Original papers
Quantification of Functional Aromatic Amino Acid Metabolites in Fermented Foods and Their Production by Food Microorganisms
2020 年 26 巻 1 号 p. 79-92
詳細
2020 年 26 巻 1 号 p. 79-92
Aromatic amino acid metabolites, including aromatic pyruvates and aromatic lactates, have been shown to be antioxidants with high radical scavenging activity. We surveyed the occurrence and distribution of these metabolites in foods and beverages, focusing on fermented products. Lactic acid bacteria (LAB)-fermented products showed high concentrations of aromatic lactates and much lower concentrations of aromatic pyruvates, while yeast-fermented products showed relatively high levels of aromatic pyruvates. The results were in accordance with the production of aromatic lactates and pyruvates by LAB and Saccharomyces cerevisiae under standard experimental culture conditions. Our findings therefore indicate the importance of utilizing the metabolic activities of fermenting microorganisms for developing foods with beneficial functions.
Free radicals generated by oxidation damage important biomolecules such as lipids, proteins, and DNA. Accumulated oxidative damage may mediate chronic diseases and is believed to be a primary determinant of aging. Antioxidants provide a first line of defense, and both endogenous antioxidants and the intake of antioxidants from food may help protect the human body from cellular damage and diseases mediated by free radicals.
In an effort to identify novel antioxidants produced by lactic acid bacteria (LAB), we found that L-indole-3-lactic acid (ILA) and L-3-(4-hydroxyphenyl)lactic acid (HPLA) from cultures of Lactobacillus plantarum exhibit 2, 2-diphenyl-1- picrylhydrazyl (DPPH) radical scavenging activity (Suzuki et al., 2013). Aromatic lactic acids (ALAs) including HPLA, ILA, and 3-phenyllactic acid (PLA) were identified as LAB metabolites produced by a transamination reaction on the aromatic amino acids, followed by subsequent dehydration of the aromatic pyruvic acids (APAs) (Yvon et al., 1997; Rijnen et al., 1999; Li et al., 2008) (See Fig. 1 for their structures). The L-aromatic amino acids (AAA) exhibited no DPPH radical scavenging activity. Consistent with these results, we have shown that pretreatment of HaCaT human keratinocytes with ILA or HPLA prevents ultraviolet B (UVB)-induced production of interleukin (IL)-6 (Aoki-Yoshida et al., 2013). The APAs 4-(3-hydroxyphenyl)pyruvic acid (HPPA), 3-phenylpyruvic acid (PPA), and indol-3-pyruvic acid (IPA) showed high DPPH radical scavenging activity, comparable to that of 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (trolox) (Suzuki et al., 2013). Therefore, HPPA, PPA, and IPA are promising antioxidant compounds with high radical scavenging activity. We explored the functional potency of APAs by observing the protective effects of APAs against UVB-derived oxidative stress in vitro in HaCaT keratinocytes and in vivo in hairless mice (HR-1) (Aoki et al., 2014). We found that the APAs antagonized cytotoxicity in UVB-irradiated HaCaT keratinocytes, and topical application of IPA to the dorsal skin of hairless mice reduced the severity of UVB-induced skin lesions (Aoki et al., 2014). Scrima et al. (2017) showed that millimolar concentrations of HPPA dose dependently prevented both nitric oxide synthase 2 expression and NO production in RAW 264.7 macrophages, triggering LPS-mediated nitro-oxidative unbalance and metabolic shift. Of the APAs investigated, IPA most effectively activated aryl hydrocarbon receptor (AHR) in vitro and in vivo, and oral administration of IPA improved chronic inflammation in an experimental colitis model, suggesting that IPA potently prevents chronic inflammation in the colon by activating AHR (Aoki et al., 2018).
Chemical structures of ALAs, APAs, and thiazolidine-derivatized APAs.
Although the beneficial activities of AAA-metabolites are now known, their occurrence and distribution in food and beverages have not been investigated. IPA in plants has been examined as a biosynthetic precursor of IAA (indole-3-acetic acid), an active constituent of the phytohormone auxin. IPA is unstable and cannot be detected with high sensitivity by LC/MS and thus requires derivatization prior to analysis (Novak et al., 2012). Bettenhausen et al. (2018) analyzed metabolites in malts and beers using UHPLC- and HILIC-MS (non-volatile metabolites), HS-SPME/GC-MS (beer volatiles), and ICP-MS (malt metals), and detected 5042 compounds in malt and 4568 compounds in beer. HPLA and PLA were detected in malts and beers whereas HPPA and PPA were not, suggesting the difficulty of detecting APAs without derivatization. We utilized thiazolidine (TAZ) derivatization to analyze APAs (Fig. 1) and obtained the highest sensitivity of any derivatization method reported to date (Novak et al., 2012). As mentioned above, LAB produce ALAs. The aminotransferase required to convert AAAs to APAs (Yvon et al., 1997; Rijnen et al., 1999) and the lactate dehydrogenase to further convert APAs to aromatic lactic acids (ALAs, Fig. 1) (Li et al., 2008) have been characterized. Saccharomyces cerevisiae also produces enzymes capable of the transamination of AAAs to produce APAs (Iraqui et al., 1998). Fermentation microorganisms can produce AAA metabolites during fermentation. We thus focused in this work on fermented foods, analyzing the concentrations of APAs and ALAs and comparing them with their concentrations in the corresponding starting materials. In addition, a variety of food samples were analyzed.
Fermented food samples Various fermented food samples and their corresponding starting materials were purchased from supermarkets in Japan: dairy products (yogurt, cheese, milk), fermented soy bean products (miso, soy sauce, natto), bread, fermented beverages (amazake, sake, beer, shochu, whisky), Menma (fermented bamboo shoot), and the starting materials of fermented foods and beverages (wheat flour, soy bean, rice grain, barley grain). Dried lactate-fermented bamboo shoots were kindly provided by the manufacturer. Raw bamboo shoots were harvested at Ibaraki University. Amazake, or sweet sake, is a non-alcoholic sweet beverage made of glycosylated koji rice malt. The processes used to produce other fermented products, including miso, soy sauce, natto, sake, and shochu are described in Kitamura et al. (2016). Vegetables, meats, and raw seafood for sashimi were purchased from supermarkets in Japan: cucumber, tomato, carrot, radish (daikon), maitake mushroom (Grifola frondosa), brown seaweed, pork, raw ham, shrimp, tuna, scallop, and sea bream.
Sample preparation for analysis of aromatic amino acid metabolites by LC-MS/MS Yogurt samples were centrifuged at 13 000 g for 5 min, and the supernatants were recovered as the analytic samples. All solid (dry or semi-dry) samples were aliquoted (20–40 mg) into 2 mL tubes. After adding 500 µL methanol and three zirconium beads (diameter, 2.8 mm), the tubes were shaken at 5 000 rpm for 3 min using a bead beater homogenizer (Precellys 24 Tissue Homogenizer, Bertin Instruments, Montigny-le-Bretonneux, France). The samples were centrifuged at 13 000 g for 5 min, and the supernatants were recovered as the analytic samples.
LAB or Saccharomyces cerevisiae culture preparation The LAB and yeast strains used in this study are listed in Table 1. Each LAB strain was pre-cultured in 3 mL of MRS broth (Becton Dickinson, Sparks, USA) liquid medium at 30 °C overnight. A 100 µL aliquot of the pre-culture was added to 10 mL of MRS liquid medium, then 1.5 mL aliquots were dispensed into six microtubes and cultured at 30 °C. The culture was sampled at 0, 3, 6, 9, 12, 24, and 48 h after inoculation, centrifuged at 13 000 g for 5 min, and the supernatant was recovered for analysis of ALAs and APAs. Each S. cerevisiae strain was pre-cultured in 1.5 mL of YPD liquid medium in a 15 mL tube at 30 °C overnight with shaking at 200 rpm. A 10 µL aliquot of the pre-culture was added to 1.5 mL of YPD liquid medium and cultured at 30 °C for 1, 2, and 3 d. The culture was centrifuged at 13 000 g for 5 min, and the supernatant was recovered for analysis of ALAs and APAs.
| Strain | Species | Source |
|---|---|---|
| D57 | Pediococcus acidilactici | Laboratory collection |
| P4 | Lactobacillus plantarum | Laboratory collection |
| L-52 | Lactobacillus rhamnosus | Laboratory collection |
| TP2 | Lactobacillus paraplantarum | Laboratory collection |
| NCYC 235 | Saccharomyces cerevisiae | NCYC |
| X2180-1A | Saccharomyces cerevisiae | ATCC |
LC-MS/MS analysis Methanol extracts of the food samples mentioned above or liquid samples (sake, beer, distilled spirits, and supernatants of the LAB and S. cerevisiae cultures) were either directly subjected to LC-MS/MS analysis or were 10-fold diluted with 10% methanol, and then subjected to LC-MS/MS analysis to quantify the ALAs. APAs (IPA, PPA, HPPA) were analyzed as thiazolidine derivatives (IPA-TAZ, PPA-TAZ, HPPA-TAZ). A 10 µL sample was mixed with 200 µL of 0.25 M cysteamine (adjusted to pH 8.0 with NH4OH), and incubated for 2 h at ambient temperature; it was then subjected to LC-MS/MS analysis.
ALAs and derivatized APAs were analyzed using a 3200 QTrap LC–MS/MS instrument (AB Sciex, Foster City, USA). The samples were separated on a Capcell Pak C18 MGIII column (2.0 mm i.d. × 250 mm; Shiseido, Tokyo, Japan) by elution with solvent A (0.1% [v/v] acetic acid in water) and solvent B (0.1% [v/v] acetic acid in methanol). The elution conditions were as follows: isocratic elution for 2 min with 10% B, a linear gradient from 10% B to 100% B over 5 min, then isocratic elution for 6 min with 100% B. This elution program gave the following retention times: ILA at 10.2 min, PLA at 10.5 min, HPLA at 9.0 min, IPA-TAZ at 9.4 min, PPA-TAZ at 9.0 min, and HPPA-TAZ at 7.8 min. The mass spectrometer was operated in the multiple reaction monitoring mode. The ion source (Turbo V Ion Source) was operated in the positive electrospray ionization mode. The source parameters were as follows: curtain gas, 10 psi for all compounds; temperature, 600 °C for APA derivatives and 400 °C for ALAs; spray gas, 70 psi for APA derivatives and 30 psi for ALAs; dry gas, 80 psi for APA derivatives and 30 psi for ALAs; and ion spray voltage, 5 500 V. The following transitions were monitored (collision energy, CE; collision cell exit potential, CXP; declustering potential, DP): IPA-TAZ: m/z 263 /132 (CE 17 V, CXP 4.0 V, DP 36 V); PPA-TAZ: m/z 224 /164 (CE 13 V, CXP 4.0 V, DP 36 V); HPPA-TAZ: m/z 240 /136 (CE 21 V, CXP 4.0 V, DP 36 V); ILA: m/z 206 /118 (CE 27 V, CXP 4.0 V, DP 31 V); PLA: m/z 167 /121 (CE 11 V, CXP 4.0 V, DP 16 V); HPLA: m/z 183 /137 (CE 11 V, CXP 4.0 V, DP 31 V). Both quadrupoles were set at unit resolution.
Statistical analysis Statistical analyses were performed using R version 3.5.1 (R Core Team, 2018). For comparison of endogenous levels of APAs and ALAs in each sample group composed of more than three samples (Figs. 2–6 and Fig. 8), significance of differences was first analyzed by one-way analysis of variance (one-way ANOVA). Where significant differences were observed, Tukey's HSD tests were performed. Statistical analysis for pairwise comparison was performed by Student's t-test (Fig. 5–7). For the time course analysis of the concentration of each APA and ALA in culture media of LAB or S. cerevisiae, one-way ANOVA was performed for detecting significant increase and/or decrease. Where significant increases and/or decreases were observed, Tukey's HSD tests were performed. The level of significance used in all tests was p < 0.05. For comparison of overall levels between APAs and ALAs in yogurts, cheeses, Menma, beers, and sake, Student's t-test was performed using total APAs (sum of HPPA, PPA and IPA), and total ALAs (sum of HPLA, PLA and ILA).
Endogenous levels of APAs (A) and ALAs (B) in yogurt and the starting material, bovine milk. Each yogurt (A–I) denotes an individual commercial product (n = 3, mean ± SD). Significance of difference was analyzed by ANOVA (P < 0.05) followed by Tukey's multiple comparison test (P < 0.05). Different lowercase letters indicate statistically significant differences.
Endogenous levels of APAs (A) and ALAs (B) in cheese and the starting material, bovine milk. The analyzed samples are commercial products from a specific company produced by different processes (A, mozzarella; B, Gouda; C, Colby-Jack; D, red cheddar; E, Parmesan; F, Camembert) (n = 3, mean ± SD). Significance of difference was analyzed by ANOVA (P < 0.05) followed by Tukey's multiple comparison test (P < 0.05). Different lowercase letters indicate statistically significant differences. Levels of PPA and HPPA did not show significant differences by ANOVA.
Endogenous levels of APAs (A) and ALAs (B) in Menma and the starting material, bamboo shoot. Menma A and Menma B denote different commercial products. Analyzed samples include dried fermented bamboo shoot used as an intermediate during the production process and was provided by a Menma-producing company (n = 3, mean ± SD). Significance of difference was analyzed by ANOVA (P < 0.05) followed by Tukey's multiple comparison test (P < 0.05). Different lowercase letters indicate statistically significant differences. Levels of HPPA and IPA did not show significant differences by ANOVA.
Endogenous levels of APAs (A) and ALAs (B) in products fermented by yeast (beer, sake, and bread) and the corresponding starting material (barley grain or malt, polished rice grain, and flour). Each beer (A–D) and sake (A–C) denotes an individual commercial product (n = 3, mean ± SD). Significance of difference for beer group and sake group was analyzed by ANOVA (P < 0.05) followed by Tukey's multiple comparison test (P < 0.05). Different lowercase letters indicate statistically significant differences. Significant differences between flour and plain bread were analysed by Student's t-test (** P<0.01, * P<0.05).
Endogenous levels of APAs (A) and ALAs (B) in koji mold-fermented products (amazake, miso, and soy source) and the corresponding starting material (polished rice grain and soybean). Miso A and miso B denote different commercial miso products (n = 3, mean ± SD). Significance of difference for miso group and soy sauce group was analyzed by ANOVA (P < 0.05) followed by Tukey's multiple comparison test (P < 0.05). Different lowercase letters indicate statistically significant differences. Levels of HPPA and PLA in miso group and HPLA in soy sauce group did not show significant differences by ANOVA. Significant differences between rice grain and amazake bread were analysed by Student's t-test (** P<0.01, * P<0.05).
Endogenous levels of APAs (A) and ALAs (B) in various non-fermented foods (n = 3, mean ± SD). Significance of difference was analyzed by ANOVA (P < 0.05) followed by Tukey's multiple comparison test (P < 0.05). Different lowercase letters indicate statistically significant differences.
Endogenous levels of APAs (A) and ALAs (B) in natto and the starting material, mature soybean (n = 3, mean ± SD). Significant differences between soybean and natto were analysed by Student's t-test (** P<0.01, * P<0.05).
The analysis results for various foods are listed in Supplemental Table 1. The results were also categorized by food group, as shown in Figs. 2–8. We focused on fermented foods and beverages produced by LAB (yogurt, cheese, Menma), yeast (beer, sake, bread), and koji mold (miso, soy source, amazake). Figure 2 shows the endogenous levels of APAs and ALAs in various types of yogurt and in bovine milk (the starting material for yogurt). APA levels in yogurt (a LAB-fermented product) were below 0.6 µg/mL (Fig. 2A), but ALA levels were high (up to 25 µg/mL HPLA) (Fig. 2B), in contrast to milk in which ALAs were hardly detectable. Statistical analysis indicated that the concentration of total ALAs is significantly higher than that of total APAs in yogurts (p = 4.5e-10; Student's t-test, Suppl. Table 2). The ALA/APA ratio for each product was at least 53.5 (Suppl. Table 2). Cheeses produced by various processes were also analyzed as LAB-fermented products (Fig. 3). High levels of ALAs were detected (Fig. 3B), whereas APA levels were negligible (below 1.5 µg/g) (Fig. 3A). The significant difference between the levels of total APAs and total ALAs was confirmed (p = 0.016; Student's t-test, Suppl. Table 2). The ratio ALAs/APAs for each product ranged from 37.9 to 222 (Suppl. Table 2). In particular, Parmesan cheese (designated “cheese E” in Fig. 3) had very high levels of ALAs (240 µg/g PLA and 230 µg/g HPLA), and its levels of IPA (1.0 µg/g) and ILA (53 µg/g) were also high. The results for these Trp metabolites in Parmesan cheese are in sharp contrast to those of all the other analyzed samples, which may be attributable to the long aging period of Parmesan cheese, and/or properties of the enzymes involved in the fermentation process. Because the levels of ALAs in Parmesan cheese were extraordinarily high, the levels in other cheeses did not show significant differences with the levels in milk by Tukey's test. However, it should be noted that the ALAs levels in all the cheeses were significantly higher than those in milk when statistical analysis was performed without the data from Parmesan cheese.
| PPA | HPPA | IPA | PLA | HPLA | ILA | |
|---|---|---|---|---|---|---|
| milk | 1.14E-03 ± 1.60E-04 | 3.48E-01 ± 5.81E-02 | 1.10E-03 ± 1.54E-04 | ND | 6.38E-02 ± 3.47E-02 | 7.93E-03 ± 7.91E-03 |
| yogurt A | 1.55E-02 ± 5.41E-04 | 5.16E-02 ± 2.70E-03 | 6.86E-04 ± 1.49E-04 | 2.02E+00 ± 1.58E-01 | 6.90E+00 ± 2.31E-01 | 1.84E-01 ± 2.49E-02 |
| yogurt B | 1.61E-01 ± 1.32E-02 | 4.99E-01 ± 5.00E-02 | 5.49E-03 ± 5.02E-04 | 1.48E+01 ± 1.17E+00 | 1.98E+01 ± 7.88E-01 | 8.49E-01 ± 7.82E-02 |
| yogurt C | 1.71E-02 ± 1.27E-03 | 5.17E-02 ± 4.67E-03 | 8.21E-04 ± 1.32E-04 | 2.85E+00 ± 4.97E-01 | 7.72E+00 ± 9.63E-01 | 2.32E-01 ± 7.82E-03 |
| yogurt D | 8.35E-03 ± 7.84E-04 | 2.65E-02 ± 3.26E-03 | 9.50E-04 ± 4.82E-05 | 5.26E+00 ± 1.21E-01 | 1.02E+01 ± 3.02E-01 | 4.23E-01 ± 2.58E-02 |
| yogurt E | 1.82E-02 ± 8.44E-04 | 2.76E-02 ± 5.32E-03 | 4.54E-04 ± 9.40E-05 | 2.56E+00 ± 3.73E-02 | 6.17E+00 ± 4.02E-01 | 1.62E-01 ± 7.43E-03 |
| yogurt F | 1.09E-02 ± 9.53E-03 | 7.32E-02 ± 6.44E-02 | 1.33E-03 ± 1.25E-03 | 1.72E+00 ± 5.32E-02 | 1.07E+01 ± 9.58E-02 | 2.71E-01 ± 2.28E-02 |
| yogurt G | ND | ND | ND | 9.44E+00 2.34E-01 | 1.63E+01 4.30E-01 | 7.02E-01 1.16E-02 |
| yogurt H | 3.56E-02 ± 7.93E-03 | 1.75E-01 ± 4.41E-02 | 3.01E-03 ± 9.06E-04 | 1.24E+01 ± 4.28E-01 | 2.36E+01 ± 7.64E-01 | 1.07E+00 ± 7.65E-02 |
| yogurt I | 2.24E-02 ± 7.19E-04 | 1.82E-01 ± 2.60E-02 | 6.34E-03 ± 3.36E-04 | 1.06E+01 ± 4.28E-01 | 1.54E+00 ± 8.63E-02 | 3.30E+00 ± 2.62E-01 |
| milk | 1.14E-03 ± 1.60E-04 | 3.48E-01 ± 5.81E-02 | 1.10E-03 ± 1.54E-04 | ND | 6.38E-02 ± 3.47E-02 | 7.93E-03 ± 7.91E-03 |
| cheese A | 3.91E-02 ± 2.62E-02 | 1.46E-01 ± 6.83E-02 | 1.81E-02 ± 2.65E-02 | 3.93E+00 ± 1.93E+00 | 3.06E+00 ± 1.67E+00 | 3.04E-01 ± 6.95E-02 |
| cheese B | 9.35E-02 ± 9.95E-03 | 9.17E-02 ± 6.07E-02 | 1.96E-03 ± 2.00E-03 | 2.88E+01 ± 2.88E+00 | 9.54E+00 ± 1.06E+00 | 4.65E-01 ± 1.26E-01 |
| cheese C | 7.04E-01 ± 7.80E-01 | 4.98E-01 ± 6.65E-01 | 3.42E-02 ± 1.80E-02 | 1.45E+01 ± 2.20E+00 | 4.41E+00 ± 8.12E-01 | 2.75E-01 ± 1.62E-02 |
| cheese D | 6.64E-01 ± 6.90E-01 | 3.95E-01 ± 3.22E-01 | 3.55E-02 ± 7.19E-03 | 2.12E+01 ± 7.08E-01 | 4.76E+00 ± 3.78E-01 | 5.30E-01 ± 2.76E-02 |
| cheese E | 1.38E+00 ± 1.27E+00 | 6.99E-01 ± 4.08E-01 | 1.03E+00 ± 1.04E-01 | 2.40E+02 ± 6.07E+01 | 2.35E+02 ± 2.72E+01 | 5.34E+01 ± 8.25E+00 |
| cheese F | 1.93E-01 ± 1.73E-02 | 3.66E-01 ± 4.02E-02 | 2.50E-03 ± 2.81E-03 | 2.30E+01 ± 2.13E+00 | 4.86E+00 ± 9.24E-01 | 9.36E-01 ± 2.32E-01 |
| bamboo shoot | 9.30E-02 ± 8.07E-02 | 6.17E-01 ± 5.69E-01 | 1.70E-02 ± 1.47E-02 | 1.15E-01 ± 3.98E-02 | 1.54E-02 ± 2.30E-02 | 8.19E-03 ± 1.13E-02 |
| Menma A | 1.24E-01 ± 9.95E-03 | 2.87E-02 ± 6.06E-03 | 5.37E-03 ± 7.36E-04 | 1.80E+00 ± 2.62E-01 | 5.65E-01 ± 8.81E-02 | 5.93E-02 ± 1.09E-02 |
| Menma B | 6.33E-03 ± 5.48E-03 | 3.92E-03 ± 3.23E-03 | 3.59E-04 ± 3.19E-04 | 1.28E-01 ± 1.09E-01 | 6.79E-02 ± 6.36E-02 | 4.85E-03 ± 3.33E-03 |
| dried fermented bamboo shoot | 1.11E-01 ± 1.92E-02 | 3.96E-02 ± 4.00E-03 | ND | 1.01E+01 ± 7.90E-01 | 6.65E+00 ± 4.66E-01 | 1.36E-03 ± 1.05E-03 |
| barley grain | 5.68E-02 ± 1.18E-03 | 3.49E-02 ± 1.04E-03 | 5.12E-03 ± 6.51E-05 | 1.62E-02 ± 1.08E-03 | 6.08E-01 ± 3.13E-02 | 1.46E-02 ± 1.11E-03 |
| malt | 1.36E-01 ± 5.50E-03 | 1.88E-01 ± 4.81E-03 | 7.67E-03 ± 4.24E-04 | 5.93E-02 ± 1.28E-02 | 1.46E-01 ± 2.79E-02 | 3.81E-02 ± 4.96E-03 |
| beer A | 1.47E-01 ± 4.26E-02 | 2.03E+00 ± 4.72E-01 | 5.12E-03 ± 8.95E-04 | 2.98E+00 ± 9.30E-02 | 4.37E-01 ± 4.48E-02 | 1.04E-01 ± 1.21E-02 |
| beer B | 3.41E-02 ± 2.48E-03 | 1.73E+00 ± 1.29E-01 | 2.56E-03 ± 2.27E-03 | 1.06E+00 ± 4.81E-02 | 3.14E-01 ± 3.60E-02 | 1.26E-01 ± 3.10E-03 |
| beer C | 1.37E-01 ± 9.66E-03 | 3.29E+00 ± 2.36E-01 | 3.45E-03 ± 8.58E-04 | 1.76E+00 ± 9.78E-03 | 3.76E-01 ± 8.82E-02 | 1.16E-01 ± 6.94E-03 |
| beer D | 3.59E-02 ± 4.35E-03 | 2.63E+00 ± 1.87E-01 | 1.10E-02 ± 7.41E-04 | 2.49E+00 ± 2.68E-01 | 3.37E-01 ± 2.01E-02 | 9.11E-02 ± 1.59E-02 |
| rice grain | 3.90E-03 ± 1.10E-03 | 3.87E-03 ± 1.01E-03 | 1.93E-03 ± 2.13E-04 | 2.56E-02 ± 1.35E-02 | 2.16E-02 ± 1.21E-02 | 2.25E-02 ± 8.26E-03 |
| sake A | 3.57E+00 ± 2.30E-01 | 1.01E+01 ± 8.58E-01 | 2.27E-02 ± 1.92E-03 | 2.21E+00 ± 4.65E-01 | 4.99E-01 ± 7.71E-03 | 8.94E-02 ± 3.48E-03 |
| sake B | 1.83E+00 ± 1.82E-01 | 3.69E+00 ± 3.32E-01 | 4.85E-03 ± 8.93E-04 | 1.01E+00 ± 1.36E-01 | 2.64E-01 ± 1.74E-02 | 8.63E-02 ± 3.48E-03 |
| shochu | ND | ND | ND | 2.45E-02 ± 1.20E-03 | 5.46E-02 ± 2.30E-03 | 1.46E-02 ± 8.00E-03 |
| whisky | ND | ND | ND | 3.91E-02 4.26E-02 | 5.07E-02 4.66E-02 | ND |
| flour | 4.93E-02 ± 9.59E-03 | 7.50E-02 ± 1.60E-02 | 8.03E-03 ± 2.35E-03 | 4.03E-01 ± 3.29E-01 | 5.92E-01 ± 2.25E-01 | 2.85E-01 ± 3.26E-02 |
| plain bread | 3.76E+00 ± 2.80E-01 | 5.22E+00 ± 4.75E-01 | 1.40E-02 ± 2.32E-03 | 1.80E+00 ± 3.05E-01 | 1.84E+00 ± 3.04E-01 | 3.45E-01 ± 8.82E-02 |
| rice grain | 3.90E-03 ± 1.10E-03 | 3.87E-03 ± 1.01E-03 | 1.93E-03 ± 2.13E-04 | 2.56E-02 ± 1.35E-02 | 2.16E-02 ± 1.21E-02 | 2.25E-02 ± 8.26E-03 |
| amazake | 1.01E-01 ± 3.65E-03 | 4.96E-01 ± 2.07E-02 | ND | 3.84E-02 ± 5.79E-02 | 5.45E-02 ± 1.47E-02 | 6.06E-02 ± 3.96E-03 |
| soybean (mature) | 2.97E-01 ± 1.92E-02 | 1.98E-01 ± 1.58E-02 | 6.13E-02 ± 1.23E-02 | 1.59E-01 ± 1.98E-02 | 1.36E+00 ± 1.59E-01 | 1.57E-01 ± 1.34E-02 |
| miso A | 4.79E-01 ± 2.78E-01 | 4.76E-01 ± 2.73E-01 | ND | ND | ND | 6.84E-03 ± 1.18E-02 |
| miso B | 1.27E+00 ± 2.53E-01 | 8.02E-01 ± 3.51E-01 | 4.66E-02 ± 3.48E-03 | 2.18E-01 ± 3.77E-01 | ND | 2.20E-02 ± 2.13E-02 |
| soybean | 2.97E-01 ± 1.92E-02 | 1.98E-01 ± 1.58E-02 | 6.13E-02 ± 1.23E-02 | 1.59E-01 ± 1.98E-02 | 1.36E+00 ± 1.59E-01 | 1.57E-01 ± 1.34E-02 |
| flour | 4.93E-02 ± 9.59E-03 | 7.50E-02 ± 1.60E-02 | 8.03E-03 ± 2.35E-03 | 4.03E-01 ± 3.29E-01 | 5.92E-01 ± 2.25E-01 | 2.85E-01 ± 3.26E-02 |
| soy sauce | 2.03E+00 ± 1.58E-01 | 1.24E-01 ± 1.14E-02 | 2.00E-01 ± 1.20E-02 | 7.90E+00 ± 5.95E-01 | 6.58E-01 ± 2.32E-01 | 1.10E-01 ± 3.34E-03 |
| soybean (mature) | 2.97E-01 ± 1.92E-02 | 1.98E-01 ± 1.58E-02 | 6.13E-02 ± 1.23E-02 | 1.59E-01 ± 1.98E-02 | 1.36E+00 ± 1.59E-01 | 1.57E-01 ± 1.34E-02 |
| natto | 9.45E+00 ± 1.23E+00 | 2.25E+00 ± 0.00E+00 | 6.03E-01 ± 2.04E-01 | 8.80E+00 ± 4.70E-01 | 2.01E+00 ± 2.32E-01 | 5.32E-01 ± 0.00E+00 |
| shrimp | 3.01E-01 ± 7.86E-03 | 5.02E+00 ± 1.31E-01 | 9.58E-02 ± 5.90E-04 | 3.13E-03 ± 5.42E-03 | 2.46E-02 ± 3.25E-03 | ND ± |
| tuna | 5.91E-02 ± 6.98E-03 | 4.75E-01 ± 3.22E-02 | 8.27E-03 ± 1.02E-03 | 2.23E-02 ± 5.53E-03 | ND | 1.15E-02 ± 2.08E-03 |
| scallop adductor | 4.44E-01 ± 9.26E-02 | 4.26E+00 ± 1.27E+00 | 1.36E-01 ± 8.16E-02 | ND | 1.69E-03 ± 2.93E-03 | 1.19E-03 ± 2.63E-04 |
| sea bream | 1.25E-01 ± 2.97E-02 | 1.14E+00 ± 2.36E-01 | 2.22E-02 ± 4.25E-03 | 2.40E-02 ± 1.01E-02 | 5.75E-02 ± 4.43E-03 | 8.39E-03 ± 9.70E-04 |
| pork | 7.43E-02 ± 7.45E-03 | 3.27E-02 ± 5.37E-03 | 1.32E-02 ± 8.37E-04 | 2.91E-02 ± 1.12E-03 | 2.60E-02 ± 8.47E-04 | 6.18E-03 ± 6.56E-04 |
| milk | 1.14E-03 ± 1.60E-04 | 3.48E-01 ± 5.81E-02 | 1.10E-03 ± 1.54E-04 | ND | 6.38E-02 ± 3.47E-02 | 7.93E-03 ± 7.91E-03 |
| rice grain | 3.90E-03 ± 1.10E-03 | 3.87E-03 ± 1.01E-03 | 1.93E-03 ± 2.13E-04 | 2.56E-02 ± 1.35E-02 | 2.16E-02 ± 1.21E-02 | 2.25E-02 ± 8.26E-03 |
| flour | 4.93E-02 ± 9.59E-03 | 7.50E-02 ± 1.60E-02 | 8.03E-03 ± 2.35E-03 | 4.03E-01 ± 3.29E-01 | 5.92E-01 ± 2.25E-01 | 2.85E-01 ± 3.26E-02 |
| barley grain | 5.68E-02 ± 1.18E-03 | 3.49E-02 ± 1.04E-03 | 5.12E-03 ± 6.51E-05 | 1.62E-02 ± 1.08E-03 | 6.08E-01 ± 3.13E-02 | 1.46E-02 ± 1.11E-03 |
| soybean (mature) | 2.97E-01 ± 1.92E-02 | 1.98E-01 ± 1.58E-02 | 6.13E-02 ± 1.23E-02 | 1.59E-01 ± 1.98E-02 | 1.36E+00 ± 1.59E-01 | 1.57E-01 ± 1.34E-02 |
| bamboo shoot | 9.30E-02 ± 8.07E-02 | 6.17E-01 ± 5.69E-01 | 1.70E-02 ± 1.47E-02 | 1.15E-01 ± 3.98E-02 | 1.54E-02 ± 2.30E-02 | 8.19E-03 ± 1.13E-02 |
| raw ham | 6.16E-01 ± 1.85E-01 | 5.10E-01 ± 1.81E-01 | 2.18E-01 ± 7.86E-02 | 4.33E-02 ± 5.88E-03 | 5.94E-02 ± 2.22E-03 | 1.12E-02 ± 2.12E-03 |
| brown seaweed | 1.55E-02 ± 1.07E-02 | 1.49E-02 ± 1.02E-02 | 7.73E-04 ± 2.16E-04 | ND | ND | 1.28E-03 ± 1.11E-03 |
| cucumber | 2.98E-01 ± 6.22E-02 | 5.53E-02 ± 3.29E-03 | 9.81E-03 ± 1.21E-03 | 7.26E-03 ± 1.26E-02 | 7.47E-03 ± 1.29E-02 | 3.33E-03 ± 1.93E-03 |
| tomato | 2.28E-02 ± 9.74E-03 | 4.70E-01 ± 2.36E-01 | 1.94E-03 ± 9.44E-04 | 6.85E-03 ± 6.31E-03 | 8.07E-02 ± 1.92E-02 | ND |
| carrot | 9.86E-01 ± 8.11E-02 | 5.74E-01 ± 1.42E-01 | 3.13E-02 ± 3.83E-03 | ND | ND | ND |
| fungus (Grifola frondosa) | 1.31E-01 ± 1.05E-01 | 2.15E+00 ± 1.81E+00 | 3.13E-02 ± 2.63E-02 | 3.45E-02 ± 8.21E-03 | 2.97E-02 ± 3.36E-03 | 7.59E-03 ± 7.31E-04 |
| Japanese radish root | 2.68E+00 ± 2.60E+00 | 9.51E-01 ± 7.07E-01 | 7.44E-03 ± 4.03E-03 | 9.42E-03 ± 1.35E-03 | 0.00E+00 ± 0.00E+00 | 3.11E-03 ± 1.22E-03 |
ND: not detected
| p value (total APAs vs total ALAs) | total ALAs/total APAs | ||
|---|---|---|---|
| yogurt A | 3.48E-04 | *** | 156 |
| yogurt B | 3.81E-04 | *** | 446 |
| yogurt C | 3.04E-03 | *** | 193 |
| yogurt D | 1.25E-04 | *** | ∞ |
| yogurt E | 3.88E-04 | *** | 135 |
| yogurt F | 5.89E-05 | *** | 53.5 |
| yogurt G | 9.24E-06 | *** | ∞ |
| yogurt H | 6.16E-05 | *** | 182 |
| yogurt I | 2.71E-04 | *** | 74.2 |
| yogurts A∼I | 4.52E-10 | *** | 123 |
| cheese A | 1.40E-03 | *** | 38.9 |
| cheese B | 1.81E-03 | *** | 222 |
| cheese C | 7.89E-03 | *** | 37.9 |
| cheese D | 3.05E-04 | *** | 38.7 |
| cheese E | 5.32E-03 | *** | 202 |
| cheese F | 1.61E-03 | *** | 51.6 |
| cheeses A∼F | 1.63E-02 | ** | 101 |
| dried fermented bamboo shoot | 7.79E-04 | *** | 113 |
| beer A | 2.14E-02 | ** | 1.67 |
| beer B | 1.37E-02 | ** | 0.848 |
| beer C | 1.27E-02 | ** | 0.661 |
| beer D | 3.13E-02 | ** | 1.09 |
| beers A∼D | 4.54E-01 | 1.01 | |
| sake A | 2.01E-03 | *** | 0.205 |
| sake B | 3.60E-03 | *** | 0.248 |
| sake A∼B | 2.32E-03 | *** | 0.216 |
Figure 4 shows the endogenous levels of AAA metabolites in Menma (LAB-fermented bamboo shoots). We analyzed two commercially available Menma (Menma A and B), fresh bamboo shoots, and dried fermented bamboo shoots (imported starting material for Menma). The dried fermented bamboo shoots showed relatively high levels of ALAs, while fresh bamboo shoots showed negligible amounts of ALAs (Fig. 4). Menma is manufactured by boiling dried fermented bamboo shoots repeatedly to rehydrate them and thus the observed low concentrations of APAs and ALAs in Menma could be attributable to this rehydration process. The level of total ALAs was significantly higher than that of total APAs in dried fermented bamboo shoot (p = 0.7e-3; Student's t-test, Suppl. Table 2) and the ratio ALAs/APAs was 113 (Suppl. Table 2).
Collectively, foods fermented by LAB contained much higher levels of ALAs than their corresponding starting materials, and the APA levels were much lower than the ALA levels. These results strongly suggest that LAB are involved in AAA metabolism. Specifically, strong lactate dehydrogenase (LDH) activity in LAB could be responsible for the high concentrations of ALAs and relatively low concentrations of APAs.
Figure 5 shows the endogenous levels of APAs and ALAs in beer, sake, and bread, which are fermented using yeast, and their corresponding starting materials. In general, products fermented by yeast (beer, sake) were relatively rich in APAs as contrasted with LAB-fermented foods, which contained negligible amounts of APAs. HPPA was the only relatively abundant APA in beer. The respective starting materials, mature barley grain or geminated seeds (malts) for beer, rice grains for sake, and wheat flour for bread, contained low amounts of APAs and very low levels of ALAs (up to 3 µg/mL or 3 µg/g). Distilled alcoholic beverages (shochu and whiskey) contained no AAA metabolites. These results strongly suggest that yeast is involved in the production of APAs in yeast-fermented products, and APAs are converted metabolically to ALAs at a low rate in yeast.
Fermentation by koji mold modestly increased the concentrations of APAs, with the highest level of APAs being 2 µg/g (Fig. 6). The levels of ALAs were also low, with the exception of PLA at 7.9 µg/mL in soy sauce. On the other hand, natto was rich in both APAs, and ALAs when compared with its starting material, soybean (Fig. 7).
In addition to the above fermented foods, scallop adductor and shrimp showed relatively high levels of HPPA, and the levels of PPA in Japanese radish root and HPPA in fungus were comparable. On the other hand, the levels of ALAs were negligible, except in flour, barley grain, and soybean (Fig. 8).
The above results strongly suggest that fermenting microbes metabolize AAAs during fermentation to form APAs and/or ALAs. The observed high relative ratios of ALAs to APAs in LAB-fermented foods also suggest that LAB efficiently convert AAAs to APAs, and APAs to ALAs, resulting in the accumulation of ALAs. In contrast, yeast actively converts AAAs to APAs but poorly further convert APAs to ALAs, resulting in higher levels of APAs than ALAs. We confirmed these metabolizing traits of LAB and S. cerevisiae using standard experimental culture conditions. LAB transiently accumulated APAs after 6–10 h of culture, then the APA levels decreased (Fig. 9, see Suppl. Table 3 for the results of statistical analysis). On the other hand, cultures of most LAB strains showed a continuous increase in ALAs up to 24 h and maintained high ALA levels up to 48 h (Fig. 10, Suppl. Table 3). The concentrations of ALAs were much higher than those of APAs. For example, Pediococcus acidilactici D57 accumulated 41.7 µg/mL PLA after 48 h of culture (Fig. 10A), whereas the highest level of any APA was only 112 ng/mL (for PPA after 6 h of culture) (Fig. 9A). The changes with time of ALA and APA concentrations in LAB culture support the above notion that LABs extensively metabolize AAAs to ALAs via APAs. The increase in ALAs was relatively slow for the initial 6 hours of incubation, which is the period of transient increase in APA concentrations, but the ALA concentrations continued to increase to very high levels thereafter. These results suggest that the LDH involved in the conversion of APAs to ALAs is induced during the initial period of incubation, presumably by its APA substrates.
Time course of the concentrations of APAs in the culture media of LAB (A, Pediococcus acidilactici D57; B, Lactobacillus plantarum P4; C, Lactobacillus rhamnosus L-52; D, Lactobacillus paraplantarum TP2). (n = 3, mean ± SD).
 |
 |
 |
Time course of the concentrations of ALAs in the culture media of LAB (A, Pediococcus acidilactici D57; B, Lactobacillus plantarum P4; C, Lactobacillus rhamnosus L-52; D, Lactobacillus paraplantarum TP2). (n = 3, mean ± SD).
In contrast to LAB, S. cerevisiae accumulated relatively high levels of APAs (Fig. 11, Suppl. Table 3). The concentrations of HPPA and PPA in cultures of NCYC 235 strain increased up to about 1.0 and 2.1 µg/mL during a 3-d incubation, respectively (Fig. 11A), whereas of the ALAs investigated, only PLA increased after inoculation, from 1.4 to 3.6 µg/mL (Fig. 11C). The level of HPLA decreased during incubation. Strain X2180-1A provided similar time courses and levels of APAs, while it showed only scarce increases in ALAs. Comparable amounts of APAs and ALAs in the culture media are in accordance with their levels in yeast-fermented products. These results strongly suggest that metabolism of AAAs by S. cerevisiae is responsible for the production of APAs in yeast-fermented products.
Temporal fluctuation of APA (A, B) and ALA (C, D) concentrations in the culture media of S. cerevisiae (A and C, NCYC 235; B and D, X2180-1A). (n = 3, mean ± SD).
The present study indicates the importance of fermentation for increasing the amounts of AAA metabolites that exhibit beneficial functions. Given the differences between the productivities of LAB and yeast for generating APAs and ALAs, it is important to know the metabolizing traits of each fermentation microorganism when developing higher functional value food products.
Acknowledgements We would like to thank Momoya Co., Ltd., for providing the dried fermented bamboo shoots. This work was supported in part by the Cross-ministerial Strategic Innovation Promotion Program (SIP) (ID: 14532924) “Technologies for creating next-generation agriculture, forestry and fisheries” from the Council for Science, Technology and Innovation (CSTI), Japan.
