International Journal of Food Microbiology 170 (2014) 61–64

Contents lists available at ScienceDirect

International Journal of Food Microbiology journal homepage: www.elsevier.com/locate/ijfoodmicro

An improved process of isomaltooligosaccharide production in kimchi involving the addition of a Leuconostoc starter and sugars Seung Kee Cho a, Hyun-Ju Eom b, Jin Seok Moon a, Sae-Bom Lim a, Yong Kook Kim c, Ki Won Lee d, Nam Soo Han a,⁎ a

Department of Food Science and Technology, Chungbuk National University, Cheongju 361-763, Republic of Korea Chungcheongbukdo Agricultural Research and Extension Services, Cheongwon 363-883, Republic of Korea Korea Food and Drug Administration, Cheongwon 363-700, Republic of Korea d Department of Agricultural Biotechnology, Seoul National University, Seoul 151-921, Republic of Korea b c

a r t i c l e

i n f o

Article history: Received 3 June 2013 Received in revised form 7 October 2013 Accepted 27 October 2013 Available online 5 November 2013 Keywords: Leuconostoc citreum Dextransucrase Isomaltooligosaccharide Prebiotics Synbiotics

a b s t r a c t Isomaltooligosaccharides (IMOs) are α-(1 → 6)-linked oligodextrans that show a prebiotic effect on Bifidobacterium spp. This study sought to improve IMO synthesis during lactate fermentation in kimchi by inoculating the kimchi fermentation mix with a starter and sugars; the psychrotrophic Leuconostoc citreum KACC 91035 strain with high dextransucrase activity was used as a starter and sucrose (58 mM) and maltose (56 mM) were added as the donor and acceptor for the glucose-transferring reaction of the dextransucrase, respectively. With the addition of both the starter and the sugars and incubation at 10 °C, IMOs were produced in kimchi after 3 d. Without the starter, the IMO production rate and maximal concentration in kimchi were 15.05 mM/d and 75.27 mM, respectively, whereas with the starter, the rate and concentration increased to 22.04 mM/d and 110.19 mM, respectively. In addition, the sucrose–maltose mix gave an appropriate level of sweetness by releasing fructose and prevented unfavorable polymer synthesis by IMO production. This result suggests that lactic acid bacteria expressing a highly active glycosyltransferase can be used for the synthesis of beneficial oligosaccharides in various fermented foods. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Lactic acid fermentation of vegetables, a valuable technique used by humans for thousands of years, is still investigated today because of the nutritional value of its products, which provide vitamins, minerals, and dietary fiber, and help the intestinal tract maintain a healthy balance of flora by increasing beneficial intestinal bacteria (Steinkraus, 1983). Among the diverse commercial products globally derived from fermented vegetables, kimchi made from cabbage is most economically relevant (Caplice and Fitzerald, 1999). Kimchi is a traditional Korean food produced by the fermentation of baechu-cabbage and other ingredients including radish, garlic, and red pepper powder (Cho et al., 1999). The growth patterns of lactic acid bacteria in kimchi and sauerkraut are similar, with the Leuconostoc bacterial genus dominating the initial lactic acid production phase during fermentation (Eom et al., 2008). During this stage, Leuconostoc produces metabolites such as dextran, lactate, acetate, alcohol, CO2, and mannitol, all of which contribute to the taste of the fermented foods (Chyun and Rhee, 1976). The ingestion of a typical watery kimchi, dongchimi, which is usually made from radish, produces 3 distinct taste sensations: a sour taste due to organic acids, a carbonated taste from CO2, and a sweet taste due to sugars and sugar alcohols. ⁎ Corresponding author. Tel.: +82 43 261 2567; fax: +82 43 271 4412. E-mail address: [email protected] (N.S. Han). 0168-1605/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijfoodmicro.2013.10.027

The enzyme dextransucrase (EC 2.4.1.5), expressed by various species of the genus Leuconostoc, transfers the glucosyl moiety of sucrose to form dextran, an α-(1 → 6)-linked D-glucan. Dextransucrase also catalyzes the transfer of glucose from sucrose (the donor) to other carbohydrates (the acceptors) by cross-linking the α-(1 → 6)-glucosyl bond. Using maltose as the acceptor molecule, several isomaltooligosaccharides (IMOs) were produced in an experiment using Leuconostoc mesenteroides NRRL B-512F (Robyt and Eklund, 1983). IMO is a representative prebiotic and an α-(1 → 6)-linked glucooligosaccharide. IMO found in commercial products shows a degree of polymerization that ranges from 2 to 6 (Crittenden and Playne, 1996; Han et al., 2002). A prebiotic resists digestion in the upper gastrointestinal tract, but is selectively metabolized in the colon by beneficial microbes and consequently offers health benefits (Gibson et al., 2004). Olano-Martin et al. (2000) reported that oligodextran can be used by bifidobacteria and lactobacilli, and it was registered as a health-promoting prebiotic by the Korean Food and Drug Administration (http://www.foodnara.go.kr/hfoodi/). In a previous study (Han et al., 2002), we applied the acceptor reaction of dextransucrase in kimchi by adding sucrose and maltose, and found that IMOs were synthesized during the lactate fermentation period as expected. The same reaction occurred in kefir-like milk fermentation when sucrose and maltose were added with a Leuconostoc starter (Seo et al., 2007). To enhance IMO production, we previously isolated psychrotrophic Leuconostoc citreum (L. citreum) KACC 91035 that has high dextransucrase activity (Eom et al., 2007). The current study

62

S.K. Cho et al. / International Journal of Food Microbiology 170 (2014) 61–64

used L. citreum KACC 91035 as a starter strain in the kimchi fermentation process to enhance the production of IMO during lactate fermentation. 2. Materials and methods 2.1. Materials, bacterial strains, and culture condition Sucrose, maltose, NaCl, and panose were purchased from Sigma Chemical Co. (St. Louis, MO, U.S.A.). L. citreum KACC 91035 used as the starter strain was cultured in MRS broth. A liquid medium to inoculate the kimchi was prepared by autoclaving and filtering the extracted cabbage. Cubic plastic jars with sealing lids were used as fermentation vessels. Radish, red pepper, green onion, and other kimchi ingredients were purchased from a local grocery store. 2.2. Preparation of kimchi (dongchimi-kimchi) To prepare dongchimi-kimchi, a whole radish (800 g) was minced and mixed with salt (40 g), green onions (20 g), garlic (10 g), and ginger (3 g). Next, the kimchi samples were sorted according to sugar and starter addition: A, kimchi with no added sugar (blank); B, kimchi with 58 mM sucrose and 56 mM maltose; and C, kimchi with 58 mM sucrose, 56 mM maltose, and the starter (107 colony forming units (CFU)/mL). The jars were filled to 4 L with drinking water and tightly sealed with plastic lids. The temperature during the fermentation process was maintained at 10 °C for 7 d, after which the temperature was dropped to 4 °C for preservation. 2.3. Microbial and chemical analyses The growth of lactic acid bacteria in the kimchi fermentation mix was measured in CFU/mL using the culture pouring method with Lactobacilli (LAB) MRS agar medium. The kimchi fermentation mix was diluted with 0.85% (w/v) physiological saline and poured on agar plates that were incubated at 28 °C for 48 h. The pH of the test solution was determined with a pH meter (IQ240; IQ Scientific Instruments, San Diego, CA, USA). 2.4. Sugar analysis The sugars present in the kimchi fermentation mix were analyzed using a High Performance Anion Exchange Chromatography (HPAEC) system (Bio-LC ICS-3000, Dionex Corp., Sunnyvale, CA, USA) fit with a CarboPac PA1 column (0.2 × 25 cm, Dionex) and a pulsed amperometric detector (ED50, Dionex). For quantitative and qualitative analyses of peaks, we used the software Chromate Window v.3.0 (Interface Engineering Inc., Portland, OR, USA). In parallel, TLC analysis was used to monitor oligosaccharide synthesis. For the analysis of mono- or di-saccharides, TLC plates (Whatman K5 TLC plates, Merck, Darmstadt, Germany) were developed thrice with acetonitrile/distilled water (85:15, v/v); the separated sugars were detected by dipping the plates in ethanol containing α-naphthol (0.5%, w/v) and sulfuric acid (5%, v/v), followed by heating at 110 °C for 5 min (Lee et al., 2008). 2.5. PCR-DGGE analysis The kimchi fermentation mix was grinded and filtered through sterilized cheese cloth twice. Genomic DNA from kimchi fermentation mix was extracted using the Genomic DNA Prep kit for bacteria (SolGent, Daejeon, Korea) according to manufacturer's instructions. The 16S rRNA gene V3 regions were amplified using the universal bacterial primers 338f and 518r. The sequences of the primers were as follows: 338f, 5′-ACTCCTACGGGAGGCAGCAG-3′ (Escherichia coli positions 338 to 357) and 518r, 5′-ATTACCGCGGCTGCTGG-3′. The forward primer, 338f, had a GC clamp. Amplified DNA sequences were resolved by electrophoresis at 60 °C, and to increase resolution, we used gels with a

30–60% urea: formamide denaturing gradient (100% corresponding to 7 M urea and 40% [wt vol−1] formamide) increasing in the direction of electrophoresis. Samples were electrophoresed for 30 min at 20 V and for 16 h at 60 V, stained with EtBr for 1 h, and photographed under UV illumination. For microbial identification, gel bands were purified with the Gel & PCR Purification System (SolGent) and sequenced after PCR re-amplification (without the GC-clamp); sequences were compared to those in the GenBank database using the BLAST algorithm (National Center Biotechnology Information, MA, USA). 3. Results 3.1. The fermentation profile of kimchi Three batches of dongchimi-kimchi were prepared: kimchi A, with no sugar; kimchi B, with sucrose and maltose; and kimchi C, with sucrose, maltose, and the starter. The batches were fermented at 10 °C for 7 d then stored at 4 °C thereafter (Fig. 1A); during this period, the biochemical changes in kimchi were monitored. The low temperature (10 °C) maintained during fermentation stage enables the Leuconostoc starter to become the dominant species and secrete dextransucrase. The initial pH in all 3 batches of kimchi was approximately 6.4, but after 1 d, the pH of kimchi C dropped rapidly to 4.2, a value lower than that in kimchi A and B; this is likely due to the increased growth of kimchi C as a result of starter addition. Over a 13-d monitoring period, the pH of kimchi A, B, and C decreased to 3.8, 3.7, and 3.7, respectively (Fig. 1B). Total LAB cell numbers of kimchi A and B increased slowly to approximately 108 CFU/mL after 7 d, whereas those numbers were reached in kimchi C after just 1 d (Fig. 1C). When PCR-DGGE analysis was used to monitor microbial dynamics during kimchi fermentation (Fig. 1D), Pseudomonas (bands a and b) and Enterobacteriaceae sp. (bands c and f) were initially found in kimchi B, but Leuconostoc gasicomitatum (band e) and Lactococcus piscium (band d) quickly became the dominant LAB species. In contrast, in kimchi C, L. citreum KACC 91035 (band g) was dominant throughout fermentation. 3.2. The effect of starter addition on IMO production The transfer of glucose to an acceptor molecule, which is catalyzed by dextransucrase, was used for IMO synthesis in kimchi. Maltose was added as the acceptor molecule during kimchi preparation, and the changes in sugar concentration were analyzed by TLC and HPAEC (Fig. 2). As expected, dextransucrase catalyzed the transfer of glucose from sucrose to the maltose acceptor, producing panose, isomaltosyl maltose (IMM), and isomaltotriosyl maltose (IM3M). The results of TLC (Fig. 2A) and HPAEC (Fig. 2B) analyses of kimchi C showed that sucrose (58 mM) was rapidly consumed within 3 d, and nearly half the maltose (28 mM), which acted as an acceptor molecule was used up. Simultaneously, the IMOs were synthesized and fructose was released from sucrose. The concentrations of oligosaccharides synthesized in kimchi B and C were also monitored (Fig. 3). IMOs were not synthesized in kimchi A to which no sugars were added (Fig. 3A). In contrast, with sucrose and maltose added in kimchi B, IMOs were immediately produced, and their concentrations peaked after 3 d and were maintained at that level for 13 d at 4 °C. Sucrose (58 mM) and maltose (56 mM) were converted into panose (50.49 mM), IMM (22.8 mM), and IM3M (1.98 mM). In kimchi C, which included the starter culture of L. citreum KACC 91035 in addition to sucrose and maltose, the production of IMO dramatically increased, with the total IMO concentration increasing to over 100 mM: panose (72.71 mM), IMM (32.53 mM), and IM3M (4.95 mM). As summarized in Table 1, in the non-starter-added kimchi B, the production rate and maximal concentration of IMOs were 15.05 mM/d and 75.27 mM, respectively, whereas in the starter-added kimchi C the rate and concentration were increased to 22.04 mM/d and 110.19 mM, respectively. After the IMOs had accumulated in kimchi, their concentrations decreased

S.K. Cho et al. / International Journal of Food Microbiology 170 (2014) 61–64

63

Fig. 2. Time course of changes in individual sugar levels during kimchi (Sample C) fermentation with 58 mM sucrose, 56 mM maltose, and starter. A, The sugar content monitored by TLC (developer solution, acetonitrile:H2O = 85:15, v/v). B, The sugar content analyzed using HPAEC at 0 d (i), 7 d (ii), and 13 d (iii). Fru, fructose; Glc, glucose; Suc, sucrose; Mal, maltose; Pan, panose; Std, standard molecules. IMM, isomaltosyl maltose; IM3M, isomaltotriosyl maltose.

Fig. 1. Profiles of temperature control (A), pH change (B), and cell growth in the microbial populations (C, D) during kimchi fermentation. ■, kimchi with no sugar; ○, kimchi with 58 mM sucrose and 56 mM maltose; ●, kimchi with 58 mM sucrose, 56 mM maltose, and starter (107 CFU/ml); (I) kimchi with 58 mM sucrose and 56 mM maltose; (II) kimchi with 58 mM sucrose, 56 mM maltose, and starter (107 CFU/mL); R, reference strain; a, Pseudomonas simiae; b, Pseudomonas strain; c, Enterobacteriaceae bacterium; d, Lactococcus piscium; e, Leuconostoc gasicomitatum; f, Enterobacteriaceae bacterium; g, Leuconostoc citreum KACC 91035.

slightly during extended storage at a low temperature, indicating a slow microbial degradation. 4. Discussion In this study, we developed a new strategy for producing oligosaccharides in kimchi using a biocatalytic reaction simultaneously with lactate fermentation. This is a simple and innovative process for manufacturing nutritious kimchi containing prebiotic oligosaccharides. To use Leuconostoc as a starter culture in the production of kimchi-like fermented foods, we isolated the psychrotrophic L. citreum KACC 91035 strain that grows rapidly and has high enzyme activity at low temperatures (Eom et al., 2007): at 8 °C, KACC 91035 showed a robust growth (0.24 h−1) and dextransucrase activity (19 units/mL). Robyt and Eklund (1983) carried out a series of reactions with dextransucrase from L. mesenteroides B-512F using different acceptors at a 1:1 acceptorto-sucrose ratio; they found maltose to be the best acceptor (100%), followed by isomaltose (89%), and then nigerose (58%), methyl-α-D-

glucopyranoside (52%), D-glucose (17%), turanose (13%), lactose (11%), cellobiose (9%), and D-fructose (6.4%). These results indicate that diverse oligosaccharides can be produced in kimchi using a method based on dextransucrase-dependent transfer of glucose to the above acceptor compounds. Microbial communities in cabbage kimchi contain several LAB species, including L. mesenteroides, Lactobacillus sakei, Wiessella koreensis, L. gasicomitatum, L. gelidum, and L. carnosum (Eom et al., 2007; Park et al., 2003, 2010), although L. citreum and L. gasicomitatum are dominant in dongchimi-kimchi (Jeong et al., 2013). L. gasicomitatum in dongchimi-kimchi is also found in cold-storage foods that are modified-atmosphere-packaged (MAP) and in nutrient-rich foods (Jääskeläinen et al., 2013; Johansson et al., 2011; Nieminen et al., 2012), but the bacterium tends to cause food spoilage. Therefore, we used the psychrotrophic L. citreum KACC 91035 strain with high dextransucrase activity, which can be isolated from various fermented foods. L. gasicomitatum has 2 dextransucrase genes, epsA (LEGAS_0699) and dsrA (LEGAS_1012), which encode 150 kDa and 170 kDa proteins, respectively, which share a 51.5% amino acid sequence identity. In contrast, L. citreum KACC 91035 (Yi et al., 2009) has a single dextransucrase gene (4431-bp long) encoding an enzyme (of 1477 amino acids, 164 kDa) that shares 45.1% and 45.5% sequence identities with epsA and dsrA, respectively. Although their similarity is not high, these enzymes all efficiently synthesize IMOs in kimchi using sucrose and maltose, indicating that the glucose-transfer reaction of dextransucrases in the 2 strains is almost the same.

64

S.K. Cho et al. / International Journal of Food Microbiology 170 (2014) 61–64

which also imparts a sweet taste (Otgonbayar et al., 2011). Maltose syrup (50% maltose) is easily found in markets and can be used as an economical source of maltose for IMO synthesis in kimchi. In conclusion, in the method presented here, the glucose-transferring reaction of L. citreum dextransucrase is used to enable the production of beneficial oligosaccharides in kimchi, while both maintaining an appropriate level of sweetness and preventing unfavorable polymer synthesis in the kimchi. The application of this method, referred to as biocatalytic synthesis of IMOs during lactate fermentation, should permit the development of lactate foods with new nutritionally beneficial molecules that may provide health benefits. Acknowledgments This work was partially supported by the grant from Korea Research Foundation (2010-0029233) funded by MEST and by the grant from Korea Rural Development Administration (Project No. PJ90715303). References

Fig. 3. Profile of sugar changes in kimchi (A), in kimchi with 58 mM sucrose and 56 mM maltose (B), and in kimchi with 58 mM sucrose, 56 mM maltose, and starter (C). ●, panose; ▲, isomaltosyl maltose; ■, isomaltotriosyl maltose; ○, total IMO. Data are from three independent experiments.

L. citreum, one of the major bacteria in kimchi, provides appropriate conditions for lactate fermentation that gives kimchi its unique taste. The bacterium grows rapidly during the initial stage of kimchi fermentation and scavenges oxygen from the liquid to create anaerobic conditions for lactic acid fermentation. The bacterium also produces lactic acid and CO2 through the hetero-lactic acid fermentation pathway, giving a sour and carbonated taste to kimchi. The sweet taste is an important attribute of kimchi, but traditionally sucrose has not been used in kimchi preparation because the dextran polymer that sucrose produces makes the kimchi viscous and sticky, and hence unsavory. To prevent this, alternative sweeteners such as saccharin are often used in kimchi production, but this raises health concerns. The method developed in this study solves the polymer problem, whi1e also meeting the sweetener requirement: maltose inhibits the synthesis of dextran; additionally sucrose, after the glucosetransfer reaction, releases an equivalent amount of fructose as a free molecule that sweetens the kimchi. During kimchi fermentation, the fructose molecules are converted by Leuconostoc spp. into mannitol,

Table 1 Comparison of IMO concentrations in kimchi after completion of the dextransucrase reaction. Types of kimchi

Pan (mM)

IMM (mM)

IM3M (mM)

Total IMOs (mM)

Production rate (mM/d)

No sugar (A) Suc + Mal (B) Suc + Mal + Starter (C)

0 50.49 72.71

0 22.8 32.53

0 1.98 4.95

0 75.27 110.19

0 15.05 22.04

Pan, panose; IMM, isomaltosyl maltose; IM3M, isomaltotriosyl maltose; Total IMOs, Total isomaltooligosaccharides.

Caplice, E., Fitzerald, G.F., 1999. Food fermentations: role of microorganisms in food production and preservation. Int. J. Food Microbiol. 50, 131–149. Cho, E.J., Rhee, S.H., Park, K.Y., 1999. Standardization of materials of baechu kimchi recipe. J. Korean Soc. Food Sci. Nutr. 30, 1456–1463. Chyun, J.H., Rhee, H.S., 1976. Studies on the volatile fatty acids and carbon dioxide produced in different kimchis. J.Korean Food Sci.Technol. 8, 90–94. Crittenden, R.G., Playne, M.J., 1996. Production, properties and applications of food-grade oligosaccharides. Trends Food Sci. Technol. 7, 353–361. Eom, H.J., Seo, D.M., Han, N.S., 2007. Selection of psychrotrophic Leuconostoc spp. producing highly active dextransucrase from lactate-fermented vegetables. Int. J. Food Microbiol. 117, 61–67. Eom, H.J., Park, J.M., Seo, M.J., Kim, M.D., Han, N.S., 2008. Monitoring of Leuconostoc mesenteroides DRC starter in fermented vegetable by random integration of chloramphenicol acetyltransferase gene. J. Ind. Microbiol. Biotechnol. 35, 953–959. Gibson, G.R., Probert, H.M., Loo, J.V., Rastall, R.A., Roberfroid, M.B., 2004. Dietary modulation of the human colonic microbiota: updating the concept of prebiotics. Nutr. Res. Rev. 17, 259–275. Han, N.S., Jung, Y.S., Eom, H.J., Koh, Y.H., Robyt, J.F., Seo, J.H., 2002. Simultaneous biocatalytic synthesis of panose during lactate fermentation in kimchi. J. Microbiol. Biotechnol. 12, 46–52. Jääskeläinen, E., Johansson, P., Kostiainen, O., Nieminen, T., Schmidt, G., Somervuo, P., Mohsina, Marzia, Vanninen, P., Auvinen, P., Björkrotha, J., 2013. Significance of heme-based respiration in meat spoilage caused by Leuconostoc gasicomitatum. Appl. Environ. Microbiol. 79, 1078–1085. Jeong, A.H., Jung, J.Y., Lee, S.H., Jin, H.M., Jeon, C.O., 2013. Microbial succession and metabolite changes during fermentation of dongchimi, traditional Korean watery kimchi. Int. J. Food Microbiol. 164, 46–53. Johansson, P., Paulin, L., Säde, E., Salovuori, N., Alatalo, E.R., Björkroth, K.J., Auvinen, P., 2011. Genome sequence of a food spoilage lactic acid bacterium, Leuconostoc gasicomitatum LMG 18811T, in association with specific spoilage reactions. Appl. Environ. Microbiol. 77, 4344–4351. Lee, M.S., Cho, S.K., Eom, H.J., Kim, S.Y., Kim, T.J., Han, N.S., 2008. Optimized substrate concentrations for production of long-chain isomaltooligosaccharides using dextransucrase of Leuconostoc mesenteroides B-512F. J. Microbiol. Biotechnol. 18, 1141–1145. Nieminen, T., Koskinen, K., Laine, P., Hultman, J., Säde, E., Paulin, L., Paloranta, A., Johansson, P., Auvinen, P., Björkrotha, J., 2012. Comparison of microbial communities in marinated and unmarinated broiler meat by metagenomics. Int. J. Food Microbiol. 157, 142–149. Olano-Martin, E., Mountzouris, K.C., Gibson, G.R., Rastall, R.A., 2000. In vitro fermentability of dextran, oligodextran and maltodextrin by human gut bacteria. Br. J. Nutr. 83, 247–255. Otgonbayar, G.-E., Eom, H.Y., Kim, B.S., Ko, J.H., Han, N.S., 2011. Mannitol production by Leuconostoc citreum KACC 91348P isolated from kimchi. J. Microbiol. Biotechnol. 21, 968–971. Park, J.A., Heo, G.Y., Oh, Y.J., Kim, B.Y., Miheen, T.I., Kim, C.K., Ahn, J.S., 2003. Change of microbial communities in kimchi fermentation at low temperature. Korean J. Microbiol. 39, 45–50. Park, J., Shin, J., Lee, D., Song, J., Suh, H., Chang, U., Kim, J., 2010. Identification of the lactic acid bacteria in kimchi according to initial and over-ripened fermentation using PCR and 16S rRNA gene sequence analysis. Food Sci. Biotechnol. 19, 541–546. Robyt, J.F., Eklund, S.H., 1983. Relative quantitative effects of acceptors in the reaction of Leuconostoc mesenteroides B-512F dextransucrase. Carbohydr. Res. 121, 279–286. Seo, D.M., Kim, S.Y., Eom, H.J., Han, N.S., 2007. Synbiotic synthesis of oligosaccharides during milk fermentation by addition of Leuconostoc starter and sugars. J. Microbiol. Biotechnol. 17, 1758–1764. Steinkraus, K.H., 1983. Lactic acid fermentation in the production of foods from vegetables, cereals and legumes. Antonie Van Leeuwenhoek 49, 337–348. Yi, A.R., Lee, S.R., Jang, M.U., Park, J.M., Eom, H.J., Han, N.S., Kim, T.J., 2009. Cloning of dextransucrase gene from Leuconostoc citreum HJ-P4 and its high-level expression in E. coli by low temperature induction. J. Microbiol. Biotechnol. 19, 829–835.