2023 年 29 巻 1 号 p. 15-25
For rare sugars to be used in fermented foods, their effect on microorganisms used in soy sauce brewing must first be determined. Here, when soy sauce-brewing microorganisms were cultured with a mixture of four rare sugars (d-allulose, d-allose, d-tagatose, and d-sorbose) with d-glucose, growth was not inhibited when the percentage of each rare sugar did not exceed 60% of the carbon source. d-Tagatose was not inhibitory, even at 100% of the carbon source, and it had the same effect as d-glucose on lactic acid fermentation by Tetragenococcus halophilus. To confirm soy sauce can be brewed with rare sugars, we conducted a small-scale brewing study with a rare sugar syrup (RSS). After aerobic fermentation by the yeast, the relative composition of each rare sugar remaining in the brew was 95% allulose, 34% allose, 68% tagatose, and 84% sorbose of the respective original concentration. Analyses of the final product found that acidic bitterness and acidic bitter aftertaste of the soy sauce were suppressed compared to the control, and this was attributed to a decrease in the amount of basic amino acids, which might be related to bitterness, during brewing.
A rare sugar syrup (RSS), developed by alkaline isomerization of high-fructose corn syrup, contains a total of 12% (w/v) rare sugars including allulose (psicose), allose, tagatose, and sorbose (Takamine et al., 2015). In 2019, RSS was submitted as a “Foods with Function Claims” by our team and approved, as the syrup suppresses the rapid rise in blood glucose level following food intakei). Thus, the syrup is expected to be widely used in foods as a functional food material to improve health.
To be effective at improving health, foods with functional ingredients must be consumed on a continual basis as part of the daily diet, ideally in small amounts within assorted foods, e.g., as seasonings. Soy sauce is widely used as a "universal seasoning" in a large number of processed soy sauce products (Okuzawa, 1994; Takashima, 1991). Some processed products containing soy sauce (e.g., dashi shoyu and tsuyu) with added RSS as a sweetener have been launched; however, these products are simply mixed with the RSS as a sweetener, and the glucose and fructose contents of the processed product are not adjusted. Therefore, if the added syrup is increased to increase the amount of the functional rare sugar, then the amount of glucose and fructose in a person's diet will also increase.
Soy sauce is a brown, salty, fermented seasoning used in China, Japan, Korea, and other Asian countries for table seasoning and cooking. Soy sauce is produced by using wheat and soybeans (Glycine max) as the raw materials, making koji with yellow koji mold (Aspergillus oryzae and/or Aspergillus sojae), brewing with 18–20% salt water, hydrolyzing the wheat or soybean material with carbohydrate-degrading enzymes and proteolytic enzymes derived from koji mold to release sugar and amino acids, and fermentation and maturation with salt-tolerant lactic acid bacteria and salttolerant yeastii) (Luh, 1995). During soy sauce brewing, sugar is used as a carbon source for microorganisms and is important for imparting color and flavor to soy sauce through the Maillard reaction (Motai, 1976; Zhao et al., 2018). When RSS is used as a raw material for soy sauce brewing, glucose and fructose in the syrup can be used to add color and flavor and may be converted to organic acids and ethanol by lactic acid fermentation and alcoholic fermentation during brewing, thereby reducing their contents. Apart from having the general properties of sugars, rare sugars can have effects such as inhibiting microbial proliferation (Mochizuki et al., 2020; Sato et al., 1993) and improving taste and physical characteristics (Innun et al., 2007; Kimoto-Nira et al., 2017); thus, it is necessary to assess microbial growth, the fermentation parameters, and the final product when rare sugars are used to ferment foods.
In yogurt, it was reported that the rare sugar d-allulose may inhibit the acid production of certain lactic acid bacteria without changing their probiotic activity (Kimoto-Nira et al., 2017); however, the effect of rare sugars on microbes during the production of high-salt fermented foods has not yet been clarified. Therefore, with the aim of utilizing rare sugars in the fermentation of foods, especially soy sauce, this study investigated the effects of four types of rare sugars in RSS on the growth of brewing microorganisms involved in fermentation. Moreover, we conducted a small-scale brewing test of soy sauce using RSS to explore the effect of rare sugars in the soy sauce brewing process and its potential for commercialization.
Chemicals Rare sugar syrup (RSS; concentration of each sugar: 44.3% glucose, 31.9% fructose, 6.0% allulose, and 17.8% other saccharides) (Hayashi et al., 2014) was purchased from Matsutani Chemical Industry Co., Ltd. (Hyogo), and the pure crystal form (100% purity) of various rare sugars, including d-allulose, d-allose, d-tagatose, d-sorbose, and L-iditol, was supplied by International Institute of Rare Sugar Research and Education, Kagawa University (Kagawa). Unless otherwise specified, all other sugars and chemicals were obtained from FUJIFILM Wako Pure Chemical Co. (Osaka).
Microorganisms and culture conditions Salt-tolerant strains Zygosaccharomyces rouxii ZR510 and Tetragenococcus halophilus No.1, 3, 6, and 9, which our research branch produces and sells to soy sauce brewing companies nationwide, were used. To make the koji for the small-batch brewing test, a commercially available soy sauce brewing seed koji, Three-dia (Higuchi Matsunosuke Shoten Co., Ltd., Osaka), was used.
Effect of various rare sugar contents on microbial growth and fermentation To prepare the basal medium for Z. rouxii strain ZR510, 14 g of NaCl, 6 mL of raw soy sauce (1.72% w/v of total nitrogen and 1.70% w/v of reducing sugar) and 0.1 g of dipotassium phosphate were added to 90 mL of distilled water to achieve pH 6.1, and then 9 mL was dispensed into test tubes and autoclaved at 121 °C for 15 min. For testing the rare sugars individually, a 20% (w/v) solution of the rare sugar was mixed with 20% (w/v) d-glucose (glucose) solution at a ratio of 0:5, 1:4, 2:3, 3:2, 4:1, or 5:0, filtered through a 0.2 µm Dismic syringe filter unit 25CS020AS (Toyo Roshi Kaisha, Ltd., Tokyo), then 1 mL of the sugar solution was aseptically added to the basal medium. Next, 0.1 mL of strain ZR510, which had been previously cultured to the logarithmic growth phase in basal medium containing 2% (w/v) glucose as the carbon source, was added and the test tubes were incubated at an angle at 30 °C for 7 d.
T. halophilus strain No.9 was cultured on basal medium for lactic acid bacteria. Twelve grams of NaCl, 20 mL of raw soy sauce, which contains growth factors (Sakaguchi, 1959), 0.5 g of dipotassium phosphate, and 0.3 g of yeast extract were dissolved in distilled water, pH 7.2, and the volume was adjusted to 90 mL, then 9 mL was dispensed into individual test tubes and autoclaved at 115 °C for 15 min. This solution was used as the basal medium, and a 20% (w/v) sugar solution, adjusted to a predetermined ratio of rare sugar to glucose, was added to the basal medium as described for the yeast. Next, 0.1 mL of strain No. 9, which had been previously cultured to the logarithmic growth phase in basic medium containing 2% (w/v) glucose as the carbon source, was added, and the cells were incubated at 30 °C for 10 d under static conditions.
Soy sauce-style fermented seasonings manufactured with or without RSS A small batch test of soy sauce-style fermented seasonings was conducted as follows based on the method of Luh (1995) with some modifications. Koji was prepared from 0.75 g of seed koji, 600 g of steamed soybeans and 750 g of roasted wheat as described above. Then, 1.35 kg of the koji was placed in a 5.4 L glass container (diameter × height: 17 × 24 cm), 1.8 L of 23% (w/v) salt water containing 20% (w/v) of RSS cooled to 10 °C was added and thoroughly mixed into the koji (test pot B). The control was similarly prepared but with only 23% (w/v) salt water (test pot A). Immediately after brewing, the soy sauce koji was separated into two layers with the brewing brine, and thus it was stirred daily for two weeks to make a uniform mash.
When the mash temperature reached 12 °C on day 17 of incubation at 10 °C, a mixture of T. halophilus No.1, 3, 6, and 9 strains, which had been cultured in the basal medium amended with 2% (w/v) glucose, was added to each glass container at 1.0 × 106 CFU/g mash, and the mash temperature was gradually increased. Since a single strain would be lysed if bacteriophages were present in the mash (Uchida and Kambe, 1993), we used a mixture of four strains (No.1, 3, 6, and 9) isolated from different breweries. On day 20, ZR510 strain cultured in basal medium with 1% (w/v) glucose was added to the mash at a final concentration of 4.8 × 107 CFU/g mash, then the temperature of the mash was gradually raised to 28 °C. The mash was stirred daily until day 27, when fermentation began to be vigorous, held at 28 °C until day 53, then the temperature was lowered to 20 °C and maintained for the rest of the brewing. The mash was collected periodically and analyzed as described in the next section. After 523 d the mash was filtered, and 40 mL of the filtrate was placed in a 100 mL glass container with a lid and tightly capped, then heated in a water bath at 60 °C for 1 h, followed by 84 °C for 15 min. The resulting precipitate was collected by centrifugation (12 000 × g, 10 min), and the supernatant was filtered through a Whatman GD/X syringe filter with 0.2 µm pore size (Cytiva Japan, Tokyo) to complete the production of soy sauce-style seasonings (soy sauces A and B).
Brewing analyses The number of colony-forming units (CFUs) of Z. rouxii or T. halophilus was measured by modifying some of the media compositions in the Methods of Analysis in Health Science 2010 (Miyoshi and Watabe, 2010). PDA (Nissui Pharmaceutical Co., Ltd., Tokyo) containing 13% NaCl (w/v), 10% raw soy sauce (v/v) and 1.0% agar (w/v) was used for the yeast count, and MRS (de MAN, Rogosa and Sharpe) agar (Merck KGaA, Darmstadt, Germany) amended with 15% (w/v) NaCl and 0.5% (w/v) agar, and adjusted to pH 7.0 was used for the lactic acid bacteria count.
The apparent growth of each microorganism was measured at certain periods based on the absorbance at 600 nm using a spectrophotometer U-2001 (Hitachi High-Tech Co., Tokyo). The pH of the culture solution and mash was measured separately using a B-212 Twin pH meter (Horiba, Ltd., Kyoto).
Ethanol content in the culture solution and mash juice was measured using a gas chromatograph GC2014 (Shimadzu Co., Kyoto) equipped with a column packed with Thermo-1000 5% Sunpak-A (Shinwa Chemical Ind., Ltd., Kyoto) and isopropanol as an internal standard according to the standard method, as described by the equipment protocol.
Various sugar contents in those samples, except for d-sorbose, were examined as described previously (Miyoshi et al., 2019), using an UHPLC system desalted with a MICRO Acilyzer G1 (Asahi Kasei Co., Tokyo) equipped with an electrodialysis membrane AC-110-10 (ASTOM Co., Tokyo). The d-sorbose content of the desalted samples was calculated by quantifying L-iditol produced after reduction with sodium borohydride in the UHPLC system according to the method of Kimura and Tajima (1998), with scyllo-inositol as an internal standard, except that the sugar sample was dissolved in distilled water instead of 1 N ammonia water as done for the other sugars.
For the analyses of amino acid and organic acid contents in brewed soy sauce, the samples were first diluted appropriately with distilled water and filtered through an Amicon Ultra 10k ultrafiltration membrane (Merck KGaA). Amino acids were analyzed using a post-column HPLC method with ortho-phthalaldehyde as a reaction solution, using a Shimadzu Prominence HPLC system (Shimadzu Co., Kyoto) equipped with a Shim-pack AMINO-NA column (4.0 × 150 mm, Shimadzu Co.). Amino Acids Mixture Standard Solution, Type H (FUJIFILM Wako Pure Chemical Co.) was used as the amino acid standard. Organic acids were analyzed according to the manufacturer's protocoliii) using the post-column HPLC method and the above HPLC system equipped with an RSpak KC811 column (8.0 × 300 mm, Showa Denko K.K., Tokyo) × 2 with bromothymol blue as the reaction solution.
General components, total nitrogen, salt, ethanol, Brix, pH, and color were measured or assessed based on the Japanese Agricultural Standards (JAS) for soy sauceiv).
The taste components umami (AAE), saltiness (CT0), sourness (CA0), bitterness (C00), and astringency (AE1) of brewed soy sauce were measured as described by Tahara and Toko (2013) using the taste sensing system SA402B (Intelligent Sensor Technology, Inc., Kanagawa) equipped with various taste sensors (Intelligent Sensor Technology). For the taste analysis, soy sauces A and B were each diluted 20 times, and conductivity was assessed with a conductivity meter AS650 (AS ONE, Inc., Osaka) to ensure accurate analysis; 30 mM KCl solution and 0.3 mM tartaric acid solution were used as the reference solutions. In addition, soy sauce A′ with equal amounts of the various sugars (d-glucose, d-fructose, d-allulose, d-sorbose, and d-allose) except d-tagatose in brewed soy sauce B was also used for the analysis.
Effect of rare sugars on fermentation by microorganisms Figure 1 shows the effect of different proportions of the four rare sugars (d-allulose, d-sorbose, d-allose, and d-tagatose) in relation to glucose on the growth and ethanol production of the osmophilic yeast Z. rouxii ZR510. Ethanol productivity by the yeast decreased linearly with the increase in the ratio of each rare sugar to glucose. However, when d-allose or d-tagatose was used as the carbon source, a trace amount (0.02% w/v) of ethanol was detected even at 100% rare sugar. On the other hand, when d-allulose or d-sorbose was used as the carbon source, no ethanol was detected. Although the ethanol productivity among the rare sugars was comparable, the apparent growth of the culture in the medium mixed with d-allose was lower than that using the other three rare sugars at all mixing ratios.
Effect of rare sugars on growth and alcoholic fermentation of Zygosaccharomyces rouxii ZR510. ×, Absorbance at 600 nm; □, Ethanol
Ethanol production was measured using a gas chromatograph, and growth of the yeast was monitored by measuring absorbance at 600 nm.
Values are mean ± SD (standard deviation) of 3 samples.
Figure 2 shows the growth and pH change of T. halophilus No.9, and lactic acid and acetic acid production in the media with different proportions of individual rare sugars and glucose. In the presence of d-allose or d-sorbose, bacterial growth was significantly increased compared to controls. The percentages of each sugar were 40% (risk ratio of p < 0.05) and 60% (risk ratio of p < 0.01) for d-allose and 60% (risk ratio of p < 0.05) for d-sorbose, respectively. With d-tagatose as the carbon source, the apparent growth rate was the same as that with glucose alone, even at 100% d-tagatose, and the lactic acid level (7.6 mg/mL) was about 98% of that with glucose alone. The lactic acid bacteria multiplied and produced lactic acid (2.3–2.6 mg/mL) in the medium when the carbon source was solely d-allulose, d-allose, or d-sorbose, except for d-tagatose. On the other hand, the amount of acetic acid in each medium was almost constant at 0.7–0.8 mg/mL for all the media, except for the amount of acetic acid (1.0 mg/mL) in the medium with 100% d-sorbose.
Effect of rare sugars on growth and lactic acid fermentation of Tetragenococcus halophilus No.9. ×, Absorbance at 600 nm; ○, pH; □, Lactic acid; ■, Acetic acid.
Growth of the lactic acid bacteria was monitored by measureing absorbance at 600 nm.
Organic acid content and pH change were monitered as described in Materials and Methods. Each value is presented as mean ± SD of 3 samples. Values with asterisks differed significantly compared with 0% rare sugar solution in Student's t-test. * p < 0.05; **p < 0.01
Small-scale brewing test using RSS with fermentation control of Tetragenococcus sp. The number of salt-tolerant lactic acid bacteria and ethanol content in the mash during brewing in the small-scale test brew with or without RSS are shown in Figure 3. As a control for this small-scale test with RSS (test pot B), the starting brine did not contain RSS (test pot A). The number of salt-tolerant lactic acid bacteria, such as T. halophilus in the mash, decreased rapidly in both test pots A and B, due to the stirred and increased ethanol content after the addition of yeast, and the number of the respective strains in each pot decreased from the initial 1.1–1.2 × 106 CFU/g mash to less than 300 CFU/g mash after 50 d. The ethanol concentration in the mash gradually increased from the time the yeast was added to 2.8 g/100 g in the mash juice of test pot A and 1.7 g/100 g in test pot B on day 18. At the end of the brewing process, the ethanol concentration in test pots A and B was 2.8 g/100 g mash juice, confirming that almost the same amount of ethanol was produced in both test pots A and B. When the mash temperature was lowered from 28 to 20 °C on day 53, the ethanol concentration reached a constant value.
Effect of ethanol on growth of Tetragenococcus sp. in the early stage of soy sauce brewing.
●, Lactic acid bacteria in test pot A; ○, Lactic acid bacteria in test pot B; ▲, Ethanol content in test pot A; △, Ethanol content in test pot B.
Test pot A is a small brewing test without RSS; test pot B is the same brewing test with RSS. The dotted line indicates the brewing temperature. Arrows indicate the time of addition of lactic acid bacteria and yeast.
Changes in sugar levels in soy sauce mash during brewing The quantitative changes of various monosaccharides during brewing are shown in Figure 4. On day 17, the major monosaccharide in both test pots was glucose; 82.76 mg/mL mash juice in test pot A and 129.17 mg/mL mash juice in test pot B. Test pot B contained 20% (w/v) RSS in terms of solids, and the glucose content in the mash juice at the start of brewing was theoretically ca. 51 mg/mL. At this time, the difference in glucose content between test pots B and A was 46.41 mg/mL, almost the same as the theoretical value. Moreover, the glucose content decreased rapidly after the addition of yeast in both test pots, and at 26 days after the addition of yeast, the glucose content in the mash juice was 1.11 mg/mL in test pot A and 46.28 mg/mL in pot B. At this time, the glucose in the mash juice was reduced by 81.65 mg/mL in test pot A and 82.89 mg/mL in test pot B.
Changes in various sugars during soy sauce brewing.
○, Fructose; ●, Glucose; △, Allulose; ▲, Allose; □, Tagatose; █, Sorbose
Test pot A is a small brewing test without RSS; test pot B is the same brewing test with RSS. The method of analysis of sugar in mash juice is described in Materials and Methods.
In test pot A, no rare sugar was detected in the mash juice at the beginning of brewing, but 1.64 mg/mL of d-tagatose was detected on day 53, when the main fermentation by Z. rouxii ZR510 was complete, and remained at that level even after 523 days. On the other hand, the amount of d-allulose, d-allose, d-tagatose, and d-sorbose in the mash juice before fermentation on day 17 in test pot B was 5.50, 1.53, 1.10, and 6.19 mg/mL, respectively. At the end of fermentation, 33 d after the preparation of Z. rouxii ZR510, the amount of d-allulose, d-allose, d-tagatose, and d-sorbose in the mash had decreased to 89, 31, 60, and 83%, respectively. The tagatose content in test pot B containing RSS was 0.66 mg/mL on day 53, compared to 1.64 mg/mL in test pot A. After 523 days of brewing, d-allulose, d-allose, d-tagatose, and d-sorbose remained in the mash juice at 95, 34, 68, and 84% of their original concentrations, respectively.
Various components in soy sauce-style fermented seasoning with RSS Table 1 shows the JAS standard analysis results for soy sauces A and B after pressing and heating the mash in test pots A and B. The chemical composition of soy sauces A and B was comparable to that of commercial soy sauces, except for the color and Brix. The total nitrogen in soy sauce B was 94% of that in soy sauce A. The ethanol content in the soy sauce was above 3% (w/v) in each test pot, indicating superior fermentation compared to the commercial soy sauce. The Brix% of soy sauce B was 9% higher than that of soy sauce A.
| Sample | T-Na (%) | Salt (%) | Ethanol (%) | Brix | pH | Color |
|---|---|---|---|---|---|---|
| Soy sauce A | 1.59 | 16.48 | 3.10 | 35.3 | 4.9 | 15 |
| Soy sauce B | 1.49 | 16.04 | 3.03 | 44.3 | 4.8 | 6 |
| Commercial soy sauceb | 1.49 | 16.20 | 2.45 | 36.6 | 4.9 | 10 |
Soy sauce A: small-scale brewing test without RSS; soy sauce B: same test but with RSS, respectively.
The pH of both soy sauces A and B was comparable to that of the commercial product. Analyses of organic acids related to the pH value of each soy sauce showed that the content of citric acid, lactic acid, and acetic acid in soy sauce A was 7.29, 0.68, and 1.15 mg/mL, respectively, compared with 6.88, 0.40, and 1.03 mg/mL, respectively, in soy sauce B.
For the color value of soy sauce, the lower the number the darker the color, and remarkably, the color of soy sauce B was darker (lower color value) than soy sauce A.
Comparison of taste sensations using a taste recognition device Soy sauces A and B, as well as soy sauce A′, which contained the same amount of various sugars as B, were analyzed using a taste sensing system (SA402B), and the results are shown in Figure 5. The system showed a significant difference between soy sauces A and B in umami, acidic bitterness, and acidic bitter aftertaste. On the other hand, the acidic bitterness and acidic bitter aftertaste of soy sauce B were lower than those of soy sauce A. Soy sauce B had significantly less acidic bitterness and acidic bitter aftertaste than soy sauces A and A' (risk rate of p < 0.01). In our soy sauces, analysis of the basic amino acids that are thought to be related to bitterness showed that the content of arginine, histidine and lysine in soy sauce A was 4.98, 1.40 mg/mL, and 3.97 mg/mL, respectively, and 3.52, 0.77, and 2.71 mg/mL in soy sauce B; thus, the respective level of each in soy sauce B was 30, 45, and 32% lower than that in soy sauce A.
Taste analysis of prototype soy sauce using a taste recognition device.
□, Soy sauce B; ■, Soy sauce A; ▨, Soy sauce A′
Soy sauce A is from test pot A, which is a small brewing test without RSS; Soy sauce B is from test pot B, which is the same brewing test with RSS; and Soy sauce A′ contains the same amount of various sugars containing RSS added after it has already been brewed. Each value is presented as mean ± SD of 3 samples. Asterisks are significantly different by Student's t-test. * p < 0.05; **p < 0.01
The effects of rare sugars on fermentation-related microorganisms, especially their stability during the soy sauce brewing process and the flavor of soy sauce brewed with RSS were investigated for the potential application of RSS, which has health-promoting effects, in soy sauce brewing. First, we investigated the effects of four types of rare sugars (d-allulose, d-allose, d-tagatose, and d-sorbose) in RSS on microorganisms related to soy sauce brewing, i.e., salt-tolerant yeast (Z. rouxii) and salt-tolerant lactic acid bacteria (T. halophilus). The results in Figure 1 show that Z. rouxii ZR510 strain does not produce ethanol from any of the rare sugars. The production of a small amount of ethanol in d-allose and d-tagatose in the medium with 100% rare sugars as the carbon source was attributed to the fermentation of 1.70% (w/v) of reducing sugars, mainly glucose, in the raw soy sauce used as the culture medium. When d-allose was used as a carbon source, the apparent growth rate was lower than that of the other rare sugars, which may be related to the fact that d-allose is an aldose and the other rare sugars are ketoses, but we were not able to further examine this. On the other hand, no ethanol production was observed in d-allulose and d-sorbose under the same conditions, suggesting the presence of these rare sugars in large amounts might inhibit glucose consumption or fermentation in the culture medium. These results suggested that Z. rouxii ZR510 is unable to utilize the four rare sugars in RSS to the same extent as glucose for growth and ethanol production. In addition, in soy sauce brewing, mixing RSS with brewing brine was not considered to inhibit yeast fermentation by rare sugars.
Next, we examined the effect of each rare sugar on the growth of T. halophilus No.9 and the production of organic acids. As shown in Figure 2, when d-allulose, d-sorbose, d-allose, and d-tagatose were used as carbon sources, bacterial growth and lactic acid production were significantly lower in the medium with 100% of each rare sugar, except for d-tagatose. The apparent lactic acid production was lower in the culture in the presence of 80% d-tagatose compared to the other concentrations, but no significant difference was observed in the culture in the presence of 100% d-tagatose as a carbon source. Although some bacterial growth and a small amount of lactic acid production were observed in the medium even with 100% rare sugars (d-allulose, d-sorbose, or d-allose), this may be caused by the use of 1.70% (w/v) reducing sugars (mainly glucose) contained in the raw soy sauce used for the medium. AI-Baarri et al. (2018) reported that glucose and d-allulose equivalently supported the growth of the lactic acid bacteria Lactobacillus acidophilus and Streptococcus thermophiles. In our study, d-allulose did not have the same effect on T. halophilus No.9 as described above, and no significant increase in growth was found with any of the sugar ratios. Uchida (1982) reported that Tetragenococcus sp. show wide variation in sugar assimilation ability.
Since T. halophilus No.9 also used d-tagatose for growth and lactic acid fermentation in this experiment, the halophilic lactic acid bacteria contained in the soy sauce mash might use rare sugars other than d-tagatose. Therefore, in order for the rare sugars to remain in the soy sauce brewing process, it is necessary to inhibit the growth of lactic acid bacteria during brewing. Inamori et al. (1984) reported that Pediococcus (Tetragenococcus) halophilus interacts with Saccharomyces (Zygosaccharomyces) rouxii during the fermentation process and that growth of P. halophilus is inhibited by S. rouxii under aerobic conditions at an early stage of fermentation. For this experiment, we used several soy sauce lactic acid bacteria to confirm the growth inhibitory effect of the interaction between Z. rouxii and T. halophilus. We explored the possibility that growth of T. halophilus is inhibited by the interaction of the brewing microbes in a small-scale brewing test with RSS, and quantified the rare sugars during the brewing process and analyzed various characteristics of the resulting soy sauce-style seasoning. When RSS is used for brewing, the sugar content in the syrup may affect the brewing process. Specifically, RSS contains 44.3% glucose and 31.9% fructose (Hayashi et al., 2014), and these sugars may affect the enzymatic degradation and fermentation of soy sauce ingredients. In particular, enzymes, such as α-amylase and glucosidase, which digest the raw materials have been reported to be inhibited by glucose (Fujita et al., 1983; Xiao et al., 2004). Therefore, it was necessary to minimize the inhibition of enzymes and maximize the content of rare sugars as functional ingredients. On the other hand, Membrĕ et al. (1999) examined the effect of sugar concentration on the degree of growth of Z. rouxii and reported that increasing the sugar concentration from 300 to 800 g/L resulted in a linear reduction in the specific growth rate. Based on these reports, a soy sauce preparation test was conducted using a starting brine containing 20% (w/v) of RSS on a dry matter basis, following the method already described.
As shown in Figure 3, aerobic alcoholic fermentation by Z. rouxii ZR510, which was added in the early stage of brewing, gradually increased the ethanol concentration in the soy sauce mash of test pots A and B, and effectively inhibited the growth of T. halophilus. The ethanol tolerance of Z. rouxii was about 3% (Kusumegi, 2001) under 15% salinity, indicating that the maximum amount of ethanol was produced under the conditions of this experiment.
In Figure 4, there was no difference in the amount of glucose liberated from the soy sauce material in pots A and B despite the addition of RSS, suggesting that the addition of RSS did not affect the glycolytic enzymes from the koji during brewing. Kanbe et al. (1981) reported that the monosaccharides in soy sauce are not sufficiently digested from koji to koji juice immediately after brewing, and were released at their maximum level at 30 d after brewing by the various carbohydrate-degrading enzymes of koji mold. Our results showed the same trend. After completion of the main fermentation by Z. rouxii ZR510, equal amounts of glucose were consumed by the yeast in test pots A and B. Among the rare sugars in test pot B, d-allulose was highly stable during brewing, with 95% of the original amount remaining even 523 days after brewing. In a previous paper, we reported that d-allulose was retained at the highest level (86.0–88.5%) under the various cooking conditions, while the other rare sugars had obviously decreased (Miyoshi et al., 2019). In addition, d-allulose showed high stability not only in the cooking process but also in fermented food production in this previous study. Furthermore, the significant decrease of d-allose, the C3 epimer of glucose, among other rare sugars in mash test pot B indicated that d-allose is less stable during the brewing process.
Interestingly, tagatose was found to be produced only in test pot A, which did not contain RSS (Fig. 4). The tagatose produced after yeast fermentation in test pot A is thought to be due to the action of Z. rouxii ZR510 fermentation, not to the processing of raw materials (Levin, 2002) or the action of koji enzymes. This is the first report of the formation of tagatose in soy sauce mash. The remaining tagatose in the mash was thought to be due to the growth suppression of Tetragenococcus sp. during brewing.
In the degradation of soy sauce ingredients, as described in Figure 4, carbohydrate-degrading enzymes were not inhibited by the addition of RSS, but the total nitrogen content of soy sauce B (Table 1) was lower than that of soy sauce A. This experimental condition confirms that the use of RSS inhibits the proteolytic enzymes involved in the degradation of raw materials during the brewing process. Analysis of organic acid compositions related to the pH value in soy sauce (Table 1) revealed citric acid as the major organic acid in both soy sauces. Kanbe et al. (1978) reported that malic acid and citric acid in soy sauce mash are derived from soy sauce raw materials and there is a high negative correlation between lactic acid and citric acid produced by lactic acid fermentation. In this experiment, lactic acid fermentation by T. halophilus was suppressed by the growth of Z. rouxii under aerobic conditions in the initial stage of fermentation, resulting in almost no lactic acid production in each soy sauce, and citric acid from the raw materials remained as the major organic acid. The darker color is thought to be due to the higher sugar content available for the Maillard reaction (Motai, 1976). In particular, soy sauce B was highly reactive and developed its color due to the high fructose and glucose contents.
By using RSS as a soy sauce ingredient in brewing and fermentation processes, not only can the amount of glucose and fructose in the RSS be reduced during the processes, but flavor differences can be generated in the soy sauces. Flavor differences are not obtained in conventional products by adding RSS directly to the soy sauce as a sweetener. Therefore, we established different brewing conditions (soy sauce A, soy sauce B, and soy sauce A′) and analyzed the flavor of each soy sauce using a taste recognition device. As the results, we found that the flavor of soy sauce A′ was different from that of soy sauce B, as shown in Figure 5. The taste sensor detects increases or decreases in the membrane potential of lipid membranes due to electrostatic or hydrophobic interactions with various taste substances in the samples, and the membrane potential is higher when more tastants are adsorbed to the membrane (Tahara and Toko, 2013). In particular, for umami, the total nitrogen content of soy sauce B was 95% that of A in the JAS analysis (Table 1), suggesting that the low amino acid content in the soy sauce affected the lower membrane potential of umami. The results showing the decrease in bitterness of soy sauce B suggest that the use of RSS as a raw material for soy sauce can effectively impart differences in the taste during the brewing process. Inagaki et al. (2017) also reported that basic amino acids (lysine, histidine, and arginine) decreased with increasing Maillard reaction time during the production of fermented alcoholic beverages. In considering the causes of a bitter taste, Yoshida et al. (1966) reported that the threshold value for detecting bitterness for arginine, histidine, and lysine is 0.05, 0.02, and 0.05 g/dL, respectively, suggesting an effect on bitterness, as the differences in the respective amino acid contents of soy sauces A and B were higher than the taste threshold. The proportions of arginine and histidine in total amino acids of soy sauce were almost the same before and after heat treatment (data not shown). Therefore, the low content of basic amino acids in soy sauce B using RSS as a raw material was considered to be due to a decrease during the brewing process. As described above, soy sauce made with RSS was affected by enzymes from the koji mold and brewing microorganisms, and the Maillard reaction during the brewing process was shown to be different from that of soy sauce without RSS. In addition, d-allulose in the RSS used in the brewing process had a high residual rate, suggesting that it could be used as a seasoning for health promotion in the future. According to the statistical data from the Soy Sauce Information Centerv), the amount of soy sauce used per person per day was 15.3 mL in 2020. Since approximately 12 mg of rare sugars/mL (calculated amount based on the data in Fig. 4) is contained in the soy sauce prototypes in this experiment, about 184 mg in total would be ingested per person per day. Yamada et al. (2017) reported that the blood glucose response after consumption was reduced when a total of 0.8 g of rare sugars (allulose, sorbose, tagatose, and allose) derived from RSS was consumed in place of sugar. Therefore, the prototype soy sauce used in this experiment provides about a quarter of the amount of rare sugars for health promotion. However, fermentation changed the composition of the four rare sugars from that of the original RSS; thus, further research is needed to clarify the health functions and to stably retain rare sugars during the fermentation process. Since this research found that RSS improved the taste of soy sauce, we are currently testing its effects on the aroma and flavor characteristics of other fermented foods.
This is the first report on the use of rare sugars as a raw material for the fermentation of foods containing salt and the quantification of changes in the levels of rare sugars as they are converted by enzymes and microorganisms during fermentation. None of the rare sugars in the RSS (d-allulose, d-allose, d-tagatose, and d-sorbose) inhibited the growth of Z. rouxii ZR510 or T. halophilus No.9. In the small-scale brewing test when lactic acid fermentation was suppressed, we confirmed that tagatose was produced in the mash without RSS on day 53 of brewing, and 95% of the initial d-allulose, 34% of d-allose, 68% of d-tagatose, and 84% of d-sorbose remained at the end of brewing. According to JAS standard analysis, the soy sauce produced in the small-scale batch test using RSS as the raw material was found to be comparable to the commercially brewed, genuine soy sauce, except for the color and Brix. Furthermore, when the soy sauce was analyzed using a taste recognition device, it showed reduced bitterness and acidic aftertaste compared to the control soy sauce. Soy sauce brewing tests using RSS suggested that the syrup could be widely used not only for brewing soy sauce but also for fermented foods such as miso, fish sauce, and Japanese pickled vegetables, leading to the development of new functional seasonings.
Acknowledgements Thanks are due to Ms. Reiko Fujii, Ms. Maya Taketoshi, and Kagawa Prefectural Industrial Technology Center, Kagawa, Japan for their helpful collaboration.
Conflict of interest There are no conflicts of interest to declare.
