Abstract
Streptomyces exfoliatus F3-2 produced an extracellular enzyme that converted levan, a β-2,6-linked fructan, into levanbiose. The enzyme was purified 50-fold from culture supernatant to give a single band on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The molecular weights of this enzyme were 54,000 by SDS-PAGE and 60,000 by gel filtration, suggesting the monomeric structure of the enzyme. The isoelectric point of the enzyme was determined to be 4.7. The optimal pH and temperature of the enzyme for levan degradation were pH 5.5 and 60°C, respectively. The enzyme was stable in the pH range 3.5 to 8.0 and also up to 50°C. The enzyme gave levanbiose as a major degradation product from levan in an exo-acting manner. It was also found that this enzyme catalyzed hydrolysis of such fructooligosaccharides as 1-kestose, nystose, and 1-fructosylnystose by liberating fructose. Thus, this enzyme appeared to hydrolyze not only β-2,6-linkage of levan, but also β-2,1-linkage of fructooligosaccharides. From these data, the enzyme from S. exfoliatus F3-2 was identified as a novel 2,6-β-d-fructan 6-levanbiohydrolase (EC 3.2.1.64).
It has been reported that oligosaccharides appear to exhibit various physiological functions according to their structures. As the functions of the oligosaccharides are recognized, the production methods become important. An attempt to synthesize a large number of novel oligosaccharides has been made by using sugar hydrolases or sugar transferases. We have been studying the effective production of novel oligosaccharides (11, 13–15, 19–22) from such unused natural fructans as levan and inulin by the microbial polysaccharide-degrading enzymes. Also we confirmed that two of these oligosaccharides, di-d-fructose-1,2′:2,3′-dianhydride (DFA III) (17) and di-d-fructose-2,6′:6,2′-dianhydride (DFA IV) (12), had effects on increasing calcium absorption in the small intestine of rats, which could be expected to have practical applications in such fields as treatment with functional food or medicine to prevent osteoporosis.
In the course of these studies, Streptomyces exfoliatus F3-2 was isolated as a strain producing a levan-degrading enzyme (LDE) that effectively produces levanbiose from levan (21). Levan is a polymer consisting of β-2,6-linked fructose residues and found in monocotyledons as a reserve carbohydrate or synthesized by microbial levansucrase (EC 2.4.1.10) from sucrose (3, 4). In contrast to a plant levan, a microbial levan generally has a lot of branching points at the C-1 position, except for a levan from Serratia levanicum (4, 5). In our previous study (21), we demonstrated that the combination of the LDE from S. exfoliatus F3-2 and levan from S. levanicum was quite efficient for the large-scale preparation of levanbiose, with a maximum production yield of levanbiose of 84% (wt/wt) from 50 mg of levan per ml. The linear nature of the levan molecule from S. levanicum seems to contribute to this high yield.
To date, three types of microbial LDEs have been reported: levanase (EC 3.2.1.65), which hydrolyzes levan into fructose (1); 2,6-β-d-fructan 6-levanbiohydrolase (EC 3.2.1.64), which hydrolyzes levan into levanbiose (2); and levan fructotransferase, which produces DFA IV from levan (11, 13, 18). The enzyme from Pseudomonas sp. was the only enzyme reported as a 2,6-β-d-fructan 6-levanbiohydrolase. However, the enzyme was not purified completely, and the enzymological properties of levanbiose-producing LDE are still unclear (2).
Thus the purification and characterization of the levanbiose-producing LDE from S. exfoliatus F3-2 were conducted in order to investigate the properties of the enzyme and to obtain more information on the effective production of levanbiose.
MATERIALS AND METHODS
Culture conditions for S. exfoliatus F3-2 and preparation of crude LDE.
The culture conditions and media for S. exfoliatus F3-2 were those of our previous report (21), except that the optimized LDE production medium was used for LDE production. The optimized LDE production medium consisted of (per liter) 10 g of levan (prepared from culture of S. levanicum), 2 g of NH4NO3, 0.5 g of MgSO4 · 7H2O, 0.5 g of KCl, 0.5 g of KH2PO4, 0.01 g of FeSO4 · 7H2O, 0.5 g of meat extract (Wako Pure Chemical Industries Ltd., Osaka, Japan), 20 g of CaCO3, and NaOH to adjust the pH to 7.0. Crude LDE was prepared as reported previously (21) by culturing strain F3-2 in optimized LDE production medium for 36 h at 32°C.
Chemicals.
Levan was prepared from a culture of S. levanicum (5) by the method described previously (21). Levanbiose was enzymatically prepared from levan by using the LDE from S. exfoliatus F3-2 (21). For preparation of levantriose, 2 g of levan was partially hydrolyzed with 10 ml of 0.01 N HCl at 80°C for 30 min. Hydrolysates were then put on an activated charcoal column (2.6 by 100 cm) washed with deionized water, and sugars in the column were eluted with deionized water. Levantriose-containing fractions (20 ml each) were detected by thin-layer chromatography (TLC), combined, concentrated under reduced pressure, and lyophilized. TLC was done by using a silica gel plate (silica gel 60; Merck) with a solvent system of 1-butanol–2-propanol–water–acetic acid (7:5:4:2 [vol/vol]) and developed at 12 cm twice. For detection of spots, the plate was sprayed with a reagent consisting of p-anisaldehyde–H2SO4–ethanol (1:1:18 [vol/vol]), dried, and kept at 120°C until spots appeared (19). Fructooligosaccharides (i.e., 1-kestose, nystose, and 1-fructosylnystose) were gifts from Meiji Seika Kaisya, Ltd., Tokyo, Japan. Protein molecular weight markers were obtained from Roche Diagnostics GmbH, Mannheim, Germany.
Enzyme assay.
The enzyme activity was assayed under the standard LDE assay conditions as follows. The reaction mixture contained 50 mM sodium phosphate buffer (pH 5.5) and 10 mg of levan per ml, and the enzyme solution was diluted appropriately in a total volume of 1 ml. The mixture was incubated at 40°C for 10 min and then heated in boiling water for 5 min to stop the enzyme reaction. Next to check the presence of the LDE activity, sugars in the reaction mixture were analyzed qualitatively by TLC. After that, for quantitative determination of enzyme activity, the reducing sugars in the reaction mixtures were assayed by the Somogyi-Nelson method (16). One unit of enzyme activity was defined as the amount of the enzyme that produced 1 μmol of reducing sugars as fructose per min under these assay conditions. The protein concentration was determined by the methods of Lowry et al. (7). Unless otherwise mentioned, this method was used for the LDE assay in the experiments.
Purification of LDE.
All purification processes were done at 4°C. Crude LDE (1,000 ml [1,789 U]) was concentrated in dialysis tubes with polyethylene glycol 20,000 for several hours. To the concentrated solution, solid ammonium sulfate was added to give 60% saturation. The precipitate formed was collected, dissolved in 20 ml of 0.1 M sodium phosphate buffer (pH 6.0), and dialyzed against equilibration buffer for column chromatography. Anion-exchange column chromatography was done on a DEAE-Toyopearl 650M column (2.2 by 18 cm) (Tosoh Co., Ltd., Tokyo, Japan) equilibrated with 10 mM sodium phosphate buffer (pH 6.0). The enzyme was eluted with a linear gradient of 0 to 0.15 M NaCl in the same buffer, and fractions were collected at 7.5 ml each. The LDE activity-positive fractions which gave single band by SDS-PAGE were combined and were used as the purified sample. SDS-PAGE was done on a 10% polyacrylamide gel by the method of Laemmli (6). The gel was stained with silver stain.
Estimation of molecular weight of LDE.
A TSKgel G3000SWXL high-performance liquid chromatography (HPLC) column (7.8 by 300 mm) (Tosoh Co., Ltd.) was used with 10 mM sodium phosphate buffer (pH 6.0) containing 0.2 M NaCl for equilibration and elution buffer at a flow rate of 0.8 ml/min. The eluted proteins were fractionated by monitoring A280, and the LDE activity-positive fractions were located.
Isoelectric point of LDE.
Isoelectric focusing was done by using a Rotofor apparatus (Bio-Rad Laboratories, Hercules, Calif.) with Bio-Lyte 3/5 (Bio-Rad Laboratories), and the LDE activity-positive fractions were located.
Reaction products from levan by the LDE.
The degradation reaction was performed under the same conditions as those of the standard LDE assay conditions, except that the final concentration of the purified LDE was 0.8 U/ml and an incubation time of up to 24 h was employed. At various time points, the reaction mixtures were heated in boiling water for 5 min to stop the enzyme reaction, and the sugars were analyzed by TLC and HPLC. HPLC analyses were done under the conditions described previously (YMC-Pack ODS-AQ column, 6 by 250 mm [YMC Co. Ltd., Kyoto, Japan]; mobile phase, water; refractive index detector) (20) to determine the concentration of the reaction products by using fructose and levanbiose as standards.
Mode of degradation by LDE.
To address whether LDE hydrolyzes levan in an endo- or exo-acting manner, the degradation reaction was conducted under standard reaction conditions, except that purified LDE concentrations of 0.04 and 0.4 U/ml and a reaction time of up to 24 h were employed. Sugars in the reaction mixtures were analyzed at appropriate intervals quantitatively by the Somogyi-Nelson method and HPLC, both by using levanbiose as the standard and qualitatively by TLC. The ratio of the amount of reducing sugars versus the amount of levanbiose determined by HPLC was then calculated. This value would be 1 if the reducing sugar in the reaction mixture consisted solely of levanbiose, while a value of higher than 1 would be expected if the reducing sugars other than levanbiose were produced in the reaction mixture. The calculated values were correlated with the distribution of sugars detected by TLC analyses of the reaction mixtures. The mode of action was deduced from the values of the initial reaction products, assuming that a value around 1 results from an exo-acting manner and a value higher than 1 results from an endo-acting manner.
Substrate specificity of LDE.
For substrate specificity of the purified LDE, 10 mg of levan per ml as a substrate under the standard LDE assay conditions was replaced with the same amount of various kinds of sugars, and a final enzyme concentration of 0.8 U/ml was employed. The substrates and the reaction products after incubation for up to 4 h were analyzed by TLC and HPLC as described above. The percentages of degradation of the substrates were determined by comparing the amounts of the decreased substrates measured by HPLC against the initial amounts of the substrates. In the substrate specificity experiments, monosaccharides produced in the reaction mixture were analyzed by using the F-kit (food analysis d-glucose/d-fructose; Roche Diagnostics GmbH).
RESULTS AND DISCUSSION
Purification of LDE.
According to the methods described in Materials and Methods, the LDE was purified about 50-fold with a recovery of 31% from the culture supernatant, and the specific activity of the purified LDE was 105.0 U/mg. LDE gave a single band on SDS-PAGE, as shown in Fig. 1A. The molecular weights of the purified LDE were estimated to be 54,000 by SDS-PAGE and 60,000 by gel filtration with Sephacryl S-200 SF (Fig. 1B). Thus, the enzyme from S. exfoliatus F3-2 was considered to be a monomer. Since the two already reported levanbiose-producing LDEs from Streptomyces sp. strain 7-3 (10) and Arthrobacter sp. strain 51A (8) are monomeric enzymes, the LDE from S. exfoliatus F3-2 was found to be structurally similar to these LDEs.
FIG. 1.
SDS-PAGE and gel chromatography of the purified LDE. (A) Lanes 1, marker proteins; 2, purified LDE. The marker proteins were fructose-6-phosphate kinase (Mr, 85,200), glutamate dehydrogenase (Mr, 55,600), aldolase (Mr, 39,200), and triosephosphate isomerase (Mr, 26,600). (B) Plots of the logarithmic molecular weight versus retention time in gel filtration. The position of the sample enzyme is shown by the solid circle. a, bovine serum albumin (Mr, 68,000); b, egg albumin (Mr, 45,000); c, chymotrypsinogen A (Mr, 25,000); d, cytochrome c (Mr, 12,500).
pH and thermal properties.
The isoelectric point of the purified LDE was estimated to be 4.7. The effects of pH and temperature on both the reaction and stability of the purified LDE were examined, and the results are shown in Fig. 2. The maximal enzyme activities were obtained at pH 5.5 (Fig. 2A) and at 60°C (Fig. 2B). More than 90% of the enzyme activities were found to remain in the pH range 3.5 to 8.0 (Fig. 2A) or at temperatures of up to 50°C (Fig. 2B).
FIG. 2.
pH and thermal properties of the purified LDE. For determination of optimal reaction pH (A [●, ▴, ■]), the standard LDE assay was employed, except that the phosphate buffer was replaced when necessary. The buffers used were 50 mM citric acid-sodium citric acid buffer (●), sodium phosphate buffer (▴), or boric acid-sodium tetraborate buffer (■). For pH stability (A [○, ▵, □]), the purified enzyme was mixed with the same volume of 0.1 M different kinds of buffers as used for the determination of optimal pH. After the incubation at 4°C for 24 h, the enzyme was diluted with 4 volumes of 0.1 M sodium phosphate buffer (pH 5.5) and used as the treated enzyme. The optimal reaction temperature (B [⧫]) was determined under standard LDE assay conditions, but at various temperatures. Temperature stability (B [◊]) was tested by heating the purified enzyme at various temperatures for 20 min. For determination of optimal reaction pH and temperature, the maximum activities obtained under the conditions tested were taken as 100%. For estimation of pH and temperature stabilities, the activities of the untreated enzyme were taken as 100%.
Reaction products from levan.
The reaction products from levan with purified LDE were examined as described in Materials and Methods. As shown in Fig. 3A, levanbiose was detected as a major product from the beginning of the reaction. Levanoligosaccharides having a degree of polymerization more than 3 are transiently detected, with a decrease in levan, especially after reaction for 1 to 2 h. The concentration of levanbiose reached its maximum, 9.3 mg/ml, after reaction for 8 h, and then degraded slowly into fructose, as can be seen in Fig. 3B.
FIG. 3.
TLC and HPLC analyses of the reaction products from levan with purified LDE. The enzyme reaction and the analyses of the sugars in the reaction mixtures by TLC and HPLC were done as described in Materials and Methods. (A) TLC analysis. S, partial HCl hydrolysates of levan. F, fructose; F2, levanbiose; F3, levantriose; F4, levantetraose; L, levan. (B) HPLC analysis of the reaction mixtures shown in panel A. ●, levanbiose; ○, fructose.
Mode of action.
The mode of action was investigated as described in Materials and Methods. Figure 4A shows the results obtained with the reaction mixture containing the purified LDE at 0.04 U/ml, where the HPLC measurement revealed that the degradation reaction yielded 2.3 mg of levanbiose per ml from 10 mg of initial levan per ml after reaction for 24 h. Thus, this reaction was designed to represent the initial event of the degradation reaction. The results indicated that the ratio (amount of reducing sugar/amount of levanbiose) at each reaction time until 24 h appeared to be almost constant and was close to the theoretical value of 1. TLC analysis of these reaction mixtures showed that only levanbiose was detected as the reaction product. The same results were also confirmed by HPLC analysis (data not shown). From these results, the enzyme was judged to degrade levan into levanbiose in an exo-acting manner.
FIG. 4.
Ratio of the amount of reducing sugar to the amount of levanbiose in the reaction mixture for levan degradation with purified LDE. The purified LDE concentrations in the reaction mixtures were 0.04 U/ml for panel A and 0.4 U/ml for panel B. The TLC analysis of the reaction mixture at each time point is shown right under the graph. Abbreviations for TLC are defined in the legend to Fig. 3. ∗, ratio of the amount of reducing sugar as levanbiose determined by the Somogyi-Nelson method versus the amount of levanbiose determined by HPLC.
On the other hand, when the purified LDE was increased up to 0.4 U/ml, the reaction proceeded rapidly, and the concentration of levanbiose reached 8.7 mg/ml after 24 h of incubation. Thus, these reaction conditions were aimed at investigating the event occurring when the reaction had proceeded almost completely. As shown in Fig. 4B, the ratios increased from 1 over time, and especially at time points later than 4 h, the ratios reached up to 1.4 until 24 h of incubation. In agreement with these findings, TLC analysis indicated the presence of fructose and/or glucose and oligosaccharides other than levanbiose, suggesting that the ratio can be used as a measure for the formation of a reducing sugar other than levanbiose.
2,6-β-d-Fructan 6-levanbiohydrolase (EC 3.2.1.64) has been defined as the enzyme that specifically cleaves levan into levanbiose (2). Therefore, the levanbiose-producing LDE from S. exfoliatus F3-2 was clearly judged to be 2,6-β-d-fructan 6-levanbiohydrolase.
Substrate specificity.
The results of the enzyme reaction to various kinds of sugars other than levan as a substrate are summarized in Table 1. All reactions were carried out for 4 h under the conditions described in Materials and Methods. As has been suggested from the results shown in Fig. 3, levanbiose was confirmed to be the smallest substrate for this enzyme, although the hydrolyzing activity was low. This slow rate of levanbiose degradation seems to result in efficient production of levanbiose from levan, as established in our previous study (21). Levantriose was degraded efficiently into levanbiose and fructose. Thus, the smallest substrate for levanbiose formation by the enzyme appeared to be levantriose.
TABLE 1.
Reaction of purified 2,6-β-d-fructan 6-levanbiohydrolase from S. exfoliatus F3-2 to various kinds of sugars
| Sugar | Reactiona (% degradation) | Reaction product(s)b |
|---|---|---|
| Levanbiose | + (3.0) | F |
| Levantriose | + (95.3) | F, F2 |
| Sucrose | + (6.5) | G/F |
| 1-Kestose | + (33.3) | G/F, GF |
| Nystose | + (73.4) | G/F, GF, GF2 |
| 1-Fructosylnystose | + (81.4) | G/F, GF, GF2, GF3 |
| Starch | − | NDc |
| Melibiose | − | ND |
| Melezitose | − | ND |
| Lactose | − | ND |
| Maltose | − | ND |
| Cellobiose | − | ND |
Checked by TLC; +, degraded; −, not degraded. Data in parentheses are the percent degradation compared to the initial amount of substrate as determined by HPLC.
All of the sugars were identified by Rf through TLC analysis. Since glucose (G) and fructose (F) were not separated under these conditions, they are presented as G/F. HPLC analysis was also done to identify sucrose and fructooligosaccharides. F, fructose; F2, levanbiose; F3, levantriose; G/F, glucose/fructose; GF, sucrose; GF2, 1-kestose; GF3, nystose.
ND, not detected.
A particularly interesting feature of this enzyme is the fact that it can hydrolyze β-2,1-fructosyl linkages of fructooligosaccharides, such as 1-kestose, nystose, and 1-fructosylnystose, with substantial reaction rates comparable to that for levantriose, as described in Table 1. Figure 5 shows the results of TLC analyses of these reactions. From Table 1 and Fig. 5, it appears that the percentage of degradation of each substrate increased as the degree of polymerization of the substrate increased. However, practically no degradation reaction was detected when inulin was used as the substrate under the same reaction conditions (data not shown). Monosaccharides produced from these fructooligosaccharides were identified as consisting mainly of fructose by enzymatic determination with the F-kit (data not shown). From these results, it was confirmed that these fructooligosaccharides were degraded into shorter oligosaccharides by liberating fructose from their fructosyl terminal. This enzyme can also hydrolyze β-2,1-linkages of sucrose (Table 1 and Fig. 5), although the reaction seemed to be slow. These data led to the conclusion that this enzyme is capable of hydrolyzing not only β-2,6-fructosyl linkages, but also β-2,1-fructosyl linkages. It was generally indicated that a bacterial levan had many branching points linked by β-2,1-fructosyl linkages at the C-1 position of fructose residues of its main β-2,6-linkage (4). It can be considered that the exo-acting LDE reaction would be slowed down or terminated at these branching points. Therefore, the results presented above indicate that the enzyme from S. exfoliatus F3-2 might have degradation activity for β-2,1-fructosyl linkages. Thus, the enzyme is expected to have the potential for high-yield production of levanbiose from bacterial levan not only from S. levanicum (5), but also from other microorganisms.
FIG. 5.
TLC analysis of the reaction products from fructooligosaccharides with purified LDE. The enzyme reaction was done as described in Materials and Methods. G/F, glucose/fructose; GF, sucrose; GF2, 1-kestose; GF3, nystose; GF4, 1-fructosylnystose.
To date, there have been few reports on levanbiose-producing LDEs (2, 8–10). Table 2 summarizes the properties of the known levanbiose-producing LDEs together with those from S. exfoliatus F3-2. As can be seen from the table, the enzyme from strain F3-2 differs from the other enzymes in terms of its range of stability (higher stability under acidic conditions), mode of action, and effect on β-2,1-fructosyl linkages. Therefore, the enzyme from S. exfoliatus F3-2 was judged to be a novel levanbiose-producing LDE that can be clearly classified as 2,6-β-d-fructan 6-levanbiohydrolase (EC 3.2.1.64). It can also be said that this is the first example of purification of the typical 2,6-β-d-fructan 6-levanbiohydrolase. These results obtained with the enzyme of S. exfoliatus F3-2 would be important for providing enzymological characteristics of LDEs, especially that of 2,6-β-d-fructan 6-levanbiohydrolase. To obtain molecular information about this LDE and to prepare it more effectively for levanbiose production, cloning of the enzyme is in progress.
TABLE 2.
Properties of purified 2,6-β-d-fructan 6-levanbiohydrolase from S. exfoliatus F3-2 and of the other levanbiose-producing LDEs
| Microorganism | Mol wt
|
Optimum condition
|
Stability range
|
pI | Mode of action on levan | Effect on β-2, 1- linkages of fructose | Products from levana | Reference | |||
|---|---|---|---|---|---|---|---|---|---|---|---|
| SDS-PAGE | Gel filtration | pH | Temp (°C) | pH | Temp (°C) | ||||||
| Arthrobacter sp. strain 51A | 60,000 | 64,000 | 5.8 | 65 | 7.0–9.0 | ≦45 | 6.4 | Endo | Yes | F, F2 | 8, 9 |
| Streptomyces sp. strain 7-3 | 54,000 | 57,000 | 6.5 | 40 | 5.5–8.5 | ≦40 | 4.7 | Exo | No | F, F2, F3 | 9, 10 |
| S. exfoliatus F3-2 | 54,000 | 60,000 | 5.5 | 60 | 3.5–8.0 | ≦50 | 4.7 | Exo | Yes | F, F2 | This study |
F, frucose; F2, levanbiose; F3, levantriose.
ACKNOWLEDGMENTS
We are grateful to Meiji Seika Kaisya, Ltd., for kindly supplying fructooligosaccharides.
This work was supported in part by a Grant-in-Aid for Scientific Research (no. 07556087) from the Ministry of Education, Science, Sports and Culture of Japan and also by a Grant-in-Aid for JSPS Fellows (no. 11-5613) to K.S. from the Japan Society for the Promotion of Science.
REFERENCES
- 1.Avigad G, Bauer S. Fructan hydrolases. Methods Enzymol. 1966;8:621–628. [Google Scholar]
- 2.Avigad G, Zelikson R. Cleavage of fructans to levanbiose by a specific hydrolase. Bull Res Counc Isr. 1963;11A4:253–257. [Google Scholar]
- 3.Hestrin S, Avineri-Shapiro S, Aschner M. The enzymic production of levan. Biochem J. 1943;37:450–456. doi: 10.1042/bj0370450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Han Y W. Microbial levan. Adv Appl Microbiol. 1990;35:171–194. doi: 10.1016/s0065-2164(08)70244-2. [DOI] [PubMed] [Google Scholar]
- 5.Kojima I, Saito T, Iizuka M, Minamiura N, Ono S. Characterization of levan produced by Serratia sp. J Ferment Bioeng. 1993;75:9–12. [Google Scholar]
- 6.Laemmli U K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680–685. doi: 10.1038/227680a0. [DOI] [PubMed] [Google Scholar]
- 7.Lowry O H, Rosebrough N J, Farr A L, Randall R J. Protein measurement with the folin phenol reagent. J Biol Chem. 1951;193:265–275. [PubMed] [Google Scholar]
- 8.Murakami H, Fukui H, Nakano H, Kitahata S. Purification and some properties of a levanase from Arthrobacter sp. no. 51A. Oyo Toshitsu Kagaku. 1994;41:173–180. [Google Scholar]
- 9.Murakami H, Kitahata S. Levan-degrading enzymes from microorganisms. Oyo Toshitsu Kagaku. 1997;44:195–202. . (In Japanese.) [Google Scholar]
- 10.Murakami H, Muroi H, Kuramoto T, Tamura Y, Mizutani K, Nakano H, Kitahata S. Purification and some properties of a levanase from Streptomyces sp. no. 7-3. Agric Biol Chem. 1990;54:2247–2255. [Google Scholar]
- 11.Saito K, Goto H, Yokota A, Tomita F. Purification of levan fructotransferase from Arthrobacter nicotinovorans GS-9 and production of DFA IV from levan by the enzyme. Biosci Biotechnol Biochem. 1997;61:1705–1709. doi: 10.1271/bbb.61.1705. [DOI] [PubMed] [Google Scholar]
- 12.Saito K, Hira T, Suzuki T, Hara H, Yokota A, Tomita F. Effects of DFA IV in rats: calcium absorption and metabolism of DFA IV by intestinal microorganism. Biosci Biotechnol Biochem. 1999;63:655–661. doi: 10.1271/bbb.63.655. [DOI] [PubMed] [Google Scholar]
- 13.Saito K, Yokota A, Tomita F. Molecular cloning of levan fructotransferase gene from Arthrobacter nicotinovorans GS-9 and its expression in Escherichia coli. Biosci Biotechnol Biochem. 1997;61:2076–2079. doi: 10.1271/bbb.61.2076. [DOI] [PubMed] [Google Scholar]
- 14.Sakurai H, Yokota A, Sumita Y, Mori Y, Matsui H, Tomita F. Metabolism of DFA III by Arthrobacter sp. H65-7: purification and properties of a DFA III hydrolysis enzyme (DFA IIIase) Biosci Biotechnol Biochem. 1997;61:989–993. [Google Scholar]
- 15.Sakurai H, Yokota A, Tomita F. Molecular cloning of an inulin fructotransferase (depolymerizing) gene from Arthrobacter sp. H65-7 and its expression in Escherichia coli. Biosci Biotechnol Biochem. 1997;61:87–92. doi: 10.1271/bbb.61.87. [DOI] [PubMed] [Google Scholar]
- 16.Somogyi M. Notes on sugar determination. J Biol Chem. 1952;195:19–23. [PubMed] [Google Scholar]
- 17.Suzuki T, Hara H, Kasai T, Tomita F. Effects of difructose anhydride III on calcium absorption in small and large intestines of rats. Biosci Biotechnol Biochem. 1998;62:837–841. doi: 10.1271/bbb.62.837. [DOI] [PubMed] [Google Scholar]
- 18.Tanaka K, Karigane T, Yamaguchi F, Nishikawa S, Yoshida N. Action of levan fructotransferase of Arthrobacter ureafaciens on levanoligosaccharides. J Biochem. 1983;94:1569–1578. [PubMed] [Google Scholar]
- 19.Yokota A, Enomoto K, Tomita F. Purification and properties of an inulin fructotransferase (depolymerizing) from Arthrobacter sp. H65-7. J Ferment Bioeng. 1991;72:262–265. [Google Scholar]
- 20.Yokota A, Hirayama S, Enomoto K, Miura Y, Takao S, Tomita F. Production of inulin fructotransferase (depolymerizing) by Arthrobacter sp. H65-7 and preparation of DFA III from inulin by the enzyme. J Ferment Bioeng. 1991;72:258–261. [Google Scholar]
- 21.Yokota A, Kondo K, Nakagawa M, Kojima I, Tomita F. Production of levanbiose by a levan-degrading enzyme from Streptomyces exfoliatus F3-2. Biosci Biotechnol Biochem. 1993;57:745–749. [Google Scholar]
- 22.Yokota A, Yamauchi O, Tomita F. Production of inulotriose from inulin by inulin-degrading enzyme from Streptomyces rochei E87. Lett Appl Microbiol. 1995;21:330–333. [Google Scholar]