- Email: [email protected]
Microbial production of inulo-oligosaccharides by an endoinulinase from Pseudomonas sp. Expressed in Escherichia coli
JOURNALOF BIOSCIENCE AND BIOENGINEERING Vol. 81, No. 3, 291-295. 1999
Microbial Production of Inulo-Oligosaccharides by an Endoinulinase from Pseudomonas sp. Expressed in Escherichia coli JONG WON YUN,‘* YONG JIN CHOI,Z CHII HYUN SONG,’ AND SEUNG KOO SONG3 Department of Biotechnology, Taegu University, Kyungbuk 712-714,’ Graduate School of Biotechnology, Korea University, Seoul 136-701,2 and Department of Chemical Engineering, Pusan National University, Pusan 609-735,3 Korea Received 4 September 199WAccepted 4 November 1998
In an attempt to utilize the whole cell as a biocatalyst for inulo-oligosaccharide @OS)production from inulin, the endoinulinase gene (inul) of Pseudomorw sp. was cloned into the plasmid pBR322 using EcoRI restriction endonuclease and Escherichia coli HBlOl as the host strain. The endoinuiinase from E. coli HBlOl/pKMG50 was constitutively expressed, producing a high yield of IOS (78%). In a batchwise reaction, the initial enzyme concentration determined the total oligosaccharide yield, and excessenzyme decreasedthe total oligosaccharide yield due to the formation of high amounts of free sugars such as glucose and fructose. The recombinant E. coli expressing endoinulinase activity were immobilized on a polystyrene carrier material, resulting in a dramatically enhanced thermal stability of the enzyme. Continuous production of 10s from inulin was also carried out at 50°C using a bioreactor packed with the immobilized cells. Under the optimal operation conditions, continuous production of IOS was achieved with a productivity of 150 g/Z-h for 17 d at SOW without significant loss of initial activity. [Key words: endoinulinase,
Escherichiacoli, immobilized
Inulin is usually found as a reserve polyfructan in the roots and tubers of various plants such as Jerusalem artichoke, chicory and dahlia. It consists of linear chains of p-2,1 -linked D-fructofuranose terminating in a glucose residue through a sucrose-type linkage at the reducing end (l-3). It has been widely investigated as a source for the production of ultra-high-fructose syrup through enzymatic hydrolysis by either exoinulinase (EC 3.2.1. 26; /3-D-fructofuranosidase) alone or synergistically with endoinulinase (EC 3.2.1.7; ,9-fructan-fructanohydrolase) (4-6). Recently, many investigators have exerted their efforts to find a selective endoinulinase which lacks invertase or exoinulinase activity, to hydrolyze the internal linkages in inulin in order to yield several oligosaccharide units such as inulotriose and inulotetraose (7-11). The resulting inulo-oligosaccharides (IOS), like other oligosaccharides, have been regarded as a type of soluble dietary fiber and have proved to be a efficient at increasing the population of resident bifidobacteria in human intestinal flora (12-17). Although most enzymes are best utilized in immobilized forms, immobilization of inulinase activity is less documented in literature than that of other enzyme systems. Several gel matrices including calcium alginate, agar and gelatin together with support materials such as cellulose derivatives and chitin have been examined as possible materials for inulinase immobilization (18-22). However, most of those immobilization systems have been used not to produce 10s but high fructose syrup (23, 24).
We have previously reported several enzymatic systems for IOS production from inulin using extracellular endoinulinase derived from a new Pseudomonas sp. isolate in a soluble or immobilized form (25-30). Most microbial endoinulinases are extracellular products, and therefore not amenable to immobilization as a whole cell form. Bearing this in mind, we have * Corresponding
author.
cells, inulo-oligosaccharides]
attempted to clone the endoinulinase gene in Pseudomonas sp. into E. coli HBlOl/pKMGSO in an attempt to confine the endoinulinase activity in the cells in order to develop an alternative process for 10s production. In the present study, both batchwise and continuous production of 10s using recombinant E. coli either intact cells or cells immobilized on a carrier material are described. MATERL4LS Materials
AND METHODS
Pure nonhydrolyzed inulin from dahlia tubers (Sigma Chemical, St. Louis, MO, USA) was used as a substrate for both cell cultures and enzyme reactions, and other chemicals used were of analytical grade. The polystyrene carrier material [poly(aminomethyl styrene), Wofatit UF93] was supplied by Chemie AG, Bitterfeld-Wolfen, Germany. The beads used of the following specifications: particle diameter 0.4-0.8 mm, average pore diameter 7 x lop6 cm, pore volume 0.8 cm3/g, content of amino groups 4 mmol/g. Cell preparation The endoinulinase gene (inul) of Pseudomonas sp. was cloned into the plasmid pBR322 after digestion with EcoRI and was transferred into the host strain E. coli HBlOl (F-, hsdS20, recA13, ara14, proA2, lacYI, gafK2, rpsL20, xy15, mtll, supE44), as previously reported (3 1). The resultant plasmid was designated as pKMG 50. The E. coli cells harboring the plasmid were cultivated at 37°C for 48 h in LB medium supplemented with 1% (w/v) inulin. The cells were harvested by centrifugation (10,OOOxg) and the resultant cells were resuspended in deionized water at an appropriate concentration and used directly in enzyme reactions or to prepare the immobilized cells. No further treatment was carried out to enhance the permeability of the cells. For a comparative study with the recombinant cells, Pseudomonas sp. was also cultivated at 45°C for 48 h in a 250-ml flask containing 50ml medium composed of 1% (w/v) inulin, 0.8% (NH&HP04, 1.5% corn steep 291
292
J. BIOSCI. BIOENG.,
YUN ET AL.
liquor, 0.05% KCl, 0.05% MgS04.7Hz0, and 0.003% FeS04. 7Hz0. After separation of the cells by centrifugation (10,000 x g), the supernatant containing endoinulinase was concentrated by dialysis and membrane filtration (M.W. cut off 30,OOODa), and the resultant crude enzyme solution was used throughout the experiments without further purification. Batch enzyme reaction Batch enzyme reactions were carried out in 15ml test tubes in a rotary shaker at 55°C and 150 rpm. Unless otherwise specified, 30 units/g inulin of intact cells or 50 units/g inulin of immobilized cells were added to 5 ml of 100 g/l inulin solution. Samples were taken at regular intervals and the reactions were terminated by heating the reaction mixtures at 100°C for 10 min in a boiling water bath. Cell immobilization The support material was activated by circulating 2.5% (v/v) glutaraldehyde using three times the bed volume of the column (1.2 x 30 cm, bed volume cu. 30ml). To ensure complete removal of glutaraldehyde, the carrier material was washed with a total amount of 500ml of water in upward as well as downward flow. For immobilization, the cell suspension was then applied to the column at a constant flow rate of 30ml/h. After immobilization for 10 h at ambient temperature, the column was washed again using 500ml of deionized water. Continuous immobilization is essentially superior to batch adsorption in terms of immobilization yield because it overcomes adsorption equilibria. To prevent channeling problems Reactor operation during column operation (due to density difference between water and inulin solution), 50ml of inulin solutions with concentrations ranging from 5 to 5Og/l were run consecutively. To evaluate reactor performance, the column was continuously operated at various flow rates using lOOg/l of inulin as substrate at 50°C. Operation of the column was facilitated by the upward flow of substrate such that self-compression of the immobilized enzyme that causes clogging was minimized. The inulooligosaccharide concentration in the effluent was analyzed after a 24-h operation, when a steady state was established. Long-term stability of the immobilized enzyme was evaluated using 100 g/l inulin at a flow rate of 45 ml/h (superficial space velocity, SV 1.5 h-l) at 5O’C. The residence time (l/SV) was calculated on the basis of the void volume of the reactor (void fraction of the reactor was 0.43). Inulin solutions were used without sterilization throughout the experiments for practical applica-
tions of the system. Enzyme activity Endoinulinase activity was assayed by mixing 2 ml of cell suspension with 8 ml of 2% (w/v) inulin in 0.05 M phosphate buffer (pH 7.5) and incubating at 55°C for 60 min. In the case of immobilized cells, the cell suspension was replaced by 1 g of immobilized cells (wet weight after filtration). One enzyme unit was defined as the amount of enzyme required for the liberation of 1 pmol of inulobiose (the first product liberated) per min under the above conditions. Carbohydrates were analyzed Analytical methods by HPLC (Shimadzu Co., Kyoto) using the Aminex HPX-42C column (0.78 x 30 cm, Bio-Rad) and a refractive index detector. The column temperature was kept constant at 85°C and water was used as the mobile phase at a flow rate of 0.6ml/min at 80°C. The total inulo-oligosaccharide production was estimated as the sum of inulobiose (FZ) and other oligofructosides (F, and some GF,, n<7, where F and G are fructose and glucose units, respectively) which have a degree of polymerization (DP) ranging from DP3 to DP7. To identify the degree of polymerization of the products, thin layer chromatographic analysis was carried out on precoated silica gel 60 plates (Merck AG, Darmstadt, Germany). The plates were developed with a solvent system of isopropanol/ethyl acetate/water (6/2/2 by volume). The sugar spots were visualized by spraying the plates with sulfuric acid solution containing phenol and heating at 120°C for 5 min. RESULTS AND DISCUSSION Batch kinetics of recombinant E. coli Recombinant E. coli was cultivated at 37°C for 2 d as described above. As shown in Fig. 1, the cell growth and enzyme production reached a maximum at 19 h, when the endoinulinase activity was 1.5 units/ml, which is slightly lower than that in a culture broth of Pseudomonas sp. (1.8 units/ml). The initially supplemented inulin was rapidly utilized within 8 h, forming mainly inulobiose via inulolytic activity of the endoinulinase (Fig. 2). Other higher oligofructosides such as inulotriose and inulotetraose were also accumulated, although their levels
looti
u”
20
0 0
0
5
10
Cultivation
15
20
25
30
time (h)
FIG. 1. Kinetics of endoinulinase production in the recombinant E. coli.
4
a Cultivation
12
16
20
time (h)
FIG. 2. Changes in carbohydrate composition during the growth of the recombinant E. coli. Symbols: 0, inulin; 0, inulobiose; n , DP3; A, DP4 oligofructosides; A, inulo-oligosaccharides higher than DP4; V’, glucose and fructose.
MICROBIAL
PRODUCTION
OF INULO-OLIGOSACCHARIDES
293
20
I
I
,
I
50
55
60
65
0 0
5
10
15
Reaction
20
25
30
45
35
Temperature
time (h)
FIG. 3. Effect of enzyme concentration on the total inulooligosaccharide yield from inulin by intact cells of recombinant E. coli expressing endoinulinase activity. Symbols: 0, 30; 0, 60; A, 90 units/g substrate.
were relatively low. It is interesting to note that the large amount of inulobiose accumulated could not be used as a carbon source even though most of the inulin was consumed after 8 h. Free fructose and glucose were released during the early culture period (from 4 to 8 h), thereafter they were utilized as carbon sources. Effect of enzyme concentration Enzyme reactions were carried out for 35 h using three enzyme concentrations (30, 60, and 90 units/g substrate) at 100 g/l inulin, which was found to be the most suitable concentration within the solubility limit at the reaction temperature (55°C). The time courses of total 10s production using intact cells under each enzyme concentration are shown in Fig. 3. The maximum total yield of 10s was around 78% irrespective of the initial enzyme concentration. Higher oligosaccharides formed during the early reaction period were further degraded to free sugars by successive enzyme actions as the reaction time increased. Conse-
70
CC)
FIG. 5. Thermal stability of the immobilized cells. Symbols: 0, intact cells; 0, immobilized cells.
quently, high amounts of free glucose and fructose were released, leading to significant decrease in the oligosaccharide content. The typical reaction profiles for each component at 30units/g substrate are illustrated in Fig. 4. Maximal IOS yield was achieved at 17.5 h, thereafter most oligosaccharides were degraded to free glucose and fructose. Thus, it is strongly recommended to terminate the reaction at a suitable time, that is, just before the oligosaccharides begin to be hydrolyzed. Stability of the immobilized cells To evaluate the thermal stability of the immobilized cells compared to that of intact cells, immobilized cells and intact cells were separately incubated at the various temperatures between 45°C to 65°C for 3 h, and then their residual activities were determined (Fig. 5). The intact cells were very unstable at temperatures over 65”C, whereas the immobilized cells exhibited enhanced stability at all the temperatures examined. Continuous production of inulo-oligosaccharides
The operational stability of the immobilized cell reactor is well correlated with the spent volume of the substrate. Thus, the stabilitv of the immobilized cell reactor should be evaluated under actual operating conditions. No steps were taken to maintain the pH at the optimum (7.5),
80
‘20x
60
Reaction
time (h)
FIG. 4. Typical reaction profile of batchwise production of inulo-oligosaccharides from inulin under optimal conditions by intact cells of recombinant E coli containing endoinulinase activity. Symbols: 0, inulin; 0, inulobiose; W, DP3; A, DP4 oligofructosides; A, inulo-oligosaccharides higher than DP4; V, glucose and fructose; 0, total inulo-oligosaccharides.
0: 0
5
IO
Operation
15
20
25
time (d)
FIG. 6. Operational stability of the immobilized cell reactor containing endoinulinase activity for continuous production of inulooligosaccharides from inulin under optimal operational conditions (100 g/l inulin, flow rate SV 1.5 h-l).
294
YUN ET AL.
J. BIOSCI. BIOENG.,
TABLE 1. Comparison of the carbohydrate compositions of inulooligosaccharides produced by the cells of recombinant E. coli with those by native enzyme from Pseudomonas SP.~ Composition (%, w/w) Recombinant E. coli Pseudomonas Intact cells Immobilized sp. cells 3.2 1.2 2.5 9.1 10.3 8.2 7.1 9.7 9.7 1.8 1.6 2.0
Carbohydrate Inulin Glucose Fructose Sucrose Inulo-oligosaccharidesb DP2 DP3 DP4 DP>4 Total inulo-oligosaccharides
36.2 22.8 9.5 10.3
44.8 17.9 8.2 6.3
29.1 21.4 9.4 17.1
78.8
77.2
17.6
a Product compositions are obtained at the reaction time when oligosaccharide production was maximum. Initial inulin concentration was 100 g/l. Enzyme concentration was 30 (intact cells and native enzyme from Pseudomonas sp.) or 50unitsIg inulin (immobilized cells). b DP means degree of polymerization.
such as the inclusion of buffers, as these would had a detrimental effect on product purity. Furthermore, efficient addition of alkalis to packed-bed reactors is very difficult. In practice, the reactor is maintained at a flow rate that allows virtually complete conversion of the inulin. Figure 6 presents the long-term stability of the immobilized cell reactor under the pre-determined optimal operational conditions (substrate concentration, 100 g/i inulin; flow rate, SV 1.5 h-r). No significant change in steady state conversion was observed during the first 17 d, thereafter, the enzyme activity started to decrease rapidly. The volumetric productivity achieved was 150 g/l. h, which is a greatly improved productivity compared to that in the immobilized endoinulinase reactor from Pseudomonas sp. (55 g/l.h, see Ref. 29). Comparison of product composition Table 1 shows the composition of the reaction products catalyzed by different forms of endoinulinase using lOOg/l inulin; me., intact cells of recombinant E. coli and immobilized cells, and native enzyme from Pseudomonas sp. There was a significant difference in composition of products catalyzed by different forms of endoinulinase. That is, higher levels of inulobiose (the smallest molecule in the product) were observed when intact cells or immobilized cells were employed. It appears that the difference in microenvironments within the cells or immobilized cells is probably responsible for the appearance of high levels of inulobiose. Furthermore, there was essentially no difference in the carbohydrate compositions between the reaction products produced by continuous and batch processing using immobilized cells (data not shown, see Table 1). ACKNOWLEDGMENTS This work was supported by a research grant from the Korea Science and Engineering Foundation (KOSEF 96-0502-02-01-3). We wish to express our sincere thanks to Dr. J. Mansfeld and A. Schellenger of Martin-Luther University, Germany, for their kind gift of the carrier material.
REFERENCES 1. Bacon, J. S.D. and Edelman, J.: The carbohydrates of the Jerusalem artichoke and other Compositae. Biochem. J., 48, 114-126 (1951). 2. Vandamme, E. J. and Derycke, D. G.: Microbial inulinases: fermentation process, properties and application. Adv. Appl. Microbial., 29, 139-176 (1983). 3. Faroworth, E. R.: Fructans in human and animal diets, p. 257-272. In Suzuki, M. and Chatterton, N. J. (ed.), Science and technology of fructans. CRC Press, Boca Raton, FL, USA (1993). 4. Zittan, L.: Enzymatic hydrolysis of inulin-an alternative way to fructose production. Starch, 33, 373-377 (1981). 5. Kierstan, M.: Production of fructose syrup from inulin. Process Biochem., May, 2-4 (1980). 6. Byun, S.M. and Nahm, B.H.: Production of fructose from Jerusalem artichoke by enzymatic hydrolysis. J. Food Sci., 43, 1871-1873 (1978). 7. Nakamura, T., Nagamoto, Y., Hamada, S., Nishino, Y ., and Ohta, K.: Occurrence of two forms of extracellular endoinulinase from Aspergillus niger mutant 817. J. Ferment. Bioeng., 78, 134-139 (1994). 8. Nakamura, T., Sbitara, A., Matsuda, S., Matsno, T., Suiko, M., and Ohta, K.: Production, purification and properties of an endoinulinase of PeniciNium sp. TN-88 that liberates inulotriose. J. Ferment. Bioeng., 84, 313-318 (1977). 9. Nakamura, T. and Nakatsu, S.: Action and production of inulinase. J. Jpn. Sot. Starch Sci., 35, 121-130 (1988). (in Japanese) 10. Onodera, S. and Shiomi, N.: Purification and substrate specificity of endo-type inulinase from Penicillium purpurogenum. Agric. Biol. Chem., 52, 2569-2576 (1988). 11. Yokota, A., Yamauchi, O., and Tomita, F.: Production of inulotriose from inulin by inulindegrading enzyme from Streptomyces rochei E87. Lett. Appl. Microbial., 21, 330-333 (1995). 12. Roberfroid, M.: Dietary fiber, inulin, and oligofructose: a review comparing their physiological effects. Crit. Rev. Food Sci. Nutr., 33, 103-148 (1993). 13. Yun, J. W.: Fructooligosaccharides-occurrence, preparation, and anulication. Enzvme Microb. Technol.. 19. 107-117 (1996): _ 14. Kuriki, T., Tsuda, M., and Imanaka, T.: Highly branched oligosaccharides production by the transglucosylation reaction of neonullulanase. J. Ferment. Bioena.. 73. 198-202 (1992). 15. Hidaka, H., Eida, T., and Saitoh, YL industrial production of fructo-oligosaccharides and its application for human and animals. Nippon Nogeikagaku Kaishi, 61, 915-923 (1987). (in Japanese) 16. Tomomatsu, H.: Health effects of oligosaccharides. Food Technol., October, 61-65 (1994). 17. Wada, K., Watanabe, J., Mizutani, J., Tomoda, M., Suzuki, H., and Saitoh, Y.: Effect of soybean oligosaccharides in a beverage on human fecal flora and metabolites. Nippon Nogeikagaku Kaishi, 66, 127-135 (1992). (in Japanese) 18. Bajpai, P. and Margartis, A.: Immobilization of Kluyveromyces marxianus cells with inulinase activity in agar gel. J. Gen. Appl. Microbial., 31, 297-304 (1985). 19. Bajpai, P. and Margartis, A.: Production of high fructose syrups from Jerusalem artichoke tubers using Kluyveromyces marxianus immobilized in agar gel. J. Gen. Appl. Microbial., 31, 305-311 (1985). 20. Bajpai, P. and Margartis, A.: Immobilization of Kluyveromyces marxianus cells containing inulinase activity in open pore gelatin matrix: II. Application for high fructose syrup production. Enzyme Microb. Technol., 7, 459-461 (1985). 21. Guiraud, J.P., Demeulle, S., and Galzy, P.: Inulin hydrolysis by the Debaryomyces phaJ7ii inulinase immobilized in DEAEcellulose. Biotechnol. Lett., 3, 683-688 (1981). 22. Guiraud, J. P., Bajon, A. M., Chautard, P., and Galzy, P.: Inulin hydrolysis by an immobilized yeast cell reactor. Enzyme Microb. Technol., 5, 185-190 (1985).
VOL. 87, 1999
MICROBIAL
23. Norman, B. E. and Hojer-Perderson, B.: The production of fructooligosaccharides from inulin or sucrose using inulinase or fructosyltransferase from Aspergillus ficuum. Denpun Kagaku, 36, 103-111 (1989). 24. Onodera, 0. and Shiomi, N.: lmmobilization of endo-type inulinase with K-carrageenan and characterization of the immobilized enzyme. .I. Rakuno Gakuen Uni., 14, 17-21 (1989). (in Japanese) 25. Kim, D. H., Choi, Y. J., Song, S. K., and Yun, J. W.: Production of inulo-oligosaccharides using endo-inulinase from a Pseudomonas sp. Biotechnol. Lett., 19, 369-371 (1997). 26. Yun, J. W., Kim, D. H., Kim, B. W., and Song, S. K.: Comparison of sugar compositions between inulo- and fructooligosaccharides produced by different enzyme forms. Biotechnol. Lett., 19, 553-556 (1997). 27. Yun, J. W., Kim, D. H., Uhm, T. B., and Song, S. K: Produc-
28.
29.
30. 31.
PRODUCTION
OF INULO-OLIGOSACCHARIDES
295
tion of high-content inulo-oligosaccharides from inulin by a purified endoinulinase. Biotechnol. Lett., 19, 935-938 (1997). Yun, J. W., Kim, D. H., Yoon, H. B., and Song, S. K.: Effect of inulin concentration on the production of inulo-oligosaccharides by soluble and immobilized endoinulinase. J. Ferment. Bioeng., 84, 365-368 (1997). Yun, J. W., Kim, D. H., Kim, B. W., and Song, S. K.: Production of inulo-oligosaccharides from inulin by immobilized endoinulinase from Pseudomonas sp. J. Ferment. Bioeng., 84, 369-371 (1997). Park, J. P., Kim, D. H., Kim, D. S., and Yun, J. W.: Enzymatic production of inulo-oligosaccharides from chicory juice. Biotechnol. Lett., 20, 385-388 (1998). Eom, S. J., Kwon, Y. M., and Choi, Y. J.: Molecular cloning of Pseudomonas sp. inulinase gene and its expression in E. coli. Kor. .I. Appl. Microbial. Biotechnol., 16, 484-488 (1988).
Microbial Production of Inulo-Oligosaccharides by an Endoinulinase from Pseudomonas sp. Expressed in Escherichia coli JONG WON YUN,‘* YONG JIN CHOI,Z CHII HYUN SONG,’ AND SEUNG KOO SONG3 Department of Biotechnology, Taegu University, Kyungbuk 712-714,’ Graduate School of Biotechnology, Korea University, Seoul 136-701,2 and Department of Chemical Engineering, Pusan National University, Pusan 609-735,3 Korea Received 4 September 199WAccepted 4 November 1998
In an attempt to utilize the whole cell as a biocatalyst for inulo-oligosaccharide @OS)production from inulin, the endoinulinase gene (inul) of Pseudomorw sp. was cloned into the plasmid pBR322 using EcoRI restriction endonuclease and Escherichia coli HBlOl as the host strain. The endoinuiinase from E. coli HBlOl/pKMG50 was constitutively expressed, producing a high yield of IOS (78%). In a batchwise reaction, the initial enzyme concentration determined the total oligosaccharide yield, and excessenzyme decreasedthe total oligosaccharide yield due to the formation of high amounts of free sugars such as glucose and fructose. The recombinant E. coli expressing endoinulinase activity were immobilized on a polystyrene carrier material, resulting in a dramatically enhanced thermal stability of the enzyme. Continuous production of 10s from inulin was also carried out at 50°C using a bioreactor packed with the immobilized cells. Under the optimal operation conditions, continuous production of IOS was achieved with a productivity of 150 g/Z-h for 17 d at SOW without significant loss of initial activity. [Key words: endoinulinase,
Escherichiacoli, immobilized
Inulin is usually found as a reserve polyfructan in the roots and tubers of various plants such as Jerusalem artichoke, chicory and dahlia. It consists of linear chains of p-2,1 -linked D-fructofuranose terminating in a glucose residue through a sucrose-type linkage at the reducing end (l-3). It has been widely investigated as a source for the production of ultra-high-fructose syrup through enzymatic hydrolysis by either exoinulinase (EC 3.2.1. 26; /3-D-fructofuranosidase) alone or synergistically with endoinulinase (EC 3.2.1.7; ,9-fructan-fructanohydrolase) (4-6). Recently, many investigators have exerted their efforts to find a selective endoinulinase which lacks invertase or exoinulinase activity, to hydrolyze the internal linkages in inulin in order to yield several oligosaccharide units such as inulotriose and inulotetraose (7-11). The resulting inulo-oligosaccharides (IOS), like other oligosaccharides, have been regarded as a type of soluble dietary fiber and have proved to be a efficient at increasing the population of resident bifidobacteria in human intestinal flora (12-17). Although most enzymes are best utilized in immobilized forms, immobilization of inulinase activity is less documented in literature than that of other enzyme systems. Several gel matrices including calcium alginate, agar and gelatin together with support materials such as cellulose derivatives and chitin have been examined as possible materials for inulinase immobilization (18-22). However, most of those immobilization systems have been used not to produce 10s but high fructose syrup (23, 24).
We have previously reported several enzymatic systems for IOS production from inulin using extracellular endoinulinase derived from a new Pseudomonas sp. isolate in a soluble or immobilized form (25-30). Most microbial endoinulinases are extracellular products, and therefore not amenable to immobilization as a whole cell form. Bearing this in mind, we have * Corresponding
author.
cells, inulo-oligosaccharides]
attempted to clone the endoinulinase gene in Pseudomonas sp. into E. coli HBlOl/pKMGSO in an attempt to confine the endoinulinase activity in the cells in order to develop an alternative process for 10s production. In the present study, both batchwise and continuous production of 10s using recombinant E. coli either intact cells or cells immobilized on a carrier material are described. MATERL4LS Materials
AND METHODS
Pure nonhydrolyzed inulin from dahlia tubers (Sigma Chemical, St. Louis, MO, USA) was used as a substrate for both cell cultures and enzyme reactions, and other chemicals used were of analytical grade. The polystyrene carrier material [poly(aminomethyl styrene), Wofatit UF93] was supplied by Chemie AG, Bitterfeld-Wolfen, Germany. The beads used of the following specifications: particle diameter 0.4-0.8 mm, average pore diameter 7 x lop6 cm, pore volume 0.8 cm3/g, content of amino groups 4 mmol/g. Cell preparation The endoinulinase gene (inul) of Pseudomonas sp. was cloned into the plasmid pBR322 after digestion with EcoRI and was transferred into the host strain E. coli HBlOl (F-, hsdS20, recA13, ara14, proA2, lacYI, gafK2, rpsL20, xy15, mtll, supE44), as previously reported (3 1). The resultant plasmid was designated as pKMG 50. The E. coli cells harboring the plasmid were cultivated at 37°C for 48 h in LB medium supplemented with 1% (w/v) inulin. The cells were harvested by centrifugation (10,OOOxg) and the resultant cells were resuspended in deionized water at an appropriate concentration and used directly in enzyme reactions or to prepare the immobilized cells. No further treatment was carried out to enhance the permeability of the cells. For a comparative study with the recombinant cells, Pseudomonas sp. was also cultivated at 45°C for 48 h in a 250-ml flask containing 50ml medium composed of 1% (w/v) inulin, 0.8% (NH&HP04, 1.5% corn steep 291
292
J. BIOSCI. BIOENG.,
YUN ET AL.
liquor, 0.05% KCl, 0.05% MgS04.7Hz0, and 0.003% FeS04. 7Hz0. After separation of the cells by centrifugation (10,000 x g), the supernatant containing endoinulinase was concentrated by dialysis and membrane filtration (M.W. cut off 30,OOODa), and the resultant crude enzyme solution was used throughout the experiments without further purification. Batch enzyme reaction Batch enzyme reactions were carried out in 15ml test tubes in a rotary shaker at 55°C and 150 rpm. Unless otherwise specified, 30 units/g inulin of intact cells or 50 units/g inulin of immobilized cells were added to 5 ml of 100 g/l inulin solution. Samples were taken at regular intervals and the reactions were terminated by heating the reaction mixtures at 100°C for 10 min in a boiling water bath. Cell immobilization The support material was activated by circulating 2.5% (v/v) glutaraldehyde using three times the bed volume of the column (1.2 x 30 cm, bed volume cu. 30ml). To ensure complete removal of glutaraldehyde, the carrier material was washed with a total amount of 500ml of water in upward as well as downward flow. For immobilization, the cell suspension was then applied to the column at a constant flow rate of 30ml/h. After immobilization for 10 h at ambient temperature, the column was washed again using 500ml of deionized water. Continuous immobilization is essentially superior to batch adsorption in terms of immobilization yield because it overcomes adsorption equilibria. To prevent channeling problems Reactor operation during column operation (due to density difference between water and inulin solution), 50ml of inulin solutions with concentrations ranging from 5 to 5Og/l were run consecutively. To evaluate reactor performance, the column was continuously operated at various flow rates using lOOg/l of inulin as substrate at 50°C. Operation of the column was facilitated by the upward flow of substrate such that self-compression of the immobilized enzyme that causes clogging was minimized. The inulooligosaccharide concentration in the effluent was analyzed after a 24-h operation, when a steady state was established. Long-term stability of the immobilized enzyme was evaluated using 100 g/l inulin at a flow rate of 45 ml/h (superficial space velocity, SV 1.5 h-l) at 5O’C. The residence time (l/SV) was calculated on the basis of the void volume of the reactor (void fraction of the reactor was 0.43). Inulin solutions were used without sterilization throughout the experiments for practical applica-
tions of the system. Enzyme activity Endoinulinase activity was assayed by mixing 2 ml of cell suspension with 8 ml of 2% (w/v) inulin in 0.05 M phosphate buffer (pH 7.5) and incubating at 55°C for 60 min. In the case of immobilized cells, the cell suspension was replaced by 1 g of immobilized cells (wet weight after filtration). One enzyme unit was defined as the amount of enzyme required for the liberation of 1 pmol of inulobiose (the first product liberated) per min under the above conditions. Carbohydrates were analyzed Analytical methods by HPLC (Shimadzu Co., Kyoto) using the Aminex HPX-42C column (0.78 x 30 cm, Bio-Rad) and a refractive index detector. The column temperature was kept constant at 85°C and water was used as the mobile phase at a flow rate of 0.6ml/min at 80°C. The total inulo-oligosaccharide production was estimated as the sum of inulobiose (FZ) and other oligofructosides (F, and some GF,, n<7, where F and G are fructose and glucose units, respectively) which have a degree of polymerization (DP) ranging from DP3 to DP7. To identify the degree of polymerization of the products, thin layer chromatographic analysis was carried out on precoated silica gel 60 plates (Merck AG, Darmstadt, Germany). The plates were developed with a solvent system of isopropanol/ethyl acetate/water (6/2/2 by volume). The sugar spots were visualized by spraying the plates with sulfuric acid solution containing phenol and heating at 120°C for 5 min. RESULTS AND DISCUSSION Batch kinetics of recombinant E. coli Recombinant E. coli was cultivated at 37°C for 2 d as described above. As shown in Fig. 1, the cell growth and enzyme production reached a maximum at 19 h, when the endoinulinase activity was 1.5 units/ml, which is slightly lower than that in a culture broth of Pseudomonas sp. (1.8 units/ml). The initially supplemented inulin was rapidly utilized within 8 h, forming mainly inulobiose via inulolytic activity of the endoinulinase (Fig. 2). Other higher oligofructosides such as inulotriose and inulotetraose were also accumulated, although their levels
looti
u”
20
0 0
0
5
10
Cultivation
15
20
25
30
time (h)
FIG. 1. Kinetics of endoinulinase production in the recombinant E. coli.
4
a Cultivation
12
16
20
time (h)
FIG. 2. Changes in carbohydrate composition during the growth of the recombinant E. coli. Symbols: 0, inulin; 0, inulobiose; n , DP3; A, DP4 oligofructosides; A, inulo-oligosaccharides higher than DP4; V’, glucose and fructose.
MICROBIAL
PRODUCTION
OF INULO-OLIGOSACCHARIDES
293
20
I
I
,
I
50
55
60
65
0 0
5
10
15
Reaction
20
25
30
45
35
Temperature
time (h)
FIG. 3. Effect of enzyme concentration on the total inulooligosaccharide yield from inulin by intact cells of recombinant E. coli expressing endoinulinase activity. Symbols: 0, 30; 0, 60; A, 90 units/g substrate.
were relatively low. It is interesting to note that the large amount of inulobiose accumulated could not be used as a carbon source even though most of the inulin was consumed after 8 h. Free fructose and glucose were released during the early culture period (from 4 to 8 h), thereafter they were utilized as carbon sources. Effect of enzyme concentration Enzyme reactions were carried out for 35 h using three enzyme concentrations (30, 60, and 90 units/g substrate) at 100 g/l inulin, which was found to be the most suitable concentration within the solubility limit at the reaction temperature (55°C). The time courses of total 10s production using intact cells under each enzyme concentration are shown in Fig. 3. The maximum total yield of 10s was around 78% irrespective of the initial enzyme concentration. Higher oligosaccharides formed during the early reaction period were further degraded to free sugars by successive enzyme actions as the reaction time increased. Conse-
70
CC)
FIG. 5. Thermal stability of the immobilized cells. Symbols: 0, intact cells; 0, immobilized cells.
quently, high amounts of free glucose and fructose were released, leading to significant decrease in the oligosaccharide content. The typical reaction profiles for each component at 30units/g substrate are illustrated in Fig. 4. Maximal IOS yield was achieved at 17.5 h, thereafter most oligosaccharides were degraded to free glucose and fructose. Thus, it is strongly recommended to terminate the reaction at a suitable time, that is, just before the oligosaccharides begin to be hydrolyzed. Stability of the immobilized cells To evaluate the thermal stability of the immobilized cells compared to that of intact cells, immobilized cells and intact cells were separately incubated at the various temperatures between 45°C to 65°C for 3 h, and then their residual activities were determined (Fig. 5). The intact cells were very unstable at temperatures over 65”C, whereas the immobilized cells exhibited enhanced stability at all the temperatures examined. Continuous production of inulo-oligosaccharides
The operational stability of the immobilized cell reactor is well correlated with the spent volume of the substrate. Thus, the stabilitv of the immobilized cell reactor should be evaluated under actual operating conditions. No steps were taken to maintain the pH at the optimum (7.5),
80
‘20x
60
Reaction
time (h)
FIG. 4. Typical reaction profile of batchwise production of inulo-oligosaccharides from inulin under optimal conditions by intact cells of recombinant E coli containing endoinulinase activity. Symbols: 0, inulin; 0, inulobiose; W, DP3; A, DP4 oligofructosides; A, inulo-oligosaccharides higher than DP4; V, glucose and fructose; 0, total inulo-oligosaccharides.
0: 0
5
IO
Operation
15
20
25
time (d)
FIG. 6. Operational stability of the immobilized cell reactor containing endoinulinase activity for continuous production of inulooligosaccharides from inulin under optimal operational conditions (100 g/l inulin, flow rate SV 1.5 h-l).
294
YUN ET AL.
J. BIOSCI. BIOENG.,
TABLE 1. Comparison of the carbohydrate compositions of inulooligosaccharides produced by the cells of recombinant E. coli with those by native enzyme from Pseudomonas SP.~ Composition (%, w/w) Recombinant E. coli Pseudomonas Intact cells Immobilized sp. cells 3.2 1.2 2.5 9.1 10.3 8.2 7.1 9.7 9.7 1.8 1.6 2.0
Carbohydrate Inulin Glucose Fructose Sucrose Inulo-oligosaccharidesb DP2 DP3 DP4 DP>4 Total inulo-oligosaccharides
36.2 22.8 9.5 10.3
44.8 17.9 8.2 6.3
29.1 21.4 9.4 17.1
78.8
77.2
17.6
a Product compositions are obtained at the reaction time when oligosaccharide production was maximum. Initial inulin concentration was 100 g/l. Enzyme concentration was 30 (intact cells and native enzyme from Pseudomonas sp.) or 50unitsIg inulin (immobilized cells). b DP means degree of polymerization.
such as the inclusion of buffers, as these would had a detrimental effect on product purity. Furthermore, efficient addition of alkalis to packed-bed reactors is very difficult. In practice, the reactor is maintained at a flow rate that allows virtually complete conversion of the inulin. Figure 6 presents the long-term stability of the immobilized cell reactor under the pre-determined optimal operational conditions (substrate concentration, 100 g/i inulin; flow rate, SV 1.5 h-r). No significant change in steady state conversion was observed during the first 17 d, thereafter, the enzyme activity started to decrease rapidly. The volumetric productivity achieved was 150 g/l. h, which is a greatly improved productivity compared to that in the immobilized endoinulinase reactor from Pseudomonas sp. (55 g/l.h, see Ref. 29). Comparison of product composition Table 1 shows the composition of the reaction products catalyzed by different forms of endoinulinase using lOOg/l inulin; me., intact cells of recombinant E. coli and immobilized cells, and native enzyme from Pseudomonas sp. There was a significant difference in composition of products catalyzed by different forms of endoinulinase. That is, higher levels of inulobiose (the smallest molecule in the product) were observed when intact cells or immobilized cells were employed. It appears that the difference in microenvironments within the cells or immobilized cells is probably responsible for the appearance of high levels of inulobiose. Furthermore, there was essentially no difference in the carbohydrate compositions between the reaction products produced by continuous and batch processing using immobilized cells (data not shown, see Table 1). ACKNOWLEDGMENTS This work was supported by a research grant from the Korea Science and Engineering Foundation (KOSEF 96-0502-02-01-3). We wish to express our sincere thanks to Dr. J. Mansfeld and A. Schellenger of Martin-Luther University, Germany, for their kind gift of the carrier material.
REFERENCES 1. Bacon, J. S.D. and Edelman, J.: The carbohydrates of the Jerusalem artichoke and other Compositae. Biochem. J., 48, 114-126 (1951). 2. Vandamme, E. J. and Derycke, D. G.: Microbial inulinases: fermentation process, properties and application. Adv. Appl. Microbial., 29, 139-176 (1983). 3. Faroworth, E. R.: Fructans in human and animal diets, p. 257-272. In Suzuki, M. and Chatterton, N. J. (ed.), Science and technology of fructans. CRC Press, Boca Raton, FL, USA (1993). 4. Zittan, L.: Enzymatic hydrolysis of inulin-an alternative way to fructose production. Starch, 33, 373-377 (1981). 5. Kierstan, M.: Production of fructose syrup from inulin. Process Biochem., May, 2-4 (1980). 6. Byun, S.M. and Nahm, B.H.: Production of fructose from Jerusalem artichoke by enzymatic hydrolysis. J. Food Sci., 43, 1871-1873 (1978). 7. Nakamura, T., Nagamoto, Y., Hamada, S., Nishino, Y ., and Ohta, K.: Occurrence of two forms of extracellular endoinulinase from Aspergillus niger mutant 817. J. Ferment. Bioeng., 78, 134-139 (1994). 8. Nakamura, T., Sbitara, A., Matsuda, S., Matsno, T., Suiko, M., and Ohta, K.: Production, purification and properties of an endoinulinase of PeniciNium sp. TN-88 that liberates inulotriose. J. Ferment. Bioeng., 84, 313-318 (1977). 9. Nakamura, T. and Nakatsu, S.: Action and production of inulinase. J. Jpn. Sot. Starch Sci., 35, 121-130 (1988). (in Japanese) 10. Onodera, S. and Shiomi, N.: Purification and substrate specificity of endo-type inulinase from Penicillium purpurogenum. Agric. Biol. Chem., 52, 2569-2576 (1988). 11. Yokota, A., Yamauchi, O., and Tomita, F.: Production of inulotriose from inulin by inulindegrading enzyme from Streptomyces rochei E87. Lett. Appl. Microbial., 21, 330-333 (1995). 12. Roberfroid, M.: Dietary fiber, inulin, and oligofructose: a review comparing their physiological effects. Crit. Rev. Food Sci. Nutr., 33, 103-148 (1993). 13. Yun, J. W.: Fructooligosaccharides-occurrence, preparation, and anulication. Enzvme Microb. Technol.. 19. 107-117 (1996): _ 14. Kuriki, T., Tsuda, M., and Imanaka, T.: Highly branched oligosaccharides production by the transglucosylation reaction of neonullulanase. J. Ferment. Bioena.. 73. 198-202 (1992). 15. Hidaka, H., Eida, T., and Saitoh, YL industrial production of fructo-oligosaccharides and its application for human and animals. Nippon Nogeikagaku Kaishi, 61, 915-923 (1987). (in Japanese) 16. Tomomatsu, H.: Health effects of oligosaccharides. Food Technol., October, 61-65 (1994). 17. Wada, K., Watanabe, J., Mizutani, J., Tomoda, M., Suzuki, H., and Saitoh, Y.: Effect of soybean oligosaccharides in a beverage on human fecal flora and metabolites. Nippon Nogeikagaku Kaishi, 66, 127-135 (1992). (in Japanese) 18. Bajpai, P. and Margartis, A.: Immobilization of Kluyveromyces marxianus cells with inulinase activity in agar gel. J. Gen. Appl. Microbial., 31, 297-304 (1985). 19. Bajpai, P. and Margartis, A.: Production of high fructose syrups from Jerusalem artichoke tubers using Kluyveromyces marxianus immobilized in agar gel. J. Gen. Appl. Microbial., 31, 305-311 (1985). 20. Bajpai, P. and Margartis, A.: Immobilization of Kluyveromyces marxianus cells containing inulinase activity in open pore gelatin matrix: II. Application for high fructose syrup production. Enzyme Microb. Technol., 7, 459-461 (1985). 21. Guiraud, J.P., Demeulle, S., and Galzy, P.: Inulin hydrolysis by the Debaryomyces phaJ7ii inulinase immobilized in DEAEcellulose. Biotechnol. Lett., 3, 683-688 (1981). 22. Guiraud, J. P., Bajon, A. M., Chautard, P., and Galzy, P.: Inulin hydrolysis by an immobilized yeast cell reactor. Enzyme Microb. Technol., 5, 185-190 (1985).
VOL. 87, 1999
MICROBIAL
23. Norman, B. E. and Hojer-Perderson, B.: The production of fructooligosaccharides from inulin or sucrose using inulinase or fructosyltransferase from Aspergillus ficuum. Denpun Kagaku, 36, 103-111 (1989). 24. Onodera, 0. and Shiomi, N.: lmmobilization of endo-type inulinase with K-carrageenan and characterization of the immobilized enzyme. .I. Rakuno Gakuen Uni., 14, 17-21 (1989). (in Japanese) 25. Kim, D. H., Choi, Y. J., Song, S. K., and Yun, J. W.: Production of inulo-oligosaccharides using endo-inulinase from a Pseudomonas sp. Biotechnol. Lett., 19, 369-371 (1997). 26. Yun, J. W., Kim, D. H., Kim, B. W., and Song, S. K.: Comparison of sugar compositions between inulo- and fructooligosaccharides produced by different enzyme forms. Biotechnol. Lett., 19, 553-556 (1997). 27. Yun, J. W., Kim, D. H., Uhm, T. B., and Song, S. K: Produc-
28.
29.
30. 31.
PRODUCTION
OF INULO-OLIGOSACCHARIDES
295
tion of high-content inulo-oligosaccharides from inulin by a purified endoinulinase. Biotechnol. Lett., 19, 935-938 (1997). Yun, J. W., Kim, D. H., Yoon, H. B., and Song, S. K.: Effect of inulin concentration on the production of inulo-oligosaccharides by soluble and immobilized endoinulinase. J. Ferment. Bioeng., 84, 365-368 (1997). Yun, J. W., Kim, D. H., Kim, B. W., and Song, S. K.: Production of inulo-oligosaccharides from inulin by immobilized endoinulinase from Pseudomonas sp. J. Ferment. Bioeng., 84, 369-371 (1997). Park, J. P., Kim, D. H., Kim, D. S., and Yun, J. W.: Enzymatic production of inulo-oligosaccharides from chicory juice. Biotechnol. Lett., 20, 385-388 (1998). Eom, S. J., Kwon, Y. M., and Choi, Y. J.: Molecular cloning of Pseudomonas sp. inulinase gene and its expression in E. coli. Kor. .I. Appl. Microbial. Biotechnol., 16, 484-488 (1988).