Ginseng powder was made with four-year-old root ginseng grown in Geumsan. Cheonggukjang was purchased as traditional Cheonggukjang manufactured in Cheongyang, Chungnam.

To isolate a pure culture of B. subtilis from the Cheonggukjang manufactured in Cheongyang, Chungnam, cell subculture was performed in a nutrient broth until a single colony was formed. 1 uL of an isolated single colony was collected using a loop and inoculated onto a slide, and then 1 uL of α-cyano-4-hydroxycinnamic acid (CHCA) Matrix solution was dropped onto the colony. The microorganism was identified by placing the slide in VITEK MS (bioMériux, Australia) Matrix-Assisted Laser Desorption/Ionization Time-of Flight Mass Spectrometry (MALDI-TOF). Escherichia coli (E. Coli) American Type Culture collection (ATCC) 8739 was used as the quality assurance (QA) strain. Based on the first identification results, the microorganism presumed to be B. subtilis was sent to BIOFACT Co., Ltd. and was confirmed as two types of B. subtilis of Cheonggukjang origin through 16S rRNA analysis.

The ginseng powder (25 g) obtained from 4-year-old root ginseng cultivated in Geumsan was mixed with distilled water (75 g) 3 times its weight in a 500 mL bottle and sterilized at 121˚C for 15 minutes to prepare the ginseng solid medium. Onto the medium, 1% (w/w) each of the two types of B. subtilis purely isolated from Cheonggukjang was inoculated and cultured in an incubator at 37˚C for up to 10 days.

To measure the bacterial count of B. subtilis isolated from Cheonggukjang, the ginseng fermentation culture medium was diluted in peptone water and smeared on the nutrient agar, and then cultured in an incubator at 37˚C for 24 hours.

The ginsenosides were analyzed by modifying the HPLC analysis method proposed by In et al. [20] . After freeze-drying the fermented ginseng solids, a 0.5 g sample was collected in a 15 mL tube, to which 10 mL of 70% MeOH was added and stirred, and then treated by ultrasound for 30 minutes. It was then centrifuged at 3000 rpm for 10 minutes, and the supernatant was filtered through a 0.45 um PTFE filter and used in the high-performance liquid chromatography (HPLC) analysis. The measurements were taken using the HPLC system (Chrompass Data System, JASCO, Japan) at 203 nm, with an auto sampler (AS-2051 plus, JASCO), Chrompass software (ver. 1.8.6.1, JASCO), and a photodiode array detector (MD-2010, JASCO). The analytical column used was a 5 μm Symmetry C18 column (250 × 4.6 mm, SunFireTM, Waters Corp, USA); the injection was 20 uL, the flow rate was 1.6 mL/min, and the column temperature was 35˚C. The mobile phase employed was (A) water and (B) acetonitrile deaerated by an ultrasonic cleaner, and the gradient system was established as shown in Table 1. Standard ginsenoside materials including Rb1, Rb2, Rb3, Rc, Rd, Rg3, Rh2, Re, Rg1, Rg2 + Rh1 and Rh1(r) were purchased from Fleton Natural Products Co., Ltd (Chengdu, China).

5 g of the ginseng medium was mixed with the same amount of distilled water and homogenized at 25˚C for 30 minutes, and then centrifuged at 3000 rpm for 15 minutes; 5 mL of the supernatant was then obtained. Afterward, it was evaporated and dried at 105˚C, and the weight was measured and expressed as the dry weight of the sample.

The viscosity of the medium was measured at 25˚C using a rheometer (Ar2000, TA instruments, New Castle, DE, USA). The shear rate showed 100 points in the range of 0 - 100 1/s.

The fibrinolytic activity of the ginseng solid medium was measured using the method proposed by Astrup and Mullertz [21] , though with some modifications. Fibrinogen was completely dissolved in a 50mM phosphorus buffer solution (pH 7.4, 0.15M NaCl included) to a final concentration of 0.3%; and then 5 mL of this fibrinogen solution and the same volume of 2% agarose (sterilized agar solution) were mixed. 500 uL of thrombin (100 NIH unit/mL) was added to the aforesaid mixed solution and fully mixed with it, and then the solution was immediately poured onto a petri-dish and left to stand at room temperature for 30 - 60 minutes for solidification to obtain the fibrin plate. To measure the activity, 7 holes with a diameter of 5 mm were made on the fibrin plate using a pasteur pipette, and then 4 mL of distilled water per 1 g of medium was added and mixed for 30 minutes and centrifuged; 20 uL of the supernatant was injected into the mixture and reacted at 37˚C for 5 hours, and then the area of the transparent

circle thus generated was calculated. For the control plot, a purified fibrinolytic enzyme, plasmin (1.0 unit/mL), was used. The fibrinolytic activity of the extract was calculated by converting the relative ratio of the fibrinolytic area of the sample into that of the control plot.

Statistical analysis was performed using SAS software (Ver. 9.0, Cary, NC, USA). The significance of the results of the study was tested using Duncan’s multiple-range test at the 5% level.

The bacterial count of B. subtilis during fermentation was measured as shown in Figure 1. The counts of the A and B strains immediately after inoculation were 3.56 and 3.74 log CFU/g, respectively, but they proliferated to over 9.00 log CFU/g after 3 days of fermentation, regardless of the kind of strain. Later, their counts were 10.47 log CFU/g and 10.25 log CFU/g, respectively, at up to 7 days

of fermentation―with a slightly higher count in the A strain but decreased slightly to 10.43 log CFU/g and 10.20 log CFU/g, respectively, after 10 days of fermentation without any significant difference. In their study involving screening for ginseng-fermented microorganisms, Kim et al. [12] reported that, when Bacillus was inoculated onto a 2.5% ginseng powder medium and cultured, the growth was more than 6 log CFU/mL for 24 hours, gradually increasing to 7 log CFU/mL or showing quiescence. The proliferation of the strain was considered sufficient because there was no significant difference in the count after 3 days of fermentation, regardless of the type of strain.

The changes in the ginsenoside content during fermentation are shown in Figure 2. Ginsenoside analysis was performed on 11 kinds of ginsenosides includ ing Rb1, Rb2, Rb3, Rc, Rd, Rg3, Rh2, Re, Rg1, Rg2+Rh1, and Rh1(r). When the A strain was inoculated and fermented in the ginseng medium, the ginsenosides Rb2, Rb3, Rd, Rg3, Rg2 + Rh1, and Rh1(r) significantly increased by 48.68%, 37.84%, 222.71%, 129.17%, 128.44%, and 59.78%, respectively, while Rc, Re, and Rg1 decreased by 60.40%, 87.16%, and 59.78%, respectively, compared to the initial amounts. In the case of the B strain, Rb1, Rc, Re, and Rg1 decreased by 26.89%, 69.33%, 89.52%, and 83.74%, respectively, whereas Rd and Rg2 + Rh1

increased by 155.40% and 65.61%, respectively, as the fermentation time increased. It was confirmed that the decrease in Rc, Re, and Rg1 and the increase Rd and Rg2 + Rh1 are common changes during fermentation, regardless of the strain type. These results are similar to those reported in the studies on ginseng fermentation using B. subtilis by Kim et al. [12] and Lim et al. [19] . In particular, Kim et al. [12] confirmed that Rb1 and Rc decreased and Rd increased in ginseng fermentation using B. subtilis. Lim et al. [19] also reported that Rb1 and Rc decreased from 4.93 and 3.03 mg/g to 2.46 and 2.62 mg/g, respectively, whereas Rd increased from 0.66 to 7.21 mg/g. Kim et al. [12] suspected that Rc is degraded by ginsenoside-α-arabinofuranase, and that the sugar part of Rb1 is degraded to Rd by γ-glucosidase. Park et al. [22] and Zhang et al. [23] also reported that Rc is considered to be converted into Rd by ginsenoside-α-arabino- furanase and Rb1 into Rd by γ-glucosidase. These results can be said to confirm the potential of probiotics when eaten raw, and both nutritional and functional benefits are expected depending on the conversion of ginsenosides.

The major component of the viscous substance in Cheonggukjang is γ-polyglu- tamate, one of the metabolites produced by microorganisms belonging to the genus Bacillus during fermentation [24] [25] and a biopolymer with functionality such as osteoporosis prevention [26] . Thus, the amount of viscous substances produced during ginseng fermentation using the strain of Cheonggukjang origin was confirmed, as shown in Figure 3. The amount of viscous substances observed in the A strain medium and the B strain medium after 3 days of fermentation was 6.253% and 7.120%, respectively, decreasing to 2.773% and 3.553%, respectively, after 5 days of fermentation. However, after 7 days of fermentation, no viscous substances were observed in the medium. Polyglutamates, the major viscous substance in Cheonggukjang, generally occur as 2.15% - 6.03% [17] . However, that the viscous substance content of fermented ginseng in the study was a maximum of 7.120%, which was 1.18 - 3.31 times higher than in Cheong-

gukjang. When the amount of viscous substances was compared, the content from the B strain fermented medium was 1.14 - 1.27 times higher than that from the A strain, suggesting a more active production of viscous substances from the medium using the B strain. Thus, it is believed that the functionality of viscous substances will be utilized more when fermented ginseng is used.

The changes in the physical properties of the ginseng medium during fermentation showed that the viscosity gradually decreased as the shear rate decreased, and then stabilized to a certain level of viscosity regardless of the type of strain (Figure 4). The viscosity of the medium fermented using the A strain was initially 399.85 Pa.s, but this gradually decreased to 224.9 Pa.s, 115.3 Pa.s, 83.4 Pa.s, and 66.9 Pa.s as fermentation progressed. In the case of the B strain, the viscosity was 362.6 Pa.s immediately after inoculation and 258.5 Pa.s after 3 days of fermentation, 96.8 Pa.s after 5 days of fermentation, 75.8 Pa.s after 7 days of fermentation, and 78.01 Pa.s after 10 days of fermentation, showing that the initial viscosity of the medium inoculated by the B strain also decreased as fermen-

tation progressed. Initially, the viscosity was high due to the viscosity of the ginseng solid medium itself, but it gradually decreased by fermentation and then stabilized at higher shear rates as the fermentation period increased. When comparing the A and B strains, the initial viscosity immediately after the Bacillus inoculation was higher in the medium inoculated by the A strain; after 3 days of fermentation, however, the initial viscosity of the B strain was 33.6 Pa.s higher compared to the A strain, meaning that the viscous substance content and the viscosity were proportional, which is thought to be a result of the production of viscous substances with polymerized fructose and glutamic acid by B. subtilis [27] . Lee et al. [28] manufactured Cheonggukjang by controlling the ratio of B. subtilis and Lactobacillus sakei with an inoculation concentration of 1% and reported that the viscosity was higher with a higher ratio of B. subtilis MC31. This latter finding is consistent with the result that the viscosity increased with an increasing amount of viscous substances. Note, however, that the decrease of viscosity after 3 days of fermentation quickly occurred in the fermented ginseng including the B strain. This result suggests that poly γ-glutamate depolymerase, a degrading enzyme of viscous substances, is generated as an extracellular enzyme, degrades viscous substances, and decreases the viscosity [29] .

Strains in the genus Bacillus used in the production of Cheonggukjang generally have fibrinolytic activity, but some strains without fibrinolytic activity actually exist. Therefore, the fibrinolytic activity of fermented ginseng by B. subtilis isolated from Cheonggukjang was identified (Figure 5). The fibrinolytic activity of the medium was not confirmed immediately after inoculation regardless of the strain, but the activity of the medium fermented by the A strain was 85.0% after 3 days of fermentation. Moreover, fibrinolytic activity was similar or increasing as fermentation continued. The fibrinolytic activity of the ginseng medium fermented by the B strain was 100.0% after 3 days of fermentation-which was the same effect as plasmin 1.0 unit/mL-and 97.5% after 5 days of fermentation, which was 2.5% lower compared to the 3-day fermentation, but fibrinolytic activity was maintained for up to 10 days of fermentation, demonstrating a 10.0% - 15.0% higher effect than that using the A strain. This result is consistent with the one reported by Mann et al. [30] wherein fibrinolytic activity was observed in Cheonggukjang fermented for 72 hours using B. subtilis MC 31. Moreover, No et al. [31] Doenjang made with Aspergillus oryzae D-2 showed 5 mm of fibrinolytic activity occurring immediately after its production showed an increase in fibrinolytic activity of up to 20 mm until day 30 of fermentation, followed by a gradual decrease after 30 days. Thus, it is expected that the fibrinolytic activity will not decrease significantly until 10 days of ginseng fermentation, but will decrease as the fermentation period is extended.