Simple and Sensitive Analysis of Histamine and Tyramine in

Jun 5, 2014 - Japanese Soy Sauces and Their Intermediates Using the Stable. Isotope Dilution HILIC−MS/MS Method. Kenichiro Todoroki,. †. Yasuhiro ...
0 downloads 0 Views 284KB Size
Article pubs.acs.org/JAFC

Simple and Sensitive Analysis of Histamine and Tyramine in Japanese Soy Sauces and Their Intermediates Using the Stable Isotope Dilution HILIC−MS/MS Method Kenichiro Todoroki,† Yasuhiro Ishii,† Chiemi Miyauchi,† Sachiyo Kitagawa,‡ Jun Zhe Min,† Koichi Inoue,† Tomoyuki Yamanaka,‡ Kuniaki Suzuki,‡ Yuko Yoshikawa,§ Norio Ohashi,§ and Toshimasa Toyo’oka*,† †

Laboratory of Analytical and Bio-Analytical Chemistry, School of Pharmaceutical Sciences, University of Shizuoka, 52-1 Yada, Suruga, Shizuoka 422-8526, Japan ‡ Shizuoka Kensan Shoyu K. K., 80-1 Takashinden, Yaizu, Shizuoka 421-0204, Japan § Laboratory of Microbiology, School of Food and Nutritional Sciences, University of Shizuoka, 52-1 Yada, Suruga, Shizuoka 422-8526, Japan ABSTRACT: We established a simple, sensitive, and reproducible method to analyze the histamine and tyramine levels in Japanese soy sauce and its mash (called moromi) using hydrophilic interaction liquid chromatography with tandem mass spectrometry (HILIC−MS/MS). Histamine and tyramine quantification was performed using their stable isotopes for electrospray ionization−tandem mass spectrometry in the selected reaction monitoring mode. The sample pretreatment process was a simple, one-step liquid−liquid extraction. HILIC separation was performed with a gradient elution of aqueous ammonium formate and acetonitrile. Because of validation tests, the linearity, the accuracies, and precisions were sufficient. The limit of detection and the limit of quantification were 0.09 and 0.29 ppm for histamine and 0.13 and 0.42 ppm for tyramine, respectively. We successfully applied this method to histamine and tyramine determination in four kinds of commercial Japanese soy sauces and also in moromi samples during soy sauce production. KEYWORDS: histamine, tyramine, high-performance liquid chromatography, HILIC−MS/MS, soy sauce, moromi



INTRODUCTION Soy sauces are fermented condiments produced from soybeans, grain, brine, and Aspergillus oryzae or Aspergillus sojae molds. In the Japanese traditional soy sauce manufacturing process, the microbial fermentation occurs in soy sauce mash (commonly called moromi). Several aromatic components of soy sauces are generated during this process. Lactic acid is produced by halophilic lactic acid bacteria (Tetragenococcus spp., primarily Tetragenococcus halophilus) to acidify the moromi; alcoholic fermentation is subsequently caused by yeast (Zygosaccharomyces rouxii). Histamine (Him) and tyramine (Tym) are produced during the lactic acid fermentation in the mash by the Tetragenococcus spp. having histidine decarboxylase (HDC) and/or tyrosine decarboxylase (TDC) genes. The HDC gene is encoded on a 30-kb plasmid, and the plasmid can be transferred among Tetragenococcus bacteria, and the bacterium that acquire the transformable element become histamine-productive.1 Excessive intake of Him and Tym can cause an allergy-like food poisoning.2 Therefore, the maximum acceptable Him levels in foods has been established as less than 200 ppm in European Union (EU) countries, Switzerland, and Norway.3−5 Although Him and Tym levels in foods have not yet been legally regulated in Japan, they will be regulated as in the EU in the near future. To manufacture soy sauces with low Him and/ or Tym levels, it is necessary to use Tetragenococcus spp. without HDC and/or TDC genes during the lactic acid fermentation of the soy sauce mash. Therefore, a simple and © XXXX American Chemical Society

sensitive monitoring method for Him and Tym during the production process of soy sauce is required. Biogenic amines, including Him and Tym, analysis in foods by LC coupled in tandem with mass spectrometry (MS) using an ODS column6−8 and a hydrophilic interaction liquid chromatography (HILIC) column9−11 have been widely reported. In the HILIC mode, analytes interact with the hydrophilic stationary phase, and the elution is generated by the hydrophobic binary mobile phase, which uses water as a strong eluting solvent; therefore, polar substances such as Him and Tym are strongly retained to minimize the potential endogenous interferences and ion suppression. Unlike other food samples, however, soy sauces contain high concentrations of sodium chloride and many stained compounds such as melanoidins.12 Time consuming pretreatments, such as ion-pair or cation-exchange solid-phase extraction, are required to remove these interfering compounds. Nevertheless, these pretreatments could not sufficiently prepare samples from soy sauce to be introduced into the mass spectrometer. In this study, we established an HILIC−MS/MS method to analyze Him and Tym in soy sauces and their intermediates. This method enables highly selective Him and Tym detection, Received: February 16, 2014 Revised: June 2, 2014 Accepted: June 5, 2014

A

dx.doi.org/10.1021/jf500767p | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

Figure 1. MS spectra of the 100 ppm standards of Tym, Tym-d4, Him and Him-d4. Sample Collection. Four kinds of Japanese soy sauces (koikuchi, usukuchi, saishikomi, and tamari)13 were purchased from supermarkets in Shizuoka city (Japan). Two moromi samples after lactic acid fermentation with pH 4.68 and pH 5.04 were obtained from Shizuoka Kensan Shoyu K. K. (Yaizu, Shizuoka, Japan). In addition, an additional soy sauce mash was prepared as an example of lower contamination with Him and Tym after lactic acid fermentation by a T. halophilus 52−47C strain that was isolated from a soy sauce mash naturally fermented by a procedure previously described.14 Polymerase chain reaction (PCR) was performed on the T. halophilus 52−47C strain in advance to confirm the absence of the genes HDC and TDC before starting the lactic acid fermentation of the soy sauce mash. PCR reactions were performed using primers of HmF (5′-TGTTTCGTATGACCGTGCGG-3′) and HmR (5′-CACCATTTTCGCCGGCAGTG-3′) for a partial 560 bp HDC gene1 and using primers of TD5 (5′-CAAATGGAAGAAGAAGTAGG-3′) and TD2 (5′-ACATAGTCAACCATRTTGAA-3′) for a 1100 bp TDC gene.15 Instrumentation. Liquid chromatography was performed using a Shimadzu (Kyoto, Japan) Nexera system coupled with a Shimadzu LCMS 8040 triple quadrupole mass spectrometer with an ESI interface. Data acquisition was controlled using the LabSolution LCMS software ver. 5.53 (Shimadzu). Chromatographic Conditions. The analytical column used was an Aeris HILIC column (2.1 mm i.d. × 150 mm, 5 μm, Phenomenex, Torrance, CA) maintained at 40 °C; the mobile phase consisted of 10 mM ammonium formate in water (solvent A) and a mixture of acetonitrile and 100 mM ammonium formate (9:1) (solvent B). The gradient profile was as follows: 85% solvent B at 0 min, 85% solvent B at 2.5 min, 50% solvent B at 4 min, 50% solvent B at 6 min, 85%

and the sample pretreatment process has been simplified to a one-step liquid−liquid extraction (LLE). For Him and Tym quantification, the stable isotopes Him-d4 and Tym-d4 were used for the electrospray ionization−tandem mass spectrometry (ESI−MS/MS) in the selected reaction monitoring (SRM) mode. The developed method was validated with regard to linearity, sensitivity, precision, accuracy, and recovery. Furthermore, we successfully applied this method to sensitive and selective Him and Tym analysis in soy sauce and its moromi samples after lactic acid fermentation during soy sauce production.



MATERIAL AND METHODS

Reagents and Solutions. Deionized and distilled water purified using an Aquerius pwu-200 automatic water distillation apparatus (Advantec, Tokyo, Japan) was used to prepare all aqueous solutions. LC grade acetonitrile was purchased from Honeywell Burdick & Jackson (Muskegon, MI). Histamine dihydrochloride, tyramine hydrochloride, magnesium sulfate, and 1 M sodium hydroxide solution were purchased from Wako Pure Chemicals (Osaka, Japan). Histamine-d4 sodium salt and tyramine-d4 were purchased from Cambridge Isotope Laboratories (Andover, MA) and CDN Isotopes (Quebec, Canada), respectively. Other chemicals and solvents were of the highest purity available and were used as received. Individual stock solutions (1000 ppm) were prepared by dissolving 5 mg of each compound in 5 mL of water and stored in polypropylene test tubes at 4 °C. Various concentrations of standard working solutions were prepared daily by appropriate dilution with water just prior to use. B

dx.doi.org/10.1021/jf500767p | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

solvent B at 6.01 min, and 85% solvent B at 12 min with a flow rate of 0.4 mL/min. The total run time was 12 min. MS Conditions. The mass spectrometer was operated in the SRM mode using positive electrospray ionization. The parameters for the SRM analysis were as follows: drying N2 gas flow at 15.0 L/min, nebulizing Ar gas flow at 2.0 L/min, desolvation line temperature at 250 °C, heat block temperature at 400 °C, interface voltage of 4.5 kV, and collision gas flow of 230 kPa. SRM transition of the precursor to product ions used for quantification were as follows: m/z 112.0 → 95.1 for Him, m/z 116.0 → 99.1 for Him-d4, m/z 138.0 → 121.0 for Tym, and m/z 142.0 → 125.0 for Tym-d4. The collision energy was 18 eV for Him and Him-d4 and 14 eV for Tym and Tym-d4. Sample Preparation. A mixture of 100 ppm of Him-d4 and Tymd4 (20 μL, for internal standard), 1 M sodium hydroxide (20 μL) and 100 mM magnesium sulfate (20 μL) were added to a 20-μL portion of each sample. After the addition of acetonitrile (320 μL), the solution was vigorously shaken for 1 min. An aliquot (20 μL) of the upper layer was transferred to another tube and then diluted 50 times with the initial mobile phase. The resulting solution (1 μL) was applied to the LC−MS/MS system. Method Validation. To obtain the validation parameters, the peak areas were estimated by the LabSolution LCMS, and the baseline-tobaseline method was used for quantification. Him and Tym concentrations were calculated by the internal standard method. For quantitative analysis, calibration standard solutions (n = 6 each) with concentrations ranging from 0.5 to 500 ppm (0.5, 1, 5, 10, 50, 100, and 500 ppm) were prepared by diluting the stock solutions. The equations of the calibration curves were determined using least-squares linear prediction. Precision of the assays was determined by the repeated measurement of four spiked samples at the lowest, low, middle, and high levels (0.5, 10, 100, and 500 ppm; n = 6). For intraday precision, these levels were analyzed six times each day, whereas for interday precision, specimens of the spiked samples at the same concentrations were analyzed twice per day for 6 days (n = 12). The spiked samples were pretreated and subsequently subjected to the HILIC−MS/MS. Him and Tym levels were determined from each calibration curve. Accuracy was defined as F/(F0 + A) × 100 (%), where F is the level of Him and Tym in the spiked sample, F0 is the level of Him and Tym in the unspiked sample, and A is the spiked concentration.

Tym were observed between the peak area and analyte concentration with good correlation coefficients (r2 > 0.999). Limit of detection (LOD) and limit of detection (LOQ) were determined from the signal-to-noise ratios of 3 and 10, respectively. LOD of Him and Tym was 0.09 and 0.13 ppm, respectively, LOQ of Him and Tym was 0.29 and 0.42 ppm, respectively. When analyzing the Him and Tym standard solution without the pretreatment process, 1000th of the above values were obtained as LOD and LOQ by the same HILIC− MS/MS system. The results of precision and accuracy of intraday and interday assays are listed in Tables 1 and 2. The intraday assay precisions Table 1. Precision and Accuracy of Intra-day Assay for Him and Tym in Soy Sauce Samples (n = 6) nominal concentration (ppm) QC QC QC QC

+ + + +

0.5 ppm (2.53) 10 ppm (12.03) 100 ppm (102.23) 500 ppm (502.03)

QC QC QC QC

+ + + +

0.5 ppm (1.78) 10 ppm (11.29) 100 ppm (101.29) 500 ppm (501.29)

precisionb (%)

accuracyc (%)

ppm ppm ppm ppm

5.57 2.09 1.55 1.11

95.1 92.2 98.5 98.6

ppm ppm ppm ppm

5.26 2.28 1.25 1.45

85.9 88.7 93.0 94.3

observed concentration (ppm) Him 2.42 ± 0.13 11.1 ± 0.23 100.5 ± 1.56 495.0 ± 5.50 Tym 1.54 ± 0.08 10.0 ± 0.23 94.2 ± 1.18 472.6 ± 6.83

a

n = 6 per day. bExpressed as relative standard deviation. cAccuracy % = Average of Observed concentration/(nominal concentration + concentration of QC sample) × 100.

Table 2. Precision and Accuracy of Inter-day Assay for Him and Tym in Soy Sauce Samples (n = 12)a nominal concentration (ppm)



RESULTS AND DISCUSSION MS Analysis of Him and Tym. Figure 1 shows SRM chromatograms of the 100 ppm standard for Him and Tym and their MS spectra. We could observe ions regarding the major fragment ions at m/z = 138.0 → 121.0 (Tym), 142.0 → 125.0 (Tym-d4), 112.0 → 95.0 (Him), and 116.0 → 99.0 (Him-d4). The retention times of Tym, Tym-d4, Him, and Him-d4 were 1.6, 1.6, 5.4, and 5.4 min, respectively. Him and Him-d4 as well as Tym and Tym-d4 were eluted at almost the same retention times. Quantitative separation of Him and Tym was achieved within 6 min on the HILIC column at the given chromatographic conditions, and selective detection of Him and Tym was also achieved with the SRM mode. Sample Pretreatment. Japanese soy sauces currently on the market are weakly acidic and contain approximately 16% sodium chloride. For the pretreatment process, soy sauce samples were alkalinized to a pH of ca. 10 by the addition of 1 M sodium hydroxide. These solutions were subsequently added to 100 mM magnesium sulfate and acetonitrile to form a bilayer. Water-soluble compounds, neutral and acidic amino acids, proteins, and melanoidins all remained in the aqueous layer. In contrast, positively charged, undissociated Him and Tym successfully partitioned into the acetonitrile phase. Method Validation. Seven different concentrations in the range 0.5−500 ppm with six replicates were chosen to draw each calibration curve. The calibration curves for both Him and

QC QC QC QC

+ + + +

0.5 ppm (2.53) 10 ppm (12.03) 100 ppm (102.23) 500 ppm (502.03)

QC QC QC QC

+ + + +

0.5 ppm (1.78) 10 ppm (11.29) 100 ppm (101.29) 500 ppm (501.29)

precisionb (%)

accuracyc (%)

ppm ppm ppm ppm

7.17 6.50 1.76 2.24

97.8 103.8 102.1 103.2

ppm ppm ppm ppm

9.72 4.66 5.52 7.09

96.6 100.2 99.2 101.5

observed concentration (ppm) Him 2.48 ± 0.18 12.5 ± 0.81 104.2 ± 1.84 518.0 ± 11.6 Tym 1.73 ± 0.17 11.3 ± 0.52 100.5 ± 5.54 508.8 ± 36.1

n = 12 (2 × 6 days). bExpressed as relative standard deviation. Accuracy % = Average of observed concentration/(nominal concentration + concentration of QC sample) × 100. a c

of Him and Tym were in the range 1.11−5.57% and 1.25− 5.76%, respectively; the interassay precisions were in the range 1.76−7.17% and 4.66−9.72%, respectively. Accuracy of the method was determined by assessing the agreement between the observed and nominal concentrations of the analyzed samples. These values for Him and Tym were in the range 85.9−98.6% and 96.6−103.8%, respectively. Recovery studies for the method were performed by spiking seven concentrations (0.5, 1, 5, 10, 50, 100, and 500 ppm) of Him and Tym into soy sauce samples. The recoveries were calculated from the slope of the calibration curve ratio of the peak areas of spiked soy sauce samples and those of standards C

dx.doi.org/10.1021/jf500767p | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

Figure 2. Typical SRM chromatograms obtained from commercially available Japanese soy sauces: (a) koikuchi, (b) usukuchi (10-fold diluted), (c) saishikomi, and (d) tamari. Upper, 100 ppm internal standards; lower, soy sauce samples.

(n = 3 each). Mean recoveries of Him and Tym were 47.7 and 59.9%, respectively. Him and Tym Determination in Commercial Japanese Soy Sauces. This method was used for Him and Tym quantification in commercial Japanese soy sauces to investigate its applicability. Figure 2 shows typical SRM chromatograms of four kinds of commercial soy sauce samples: (a) koikuchi, (b) usukuchi, (c) saishikomi, and (d) tamari. The koikuchi, usukuchi, and tamari-type soy sauces are produced from moromi, which is fermented with rice malt and brine, whereas nama shoyu (raw soy sauce) is used in the production of the saishikomi soy sauce instead of brine. The nama shoyu is rich in color, taste, aroma, and nitrogen contents, and it may contain significant amounts of Him and Tym. Results of the Him and Tym levels obtained by the present method were summarized in Table 3. Him levels of the commercial soy sauce samples used were 1.2−775 (koikuchi), 1.9−364 (usukuchi), 46.2−173.9 (saishikomi), and 131 ppm (tamari). Tym levels of the above samples were 1.2− 2040 (koikuchi), 4.3−1594 (usukuchi), 171−575 (saishikomi), and 336 ppm (tamari). Each sample was purchased from a different manufacturer; therefore, Him and Tym levels were significantly different because of differences in the manufacturers. Him and Tym Analysis in Moromi Samples after Lactic Acid Fermentation. Figure 3 shows the SRM chromatograms of three moromi samples after lactic acid fermentation during the manufacturing process of koikuchi soy sauces. When the soy sauce mash was fermented at pH 4.68, 183.4 ppm of Him and 108.7 ppm Tym were detected (Figure 3a). When the mash was fermented at a slightly higher pH (5.04), 116.7 ppm of

Table 3. Him and Tym Contents in Commercial Japanese Soy Sauces concentration (ppm) sample no.

kind of soy sauce

Tym

Him

1 2 3 4 5 6 7 8 9 10 11 12 13

koikuchi (bland A) koikuchi (bland B) koikuchi (bland C) koikuchi (bland D) koikuchi (bland E) koikuchi (bland F) usukuchi (bland B) usukuchi (bland G) usukuchi (bland H) saishikomi (bland B) saishikomi (bland I) saishikomi (bland G) tamari (bland J)

2.0 234.5 14.1 32.6 774.6 1.2 363.7 1.9 11.1 62.5 173.9 46.2 131.1

1.2 277.0 29.0 134.3 2040 11.4 1594 17.5 4.3 171.2 575.1 258.3 336.0

Him and 150.3 ppm Tym were observed (Figure 3b). Thus, lactic acid fermentations under different pHs appears to influence the production quantities of Him and Tym; this probably depends on the growth of contaminated Tetragenococcus spp. that contains HDC and TDC genes during the fermentation, although no significant difference was observed in the Him and Tym levels between the two pH samples. We further analyzed Him and Tym levels in a soy sauce mash after lactic acid fermentation using a T. halophilus 52−47C strain that did not have HDC and TDC genes. Extremely low Him and Tym levels were detected (less than 8.29 and 14.3 D

dx.doi.org/10.1021/jf500767p | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

Figure 3. Chromatograms of three moromi samples after the lactic fermentation manufacturing process of koikuchi soy sauces. (a) Lactic acid fermentation at pH 4.68; (b) lactic acid fermentation at pH 5.04; (c) lactic acid fermentation using a T. halophilus 52−47C strain without HDC and TDC genes. Upper, 100 ppm internal standards; lower, moromi samples.

ppm, respectively) in the mash (Figure 3c), suggesting that T. halophilus without HDC and TDC genes available in lactic acid fermentation can be used to manufacture soy sauce with minimal Him and Tym contamination. As mentioned above, our method would be useful for monitoring Him and Tym contaminations during the manufacturing of soy sauce.



various kinds of foods and beverages beside Japanese soy sauces, such as fish sauces, seafood, and lactic fermented foodstuffs.



AUTHOR INFORMATION

Corresponding Author

*Tel: +81-54-264-5656. Fax: +81-54-264-5593. E-mail: [email protected].

CONCLUSION

Notes

The authors declare no competing financial interest.



This paper reports a simple and sensitive analysis method for Him and Tym in Japanese soy sauce and moromi samples. The total run time was satisfactorily short for the routine analyses. The sample pretreatment procedure was also simplified to a one-step LLE without concentration of the extract. In this procedure, solid-phase extraction cartridge was not necessary, and general-use reagents, which are magnesium sulfate, sodium hydroxide and acetonitrile, were only required for the LLE. Therefore, it would suppress the analytical cost and be useful in routine analyses. The developed method was validated and successfully applied for Him and Tym quantification in the four kinds of Japanese commercial soy sauces and moromi samples. The proposed method would be useful as a simple assay method for soy sauce manufacturers, quality control units in food companies, and related analytical laboratories. Our simple pretreatment procedure enabled the extraction of Him and Tym from high-salting and protein-containing matrices selectively and efficiently. In addition, the HILIC−MS/MS analysis provided simple and clean SRM chromatograms for Him and Tym with high sensitivity. Therefore, our proposed method would be applicable to Him and Tym analysis in

ACKNOWLEDGMENTS This work was supported in part by Shizuoka Foundation for Industry Promotion. The authors greatly thank to Shimadzu Corporation for useful technical assistance regarding an LC− MS/MS instrument.



REFERENCES

(1) Satomi, M.; Furushita, M.; Oikawa, H.; Yoshikawa-Takahashi, M.; Yano, Y. Analysis of a 30 kbp plasmid encoding histidine decarboxylase gene in Tetragenococcus halophilus isolated from fish sauce. Int. J. Food Microbiol. 2008, 126, 202−209. (2) Taylor, S. L. Histamine food poisoning - Toxicology and clinical aspects. CRC Cr. Rev. Toxicol. 1986, 17, 91−128. (3) Directive, E. U. Council Directive 91/493/EEC of 22 July 1991 laying down the health conditions for the production and the placing on the market of fishery products. Official Journal L 268.24/09, 1991, 0015−0034. (4) Fish, F. D. A. Fishery Products Hazards Controls Guide 1998. (5) Lehane, L.; Olley, J. Histamine fish poisoning revisited. Int. J. Food Microbiol. 2000, 58, 1−37. E

dx.doi.org/10.1021/jf500767p | J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry

Article

(6) Ö nal, A. A review: Current analytical methods for the determination of biogenic amines in foods. Food Chem. 2007, 103, 1475−1486. (7) Romero-Gonzalez, R.; Alarcon-Flores, M. I.; Vidal, J. L. M.; Frenich, A. G. Simultaneous determination of four biogenic and three volatile amines in anchovy by ultra-high-performance liquid chromatography coupled to tandem mass spectrometry. J. Agric. Food Chem. 2012, 60, 5324−5329. (8) Mayr, C. M.; Schieberle, P. Development of stable isotope dilution assays for the simultaneous quantitation of biogenic amines and polyamines in foods by LC-MS/MS. J. Agric. Food Chem. 2012, 60, 3026−3032. (9) Gianotti, V.; Chiuminatto, U.; Mazzuccoa, E.; Gosetti, F.; Bottaro, M.; Frascarolo, P.; Gennaro, M. C. A new hydrophilic interaction liquid chromatography tandem mass spectrometry method for the simultaneous determination of seven biogenic amines in cheese. J. Chromatogr., A 2008, 1185, 296−300. (10) Kakigi, Y.; Yamashita, A.; Miyamoto, Y.; Icho, T.; Mochizuki, N. Quantification of non-volatile amines in beer using hydrophilic interaction liquid chromatograph-tandem mass spectrometer. Bunseki Kagaku 2011, 60, 157−162. (11) Self, R. L.; Wu, W. H.; Marks, H. S. Simultaneous quantification of eight biogenic amine compounds in tuna by matrix solid-phase dispersion followed by HPLC−orbitrap mass spectrometry. J. Agric. Food Chem. 2011, 59, 5906−5913. (12) Obretenov, T.; Vernin, G. Melanoidins in the Maillard reaction. Dev. Food Sci. 1998, 40, 455−482. (13) Lioe, H. N.; Wada, K.; Aoki, T.; Yasuda, M. Chemical and sensory characteristics of low molecular weight fractions obtained from three types of Japanese soy sauce (shoyu)−Koikuchi, tamari and shiro shoyu. Food Chem. 2007, 100, 1669−1677. (14) Justé, A.; Van Trappen, S.; Verreth, C.; Cleenwerck, I.; De Vos, P.; Lievens, B.; Willems, K. A. Characterization of Tetragenococcus strains from sugar thick juice reveals a novel species, Tetragenococcus osmophilus sp. nov., and divides Tetragenococcus halophilus into two subspecies, T. halophilus subsp. nov. and T. halophilus subsp. f landriensis subsp. nov. Int. J. Syst. Evol. Micr. 2012, 62, 129−137. (15) Coton, M.; Coton, E.; Lucas, P.; Lonvaud, A. Identification of the gene encoding a putative tyrosine decarboxylase of Carnobacterium divergens 508. Development of molecular tools for the detection of tyramine-producing bacteria. Food Microbiol. 2004, 21, 125−130.

F

dx.doi.org/10.1021/jf500767p | J. Agric. Food Chem. XXXX, XXX, XXX−XXX