured at 260 nm using a 96-well plate ELISA reader (Bajpai et al., 2013). Controls were treated in the same manner without CFS of P. pentosaceus 4I1. Results were expressed in terms of optical density (OD) of 260-nm absorbing materials in each interval with respect to the ultimate time.

The effects of CFS of P. pentosaceus 4I1 on cell membrane permeability of S. aureus KCTC-1621 and E. coli O157:H7 were determined as described previously (Patra et al., 2015), and expressed in terms of relative electrical conductivity. Upon action of CFS on the cell membrane of tested pathogens, it may cause drastic release of cytosolic materials such as protein, DNA and other essential metabolites, resulting in the cell death. Prior to the assay, cultures of test pathogens were incubated at 37°C for 10 h, followed by centrifugation (5,000 × g) for 10 min, and washed with 5% glucose solution (w/v) until their electrical conductivities reached close to 5% glucose solution to induce an isotonic condition. MICs of CFS of P. pentosaceus 4I1 acquired for both the tested pathogens were added to 5% glucose (isotonic solution), incubated at 37°C for 8 h, and the electrical conductivities (L_a) of the reaction mixtures were determined. Further, electrical conductivities of the bacterial solutions were measured at 2 h of intervals for a total duration of 8 h (L_b). The electrical conductivity of each test pathogen in isotonic solution killed by boiling water for 5 min served as a control (L_c). The relative electrical conductivity was measured using an electrical conductivity meter. The permeability of bacterial membrane was calculated according to the following formula:

Scanning electron microscopic (SEM) study was executed according to Kim et al. (2007) to examine the effects of CFS of P. pentosaceus 4I1 on the morphological changes in the cell wall of the selected pathogens, S. aureus KCTC-1621 and E. coli O157:H7. Control samples were prepared without CFS of P. pentosaceus 4I1. Microscopic examination was performed using a S-4300 SEM Analyzer (Hitachi, Japan).

To determine whether CFS of P. pentosaceus 4I1 has growth phase-independent inhibitory effects, CFS of P. pentosaceus 4I1 at MIC + pathogenic bacteria (overnight grown single bacterial colony) were inoculated in their respective culture tubes containing 20 mL sterile NB medium followed by incubation at 37°C until 24. After every 4 h, samples were withdrawn and analyzed for the bacterial counts of pathogenic strains. Colonies were counted after growth at 37°C up to 24 h, and the log10 CFU was plotted against incubation period to prepare growth curves of individual strains. As a control, pathogenic bacterial strains without CFS of P. Pentosaceus 4I1 were used.

Pathogenic bacterial strains were inoculated and grown in NB broth and evaluated for their biofilim formation ability, and effects of CFS of P. pentosaceus 4I1 on biofilm formation by pathogenic bacteria was determined following the modified method of Aoudia et al. (2016). A 2 mL of bacterial cultures of S. aureus KCTC-1621 and E. coli O157:H7 (10⁶ CFU/ml) grown in NB broth were added to each glass test-tube. On the other hand, to determine the effect of CFS of P. pentosaceus 4I1 on biofilm formation, CFS at MIC mixed with 2 mL of bacterial cultures of S. aureus KCTC-1621 and E. coli O157:H7 (10⁶ CFU/ml) grown in NB broth and was added to each glass test-tube. While, 2 mL of NB broth was added in blank test-tubes without bacterial culture (negative control). The tubes were incubated for 48 h at 30°C. To quantify the biofilm formation, the tubes were gently washed three times with 2 mL of sterile distilled water, and attached bacteria were fixed with 2 mL of methanol for 15 min, and then, tubes were emptied and air dried at room temperature or oven dried at 60°C for 30–45 min. Subsequently, 2 mL of a 2% (v/v) crystal violet solution was added to each well and held at ambient temperature for 15 min. Excess stain was then removed by placing the test tubes under gently running tap water. A 2 mL of ethanol was used for destaining. The OD of released adherent cells was measured at 595 nm. Each assay was performed in three individual times on different days under the same conditions, and the negative control was performed in uninoculated NB broth. The cut-off OD was defined as the mean OD value of the negative control. Based on the OD, strain was confirmed for biofilm production ability in presence of CFS as a no-biofilm producer or biofilm producer (strong or week) (OD < OD of negative control confirmed as no-biofilm producer), (ODC < OD × 2 OD of negative control confirmed as week biofilm producer), moderate (2 × OD of negative control < OD ≤ 4 × OD of negative control confirmed as moderate biofilm producer) (4 × OD of negative control < OD confirmed as strong biofilm producer).

All experiments were performed in triplicate and results were expressed the mean ± SD following one-way ANOVA coupled with Duncan’s multiple test.

Small yellow colonies of similar sizes that appeared on BCP agar using pour-plating method confirmed the presence of the LAB isolate 4I1 which was confirmed to be coccus-shaped by microscopic evaluation. Biochemical analysis of 4I1 was performed using the API 50 CHL strip kit and a selected strain was identified as a gram-positive and rod-shaped isolate (Table 1). API web software confirmed that strain 4I1 utilized carbohydrates, including L-arabinose, D-ribose, D-xylose, D-galactose, D-glucose, D-fructose, D-mannitol, D-sorbitol, N-acetylglucosamine, amygdalin, arbutin, salicin, D-cellobiose, D-maltose, D-lactose, D-melibiose, D-saccharose, D-trehalose, D-raffinose, gentiobiose, and D-turanose (Table 1). Color change from violet to yellow in the strip capsule indicated complete fermentation of sugar by 4I1. Molecular analysis using partial 16S rDNA gene sequencing showed the selected strain displayed 99.9% similarity with different Pediococcus spp. (Figure 1), thus, the strain was finally characterized as P. pentosaceus 4I1. The derived sequence was submitted to GenBank with nucleotide accession number KT372700.

The GC–MS analyses of the CFS of P. pentosaceus 4I1 identified seven different components accounted for 99.98% of total CFS. The CFS of P. pentosaceus 4I1 yielded compounds largely amino acids, organic acids, and fatty acids, as well as pyrrol derivatives which included (S)-2-Hydroxypropanoic acid (24.28%), caprolactam (8.83%), D-valine (28.53%), D-leucine (35.71%), 3-pyrrolidin-2-yl-propionic acid (2.43%), pyrrolo [1,2-a]pyrazine-1,4-dione, hexa (19.34%), and 9-octadecenoic acid (8.29%).

The antibacterial activity of CFS of P. pentosaceus 4I1 against the tested foodborne pathogenic bacteria was confirmed by the presence or absence of inhibition zones on the agar well plates. As presented in Figure 2, CFS of P. pentosaceus 4I1 exhibited potent inhibitory effects against all the tested foodborne pathogenic bacteria. In this assay, P. pentosaceus 4I1 exerted consistent antibacterial effects against both gram-positive and gram-negative bacteria, with zone of inhibition diameters ranging from 16.5 to 20.4 mm (Figure 2). Sterilized distilled water and/or MRS medium used as a negative control had no inhibitory effect.

The MIC assay revealed different susceptibilities of tested pathogens to the CFS of P. pentosaceus 4I1, and exhibited potent inhibitory effect as MIC and MBC values. In this assay, the MIC and MBC values of 4I1 CFS against the tested foodborne pathogens were ranged from 250 to 1,000 μg/mL (Figure 3). Furthermore, the CFS of 4I1 exhibited potential antibacterial effects as reflected by MIC and MBC values against all the tested pathogens (Figure 3). Interestingly, one of the pathogens S. enterica ATCC-4731 was found to be highly susceptible pathogen to the CFS of P. pentosaceus 4I1. Notably, both gram-positive and gram-negative bacteria were inhibited by the CFS of P. pentosaceus 4I1.

In this assay, the CFS of P. pentosaceus 4I1 when inoculated at MIC, exhibited significant inhibitory effects against the growth of tested bacterial pathogens, S. aureus KCTC-1621 and E. coli O157:H7, as confirmed by reduced bacterial cell viability when inoculated at MIC (Figure 4). Exposure to CFS of P. pentosaceus 4I1 for 0 to 80 min did not elicit severe inhibition of cell viability, but remarkable declines in the cell viable counts of S. aureus KCTC-1621 and E. coli O157:H7 was observed after exposure to the CFS of P. pentosaceus 4I1 for 160 min. Interestingly, the exposure to CFS of P. pentosaceus 4I1 for 200 min completely inhibited the cell viabilities of both tested pathogens (Figure 4B).

This test assay confirmed the antibacterial effect of the CFS of P. pentosaceus 4I1 by revealing K⁺ release from S. aureus KCTC-1621 and E. coli O157:H7 versus the control (Figure 5). The CFS of P. pentosaceus 4I1 added at MIC to cell suspensions of tested pathogenic bacteria resulted in rapid K⁺ release from the bacterial cells following protracted steady loss (Figures 5A,B). However, no leakage of K⁺ was observed from S. aureus KCTC-1621 and E. coli O157:H7 controls (Figure 5).

Release of 260-nm absorbing materials (DNA and RNA) from the cells treated with specific antimicrobials may be an indication of bacterial cell death. Hence, we evaluated the effects of the CFS of P. pentosaceus 4I1 on the release of 260-nm absorbing materials from S. aureus KCTC-1621 and E. coli O157:H7 treated at MIC. Interestingly, exposure of CFS of P. pentosaceus 4I1 to S. aureus KCTC-1621 and E. coli O157:H7 caused rapid loss of 260-nm absorbing materials from the bacterial cells. The ODs of culture filtrates of S. aureus KCTC-1621 and E. coli O157:H7 cells exposed to the CFS of P. pentosaceus 4I1 at 260 nm, revealed a significant time-dependent increase in the release of 260-nm-absorbing materials (Figure 6). However, no changes in the OD of control cells of tested pathogens were observed. Notably, exposure for 60 min to the CFS of P. pentosaceus 4I1 caused about a twofold increase in the OD of treated bacterial cell culture filtrates as compared with their respective controls (Figures 6A,B).

This assay visualized the effect of the CFS of P. pentosaceus 4I1 at MIC on the membrane permeabilities of tested pathogens as determined by their relative electrical conductivities. As was expected, the CFS of P. pentosaceus 4I1 exhibited time-dependent inhibitory effect on the membrane permeabilities of the tested pathogens, and the relative electrical conductivity of each tested pathogen was increased time-dependently. Furthermore, the CFS of P. pentosaceus 4I1 exhibited a greater inhibitory effect on cell membrane of S. aureus KCTC-1621 (Figure 7A) than E. coli O157:H7 (Figure 7B) as indicated by their relative electrical conductivity values (Figure 7). No relative electrical conductivity was observed in untreated controls. In this assay, the CFS of P. pentosaceus 4I1 showed an ability to disrupt the plasma membranes of both tested bacteria as confirmed by the changes observed in the relative electrical conductivity values.

Since exposure to an antimicrobial agent may lead to disruption of bacterial cell wall, we turned to SEM analysis to investigate further the effect of the CFS of P. pentosaceus 4I1 on the cell wall physiologies and morphologies of S. aureus KCTC-1621 and E. coli O157:H7 cells (Figure 8). As was expected, control bacterial cells not exposed the result of CFS of 4I1 had regular smooth surfaces (Figures 8A,B), whereas those treated with CFS of 4I1 at MIC showed cell wall damage and lysis (Figures 8C,D).

The growth phase-dependent behavior of CFS of P. pentosaceus 4I1 on the tested pathogens was measured at different time intervals as demonstrated in Supplementary Figure S1. In both, mono- and co-culture tubes, the lag phase for P. pentosaceus 4I1 lasted about 3–4 h, whereas pathogenic strains grew rapidly (Supplementary Figure S1). However, decreases in pathogen growths were observed in the presence of CFS of P. pentosaceus 4I1 after incubation for 6–12 h, and pathogen growth was reduced by almost four log cycles within 24 h of incubation as compared with the control. The effect of CFS of P. pentosaceus 4I1 on the growth of S. aureus KCTC-1621 was significantly more marked than on that of E. coli O157:H7.

The effects of CFS of P. pentosaceus 4I1 on biofilim formation by S. aureus KCTC-1621 and E. coli O157:H7 were analyzed by using crystal violet assay, which showed a potent inhibitory effect of the CFS of 4I1 on the biofilm formation ability of both the tested pathogenic bacteria (Supplementary Figure S2). As a result, the CFS of P. pentosaceus 4I1 induced inhibition of biofilm formation by S. aureus KCTC-1621 and E. coli O157:H7 (Supplementary Figure S2) as confirmed by a significant (p < 0.05) reduction in OD values, while in control without CFS of P. pentosaceus 4I1, S. aureus KCTC-1621 and E. coli O157:H7 showed a strong biofilm formation ability (Supplementary Figure S2). Our results indicate that CFS of P. pentosaceus 4I1 has potential effect against biofilm formation ability of S. aureus KCTC-1621 and E. coli O157:H7.