Biotechnology Letters (2005) 27: 103106

Springer 2005

Simultaneous and selective production of levan and poly(c-glutamic acid) by Bacillus subtilis
Ing-Lung Shih*,1 & Yun-Ti Yu
Department of Environmental Engineering, Da-Yeh University, Chang-Hwa, Taiwan 1 Present address: 112 Shan-Jiau Rd, Da-Tsuen, Chang-Hwa, Taiwan 51505 *Author for correspondence (Fax:+886-4-8511344; E-mail: ils@mail.dyu.edu.tw)
Received 30 July 2004; Revisions requested 28 August 2004/14 October 2004; Revisions received 11 October 2004/22 November 2004; Accepted 24 November 2004

Key words: Bacillus subtilis, levan, poly(c-glutamic acid) Abstract Bacillus subtilis (natto) Takahashi, used to prepare the fermented soybean product natto, was grown in a basal medium containing 5% (w/w) sucrose and 1.5% (w/w) L -glutamate and produced 58% (w/w) poly(cglutamic acid) and 42% (w/w) levan simultaneously. After 21 h, 4050 mg levan ml)1 had been produced in medium containing 20% (w/w) sucrose but without L -glutamate. In medium containing L -glutamic acid but without sucrose, mainly poly(c-glutamic acid) was produced.

Introduction Poly(c-glutamic acid), c-PGA, is an anionic, naturally-occurring, water-soluble polyamide composed solely of glutamic acid units (Kunioka 1997, Shih et al. 2002). It is biodegradable, edible and non-toxic toward humans and the environment. It has applications in a broad range of elds including food, cosmetics, medicine and watertreatment (Shih & Van 2001). Levan is a polymer of fructose linked by b-(26)fructofuranosidic bond and is found in many plants and microbial products (Han 1990). Microbial levans are produced from sucrose-based substrates by the transfructosylation reaction of levansucrase (b-2,!6 fructan:D-glucose-1-fructosyltransferase, E.C. 2.4.1.10) by a variety of microorganisms (Han 1990). Levan has a variety of industrial applications in cosmetics, foods and pharmaceuticals; it can be used as a gum, blood plasma extender or sweetener (Han 1990). Bacillus subtilis (natto) Takahashi, which is used for natto fermentation from soybean, has now been used to simultaneous or selective production of poly(c-glutamic acid) and levan.

Materials and methods Microorganism and growth Bacillus subtilis (natto) Takahashi, from Takahashi Yuzo research facility Japan, was grown in a medium composed of MgSO4 7H2O (0.5 g l)1), NaH2PO4 2H2O (3 g l)1), Na2HPO4 12H2O (3 g l)1) and supplemented with sucrose (50 g l)1) and L -glutamic acid (15 g l)1). The bacteria were inoculated into 100 ml medium in a 250 ml ask and incubated at 37 C, pH 7 with shaking at 150 rpm for 21 h. Poly(c-glutamic acid) and levan mixture was harvested by precipitation from the culture broth by addition of 4 vols. cold ethanol, followed by dialysis through a membrane with 10 kDa cuto. The products were characterized by amino acid analysis and C13-NMR and gel permeation chromatography (GPC). For selective production of c-PGA, the bacteria were grown in Medium E (Leonard et al. 1958) composed of glutamic acid (20 g l)1), citric acid (12 g l)1), glycerol (80 g l)1), NH4Cl (7 g l)1), MgSO4 7 H2O (0.5 g l)1), FeCl3 6 H2O (0.04 g l)1), K2HPO4 (0.5 g l)1), and CaCl2 7 H2O (0.15 g l)1).

104 The c-PGA was prepared and puried by a method described previously (Shih et al. 2002). Analytical methods The number-average molecular weights (Mn) of the c-PGA and levan were measured by GPC using a Hitachi L6200 system controller equipped with Shodex KB800 series columns and a refractive index (RI) detector (Shih et al. 2002). Dextran standards (Phenomenex, USA) were used to construct a calibration curve. The eluant used was deionized water at 1 ml min)1, the column was at 50 C. The total carbohydrate contents of the products were determined by the phenol/sulfuric acid method (Dubois et al. 1956) and expressed as glucose equivalents. C13-NMR spectroscopy was performed at 56 MHz with a Varain Unity Inova 600 spectrometer. Samples for NMR were dissolved in D2O. Optical rotation was measured on a polarimeter with a sodium lamp and a 100 mm sample tube. amino acid analysis suggesting the presence of poly(glutamic acid). The C13-NMR spectrum showed 11 main resonances (60.1, 63.6, 75.4, 76.5, 80.5, and 104.4 ppm for levan; 178, 174, 55, 33, and 28 ppm for c-PGA), which further conrmed the co-existence of both c-PGA and levan in the product. The eects of the concentrations of L -glutamic acid and sucrose on the total product (c-PGA plus levan) yields, and the proportion of each product in the total are shown in Tables 1 and 2. The results suggest that selective production of c-PGA or levan is possible if the bacteria are grown in a medium with the absence of sucrose or L -glutamate, respectively. Selective production of levan Bacillus subtilis (natto) Takahashi selectively produced levan when it was grown in a basal medium containing 20% (w/w) sucrose but without L -glutamate. Usually cultivation for 21 h was needed to give the maximum yields (Figure 1). The optimum pH for growth and levan production was 6, the suitable temperature range for growth and levan production was from 2540 C. The bacteria produced about 45 g of levan in 21 h (about 50% yields on available fructose) in 100 ml medium after precipitation by 4 vols. cold ethanol. The levan produced by the Takahashi strain was readily soluble in water at room temperature but was non-hygroscopic. It was easily hydrolyzed by boiling in 0.5% oxalic acid and its specic optical rotation is )42.0. The C13-NMR spectrum showed six main resonances at 60.1, 63.6, 75.4, 76.5, 80.5, and 104.4 ppm.

Results and discussion Simultaneous production of c-PGA and levan Bacillus subtilis (natto) Takahashi grown in a basal medium containing 5% (w/w) sucrose and 1.5% (w/w) L -glutamate produced poly(c-glutamic acid) and levan simultaneously. The presence of polysaccharide, 42% (w/w), in the product was conrmed by the phenol/sulfuric acid method. The hydrolysate of the dialyzed product obtained with 6 M HCl, was composed solely of glutamic acid by

Table 1. The eects of sucrose concentrations on the total product (levan and c-PGA) yields, and the proportion of each component in the totala. Sucrose concentration (g l)1) Total product yield (g l)1) Proportion of each component Levan (%) 0 20 50 70 100 200 1 7 12 17 30 64 2 21 42 67 86 89 c-PGA (%) 98 79 58 33 14 11

a B. subtilis (natto) Takahashi was grown in a medium composed of MgSO4 7H2O (0.5 g l)1), NaH2PO4 2H2O (3 g l)1), Na2HPO4 12H2O (3 g l)1) and supplemented with L -glutamic acid (15 g l)1) and various concentrations of sucrose at 37 C, pH 7 with shaking at 150 rpm.

105
Table 2. The eects of L -glutamate concentrations on the total product (levan and c-PGA) yields, and the proportion of each component in the totala.
L -Glutamate

concentration (g l)1)

Total product yield (g l)1)

Proportion of each component Levan (%) c-PGA (%) 9 16 21 23 26 35 31 36

0 1 5 10 15 20 30 40

8 8 9 11 12 12 14 15

91 84 79 77 74 65 69 64

a B. subtilis (natto) Takahashi was grown in a medium composed of MgSO4 7H2O (0.5 g l)1), NaH2PO4 2H2O (3 g l)1), Na2HPO4 12H2O (3 g l)1) and supplemented with sucrose (50 g l)1) and various concentrations of L -glutamic acid at 37 C, pH 7 with shaking at 150 rpm.

0.5 0.45 0.4

50 45 40 35 30 25 20 15 Cell dry weight (g l1) Levan yield (g 0 7 14 21 31 46 Time (h) 10 5 0 54 69 78

Selective production of c-PGA Bacillus subtilis (natto) Takahashi selectively produced c-PGA without any polysaccharide byproduct in Medium E (Leonard et al. 1958). This was evidenced by the HCl hydrolysate of the product being composed solely of glutamic acid. Less than 1% (w/w) polysaccharide was detected in the product by the phenol/sulfuric acid method. The number-average molecular weight (Mn) of c-PGA produced by Takahashi strain was 720 kDa as determined by GPC. In contrast, the number-average molecular weight of c-PGA produced by Bacillus licheniformis or B. subtilis IFO3335 was over 106 Da (Kunioka & Goto 1994). The C13-NMR spectrum shows ve main resonances at 178, 174, 55, 33, and 28 ppm which are almost identical with that of the authentic sample previously produced by B. licheniformis (Shih et al. 2002). References
Dubois M, Gilles KA, Hamilton JK, Rebers PA, Smith F (1956) Colorimetric method for determination of sugars and related substances. Anal. Chem. 28: 350356. Han YW (1989) Levan production by Bacillus polymyxa. J. Ind. Microbiol. 4: 447451. Han YW (1990) Microbial levan. Adv. Appl. Microbiol. 35: 171194. Han YW, Clarke MA (1990) Production and characterization of microbial levan. J. Agric. Food Chem. 38: 393396. Kunioka M (1997) Biosynthesis and chemical reactions of poly(amino acid)s from microorganisms. Appl. Microbiol. Biotechnol. 47: 469475.

Dry cell weight (g l1)

0.35 0.3 0.25 0.2 0.15 0.1 0.05 0

l1)

Fig. 1. Time course of cell growth and levan production by B. subtilis (natto) Takahashi, which was grown in a medium composed of MgSO4 7H2O (0.5 g l)1), NaH2PO4 2H2O (3 g l)1), Na2HPO4 12H2O (3 g l)1) and sucrose (200 g l)1) at 37 C, pH 7 with shaking at 150 rpm.

The above properties are consistent with what were reported for the levan produced by Bacillis polymyxa (Han 1989, Han & Clarke 1990). The levan product, after dialysis through a membrane with 10 kDa cuto, gave two sharp clean peaks on GPC chromatogram (not shown); one has molecular weight of 1794 kDa, the other has molecular weight of 11.9 kDa. In contrast, the levan product produced by B. polymyxa gave only a single, sharp clean peak just below 2 106 Da on Sephacryl S-500 after dialysis through a membrane with 12 kDa cuto (Han 1989, Han & Clarke 1990).

Levan yield (g l1)

106
Kunioka M, Goto A (1994) Biosynthesis of poly(c-glutamic acid) from L -glutamic acid, citric acid, and ammonium sulfate in Bacillus subtilis IFO3335. Appl. Microbiol. Biotechnol. 40: 867872. Leonard CG, Housewright RD, Thorne, CB (1958) Eects of some metallic ions on glutamyl polypeptide synthesis by Bacillus subtilis. J. Bacteriol. 76: 499503. Shih IL, Van YT (2001) The production of poly(c-glutamic acid) from microorganisms and its various applications. Bioresour Technol. 79: 207225. Shih IL, Van YT, Chang YN (2002) Application of statistical experimental methods to optimize production of poly(cglutamic acid) by Bacillus licheniformis CCRC 12826. Enzyme Microb. Technol. 31: 213220.