Skip to main content
NCBI home page
Food Science and Biotechnology logoLink to Food Science and Biotechnology
. 2017 Aug 3;26(4):871–882. doi: 10.1007/s10068-017-0115-1

Change in profiles of volatile compounds from two types of Fagopyrum esculentum (buckwheat) soksungjang during fermentation

Min-Kyung Park 1, Hye-Sun Choi 2, Young-Suk Kim 1, In Hee Cho 3,✉
PMCID: PMC6049563  PMID: 30263615

Abstract

Fagopyrum esculentum (buckwheat) soksungjang is one of the traditional soybean pastes in Korea. This study profiled and compared volatile compounds between traditionally manufactured (TBS) and commercially modified buckwheat soksungjang (CBS) according to their fermentation periods. More volatile compounds were generated and non-uniform increases or decreases in volatiles were more common during TBS fermentation. In addition, the changes in and differences between the volatiles from TBS and CBS during the fermentation process (after 0, 1, 2, and 5 weeks) were investigated in partial least squares-discriminant analysis models. The changes were accelerated during CBS fermentation in comparison with TBS fermentation. Several major volatile compounds, such as methyl decanoate, 3-hydroxy-2,6-dimethylpyran-4-one, and methyl heptanoate were found in the final stage of fermentation in TBS, in contrary, tridecane, (Z)-hex-3-en-1-ol, furan-2-carbaldehyde, and ethyl tetradecanoate were contributed to the latest of fermentation in CBS.

Keywords: Buckwheat soksungjang, Volatile profile, Fermentation, Novel starter, Partial least squares-discriminant analysis

Introduction

Various fermented soybean products (e.g., Korean doenjang and gochujang, Japanese miso and natto, and Chinese douchi) have been enjoyed for more than thousands of years in Eastern Asia. Their consumption has recently been increasing due not only to their own characteristic flavors but also their various nutritional (high protein) and functional benefits, such as anti-oxidant [1, 2], fibrinolytic [3, 4], and anti-cancer [5, 6] effects. In particular, doenjang is the oldest and the most popular Korean fermented soybean product. Traditional doenjang is made with meju, which is prepared by soaking, streaming, and crushing soybeans, forming them into a rectangular block, and then hanging this up with rice straw for 1–3 months to facilitate the growth of wild microorganisms (e.g., Bacillus, Lactobacillus, Aspergillus species, and yeasts) [7, 8]. Meju is then brined and ripened for another 2–3 months. After fermentation, it is separated into two parts (e.g., liquid and solid residue) and the solids are additionally ripened for more than 2-month (doenjang).

Soksungjang (literally, ‘soksung’ of soksungjang in Korean means ‘quick’ in English) is similar to doenjang, except that it is made by mixing, soaking, steaming, molding, and fermenting soybeans together with grains (e.g., barley, rice, wheat, and buckwheat). It requires a shorter (e.g., 5–6 weeks) fermentation period than most fermented soybean products due to this, which generally require 6–12 months of fermentation. Furthermore, brining and ripening processes are applied, but not the separating process. In this study, Fagopyrum esculentum (buckwheat) soksungjang (BS) was prepared. Buckwheat is a rich source of starch, protein, dietary fiber, vitamin B, and minerals [9, 10]. In particular, its beneficial effects are mainly due to its abundance of flavonoids (especially, rutin) [11]. Rutin is the glycoside between the flavonol quercetin and the disaccharide rutinose (α-l-rhamnopyranosyl(1 → 6)-β-d-glucopyranose) that exhibits anti-oxidative, anti-inflammatory, and anti-carcinogenic effects, and can also reduce the fragility of blood vessels related to hemorrhagic disease and hypertension in humans [12].

In general, fermented soybean products have strong and characteristic flavors which can be caused by generating specific volatile compounds, such as butanoic acid (sweaty), 3-methylbutanoic acid (rancid, sweaty), 2-pentylfuran (beany), methyl 2-methyl- butanoate (fruity), 2- and 3-methylbutanal (malty), and 2,3-dimethylpyrazine (nutty). Therefore, it is important to profile volatile compounds of fermented soybean products in order to predict their qualities [13, 14]. In addition, during fermentation, the volatile compounds were changed according to a starter which is a predominant microorganism due to its type and activity of enzymes. In this study, there are two types of BSs: (1) traditionally manufactured buckwheat soksungjang (TBS) is non-uniformly fermented by natural microorganisms, and thus unequal qualities can be produced during fermentation, while (2) commercial modified buckwheat soksungjang (CBS) is generally made by inoculating certain microorganisms (e.g., Bacillus, Rhizopus, Mucor, and Aspergillus species) that known to be the dominant microflora of fermented soybean products [15, 16]. Microorganisms in soksungjang degrade macromolecules (e.g., proteins, sugars, and fatty acids) in soybeans with various enzymes and provide a substrate pool for further biological and chemical reactions, which can produce the characteristic tastes and odors of fermented foods [17]. Aspergllus oryzae (PS03) and Bacillus amyloliquefaciens (RD7-7) previously isolated from meju were applied for CBS as the novel multi-starter in this study.

The aims of this study were (1) to profile and compare volatile compounds between two BSs, since the microorganisms involved in fermentation could contribute to the formation of unique flavor characteristics, and (2) to observe their changes and differences according to fermentation periods by multivariate statistical analysis.

Materials and methods

Sample preparation

BS samples were prepared by two different types of method, traditional and commercial modified methods. The procedures for manufacturing BS are as follows; soybeans were soaked in water and steamed at 121 °C for 40 min and mixed with buckwheat at 7:3 ratio (w/w). In CBS, the novel multi-starter (Aspergillus oryzae PS03 : Bacillus amyloliquefaciens RD7-7 = 1:1) was inoculated, and then mixtures were formed in block shape (500 g). On the other hands, traditional BS (TBS) was only exposed to the natural strains. Their surfaces were dried at 25 ± 5 °C for 2 h using a drying oven (DS-80-2; Dasol Scientific Co., Ltd., Hwaseong-si, Korea), then put them at 28–30 °C for 2 weeks. The fermented soybeans were called, BS meju. Finally, BS samples were prepared by crushing dried BS meju submerged in 15.5% sea salt solution (1.0 kg dry BS meju plus 1.4 kg sea salt solution) at 20–25 °C during 5 weeks. Samples were collected at the 0, 1, 2, and 5 weeks of fermentation to analyze their volatile compounds.

Extraction of volatile compounds in BS

Solid phase micro-extraction (SPME) with carboxen/polydimethylsiloxane fiber (CAR/PDMS, 75 μm, Supelco, Bellefonte, PA, USA) was used to extract volatile compounds in BS. BS sample (5.0 g) was added into a 20 mL vial (headspace screw top clear vial, Agilent technologies, Santa Clara, CA, USA) and sealed with a cap (Ultraclean 18 mm screw-cap w/septa, Agilent Technologies). The vial was kept at 60 °C with agitating at 250 rpm for 30 min to obtain an equilibrium state. SPME fiber (22 mm) was then exposed into the headspace of the vial to adsorb volatile compounds for 30 min. Then it was desorbed in the GC injection port (230 °C, splitless mode) for 5 min. All experiments were conducted in triplicate.

Gas chromatography–mass spectrometry (GC–MS) analysis

GC–MS analysis was performed using a 7890B series gas chromatograph connected to a 5977A mass selective detector (MSD) (Agilent Technologies) and multi-purpose sampler MPS 2 (Gerstel, Mülheim an der Ruhr, Germany) were used in the present study. DB-FFAP capillary column (30 m length × 0.25 mm i.d. × 0.25 µm film thickness, J&W Scientific, Folsom, CA, USA) was equipped and the oven temperature was initially held at 40 °C (6 min) and followed by ramping to 200 °C (10 min) at a rate of 3 °C/min. The other GC–MS conditions were as follows: Both front inlet and detector transfer line temperatures were 230 °C. Helium, as a carrier gas, was used at a constant flow rate of 0.8 mL/min. Mass scanning range were 35–550 m/z and mass spectra were obtained at 70 eV in the electron ionization (EI) mode.

Identification and quantification

Volatile compounds were identified by comparing their mass spectral data with those of the National Institute of Standards and Technology (NIST) version 08 mass spectral database. Also retention index (RI) of each compound were determined with n-paraffins from C7 to C22 as external standards. The relative level of each volatile compound was performed by comparing their peak areas to that of the internal standard compound (300 μl of 200 ppm methyl cinnamate in methanol) on the GC–MS total ion chromatogram (TIC).

Statistical analysis

Analysis of variance (ANOVA) using the Statistical Package for the Social Sciences (SPSS, version 12.0, Chicago, IL, USA) program was conducted to evaluate statistical change/or difference of the volatile compounds in each BSs during fermentation, respectively. The result of Duncan’s multi-range test presented the significant different level at p < 0.05. The values of volatile compounds were presented by average ± standard deviation of the three repetitive results. Partial least squares-discriminant analysis (PLS-DA) was conducted to the raw values (n = 3) of the relative peak areas obtained by GC–MS using SIMCA-P (version 11.0, Umetrics, Umeå, Sweden).

Results and Discussion

Volatile profiles from two BSs during the fermentation

The volatiles in the two BSs (TBS and CBS) were extracted using SPME and then analyzed by GC–MS. The volatile compounds identified in TBS and CBS according to their fermentation periods, their relative peak areas, and RIs on the DB-FFAP column were listed in Tables 1 and 2. A total of 91 and 80 volatile compounds were identified in TBS and CBS during fermentation, respectively. The volatile compounds of TBS encompassed 3 acids, 14 alcohols, 13 benzenes, 5 carbonyls, 32 esters, 4 furans, 6 hydrocarbons, 6 phenols, 2 pyrazines, 2 sulfur-containing compounds, and 4 miscellaneous compounds. On the other hand, volatiles of CBS were composed of 3 acids, 10 alcohols, 13 benzenes, 4 carbonyls, 28 esters, 3 furans, 6 hydrocarbons, 4 phenols, 3 pyrazines, 1 sulfur-containing compound, and 4 miscellaneous compounds. During the fermentation process of soybean products, various enzymes (predominantly protease, lipase, and amylase) can generate free amino acids, organic acids, mono- or di-saccharides, and fatty acids, which can then produce characteristic flavor compounds [18]. In particular, volatile organic acids, alcohols, and esters have been considered important contributors to the unique flavor properties of fermented soybean products [19]. As indicated in Tables 1 and 2, these were also the predominant volatile compounds in the BSs in this study.

Table 1.

The volatile compounds identified in TBS according to fermentation periods

No. Possible compound RI1 Relative Peak Area2 ID3
0w 1w 2w 5w
Acids
v1 Acetic acid 1455 1.519 ± 0.347a4 2.028 ± 0.340a 1.247 ± 0.180a 4.091 ± 0.892 B
v2 3-Methylbutanoic acid 1672 0.153 ± 0.018b 0.332 ± 0.088c ND5a NDa B
v3 Hexanoic acid 1851 0.028 ± 0.003b NDa NDa NDa B
Alcohols
v4 Ethanol 935 NDa 5.650 ± 0.741a 21.437 ± 5.471b 149.245 ± 13.491c B
v5 4-Methoxybutan-1-ol 936 1.400 ± 0.076b NDa NDa 1.415 ± 0.312b B
v6 2-Methylpropan-1-ol 1095 0.151 ± 0.030b 0.561 ± 0.110c 0.841 ± 0.096d NDa B
v7 Butan-1-ol 1102 NDa NDa NDa 0.480 ± 0.130b B
v8 3-Methylbutan-1-ol 1206 3.589 ± 0.654a 8.273 ± 2.247a 8.027 ± 1.040a 22.876 ± 7.077b B
v9 3-Methylbut-3-en-1-ol 1248 0.136 ± 0.014b 0.316 ± 0.092c 0.091 ± 0.027a NDa B
v10 2-Methylbut-2-en-1-ol 1321 0.263 ± 0.056b 0.842 ± 0.198c 0.328 ± 0.056b NDa B
v11 Hexan-1-ol 1354 0.073 ± 0.015a 0.300 ± 0.050b 0.244 ± 0.045b 0.262 ± 0.074b B
v13 Oct-1-en-3-ol 1450 0.647 ± 0.171a 8.956 ± 1.412d 4.719 ± 0.722b 7.284 ± 0.554c B
v15 Butane-2,3-diol 1542 1.529 ± 0.291a 1.232 ± 0.284a 1.571 ± 0.236a 2.690 ± 0.212b B
v16 Butane-1,3-diol 1580 1.729 ± 0.196ab 2.498 ± 0.521b 1.143 ± 0.157a 2.070 ± 0.267ab B
v17 4-Methyl-1-propan-2-ylcyclohex-3-en-1-ol
(4-Terpineol)		 1604 0.063 ± 0.012a 0.709 ± 0.141c 0.450 ± 0.112b NDa B
v18 3-Hydroxy-2,6-dimethylpyran-4-one (Methyl maltol) 1970 0.071 ± 0.012a 0.846 ± 0.074c 0.591 ± 0.177b 0.672 ± 0.090b B
v19 3-Hydroxy-2-methylpyran-4-one (Maltol) 1976 0.045 ± 0.010a NDa 0.308 ± 0.087b 0.725 ± 0.218c B
Benzens and benzene derivatives
v20 1,3-Xylene 1104 NDa NDa NDa 0.414 ± 0.085b B
v22 Benzaldehyde 1532 0.976 ± 0.127a 3.242 ± 0.406c 1.915 ± 0.153b 7.564 ± 0.725d B
v24 Methyl benzoate 1630 1.285 ± 0.361a 11.987 ± 1.590b 8.945 ± 2.180b 16.327 ± 2.351c B
v25 2-Phenylacetaldehyde 1654 0.855 ± 0.222a 9.236 ± 1.689c 6.464 ± 0.805b 13.536 ± 2.081d B
v26 2-Hydroxy-benzaldehyde 1688 0.089 ± 0.011a 0.842 ± 0.088b 0.606 ± 0.133b 1.340 ± 0.197c B
v27 1,2-Dmethoxybenzene 1734 0.037 ± 0.009a 0.465 ± 0.082bc 0.310 ± 0.043b 0.562 ± 0.144c B
28 Methyl 2-phenylacetate 1766 0.107 ± 0.031a 2.377 ± 0.563b 1.737 ± 0.348b 3.411 ± 0.861c B
v29 2-Phenylprop-2-enal 1810 0.172 ± 0.040a 3.655 ± 0.726bc 2.648 ± 0.659b 4.063 ± 0.764c B
v30 Methyl 3-phenylpropanoate 1854 0.046 ± 0.010b NDa NDa NDa B
v31 Phenylmethanol 1884 0.197 ± 0.054a 2.517 ± 0.366c 1.707 ± 0.166b 2.912 ± 0.084d B
v32 2-Phenylethanol 1915 0.366 ± 0.097a 7.096 ± 1.460b 8.040 ± 0.989b 25.367 ± 3.971c B
v34 2-Phenylbut-2-enal 1939 NDa 0.102 ± 0.011a 0.452 ± 0.080a 1.894 ± 0.476b B
v35 Benzoic acid >2000 0.059 ± 0.011a 9.868 ± 0.211c 0.126 ± 0.024a 1.087 ± 0.778b A
Carbonyls
v37 2-Methylbutanal 915 0.523 ± 0.038b 1.452 ± 0.302c 0.587 ± 0.096b NDa B
v38 3-Methylbutanal 919 4.244 ± 0.419b 4.992 ± 1.296b 1.543 ± 0.297a 3.604 ± 1.165b B
v39 Octan-3-one 1252 NDa 0.668 ± 0.147c 0.200 ± 0.028b NDa B
v40 3-Hydroxybutan-2-one 1288 0.115 ± 0.013b NDa NDa NDa B
v41 6-Methylhept-5-en-2-one 1337 NDa 0.419 ± 0.022c 0.279 ± 0.103b NDa B
Esters
v42 Methyl acetate <900 8.202 ± 1.219a 17.967 ± 3.742c 8.083 ± 1.418a 13.084 ± 2.259b A
v43 Ethyl acetate <900 NDa NDa 0.878 ± 0.090a 4.905 ± 1.764b A
v44 Methyl 2-methylpropanoate 922 NDa NDa 0.532 ± 0.071b NDa B
v45 Methyl butanoate 989 0.243 ± 0.028b 0.655 ± 0.169d 0.463 ± 0.058c NDa B
v46 Dimethyl carbonate 999 0.144 ± 0.011b NDa NDa NDa B
v47 Methyl 2-methylbutanoate 1009 NDa NDa 0.938 ± 0.150b NDa B
v50 Methyl hexanoate 1186 0.284 ± 0.061a 3.467 ± 0.160d 1.619 ± 0.216b 2.920 ± 0.229c B
v53 Ethyl hexanoate 1286 NDa NDa NDa 0.974 ± 0.137b B
v54 Methyl heptanoate 1390 NDa 0.845 ± 0.151c 0.407 ± 0.044b 0.468 ± 0.114b B
v55 Methyl octanoate 1431 0.257 ± 0.050a 3.468 ± 0.425d 2.016 ± 0.324b 2.719 ± 0.422c B
v56 Methyl oct-4-enoate 1434 NDa 0.768 ± 0.124c 0.353 ± 0.048b NDa B
v57 Ethyl octanoate 1492 NDa NDa 0.190 ± 0.046ab 0.643 ± 0.186b B
v58 Methyl nonanoate 1533 0.098 ± 0.011a 1.146 ± 0.181c 0.481 ± 0.086b NDa B
v59 Methyl decanoate 1630 0.063 ± 0.015a 1.620 ± 0.177c 1.095 ± 0.256b 0.870 ± 0.120b B
v60 Methyl pyridine-3-carboxylate 1767 NDa NDa NDa 0.153 ± 0.033b B
v61 Ethyl 2-phenylacetate 1804 NDa NDa 0.270 ± 0.058b 0.843 ± 0.202c B
v62 Methyl dodecanoate 1811 0.093 ± 0.021a 1.279 ± 0.196b 1.188 ± 0.377b NDa B
v63 Ethyl dodecanoate >2000 NDa NDa NDa 0.375 ± 0.049b A
v64 Methyl tetradecanoate >2000 0.386 ± 0.098a 3.960 ± 0.804b 3.780 ± 1.663b 3.614 ± 1.165b A
v65 Ethyl tetradecanoate >2000 NDa NDa 0.152 ± 0.077a 1.135 ± 0.422b A
v66 Methyl 12-methyltetradecanoate >2000 0.075 ± 0.021a 0.434 ± 0.078b 0.210 ± 0.100a 0.435 ± 0.105b A
v67 Methyl pentadecanoate >2000 0.129 ± 0.038a 1.358 ± 0.124b 1.176 ± 0.578b NDa A
v68 Methyl 14-methylpentadecanoate >2000 0.015 ± 0.013b NDa NDa NDa A
v69 Methyl hexadecanoate >2000 6.453 ± 1.844a 80.933 ± 12.665c 65.240 ± 6.335b NDa A
v70 Methyl hexadec-7-enoate >2000 0.089 ± 0.023a 1.356 ± 0.407a 2.044 ± 0.350a 19.767 ± 5.816b A
v71 Methyl (Z)-hexadec-9-enoate >2000 0.093 ± 0.018a 0.692 ± 0.169b 1.199 ± 0.246c 0.418 ± 0.075b A
v72 Ethyl hexadecanoate >2000 0.048 ± 0.012a 0.576 ± 0.175b 2.874 ± 0.480c NDa A
v73 Methyl (7E,10E)-hexadeca-7,10-dienoate >2000 NDa NDa 0.361 ± 0.082b NDa A
v75 Methyl octadecanoate >2000 0.156 ± 0.053a 3.418 ± 0.900b 3.450 ± 0.926b 1.240 ± 0.190a A
v76 Methyl octadec-9-enoate >2000 1.846 ± 0.505a 21.875 ± 6.452b 19.762 ± 4.565b 0.497 ± 0.089a A
v77 Methyl (9Z,12Z)-octadeca-9,12-dienoate >2000 2.040 ± 0.638a 18.006 ± 4.342b 1.986 ± 0.201a 3.235 ± 0.275a A
v79 Ethyl octadec-9-enoate >2000 NDa NDa 0.456 ± 0.170b 0.166 ± 0.057a A
Furans and furan derivatives
v80 Furan-2-carbaldehyde 1472 NDa NDa 0.092 ± 0.008a 0.342 ± 0.099b B
v81 Oxolan-2-one 1641 0.045 ± 0.007b NDa NDa NDa B
v82 Furan-2-ylmethanol 1667 0.032 ± 0.009b 0.189 ± 0.026c 0.222 ± 0.018d NDa B
v83 4-Methyl-2H-furan-5-one 1722 NDa NDa NDa 0.748 ± 0.108b B
Hydrocarbons
v85 2-Methylnonane 956 NDa 0.817 ± 0.055b NDa NDa B
v87 Decane 1000 0.522 ± 0.099a 6.982 ± 0.528b 4.072 ± 0.864c NDa B
v88 Dodecane 1200 NDa NDa 0.184 ± 0.035b NDa B
v89 Tridecane 1300 NDa NDa 0.428 ± 0.083a 21.686 ± 4.288b B
v90 Tetradecane 1400 NDa NDa 0.172 ± 0.046b NDa B
v92 Heptadecane 1700 0.039 ± 0.008b NDa NDa NDa B
Phenols
v93 2-Methoxyphenol 1868 0.028 ± 0.005a 0.502 ± 0.086b 0.524 ± 0.026b 0.464 ± 0.026b B
v94 4-Ethyl-2-methoxyphenol >2000 NDa NDa NDa 0.568 ± 0.196b A
v95 2-Methoxy-4-prop-2-enylphenol >2000 NDa 0.123 ± 0.019c 0.089 ± 0.020b NDa A
v96 3-Ethylphenol >2000 NDa 0.077 ± 0.015a 0.046 ± 0.004a 58.997 ± 3.688b A
v97 4-Ethenyl-2-methoxyphenol >2000 0.053 ± 0.007a 1.135 ± 0.099b 1.535 ± 0.100c 2.505 ± 0.246d A
v98 4-Ethenylphenol >2000 0.052 ± 0.007a 0.497 ± 0.078a 0.386 ± 0.039a 16.190 ± 7.193b A
Pyrazines
v99 2,6-Dimethylpyrazine 1329 NDa 0.477 ± 0.127b 0.339 ± 0.094b NDa B
v101 2,3,5,6-Tetramethylpyrazine 1475 0.113 ± 0.015a 0.267 ± 0.014b 0.133 ± 0.015a 0.375 ± 0.054c B
Sulfur-containing compounds
v102 (Methyltrisulfanyl)methane (Dimethyl trisulfide) 1380 0.031 ± 0.011a NDa 0.108 ± 0.041b NDa B
v103 3-Methylsulfanyl-propan-1-ol (Methionol) 1722 NDa NDa 0.164 ± 0.020a 1.890 ± 0.245b B
Miscellaneous compounds
v104 Styrene 1208 NDa NDa NDa 4.448 ± 1.060b B
v106 Alpha-cubebene 1435 NDa NDa NDa 0.267 ± 0.047b B
v107 Alpha-muurolene 1723 0.110 ± 0.022a 1.765 ± 0.220c 0.112 ± 0.025a 0.388 ± 0.045b B
v108 Delta-cadinene 1755 0.153 ± 0.031a 2.194 ± 0.189b 1.537 ± 0.402b 1.806 ± 0.592b B
Open in a new tab

1Retention indices were determined using n-paraffins C7–C22 as external standards

2Average of contents compared to that of the internal standard ± standard deviation

3Tentative identification was performed as follow: A, mass spectrum was consistent with that of Wiley mass spectrum database; B, mass spectrum and retention index were consistent with those of the mass spectrum database and literatures

4There are significant differences (p < 0.05) among samples collected according to fermentation period by using Duncan’s multiple comparison test between the samples having different letter in low

5 Not detected

Table 2.

The volatile compounds identified in CBS according to fermentation periods

No. Possible compound RI1 Relative Peak Area2 ID3
0w 1w 2w 5w
Acids
v1 Acetic acid 1454 1.467 ± 0.735a4 1.609 ± 0.484a 1.737 ± 0.156a 1.269 ± 0.063a
v2 3-Methylbutanoic acid 1672 0.133 ± 0.026a 0.193 ± 0.024a 0.730 ± 0.096c 0.380 ± 0.023b
v3 Hexanoic acid 1849 ND5a NDa 0.073 ± 0.015b NDa
Alcohols
v5 4-Methoxybutan-1-ol 937 3.438 ± 0.445b NDa NDa NDa
v8 3-Methylbutan-1-ol 1207 1.966 ± 0.568b 0.472 ± 0.169a 2.526 ± 0.390b 0.716 ± 0.077a
v11 Hexan-1-ol 1353 NDa 0.046 ± 0.007b 0.184 ± 0.037c 0.072 ± 0.019b
v12 (Z)-Hex-3-en-1-ol 1383 NDa NDa NDa 0.067 ± 0.007b
v13 Oct-1-en-3-ol 1451 0.106 ± 0.002a 0.067 ± 0.016a 1.873 ± 0.277c 1.000 ± 0.131b
v14 2-Ethylhexan-1-ol 1492 0.018 ± 0.003a 0.230 ± 0.037b NDa NDa
v15 Butane-2,3-diol 1544 4.424 ± 1.169c 0.111 ± 0.019a 2.805 ± 0.728b 0.614 ± 0.069a
v16 Butane-1,3-diol 1581 0.736 ± 0.174a 3.158 ± 0.720b 0.839 ± 0.209a 0.149 ± 0.011a
v17 4-Methyl-1-propan-2-ylcyclohex-3-en-1-ol
(4-Terpineol)		 1603 NDa NDa 0.230 ± 0.012c 0.099 ± 0.001b
v19 3-Hydroxy-2-methylpyran-4-one (Maltol) 1972 0.034 ± 0.003a NDa 0.439 ± 0.182b 0.167 ± 0.008a
Benzens and benzene derivatives
v21 1,4-Xylene 1138 NDa 1.138 ± 0.527b 0.180 ± 0.030a NDa
v22 Benzaldehyde 1532 0.706 ± 0.095a 2.565 ± 0.363b 3.532 ± 0.340c 2.569 ± 0.348b
v23 Benzonitrile 1613 NDa NDa 0.075 ± 0.017c 0.054 ± 0.009b
v24 Methyl benzoate 1629 0.289 ± 0.023a 0.029 ± 0.026a 7.844 ± 2.147c 3.229 ± 0.338b
v25 2-Phenylacetaldehyde 1653 0.330 ± 0.041a NDa 7.435 ± 1.010c 5.072 ± 0.574b
v26 2-Hydroxybenzaldehyde 1679 0.067 ± 0.005a NDa 1.062 ± 0.253c 0.672 ± 0.013b
v27 1,2-Dimethoxybenzene 1757 NDa 0.065 ± 0.008c NDa 0.055 ± 0.001b
v28 Methyl 2-phenylacetate 1768 0.038 ± 0.001a 0.065 ± 0.006a 1.550 ± 0.349c 0.456 ± 0.035b
v29 2-Phenylprop-2-enal 1812 0.045 ± 0.004a 0.028 ± 0.006a 1.194 ± 0.246c 0.860 ± 0.059b
v30 Methyl 3-phenylpropanoate 1853 0.008 ± 0.013a 0.625 ± 0.153b 0.176 ± 0.032c 0.061 ± 0.011ab
v31 Phenylmethanol 1886 0.025 ± 0.001a 0.116 ± 0.003a 0.550 ± 0.140b 0.293 ± 0.017b
v32 2-Phenylethanol 1920 0.106 ± 0.009a NDa 1.769 ± 0.312c 1.356 ± 0.065b
v35 Benzoic acid >2000 0.040 ± 0.008b NDa 0.068 ± 0.060c 0.113 ± 0.005c
Carbonyls
v36 Butanal <900 NDa 0.075 ± 0.032b 0.159 ± 0.033c NDa
v37 2-Methylbutanal 919 0.399 ± 0.078ab 0.559 ± 0.128b 1.037 ± 0.231c 0.179 ± 0.006a
v38 3-Methylbutanal 921 3.475 ± 0.597c 1.674 ± 0.690ab 2.557 ± 0.248b 0.920 ± 0.061a
v40 3-Hydroxybutan-2-one 1288 0.171 ± 0.042c 0.077 ± 0.009b NDa 0.088 ± 0.005b
Esters
v42 Methyl acetate <900 4.944 ± 0.506a 7.113 ± 0.886a 16.428 ± 2.521b 6.374 ± 0.043a
v45 Methyl butanoate 988 NDa 0.138 ± 0.032b 0.294 ± 0.013c 0.146 ± 0.008b
v46 Dimethyl carbonate 999 0.180 ± 0.048b NDa NDa NDa
v47 Methyl 2-methylbutanoate 1010 NDa 0.707 ± 0.105b 1.547 ± 0.254c NDa
v48 Methyl pentanoate 1019 NDa NDa 0.286 ± 0.069b NDa
v49 Methyl 4-methylpentanoate 1145 NDa 0.049 ± 0.015a 2.453 ± 0.337c 0.836 ± 0.080b
v50 Methyl hexanoate 1194 NDa 0.583 ± 0.099b 2.161 ± 0.302c 1.041 ± 0.071b
v51 Methyl 4-methylhexanoate 1255 NDa 0.218 ± 0.036a 1.993 ± 0.400c 0.624 ± 0.029b
v52 Methyl 4-methylpent-3-enoate 1277 NDa 0.209 ± 0.063b 0.182 ± 0.048b 0.068 ± 0.006a
v55 Methyl octanoate 1391 0.081 ± 0.010a 0.521 ± 0.077b 1.787 ± 0.319c 0.705 ± 0.042b
v56 Methyl oct-4-enoate 1432 NDa 0.272 ± 0.054c 0.476 ± 0.054d 0.122 ± 0.006b
v58 Methyl nonanoate 1494 0.068 ± 0.014a NDa 0.713 ± 0.143c 0.534 ± 0.098b
v59 Methyl decanoate 1598 0.060 ± 0.017a 0.337 ± 0.042b 0.753 ± 0.139c 0.365 ± 0.062b
v62 Methyl dodecanoate 1805 NDa 0.020 ± 0.004a 0.609 ± 0.128b NDa
v64 Methyl tetradecanoate >2000 0.126 ± 0.023a 0.437 ± 0.019a 2.647 ± 0.360c 2.106 ± 0.163b
v65 Ethyl tetradecanoate >2000 NDa NDa NDa 0.057 ± 0.005b
v66 Methyl 12-methyltetradecanoate >2000 0.070 ± 0.017ab 0.079 ± 0.068b 1.068 ± 0.108c NDa
v67 Methyl pentadecanoate >2000 0.024 ± 0.002a 0.129 ± 0.012b 0.567 ± 0.058d 0.405 ± 0.041c
v68 Methyl 14-methylpentadecanoate >2000 NDa 0.052 ± 0.002b 0.237 ± 0.012c 0.317 ± 0.031d
v69 Methyl hexadecanoate >2000 1.109 ± 0.059a 0.033 ± 0.008a 29.559 ± 2.841c 16.946 ± 0.849b
v70 Methyl hexadec-7-enoate >2000 NDa 0.287 ± 0.003b 0.315 ± 0.062b 0.273 ± 0.026b
v71 Methyl (Z)-hexadec-9-enoate >2000 NDa 2.742 ± 0.293c 0.345 ± 0.044b 0.365 ± 0.028b
v72 Ethyl hexadecanoate >2000 NDa 0.269 ± 0.063c 0.151 ± 0.017b 1.504 ± 0.092d
v74 Methyl heptadecanoate >2000 NDa 1.971 ± 0.424b 0.049 ± 0.004a 0.048 ± 0.002a
v75 Methyl octadecanoate >2000 NDa NDa 1.266 ± 0.342c 0.345 ± 0.036b
v76 Methyl octadec-9-enoate >2000 0.199 ± 0.030a NDa 7.653 ± 1.647c 3.363 ± 0.147b
v77 Methyl (9Z,12Z)-octadeca-9,12-dienoate >2000 0.212 ± 0.024a NDa 5.061 ± 0.347c 3.661 ± 0.189b
v78 Methyl octadeca-9,12,15-trienoate >2000 NDa NDa 0.365 ± 0.039c 0.281 ± 0.030b
Furans and furan derivatives
v80 Furan-2-carbaldehyde 1460 NDa NDa NDa 0.107 ± 0.011b
v81 Oxolan-2-one 1639 0.075 ± 0.020b 0.193 ± 0.033c 0.072 ± 0.016b NDa
v82 Furan-2-ylmethanol 1667 NDa 0.023 ± 0.020b 0.203 ± 0.010d 0.147 ± 0.007c
Hydrocarbons
v84 Hexane <900 1.016 ± 0.256b 2.694 ± 0.322c 0.768 ± 0.014b NDa
v85 2-Methylnonane 956 NDa 0.199 ± 0.040b 0.465 ± 0.041d 0.313 ± 0.006c
v86 3-Methylnonane 966 NDa 0.153 ± 0.043b 0.261 ± 0.039c 0.189 ± 0.007b
v87 Decane 1000 0.405 ± 0.082a 1.669 ± 0.037b 4.194 ± 0.132c 4.480 ± 0.576c
v89 Tridecane 1300 nda NDa NDa 0.148 ± 0.006b
v91 Hexadecane 1600 0.050 ± 0.006b 0.151 ± 0.043c NDa NDa
Phenols
v93 2-Methoxyphenol 1870 0.040 ± 0.002a 0.092 ± 0.010a 4.879 ± 0.919c 3.183 ± 0.335b
v95 2-Methoxy-3-prop-2-enylphenol >2000 NDa NDa 0.053 ± 0.012b NDa
v97 4-Ethenyl-2-methoxyphenol >2000 0.023 ± 0.000a 0.051 ± 0.014a 0.599 ± 0.015c 0.468 ± 0.058b
v98 4-Ethenylphenol >2000 0.050 ± 0.007a 71.482 ± 6.445b NDa 0.276 ± 0.028a
Pyrazines
v99 2,6-Dimethylpyrazine 1330 NDa 0.027 ± 0.000a 0.181 ± 0.041b 0.150 ± 0.010b
v100 2,3,5-Trimethylpyrazine 1406 NDa 0.084 ± 0.021b 0.218 ± 0.049c 0.189 ± 0.009c
v101 2,3,5,6-Tetramethylpyrazine 1477 0.091 ± 0.017b 0.039 ± 0.017a 0.162 ± 0.009d 0.116 ± 0.001c
Sulfur-containing compounds
v102 (Methyltrisulfanyl)methane 1380 NDa 0.046 ± 0.011b 0.157 ± 0.007d 0.084 ± 0.014c
Miscellaneous compounds
v105 1-(1H-Pyrrol-2-yl)ethanone
(2-Acetyl-1-pyrroline)		 1338 NDa 0.224 ± 0.037b 0.222 ± 0.012b NDa
v106 Alpha-cubebene 1434 NDa NDa NDa 0.073 ± 0.008b
v107 Alpha-muurolene 1725 0.031 ± 0.006a 0.307 ± 0.054b 0.702 ± 0.079c 0.000 ± 0.000a
v108 Delta-cadinene 1758 0.038 ± 0.009a 0.035 ± 0.007a 0.829 ± 0.099c 0.293 ± 0.023b
Open in a new tab

1Retention indices were determined using n-paraffins C7–C22 as external standards

2Average of contents compared to that of the internal standard ± standard deviation

3Tentative identification was performed as follow: A, mass spectrum was consistent with that of Wiley mass spectrum database; B, mass spectrum and retention index were consistent with those of the mass spectrum database and literatures

4There are significant differences (p < 0.05) among samples collected according to fermentation period by using Duncan’s multiple comparison test between the samples having different letter in low

5Not detected

The relative levels of acetic acid and 3-methylbutanoic acid in CBS increased until 2 weeks of fermentation and then decreased after 5 weeks of fermentation. In contrast, the amount of acetic acid in TBS increased until 1 week of fermentation, decreased after 2-weeks of fermentation, and then again increased from 5 weeks of fermentation. Notably, 3-methylbutanoic acid was not detected in TBS from 2 weeks of fermentation. Previous studies reported that the content of volatile organic acids varied according to the predominant microorganism in fermented products, with Bacillus species being mainly associated with the production of these compounds [20, 21]. The irregular aspect of 3-methylbutanoic acid not being detected in TBS could have been due to the unequal progression of fermentation with natural microflora, in contrast to CBS being fermented with a single predominant microorganism (Bacillus amyloliquefaciens RD7-7).

3-Methylbutan-1-ol, butane-2,3-diol, butane-1,3-diol, and 4-methoxybutan-1-ol were dominant alcohols in both two BSs during the initial stage of fermentation, and their contents or aspects tended to vary with the progression of fermentation. In additional, ethanol was the most abundant compound in the later state of fermentation in TBS, while this was not found in CBS. A previous study [22, 23] found that ethanol could be generated in soybean by the metabolism of yeast. In the present study, the higher amount of ethanol in TBS could also have resulted from the higher activities of yeast and/or bacteria during TBS fermentation.

Esters contribute to positive aroma properties of fermented products [24, 25], and they can be produced by the esterification of alcohols with fatty acids [26]. In particular, esters with low molecular weights have floral, sweet, and fruity aroma properties and help to reduce the sharpness of fatty acids and the bitterness of amines [25]. In contrast, esters with high molecular weights commonly exhibit fatty and oily aroma notes [25]. In the present study, the contents of esters with low molecular weights (RI < 1600) was higher in TBS, which could have been due to the higher amount of ethanol in TBS facilitating the generation of esters.

Changes in volatile compounds in two BSs according to fermentation periods by PLS-DA

The PLS-DA is one of efficient statistical techniques to clarify the separation between samples, to build their classification models, and to see how the underlying variables affect the separation between samples. Recently, the models have been successfully applied to discriminate grape musts based on their volatile compositions during fermentation [27]; water-soluble metabolites of meju according to fermentation periods [28]; olive oil samples from the Catalonia region with the presence of a variety of defects [29], and so on. In this study, the discriminations of BSs according to the fermentation periods were accomplished by applying their volatile profiles to PLS-DA, to ensure the objective interpretation of the results in Tables 1 and 2.

In a PLS-DA model, the X variables (the predictors) are reduced to scores of component (estimates of the latent variables or their rotation), which are denoted by ta (a = 1,2,3,…). In other words, the variables are correlated and summarized by new variables ta scores whose values explain the variation represented by each principal component. The score t1 (first component) explains the largest variation of the X space, followed by t2, and so on [30]. The PLS-DA models revealed clear clustering among each BS groups based on the volatile compounds of BSs during fermentation (0, 1, 2, and 5 weeks) in this study (Figs. 1, 2). Their internal cross-validations on PLS-DA were performed and component 4 and parameters, such as R2X = 0.931, R2Y = 0.982, and Q2Y = 0.957 in TBS and R2X = 0.969, R2Y = 0.981, and Q2Y = 0.964 in CBS, respectively. The score plot in Fig. 1 indicates that the TBS samples were located on the positive t[1] axis during the initial and middle stages of fermentation (up to 2 weeks), while these samples were on the negative t[2] axis during the latest stage of fermentation (5 weeks). On the other hand, the score plot for CBS in Fig. 2 shows that CBS samples during the initial and first middle stages of fermentation (up to 1 week) and during the second middle and latest stage of fermentation (2–5 weeks) could be clearly separated by t[1]. Meanwhile, the discrimination between 0- and 1-week CBS samples and the separation between 2- and 5-week CSB samples could be done by t[2]: the sample at the initial stage of fermentation sample (0 week) was located on the positive t[1] and t[2] axes, that at the first middle stage of fermentation (1 week) was on the positive t[1] and negative t[2] axes, that at the second middle stage of fermentation (2 weeks) being on the negative t[1] and t[2], and the latest stage of fermentation sample (5 weeks) was on the negative t[1] and positive t[2] axes. It seems that the change in the volatiles of CBS was accelerated relative to TBS because CBS had initially been fermented with a novel multi-starter. A starter culture could induce a high biosynthetic capacity of fermented soybean products and uniformly control them to produce flavor compounds [31, 32].

Fig. 1.

Fig. 1

Open in a new tab

PLS-DA score plot of TBS samples during fermentation

Fig. 2.

Fig. 2

Open in a new tab

PLS-DA score plot of CBS samples during fermentation

Loading plots were produced to classify which variables contribute most to the discrimination ability. Tables 3 and 4 show their contribution on variables with loadings in a loadings plot, denoted by “pa”, contribute heavily to observations whose scores are found in a similar position in a scores plot [30]. The major volatile compounds that contribute to the t[1] dimension during TBS fermentation in the following order in Table 3: ethyl dodecanoate (v63) > 4-methyl-2H-furan-5-one (v83) > 3-methylsulfanylpropan-1-ol (v103) > tridecane (v89) > α-cubebene (v106) > ethanol (v4) > methyl pyridine-3-carboxylate (v60) > ethyl hexanoate (v53) > styrene (v104) > methyl hexadec-7-enoate (v70) > 1,3-xylene (v20) > 2-phenylbut-2-enal (v34) > 4-ethyl-2-methoxyphenol (v94). In particular, since these compounds where on the negative t[1] axis, they were strongly associated with the latest stage of TBS fermentation. This also means that their amounts were increased after 5 weeks of TBS fermentation. On the other hand, the following compounds were important contributors to the PC 2 dimension (also in decreasing order): methyl decanoate (v59) > 3-hydroxy-2,6-dimethylpyran-4-one (v18) > methyl heptanoate (v54) > delta-cadinene (v108) > oct-1-en-3-ol (v13) > 2-methoxyphenol (v93) > methyl 3-phenylpropanoate (v30) > heptadecane (v92) > oxolan-2-one (v81) > 3-hydroxybutan-2-one (v40) > hexanoic acid (v3) > dimethyl carbonate (v46). Methyl 3-phenylpropanoate (v30), heptadecane (v92), oxolan-2-one (v81), 3-hydroxybutan-2-one (v40), hexanoic acid (v3), and dimethyl carbonate (v46) were located on the positive t[1] and negative t[2] axes and were mainly associated with the initial stage of fermentation in TBS samples, whereas methyl decanoate (v59) on the positive t[1] and t[2] axes, was the dominant compound during the middle stage of TBS fermentation.

Table 3.

The contribution of volatile variances in TBS according to fermentation period

No. p[1] p[2] No. p[1] p[2] No. p[1] p[2]
v1 −0.143 0.020 v38 0.005 0.001 v73 0.047 0.038
v2 0.085 0.051 v39 0.076 0.139 v75 0.050 0.148
v3 0.053 −0.156 v40 0.053 −0.156 v76 0.094 0.128
v4 −0.159 0.012 v41 0.089 0.140 v77 0.046 0.118
v5 −0.092 −0.132 v42 −0.019 0.116 v79 0.005 0.039
v6 0.105 0.095 v43 −0.149 0.006 v80 −0.149 0.010
v7 −0.153 −0.001 v44 0.048 0.037 v81 0.053 −0.155
v8 −0.141 0.040 v45 0.126 0.100 v82 0.105 0.117
v9 0.108 0.073 v46 0.053 −0.156 v83 −0.161 0.003
v10 0.109 0.101 v47 0.048 0.036 v85 0.059 0.120
v11 −0.038 0.150 v50 −0.059 0.151 v87 0.096 0.139
v13 −0.052 0.160 v53 −0.158 0.001 v88 0.047 0.038
v15 −0.134 −0.015 v54 −0.007 0.166 v89 −0.160 0.004
v16 −0.021 0.045 v55 −0.043 0.164 v90 0.047 0.038
v17 0.097 0.137 v56 0.084 0.145 v92 0.053 −0.154
v18 −0.039 0.168 v57 −0.111 0.018 v93 −0.039 0.160
v19 −0.143 0.010 v58 0.091 0.135 v94 −0.156 0.004
v20 −0.156 0.000 v59 0.005 0.175 v95 0.090 0.143
v22 −0.151 0.056 v60 −0.159 0.003 v96 −0.154 0.004
v24 −0.110 0.126 v61 −0.152 0.017 v97 −0.128 0.093
v25 −0.118 0.117 v62 0.096 0.131 v98 −0.151 0.008
v26 −0.126 0.109 v63 −0.161 0.002 v99 0.088 0.140
v27 −0.097 0.134 v64 −0.039 0.148 v101 −0.127 0.078
v28 −0.110 0.121 v65 −0.153 0.010 v102 0.063 −0.013
v29 −0.084 0.145 v66 −0.081 0.131 v103 −0.161 0.006
v30 0.053 −0.154 v67 0.095 0.129 v104 −0.158 0.002
v31 −0.094 0.141 v68 0.042 −0.120 v106 −0.160 0.003
v32 −0.152 0.055 v69 0.100 0.133 v107 0.033 0.127
v34 −0.156 0.020 v70 −0.157 0.015 v108 −0.012 0.164
v35 0.043 0.124 v71 0.037 0.119
v37 0.115 0.094 v72 0.063 0.060
Open in a new tab

Table 4.

The contribution of volatile variances in CBS according to fermentation period

No. p[1] p[2] No. p[1] p[2] No. p[1] p[2]
v1 −0.024 −0.106 v37 −0.088 −0.157 v75 −0.139 −0.030
v2 −0.143 −0.024 v38 0.023 −0.034 v76 −0.143 0.003
v3 −0.126 −0.075 v40 0.117 0.099 v77 −0.139 0.057
v5 0.083 0.076 v42 −0.127 −0.096 v78 −0.139 0.058
v8 −0.069 −0.002 v45 −0.129 −0.089 v80 −0.024 0.175
v11 −0.139 −0.059 v46 0.081 0.075 v81 0.057 −0.216
v12 −0.024 0.175 v47 −0.101 −0.165 v82 −0.139 0.038
v13 −0.143 0.018 v48 −0.124 −0.074 v84 0.071 −0.204
v14 0.072 −0.179 v49 −0.142 −0.020 v85 −0.134 −0.036
v15 −0.001 0.050 v50 −0.141 −0.046 v86 −0.123 −0.060
v16 0.056 −0.208 v51 −0.140 −0.045 v87 −0.122 0.053
v17 −0.142 0.002 v52 −0.055 −0.199 v89 −0.024 0.176
v19 −0.137 −0.007 v55 −0.138 −0.068 v91 0.095 −0.166
v21 0.043 −0.195 v56 −0.109 −0.155 v93 −0.142 0.036
v22 −0.116 −0.086 v58 −0.138 0.063 v95 −0.126 −0.075
v23 −0.140 0.048 v59 −0.133 −0.082 v97 −0.137 0.053
v24 −0.142 −0.003 v62 −0.125 −0.082 v98 0.065 −0.182
v25 −0.140 0.049 v64 −0.139 0.041 v99 −0.137 0.042
v26 −0.141 0.040 v65 −0.024 0.175 v100 -0.131 0.004
v27 0.041 −0.030 v66 −0.119 −0.094 v101 −0.122 0.075
v28 −0.139 −0.033 v67 −0.141 0.016 v102 −0.139 −0.039
v29 −0.140 0.049 v68 −0.113 0.096 v105 −0.050 −0.216
v30 0.030 −0.205 v69 −0.143 0.030 v106 −0.024 0.175
v31 −0.142 −0.021 v70 −0.094 −0.092 v107 −0.100 −0.163
v32 −0.139 0.062 v71 0.049 −0.183 v108 −0.141 −0.019
v35 −0.113 0.144 v72 −0.028 0.146
v36 −0.098 −0.169 v74 0.061 −0.183
Open in a new tab

In addition, the main volatiles on the t[1] dimension contributing to discrimination among CBS samples were methyl hexadecanoate (v69), methyl octadec-9-enoate (v76), 3-methylbutanoic acid (v2), oct-1-en-3-ol (v13), methyl benzoate (v24), methyl 4-methylpentanoate (v49), 2-methoxyphenol (v93), phenylmethanol (v31), and 4-methyl-1-propan-2-ylcyclohex-3-en-1-ol (v17) (Table 4). In particular, these compounds were located on the negative t[1] axis, suggesting that they were predominantly produced during the second middle stage of CBS fermentation process (first middle stage → second middle stage). On the other hand, oxolan-2-one (v81), 1-(1H-pyrrol-2-yl) ethanone (v105), butane-1,3-diol (v16), methyl 3-phenylpropanoate (v30), hexane (v84), methyl 4-methylpent-3-enoate (v52), 1,4-xylene (v21), methyl (Z)-hexadec-9-enoate (v71), methyl heptadecanoate (v74), 4-ethenylphenol (v98), 2-ethylhexan-1-ol (v14), tridecane (v89), (Z)-hex-3-en-1-ol (v12), furan-2-carbaldehyde (v80), ethyl tetradecanoate (v65), and α-cubebene (v106) were important volatile variables on the t[2] dimension. In particular, 3-hydroxybutan-2-one (v40), located on the positive t[1] and t[2] axes, was related to the initial stage of CBS fermentation, and oxolan-2-one (v81), butane-1,3-diol (v16), methyl 3-phenylpropanoate (v30), hexane (v84), 1,4-xylene (v21), methyl (Z)-hexadec-9-enoate (v71), methyl heptadecanoate (v74), 4-ethenylphenol (v98), and 2-ethylhexan-1-ol (v14), being on the positive t[1] and negative t[2] axes, were strongly associated with the first middle stage of CBS fermentation. In addition, 1-(1H-pyrrol-2-yl) ethanone (v105) and methyl 4-methylpent-3-enoate (v52), being on the negative t[1] and t[2] axes, were especially produced between the first (1 week) and second (2 weeks) middle stages of CBS fermentation. Lastly, tridecane (v89), (Z)-hex-3-en-1-ol (v12), furan-2-carbaldehyde (v80), ethyl tetradecanoate (v65), and α-cubebene (v106) which were located on the negative t[1] and positive t[2] axes were associated with the latest stage of CBS fermentation.

In conclusion, the volatile compounds in BSs changed or differed according to the fermentation type (TBS or CBS) and the fermentation period (0, 1, 2, and 5 weeks). In particular, the PLS-DA models identified different trends between TBS and CBS fermentation. During TBS fermentation, 0-, 1-, and 2-week samples and 5-week samples were preferentially separated, whereas 0- and 1-week samples and 2- and 5-week samples were discriminated during CBS fermentation. It seems that the generation of flavors in BS could be closely related to the microorganisms involved in fermentation. The use of a starter culture at the beginning of fermentation could facilitate the biosynthetic and chemical reactions in fermented products and subsequently increase the production of secondary metabolites, including flavor compounds.

Acknowledgements

This paper was supported by Wonkwang University in 2015.

Compliance with ethical standards

Conflict of interest

The authors declare no conflict of interest.

References

  • 1.Kwak CS, Lee MS, Park SC. Higher antioxidant properties of Chungkookjang, a fermented soybean paste, may be due to increased aglycone and malonylglycoside isoflavone during fermentation. Nutr. Res. 2009;27:719–727. doi: 10.1016/j.nutres.2007.09.004. [DOI] [Google Scholar]
  • 2.Wang D, Wang LJ, Zhu FX, Zhu JY, Chen XD, Zou L, Saito M, Li LT. In vitro and in vivo studies on the antioxidant activities of the aqueous extracts of Douchi (a traditional Chinese salt-fermented soybean food) Food Chem. 2008;107:1421–1428. doi: 10.1016/j.foodchem.2007.09.072. [DOI] [Google Scholar]
  • 3.Fujita M, Nomura K, Hong K, Ito Y, Asada A, Nishimuro S. Purification and characterization of a strong fibrinolytic enzyme (Nattokinase) in the vegetable cheese natto, a popular soybean fermented food in Japan. Biochem. Bioph. Res. Com. 1993;197:1340–1347. doi: 10.1006/bbrc.1993.2624. [DOI] [PubMed] [Google Scholar]
  • 4.Choi NS, Chung DM, Han YJ, Kim SH, Song JJ. Purification and characterization of a subtilisin D5, a fibrinolytic enzyme of Bacillus amyloliquefaciens DJ-5 isolated from Doenjang. Food Sci. Biotechnol. 2009;18:500–505. [Google Scholar]
  • 5.Jung KO, Park SY, Park KY. Longer aging time increases the anticancer and antimetastatic properties of doenjang. Nutrition. 2006;22:539–545. doi: 10.1016/j.nut.2005.11.007. [DOI] [PubMed] [Google Scholar]
  • 6.Kerwin SM. Soy Saponins and the anticancer effects of soybeans and soy-based foods. Curr. Med. Chem. Anti-Cancer Agents. 2004;4:263–272. doi: 10.2174/1568011043352993. [DOI] [PubMed] [Google Scholar]
  • 7.Kim YS, Kim MC, Kwon SW, Kim SJ, Park IC, Ka JO, Weon HY. Analyses of bacterial communities in meju, a Korean traditional fermented soybean bricks, by cultivation-based and pyrosequencing methods. J. Micobiol. 2011;49:340–348. doi: 10.1007/s12275-011-0302-3. [DOI] [PubMed] [Google Scholar]
  • 8.Lee JH, Kim TW, Chang HC, Kim HY. Determination of microbial diversity in meju, fermented cooked soya beans, using nested PCR-denaturing gradient gel electrophoresis. Lett. Appl. Microbiol. 2010;51:388–394. doi: 10.1111/j.1472-765X.2010.02906.x. [DOI] [PubMed] [Google Scholar]
  • 9.Park CH, Kim YB, Choi YS, Heo K, Kim SL, Lee KC. Rutin content in food products processed from groats, leaves and flowers of buckwheat. Fagopyrum. 2000;17:63–66. [Google Scholar]
  • 10.Kreft S, Knapp M, Kreft I. Extraction of rutin from buckwheat (Fagopyrum esculentum Moench) seeds and determination by capillary electrophoresis. J. Agr. Food Chem. 1999;47:4649–4652. doi: 10.1021/jf990186p. [DOI] [PubMed] [Google Scholar]
  • 11.Sun T, Ho CT. Antioxidant activities of buckwheat extracts. Food Chem. 2005;90:743–749. doi: 10.1016/j.foodchem.2004.04.035. [DOI] [Google Scholar]
  • 12.Oomah BD, Mazza G. Flavonoids and antioxidative activities in buckwheat. J. Agr. Food Chem. 1996;44:1746–1750. doi: 10.1021/jf9508357. [DOI] [Google Scholar]
  • 13.Jo YJ, Cho IH, Song CK, Shin HW, Kim YS. Comparison of fermented soybean paste (Doenjang) prepared by different methods based on profiling of volatile compounds. J. Food Sci. 2011;76:368–379. doi: 10.1111/j.1750-3841.2011.02068.x. [DOI] [PubMed] [Google Scholar]
  • 14.Cho KM, Lim HJ, Kim MS, Hwang CE, Nam SH, Joo OS, Lee BW, Kim JK, Shin EC. Time course effects of fermentation on fatty acid and volatile compound profiles of Cheonggukjang using new soybean cultivars. J. Food Drug Anal: In press; 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Yoo SK, Cho WH, Kang SM, Lee SH. Isolation and identification of microorganisms in Korean traditional soybean paste and soybean sauce. Med. Microbiol. Lett. 1999;27:113–117. [Google Scholar]
  • 16.Kim TW, Lee JW, Kim SE, Park MH, Chang HC, Kim HY. Analysis of microbial communities in doenjang, a Korean fermented soybean paste, using nested PCR-denaturing gradient gel electrophoresis. Int. J. Food Microbiol. 2009;131:265–271. doi: 10.1016/j.ijfoodmicro.2009.03.001. [DOI] [PubMed] [Google Scholar]
  • 17.Park MK, Cho IH, Lee SR, Choi HK, Kwon DY, Kim YS. Metabolite profiling of Cheinggukjang, a fermented soybean paste, during fermentation by gas chromatography–mass spectrometry and principal component analysis. Food Chem. 2010;122:1313–1319. doi: 10.1016/j.foodchem.2010.03.095. [DOI] [Google Scholar]
  • 18.Kim GE, Kim MH, Choi BD, Kim TS, Lee JH. Flavor compounds of domestic meju and Doenjang. J. Korean Soc. Food Nutr. 1992;21:557–565. [Google Scholar]
  • 19.Tavaria FK, Ferreira SF, Malcata X. Volatile Free Fatty Acids as Ripening Indicators for Serra da Estrela Cheese. J. Dairy Sci. 2004;87:4064–4072. doi: 10.3168/jds.S0022-0302(04)73548-1. [DOI] [PubMed] [Google Scholar]
  • 20.Ji WD, Yang SH, Choi MR, Kim JK. Volatile components of Korean soybean paste produced by B. subtilis PM3. J. Micobiol Biotechnol. 1995;5:143–148. [Google Scholar]
  • 21.Park HK, Kim JK. Optimal manufacturing conditions for Korean soybean paste and soy sauce, using Aspergillus oryzae AJ 100 as a flavor improver. Food Sci. Biotechnol. 2008;17:208–211. [Google Scholar]
  • 22.Fleet GH, Lafon-lafourcade S, Ribereau-Gayon P. Evolution of yeasts and bacteria during fermentation and storage of bordeaus wines. Appl. Environ. Microbiol. 1984;48:1034–1038. doi: 10.1128/aem.48.5.1034-1038.1984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Bontinis TG, Mallatou H, Pappa EC, Massouaas T, Alichanidis E. Study of proteolysis, lipolysis and volatile profile of a traditional Greek goat cheese (Xinotyri) during ripening. Small Rumin Res. 2012;105:193–201. doi: 10.1016/j.smallrumres.2012.01.003. [DOI] [Google Scholar]
  • 24.Francis IL, Newton JL. Determining wine aroma from compositional data. Aust. J. Grape Wine Res. 2005;11:114–126. doi: 10.1111/j.1755-0238.2005.tb00283.x. [DOI] [Google Scholar]
  • 25.Reineccius G. Flavor Chemistry and Technology. 2. New York, USA: Taylor & Francis; 2006. pp. 73–98. [Google Scholar]
  • 26.Zhao Y, Xu Y, Li J, Fan W, Jiang W. Profile of volatile compounds in 11 brandies by headspace solid-phase microextraction followed by gas chromatography–mass spectrometry. J. Food Sci. 2009;74:90–99. doi: 10.1111/j.1750-3841.2008.01029.x. [DOI] [PubMed] [Google Scholar]
  • 27.Novel application to investigate problem fermentations Malherbe S, Watts VH. Nieuwoudt H, F. Bauer F, Du Toit M. Analysis of volatile profiles of fermenting grape must by headspace solid-phase dynamic extraction coupled with gas chromatography–mass spectrometry (HS-SPDE GC–MS) J. Agr. Food Chem. 2009;57:5161–5166. doi: 10.1021/jf900532v. [DOI] [PubMed] [Google Scholar]
  • 28.Kang HJ, Yang HJ, Kim MJ, Han ES, Kim HJ, Kwon DY. Metabolomic analysis of meju during fermentation by ultra performance liquid chromatography-quadrupole-time of flight mass spectrometry (UPLC-Q-TOF MS) Food Chem. 2011;127:1056–1064. doi: 10.1016/j.foodchem.2011.01.080. [DOI] [PubMed] [Google Scholar]
  • 29.Borràs E, Ferré J, Boqué R, Mestres M, Aceña L, Calvo A, Busto O. Olive oil sensory defects classification with data fusion of instrumental techniques and multivariate analysis (PLS-DA) Food Chem. 2016;203:314–322. doi: 10.1016/j.foodchem.2016.02.038. [DOI] [PubMed] [Google Scholar]
  • 30.Szymańska E, Saccenti E, Smilde AK, Westerhuis JA. Double-check: validation of diagnostic statistics for PLS-DA models in metabolomics studies. Metabolomics. 2012;8:3–16. doi: 10.1007/s11306-011-0330-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Guldfeldt LU, Sørensen KI, Strøman P, Behrndt H, Williams D, Johansen E. Effect of starter cultures with a genetically modified peptidolytic or lytic system on Cheddar cheese ripening. Int. Dairy J. 2001;11:373–382. doi: 10.1016/S0958-6946(01)00066-8. [DOI] [Google Scholar]
  • 32.Kum SJ, Yang SO, Lee SM, Chang PS, Choi YH, Lee JJ, Hurh BS, Kim YS. Effects of Aspergillus Species Inoculation and Their Enzymatic Activities on the Formation of Volatile Components in Fermented Soybean Paste (doenjang) J. Agr. Food Chem. 2015;63:1401–1418. doi: 10.1021/jf5056002. [DOI] [PubMed] [Google Scholar]

Articles from Food Science and Biotechnology are provided here courtesy of Springer

RESOURCES