Introduction

Doenjang is a protein-rich traditional food product and flavouring ingredient in Korea that is produced by the fermentation of soybeans. Traditionally, in the preparation of starter culture (meju) for doenjang, soybeans are fermented by Bacillus species in the early stage, followed by the secondary microflora of Aspergillus species such as Aspergillus oryzae and Aspergillus niger1. Processing of traditional doenjang differs depending on its region of origin1. In addition, traditional meju can be made from various salt solutions and is fermented for more than 2 months2. The taste and colour of doenjang are considered as the main factors responsible for its overall acceptability, which is highly influenced by the starter culture and the presence of oxygen1. A standardised starter culture would be advantageous to ensure the rapid production of doenjang with improved quality and safety.

Several microorganisms, mainly filamentous fungi including Rhizopus species, Bacillus species, and lactic acid bacteria are commonly found during the fermentation of soybeans3. Rhizopus species have an especially vital role in the food industry due to their ability to produce thermostable polygalacturonases, which hydrolyses α-1,4 glycosidic bonds between galacturonic acid residues and thus affect the plant cell wall4. In particular, R. oligosporus is used to ferment soybeans to make tempeh, which has many health benefits including potent anti-radical activity5. R. nigricans (also named as Rhizopus stolonifera) ferments the ground nut flour, which improves its nutritional value6. R. nigricans has received increased interest in the fermentation industry because of its ability to produce fumaric acid abundantly, which has acidifying and gelling properties7,8.

Safety concerns with doenjang include microbial contamination, formation of biogenic amines, and production of aflatoxins, which depend on various factors, such as raw materials, microbial flora, fermentation conditions, and salt content9. Several studies have been carried out to produce commercially modified traditional Korean doenjang1. However, no studies have so far been carried out on commercially modified meju prepared by the inoculation of a mixture of various starter cultures, especially Rhizopus strains. The present study sought to develop a novel starter culture in combination with R. nigricans for the preparation of doenjang. Prepared doenjang samples were evaluated for their nutritional and safety parameters. Additionally, the cellular antioxidant activity of the selected doenjang sample was carried out to demonstrate its potential against oxidative stress induced by various reactive oxygen species (ROS).

Materials and Methods

Chemicals and reagents

All chemicals and solvents were of the analytical and chromatographic grade. Standard biogenic amines (agmatine sulfate, tryptamine hydrochloride, 2-phenylethylamine, putrescine dihydrochloride, cadaverine dihydrochloride, histamine dihydrochloride, tyramine hydrochloride, spermidine trihydrochloride, and spermine tetrahydrochloride), dansyl chloride, and acetone were purchased from Sigma-Aldrich (St. Louis, MO, USA). Sodium hydroxide, sodium hydrogen carbonate, ammonium hydroxide, and perchloric acid were purchased from Junsei Chemicals (Tokyo, Japan). High-performance liquid chromatography (HPLC) grade acetonitrile and ammonium acetate were obtained from Merck (Darmstadt, Germany). Nutrient agar was purchased from Difco (Sparks, MD, USA). API kits were purchased from BioMerieux (Marcy I’ Etoile, France).

Microbial strains

Bacillus subtilis TKSP 24, A. oryzae complex, Mucor racemosus 15, and R. nigricans were isolated from traditional meju and identified using 16 S rRNA and internal transcribed spacer gene sequence analyses. All microorganisms were grown individually under aseptic conditions and used as a starter culture. The microbial strains were deposited in National Center for Biotechnology Information (NCBI) GenBank and Korean Collection for Type Cultures (KCTC), Republic of Korea under the following accession numbers: B. subtilis TKSP 24 (KCTC 11840BP; NCBI GenBank accession No. MN784431), A. oryzae complex (KCTC 11841BP; NCBI GenBank accession No. MN784429), M. racemosus 15 (KCTC 11842BP; NCBI GenBank accession No. MN784430), R. nigricans (R. stolonifera) (KCTC 12107BP; NCBI GenBank accession No. MN791090).

Preparation of meju and doenjang

Previously prepared starter cultures were mixed separately with boiled soybeans at a concentration of 1% (w/w) on a stainless-steel square plate (36 cm × 32 cm × 6 cm). At 70% relative humidity, A. oryzae complex, M. racemosus 15, and R. nigricans seeded soybeans were incubated for 3 days at 28 °C. They were designated A meju, M15 meju, and R meju, respectively. Soybeans seeded with B. subtilis TKSP 24 were incubated at 30 °C for 2 days and named as B meju. Six different doenjang samples were prepared by mixing these four starter cultures in different proportions. Doenjang samples, BAR-1, BAR-2, and BAR-3, were prepared by mixing B. subtilis meju, A. oryzae complex meju, and R. nigricans meju at 1:1:1 (w/w), 1:0.5:1.5 (w/w), and 1:1.5:0.5 (w/w), respectively. Similarly, doenjang samples, BM15R-1, BM15R-2, and BM15R-3, were prepared by mixing B. subtilis meju, M. racemosus 15 meju, and R. nigricans meju at 1:1:1 (w/w), 1:0.5:1.5 (w/w), and 1:1.5:0.5 (w/w), respectively (Table 1). All meju preparations were fermented at 30 °C for 30 days in the presence of a 15% salt solution10. After 30 days of fermentation, the solid portion was separated from the liquid and re-fermented at 30 °C for 30 days to obtain the final doenjang product. Traditional Korean doenjang prepared with natural microflora served as a control. The above described doenjang and meju production processes were repeated several times to confirm reproducibility.

Table 1 List of doenjang samples fermented with various meju products and their pH values.
Full size table

Measurement of pH

The pH levels of doenjang samples was determined as previously described method2 using a pH meter (Thermo Electroncorporation; Beverly, MA, USA).

Colour analysis

The colour values of doenjang samples was determined by using a chromameter (Minolta; Osaka, Japan). Doenjang samples were placed on a white surface and the Hunter L*, a*, and b* colour values were measured as previously described method11.

Analysis of taste components

Sample preparation

A previously described method12 was adopted for the extraction of free amino acids, free sugars, and organic acids (volatile and non-volatile) from doenjang samples. Briefly, 100 g of doenjang was refluxed with 150 ml of 85% ethanol at 65 °C for 1 h, followed by filtration. The residual was re-extracted twice using same conditions. The collected filtrates were pooled and centrifuged at 4000 × g for 20 min at 4 °C. The supernatant was collected and vacuum-dried.

Quantification of free sugar contents

Free sugars were separated by passing the extracted doenjang samples on ion exchange column as previously described12. Amberlite IR-120 and amberlite IRA-400 columns were used as cation and anion exchangers for chromatographic separation, respectively. Quantitative analysis of the eluted free sugars was carried out by HPLC using a model 600E HPLC unit (Waters; Milliford, MA, USA), equipped with an RI-Model 410 refractive index detector. Separation was achieved using a Sugar-Pak I column (6.5 mm internal diameter × 300 mm length; Waters) using Ca-EDTA buffer solution (50 mg of Ca-EDTA per litre of distilled water) as the mobile phase at a constant flow rate of 0.5 ml/min. Free sugars were identified and quantified in doenjang samples by comparison with standard sugar samples of known strength.

Quantification of free amino acids and non-volatile organic acids

Quantification of free amino acids and non-volatile organic acids in doenjang samples were carried out by the method adopted by Shukla et al.12. Briefly, the amberlite IR-120 and IRA-400 columns used earlier for free sugar collection were washed with 2 N NH4OH and 2 N (NH4)2CO3 for the recovery of bound free amino acids and non-volatile organic acids. Non-volatile organic acids were quantified by HPLC using an RS Pak KC-811column (8.0 mm internal diameter × 300 mm length) at a constant column temperature of 40 °C. H3PO4 solution (0.1% in distilled water) was used as a mobile phase with a constant flow rate of 1 ml/min.

Separation and quantification of amino acids was made using an amino acid analyser (Hitachi; Tokyo, Japan), equipped with a cation exchange resin in a column (4.6 mm internal diameter × 60 mm length) at a temperature of 30 °C to 70 °C as a method adopted by Shukla et al.12.

Quantification of volatile organic acids

Extraction and quantification of volatile organic acids were carried as previously described method13. In brief, 2 µl of each extracted doenjang sample and standard were separately injected into a gas chromatography system (Hewlett-Packard; Palo Alto, CA, USA). Separation was carried out in a glass column filled with 10% polyethylene glycol 6000 (Shimazu; Tokyo, Japan) using helium as a carrier gas at a constant flow rate of 0.9 ml/min. The column was maintained at 170 °C throughout the operation process.

Determination of biogenic amines

Extraction of biogenic amines

For extraction of biogenic amines, 5 g of doenjang sample was homogenized with 10 ml of 0.4 M perchloric acid, followed by centrifugation at 4000 × g for 10 min. The supernatant was collected, and the pellet was re-extracted twice with 10 ml of 0.4 M perchloric acid. All the collected supernatant was pooled and processed for derivatisation before HPLC analysis.

Derivatisation of biogenic amines

Derivatisation of biogenic amines was carried out by mixing 1 ml of extracted doenjang sample with 0.2 ml of 2 M NaOH, 0.3 ml of saturated NaHCO3, and 2 ml of 10 mg/ml dansyl chloride, followed by 45 min incubation at 40 °C. After incubation, the reaction was stop by adding 100 µl of 25% NH4OH, and the volume was adjusted to 5 ml by acetonitrile, followed by centrifugation at 2,500 × g for 5 min. Finally, samples were filtered through a 0.2 µm syringe filter (Woongki Science Co.; Seoul, Korea) and processed for HPLC analysis.

Separation and quantification of biogenic amines

Separation of biogenic amines was carried out using HPLC (Agilent Technologies; Santa Clara, CA, USA), equipped with an ultraviolet-visible spectrophotometer as a detector. A 20 µl volume of derivatised extracted biogenic amines sample was injected into an Agilent C-18 column (4.6 mm internal diameter × 150 mm length, pore size of 5 µm). Separation was carried by gradient elution using 0.1 M ammonium acetate and acetonitrile mobile phase at a constant flow rate of 1 ml/min. Biogenic amines were analysed at 254 nm. Quantification and identification of biogenic amines in samples were made by comparison with reference standards.

Cellular antioxidant potential

To evaluate the functional property of doenjang samples, cellular antioxidant activity was carried out using BAR-1 as a representative candidate. For the evaluation of cellular antioxidant activity, 10 g of BAR-1 doenjang was extracted with 100 ml distilled water, followed by filtration and freeze-drying to obtain the dry powder.

Cell and cell culture

NIH 3T3 mouse fibroblast cells were purchased from American Type Culture Collection (Manassas, VA, USA). Cells were grown and maintained in Dulbecco’s modified Eagle’s medium [10% (v/v) fetal calf serum and 1% (v/v) penicillin-streptomycin mix] in a 5% CO2 incubator at 37 °C.

Cell viability assay

Effect of BAR-1 doenjang on the cell viability of NIH 3T3 cells was evaluated by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay14 using different concentrations of BAR-1 (0–10 mg/ml).

Determination of cellular ROS

Cellular ROS were determined by 2′,7′-dichlorofluorescein diacetate (DCFDA) assay15. In brief, NIH 3T3 cells were grown in 24-well culture plates, followed by the treatment with various concentrations of BAR-1 (0 mg/ml, 0.5 mg/ml, 1.0 mg/ml, and 2.5 mg/ml). After 24 h incubation, oxidative stress in the cells was induced by treatment with 0.25 mM hydrogen peroxide (H2O2) for 2 h. Finally, cells were washed twice with 0.1 M phosphate buffered saline (PBS, pH 7.4), followed by the addition of DCFDA (20 µM). After 30 min of incubation, the stained cells were washed with 0.1 M PBS and immediately visualised by fluorescence microscopy (Nikon; Tokyo, Japan). The fluorescence intensity was quantified using Image J software (NIH; Bethesda, MD, USA).

Analysis of microbial counts

Based on profiling of amino acids, volatile and non-volatile organic acids, and sugar derivatives, two doenjang samples (BAR-1 and BM15R-1) were selected as representatives and analysed for microbial food pathogens. Pathogenic microbes were analysed as previously described methods16. The analysis was carried out for the presence or absence of foodborne pathogens, including total coliforms, Staphylococcus aureus, Escherichia coli O157, and Salmonella species, using selective culture-dependent methods. Isolates showing typical pathogen morphologies were further identified using a microbial pathogen detection API kit.

Determination of total aflatoxin levels

Analysis of aflatoxins in doenjang samples of BAR-1 and BM15R-1 was performed according to the instructions of the manufacturer of the quantitative aflatoxin detection kit (Neogen Corp.; Lansing, MI, USA). Aflatoxin was extracted by homogenizing 10 g of each sample in 50 ml of 70% methanol, followed by filtration. Filtered samples were transferred into wells of an ELISA plate pre-coated with an antibody against aflatoxin. After sufficient antigen (aflatoxin)-antibody interaction, a chromogenic substrate was added, and the intensity of the colour was measured at 650 nm using ELISA reader (Tecan; Mannedorf, Switzerland). Quantification of aflatoxin was made using Veratox data reduction software (Neogen Corp.) by converting absorbance into concentration against reference standards.

Statistical analyses

Results were presented as mean ± standard deviation (SD) of three independent experiments. Statistical analyses involved one-way ANOVA, followed by Duncan test for post-hoc analysis using IBM SPSS 19 program (SPSS Inc; Chicago, IL, USA). P < 0.05 was considered to be significantly different.

Results and Discussion

Doenjang production

Six different kinds of doenjang samples were prepared using meju starter culture, formed by four different types microorganisms (Table 1). The doenjang sample designated BAR was prepared by mixing different ratios of meju starter culture made up of B. subtilis, A. oryzae complex, and R. nigricans. While doenjang samples designated as BM15R were prepared by mixing meju starter culture comprised of B. subtilis, M. racemosus 15, and R. nigricans. Three different types of doenjang designated as BAR-1, BAR-2, and BAR-3 were prepared by mixing B. subtilis meju, A. oryzae complex meju, and R. nigricans meju at ratio of 1:1:1 (w/w), 1:0.5:1.5 (w/w), and 1:1.5:0.5 (w/w), respectively. Similarly, doenjang samples, BM15R-1, BM15R-2, and BM15R-3, were prepared by blending B. subtilis meju, M. racemosus 15 meju, and R. nigricans meju at a ratio of 1:1:1 (w/w), 1:0.5:1.5 (w/w), and 1:1.5:0.5 (w/w), respectively.

Measurement of pH

pH is one of the most important parameters influencing proper fermentation and many biochemical events in fermented foods. The pH values of all the prepared doenjang samples are shown in Table 1. All doenjang samples displayed a slightly acidic pH, ranging from 4.97 ± 0.01 to 5.60 ± 0.11. The pH was the highest in control (5.60 ± 0.11) and the lowest in BAR-1 (4.97 ± 0.01). In fermented foods, a pH below 4 to 5 is usually considered to be safe, as most microbial pathogens are unable to survive in such a low pH17. The results were consistent with our previous findings in which doenjang prepared using different meju starter cultures displayed an acidic pH10.

Colour characteristics

Colour is an important physical characteristic of food that catches the attention of the consumer and promotes acceptance. Presently, all six doenjang samples showed the yellow surface colour. The maximum yellow surface colour was observed for BAR-1 (24.70 ± 1.40) and minimum for BAR-3 (10.74 ± 0.44). The inner region of all doenjang samples displayed a highly intense yellow colour, compared to the outer surface (Table 2). The most likely reason for the more intense inner colour is the Maillard reaction that occur during fermentation and involves several amino acids, sugars, and other components. During fermentation, the reducing sugar enhanced by the action of microbes on the complex carbohydrate is responsible for browning due to the Maillard reaction18. The enhanced inner yellow colour value (b* value) of 25.64%, 88.64%, and 80.07% was observed in BAR-1, BAR-2, and BAR-3, respectively, compared to the surface colour value. A similar trend was observed in doenjang samples BM15R-1, BM15R-2, and BM15R-3, where the inner colour value was enhanced by 25.55%, 36.20%, and 27.85%, respectively, compared to that of the surface. Both the major preparations, BAR and BM15R, displayed inner colour changes as compared to the surface, however, the most prominent changes were observed in doenjang prepared by BAR meju. These findings suggested an important role of A. oryzae complex in combination with B. subtilis and R. nigricans in yellow colour development. These results agreed well with the published report, indicating the different colours of doenjang samples, based on the starter culture used12.

Table 2 Colour values of doenjang samples fermented with various meju.
Full size table

Free amino acid contents

The contents of individual and total free amino acids of doenjang samples are listed in Table 3. Free amino acids present in food products are responsible for the special flavours and are categorised on the basis of their taste characteristics such as sweet, savoury, sour, and bitter19. Apart from imparting the flavour to food, amino acids are vital nutrients that have a role as growth factors and stabilizer of nucleic acid20. In this study, traditional Korean doenjang (control) displayed the highest content of free amino acids responsible for sweet taste (390.20 mg/100 g), followed by BAR-2 (322.50 mg/100 g), and BM15R-3 (320.07 mg/100 g). Of the doenjang samples tested for free amino acids related to savoury taste, BAR-2 showed a higher score (285.65 mg/100 g) than traditional doenjang (263.20 mg/100 g), followed by BAR-3 (234.49 mg/100 g) and BAR-1 (228.64 mg/100 g). For free amino acids related to bitter taste, BAR-2 (221.73 mg/100 g) showed the highest contents among all tested samples of doenjang, followed by BM15R-3 (220.46 mg/100 g) and traditional doenjang (209.50 mg/100 g). BM15R-1 displayed the lowest amino acid contents related to bitter taste (125.71 mg/100 g). The highest amount of amino acids related to other tastes was in BAR-2 (226.42 mg/100 g), followed by BAR-1 (196.68 mg/100 g), with lowest content in BM15R-1 (113.63 mg/100 g). The results agreed well with the previous findings of the varied amount of free amino acids in doenjang samples and other fermented products21. The amount of free amino acids is strongly dependent on the raw material and the microorganisms that participate in fermentation. The variation in the content of free amino acids in different doenjang samples can be explained by the fact that they use a different meju as a starter culture, and thus the involvement of diverse microorganisms results in consequently varied free amino acids. The results are consistent with our previous report that demonstrated diversification in the free amino acids in different doenjang samples with the difference of meju as a starter culture12.

Table 3 Compositions of free amino acids in doenjang samples fermented with various meju.
Full size table

Free sugar contents

During fermentation of doenjang, various microbial enzymes act on the complex food material and convert it into simple biomolecules such as amino acids, sugar, organic acids, and fatty acids, which are precursors of various volatile compounds22. Free sugar content is not only responsible for the nutritive value of food but also important for flavour characteristics. Figure 1a depicts the total free sugar contents of the tested doenjang samples. BM15R-1 and BAR-1 displayed the highest total free sugar contents (55.0 mg/100 g and 47.65 mg/100 g, respectively). Among the five tested sugars, fructose was present in all the doenjang samples. However, the sugar sorbitol was present only in BAR-2. We speculate that the variations in free sugar contents were due to differences in the meju starter culture as a result of the involvement of different microorganisms and the enzyme produced by them23. Many reports have demonstrated that the varied amount of free sugars in fermented foods are associated with the involvement of microorganisms. One such a study conducted by our group demonstrated the influence of different meju starter culture on the variation in sugar and other metabolites in doenjang12.

Figure 1
figure 1

Compositions of free sugars and non-volatile and volatile organic acids in doenjang samples fermented with various meju.

Full size image

Non-volatile and volatile organic acid contents

During fermentation, microorganisms convert simple organic biomolecules into various useful products such as organic acids, which impart characteristic tastes, aromas, and functional properties to foods. Organic acids (citric acid and lactic acid) are well known for the improvement of the taste of soy sauce24. Knowing the importance of organic acids in fermented foods, we quantified organic acids in the doenjang samples, and results are presented in Fig. 1b,c. Among the two non-volatile organic acids tested, oxalic acid was present only in BAR-2 and BM15R-3. On the other side, glutaric acid was not detected in any of the doenjang sample. However, glutaric acid was abundant in the control doenjang sample.

Among the four volatile organic acids that were tested, acetic acid was present in all the doenjang samples with a significantly high amount in BAR-1. In contrast, a high amount of butyric acid was present in BAR-2. Overall, the total volatile organic acids were highest and lowest in BAR-1 and BM15R-1, respectively. These findings indicate the influence of the meju starter culture on the production of organic acids, consistent with a previous report concerning the role of starter culture on volatile and non-volatile organic acids in doenjang12.

Biogenic amines contents

The most common contamination of soybean-based fermented foods is biogenic amines25, which are formed by the decarboxylation of amino acids and serve as a good indicator of food quality26. Some biogenic amines, such as putrescine, cadaverine, and agmatine, react with nitrites and are converted into a potential carcinogenic (nitrosamine) and thus pose a serious threat to health26. Other than this, aromatic biogenic amines, such as tyramine and 2-phenylethylamine, are the main agents of a dietary-induced migraine and hypertensive crisis26. Knowing the importance of biogenic amines in the food, we quantified them in the doenjang samples. The results are provided in Table 4. The six doenjang samples varied in the amount of biogenic amines. The maximum amount of total biogenic amines was observed in BAR-2 (6.79 mg/100 g), followed by BAR-1 (5.27 mg/100 g). Biogenic amines were lowest in BM15R-3 (0.80 mg/100 g). A negative correlation (y = −3.02 × + 8.40, r2 = 0.99) between the amount of biogenic amines and the addition of A. oryzae complex meju was established in doenjang samples (Table 4). Among the different biogenic amines, histamine is considered a serious food toxicants and is the cause of histamine poisoning26. In all six doenjang samples, the histamine level ranged from 0.00 ± 0.00 (BAR-1 and BAR-2) to 0.37 ± 0.02 mg/100 g (BM15R-1). Intermediate values were 0.17 ± 0.03 mg/100 g (BAR-3) and 0.13 ± 0.01 mg/100 g (BM15R-2). The amount of histamine in all the doenjang samples was well below the toxicity level26, suggesting that the preparations were safe to consume. Biogenic amines in all the samples were also well below the established toxic level26. The results are markedly lower than the reported content of biogenic amines in doenjang9,27. The present results are consistent with our earlier report of a similar profile of biogenic amines in doenjang samples prepared by the combination of different meju12. The present findings also reveal that none of the biogenic amines in the doenjang samples were present in high amounts, which might be due to the use of specific starter cultures that have an impact on the inhibition of biogenic amines formation. Results are well supported by the findings of Zaman et al.28, demonstrating the role of starter culture on the inhibition of biogenic amines.

Table 4 Determination of biogenic amines in doenjang fermented with various meju.
Full size table

Cellular antioxidant potential

The effect of BAR-1 on the viability of NIH 3T3 cells was evaluated by the MTT assay. The results are depicted in Fig. 2. BAR-1 applied at concentrations up to 2.5 mg/ml had no significant effect on the viability of cells. However, at higher concentrations, the cell viability decreased significantly. The results of cellular antioxidant activity evaluated by DCFDA staining revealed a dose-dependent effect of BAR-1on the inhibition of ROS production in NIH 3T3 cells challenged by H2O2. A dose-dependent decreased fluorescent intensity of 97.53 ± 7.40%, 86.71 ± 6.40%, and 77.61 ± 4.10% was observed in the cells treated with 0.5 mg/ml, 1.0 mg/ml, and 2.5 mg/ml of BAR-1 respectively, as compared to 100% in only H2O2 stimulated cells (Fig. 3a,b). Previously, many reports have been described the abundance of phenolics and flavonoids in various soy fermented foods29,30,31. We speculate that the cellular antioxidant activity of BAR-1 is due to the presence of these diverse phytochemicals. Consistent with this, the role of different phenolics and flavonoids as antioxidants by direct scavenging of ROS or by inducing cellular antioxidants has been reported32. To the best of our knowledge, no prior study has indicated the cellular antioxidant nature of doenjang. However, few reports showing the in vitro antioxidant potential of doenjang33.

Figure 2
figure 2

Effect of doenjang sample BAR-1 (1–10 mg/ml) on the viability of NIH 3T3 fibroblast cells evaluated by the MTT assay. Results are depicted as mean ± SD of three independent experiments. Values with different letters differ significantly from each other (P < 0.05).

Full size image
Figure 3
figure 3

Effect of doenjang sample BAR-1 on reactive oxygen species in the NIH 3T3 fibroblast cells challenged by H2O2. (a) DCFDA staining (I) without H2O2 and BAR-1, (II) with only 0.25 mM H2O2, (III) with 0.5 mg/ml BAR-1 + 0.25 mM H2O2, (IV) with 1 mg/ml BAR-1 + 0.25 mM H2O2, and (V) with 2.5 mg/ml BAR-1 + 0.25 mM H2O2. All images were taken using an EPI florescence microscope at 40 × magnifications (scale bar = 0.1 mm). (b) Fluorescence intensity was determined using Image J software. Each value represents mean ± SD of three independent experiments. Values with different letters differ significantly from each other (P < 0.05).

Full size image

Determination of microbial counts

BAR-1 and BM15R-1 were examined for the viable counts of specific food pathogens. According to Gilbert et al.34, satisfactory, acceptable, and unsatisfactory limits of detection for coliform bacteria are >2 log CFU/g, 2–3 log CFU/g, and <3 log CFU/g, respectively, in ready-to-eat foods. In both doenjang samples, S. aureus, E. coli O157: H7, Salmonella species, and members of the coliform group were not detected (Table 5). The results indicated that the microbial safety of doenjang preparations and their good hygienic condition. There are various reasons of the absence of pathogenic microbes. The most prominent is the hygienic manufacturing conditions. Other than this the metabolic profile of prepared doenjang, that contains organic acids, which are well known for their antimicrobial properties35.

Table 5 Determination of foodborne pathogenic bacteria and aflatoxins in various doenjang samples.
Full size table

Determination of total aflatoxins

Aflatoxin is a fungal secondary metabolite and the most serious toxicant in food with an adverse health effect even at the lower concentration. According to the guidelines of the Korea Food and Drug Administration, an aflatoxin level of 10 µg/kg (ppb) in food samples is considered hazardous16. Food is one of the major sources of aflatoxin poisoning. Hence, we examined the aflatoxin level in the prepared doenjang. The level was below the limit of quantification (<5 µg/kg), as described in the Veratox total aflatoxin test kit instructions (Table 5). The absence of pathogenic microbes and the nontoxic level of aflatoxin suggest the prepared doenjang to meet the current food safety demand.

Conclusion

In summary, this study focused on the use of R. nigricans-based starter culture, featuring the combined inoculation of bacteria and fungus to improve the quality of doenjang. Six different types of doenjang samples were manufactured and analysed for their taste and safety characteristics. Furthermore, BAR-1 produced substantial ROS inhibition at the cellular level and was effective in mitigating the harmful effect exerted by ROS. None of the doenjang samples contained the biogenic amines above their safety limits. Therefore, it can be concluded that selection of appropriate microorganisms in the proper ratio with R. nigricans could be used as a novel starter culture for the production of doenjang with enhanced levels of nutritional and functional components and quality characteristics.