- Email: [email protected]
Effects of different fermentation temperatures on metabolites of Kimchi
Author’s Accepted Manuscript Effects of different fermentation temperatures on metabolites of Kimchi Seong-Eun Park, Seung-Ho Seo, Eun-Ju Kim, Chang-Su Na, Hong-Seok Son www.elsevier.com/locate/sdj
PII: DOI: Reference:
S2212-4292(17)30822-2 https://doi.org/10.1016/j.fbio.2018.03.009 FBIO290
To appear in: Food Bioscience Received date: 3 November 2017 Revised date: 19 March 2018 Accepted date: 24 March 2018 Cite this article as: Seong-Eun Park, Seung-Ho Seo, Eun-Ju Kim, Chang-Su Na and Hong-Seok Son, Effects of different fermentation temperatures on metabolites of Kimchi, Food Bioscience, https://doi.org/10.1016/j.fbio.2018.03.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Effects of different fermentation temperatures on metabolites of Kimchi Seong-Eun Park, Seung-Ho Seo, Eun-Ju Kim, Chang-Su Na, Hong-Seok Son* Department of Korean Medicine, Dongshin University, Naju, Jeonnam 58245, Republic of Korea * Corresponding author. Tel.: +82 61 330 3513; Fax: +82 61 330 3519. [email protected] (H.-S. Son).
Abstract The influences of fermentation temperature on metabolic changes in Kimchi using a gas chromatography–mass spectrometry (GC-MS) method were studied. Kimchi mixtures fermented at 4, 12, and 20°C were taken on the 1st, 5th, 10th, and 50th day of fermentation to determine overall changes in metabolites of Kimchi during fermentation. Based on the principal component analysis (PCA) score plot, Kimchi samples were confirmed to have separation trends by the first principal component 1 (PC1). One group included all samples on fermentation day 0 and day 1, and samples fermented at 4°C for up to 10 days after fermentation, demonstrating that metabolites of Kimchi fermented at 4°C changed slowly. The partial least squares discriminant analysis (PLS-DA) score plot also showed clear differences in metabolites among Kimchi samples with fermentation temperature. The changing metabolites were identified to be alanine, propylene glycol, fumaric acid, malic acid, citric acid, and galactaric acid. These results highlight that a GC-MS-based metabolomics approach can be used to monitor distinct metabolite changes in Kimchi with fermentation temperature.
Keywords: Kimchi, Chinese cabbage, fermentation, metabolomics, GC-MS
1
1. Introduction Kimchi is Korea's representative ethnic food. It is a unique food made by fermenting vegetables (such as Chinese cabbage and radish), fish sauce, and spices (such as garlic and ginger) (Jang et al., 2015). According to the Korean Nutrition survey, Korean adults consume 50-200 g/day/person, accounting for 12.5% of their total daily food intake (Mheen and Kwon, 1984). Kimchi is usually manufactured by spontaneous fermentation without using a starter culture. Lactic acid bacteria (LAB) such as Lactobacillus, Leuconostoc, Weissella, Lactococcus, and Pediococcus are involved in Kimchi fermentation (Park et al., 2003). Recently, many studies have determined the dynamics of microbial communities in Kimchi using a metagenomics approach (Jeong et al., 2013; Jung et al., 2012; Jung et al., 2013; Park et al., 2003). Microbial communities in Kimchi during its fermentation are affected by several factors, including raw material (Chinese cabbage), temperature, pH, salt concentration, and fermentation period (Kim et al., 2017; Lee et al., 2017). Among these, fermentation temperature is a critical factor for determining its final quality. To maintain the quality of Kimchi and store it for a long time, controlling fermentation rate through temperature regulation is well-known. Many previous studies have reported the relationship between fermentation temperature and Kimchi quality. Mheen and Kwon (1984) have reported that Kimchi samples stored at high temperatures will reach the ripening stage faster than those stored at low temperatures. Chang and Kim (2000) have reported that the edible period of Kimchi depended on storage temperature. In addition, it has been reported that low storage temperatures are better for the stability and quality of Kimchi than high storage temperatures (Choi et al., 1990; Park et al., 2008). Therefore, differences in Kimchi quality might be simply due to difference in fermentation speed depending on fermentation temperature. However, recent studies have 2
shown that differences in fermentation microbial community also depended on temperature (Park et al., 2003; Tabatabaei Yazdi., 2013). For example, Lee et al. (2008) have reported that Lactobacillus sake was predominant in Kimchi due to proper fermentation temperature (5-9℃) and storage temperature (-2℃). However, few studies have shown the overall fermentation pattern or metabolite differences of Kimchi according to fermentation temperature. Metabolomics refers to systematic identification and quantification of metabolites present within an organism, cell, or tissue. Many metabolomics studies have been done for component analysis (Farag et al., 2012), quality evaluation (Vikram et al., 2006), safety evaluation (Castro-Puyana and Herrero, 2013), or microbial monitoring (MacKenzie et al., 2008). A previous metabolomics study using GC-MS with multivariate statistical analysis has shown the metabolic characteristics of Kimchi according to starter cultures (Park et al., 2016). Kimchi metabolites are important factors that determine the taste of Kimchi. However, little is known about changes of Kimchi metabolites with temperature during the fermentation process. Thus, the objective of this study was to monitor overall changes in metabolites of Kimchi with fermentation temperature using a GC-MS-based metabolomics approach.
2. Materials and methods 2.1. Kimchi preparation Kimchi was prepared according to a previous report (Jeong et al., 2013). Briefly, a seasoning mixture was prepared as follows: Chinese cabbage (90%); red pepper (2.2%); leek (2.8%); garlic (1.4%); ginger (0.4%); water (3.2%). The raw materials, including Chinese cabbage (Brassica rapa L. ssp. pekinensis), were purchased at a fully ripe state from a local market in Naju (Korea) in 2016. The final salinity of Kimchi was 2.5%. The Kimchi mixture was then fermented in 4, 12, and 20℃ incubators. Samples were taken on the 1st, 5th, 10th, and 50th day of fermentation for analyses. 3
2.2. Total cell counts Kimchi samples were diluted ten-fold with 0.1% peptone water (Difco, Detroit, MI, USA). Total bacteria and LAB counts were measured after incubating samples at 37℃ for 48 h in plate count agar (PCA, Difco) and de Man, Rogosa and Sharpe (MRS) agar (Difco), respectively. Total cell counts were carried out in triplicates. Results were expressed as log CFU/g.
2.3. Measurement of pH and titratable acidity Kimchi samples (10 g) were ground using a hand mixer (model: HBL-3500S, Samyang Electronic Co., Gimpo, Korea). After centrifuging (Union 55R, Hanil, Seoul, Korea) Kimchi samples at 14,300×g for 15 min at 4℃, pH was measured using a pH meter (pH-250L, ISTEK, Seoul, Korea). Titratable acidity was expressed as lactic acid (%) by the volume of 0.1 N NaOH needed to reach pH 8.3.
2.4. Derivatization The sample derivatization protocol and GC-MS analysis conditions were the same as described in a previous study (Park et al., 2016) with slight modifications. Briefly, 150 µL of Kimchi sample supernatant and 30 µL of internal standard (ribitol in water, 20 mg/mL) were lyophilized (FD 8505, Ilshinlab, Daejeon, Korea). After adding 80 µL of methoxyamine hydrochloride in pyridine (20 mg/mL) to the freeze-dried sample residue, the mixture was sonicated (Powersonic 520, Hwashin, Seoul, Korea) for 5 min followed by incubation at 37℃ for 12 h for oximation. Next, 100 µL of N,O-bis-(trimethylsilyl)-trifluoroacetamide (containing 1% trimethylchlorosilane) was added to the reaction and incubated at 70℃ for 1 h. These samples were cooled at 20℃ in the dark for 1 h. After centrifuging samples at 12,500 ×g for 10 min at 4℃, supernatants were transferred to glass vials for GC-MS analysis. The 4
reagents used in this experiment were purchased from Sigma-Aldrich (Sigma Aldrich, St. Louis, MO, USA).
2.5. GC-MS analysis One µL of each derivatized sample was injected using a splitless injector with an Agilent 6890N GC (Agilent, Santa Clara, CA, USA) equipped with a DB-5MS capillary column from Agilent (30 m × 0.25 mm i.d., 0.25 μm film thickness). The injector temperature was set at 250℃. Flow rate of carrier gas (He, 99.9%) was at 1 mL/min. GC-MS temperatures were as follows: injector, 250℃; transfer line, 280℃; ion source, 230℃; and quadrupole temperature, 150℃. The oven temperature was maintained at 60℃ for 1 min, increased to 300℃ at 10℃/min, and then held at 300℃ for 10 min. Column effluent was introduced into the ion source of an Agilent 5973N MS (Agilent). The mass spectrometer was programmed using electron impact in a full scan mode at m/z 50–550 with a scanning rate of 2 scans/sec.
2.6. Data processing and multivariate analysis GC-MS raw data in a netCDF format obtained from MSD ChemStation (Agilent) were imported to the XCMS website (https://xcmsonline.scripps.edu) for peak detection, integration, and statistical analysis. A default Centwave method (Tautenhahn et al., 2008) for GC Single Quadruple was selected for peak detection and alignment with the following parameters: signal/noise threshold, 2; mzdiff, 0.1; integration methods, 1; prefilter peaks, 3; prefilter intensity, 100; mzwid, 0.25; minfrac, 0.5; and bandwidth, 3. Metabolites were identified by comparing their accurate mass and MS fragments with spectra in the NIST ver. 14.0 library (http://www.sisweb.com/software/ms/nist.htm). Peak intensities were normalized against an internal standard (ribitol). Multivariate statistical analysis was done with principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) using 5
SIMCA-P version 14.0 (Umetrics, Umea, Sweden). Hotelling’s T2 region, shown as an ellipse in the PCA score diagram, defines the 95% confidence interval of the modeled variation (Hotelling, 1931). The quality of PCA models was expressed by R2 (the variance in the data) and Q2 values (the prediction of model). A permutation test of two hundred iterations with a cross-validation step was done to avoid model overfitting. After processing PLS-DA, peaks with variable importance in projection (VIP) score > 1.5 and p < 0.05 were considered to be responsible for differences.
3. Results and Discussion 3.1. Microbial and physicochemical changes during Kimchi fermentation Growth curves for total microbial cells in Kimchi according to fermentation temperature is shown in Fig. 1A. Total cell number in Kimchi before fermentation was log 6.42 CFU/g. Total cell counts of Kimchi decreased slightly on the 1st day of fermentation, but increased significantly at 12 and 20℃. On the 10th day of fermentation, Kimchi samples fermented at 12 and 20℃ showed maximum total cell counts. On the other hand, total cell number in Kimchi stored at 4℃ increased steadily from the 1st day. These results confirmed that higher fermentation temperature would lead to faster growth rate and higher total number of bacteria during the entire fermentation period. The growth curve for LAB showed a similar trend as total cell count (data not shown) and were of similar magnitude, indicating that most bacteria during Kimchi fermentation were LAB. Changes in pH and titratable acidity (TA) of Kimchi with different fermentation temperature are shown in Fig. 1B and 1C, respectively. On the 5th day of fermentation, the pH of Kimchi was rapidly decreased at 12 or 20℃. However, the pH value of Kimchi fermented at 4℃ was decreased slower compared to that of Kimchi fermented at 12 or 20℃, suggesting that fermentation progressed slowly at 4℃. Generally, TA values and pH of 6
Kimchi showed opposite tendencies during Kimchi fermentation. As expected, a large and rapid increase of TA was observed for Kimchi stored at 20℃. This clearly demonstrates that high temperature accelerates the fermentation process of Kimchi. It has been reported that pH and TA of Kimchi after the optimum ripening period are 4.2 and 0.6%, respectively (Mheen and Kwon, 1984). Based on such criteria, the optimum fermentation period for Kimchi at 12 and 20℃ would be 5 to 10 days. However, the optimal fermentation period for Kimchi at 4℃ was found to be 50 days or possibly a few days longer because pH and TA at that point were 4.3 and 0.55%, respectively (Fig. 1).
3.2. Metabolite changes during Kimchi fermentation GC-MS was used to monitor metabolic changes during the Kimchi fermentation process. A total of 1,851 variables were obtained based on mass peaks on chromatograms. Changes in total metabolites of Kimchi samples during 50 days of fermentation were expressed as score plots using PCA (R2 = 0.695, Q2 = 0.674) (Fig. 2). As fermentation progressed, points moved from quadrant 4 to quadrant 1. They then moved to quadrant 2 or 3 according to fermentation period, indicating continuous metabolic changes during fermentation. The Kimchi samples in the PCA score plot showed a clear separation into two groups for PC1. One group included all Kimchi samples at fermentation day 0 and 1, and samples fermented at 4℃ until 10 days after fermentation, implying that metabolites of Kimchi fermented at 4℃ changed slowly during fermentation. The other group included the rest of the Kimchi samples during 50 days of fermentation. Kimchi samples fermented at 20℃ on the 50th day of fermentation were found to be out of Hotelling's T2 range, suggesting that their metabolite profiles were significantly different from those of other samples. To investigate metabolic changes of Kimchi over the fermentation period, PCA score plots were regenerated at each temperature (Fig. 3). As shown in the PCA score plot (Fig. 3A), 7
points of Kimchi samples fermented at 4℃ showed a noticeable movement between 10 and 50 days of fermentation, indicating that fermentation proceeded rapidly during this period. Meanwhile, Kimchi samples stored at 12℃ showed a large metabolic change between days 1 and 5 (Fig. 3B), suggesting that early Kimchi fermentation progressed very fast. The PCA score plot of Kimchi samples fermented at 20℃ also showed clear separation of PC1 between samples obtained at days 1 and 5 (R2 = 0.760, Q2 = 0.725) (Fig. 3C).
3.3. Metabolite differences by fermentation temperature To determine metabolic differences of final Kimchi samples fermented at different temperatures, PCA models were generated using variables obtained from GC-MS data at 50 days after fermentation (Fig. 4A). Kimchi samples fermented at 4℃ were successfully separated from other Kimchi samples using PC1. However, Kimchi samples were not fully distinguishable between 12 and 20℃ in the PCA score plot, implying that metabolites of these Kimchi samples were similar to each other, but different from those of Kimchi samples stored at 4℃. This might be due to differences in bacterial community structure depending on fermentation temperature. To maximize separation between samples, the PLS-DA (supervised pattern recognition method) was applied (Fig. 4B). The PLS-DA score plot of Kimchi samples showed clear differentiation. When two components were calculated, cumulative R2X, R2Y, and Q2 values were 0.650, 0.961, and 0.892, respectively. A permutation test (200 permutations) was done to verify the PLS-DA model. Through this test, Q2 and R2 values were found to be higher than their original values, proving the suitability and validity of this model (Fig. 4C). To determine which metabolites caused segregation, the VIP score of > 1.5 and p < 0.01 were used. A total of six metabolites were identified based on fragmentation patterns of GC–MS library in NIST 14 and other researchers’ experimental data. Metabolites responsible for 8
metabolic differences among these groups included some organic acids (fumaric acid, malic acid, citric acid, and galactaric acid), propylene glycol, and alanine. Changes of these six metabolites according to fermentation temperature are shown in Fig. 5. Levels of fumaric acid, malic acid, and citric acid in Kimchi samples fermented at 20℃ decreased rapidly during the early fermentation period. However, levels of malic acid and citric acid in Kimchi samples fermented at 4℃ decreased after 10 days of fermentation. These results were consistent with large changes in metabolites of Kimchi fermented at 4℃ between 10 and 50 days of fermentation as shown in Fig. 3A. Reduction in malic acid and citric acid levels indicated the presence of malolactic fermentation since LAB could convert malic acid and citric acid into lactic acid (Son et al., 2009). When lactic acid is produced in Kimchi, the growth of aerobic bacteria is suppressed (Jang and Kim, 2013). Park et al. (2012) have reported changes in bacterial community structure of Kimchi after a long period of incubation at room temperature (22℃). These changes in microbial communities might cause differences in metabolites, thus affecting the quality of Kimchi. Levels of propylene glycol and galactaric acid in Kimchi stored at 4℃ were increased sharply between the 10th and 50th day of fermentation. Interestingly, these metabolites showed very little changes during the entire fermentation period in Kimchi samples stored at 12 or 20℃. In this study, different contents of metabolites in Kimchi with fermentation temperatures indicated that fermentation temperature can affect microbial community and fermentation rate. Kimchi generally undergoes fermentation at low temperature (about 4-6℃). However, it will take weeks to finish the process. Commercially produced Kimchi is usually fermented for a short period (1-2 days) at 20℃. It is then stored at low temperature (5-8℃) to improve organoleptic quality (Hong et al., 1994; Lee et al., 2017). It is difficult to clearly determine the optimum duration of ripening time according to the fermentation temperature of Kimchi. Cheigh et al. (1994) have reported that optimized time for Kimchi fermented at 5 9
and 20℃ were 35-180 days and 2-3 days, respectively. According to a study of Chang and Kim (2000), optimized time for Kimchi ripening at 15, 10, and 5℃ were 4, 10, and 18 days, respectively. Similar to these studies, Lee et al. (1991) have reported that Kimchi’s shelf life 4 and 28℃ were 33 and 3 days, respectively, based on kinetic modeling of total acidity for quality prediction. Hong et al. (2016) have reported that the optimum ripening time for Kimchi fermented at 4 and 20℃ was 35 and 2 days, respectively, after analyzing volatile compositions of Kimchi. Nowadays, it is common to store homemade or commercially produced Kimchi products in a Kimchi refrigerator in Korea to reduce microbial growth and avoid over-fermenting or souring by rapid shifting to subzero temperatures (Hongu et al., 2017; Lee et al., 2008). Although the optimum ripening time for Kimchi is different, it obviously depends on storage temperature. Storage temperature of Kimchi is directly related to the growth of microorganisms which will affect Kimchi metabolites. Therefore, it is important to know the metabolic difference in Kimchi because it affects the quality of Kimchi. These results suggest that the optimum duration for Kimchi fermentation can be determined using a metabolomics approach to simultaneously observe metabolites during fermentation.
4. Conclusion A GC-MS based metabolomics approach coupled with multivariate statistics were applied in this study to understand relationships between fermentation temperature and metabolite differences of Kimchi. Metabolites of Kimchi fermented at 4℃ changed more slowly during early fermentation than those of Kimchi fermented at higher temperatures (4 and 20℃). After fermentation, different concentrations of metabolites were observed in Kimchi samples depending on fermentation temperature. It is possible to determine the optimum duration of Kimchi fermentation by metabolomics approach which can simultaneously observe 10
metabolites during fermentation. However, more detailed studies are needed to better understand relationships of sensory characteristics, microbial communities, and metabolites of Kimchi with fermentation temperature.
Acknowledgements This work was supported by a grant (NRF- 2014R1A1A1002817) from the National Research Foundation funded by the Korean Government.
Conflict of interest The authors declare that there are no conflicts of interest.
References Castro-Puyana, M., Herrero, M. Metabolomics approaches based on mass spectrometry for food safety, quality and traceability. TrAC Trends in Analytical Chemistry, 52, 74-87. Chang, M. J., Kim, M. H. (2000). Fermentation property of Chinese cabbage Kimchi by fermentation temperature and salt concentration. Applied Biological Chemistry, 43, 7-11. Cheigh, H. S., Park, K. Y., Lee, C. (1994). Biochemical, microbiological, and nutritional aspects of Kimchi (Korean fermented vegetable products). Critical Reviews in Food Science and Nutrition, 34, 175-203. Choi, S. Y., Kim, Y. B., Yoo, J. Y., Lee, I. S., Chung, K. S., Koo, Y. J. (1990). Effect of temperature and salts concentration of Kimchi manufacturing on storage. Korean Journal of Food Science and Technology, 22, 707-710. Farag, M. A., Porzel, A., Wessjohann, L. A. (2012). Comparative metabolite profiling and fingerprinting of medicinal licorice roots using a multiplex approach of GC–MS, LC–MS and 1D NMR techniques. Phytochemistry, 76, 60-72. Hong, S. P., Lee, E. J., Kim, Y. H., Ahn, D. U. (2016). Effect of fermentation temperature on the volatile composition of Kimchi. Journal of Food Science, 81, C2623-C2629. Hongu, N., Kim, A. S., Suzukim, A., Wilson, H., Tsui, K. C., Park, S. (2017). Korean Kimchi: Promoting healthy meals through cultural tradition. Journal of Ethnic Foods, 4, 172-180. 11
Hotelling H. (1931). The generalization of Student's ratio. The Annals of Mathematical Statistics, 2, 360-378. Jang, D. J., Chung, K. R., Yang, H. J., Kim, K. S., Kwon, D. Y. (2015). Discussion on the origin of Kimchi, representative of Korean unique fermented vegetables. Journal of Ethnic Foods, 2, 126-136. Jang, J. Y., Kim, T. W. (2013). Lactic acid bacteria in Kimchi and their immunomodulatory activities. Current Topic in Lactic Acid Bacteria and Probiotics, 1, 28-37. Jeong, S. H., Lee, S. H., Jung, J. Y., Choi, E. J., Jeon, C. O. (2013). Microbial succession and metabolite changes during long-term storage of Kimchi. Journal of Food Science, 78, M763-M769. Jung, J. Y., Lee, S. H., Jin, H. M., Hahn, Y., Madsen, E. L., Jeon, C. O. (2013). Metatranscriptomic analysis of lactic acid bacterial gene expression during Kimchi fermentation. International Journal of Food Microbiology, 163, 171-179. Jung, J. Y., Lee, S. H., Lee, H. J., Seo, H. Y., Park, W. S., Jeon, C. O. (2012). Effects of Leuconostoc mesenteroides starter cultures on microbial communities and metabolites during Kimchi fermentation. International Journal of Food Microbiology, 153, 378-387. Kim, D. W., Kim, B. M., Lee, H. J., Jang, G. J., Song, S. H., Lee, J. I., Lee, S. B., Shim, J. M., Lee, K. W., Kim, J. H., Ham, K. S., Chen, F., Kim, H. J. (2017). Effects of different salt treatments on the fermentation metabolites and bacterial profiles of Kimchi. Journal of Food Science, 82, 1124-1131. Lee, D., Kim, S., Cho, J., Kim, J. (2008). Microbial population dynamics and temperature changes during fermentation of kimjang Kimchi. The Journal of Microbiology, 46, 590-593. Lee, K. H., Cho, H. Y., Pyun, Y. R. (1991). Kinetic modelling for the prediction of shelf-life of Kimchi based on total acidity as a quility index. Korean Journal of Food Science and Technology, 23, 306-310. Lee, M., Song, J. H., Jung, M. Y., Lee, S. H., Chang, J. Y. (2017). Large-scale targeted metagenomics analysis of bacterial ecological changes in 88 Kimchi samples during fermentation. Food Microbiology, 66, 173-183. MacKenzie, D. A., Defernez, M., Dunn, W. B., Brown, M., Fuller, L. J., de Herrera, S. R., Günther, A., James, S. A., Eagles, J., Philo, M. (2008). Relatedness of medically important strains of Saccharomyces cerevisiae as revealed by phylogenetics and metabolomics. Yeast, 25, 501-512. Mheen, T. I., Kwon, T. W. (1984). Effect of temperature and salt concentration on Kimchi fermentation. Korean Journal of Food Science and Technology, 16, 443-450. Park, E. J., Chun, J., Cha, C. J., Park, W. S., Jeon, C. O., Bae, J. W. (2012). Bacterial community analysis during fermentation of ten representative kinds of Kimchi with barcoded pyrosequencing. Food Microbiology, 30, 197-204. 12
Park, J., Heo, G., Oh, Y., Kim, B., Miheen, T., Kim, C., Ahn, J., Lee, J. (2003). Change of microbial communities in Kimchi fermentation at low temperature. Korean Journal of Microbiology, 39, 45-50. Park, J. G., Kim, J. H., Park, J. N., Kim, Y. D., Kim, W. G., Lee, J. W., Hwang, H. J., Byun, M. W. (2008). The effect of irradiation temperature on the quality improvement of Kimchi, Korean fermented vegetables, for its shelf stability. Radiation Physics and Chemistry, 77, 497-502. Park, S. E., Yoo, S. A., Seo, S. H., Lee, K. I., Na, C. S., Son, H. S. (2016). GC–MS based metabolomics approach of Kimchi for the understanding of Lactobacillus plantarum fermentation characteristics. LWT-Food Science and Technology, 68, 313-321. Son, H. S., Hwang, G. S., Park, W. M., Hong, Y. S., Lee, C. H. (2009). Metabolomic characterization of malolactic fermentation and fermentative behaviors of wine yeasts in grape wine. Journal of Agricultural and Food Chemistry, 57, 4801-4809. Tabatabaei Yazdi, F., Alizadeh Behbahani, B., Mohebbi, M., Mortazavi, A., Ghaitaranpour, A. (2013). Effect of temperature on microbial changes during Kimchi fermentation. Scientific Journal of Microbiology, 2, 9-14. Tautenhahn, R., Boettcher, C., Neumann, S. (2008). Highly sensitive feature detection for high resolution LC/MS. BMC Bioinformatics, 9, 504. Vikram, A., Lui, L., Hossain, A., Kushalappa, A. (2006). Metabolic fingerprinting to discriminate diseases of stored carrots. Annals of Applied Biology, 148, 17-26.
Fig. 1. Changes in viable cells of total bacteria (A), pH (B), and titratable acidity (C) of Kimchi during ripening at different temperatures. Values are presented as mean ± standard deviations (n = 5). Fig. 2. PCA score plot derived from GC-MS data of Kimchi during ripening at different temperatures (4, 12, and 20℃). Each symbol (point) in the score plot represents a Kimchi sample. Symbols with different colors and shapes denote Kimchi samples taken at different temperatures on different days of fermentation (Color figure online). Fig. 3. PCA score plots derived from GC-MS data of Kimchi during ripening at 4℃ (A), 12℃ (B), and 20℃ (C), showing different metabolic changes (fermentation rates) 13
depending on different temperature. The color and shape of each point are the same as shown in Fig. 1 (Color figure online). Fig. 4. PCA (A) and PLS-DA (B) score plots derived from GC-MS data of Kimchi after 50 days of fermentation at different temperatures. A permutation test was carried out with 200 random permutations in the PLS-DA model (C). The color and shape of each point are the same as shown in Fig. 1. Fig. 5. Changes in identified metabolites of Kimchi during ripening at different temperatures. Metabolites contributing to differentiation in the PLS-DA model (VIP score > 1.5, p < 0.05) were selected. Values are presented as mean ± standard deviations (n = 5).
Highlights ● Effect of fermentation temperature on metabolic changes in Kimchi was investigated. ● Metabolites of Kimchi fermented at low temperature changed slowly during fermentation. ● Alanine, propylene glycol, fumaric acid, malic acid, citric acid, and galactaric acid concentrations were different depending on temperature. ● GC-MS based metabolomics approach is a valuable tool to determine the optimum duration of Kimchi fermentation
14
Fig. 1.
Total cell counts (log CFU/g)
(A) Total bacteria
11.0
4℃
12 ℃
20 ℃
10.0 9.0 8.0 7.0 6.0 5.0 4.0 0
5
10
15
20
25
30
35
40
45
50
Fermentation time (days)
(B) pH
7.0
4℃
12℃
20℃
6.0
pH
5.0 4.0 3.0 2.0 0
5
10
15
20
25
30
Fermentation time (days)
15
35
40
45
50
(C) TA
Titratable acidity (%)
2.0
4℃
12℃
20℃
1.5 1.0 0.5 0.0 0
5
10
15
20
25
30
35
Fermentation time (days)
Fig. 2.
16
40
45
50
Fig. 3. (A)
(B)
17
(C)
Fig. 4. (A)
18
(B)
(C)
19
Fig. 5. Fumaric acid
Malic acid
0.0050
0.012 4℃ 12℃ 20℃
0.010
Intensity
Intensity
0.0040 0.0030 0.0020
0.008 0.006 0.004
0.0010
0.002 0.000
0.0000 0
10
20
30
40
0
50
10
Fermentation time (days)
20
30
40
50
40
50
40
50
Fermentation time (days)
Citric acid
Galactaric acid
0.012
0.0020
0.010 0.0015
Intensity
Intensity
0.008 0.006
0.0010
0.004 0.0005 0.002 0.0000
0.000 0
10
20
30
40
0
50
10
β-alanine
30
Propylene glycol
0.14
0.006
0.12
0.005
0.10
0.004
Intensity
Intensity
20
Fermentation time (days)
Fermentation time (days)
0.08 0.06 0.04
0.003 0.002 0.001
0.02
0
0.00 0
10
20
30
40
0
50
10
20
30
Fermentation time (days)
Fermentation time (days)
20
21
PII: DOI: Reference:
S2212-4292(17)30822-2 https://doi.org/10.1016/j.fbio.2018.03.009 FBIO290
To appear in: Food Bioscience Received date: 3 November 2017 Revised date: 19 March 2018 Accepted date: 24 March 2018 Cite this article as: Seong-Eun Park, Seung-Ho Seo, Eun-Ju Kim, Chang-Su Na and Hong-Seok Son, Effects of different fermentation temperatures on metabolites of Kimchi, Food Bioscience, https://doi.org/10.1016/j.fbio.2018.03.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Effects of different fermentation temperatures on metabolites of Kimchi Seong-Eun Park, Seung-Ho Seo, Eun-Ju Kim, Chang-Su Na, Hong-Seok Son* Department of Korean Medicine, Dongshin University, Naju, Jeonnam 58245, Republic of Korea * Corresponding author. Tel.: +82 61 330 3513; Fax: +82 61 330 3519. [email protected] (H.-S. Son).
Abstract The influences of fermentation temperature on metabolic changes in Kimchi using a gas chromatography–mass spectrometry (GC-MS) method were studied. Kimchi mixtures fermented at 4, 12, and 20°C were taken on the 1st, 5th, 10th, and 50th day of fermentation to determine overall changes in metabolites of Kimchi during fermentation. Based on the principal component analysis (PCA) score plot, Kimchi samples were confirmed to have separation trends by the first principal component 1 (PC1). One group included all samples on fermentation day 0 and day 1, and samples fermented at 4°C for up to 10 days after fermentation, demonstrating that metabolites of Kimchi fermented at 4°C changed slowly. The partial least squares discriminant analysis (PLS-DA) score plot also showed clear differences in metabolites among Kimchi samples with fermentation temperature. The changing metabolites were identified to be alanine, propylene glycol, fumaric acid, malic acid, citric acid, and galactaric acid. These results highlight that a GC-MS-based metabolomics approach can be used to monitor distinct metabolite changes in Kimchi with fermentation temperature.
Keywords: Kimchi, Chinese cabbage, fermentation, metabolomics, GC-MS
1
1. Introduction Kimchi is Korea's representative ethnic food. It is a unique food made by fermenting vegetables (such as Chinese cabbage and radish), fish sauce, and spices (such as garlic and ginger) (Jang et al., 2015). According to the Korean Nutrition survey, Korean adults consume 50-200 g/day/person, accounting for 12.5% of their total daily food intake (Mheen and Kwon, 1984). Kimchi is usually manufactured by spontaneous fermentation without using a starter culture. Lactic acid bacteria (LAB) such as Lactobacillus, Leuconostoc, Weissella, Lactococcus, and Pediococcus are involved in Kimchi fermentation (Park et al., 2003). Recently, many studies have determined the dynamics of microbial communities in Kimchi using a metagenomics approach (Jeong et al., 2013; Jung et al., 2012; Jung et al., 2013; Park et al., 2003). Microbial communities in Kimchi during its fermentation are affected by several factors, including raw material (Chinese cabbage), temperature, pH, salt concentration, and fermentation period (Kim et al., 2017; Lee et al., 2017). Among these, fermentation temperature is a critical factor for determining its final quality. To maintain the quality of Kimchi and store it for a long time, controlling fermentation rate through temperature regulation is well-known. Many previous studies have reported the relationship between fermentation temperature and Kimchi quality. Mheen and Kwon (1984) have reported that Kimchi samples stored at high temperatures will reach the ripening stage faster than those stored at low temperatures. Chang and Kim (2000) have reported that the edible period of Kimchi depended on storage temperature. In addition, it has been reported that low storage temperatures are better for the stability and quality of Kimchi than high storage temperatures (Choi et al., 1990; Park et al., 2008). Therefore, differences in Kimchi quality might be simply due to difference in fermentation speed depending on fermentation temperature. However, recent studies have 2
shown that differences in fermentation microbial community also depended on temperature (Park et al., 2003; Tabatabaei Yazdi., 2013). For example, Lee et al. (2008) have reported that Lactobacillus sake was predominant in Kimchi due to proper fermentation temperature (5-9℃) and storage temperature (-2℃). However, few studies have shown the overall fermentation pattern or metabolite differences of Kimchi according to fermentation temperature. Metabolomics refers to systematic identification and quantification of metabolites present within an organism, cell, or tissue. Many metabolomics studies have been done for component analysis (Farag et al., 2012), quality evaluation (Vikram et al., 2006), safety evaluation (Castro-Puyana and Herrero, 2013), or microbial monitoring (MacKenzie et al., 2008). A previous metabolomics study using GC-MS with multivariate statistical analysis has shown the metabolic characteristics of Kimchi according to starter cultures (Park et al., 2016). Kimchi metabolites are important factors that determine the taste of Kimchi. However, little is known about changes of Kimchi metabolites with temperature during the fermentation process. Thus, the objective of this study was to monitor overall changes in metabolites of Kimchi with fermentation temperature using a GC-MS-based metabolomics approach.
2. Materials and methods 2.1. Kimchi preparation Kimchi was prepared according to a previous report (Jeong et al., 2013). Briefly, a seasoning mixture was prepared as follows: Chinese cabbage (90%); red pepper (2.2%); leek (2.8%); garlic (1.4%); ginger (0.4%); water (3.2%). The raw materials, including Chinese cabbage (Brassica rapa L. ssp. pekinensis), were purchased at a fully ripe state from a local market in Naju (Korea) in 2016. The final salinity of Kimchi was 2.5%. The Kimchi mixture was then fermented in 4, 12, and 20℃ incubators. Samples were taken on the 1st, 5th, 10th, and 50th day of fermentation for analyses. 3
2.2. Total cell counts Kimchi samples were diluted ten-fold with 0.1% peptone water (Difco, Detroit, MI, USA). Total bacteria and LAB counts were measured after incubating samples at 37℃ for 48 h in plate count agar (PCA, Difco) and de Man, Rogosa and Sharpe (MRS) agar (Difco), respectively. Total cell counts were carried out in triplicates. Results were expressed as log CFU/g.
2.3. Measurement of pH and titratable acidity Kimchi samples (10 g) were ground using a hand mixer (model: HBL-3500S, Samyang Electronic Co., Gimpo, Korea). After centrifuging (Union 55R, Hanil, Seoul, Korea) Kimchi samples at 14,300×g for 15 min at 4℃, pH was measured using a pH meter (pH-250L, ISTEK, Seoul, Korea). Titratable acidity was expressed as lactic acid (%) by the volume of 0.1 N NaOH needed to reach pH 8.3.
2.4. Derivatization The sample derivatization protocol and GC-MS analysis conditions were the same as described in a previous study (Park et al., 2016) with slight modifications. Briefly, 150 µL of Kimchi sample supernatant and 30 µL of internal standard (ribitol in water, 20 mg/mL) were lyophilized (FD 8505, Ilshinlab, Daejeon, Korea). After adding 80 µL of methoxyamine hydrochloride in pyridine (20 mg/mL) to the freeze-dried sample residue, the mixture was sonicated (Powersonic 520, Hwashin, Seoul, Korea) for 5 min followed by incubation at 37℃ for 12 h for oximation. Next, 100 µL of N,O-bis-(trimethylsilyl)-trifluoroacetamide (containing 1% trimethylchlorosilane) was added to the reaction and incubated at 70℃ for 1 h. These samples were cooled at 20℃ in the dark for 1 h. After centrifuging samples at 12,500 ×g for 10 min at 4℃, supernatants were transferred to glass vials for GC-MS analysis. The 4
reagents used in this experiment were purchased from Sigma-Aldrich (Sigma Aldrich, St. Louis, MO, USA).
2.5. GC-MS analysis One µL of each derivatized sample was injected using a splitless injector with an Agilent 6890N GC (Agilent, Santa Clara, CA, USA) equipped with a DB-5MS capillary column from Agilent (30 m × 0.25 mm i.d., 0.25 μm film thickness). The injector temperature was set at 250℃. Flow rate of carrier gas (He, 99.9%) was at 1 mL/min. GC-MS temperatures were as follows: injector, 250℃; transfer line, 280℃; ion source, 230℃; and quadrupole temperature, 150℃. The oven temperature was maintained at 60℃ for 1 min, increased to 300℃ at 10℃/min, and then held at 300℃ for 10 min. Column effluent was introduced into the ion source of an Agilent 5973N MS (Agilent). The mass spectrometer was programmed using electron impact in a full scan mode at m/z 50–550 with a scanning rate of 2 scans/sec.
2.6. Data processing and multivariate analysis GC-MS raw data in a netCDF format obtained from MSD ChemStation (Agilent) were imported to the XCMS website (https://xcmsonline.scripps.edu) for peak detection, integration, and statistical analysis. A default Centwave method (Tautenhahn et al., 2008) for GC Single Quadruple was selected for peak detection and alignment with the following parameters: signal/noise threshold, 2; mzdiff, 0.1; integration methods, 1; prefilter peaks, 3; prefilter intensity, 100; mzwid, 0.25; minfrac, 0.5; and bandwidth, 3. Metabolites were identified by comparing their accurate mass and MS fragments with spectra in the NIST ver. 14.0 library (http://www.sisweb.com/software/ms/nist.htm). Peak intensities were normalized against an internal standard (ribitol). Multivariate statistical analysis was done with principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA) using 5
SIMCA-P version 14.0 (Umetrics, Umea, Sweden). Hotelling’s T2 region, shown as an ellipse in the PCA score diagram, defines the 95% confidence interval of the modeled variation (Hotelling, 1931). The quality of PCA models was expressed by R2 (the variance in the data) and Q2 values (the prediction of model). A permutation test of two hundred iterations with a cross-validation step was done to avoid model overfitting. After processing PLS-DA, peaks with variable importance in projection (VIP) score > 1.5 and p < 0.05 were considered to be responsible for differences.
3. Results and Discussion 3.1. Microbial and physicochemical changes during Kimchi fermentation Growth curves for total microbial cells in Kimchi according to fermentation temperature is shown in Fig. 1A. Total cell number in Kimchi before fermentation was log 6.42 CFU/g. Total cell counts of Kimchi decreased slightly on the 1st day of fermentation, but increased significantly at 12 and 20℃. On the 10th day of fermentation, Kimchi samples fermented at 12 and 20℃ showed maximum total cell counts. On the other hand, total cell number in Kimchi stored at 4℃ increased steadily from the 1st day. These results confirmed that higher fermentation temperature would lead to faster growth rate and higher total number of bacteria during the entire fermentation period. The growth curve for LAB showed a similar trend as total cell count (data not shown) and were of similar magnitude, indicating that most bacteria during Kimchi fermentation were LAB. Changes in pH and titratable acidity (TA) of Kimchi with different fermentation temperature are shown in Fig. 1B and 1C, respectively. On the 5th day of fermentation, the pH of Kimchi was rapidly decreased at 12 or 20℃. However, the pH value of Kimchi fermented at 4℃ was decreased slower compared to that of Kimchi fermented at 12 or 20℃, suggesting that fermentation progressed slowly at 4℃. Generally, TA values and pH of 6
Kimchi showed opposite tendencies during Kimchi fermentation. As expected, a large and rapid increase of TA was observed for Kimchi stored at 20℃. This clearly demonstrates that high temperature accelerates the fermentation process of Kimchi. It has been reported that pH and TA of Kimchi after the optimum ripening period are 4.2 and 0.6%, respectively (Mheen and Kwon, 1984). Based on such criteria, the optimum fermentation period for Kimchi at 12 and 20℃ would be 5 to 10 days. However, the optimal fermentation period for Kimchi at 4℃ was found to be 50 days or possibly a few days longer because pH and TA at that point were 4.3 and 0.55%, respectively (Fig. 1).
3.2. Metabolite changes during Kimchi fermentation GC-MS was used to monitor metabolic changes during the Kimchi fermentation process. A total of 1,851 variables were obtained based on mass peaks on chromatograms. Changes in total metabolites of Kimchi samples during 50 days of fermentation were expressed as score plots using PCA (R2 = 0.695, Q2 = 0.674) (Fig. 2). As fermentation progressed, points moved from quadrant 4 to quadrant 1. They then moved to quadrant 2 or 3 according to fermentation period, indicating continuous metabolic changes during fermentation. The Kimchi samples in the PCA score plot showed a clear separation into two groups for PC1. One group included all Kimchi samples at fermentation day 0 and 1, and samples fermented at 4℃ until 10 days after fermentation, implying that metabolites of Kimchi fermented at 4℃ changed slowly during fermentation. The other group included the rest of the Kimchi samples during 50 days of fermentation. Kimchi samples fermented at 20℃ on the 50th day of fermentation were found to be out of Hotelling's T2 range, suggesting that their metabolite profiles were significantly different from those of other samples. To investigate metabolic changes of Kimchi over the fermentation period, PCA score plots were regenerated at each temperature (Fig. 3). As shown in the PCA score plot (Fig. 3A), 7
points of Kimchi samples fermented at 4℃ showed a noticeable movement between 10 and 50 days of fermentation, indicating that fermentation proceeded rapidly during this period. Meanwhile, Kimchi samples stored at 12℃ showed a large metabolic change between days 1 and 5 (Fig. 3B), suggesting that early Kimchi fermentation progressed very fast. The PCA score plot of Kimchi samples fermented at 20℃ also showed clear separation of PC1 between samples obtained at days 1 and 5 (R2 = 0.760, Q2 = 0.725) (Fig. 3C).
3.3. Metabolite differences by fermentation temperature To determine metabolic differences of final Kimchi samples fermented at different temperatures, PCA models were generated using variables obtained from GC-MS data at 50 days after fermentation (Fig. 4A). Kimchi samples fermented at 4℃ were successfully separated from other Kimchi samples using PC1. However, Kimchi samples were not fully distinguishable between 12 and 20℃ in the PCA score plot, implying that metabolites of these Kimchi samples were similar to each other, but different from those of Kimchi samples stored at 4℃. This might be due to differences in bacterial community structure depending on fermentation temperature. To maximize separation between samples, the PLS-DA (supervised pattern recognition method) was applied (Fig. 4B). The PLS-DA score plot of Kimchi samples showed clear differentiation. When two components were calculated, cumulative R2X, R2Y, and Q2 values were 0.650, 0.961, and 0.892, respectively. A permutation test (200 permutations) was done to verify the PLS-DA model. Through this test, Q2 and R2 values were found to be higher than their original values, proving the suitability and validity of this model (Fig. 4C). To determine which metabolites caused segregation, the VIP score of > 1.5 and p < 0.01 were used. A total of six metabolites were identified based on fragmentation patterns of GC–MS library in NIST 14 and other researchers’ experimental data. Metabolites responsible for 8
metabolic differences among these groups included some organic acids (fumaric acid, malic acid, citric acid, and galactaric acid), propylene glycol, and alanine. Changes of these six metabolites according to fermentation temperature are shown in Fig. 5. Levels of fumaric acid, malic acid, and citric acid in Kimchi samples fermented at 20℃ decreased rapidly during the early fermentation period. However, levels of malic acid and citric acid in Kimchi samples fermented at 4℃ decreased after 10 days of fermentation. These results were consistent with large changes in metabolites of Kimchi fermented at 4℃ between 10 and 50 days of fermentation as shown in Fig. 3A. Reduction in malic acid and citric acid levels indicated the presence of malolactic fermentation since LAB could convert malic acid and citric acid into lactic acid (Son et al., 2009). When lactic acid is produced in Kimchi, the growth of aerobic bacteria is suppressed (Jang and Kim, 2013). Park et al. (2012) have reported changes in bacterial community structure of Kimchi after a long period of incubation at room temperature (22℃). These changes in microbial communities might cause differences in metabolites, thus affecting the quality of Kimchi. Levels of propylene glycol and galactaric acid in Kimchi stored at 4℃ were increased sharply between the 10th and 50th day of fermentation. Interestingly, these metabolites showed very little changes during the entire fermentation period in Kimchi samples stored at 12 or 20℃. In this study, different contents of metabolites in Kimchi with fermentation temperatures indicated that fermentation temperature can affect microbial community and fermentation rate. Kimchi generally undergoes fermentation at low temperature (about 4-6℃). However, it will take weeks to finish the process. Commercially produced Kimchi is usually fermented for a short period (1-2 days) at 20℃. It is then stored at low temperature (5-8℃) to improve organoleptic quality (Hong et al., 1994; Lee et al., 2017). It is difficult to clearly determine the optimum duration of ripening time according to the fermentation temperature of Kimchi. Cheigh et al. (1994) have reported that optimized time for Kimchi fermented at 5 9
and 20℃ were 35-180 days and 2-3 days, respectively. According to a study of Chang and Kim (2000), optimized time for Kimchi ripening at 15, 10, and 5℃ were 4, 10, and 18 days, respectively. Similar to these studies, Lee et al. (1991) have reported that Kimchi’s shelf life 4 and 28℃ were 33 and 3 days, respectively, based on kinetic modeling of total acidity for quality prediction. Hong et al. (2016) have reported that the optimum ripening time for Kimchi fermented at 4 and 20℃ was 35 and 2 days, respectively, after analyzing volatile compositions of Kimchi. Nowadays, it is common to store homemade or commercially produced Kimchi products in a Kimchi refrigerator in Korea to reduce microbial growth and avoid over-fermenting or souring by rapid shifting to subzero temperatures (Hongu et al., 2017; Lee et al., 2008). Although the optimum ripening time for Kimchi is different, it obviously depends on storage temperature. Storage temperature of Kimchi is directly related to the growth of microorganisms which will affect Kimchi metabolites. Therefore, it is important to know the metabolic difference in Kimchi because it affects the quality of Kimchi. These results suggest that the optimum duration for Kimchi fermentation can be determined using a metabolomics approach to simultaneously observe metabolites during fermentation.
4. Conclusion A GC-MS based metabolomics approach coupled with multivariate statistics were applied in this study to understand relationships between fermentation temperature and metabolite differences of Kimchi. Metabolites of Kimchi fermented at 4℃ changed more slowly during early fermentation than those of Kimchi fermented at higher temperatures (4 and 20℃). After fermentation, different concentrations of metabolites were observed in Kimchi samples depending on fermentation temperature. It is possible to determine the optimum duration of Kimchi fermentation by metabolomics approach which can simultaneously observe 10
metabolites during fermentation. However, more detailed studies are needed to better understand relationships of sensory characteristics, microbial communities, and metabolites of Kimchi with fermentation temperature.
Acknowledgements This work was supported by a grant (NRF- 2014R1A1A1002817) from the National Research Foundation funded by the Korean Government.
Conflict of interest The authors declare that there are no conflicts of interest.
References Castro-Puyana, M., Herrero, M. Metabolomics approaches based on mass spectrometry for food safety, quality and traceability. TrAC Trends in Analytical Chemistry, 52, 74-87. Chang, M. J., Kim, M. H. (2000). Fermentation property of Chinese cabbage Kimchi by fermentation temperature and salt concentration. Applied Biological Chemistry, 43, 7-11. Cheigh, H. S., Park, K. Y., Lee, C. (1994). Biochemical, microbiological, and nutritional aspects of Kimchi (Korean fermented vegetable products). Critical Reviews in Food Science and Nutrition, 34, 175-203. Choi, S. Y., Kim, Y. B., Yoo, J. Y., Lee, I. S., Chung, K. S., Koo, Y. J. (1990). Effect of temperature and salts concentration of Kimchi manufacturing on storage. Korean Journal of Food Science and Technology, 22, 707-710. Farag, M. A., Porzel, A., Wessjohann, L. A. (2012). Comparative metabolite profiling and fingerprinting of medicinal licorice roots using a multiplex approach of GC–MS, LC–MS and 1D NMR techniques. Phytochemistry, 76, 60-72. Hong, S. P., Lee, E. J., Kim, Y. H., Ahn, D. U. (2016). Effect of fermentation temperature on the volatile composition of Kimchi. Journal of Food Science, 81, C2623-C2629. Hongu, N., Kim, A. S., Suzukim, A., Wilson, H., Tsui, K. C., Park, S. (2017). Korean Kimchi: Promoting healthy meals through cultural tradition. Journal of Ethnic Foods, 4, 172-180. 11
Hotelling H. (1931). The generalization of Student's ratio. The Annals of Mathematical Statistics, 2, 360-378. Jang, D. J., Chung, K. R., Yang, H. J., Kim, K. S., Kwon, D. Y. (2015). Discussion on the origin of Kimchi, representative of Korean unique fermented vegetables. Journal of Ethnic Foods, 2, 126-136. Jang, J. Y., Kim, T. W. (2013). Lactic acid bacteria in Kimchi and their immunomodulatory activities. Current Topic in Lactic Acid Bacteria and Probiotics, 1, 28-37. Jeong, S. H., Lee, S. H., Jung, J. Y., Choi, E. J., Jeon, C. O. (2013). Microbial succession and metabolite changes during long-term storage of Kimchi. Journal of Food Science, 78, M763-M769. Jung, J. Y., Lee, S. H., Jin, H. M., Hahn, Y., Madsen, E. L., Jeon, C. O. (2013). Metatranscriptomic analysis of lactic acid bacterial gene expression during Kimchi fermentation. International Journal of Food Microbiology, 163, 171-179. Jung, J. Y., Lee, S. H., Lee, H. J., Seo, H. Y., Park, W. S., Jeon, C. O. (2012). Effects of Leuconostoc mesenteroides starter cultures on microbial communities and metabolites during Kimchi fermentation. International Journal of Food Microbiology, 153, 378-387. Kim, D. W., Kim, B. M., Lee, H. J., Jang, G. J., Song, S. H., Lee, J. I., Lee, S. B., Shim, J. M., Lee, K. W., Kim, J. H., Ham, K. S., Chen, F., Kim, H. J. (2017). Effects of different salt treatments on the fermentation metabolites and bacterial profiles of Kimchi. Journal of Food Science, 82, 1124-1131. Lee, D., Kim, S., Cho, J., Kim, J. (2008). Microbial population dynamics and temperature changes during fermentation of kimjang Kimchi. The Journal of Microbiology, 46, 590-593. Lee, K. H., Cho, H. Y., Pyun, Y. R. (1991). Kinetic modelling for the prediction of shelf-life of Kimchi based on total acidity as a quility index. Korean Journal of Food Science and Technology, 23, 306-310. Lee, M., Song, J. H., Jung, M. Y., Lee, S. H., Chang, J. Y. (2017). Large-scale targeted metagenomics analysis of bacterial ecological changes in 88 Kimchi samples during fermentation. Food Microbiology, 66, 173-183. MacKenzie, D. A., Defernez, M., Dunn, W. B., Brown, M., Fuller, L. J., de Herrera, S. R., Günther, A., James, S. A., Eagles, J., Philo, M. (2008). Relatedness of medically important strains of Saccharomyces cerevisiae as revealed by phylogenetics and metabolomics. Yeast, 25, 501-512. Mheen, T. I., Kwon, T. W. (1984). Effect of temperature and salt concentration on Kimchi fermentation. Korean Journal of Food Science and Technology, 16, 443-450. Park, E. J., Chun, J., Cha, C. J., Park, W. S., Jeon, C. O., Bae, J. W. (2012). Bacterial community analysis during fermentation of ten representative kinds of Kimchi with barcoded pyrosequencing. Food Microbiology, 30, 197-204. 12
Park, J., Heo, G., Oh, Y., Kim, B., Miheen, T., Kim, C., Ahn, J., Lee, J. (2003). Change of microbial communities in Kimchi fermentation at low temperature. Korean Journal of Microbiology, 39, 45-50. Park, J. G., Kim, J. H., Park, J. N., Kim, Y. D., Kim, W. G., Lee, J. W., Hwang, H. J., Byun, M. W. (2008). The effect of irradiation temperature on the quality improvement of Kimchi, Korean fermented vegetables, for its shelf stability. Radiation Physics and Chemistry, 77, 497-502. Park, S. E., Yoo, S. A., Seo, S. H., Lee, K. I., Na, C. S., Son, H. S. (2016). GC–MS based metabolomics approach of Kimchi for the understanding of Lactobacillus plantarum fermentation characteristics. LWT-Food Science and Technology, 68, 313-321. Son, H. S., Hwang, G. S., Park, W. M., Hong, Y. S., Lee, C. H. (2009). Metabolomic characterization of malolactic fermentation and fermentative behaviors of wine yeasts in grape wine. Journal of Agricultural and Food Chemistry, 57, 4801-4809. Tabatabaei Yazdi, F., Alizadeh Behbahani, B., Mohebbi, M., Mortazavi, A., Ghaitaranpour, A. (2013). Effect of temperature on microbial changes during Kimchi fermentation. Scientific Journal of Microbiology, 2, 9-14. Tautenhahn, R., Boettcher, C., Neumann, S. (2008). Highly sensitive feature detection for high resolution LC/MS. BMC Bioinformatics, 9, 504. Vikram, A., Lui, L., Hossain, A., Kushalappa, A. (2006). Metabolic fingerprinting to discriminate diseases of stored carrots. Annals of Applied Biology, 148, 17-26.
Fig. 1. Changes in viable cells of total bacteria (A), pH (B), and titratable acidity (C) of Kimchi during ripening at different temperatures. Values are presented as mean ± standard deviations (n = 5). Fig. 2. PCA score plot derived from GC-MS data of Kimchi during ripening at different temperatures (4, 12, and 20℃). Each symbol (point) in the score plot represents a Kimchi sample. Symbols with different colors and shapes denote Kimchi samples taken at different temperatures on different days of fermentation (Color figure online). Fig. 3. PCA score plots derived from GC-MS data of Kimchi during ripening at 4℃ (A), 12℃ (B), and 20℃ (C), showing different metabolic changes (fermentation rates) 13
depending on different temperature. The color and shape of each point are the same as shown in Fig. 1 (Color figure online). Fig. 4. PCA (A) and PLS-DA (B) score plots derived from GC-MS data of Kimchi after 50 days of fermentation at different temperatures. A permutation test was carried out with 200 random permutations in the PLS-DA model (C). The color and shape of each point are the same as shown in Fig. 1. Fig. 5. Changes in identified metabolites of Kimchi during ripening at different temperatures. Metabolites contributing to differentiation in the PLS-DA model (VIP score > 1.5, p < 0.05) were selected. Values are presented as mean ± standard deviations (n = 5).
Highlights ● Effect of fermentation temperature on metabolic changes in Kimchi was investigated. ● Metabolites of Kimchi fermented at low temperature changed slowly during fermentation. ● Alanine, propylene glycol, fumaric acid, malic acid, citric acid, and galactaric acid concentrations were different depending on temperature. ● GC-MS based metabolomics approach is a valuable tool to determine the optimum duration of Kimchi fermentation
14
Fig. 1.
Total cell counts (log CFU/g)
(A) Total bacteria
11.0
4℃
12 ℃
20 ℃
10.0 9.0 8.0 7.0 6.0 5.0 4.0 0
5
10
15
20
25
30
35
40
45
50
Fermentation time (days)
(B) pH
7.0
4℃
12℃
20℃
6.0
pH
5.0 4.0 3.0 2.0 0
5
10
15
20
25
30
Fermentation time (days)
15
35
40
45
50
(C) TA
Titratable acidity (%)
2.0
4℃
12℃
20℃
1.5 1.0 0.5 0.0 0
5
10
15
20
25
30
35
Fermentation time (days)
Fig. 2.
16
40
45
50
Fig. 3. (A)
(B)
17
(C)
Fig. 4. (A)
18
(B)
(C)
19
Fig. 5. Fumaric acid
Malic acid
0.0050
0.012 4℃ 12℃ 20℃
0.010
Intensity
Intensity
0.0040 0.0030 0.0020
0.008 0.006 0.004
0.0010
0.002 0.000
0.0000 0
10
20
30
40
0
50
10
Fermentation time (days)
20
30
40
50
40
50
40
50
Fermentation time (days)
Citric acid
Galactaric acid
0.012
0.0020
0.010 0.0015
Intensity
Intensity
0.008 0.006
0.0010
0.004 0.0005 0.002 0.0000
0.000 0
10
20
30
40
0
50
10
β-alanine
30
Propylene glycol
0.14
0.006
0.12
0.005
0.10
0.004
Intensity
Intensity
20
Fermentation time (days)
Fermentation time (days)
0.08 0.06 0.04
0.003 0.002 0.001
0.02
0
0.00 0
10
20
30
40
0
50
10
20
30
Fermentation time (days)
Fermentation time (days)
20
21