Succinylation of cyclodextrin glycosyltransferase from Thermoanaerobacter sp. 501 enhances its transferase activity using starch as donor

Journal of Biotechnology 86 (2001) 71 – 80 www.elsevier.com/locate/jbiotec

Succinylation of cyclodextrin glycosyltransferase from Thermoanaerobacter sp. 501 enhances its transferase activity using starch as donor Miguel Alcalde a, Francisco J. Plou a,*, M. Teresa Martı´n a, Israel Valde´s b, Enrique Me´ndez b, Antonio Ballesteros a a

Departamento de Biocata´lisis, Instituto de Cata´lisis, C.S.I.C., Cantoblanco, 28049 Madrid, Spain b Centro Nacional de Biotecnologı´a, C.S.I.C., Cantoblanco, 28049 Madrid, Spain

Received 20 June 2000; received in revised form 23 November 2000; accepted 1 December 2000

Abstract A simple modification procedure, the succinylation of amino groups, was suitable to increase the transferase (disproportionation) activity of cyclodextrin glycosyltransferase (CGTase) from Thermoanaerobacter sp. 501 using different linear oligosaccharides as acceptors. On the contrary, the synthesis of cyclodextrins (CDs), the coupling of CDs with oligosaccharides, and the hydrolysis of starch decreased after chemical modification. The degree of succinylation of amino groups (45%) was accurately determined by MALDI-TOF mass spectrometry. The formation of CDs under industrial conditions was analyzed for native and succinylated CGTases, showing similar selectivity to a-, b-, g-CD. The acceptor reaction with D-glucose using soluble starch as glucosyl donor was studied at 60°C and pH 5.5. Malto-oligosaccharides (MOS) production was notably higher using the semisynthetic enzyme at different ratios (w/w) starch:D-glucose. Thus, more than 90% of the initial starch was converted into MOS (G2– G7) in 48 h employing a ratio donor:acceptor 1:2 (w/w). © 2001 Elsevier Science B.V. All rights reserved. Keywords: Acceptors; CGTase; Chemical modification; Disproportionation; Oligosaccharides

1. Introduction Cyclodextrin glycosyltransferase (CGTase; EC 2.4.1.19) catalyzes the conversion of starch and Abbre6iations: CDs, cyclodextrins; CGTase, cyclodextrin glycosyltransferase; G2, maltose; MOS, malto-oligosaccharides. * Corresponding author. Tel. + 34-91-5854814; fax: + 3491-5854760. E-mail address: [email protected] (F.J. Plou).

related a(1“ 4) glucans to cyclodextrins (CDs) through an intramolecular transglycosylation reaction (Tonkova, 1998). Besides cyclization, the enzyme also catalyzes several intermolecular transglycosylations —coupling (opening of CD rings and transfer of resulting linear maltooligosaccharides (MOS) to acceptors), disproportionation (transfer of linear MOS to acceptors) and saccharifying (hydrolysis of starch) (Nakamura et al., 1993).

0168-1656/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 6 5 6 ( 0 0 ) 0 0 4 2 2 - 3

72

M. Alcalde et al. / Journal of Biotechnology 86 (2001) 71–80

A wide range of compounds including carbohydrates (Vetter et al., 1992), glucosides (Lee et al., 1993), sugar alcohols (Sato et al., 1991), vitamins (Aga et al., 1991) or flavonoids (Kometani et al., 1994) may act as acceptors in intermolecular transglycosylations (acceptor reaction) by disproportionation or coupling. New CGTases characterized by a high thermostability have been found in a group of thermophilic anaerobic microorganisms belonging to the genus Thermoanaerobacter. The CGTase from Thermoanaerobacter sp. has 683 residues, and a molecular mass of 75.5 kDa (Norman and Jorgensen, 1992). The optimum temperature for this enzyme is 85°C at pH 5.5, and it is sufficiently stable at 105°C in the presence of starch for use in the industrial jet-cooking process. This fact allows the enzymatic synthesis of CDs using one single enzyme (Zamost et al., 1991). The explosion in commercial and synthetic applications of CGTases has stimulated much of the interest in enhancing enzyme functionality and stability. An extraordinary effort is being done on site-directed mutagenesis of CGTase, trying to get enzymes with enhanced selectivity towards a particular cyclodextrin or trying to minimize side reactions (esp. coupling) (Klein et al., 1992; Sin et al., 1994; Dijkhuizen et al., 1995, 1999). On the other hand, covalent chemical modification — the original method available for studying enzyme mechanisms — has proven to be an interesting approach to enhance the activity, specificity, selectivity, stability and/or resistance to denaturants of numerous CGTases (Mattson et al., 1992; Villette et al., 1992, 1993; Jeang and Lin, 1994; Ohnishi et al., 1994a,b; Alcalde et al., 1998, 1999). Chemical modification is a generally applicable tool because modified biomolecules are easy to prepare in large scale from commercial material. In the present work, chemical modification of amino groups (lysines+ amino terminal) of CGTase from Thermoanaerobacter sp. 501 with succinic anhydride was carried out. Succinic anhydride reacts with amino groups to give a stable modified derivative, which has negative charge at pH\5. Our aim was to understand whether a simple modification procedure was suitable for a significant improvement of enzyme

properties. The effect of succinylation on the different activities of CGTase was evaluated.

2. Methods CGTase from Thermoanaerobacter sp. 501 was kindly provided by Novo Nordisk. Succinic anhydride, phenolphthalein, bromcresol green, MOS (G1 –G7), a-, b- and g-CD, methyl a-D-glucopyranoside, amyloglucosidase (EC 3.2.1.3) from Aspergillus niger, sinapinic acid and hydroxylamine were purchased from Sigma. Glucose GOD-Perid reagent, p-nitrophenyl-a-D-maltoheptaoside-4,6O-ethylidene (EPS) and a-glucosidase (EC 3.2.1.20) from Saccharomyces cere6isiae, were from Boehringer Mannheim. 3,5-Dinitrosalicylic acid (DNS) was from Fluka. Methyl orange was from Aldrich. Ampholites (pH 3–9) and pI markers for isoelectric focusing were from Bio-Rad. Potato soluble starch (Paselli SA2) was donated by Avebe (Foxhol, The Netherlands).

2.1. Enzyme purification CGTase was purified by FPLC using an affinity chromatography column (1× 12 cm) packed with Sepharose-6FF (Pharmacia) covalently coupled to a-cyclodextrin as described (Sundberg and Porath, 1974). Once the crude enzyme was loaded, the column was washed with 200 ml of 10 mM sodium acetate buffer (pH 5.5) — buffer 1 — at 2.5 ml min − 1, and the bound CGTase was further eluted with 50 ml of the same buffer containing a-cyclodextrin (10 mg ml − 1) — buffer 2. Fractions containing the CGTase were pooled and stored at − 20°C (in presence of a-CD). The concentration of protein was determined by Bradford Bio-Rad micro-assay with immunoglobulin G (IgG) as standard.

2.2. Chemical modification of amino groups CGTase was incubated in 10 mM phosphate buffer (pH 8.0) at 4°C with magnetic stirring at 250 rpm. Succinic anhydride dissolved in dry acetone (27 mg ml − 1) was added in two portions of 40 ml. A molar ratio of 20:1 of reagent with

M. Alcalde et al. / Journal of Biotechnology 86 (2001) 71–80

respect to the sum of lysine and tyrosine residues was used. The reaction was maintained in a pHstat (Radiometer) at pH 8.0 with 0.2 N NaOH. Once the reaction was finished (approx. 1 h), the mixture was dialyzed against 10 mM phosphate buffer (pH 7.0) at 4°C in order to remove salt and excess reagents. Succinylated enzyme was treated with 0.5 M hydroxylamine (pH 7.0) for 5 h at 25°C, and dialyzed again. A blank was also performed in the absence of succinic anhydride.

2.3. Sample preparation for the analysis of nati6e and succinylated CGTases by MALDI-TOF mass spectrometry About 5 ml of native (0.4 mg ml − 1) or succinylated (0.2 mg ml − 1) CGTase were mixed in a Eppendorf tube with 5 ml of a matrix solution comprised of saturated sinapinic acid in 30% aqueous acetonitrile and 0.1% trifluoroacetic acid (TFA). Then 1 ml of these solutions were deposited on a stainless steel probe tip and allowed to dry at room temperature for 5 min. Samples were measured on a Bruker (Bremen, Germany) Reflex II MALDI-TOF mass spectrometer equipped with an ion source with visualization optics and a N2 laser (337 nm). Mass spectra were recorded in linear positive mode at 25 kV acceleration voltage with a delayed extraction time of 750 nanoseconds. A total of 200 spectra of single laser shots under threshold irradiance were accumulated. Only highly intense, well-resolved mass signals arising from five to seven selected target spots were considered. The equipment was externally calibrated employing single, double and triple charged signals from a mixture of bovine albumin (66 430 Da) and cytochrome C (12 360 Da).

2.4. Circular dichroism spectra Circular dichroism spectra were obtained using a Jasco J-720 Spectropolarimeter in the range 190 – 260 nm. Then 150 ml of enzyme solution was deposited into the cell (0.5 mm). The protein concentration was in the range of 0.15 – 0.25 mg ml − 1.

73

2.5. Assays of initial acti6ities The different CGTase activities were measured at 85°C by incubating the enzyme with substrate solutions in 10 mM sodium citrate (pH 5.5) containing 0.15 mM CaCl2. At time intervals of 1 min, aliquots were removed and added to the corresponding reagent.

2.5.1. Cyclization acti6ity The production of CDs was followed colorimetrically via the formation of inclusion complexes with different organic compounds. Paselli SA2 (partially hydrolyzed potato starch, with an average degree of polymerization of 50) was used as substrate at a concentration of 5% (w/v) for band g-CD assays, and 2% (w/v) for a-CD. The formation of a-CD was assessed at 490 nm on the basis of its ability to form a stable, colorless inclusion complex with methyl orange (Hirai et al., 1981). b-CD was determined at 552 nm on the basis of its ability to form a stable, colorless inclusion complex with phenolphthalein (Penninga et al., 1995). g-CD was determined measuring the color increase at 630 nm due to the formation of an inclusion complex with bromcresol green (Kato and Horikoshi, 1984). One unit of activity was defined as the amount of enzyme able to produce 1 mmol of a/b/g-CD per minute under the corresponding conditions. 2.5.2. Coupling a- or b-CDs (2.5 mM) were used as donors and methyl-a-D-glucopyranoside (10 mM) as acceptor (Nakamura et al., 1993). The linear oligosaccharide formed was converted into single glucose units by the action of amyloglucosidase. The concentration of glucose was accurately detected with the glucose/GOD-Perid reagent. One unit of activity was defined as the amount of enzyme able to convert 1 mmol of cyclodextrin per minute under the above conditions. 2.5.3. Disproportionation The reaction was based on the method described by Nakamura et al. (1994). EPS (p-nitrophenyl-a-D-maltoheptaoside-4-6-O-ethylidene, 3 mM) was used as donor and maltose (G2) (10

M. Alcalde et al. / Journal of Biotechnology 86 (2001) 71–80

74

mM) as acceptor. In this reaction, EPS is cleaved and the G2 is coupled to the free reducing end. The p-nitrophenol may be cleaved from the reaction product by the action of a-glucosidase. One unit of activity was defined as the amount of enzyme able to release 1 mmol of p-nitrophenol per minute under these conditions.

2.5.4. Saccharifying Hydrolytic activity was assessed using 2% potato soluble starch as substrate, measuring the increase in reducing ends. The reducing power was measured with dinitrosalicylic acid (Bernfeld, 1955). One unit of activity was defined as the amount of enzyme able to release 1 mmol of reducing end per minute under the corresponding conditions.

Varian 9012 pump and two Aminex HPX-42A columns (300× 7.8 mm, Bio-Rad) put in series. Water was used as mobile phase (0.7 ml min − 1). The column temperature was kept constant at 85°C. Detection was performed using a refractionindex detector (Varian). Integration was carried out using the Varian Star 4.0 software.

2.9. Thin-Layer Chromatography Acceptor reaction was also followed by analytical TLC performed on silica gel 60 plates (Merck) using water/ethanol/butanol (2:5:3, v/v/v) as eluent (Del Rio et al., 1997). Spots were detected by inmersion in a solution of orcinol/ferric chloride (Bial’s reagent) diluted with ethanol (4 vol), drying and heating at 120°C for 5 min.

2.6. Production assay The production of CDs and oligosaccharides was also assayed under similar conditions to those used in industry. CGTase (0.7 mg ml − 1) was incubated during 8 days at 85°C with 25% (w/v) soluble starch in 10 mM sodium citrate buffer (pH 5.5) containing 0.15 mM CaCl2. At different times aliquots of 300 ml were removed and mixed with 300 ml of 0.4 N NaOH in order to quench the reaction. Samples were centrifuged during 15 min at 8500×g, and analyzed by thin layer chromatography (TLC) and high-performance liquid chromatography (HPLC).

2.7. Acceptor reaction The acceptor reaction of CGTase using soluble starch as donor and D-glucose as acceptor was carried out. CGTase (0.7 mg ml − 1) was incubated at 60°C with 10% (w/v) soluble starch and different ratios (w/w) donor:acceptor (from 1:2 to 2:1) in 10 mM sodium citrate buffer (pH 5.5) containing 0.15 mM CaCl2. At different times, aliquots were removed and submitted to the same process described in the production assay.

2.8. HPLC analysis Samples were analyzed by HPLC using a

3. Results and discussion

3.1. Chemical modification of CGTase CGTase was purified by affinity chromatography yielding pure enzyme (assessed by SDSPAGE and isoelectrofocusing) with a specific b-cyclodextrin-forming activity of 252 U mg − 1. The amino groups of CGTase were subjected to chemical modification with succinic anhydride. A 20-fold molar excess of succinic anhydride to the sum of amino and phenolic groups was used because the reaction of the dicarboxylic anhydride with the protein takes place in competition with its hydrolysis to succinic acid. Succinylation may radically alter the solubility of a protein or its ability to participate in intermolecular associations (Geoghegan, 1996). Short-chain acid anhydrides display a low selectivity towards the acylation of amino groups. Thus, the modification of side-chains, such as hydroxy groups of Tyr, thiol groups of Cys and imino groups of His is also possible. However, His and Cys residues are spontaneously deacylated under reaction conditions; O-succinyltyrosine residues suffer a rapid intramolecularly catalyzed hydrolysis at pH\5,

M. Alcalde et al. / Journal of Biotechnology 86 (2001) 71–80

which results in the regeneration of the original tyrosyl residues. Succinyl derivates of aliphatic hydroxy amino acids (Ser, Thr) do not spontaneously hydrolyze but can be cleaved by treatment with hydroxylamine (Lundblad, 1991; Means and Feeney, 1971).

3.2. Determination of modification degree The CGTase from Thermoanaerobacter sp. 501 contains 25 amino groups (24 lysines and the N-terminal a-amino group – Ala1-). The number of amino groups modified per mole of enzyme was accurately determined by MALDI-TOF mass spectrometry (each succinylation adds 100.1 Da). The number of succinylation sites was determined by simply obtaining the mass difference between native and succinylated CGTase molecules. The mass of 75 624 Da (Fig. 1) obtained for native CGTase was slightly lower than value of 75 626 Da from the protein sequence (calculated by the Expasy Molecular Biology Server). On the other hand, the mass of succinylated CGTase molecule was 76 751 Da (Fig. 1). The mass increase of 1127

75

Da with respect to the native molecule corresponds to incorporation of 11 succinic anhydride molecules in the CGTase (degree of substitution around 45%). This implies that approximately 11 amino groups (protonated at pHB 9) were converted into stable modified derivatives, which have a negative charge at pH\ 5. As a consequence of this change of charge, a shift of the isoelectric point of CGTase from neutral (approx. pI 6.3) to moderately acid values (approx. pI 5.2) was observed (data not shown). To evaluate whether the secondary structure of the protein had been altered by the chemical modification process, circular dichroism analysis of native and modified CGTases was carried out. The native structure proved to be maintained throughout the succinylation reaction, since no significant change was recognized in CD spectra after the modification (data not shown). This implied that possible changes in enzyme activity should not be a consequence of a substantial alteration of the protein structure.

Fig. 1. MALDI-TOF mass spectra of native and succinylated CGTases. A region of the mass spectra of both proteins has been combined in an inset for comparison. Conditions of MALDI-TOF analysis in Section 2.

M. Alcalde et al. / Journal of Biotechnology 86 (2001) 71–80

76

Table 1 Initial activities of native and succinylated CGTases from Thermoanaerobacter sp. 501a CGTase

Activity (U mg−1 protein) Cyclization

Native Succinylated a

Coupling

Disproportionation

a

b

g

a

b

290 225

250 220

150 115

270 115

48 23

Saccharifying

1530 1980

82 71

Reaction conditions described in Section 2.

Table 2 Disproportionation activities of native and succinylated CGTases from Thermoanaerobacter sp. 501 using different malto-oligosaccharides (G1–G7) as acceptors CGTase

Disproportionation activity (U mg−1 protein)a G1

G2

G3

G4

G5

G6

Native

385

725

690

435

350

310

240

Succinylated

475

835

800

500

405

370

275

a

G7

Conditions described in Section 2 (Section 2.5), but performing the reactions at 60°C with both enzymes.

3.3. Effect of succinylation on acti6ity The different activities of native and chemically-modified CGTases were evaluated (Table 1). These assays of activity are referred to a short time (0–10 min) as described in methodology section. The g-coupling assay was not carried out due to the interaction between g-CD and reagents (amyloglucosidase) of this assay, which disturbs the measurement. Interestingly, Table 1 shows that, with the exception of disproportionation activity, the succinylation of CGTase amino groups produced a moderate decrease of the specific activities. The cyclization activities of succinylated CGTase were around 1.3-fold lower than those of the native glycosyltransferase, and a similar reduction was observed for the saccharifying activity. Furthermore, the coupling of a- and b-cyclodextrin with oligosaccharides was 2.4 and 2.1 times faster, respectively for the native enzyme compared with the succinylated preparation. In contrast, the disproportionation activity was higher after modifi-

cation. This disproportionation activity (1980 U mg − 1) is, to our knowledge, the highest value ever reported for a CGTase [e.g. Thermoanaerobacterium thermosulfurigenes CGTase, — a CGTase very closely related to the CGTase from Thermoanaerobacter — displays a disproportionation activity of 330 U mg − 1 (Wind et al., 1998)]. In order to evaluate the behavior of modified CGTase in the disproportionation reaction, we analyzed the influence of the chain length of the MOS acceptor (G1 –G7, G1 being glucose) at 60°C. In all cases, the disproportionation activities of succinylated CGTase were higher than those using the native enzyme (Table 2). On the other hand, the best MOS acceptor for both CGTases was G2. When increasing the degree of polymerization of the acceptor, the disproportionation activity decreased, probably due to steric hindrance in the acceptor binding site. The succinylated CGTase was assayed under similar conditions (starch concentration, high temperature, pH) to those used in industry. The time-course of CD and oligosaccharide formation

M. Alcalde et al. / Journal of Biotechnology 86 (2001) 71–80

was analyzed. At the first stages of the reaction, the appearance of a-, b-, and g-CD is clearly the main process with respect to the formation of short oligosaccharides, but this ratio is inverted at the final stages due to the rest of activities. Fig. 2 shows the production of a-, b-, and g-CD with time for native and modified CGTases. These results seem to be related to the initial activities

Fig. 2. Production of a- ( ) b- ( ) and g-CD ( ) with time for native and succinylated CGTases. Conditions, 25% (w/v) soluble starch, 0.7 mg ml − 1 CGTase, 85°C, 10 mM sodium citrate buffer (pH 5.5) containing 0.15 mM CaCl2.

Fig. 3. Selectivity of CGTase from Thermoanaerobacter with respect to the formation of cyclodextrins in the range of pH 3.5–8.5 after 6 h of reaction. Conditions, 25% (w/v) soluble starch, 0.7 mg ml − 1 CGTase, 85°C, 10 mM buffer (pH 5.5) containing 0.15 mM CaCl2. Buffers employed, sodium citrate (pH 3.5– 5.5); MES (pH 6.5); Tris–HCl (pH 7.5–8.5).

77

shown in Table 1. Both native and succinylated CGTases reached the maximum production of CDs after around 24–48 h. Using the native enzyme, at the point of maximal productivity (24 h) the molar ratio a-: b-: g-CD was 2.9:5.6:1 (with a conversion of 20.5% starch into CDs); for the chemically-modified CGTase the maximum conversion to CDs (21.1%) was at 48 h, with a molar ratio 2.2:4.1:1. As shown, the selectivity a-: b-: g-CD is only slightly affected by the substitution of about 11 positive charges by negative ones. The above results are indicating that the enzyme selectivity is difficult to manipulate by changing the ionization state of residues, and that the ratio a-: b-: g-CD obtained for each CGTase depends intrinsically on the geometry of the active site. In this context, we analyzed the CGTase selectivity in the range of pH 3.5 –8.5 (Fig. 3). It remains clear that in the pH range 4.5 –8.5, only slight differences in selectivity are found, despite the great number of residues that change their ionization state throughout this pH interval. On the other hand, the oligosaccharide production after 8 days was higher with the succinylated preparation (4.4 vs. 3.1% starch converted into oligosaccharides), because of the increase in disproportionation activity after modification. In order to exploit the high transferase efficiency of this semisynthetic CGTase, the acceptor reaction using 10% (w/v) soluble starch as donor and D-glucose as acceptor was tested. Different weight ratio starch:D-glucose was studied. It has been demonstrated that D-glucose is one of the best acceptors for CGTase (Norman and Jorgensen, 1992). In fact, the basic requirement for a good acceptor of CGTase is the presence of a Dor L-glucopyranoside structure with equatorial unsubstituted hydroxyl groups at C-2, C-3 and C-4 (Rendleman, 1996). Although the optimal temperature of CGTase from Thermoanaerobacter sp. 501 for production of CDs is 85°C, the acceptor reaction was carried out at 60°C because the yields of acceptor products are comparable, but the formation of byproducts (furfurals) and CDs are considerably reduced at 60°C (data not shown). In the acceptor reaction, a mixture of homologous MOS is produced ranging from G2 to

M. Alcalde et al. / Journal of Biotechnology 86 (2001) 71–80

78

cose (from 1:2 to 2:1) on product distribution was tested. Table 3 summarizes the conversion of starch into MOS by native and succinylated CGTases. It is noteworthy that MOS production with semisynthetic CGTase was higher than with native enzyme, converting 91% of the initial starch when the ratio donor/acceptor was 1:2. On the other hand, the behavior of both CGTases as a function of the ratio donor/acceptor was similar. The lower the percentage of D-glucose, the higher the degree of polymerization of the acceptor products. Using a ratio donor/acceptor 2:1, a notable yield of MOS ranging G2 –G10 (76%) was obtained with succinylated CGTase. This data indicates that the initial concentrations of substrates determine the number of MOS formed and the yield of the overall process.

Fig. 4. HPLC analysis of the acceptor (disproportionation) reaction after 5 days with the succinylated CGTase from Thermoanaerobacter. Conditions, 10% (w/v) soluble starch, 5% D-glucose, 0.7 mg ml − 1 CGTase, 60°C, 10 mM sodium citrate buffer (pH 5.5) containing 0.15 mM CaCl2.

4. Conclusion The highest specific activity of the reactions catalyzed by Thermoanaerobacter sp. 501 CGTase corresponds to disproportionation, not cyclization. Furthermore, we have manipulated the CGTase by chemical modification in such a manner that just one of the reactions (in this case the disproportionation) is selectively enhanced. To our knowledge, succinylated CGTase from Thermoanaerobacter sp. 501 displays the highest disproportionation activity ever reported for a CGTase. This enzyme was obtained in a simple, easy to scale-up process, and is able to work in

maltodecaose (G10). This is because the acceptor products may also work as acceptors. Fig. 4 shows the HPLC chromatogram of the reaction mixture after 5 days, employing the chemicallymodified CGTase and a weight ratio (w/w) donor/ acceptor 2:1. The maximum conversion was achieved after 48 h regardless of the ratio employed. The effect of the ratio (w/w) starch:D-glu-

Table 3 Acceptor (disproportionation) reaction catalyzed by native and succinylated CGTases from Thermoanaerobacter sp. 501 using starch as donor and D-glucose as acceptor Ratio (w/w) Starch:D-glucose

1:2 1:2 1:1 1:1 2:1 2:1 a

CGTase

Native Succinylated Native Succinylated Native Succinylated

Conversion of starch into oligosaccharides (%)a G2

G3

G4

G5

G6

G7

G8

G9

G10

30 36 21 22 10 19

27 31 18 21 16 15

11 12 14 17 13 13

5 7 9 10 9 8

4 3 6 7 7 7

2 2 4 5 4 5

– – 2 3 2 4

– – 1 2 1 3

– – 0.5 1 1 2

Total (%) 79 91 75 88 63 76

Percentage of initial soluble starch transformed into malto-oligosaccharides at 48 h. Reaction conditions described in Section 2.

M. Alcalde et al. / Journal of Biotechnology 86 (2001) 71–80

the acceptor reaction with D-glucose producing MOS with high yield. We are currently exploring the transferase activity of this semisynthetic CGTase using acceptors of different nature.

Acknowledgements We thank Professor Juan Manuel Ramirez (C.I.B., C.S.I.C., Madrid) for help with the circular dichroism spectra. We thank our European colleagues Dr Sven Pedersen and Dr Carsten Andersen (Novo Nordisk, Denmark) and Professor Lubbert Dijkhuizen and Professor Bauke W. Dijkstra (University of Groningen) for help with this work. This work was funded by the European Union (project BIO2-CT94-3071) and the Spanish CICYT (project BIO98-0793). We thank Instituto Danone and Ministerio de Educacio´n y Cultura for financial support.

References Aga, H., Yoneyama, M., Sakai, S., Yamamoto, I., 1991. Synthesis of 2-O-a-glucopyranosyl L-ascorbic acid by cyclomaltodextrin glucanotransferase from Bacillus stearothermophilus. Agric. Biol. Chem. 55, 1751–1756. Alcalde, M., Plou, F.J., Pastor, E., Ballesteros, A., 1998. Effect of the chemical modification of cyclodextrin glycosyltransferase (CGTase) from Thermoanaerobacter sp. on its activity and product selectivity. Ann. New York Acad. Sci. 864, 183– 187. Alcalde, M., Plou, F.J., Andersen, C., Martı´n, M.T., Pedersen, S., Ballesteros, A., 1999. Chemical modification of lysine side chains of cyclodextrin glycosyltransferase (CGTase) from Thermoanaerobacter causes a shift from CGTase to a-amylase specificity. FEBS Lett. 445, 333–337. Bernfeld, P., 1955. Amylases, a and b. In: Colwick, S.P., Kaplan, N.O. (Eds.), Methods in Enzymology, vol. III. Academic Press, New York, pp. 149–158. Del Rio, G., Morett, E., Soberon, X., 1997. Did cyclodextrin glycosyltransferases evolve from a-amylases? FEBS Lett. 416, 221– 224. Dijkhuizen, L., Penninga, D., Rozeboom, H.J., Strokopytov, B., Dijkstra, B.W., 1995. Protein engineering of cyclodextrin glycosyltransferase from Bacillus circulans strain 251. In: Petersen, S.B., Svensson, B., Pedersen, S. (Eds.), Carbohydrate Bioengineering. Elselvier Science, Amsterdam, pp. 165– 174. Dijkhuizen, L., van der Veen, B.A., Uitdehaag, J., Dijkstra, B.W., 1999. Engineering of cyclodextrin glycosyltrans-

79

ferase. In: Gilbert, G.J., Davies, G.J., Henrissat, B., Svensson, B. (Eds.), Recent Advances in Carbohydrate Bioengineering. The Royal Society of Chemistry, Cambridge, pp. 263– 271. Geoghegan, K.F., 1996. Chemical modification of proteins. In: Coligan, J.E., Dunn, B.M., Ploegh, H.L., Speicher, D.W., Wingfield, P.T. (Eds.), Current Protocols in Protein Science. Wiley, New York Chapter 15. Hirai, H., Toshima, N., Uenoyama, S., 1981. Inclusion complex formation of cyclodextrin with large dye molecule. Polym. J. 13, 607– 610. Jeang, C.L., Lin, Y.W., 1994. Evidence for the presence of essential histidines on cyclodextrin glucanotransferase from Bacillus macerans. Biotechnol. Appl. Biochem. 19, 85 – 92. Kato, T., Horikoshi, K., 1984. Colorimetric determination of a-cyclodextrin. Anal. Chem. 56, 1738– 1740. Klein, C., Hollender, J., Bender, H., Schulz, G.E., 1992. Catalytic center of cyclodextrin glycosyltransferase derived from X-ray structure analysis combined with site-directed mutagenesis. Biochemistry 31, 8740– 8746. Kometani, T., Terada, Y., Nishimura, T., Takh, H., Okada, S., 1994. Transglycosylation to hesperidin by cyclodextrin glucanotransferase from an alkalophilic Bacillus species and in alkaline pH and properties of hesperidin glycosides. Biosci. Biotechnol. Biochem. 58, 1990– 1994. Lundblad, R.L., 1991. Chemical Reagents for Protein Modification. CRC Press, Boca Raton, Florida. Lee, Y.H., Baek, S.G., Shin, H.D., Park, D.C., 1993. Transgycosylation reaction of cyclodextrin glucanotransferase in the attrition coupled reaction system using raw starch as a donor. Kor. J. Appl. Microbiol. Biotechnol. 21, 461– 467. Mattson, P., Pohjalainen, T., Korpela, T., 1992. Chemical modification of cyclomaltodextrin glucanotransferase from Bacillus circulans var. Alkalophilus. Biochim. Biophys. Acta 1122, 33 – 40. Means, G.E., Feeney, R.E., 1971. Chemical modification of proteins. Holden-Day Inc, San Francisco. Nakamura, A., Haga, K., Yamane, K., 1993. Three histidine residues in the active center of cyclodextrin glucanotransferase from alkalophilic Bacillus sp. 1011. Effects of the replacement on pH dependence and transition-state stabilization. Biochemistry 32, 6624– 6631. Nakamura, A., Haga, K., Yamane, K., 1994. The transglycosylation reaction of cyclodextrin glucano transferase is operated by a ping-pong mechanism. FEBS Lett. 337, 66 – 70. Norman, B.E., Jorgensen, S.T., 1992. Thermoanaerobacter sp. CGTase: its properties and application. Denpun Kagaku 39, 101– 108. Ohnishi, M., Abe, M., Azuma, T., Kubota, M., Rokushika, S., 1994a. Tryptophan residues of Bacillus cycloamylase glucanotransferase: effect of modification with N-bromosuccinimide on the enzyme-catalyzed synthesis of cyclomaltoheptaose from maltotriose. Starch 46, 272– 275. Ohnishi, M., Azuma, T., Kubota, M., 1994b. Cycloamylose glucanotransferase-catalyzed cyclization for a substrate maltose. Modification with N-bromosuccinimide on the tryptophan residues. Starch 46, 276– 279.

80

M. Alcalde et al. / Journal of Biotechnology 86 (2001) 71–80

Penninga, D., Strokopytov, B., Rozeboom, J., Lawson, L., Dijkstra, W., Bergsma, J., Dijkhuizen, L., 1995. Site-directed mutations in Tyrosine 195 of cyclodextrin glycosyltransferase from Bacillus circulans strain 251 affect activity and product specificity. Biochemistry 34, 3368–3376. Rendleman, J.A., 1996. Influence of saccharide as inhibitors of cyclodextrin production. Biotechnol. Appl. Biochem. 24, 121– 127. Sato, M., Matsuo, T., Orita, N., Yagi, Y., 1991. Synthesis of novel sugars, oligoglucosyl-inositols, and their growth stimulating effect for Bifidobacterium. Biotechnol. Lett. 13, 69– 74. Sin, K.A., Nakamura, A., Masaki, H., Matsuura, Y., Uozumi, T., 1994. Replacement of an amino acid residue of cyclodextrin glucanotransferase of Bacillus ohbensis doubles the production of gamma-cyclodextrin. J. Biotechnol. 32, 283– 288. Sundberg, L., Porath, J., 1974. Preparation of adsorbents for biospecific affinity chromatography. J. Chromatogr. 90, 87– 98. Tonkova, A., 1998. Bacterial cyclodextrin glucanotransferase. Enzyme Microb. Technol. 22, 678–686.

Vetter, D., Thorn, W., Brunner, H., Koning, W.A., 1992. Directed enzymatic synthesis of linear and branched glucooligosaccharides using cyclodextrin-glucanosyltransferase. Carbohydr. Res. 223, 61 – 69. Villette, J.R., Sicard, P.J., Bouquelet, S.J.L., 1992. Cyclomaltodextrin glucanotransferase from Bacillus circulans E192: III. Chemical modification by diethylpyrocarbonate: evidence for an induced fit at the active site resulting from the binding of an acceptor. Biotechnol. Appl. Biochem. 15, 69 – 79. Villette, J.R., Helbecque, N., Albani, J.R., Sicard, P.J., Bouquelet, S.J.L., 1993. Cyclomaltodextrin glucanotransferase from Bacillus circulans E 192: nitration with tetranitromethane. Biotechnol. Appl. Biochem. 17, 205– 216. Wind, R.D., Buitelaar, R.M., Dijkhuizen, L., 1998. Engineering of factors determining a-amylase and cyclodextrin glycosyltransferase specificity in the cyclodextrin glycosyltransferase from Thermoanaerobacterium thermosulfurigenes EM1. Eur. J. Biochem. 253, 598– 605. Zamost, B.L., Nielsen, H.K., Starnes, R.L., 1991. Thermostable enzymes for industrial applications. J. Ind. Microbiol. 8, 71 – 82.

.