Probiotic Properties and Antioxidant Activities of Pediococcus pentosaceus SC28 and Levilactobacillus brevis KU15151 in Fermented Black Gamju

Black gamju is Korean traditional beverage fermented with molds. The aim of this study was to assess the probiotic properties and antioxidant activities of novel Pediococcus pentosaceus SC28 and Levilactobacillus brevis KU15151 to develop black gamju with bioactive properties for health. Tolerance against artificial gastric juice and bile salts, adhesion ability on HT-29 cells of strains, and antibiotics susceptibility were evaluated as probiotics, and various enzyme productions were detected. The 2,2-diphenyl-1-picrylhydrazyl assay, 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonate, and β-carotene bleaching assay were used for antioxidant activity of samples. The tolerance of both strains to artificial gastric juice and bile salts (Oxgall) was more than 90%. Additionally, both strains did not produce β-glucuronidase and were resistant to gentamicin, kanamycin, streptomycin, and ciprofloxacin. After fermentation of black gamju with each strain, the number of viable lactic acid bacteria increased to 8.25–8.95 log colony forming unit/mL, but the pH value of fermented samples decreased more (to pH 3.33–3.41) than that of control (pH 4.37). L. brevis KU15151 showed higher adhesion activity to HT-29 cells and antioxidant effects than P. pentosaceus SC28 in three antioxidant assays.


Introduction
Nowadays, consumers are increasingly demanding functional foods that are rich in probiotic bacteria [1,2]. Probiotics are defined as live microorganisms that, when administered in adequate amounts, confer a health benefit on the host such as immunomodulatory, antimicrobial, anticancer, and hypolipidemi effects [3]. In particular, they have been known to have antioxidant effects due to exopolysaccharides on cell wall; their antioxidant enzymes such as catalase, glutathione peroxidase, superoxide dismutase (SOD), catalase; various antioxidant compounds such as glutathione, tocopherol, and ascorbic acid [4]. Most lactic acid bacteria (LAB) and some Bacillus strains [5] have been used as probiotics and are generally recognized as safe [6]. Lactobacillus, Leuconostoc, Streptococcus, and Pediococcus are most popular as probiotic strains. However, as probiotics, LAB should safely pass through and attach to the gastrointestinal (GI) tract to have beneficial effects [7][8][9]. In addition, lactic acid fermentation should increase the nutritional value and prolong shelf life of food products [10]. The most commonly consumed probiotic foods are milk-based formulations such as yogurt and cheese. However, these milk-based formulations are not consumed by individuals with lactose intolerance and

Isolation and Identification of Probiotic Strains
The LAB strains used in this study were isolated from traditional Korean food (octopus jeotgal and radish kimchi). One gram of each food sample was serially diluted and spread on de Man, Rogosa, and Sharpe (MRS; BD Biosciences, Franklin Lakes, NJ, USA) agar at 37 • C for 24 h. Then, every colony was inoculated and incubated in MRS broth at 37 • C for 24 h. Potential probiotic LAB strains were identified as Pediococcus pentosaceus SC28 and Levilactobacillus brevis KU15151 by using 16S rRNA sequencing performed by Bionics Inc. (Seoul, Korea). The results on sequencing were analyzed by comparing with GENBANK database using the Basic Local Alignment Search Tool (BLAST) website (http://blast.ncbi.nlm.nih.gov). As a reference probiotic strain, Lactobacillus rhamnosus GG was taken from the Korean Collection for Type Cultures (KCTC, Daejeon, Korea).

Tolerance to Artificial Gastric Juice and Bile Salts
The resistances of the isolated strains to artificial gastric juice and bile salts were evaluated as described by Guo et al. [20] with some modifications. The resistance to artificial gastric juice was measured after incubation in 50 mM sodium carbonate buffer including 0.3% (w/v) pepsin (Sigma-Aldrich, St. Louis, MO, USA), and the pH was controlled at 2.5 for 3 h at 37 • C. The resistance to bile salts was measured by incubation in MRS broth including 0.3% (w/v) Oxgall (BD Biosciences, Oxford, UK) at 37 • C for 24 h. The survival rate (%) of the strains was measured by enumerating the viable cells on the MRS agar plates.

Adhesion Ability to HT-29 Cells
In case of measurement of the adhesion ability of the LAB strains, HT-29 (KCLB 30038, a human colon adenocarcinoma cells) were used [21]. HT-29 cells (2 × 10 5 cells/mL) were incubated in a 24-well Foods 2020, 9, 1154 3 of 12 plate at 37 • C for 24 h. The intact bacterial cells (1 × 10 7 CFU/mL) of each LAB strain were added to the HT-29 cells and were incubated in Roswell Park Memorial Institute (RPMI) 1640 medium (Hyclone, Logan, UT, USA) at 37 • C for 2 h. Non-adherent bacterial cells were removed by washing three times with PBS. The adherent bacterial cells were detached using 1% Triton X-100 (Sigma-Aldrich, St. Louis, MO, USA) solution. Then, 100 µL of a detached cell broth was used for MRA plate counting. The adhesion ability was evaluated by counting the number of initial cells and adherent cells on the MRS plates as follows: The adhesion ability (%) = Adherent bacterial cells (CFU/mL) Initial number of bacterial cells (CFU/mL) × 100% (1)

Enzyme Production
To measure the various enzyme productions of LAB strains, the API ZYM kit (BioMerieux, Lyon, France) was used. The strains were suspended in phosphate buffered saline (PBS) (Gibco, Grand Island, NY, USA) at 10 6 CFU/mL, added to each cupule, and incubated at 37 • C for 4 h. One drop of ZYM test reagent was inoculated, and the level of enzyme activity was determined as the degree of color change.

Antibiotic Susceptibility
The antibiotic susceptibility of the LAB strains was evaluated using the disk diffusion assay, according to the method of Yang et al. [21]. The culture solution (100 µL) of each LAB strain was spread onto MRS agar. Then, paper disks containing antibiotics were placed on the MRS agar and incubated at 37 • C for 24 h. The clear zone (mm) was measured and compared with the standards set by the Clinical and Laboratory Standards Institute (CLSI). Eight types of antibiotics were used in this test: ampicillin (0.2 g/L), chloramphenicol (0.6 g/L), ciprofloxacin (0.1 g/L), doxycycline (0.6 g/L), gentamicin (0.2 g/L), kanamycin (0.6 g/L), streptomycin (0.2 g/L), and tetracycline (0.6 g/L).

Black Gamju Fermentation and Investigation of Viable Cell Number and pH Value
Black gamju saccharified by Aspergillus awamori was obtained from Hansan F&G Co. (Eumseong, Korea). The black gamju was prepared as follows: black koji was made using A. awamori. Rice (10 kg) was steeped in distilled water at 20 • C for 2 h and steamed for 40 min. After cooling to 35 • C, the steamed rice was inoculated with 0.2% (w/w) A. awamori spores and incubated at 35 • C for 3 days. To produce black gamju, steeped rice, distilled water, and black koji were mixed at a ratio of 1:4:1 (weight) and incubated at 60 • C for saccharification. The saccharified black gamju was sterilized at 121 • C for 15 min.
P. pentosaceus SC28 and L. brevis KU15151 were used to ferment the black gamju. The fermented black gamju was prepared as follows: each LAB strain was separately inoculated into the black gamju at a final concentration of 10 5 CFU/mL and fermented at 37 • C for 24 h in the incubator. Non-fermented black gamju was used as the negative control. The number of LAB strains in the black gamju was enumerated by counting viable cells on the MRS plates. The pH value of the fermented black gamju was measured using a pH meter (Model: pH7110, Xylem Analytics Germany GmbH, Weilheim, Germany).

Sample Preparation of Bacterial Cells and Extraction of Black Gamju Samples
LAB strains were cultured in MRS broth at 37 • C for 18 h. Intact cells were harvested by centrifuging (14,000× g) at 4 • C for 10 min. The harvested cells were washed three times and re-suspended in PBS (Summer Scientific, Waltham, MA, USA).
The fermented and non-fermented black gamju extracts were prepared as described by Ghosh et al. [22] as follows: the black gamju (50 mL) samples were extracted with 150 mL of methanol:acetone:water (4:3:3) mixture. The mixture was filtered with Whatman No. 2 filter paper and evaporated with a rotary vacuum evaporator (EYELAN-1000V; Tokyo, Japan) at 50 • C. The resultant Foods 2020, 9, 1154 4 of 12 products were lyophilized in a freezing dryer (Benchtop FDB; Operon, Gimpo, Korea) and stored at −18 • C until further used.

Antioxidant Activity of LAB Strains and Black Gamju Extracts
2.8.1. 2,2-Diphenyl-1-picrylhydrazyl (DPPH) Radical Scavenging Activity 2,2-Diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity was evaluated as described by Das and Goyal [23] with some modifications. To evaluate the antioxidant activity of the LAB strains, 2 mL of DPPH solution (0.4 mM) in methanol and 2 mL of bacterial samples were mixed and incubated at 37 • C for 30 min in the dark. Ascorbic acid (1 mg/mL) was used as the positive control. The DPPH radical scavenging activity was calculated by measuring the absorbencies of the supernatant at 517 nm and using the following equation: where A sample and A control are the absorbance values of the bacterial sample and distilled water at 30 min, respectively.
2.8.2. 2,2 -Azinobis(3-ethylbenzothiazoline-6-sulfonate) (ABTS) Radical Scavenging Activity 2,2 -Azinobis(3-ethylbenzothiazoline-6-sulfonate) (ABTS) radical scavenging activity was evaluated as described by Verón et al. [24] with some modifications. To make the ABTS solution, 14 mM ABTS and 5 mM potassium persulfate dissolved in 0.1 M potassium phosphate buffer (pH 7.4) were mixed and diluted until that the absorbance at 734 nm was adjusted to 0.7 ± 0.02. One hundred and fifty microliters each of the bacterial samples, gamju extracts dissolved in distilled water, and 1 mg/mL ascorbic acid (positive control) were mixed with the same amount of ABTS solution and incubated at 37 • C for 10 min. Absorbance of the supernatant was measured at 734 nm and calculated using the following equation: where A sample and A control are the absorbance values of the sample (bacterial or gamju) and distilled water after reaction for 10 min, respectively.

β-Carotene Bleaching Method
The β-carotene bleaching method was used as described by Kassim et al. [25] with some modifications. The β-carotene solution was composed of linoleic acid (132 µL), β-carotene (6 mg), and Tween 80 (600 µL) (Samchun Co., Seoul, Korea) with 20 mL of chloroform. The chloroform in the solution was removed using a rotary evaporator. The absorbance of solution at 470 nm was adjusted to 1.20 with distilled water. Two hundred microliters of bacterial samples, gamju extracts dissolved in distilled water (50 mg/mL), and 1 mg/mL ascorbic acid (positive control) was mixed with 4 mL of the solution and incubated at 50 • C for 2 h. The absorbance of the supernatant was measured at 470 nm for 0 and 2 h and calculated using the following equation: where A sample, 2 h and A control, 2 h are absorbance values of the sample (bacterial or gamju) and distilled water after 2 h of the reaction, respectively. A control, 0 h is the absorbance of distilled water at the initial time.

Statistical Analysis
All experiments were triplicated, the significant differences were determined using one-way analysis of variance (ANOVA), and Duncan's multiple range tests were performed by using SPSS software (Version 24; SPSS, Inc., Chicago, IL, USA).

Tolerance to Artificial Gastric Juice and Bile Salts
The tolerances to artificial gastric juice and bile salts of two strains were tested (Table 1). All the tested LAB strains showed more than 90% survival rate and no significant differences (p > 0.05) under acidic conditions. In contrast, significant differences (p < 0.001) were observed under basic conditions. The data showed that the survival rates of L. rhamnosus GG and P. pentosaceus SC28 against bile salts were 101.83% and 100.05%, respectively, whereas L. brevis KU15151 showed a tolerance of 97.96%.

Adhesion Ability of LAB Strains to HT-29 Cells
To have beneficial effects on the host, LAB should be adherent and colonize the intestinal tissues of the host [26]. The adhesion ability of the strains to HT-29 cells is presented in Table 1. The adhesion ability of the tested strains appeared to be variable (<4.45-6.87%), depending on the LAB. P. pentosaceus SC28 and L. rhamnosus GG showed adhesion rates of 4.45% and 6.30%, respectively. L. brevis KU15151 had a higher adhesion rate of 6.87% than the other strains.

Enzyme Production by LAB Strains
In this study, production of bacterial enzymes was estimated as a biological property of LAB. The production of 19 enzymes by the LAB strains was evaluated using the API ZYM kit (Table 2). P. pentosaceus SC28 and L. brevis KU15151 did not produce β-glucuronidase, which is detrimental to the human intestine because it hydrolyzes glucuronides [27]. In addition, these LAB strains commonly produce acid phosphatase, β-galactosidase, leucine arylamidase, lipase, naphthol-AS-BI-phosphohydrolase, and valine arylamidase. In addition, L. brevis KU15151 produces cystine arylamidase, esterase, esterase lipase, α-galactosidase, α-glucosidase, and β-glucosidase. Lactose is hydrolyzed into glucose and galactose by β-galactosidase, which alleviates lactose intolerance [27]. Table 2. Analysis of enzyme production by two strains with the API ZYM kit.

Antibiotic Susceptibility of LAB Strains
As shown in Table 3, all the strains were resistant to four antibiotics, namely, gentamicin, kanamycin, streptomycin, and ciprofloxacin, but both P. pentosaceus SC28 and L. brevis KU15151 were sensitive to ampicillin, tetracycline, chloramphenicol, and doxycycline.

Antioxidant Activity of LAB Strains
In this study, three methods were performed to evaluate the antioxidant effects of the LAB strains ( Figure 1). The DPPH radical scavenging assay results are presented in Figure 1A. The DPPH radical scavenging activity (%) was in the following order: L. brevis KU15151 (31.14%), L. rhamnosus GG (27.89%), and P. pentosaceus SC28 (17.83%). According to the ABTS radical scavenging assay ( Figure 1B), the antioxidant activity of L. brevis KU15151 (27.50%) was similar to that of L. rhamnosus GG (27.55%) but greater than that of P. pentosaceus SC28 (8.65%). The β-carotene bleaching assay results are presented in Figure 1C. The inhibitory effects of L. rhamnosus GG and L. brevis KU15151 were 27.60% and 23.82%, respectively, whereas P. pentosaceus SC28 (14.18%) showed a lower inhibition rate. Thus, P. pentosaceus SC28 showed the lowest antioxidant effects, and L. brevis KU15151 was superior as an antioxidant.

Fermentation of Black Gamju Using LAB Strains
The samples of black gamju were separately fermented using P. pentosaceus SC28 and L. brevis KU15151. The viable cell numbers and pH values are presented in Table 4. The viable cell numbers in all fermented black gamju increased from 5 to 8.25-8.95 log CFU/mL for 1 day. Similarly, the pH value of fermented black gamju (pH 3.32-3.41) decreased when compared with non-fermented black gamju (pH 4.37). Due to the reaction of hydrolytic enzymes, the starch in rice was converted to sugars, and gamju contains high concentrations of glucose, maltose, and malto-oligomers [28]. The LAB strains were assumed to grow by consuming sugars in the gamju, and they produced organic acids such as lactic acid.  KU15151; (2) a-c Different superscript letters in the same column present significant differences in each characteristic (p < 0.001). All values are mean ± standard deviation (SD) of triplicated experiments.

Fermentation of Black Gamju Using LAB Strains
The samples of black gamju were separately fermented using P. pentosaceus SC28 and L. brevis KU15151. The viable cell numbers and pH values are presented in Table 4. The viable cell numbers in all fermented black gamju increased from 5 to 8.25-8.95 log CFU/mL for 1 day. Similarly, the pH value of fermented black gamju (pH 3.32-3.41) decreased when compared with non-fermented black gamju (pH 4.37). Due to the reaction of hydrolytic enzymes, the starch in rice was converted to sugars, and gamju contains high concentrations of glucose, maltose, and malto-oligomers [28]. The LAB strains were assumed to grow by consuming sugars in the gamju, and they produced organic acids such as lactic acid.

Antioxidant Activity of Black Gamju Extracts
The ABTS radical scavenging assay and β-carotene bleaching assay were conducted for evaluating the antioxidant potential of the black gamju extracts. As shown in Figure 2, IC 50 values of non-fermented black gamju (B), black gamju fermented using P. pentosaceus SC28 (B (SC28)), and black gamju fermented using L. brevis KU15151 (B (151)) were 18.10, 15.33, and 14.74 mg/mL, respectively, and the radical scavenging effect against ABTS + · was increased in the black gamju extract fermented by P. pentosaceus SC28 and L. brevis KU15151. The non-fermented and fermented black gamju extracts showed a high inhibition rate for βcarotene and linoleic acid oxidant. Especially, the inhibition rate of black gamju fermented using L. brevis KU15151 (75.49%) was higher than that of non-fermented black gamju (74.17%) and black gamju fermented using P. pentosaceus SC28 (74.57%). However, no significant differences (p > 0.05) were observed between the non-fermented black gamju and fermented black gamju extracts. Prevention of lipid peroxidation was not increased significantly by fermentation using LAB.

Discussion
The human stomach is an acidic environment in which the pH is 1.0-4.5 and ingested substances stay in the stomach for 3 h [29]. The intestinal tract is a basic environment in which 0.3% (w/v) bile salts are found [30]. Therefore, when the ingested LAB strains pass through the GI tract, the tolerance of the LAB strains to artificial gastric juice and bile salts is a very important factor. Some researchers have shown that the viable cell number of P. pentosaceus OZF decreased from 8.74 to 7.21 log CFU/mL when the strain was exposed to 0.3% oxgall for 4 h [31]. Bujnakova et al. [27] also reported that the survival rates of Lactobacillus reuteri L4/1 under acidic (0.3% pepsin, pH 2.5) and basic (0.3% oxgall, pH 7) conditions were 84% and 90%, respectively. Vidhyasagar et al. [26] reported that the cell number of potential probiotic P. pentosaceus VJ13 decreased to about 50% against gastric juice (pH 2 for 4 h). On the basis of our results, P. pentosaceus SC28 and L. brevis KU15151 could pass safely through the human GI tract because these two strains have superior tolerance to gastric acid and bile salts.
In addition, previous studies have reported that the adhesion ability to intestinal epithelium is related to hydrophobicity and auto-aggregation ability because the adhesion process involves interaction between the intestinal cells and bacterial cell surface [32][33][34]. Especially, Han et al. [32] and Bengoa et al. [33] reported that hydrophobic interaction allows bacterial cells and intestinal cells to bind more strongly and the auto-aggregation is related to charge and compounds such as proteins (SlpA) and exopolysaccharides of the LAB cell surface; L. brevis R4 and L. acidophilus AD1, which have the greatest hydrophobicity and auto-aggregation ability, have higher adhesion abilities to Caco-2 cells [32]. Jeon et al. [35] reported that 2-10% Lactobacillus strains can attach to the intestine. Potential probiotic Lactobacillus plantarum C182 showed 1.2% adhesion ability to HT-29 cells [36].
In this study, the adhesion ability of P. pentosaceus SC28 and L. brevis KU15151 to HT-29 cells showed that they have high adhesion and colonization ability for human intestine cells.
In this study, production of bacterial enzymes was estimated as a biological property of LAB. α-Glucosidase hydrolyzes disaccharides to glucose [37]. β-Glucosidase is used for the bioconversion of The non-fermented and fermented black gamju extracts showed a high inhibition rate for β-carotene and linoleic acid oxidant. Especially, the inhibition rate of black gamju fermented using L. brevis KU15151 (75.49%) was higher than that of non-fermented black gamju (74.17%) and black gamju fermented using P. pentosaceus SC28 (74.57%). However, no significant differences (p > 0.05) were observed between the non-fermented black gamju and fermented black gamju extracts. Prevention of lipid peroxidation was not increased significantly by fermentation using LAB.

Discussion
The human stomach is an acidic environment in which the pH is 1.0-4.5 and ingested substances stay in the stomach for 3 h [29]. The intestinal tract is a basic environment in which 0.3% (w/v) bile salts are found [30]. Therefore, when the ingested LAB strains pass through the GI tract, the tolerance of the LAB strains to artificial gastric juice and bile salts is a very important factor. Some researchers have shown that the viable cell number of P. pentosaceus OZF decreased from 8.74 to 7.21 log CFU/mL when the strain was exposed to 0.3% oxgall for 4 h [31]. Bujnakova et al. [27] also reported that the survival rates of Lactobacillus reuteri L4/1 under acidic (0.3% pepsin, pH 2.5) and basic (0.3% oxgall, pH 7) conditions were 84% and 90%, respectively. Vidhyasagar et al. [26] reported that the cell number of potential probiotic P. pentosaceus VJ13 decreased to about 50% against gastric juice (pH 2 for 4 h). On the basis of our results, P. pentosaceus SC28 and L. brevis KU15151 could pass safely through the human GI tract because these two strains have superior tolerance to gastric acid and bile salts.
In addition, previous studies have reported that the adhesion ability to intestinal epithelium is related to hydrophobicity and auto-aggregation ability because the adhesion process involves interaction between the intestinal cells and bacterial cell surface [32][33][34]. Especially, Han et al. [32] and Bengoa et al. [33] reported that hydrophobic interaction allows bacterial cells and intestinal cells to bind more strongly and the auto-aggregation is related to charge and compounds such as proteins (SlpA) and exopolysaccharides of the LAB cell surface; L. brevis R4 and L. acidophilus AD1, which have the greatest hydrophobicity and auto-aggregation ability, have higher adhesion abilities to Caco-2 cells [32]. Jeon et al. [35] reported that 2-10% Lactobacillus strains can attach to the intestine. Potential probiotic Lactobacillus plantarum C182 showed 1.2% adhesion ability to HT-29 cells [36].
In this study, the adhesion ability of P. pentosaceus SC28 and L. brevis KU15151 to HT-29 cells showed that they have high adhesion and colonization ability for human intestine cells.
In this study, production of bacterial enzymes was estimated as a biological property of LAB. α-Glucosidase hydrolyzes disaccharides to glucose [37]. β-Glucosidase is used for the bioconversion of isoflavone, and substances produced by β-glucosidase have high absorption and bio-activity in the intestines [36]. In addition, P. pentosaceus SC28 and L. brevis KU15151 could be safe from the absence of β-glucuronidase. Generally, some probiotics that produce glucuronidase have been restricted because many toxic compounds (such as carcinogens) bound to glucuronic acid in the liver by phase II detoxification may be re-dissociated in the colon by glucuronidase and re-absorbed into the human body [38].
In this study, the susceptibility of LAB strains to various antibiotics was tested as a probiotic property. As probiotics, the strains that have beneficial effects on the health particularly of medicated patients should be resistant against various antibiotics. However, susceptibility to antibiotics may be different according to the LAB subspecies. According to Wang et al. [39], 10 L. brevis strains showed resistance to gentamicin, kanamycin, streptomycin, ciprofloxacin, and chloramphenicol. Some researchers reported that P. pentosaceus VJ13 was resistant to ampicillin, gentamicin, streptomycin, and ciprofloxacin and P. pentosaceus OZF was resistant to kanamycin [26,31]. In addition, the presence of antibiotic resistance determinants in probiotics has been concerned for safety because it is possible to be transferred to other pathogens in the human intestine by a genetic mechanism such as chromosomal recombination. In the intestine, bacteria containing antibiotic resistance genes may create other antibiotic-resistant bacteria [31]. According to the clinical and laboratory standards institute (CLSI) guideline, these strains were found to be safe for human health.
In previous studies, free radical scavenging activity, chelating activity, inhibitory activity of lipid peroxidation, and antioxidant-related enzyme activity were used to assess the antioxidant effects of LAB strains [32,40]. Han et al. [32] and Tang et al. [40] presented that the antioxidant substances of LAB strains are NADH, NADPH, antioxidant enzymes, Mn 2+ , bioactive compounds, and exopolysaccharides. Das and Goyal [23] found higher DPPH radical scavenging activity in L. plantarum DM5, L. plantarum B-4496, and L. acidophilus B-4495. Han et al. [32] reported that intact cells of LAB strains containing P. pentosaceus R1 and L. brevis R4 had significantly higher ABTS radical scavenging activity than cell-free extract and supernatant. Li et al. [41] showed that the free radical scavenging ability of intact cells is related to the cell surface materials of bacteria, e.g., proteins, polysaccharides, and lipoteichoic acid. In addition, Tang et al. [40] reported that the intact cells of L. plantarum MA2 prevented lipid peroxidation. On the basis of the results of the radical scavenging and β-carotene bleaching assays, L. brevis KU15151 showed superior antioxidant ability.
Various studies have reported non-dairy probiotic beverages. Chavan et al. [10] reported that non-dairy probiotic drinks that consist of cereal powder (barley, millet, and bean) and soymilk or almond milk showed an increase in acidity, decrease in pH, and increase in viable cell number. In addition, it was reported that a rice-based beverage fermented by Lactobacillus fermentum KKL1 showed a decrease in pH from 6.72 to 4.03 and increase in total titratable acidity from 0.01 to 0.84 and bacterial content [22].
Ghosh et al. [6] reported that haria, an Indian rice-based fermented beverage, showed higher DPPH radical scavenging activity as the fermentation time increased. In addition, the rice-based beverage fermented by L. fermentum KKL1 had a significantly higher amount of phenolic and flavonoid compounds and free radical scavenging ability [22]. These antioxidant effects were related to the higher concentration of oligosaccharides, flavonoids, and phenolic compounds in the fermented rice. On the basis of the antioxidant activities, the fermented black gamju has superior antioxidant ability than non-fermented black gamju. Therefore, the fermented black gamju could be used as a source of probiotic drinks.

Conclusions
P. pentosaceus SC28 and L. brevis KU15151 isolated from Korean traditional foods showed higher gastric acid and bile salts tolerances and adhesion abilities on HT-29 cells. Additionally, both P. pentosaceus SC28 and L. brevis KU15151 are safe because they do not produce β-glucuronidase, which causes toxin re-cycling in the human body, and have good antibiotic resistance. L. brevis KU15151 exhibited greater radical scavenging activity and lipid peroxidation inhibitory activity than P. pentosaceus SC28. After the black gamju was inoculated with the LAB strains, the viable cell number increased, and pH value decreased when compared with non-fermented black gamju. Furthermore, the fermented black gamju extracts showed higher antioxidant activities than the non-fermented black gamju extract except for β-carotene and linoleic acid anti-oxidation. Therefore, P. pentosaceus SC28 and L. brevis KU15151 can be used as potential probiotics, and the fermented black gamju can be used as a probiotic drink.

Conflicts of Interest:
The authors declare no conflict of interest.