Evaluation of non-volatile taste components in commercial soy sauces

ABSTRACT The dominating non-volatile taste compounds in commercial brewed soy sauces were determined by HPLC and evaluated on the contributions to overall taste. Aspartic acid and glutamic acid accounted for 8.77 to 147.98 mg/mL in ten commercial soy sauces samples. Lactic acid (ranging from 0.83 to 13.19 mg/mL) and pyroglutamic acid (ranging from 0 to 12.80 mg/mL) were the dominant organic acids, contributing to the acidity and ensuring a balance in taste of soy sauces. 5ʹ-Inosine monophosphate was the most abundant nucleotide, followed by 5ʹ-guanosine monophosphate, and they accounted for 0.30 to 3.54 mg/mL in ten soy sauces. According to the determination of non-volatile taste compounds in soy sauce samples, taste activity value (TAV) and equivalent umami concentration (EUC) of different soy sauces were calculated and compared. An exclusive cluster analysis based on TAV was proposed to classify the commercial soy sauces. The EUC value of new class A is much higher than other classes.


Introduction
As a traditional seasoning, soy sauce is widely consumed across the globe and predominantly in Asian countries. Soy sauces can be roughly divided into Chinese type and Japanese type categories. [1] Chinese soy sauce has been popular for centuries due to its unique and pleasant flavour. [2] Soy sauce not only contributes to flavour but also enhances the taste of other ingredients to which it is added, especially the umami taste. [3,4] The flavor of soy sauce is generated from grain fermentation. [5] Brewed soy sauce is traditionally made from soybeans and may be fortified with wheat grains over several months after seeding with Aspergillus. Vegetable proteins can be degraded into peptides and amino acids that are readily utilised by the bacteria produced during fermentation. Throughout this time, bacteria can also synthesise amino acids from metabolic intermediates. So free amino acids are released during the fermentation process. [6] Most of the organic acids and nucleotides in the sauce generate at this time, while some originate in the raw material. [7] Free amino acids, organic acids, nucleotides, and sodium salt are believed to significantly contribute to the sensory characteristics of soy sauce. [8,9] Zhang et al. analysed taste compounds in Chinese solid fermented soy sauce and found that free amino acids, such as Glu, Asp, Arg, Leu, and Ala, were abundant and are recognised as important contributors to taste of the soy sauce. [4] Lioe et al. compared three typical forms of Japanese soy sauce (shoyu), koikuchi, tamari, and shiro shoyu, and concluded all of them contain sodium salt, free Glu, and several other free amino acids, especially sweet taste-eliciting amino acids. [10] The taste of soy sauce is probably a result of the interaction and balance among different taste components. The synergistic effect among Glu, 5ʹ-nucleotides and its sodium salt has increased the umami taste of Chinese solid fermented soy sauce. [11] Potential synergistic effect among free Glu, salt, and phenylalanine has been observed in some Chinese, Japanese, and Indonesian soy sauce. [10,11] Even though soy sauce is an essential seasoning in Chinese cuisine, researches on the relationship between classification and taste active compounds of Chinese soy sauce were limited. [12] Jiang et al. analysed nine organic acids in commercial soy sauces and proposed a classification method for soy sauce based on organic acids. [13] In this work, high performance liquid chromatography was used to determine the free amino acids, organic acids, and nucleotides in commercial soy sauce. The soy sauce samples were classified according to the established comprehensive composition characterisation of Chinese brewed soy sauce. The contributions of different components to the overall taste were evaluated. The new classification was more significance for the evaluation of soy sauce.

Soy sauce samples
All soy sauce samples are representative products of Chinese-style brewed soy sauces available in Chinese markets in 2016. Ten brands of soy sauce were randomly numbered 1 to 10. Based on the classification method in GB18186-2000, the samples were divided into three Special-grade soy sauce (samples 3, 5, and 9), three First-grade soy sauces (samples 1, 4, and 10), one Second-grade soy sauce (sample 7), and three Third-grade soy sauces (samples 2, 6, and 8). The origins and ingredients of each sample were listed in Table 1. Two typical fermentation methods were used in the processing of samples, which were high-salt solid-state fermentation and low-salt liquid-state fermentation. Samples of each type were obtained from a well-known area with high production, which aslo represented the most well-accepted products by consumers.
In a 50 mL volumetric flask, 2.5 mL soy sauce sample, 1 mL 100 mg/mL K 4 [Fe(CN) 6 ], and 1 mL 300 mg/mL ZnSO 4 were mixed, shaken vigorously, adjusted to the graduation line with ultra-pure water, and then left to stand for 30 min to participate. The clear supernatant extraction was filtered through a 0.22 µm syringe filter (Cleman, Beijing, China) twice and stored at 4°C before use. Analysis of the general components The total nitrogen content was determined by the Kjeldahl method, and the crude protein content was calculated by multiplying the total nitrogen content by 6.25. [14] The crude fat content was analysed by the Soxhlet extraction method using diethyl ether as the solvent. [15] Ash levels were determined from the weight after heating at 550°C for 4 h. The sodium content was determined through flame atomic emission spectrometry analysis. Moisture levels were measured per the change in weight after drying for 20 h via an electric incubator (WG9220A, Tianjin Tonglixinda Instrument Factory, Tianjin, China) and analytical balance (ME104, Mettler Toledo, Shang, China). The samples were heated to constant weight at 105°C, and then the weight of water in the samples was measured.

Free amino acid analysis
The analysis of free amino acids was performed using an AA analysis kit (Agela Technologies, Tianjin, China). Original soy sauce samples of 20 mL were centrifuged in 50 mL centrifuge tubes for 10 min at 4267 g (RCF) under 4°C. The supernatants were filtered through 0.22 µm syringe filters, and the volume was measured before diluted by 0.1 mol/L HCl. The total amino acid concentration after dilution was 1 to 2 mg/mL before mixing with internal standard solution (v:v, 10:1) for HPLC analysis. An Agilent 1260 HPLC system coupled with a DAD detector (Agilent Corp., Karlsruhe, Germany) was adopted for free amino acid analysis. The chromatography column (Durashell AA, 150 mm× 4.6 mm i.d., 3 µm) was from Agela Technologies (Tianjin, China). A gradient solvent system consisting of mobile phase A (4.50 mg/mL Na 2 HPO 4 and 4.75 mg/mL Na 2 B 4 O 7 , pH = 8.2) and mobile phase B (methanol: acetonitrile: water = 9: 9: 2, v/v/v) was utilised. A and B were mixed according to certain volumetric ratios, and the elution conditions were 6% to 10% B from 0 to 6 min, 10% B from 6 to 8 min, 10% to 16% B from 8 to 10 min, 16% to 40% B from 10 to 23 min, 40% to 50% B from 23 to 30 min, 100% B from 31 to 34 min, and 6% B from 35 to 38 min. The flow rate was 1.6 mL/min. The column temperature was 45°C. The amino acids were detected at 338 and 262 nm with injection volume 2 µL. Quantitative analysis of the amino acids was calibrated by the internal standard method, and calibration curves were generated by comparing the amino acid peak areas from the standard mixture solution (concentration of Cys-Cys was 0.014-0.341 mol/L, and other amino acids were in 0.027-0.682 mol/L) with the Nva and Sar peak area from the internal standard solution. The standard curves (five data points, n = 3) were all linear with R 2 values higher than 0.999.

Organic acid analysis
Organic acid analysis was performed on a Thermo U3000 UPLC system (Thermo Scientific, Waltham, Mass., USA) equipped with chromatography column (MP C18(2) 250 mm× 4.6 mm i.d., 5 µm) from Bonna-Agela Technologies Inc. (Tianjin, China) at 25°C. All organic acids were detected at 205 nm. The mobile phase was methanol-Buffer salt I (5:95, v/v) with a flow rate of 1.0 mL/min and the injection volume was 20 µL. Data were collected, processed, and analysed using Chromeleon software (Shimadzu Corp., Kyoto, Japan). The organic acids were quantified using external calibration curves. A mixed organic acid calibration standard solution was diluted into five gradients with ultra-pure water. The standard curves (five data points, n = 3) were all linear with R 2 values higher than 0.999.

Nucleotide analysis
Nucleotide analysis was carried out using the same UPLC system as the organic acid analysis [16,17] ; the nucleotides were detected at 254 nm. Four different nucleotides were separated by HPLC, including 5ʹ-AMP, 5ʹ-GMP, 5ʹ-IMP, and 5ʹ-CMP. The mobile phase was methanol-Buffer salt II (5:95, v:v) with a flow rate of 1.0 mL/min. The nucleotides were quantitated using external calibration curves. A mixed nucleotide calibration standard solution ranging from 5 to 100 μg/mL was prepared for this purpose. The standard curves (five data points, n = 3) were all linear with R 2 values higher than 0.999.

Statistical analysis
The SPSS v22.0 software (SPSS Inc., Chicago, IL, USA) was used for statistical analysis. Significant differences among the samples were calculated using one-way ANOVA followed by Duncan's multiple-range test at the 5% level (P < 0.05). [18] Cluster analysis was performed in SPSS v22.0 using taste activity value (TAV) of all taste compounds as variables to differentiate the ten types of commercial soy sauce samples. All samples were hierarchically clustered by betweengroups linkage, and a dendrogram was drawn automatically. The distance between samples was calculated as a squared Euclidean distance, which is the most commonly used unit of distance in cluster analyses.

Contents of general components in commercial soy sauce
The composition of Chinese commercial soy sauce is quite complex. The information of commercial soy sauces samples was listed in Table 1, including the grades, general component contents, origins, and ingredients formula. There were notable differences and similarities between different grades. In addition to moisture and salt, the major component of the commercial soy sauces was crude protein.
Among the four grades of soy sauces, the crude protein content was significantly higher in the Special grade, followed by the First grade, Second grade, and finally Third grade. The crude fat content in soy sauces was also higher in the Special grade and lower in the Third grade. The ash and NaCl content of the Third grade samples was higher than those in the Special grade or First grade, which was consistent with the classification methods used in previous studies. [19,20] The moisture content of the ten samples ranged from 62.27 to 74.54 g/100 g wet basis; the average moisture of Third-grade soy sauces was higher than that of other grades.

Absorbance of commercial soy sauces
Soy sauces contain some strong UV absorbents, such as tyrosine, phenylalanine, nucleotides, and other components which vary by different brands and grades. [21] The UV absorption spectra of different soy sauce samples markedly differed from each other in our study. The soy sauce samples were diluted 1000 times, and their absorbance measured under the maximum absorption wavelength was reported in Table 2. All soy sauces showed the largest absorption peak in the ultraviolet region, except sample 7 (Second-grade soy sauce). The absorbance of sample 6 (0.71), a Third-grade soy sauce, was significantly higher than that of other grades ranging from 0.27 to 0.46this might be explained by the fact that sample 6 is the only low-salt solid-state fermented soy sauce. The highest average absorbance was observed in Third-grade samples (0.44), followed by Special grade (0.43) and First grade (0.36).

Amino acid in commercial soy sauces
Free amino acids play a crucial role to the development of organisms [22] and enhance the taste of food. The 17 free amino acids in ten kinds of soy sauce were analysed by HPLC. The contents of free amino acid (mg/mL) in various commercial soy sauces were shown in Table 3.
Among the 17 amino acids, Asp and Glu are responsible for umami taste [23] ; Ser, Pro, Gly, Thr, and Ala create sweet tastes; Val, Met, Ile, Phe, Lys, Leu, Arg, His and Tyr taste bitter; Cys-Cys, which accounted for less than 0.49 mg/mL in samples, has no contribute to the taste. The content of umami-taste amino acids was much higher than sweet and bitter taste amino acids, ranging from 8.77 to 147.98 mg/mL. The data indicated that umami-taste amino acids were the dominant amino acids in 8 kinds of soy sauces. The amount of sweet-taste amino acid ranged from 3.43 to 12.19 mg/ mL. Bitter-taste amino acid contents varied from 5.72 to 18.99 mg/mL. As for different grades, the amount of umami-taste amino acid and total amino acid in 5 of 6 Special and First-grade soy sauces were much higher than those in lower grades, while the contents in sample 1 were not related to its quality and grades.

Organic acid in commercial soy sauce
Many previous studies have shown that organic acids in soy sauce primarily contain oxalic acid, tartaric acid, malic acid, lactic acid, acetic acid, pyroglutamic acid, and succinic acid. In our study, nine organic acids, including oxalic acid, tartaric acid, pyruvic acid, malic acid, lactic acid, acetic acid, pyroglutamic acid, citric acid, and succinic acid, were identified in the tested soy sauces. Table 4 showed the distribution of different organic acids in ten commercial soy sauces. Lactic acid was the dominant organic acid in sample 3, 6, 7, and 8. Similar result was reported by Li et al. in a previous research. [24] Pyroglutamic acid was the largest amount of organic acid in the three tested soy sauces, sample 5, 9, and 10. Succinic acid was less than 2.56 mg/mL in the sauces. The pyruvate and citric acid contents were too low to be detected. Pyruvate might be used as the beginning of the TCA cycle and as a raw material of free amino acid. [25] Acetylpropionic acid was not detected either, which was reported to be absence from brewed soy sauce, and was considered a judgement factor for Japanesestyle fermented soy sauce. [26] These organic acids influence the acidity of soy sauces and ensure a balance in taste. The amount of total organic acid were basically increased by the grades. However, there were still three samples (sample 1, 5, and 8) that did not follow the rule. Regardless the low concentration of organic acid, it is still one of the most important taste active substances in soy sauces.

Nucleotide in commercial soy sauces
To evaluate the taste active substances in soy sauce, four types of nucleotide in the standard solution and commercial soy sauce samples were determined by HPLC. Concentrations of 5ʹ-CMP, 5ʹ-GMP, 5ʹ-IMP, and 5ʹ-AMP in the soy sauce samples were listed in Table 5. The flavour nucleotides, 5ʹ-IMP nd: not detectable. Relative: ratio of the content of certain taste amino acids and total amino acid. The content of amino acid was mean ± standard deviation (n = 3). Different lowercase letters between columns represent significant differences (p < 0.05). nd: not detectable. The content of organic acid was mean ± standard deviation (n = 3). Different lowercase letters between columns represent significant differences (p < 0.05).
and 5ʹ-GMP, are responsible for umami taste. [27] Adding certain proportions of 5ʹ-IMP and 5ʹ-GMP in soy sauce can significantly improve the quality of the product by increasing of flavour intensity and enhancement of freshness. In ten samples, the amount of 5ʹ-IMP and 5ʹ-GMP occupied 0.30 to 3.54 mg/mL. The content of 5ʹ-IMP was the most abundant, followed by 5ʹ-GMP. The concentration of 5ʹ-IMP ranged from 0 to 2.85 mg/mL, while the concentration of 5ʹ-GMP ranged from 0.30 to 1.00 mg/mL. It was hard to find the relationship between grades and concentrations of these nucleotides.

TAV
TAV value of taste active compounds was calculated as the ratio of the concentration determined in the soy sauces and its threshold value, which was generally measured in water or in a simple matrix. [28][29][30] The taste threshold value (mg/100 ml) and taste of free amino acids in water were preformed according to revious studies. [31] The compounds with TAV greater than 1 were considered as 'active' in food taste. [32] Although some taste-active compounds were present in smaller amounts in the soy sauces, they have higher influence to the taste feeling due to their low threshold values. Thus, TAV value is a very useful index for evaluating these compounds and their actual contribution to taste. As shown in Table 6, the umami taste amino acid, Glu, had both high concentration and high TAV of more than 97.69 in Special-grade samples, followed by Asp, whose TAV was higher than 2.18 in the same grade. The sweet taste amino acids, Ser, Gly, and Ala also had high TAV of 1.37, 1.34, and 6.47, respectively, revealing their significant contribution to the sweet taste of soy sauces.
TAVs of six organic acid were calculated in the soy sauce samples [33] , and all were greater than 1. Among these organic acid TAVs, the value of tartaric acid was highest in nine samples, except sample 6. TAVs of GMP (0.92-6.19) and IMP (0-11.39) were the highest among nucleotides, indicating that the two nucleotides were the main taste active nucleotides in the sauce samples. Na + also had a high TAV, 31.78-43.44, suggesting a notable contribution to the taste of soy sauce. This was consistent with previous studies [34] demonstrating that inorganic ions like Na + , K + , PO 4 3-, and Cl − play an important role in strengthening the taste of food.

Equivalent umami concentration (EUC)
The equivalent umami concentration (EUC, g MSG/100 g) is the concentration of MSG equivalent to the umami intensity given by a mixture of MSG-like amino acids and 5ʹ-nucleotide. It is represented by the following equation [35] : where Y is the EUC of the mixture in terms of g MSG/100 g; a i is the concentration (g/100 g) of each umami amino acid (Asp or Glu); b i is the relative umami concentration (RUC) for each umami amino acid to MSG (1 for Glu and 0.077 for Asp); a j is the concentration (g/100 g) of each umami 5ʹnucleotide (5ʹ-IMP, 5ʹ-GMP, or 5ʹ-AMP); b j is the RUC for each umami 5ʹ-nucleotide to 5ʹ-IMP (1 for 5ʹ-IMP, 2.3 for 5ʹ-GMP and 0.18 for 5ʹ-AMP); and 1218 is a synergistic constant based on the concentration of g/100 g. [36] The high levels of Glu, 5ʹ-GMP, and 5ʹ-IMP led to significantly higher EUC values in our soy sauce samples. The EUC values of ten samples in order were 138.24, 46.08, 1238.20, 1881.18, 1223.49, 141.31, 238.11, 220.76, 2420.36, and 2929.82 (Sample 1 to 10, in that order). Among four grades of sauces, the EUC of sample 9 from Special grade and sample 10 from First grade reached 2420.36 and 2929.83 g MSG/100 gsignificantly higher than other grades of sauces (46.08-238.11 g MSG/100 g).

Cluster analysis of soy sauces
Soy sauce is typically classified based on amino acid nitrogen content, which doesn't take account of the taste feeling of soy sauces. The analysis also showed that the typical classification was not related to the content of taste components. Hence, our motivation in this study is to establish an exclusive method for the evaluation of commercial soy sauces based on taste active components, which would be a good complement to traditional grading method. Cluster analysis has been applied as a workable basis for the classification of soy sauce based on organic acids in a previous study. [13] In our study, more taste active compounds in addition to organic acids were observed with quantifiable taste activity. Cluster analysis was performed based on the TAVs of all taste active compounds tested in our sauce samples (Figure 1), resulting in three class of soy sauces: class A (samples 9 and 10), class B (sample 4), and class C (samples 1, 2, 3, 5, 6, 7, and 8). The EUC of class A was highest nd: not detected. a Taste threshold value (mg/100 mL). [30] (2420.36 and 2929.82), followed by class B and class C, and so the new classes seem to relate to EUC. These results may facilitate the analysis and discussion of soy sauces regarding to their respective taste, providing a new proposal for the classification of all types of soy sauces, but not restricted to the brewed Chinese-style soy sauces. [37] Conclusion Non-volatile taste-active compounds tested in this study, including amino acids, organic acids, and nucleotides, play an important role in the taste of soy sauce. Glu and Asp were the dominant amino acid in the sauce, contributing to the umami taste of soy sauce. Lactic acid and pyroglutamic acid were the predominant organic acid, contributing to the acidity of soy sauces. 5ʹ-IMP and 5ʹ-GMP contributed substantially to the umami taste of the sauce owning to the high levels of concentration. Besides, sodium also had an influence on overall taste by enhancing umami taste synergistically with amino acids and nucleotides. According to our results, soy sauces of higher grade tended to have a greater amount of taste active compounds, higher absorbance, and lower moisture. Commercial soy sauces were re-assigned to different classes using cluster assay based on TAV. This work provided an intensive evaluation of active compounds in soy sauce and might be helpful in understanding the formation of special taste in the brewed soy sauces.