Introduction

A wide range of traditional or indigenous fermented foods and beverages, representing approximately 5-40% of the total daily food intake, are produced and consumed globally. Fermented foods can play an important role in strengthening the livelihoods and improving the nutrition and social well-being of millions of people worldwide, and can provide food security for vulnerable and marginalized populations, in particular. For example, doenjang (fermented soy paste), a traditional Korean fermented food, is an excellent source of nutrition that is used as a flavoring ingredient in Korea, similar to miso (Japanese fermented soybean paste) in Japan and tempeh (Indonesian fermented soybean food) in Indonesia [14]. Doenjang has received considerable attention owing to its beneficial health-promoting properties, such as its anticancer, antioxidant, and fibrinolytic activities [23]. However, concerns remain about the safety of consuming fermented soybean products such as doenjang because these products are often contaminated with various foodborne pathogens, including Bacillus cereus [18].

B. cereus is a spore-forming, gram-positive bacterium that is responsible for diarrheal (heat-labile) and emetic (heatstable) food poisoning, which are caused by enterotoxins and emetic toxins, respectively [17,28]. Its ability to cause diarrhea is attributed to various enterotoxins and virulence factors such as non-hemolytic enterotoxin (Nhe), hemolysin BL, cytotoxin-K, and enterotoxin FM (EntFM), a group of heat-labile proteins that cause abdominal cramps, nausea, and, rarely, vomiting and watery diarrhea [7,8]. The Nhe complex is composed of the NheA, NheB, and NheC proteins, which are encoded by nheA, nheB, and nheC [33]. The heat shock protein GroEL, which provides protection against physiological and heat stresses and is required for survival of B. cereus, has been used to reveal phylogenetic relationships between bacteria [12,26]. Emetic foodborne illness is induced by the small cyclic heat-stable toxin cereulide, which causes vomiting and nausea; the cereulide synthetase gene (ces) is only found in emetic toxin-producing B. cereus [11,17].

B. cereus is commonly found in contaminated foods containing fermented soybeans, such as doenjang, and the South Korean food authority has reported that ingestion of more than 104 CFU of B. cereus per gram of fermented soybean products may cause food poisoning [27,30]. Because spores of B. cereus are highly resistant to various stresses (heat, cold, radiation, desiccation, and disinfectants) and show excellent adhesion to food surfaces, B. cereus contamination is very difficult to control in the fermented soybean food industry [10]. However, some strains of bacteria produce antibacterial substances (bacteriocin-like peptides and antimicrobial lipopeptides) that have few or no undesirable effects, such as inhibition of the growth of fermenting bacteria or decrement of food quality. These strains have been investigated as beneficial additions to starter cultures in the industrial-scale production of fermented soybean foods [5,29]. For example, B. amyloliquefaciens strain RD7-7, isolated from traditional fermented soybean foods, exhibits high enzymatic and antibacterial activities against foodborne pathogens.

Previous studies have characterized the prevalence of B. cereus in fermented soybean products, but little research has been conducted on the detection or characterization of the expression of genes encoding B. cereus toxins in these products [19,25]. Therefore, we investigated the toxin gene profiles and toxin expression levels of B. cereus in fermented soybean samples co-cultured with B. amyloliquefaciens RD7-7. Strain RD7-7 exhibited strong antibacterial activity against B. cereus and reduced its toxin expression. Thus, B. amyloliquefaciens RD7-7 shows potential for use as a starter strain to prevent the growth of B. cereus during the industrial production of fermented soybean foods and has numerous potential applications for food preservation.

Materials and Methods

Bacterial Strains and Culture Conditions

B. amyloliquefaciens RD7-7 KACC 92071P, B. amyloliquefaciens KACC 15866 (used as a reference strain), and B. cereus KACC 10004 (indicator strain) were grown in Luria-Bertani (LB) broth (Difco, Becto n Dickinso n, Sparks, MD, USA) or on LB agar medium at 30℃. The strains were subcultured at 30℃ for 24 h in LB broth and streaked on nutrient agar (NA; Difco, Becton Dickinson) plates, and then incubated at 30℃ for 24 h before use.

Transmission Electron Microscopy (TEM)

Bacterial cultures were centrifuged at 930 ×g for 20 min and then washed twice with distilled water (DW). A carbon Formvar-coated 200-mesh copper grid was rendered hydrophilic by high-voltage glow discharge (JFC-1100E Ion Sputter, Jeol Co., Tokyo, Japan). Bacteria on the grid were negatively stained with 2% uranyl acetate for 315 sec and then rinsed three times with DW. The sample was examined under a Tecnai 12 transmission electron microscope (Philips, Eindhoven, The Netherlands) at an acceleration voltage of 120 kV.

Scanning Electron Microscopy (SEM)

Bacterial cells grown on LB for 24 h were fixed with Karnovsky’s fixative solution at 4℃ for 24 h and then washed three times for 10 min each with 0.05 M cacodylate buffer (pH 7.2). The fixed specimens were post-fixed with 1% osmic acid for 2 h at 4℃, and then washed three times with DW for 10 min each. Specimens were dehydrated using a graded ethanol series (50%, 75%, 90%, and 95%) for 30 min each, with two final 30 min treatments with 100% ethanol. The specimen was transited with 100% amyl acetate at room temperature (RT) two times for 30 min each. After critical-point drying and gold coating, the sample was observed using the scanning electron microscope (Hitachi S-2460N; Hitachi. Ltd., Tokyo, Japan) at an acceleration voltage of 20 kV.

Activity Against Pathogenic Bacteria

Antimicrobial activity against several pathogenic bacteria was measured using the agar well diffusion method [13,31]. Bacterial cultures grown for 24 h in LB broth were inoculated (3% (v/v)) to soft nutrient agar containing 0.7% agar, which was melted and then cooled to approximately 45–50℃. After a very vigorous homo genizatio n, the ino culated agar w as p oured i nto standard plastic Petri dishes, and 3-mm diameter wells were bored into the agar plates. The plates were incubated for 24 h at 37℃ and then the diameters of the inhibition halos were measured in centimeters.

Co-Inoculation of B. cereus and B. amyloliquefaciens

For co-culture experiments, bacterial strains were cultured at 30℃ for 24 h in LB broth. When the optical density at 600 nm (OD600) reached approximately 0.4, which is indicative of a bacterial density of 107 CFU/ml, B. cereus 5 × 104 CFU/ml (0.5%) was inoculated with different concentrations of B. amyloliquefaciens RD7-7 (1.25 × 104 to 1.0 × 105 CFU/ml; 0.125%, 0.25%, 0.5%, and 1%) or B. amyloliquefaciens KACC 15866 (1.25 × 104 to 1.0 × 105 CFU/ml; 0.125%, 0.25%, 0.5%, and 1%) in LB broth at 30℃ for 24 h.

Isolation and Enumeration of B. cereus Cells by Selective Cultivation and Real-Time qPCR

Following 10-fold serial dilution in 0.85% sterile saline, 100 μl aliquots of each diluted sample were spread on B. cereus chromogenic medium, CHROMagar B. cereus (CHROMagar Microbiology, Paris, France), and incubated at 30℃ for 24 h for isolation and identification of B. cereus. The co-cultured cells were harvested by centrifugation at 8,000 ×g a t 4℃ fo r 10 min and the cell pellets were washed with 1 ml of sterile DW and centrifuged. After centrifugation, total genomic DNA was extracted using a QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. The purified DNA was eluted in 100 μl of sterile DW. DNA isolated from B. cereus was detected using a B. cereus Detection Real-time PCR Kit (JSBC050; Jinsung Uni-Tech, Korea). The reactions were performed using a C1000 Thermal Cycler equipped with a CFX96 Real-Time System (Bio-Rad, CA, U SA) in a f inal v o lume o f 20 μl containing 1 μl of genomic DNA, an appropriate amount of each primer set (10 pmol for groEL) and probe (10 pmol for groEL; HEX), 10 μl of TaqMan universal PCR master mix, and DW. The conditions for the PCR amplifications were as follows: 50℃ for 2 min, 95℃ for 10 min, followed by 40 cycles of 95℃ for 30 sec and 55℃ for 50 sec. A calibration curve for B. cereus KACC10004 was prepared for quantification. The curves were constructed using genomic DNA from B. cereus cultures in LB broth at different concentrations (104, 105, 106, and 107 CFU/ml) as determined by microbial counts in Plate Count Agar (Difco). DNA was isolated from 1 ml of each dilution. Cycle threshold (Ct) values were plotted against the colony forming units (CFUs) [12].

Quantitation of Expression of Toxin-Related Genes in B. cereus by Real-Time qPCR

Total RNA was isolated from cells using the RNeasy Plus Mini Kit (Qiagen) according to the manufacturer’s instructions. Isolated RNA was quantified on a Synergy Mx microplate reader (BioTek Instruments, Winooski, VT, USA). The first-strand cDNA was synthesized from 1 μg of isolated RNA template and amfiRivert Platinum cDNA synthesis Master Mix (GenDEPOT; Barker, TX, USA). A subset of genes was amplified with amfiEco Taq DNA polymerase (GenDEPOT) and the gene-specific primers listed in Table 1. The qPCR analyses were performed with the C1000 Thermal Cycler equipped with a CFX96 Real-Time System (Bio-Rad) in a total volume of 10 μl containing 5 μl of iQ SYBR Green Supermix (Bio-Rad), 200 nM of each of the primers listed in Table 1, and 2 μl of cDNA. The PCR cycling conditions were as follows: 50℃ for 2 min, 95℃ for 3 min, followed by 45 cycles of 95℃ for 30 sec, 55℃ for 30 sec, and 72℃ for 30 sec. 16S rRNA was used as an internal control for data normalization.

Table 1.aPrimer sequences have been described previously [12]. bPrimer sequences have been described previously [12,23]. cPrimer sequences have been described previously [12,17].

Table 2.Individual cycle threshold (Ct) values correspond to standard curves derived from the B.cereu s PCR detection kit with serial 10-fold dilutions. 1)B. cereus KACC 10004. 2)Standard curve of B. cereus KACC 10004. 3)B. amyloliquefaciens KACC 15866. 4)Each value indicates the mean ± SD of three replicates. Within each row, means indicated with different superscript letters differ significantly p < 0.05 (one-way ANOVA, followed by Duncan’s multiple comparison test).

Production of Fermented Soybean Products and Isolation of Bacteria

The soybeans (500 g) were washed and soaked in potable water for 16 h at RT. After the soaking water was drained, the weight of the soaked soybeans had increased by approximately 2-fold. The drained soybeans were steamed for 3 h at 100℃ and cooled below 40℃. B. cereus and B. amyloliquefaciens RD7-7 cultures were inoculated to a final cell density of approximately 107 CFU/ml (OD600 = 0.4) after 24 h of incubation in LB broth at 30℃. After cooling, the surfaces of the cooked soybeans were inoculated with 1% (v/w) (105 CFU/g) inoculum (mixed culture of B. amyloliquefaciens RD7-7 and B. cereus at a ratio of 10:0 (sample 2), 0:10 (sample 3), 9:1 (sample 4), or 5:5 (sample 5)). The sample 1 group was treated with sterile saline solution (0.85% NaCl), as a negative control. The mixtures were fermented at 35℃ for 36 h. To extract bacterial DNA, RNA, and protein, 50 g of fermented soybeans was mixed with 450 ml of sterile saline and shaken for 15 min and then filtered through No. 2 Whatman filter paper. The extracts were centrifuged at 8,000 ×g for 10 min and the residue was collected in 1 ml of sterile saline solution.

Production of Fermented Buckwheat Soksungjang Products and Isolation of Bacteria

To prepare buckwheat meju (Fermentation agent for Deonjang), 2.5 kg of soybeans was cleaned and soaked in drinking water for 24 h at RT and then drained. The drained soybeans were steamed at 100℃ for 5 h and then crushed. Then, 1.4 kg each of crushed soybean and buckwheat (soybean: buckwheat = 7:3) was mixed in water to make a paste. The surfaces of the mixed samples were inoculated with 1% (v/w) (105 CFU/g) inoculum (mixed culture of B. amyloliquefaciens RD7-7 and B. cereus at a ratio of 0:10 (sample 2) or 9:1 (sample 3)). The sample 1 group (NT) was treated with sterile saline solution (0.85% NaCl), as a negative control. The paste was molded into 500 g discs with a consistent diameter of 12 cm and a thickness of 3 cm. After the discs were dried in the shade for 24 h, they were fermented for 7 days at 30℃ with 80% relative humidity, and dried for 1 day. Then, the discs were crushed, and 1 kg of buckwheat meju, 220 g of sun-dried salt, and 1.2 L of DW were mixed and naturally fermented in a jar (Kalsantoki, Hongsung, Korea). Samples were collected at 7-day intervals for 8 weeks. To extract bacterial RNA and protein, 50 g samples of buckwheat soksungjang were mixed with 450 ml of sterile saline and shaken for 15 min, and then filtered by suction through a Whatman No. 2 filter paper. Samples were harvested by centrifugation at 8,000 ×g for 10 min and the residue was resuspended in 1 ml of saline solution.

Total Protein Extraction and Western Blot Analysis

To measure expression of the B. cereus atpB gene, co-cultures of B. cereus and B. amyloliquefaciens were incubated under the culture conditions described above. The cultured cells were collected by centrifugation at 10,000 ×g for 5 min, and then the resulting pellet was resuspended in 30 μl of sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer. For extraction ofcellular proteins, bacteria isolated from fermented soybean products and buckwheat soksungjang samples were centrifuged at 10,000 ×g for 10 min and the collected pellets were resuspended in 30 μl of SDS-PAGE sample buffer, boiled for 10 min, and analyzed by western blotting. Equal amounts of protein from each sample were resolved on 12% SDS-PAGE gels and then subjected to western blot analysis. The separated proteins were transferred to a nitrocellulose membrane by electroblotting (120 V, 1 h) using a Mini Trans-Blot Electrophoretic Transfer Cell (Bio-Rad). The nitrocellulose membranes were blocked with 3% bovine serum albumin in Tris-buffered saline with 0.1% Tween-20 (TBS-T) for 1 h and then incubated for 1 h with a primary antibody, polyclonal anti-AtpB (the beta subunit of ATP synthase) (1:3,000; Agrisera, Vannas, Sweden). After washing with TBS-T, the membranes were incubated with horseradish-peroxidase-conjugated goat IgG secondary antibodies (1:3,000; Bio-Rad) for 1 h. The blots were analyzed using an enhanced BM chemiluminescence blotting substrate (POD; Roche, Mannheim, Germany).

Statistical Analysis

All statistical analyses were performed in duplicate, using three replicates of each experiment. Using Statistical Package for the Social Sciences (SPSS), ver. 12.0, one-way analysis of variance (ANOVA) was applied to determine whether differences between treatments were significant. The means were compared using Duncan’s multiple comparison test, and p < 0.05 was considered to indicate statistical significance.

Results and Discussion

Morphological Characteristics of B. amyloliquefaciens RD7-7 and Its Activity Against Pathogenic Bacteria

We previously demonstrated that B. amyloliquefaciens RD7-7, isolated from rice doenjang, a traditional fermented soybean paste, has the potential to improve the quality of fermented soybean food products. When this strain was added to the starter culture, the resulting food products showed high enzymatic activity and amino-type nitrogen contents [19]. To further characterize B. amyloliquefaciens RD7-7, we performed a morphological analysis using TEM and SEM techniques. As shown in Figs. 1A and 1B, B. amyloliquefaciens RD7-7 is a rod-shaped bacterium with flagella, similar to other B. amyloliquefaciens strains. We next evaluated the antibacterial activity of B. amyloliquefaciens RD7-7 against various pathogenic strains of bacteria, such as B. cereus, Staphylococcus aureus, Escherichia coli, Salmonella enterica, and Listeria monocytogenes. B. amyloliquefaciens RD7-7 had antagonistic activity towards all five pathogenic bacteria and produced inhibition zones >1.3 cm; in particular, it produced the largest zone of inhibition (1.51-1.80 cm) against B. cereus (Fig.1C ). This result indicates that B. amyloliquefaciens RD7-7 has broad-spectrum antimicrobial activity against gram-positive and gram-negative bacteria and the highest antimicrobial potential against B. cereus. Thus, B. amyloliquefaciens RD7-7 shows potential as a candidate novel antimicrobial agent to prevent or treat many important bacterial diseases.

Fig. 1.Microscopy and antimicrobial activity analyses of B. amyloliquefaciens RD7-7.

B. cereus Survival and Toxin Gene Expression in Co-Culture with B. amyloliquefaciens RD7-7

To test whether B. amyloliquefaciens RD7-7 could inhibit the growth and toxin production of B. cereus, we co-cultured the two strains in broth and quantified B. cereus survival and toxin gene expression.

As shown in Fig. 2, the growth of B. cereus was inhibited when it was grown in co-culture with B. amyloliquefaciens RD7-7, but the survival of B. cereus was not affected by co-culture with B. amyloliquefaciens KACC 15866. The survival and growth of B. cereus decreased to approximately 7.01, 6.01, and 3.80 log CFU/ml in the presence of 0.25%, 0.5%, and 1% B. amyloliquefaciens RD7-7, respectively, compared with co-culture with B. amyloliquefaciens KACC 15866 (8.31, 8.37, and 8.42 log CFU/ml) and the control (8.52 log CFU/ml). The initial bacterial populations of B. amyloliquefaciens KACC 15866 and B. amyloliquefaciens RD7-7 did not differ significantly after co-cultivation (data not shown). Thus, B. amyloliquefaciens RD7-7 showed potent antibacterial activity against B. cereus.

Fig. 2.Survival of Bacillus cereus KACC10004 incubated with B. amyloliquefaciens RD7-7 and B. amyloliquefaciens KACC 15866, after 24 h.

Detection of B. cereus was performed using a B. cereus real-time PCR kit using the groEL probes, and a linear standard curve for real-time PCR amplification was achieved for purified DNA of the B. cereus KACC10004 strain at concentrations ranging from 2 × 105 to 2 × 108CFU/ml. The Ct values for B. cereus groEL after co-culture with 0.25%, 0.5%, and 1% B. amyloliquefaciens RD7-7 at 6.68, 5.90, and 5.23 log CFU/ml were 23.14 ± 0.02, 26.24 ± 0.07, and 28.92 ± 0.01, respectively; thus, the Ct values increased as the template quantity decreased. The absolute value of the correlation coefficient of the calculated standard curve generated from B. cereus KACC10004 groEL was 0.9641. In contrast, the Ct values for groEL from B. cereus co-cultures with B. amyloliquefaciens KACC 15866 were similar to those of B. cereus KACC10004 (control strain).

To examine the expression of B. cereus toxin-related genes in co-culture with B. amyloliquefaciens RD7-7, we erformed in vitro transcription and translation reactions.

Expression levels of transcripts of the B. cereus toxin-related genes groEL, nheA, nheC, and entFM were downregulated in B. cereus co-cultured with B. amyloliquefaciens RD7-7, whereas the B. cereus KACC10004 control and co-culture with B. amyloliquefaciens KACC 15866 did not differ significantly (Fig. 3A). The ces gene encoding the emetic toxin was not detected by qPCR analysis in any of the bacterial strains (data not shown); thus, B. cereus KACC10004 is a diarrheal toxin-producing strain. groEL is common to all strains of B. cereus, whereas ces is detected only in emetic toxin-producing B. cereus; therefore, simultaneous amplification of groEL and ces could facilitate detection and differentiation of non-emetic and emetic B. cereus in food products [23]. The nheA and nheC genes of the Nhe complex was detected in most B. cereus strains tested in this study, consistent with previous findings that most isolates from food products contained nhe. The enterotoxin EntFM contributes to the cytotoxic and hemolytic activities of B. cereus and to its adhesion to host epithelial cell monolayers [9,16].

Fig. 3.Expression of toxin-associated genes and AtpB levels of Bacillus cereus in the presence of Bacillus amyloliquefaciens strains.

We subsequently performed western blot analysis using an anti-AtpB antibody for the detection and quantification of B. cereus. The beta subunit of ATP synthase, AtpB, which has been reported to control growth in B. cereus, is the universal enzyme that synthesizes ATP from ADP and phosphate using the energy stored in a transmembrane ion gradient [1,12]. As shown in Fig. 3B, AtpB expression levels were markedly lower in B. cereus co-cultured with B. amyloliquefaciens RD7-7 than in the control or in B. cereus in co-culture with B. amyloliquefaciens KACC 15866. These results demonstrate that inhibition of B. cereus growth was mediated by suppression of AtpB expression, and that B. amyloliquefaciens RD7-7 may inhibit the growth of B. cereus by inhibiting expression of toxin-encoding genes.

Arguelles-Arias et al. [3] investigated the effect of B. amyloliquefaciens GA1, which produces antibiotic compounds, for the development of biocontrol agents for use as “green” pesticides. Thus, it seems likely that B. amyloliquefaciens RD7-7 could produce an antibacterial substance that could be used as an agent for biocontrol of B. cereus.

Inhibition of B. cereus Growth and Toxin Production by B. amyloliquefaciens RD7-7 in Fermented Soybean Products

We tested whether B. amyloliquefaciens RD7-7, which showed significant antimicrobial activity against B. cereus in broth, had the same antimicrobial effect in a fermented soybean product. Consistent with the results shown in Fig.4, growth of B. cereus was significantly inhibited in a soybean product fermented with B. amyloliquefaciens RD7-7. The growth of B. cereus decreased to approximately 4.71 log CFU/ml (a 1.9-fold reduction) and 4.33 log CFU/ml (a 2.1-fold reduction) at co-culture ratios of 5:5 and 1:9 B. cereus KACC 10004 to B. amyloliquefaciens RD7-7, respectively, compared with control B. cereus (9.1 log CFU/ml) at 24 h (Fig.4 B). Thus, B. amyloliquefaciens RD7-7 effectively inhibited the growth of B. cereus. Fermented soybean foods that contain many beneficial microorganisms (bacteria andfungi) are commonly contaminated with B. cereus during fermentation [15]. Fig. 4A shows that the surface of a fermented soybean product inoculated with B. amyloliquefaciens RD7-7 was almost completely covered by a gelatinous, slippery, and viscous slime that was not present in the product inoculated with the control B. cereus KACC 10004 strain (sample 3), suggesting that slime production may play an important role in the bactericidal activity of B. amyloliquefaciens RD7-7.

Fig. 4.Photographic images of fermented soybean products (A) and growth inhibition of Bacillus cereus KACC10004 (B) in soybean products fermented with B. cereus KACC10004 and B. amyloliquefaciens RD7-7.

Slime, which is also referred to as a self-produced matrix of extracellular polymeric substances, is a polymeric conglomeration generally composed of extracellular biopolymers in various structural forms [6]. Mucilage or slime on fermented soybean food products has been shown to inhibit the growth of bacteria by damaging the cellular membrane of pathogenic bacteria and some harmful yeast [24]. The γ-polyglutamic acid (γ-PGA) produced by B. amyloliquefaciens C06 enhances colonization of Bacillus cells by improving biofilm formation, surface adhesion, and swarming motility; these findings suggest that γ-PGA producers might be used to improve biological control of Bacillus strains with enhanced surface colonization abilities [22]. In addition, a bacteriocin-like substance produced by a strain of B. amyloliquefaciens isolated from the Brazilian Atlantic forest was shown to inhibit pathogenic and food-spoilage bacteria, such as L. monocytogenes, B. cereus, and Serratia marcescens [21]. Thus, gluey, stringy, and mucilaginous substances, such as the bacteriocin-like substance produced by B. amyloliquefaciens RD7-7, may be good candidates in the search for natural antimicrobial agents.

Growth of B. cereus co-cultured with B. amyloliquefaciens RD7-7 in fermented soybeans was measured by quantitative real-time PCR analysis of groEL. The growth of B. cereus was reduced to approximately 3.24 log CFU/ml (36.80 Ct) and 3.61 log CFU/ml (35.33 Ct) at co-culture ratios of 1:9 and 5:5 B. cereus KACC 10004 to B. amyloliquefaciens RD7-7, respectively, compared with the control B. cereus (7.4 log CFU/ml, 20.26 Ct); thus, high Ct values represent low numbers of B. cereus (Table 3). Therefore, reduced expression of groEL in B. amyloliquefaciens RD7-7–treated fermented soybean products indicates inhibition of B. cereus growth.

Table 3.Individual cycle threshold (Ct) values correspond to standard curves derived from the B. cereus PCR detection kit with 10-fold serial dilutions of B. cereus. 1)B. cereus KACC 10004 2)Standard curve of B. cereus KACC 10004 3)Each value indicates the mean ± SD of three replicate analyses. Within each row, means indicated with different superscript letters differ significantly p < 0.05 (oneway ANOVA, followed by Duncan's multiple comparison test).

For detection of expression of B. cereus toxin genes at the transcript level in fermented soybean products, we used qPCR for sensitive detection of groEL, nheA, nheC, and entFM transcripts in B. cereus co-cultured with B. amyloliquefaciens RD7-7. As shown in Fig. 5A, groEL, nheA, nheC, and entFM transcript levels were reduced in the 5:5 and 1:9 co-cultures of B. cereus KACC10004 and B. amyloliquefaciens RD7-7, compared with monoculture fermentation with B. cereus KACC10004. These results indicate that the inhibition of B. cereus growth in the presence of B. amyloliquefaciens RD7- 7 may be regulated by groEL, nheA, nheC, and entFM.

Fig. 5.Expression of Bacillus cereus spore-related genes and AtpB levels of fermented soybean products in the presence of Bacillus amyloliquefaciens strains.

We next performed western blot analysis using an anti- AtpB antibody for the detection and quantification of B. cereus in fermented soybean products. Fig. 5B shows that the fermented soybean product (5:5 and 9:1 ratio) inoculated with B. amyloliquefaciens RD7-7 showed low or undetectable levels of AtpB expression compared with the B. cereus KACC10004 control, suggesting a significantly lower expression level of the AtpB protein in the B. amyloliquefaciens RD7-7–treated fermented soybean samples, and indicating inhibition of the growth of B. cereus (Fig. 5B). Together, these results suggest that B. amyloliquefaciens RD7-7 may decrease the survival of B. cereus by reducing the expression of toxin-related genes in fermented soybean products.

Growth and Toxin Expression by B. cereus Co-Cultured with B. amyloliquefaciens RD7-7 in a Buckwheat Soksungjang Product

We measured the growth and toxin gene expression of B. cereus in a buckwheat soksungjang product inoculated with B. amyloliquefaciens RD7-7.

Buckwheat soksungjang is a bealmijang manufactured with buckwheat and soybeans. Buckwheat protein, which is mainly composed of lysine, arginine, albumin, globulin, and glutelin, shows various health benefits against several diseases, such as hypertension, hypercholesterolemia, diabetes, and obesity; thus, buckwheat, used as a protein source complementary to grain (cereal) foods and vegetables, has gained an excellent reputation for its nutritious qualities [2,20]. Therefore, we tested the efficacy of B. amyloliquefaciens RD7-7 in preventing B. cereus contamination of a soksungjang product made using buckwheat.

As s hown in Fig. 6A, t he g rowth o f B. cereus was inhibited in the buckwheat soksungjang product fermented with B. amyloliquefaciens RD7-7, but did not differ significantly from the growth of control B. cereus. The survival and growth of B. cereus decreased to approximately 2.74 log CFU/ml (a 2.4-fold reduction) and 0.30 log CFU/ml (a 22.2-fold reduction) at co-culture ratios of 1:9 B. cereus KACC 10004 to B. amyloliquefaciens RD7-7, respectively, in comparison with control B. cereus (6.66 and 6.68 log CFU/ml) at 21 and 49 days.

Fig. 6.Growth and toxin expression by Bacillus cereus co-cultured with Bacillus amyloliquefaciens RD7-7 in a buckwheat soksungjang product.

Expression of transcripts of B. cereus toxin-related genes (groEL, nheA, nheC, and entFM) was downregulated in B. cereus co-cultured with B. amyloliquefaciens RD7-7 at 21 and 49 days, and did not differ at the three time points in the B. cereus control (Fig. 6B).

In addition, AtpB expression levels were markedly lower in B. cereus co-cultured with B. amyloliquefaciens RD7-7 compared with the control strain at 21 and 49 days. Together, these results suggest that B. amyloliquefaciens RD7-7 decreases the survival of B. cereus by reducing expression of toxin-related genes in a buckwheat soksungjang product. The use of antagonistic substances such as bacteriocins, bacteriocin-like substances, and antibacterial lipopeptides (surfactin, fengycin, and iturin) produced by B. amyloliquefaciens strains has been reported to prevent growth of B. cereus in cheese, milk, rice-based foods, cooked rice, beef gravy, and chilled dairy products [15,22,32]. Recent studies have shown that bacteriocin J4 produced by B. amyloliquefaciens J4, isolated from traditional fermented soybean paste, exhibits specific antagonistic activity against various foodborne pathogens such as Micrococcus luteus, Vibrio parahaemolyticus, S. aureus, and L. monocytogenes [22]. Arguelles-Arias et al. [4] observed that the B. amyloliquefaciens GA1 strain produces an antimicrobial peptide, named amylolysin (a novel antibiotic), active against an array of gram-positive bacteria, including methicillin-resistant S. aureus. Our results indicate that the novel antibacterial peptides produced by B. amyloliquefaciens RD7-7 will play important roles as bacterial biocontrol agents for the inhibition of B. cereus growth in food production.

In conclusion, B. amyloliquefaciens RD7-7 isolated from rice doenjang exhibited a significant antibacterial effect against B. cereus and reduced the expression of its enterotoxin genes. B. amyloliquefaciens RD7-7 shows potential for use as an efficient biological control agent in fermented soybean products to exclude pathogenic B. cereus during manufacturing, without decreasing food quality or inhibiting fermentation by Bacillus spp. It may also have other uses in the food, agricultural, biotechnology, and pharmaceutical industries. Hence, further studies are required to identify and characterize the novel antagonistic substances such as bacteriocins, bacteriocin-like substances, and antibacterial lipopeptides produced by B. amyloliquefaciens RD7-7.