G Model

ARTICLE IN PRESS

BIOTEC-6633; No. of Pages 5

Journal of Biotechnology xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Journal of Biotechnology journal homepage: www.elsevier.com/locate/jbiotec

Short communication

Structural basis of a mutant Y195I ␣-cyclodextrin glycosyltransferase with switched product specificity from ␣-cyclodextrin to ␤-/␥-cyclodextrin Ting Xie a,c,1 , Yanjie Hou b,1 , Defeng Li b , Yang Yue a,d , Shijun Qian a,c , Yapeng Chao a,c,∗ a

State Key Laboratories of Transducer Technology, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China National Laboratory of Biomacromolecules, Institute of Biophysics, Chinese Academy of Sciences, 15 Datun Road, Chaoyang District, Beijing 100101, China c CAS Key Laboratory of Microbial Physiological and Metabolic Engineering, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China d Key Laboratory for Biomechanics and Mechanobiology of Ministry of Education, School of Biological Science and Medical Engineering, Beihang University (BUAA), Beijing 100191, China b

a r t i c l e

i n f o

Article history: Received 22 November 2013 Received in revised form 17 February 2014 Accepted 4 March 2014 Available online xxx Keywords: Y195I ␣-CGTase Crystallization Crystal structure Product specificity

a b s t r a c t Cyclodextrin glycosyltransferase (EC 2.4.1.19) (CGTase) is an extracellular bacterial enzyme which has the unique capability of forming cyclodextrins from starch. Our previous investigation revealed that a mutant Y195I ␣-CGTase drastically altered the cyclodextrin specificity by switching toward the synthesis of both ␤- and ␥-CDs (Xie et al., 2013a,b). In this study, we determined one X-ray structure of the mutant Y195I ˚ The overall structure was similar to that of the typical ␤-CGTase from Bacillus circulans ␣-CGTase at 2.3 A. 251, with minor difference in flexible domains since they showed about 70% homogeneity of amino acid sequences. The central site with isoleucine tended to be more flexible than tyrosine thus made the sugar chain, during the cyclization process, form a larger cyclodextrin like ␤- and ␥-CDs surrounding the central site instead of ␣-CD. Superposition of the structure of Y195I ␣-CGTase with those of ␤-CGTase and ␥CGTase showed that residues Lys232, Lys89 and Arg177 at subsites +2, −3 and −7 could form smaller substrate binding cavity. In summary, the crystal structure revealed that moderate increase of mobility of the central site resulted in the switched product specificity from ␣-CD to ␤- and ␥-CDs of the mutant Y195I ␣-CGTase. The space differences alongside the active domain may be another factor that impacts the product specificity of the CGTase. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Cyclodextrin glycosyltransferase (EC 2.4.1.19, CGTase) can convert starch to cyclodextrin through cyclization reactions. It can be classified into ␣-CGTase, ␤-CGTase, or ␥-CGTase according to their main products. The first structure of ␤-CGTase was elucidated at a resolution of 3.4 A˚ from Bacillus circulans 251, using multiple isomorphous replacement and solvent flattening for phase determination by Hofmann et al. in 1989 (Birgit et al., 1989). After that, a series of structures of ␤-CGTases and their mutant enzymes were determined (Strokopytov et al., 1995; Uitdehaag et al., 2000; Kanai et al.,

∗ Corresponding author at: State Key Laboratories of Transducer Technology, National Engineering Lab for Industrial Enzymes, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101, China. Tel.: +86 10 64807428; fax: +86 10 64807428. E-mail address: [email protected] (Y. Chao). 1 These authors contributed equally to this paper.

2001; Dauter et al., 1999). According to them, CGTases consist of five domains A–E (Klein et al., 1992). Domain A and B constitute the catalytic domains, whereas C and E are specialized in binding to raw starch granules (Lawson et al., 1994; Penninga et al., 1996). The function of domain D remains to be known (Uitdehaag et al., 1999a). The active site of Bacillus circulans strain 251 CGTase is comprised of at least nine sugar binding subsites, labeled −7 to +2 (Strokopytov et al., 1996). Bond cleavage located between subsites −1 and +1, results in an intermediate that is covalently linked to Asp-229 at subsite −1 (Uitdehaag et al., 1999b). However, most of the existing X-ray structures of CGTases are ␤-CGTases and a few of ␥-CGTases (Strokopytov et al., 1995; Uitdehaag et al., 2000). The typical structure of ␣-CGTase is rare (Choe et al., 2003). One major issue in cyclodextrin (CD) production is that CGTases always produce a mixture of CDs (including ␣-, ␤-, and ␥-CDs). Selective purification step is thus required to obtain pure ␣-CD, ␤-CD, or ␥-CD, through the use of complexing agents during CD synthesis to obtain high yield of the aimed products (Son et al., 2008).

http://dx.doi.org/10.1016/j.jbiotec.2014.03.014 0168-1656/© 2014 Elsevier B.V. All rights reserved.

Please cite this article in press as: Xie, T., et al., Structural basis of a mutant Y195I ␣-cyclodextrin glycosyltransferase with switched product specificity from ␣-cyclodextrin to ␤-/␥-cyclodextrin. J. Biotechnol. (2014), http://dx.doi.org/10.1016/j.jbiotec.2014.03.014

G Model BIOTEC-6633; No. of Pages 5

ARTICLE IN PRESS T. Xie et al. / Journal of Biotechnology xxx (2014) xxx–xxx

2

Thus, exploration of the mechanism by which the enzyme determines the size of synthesized CDs, as well as the strategies to improve the CDs specificity, became intriguing. Many efforts were taken to change the product specificities. Amino acid residues at +2 subsite (183), +1 subsite (194, 195), −3 subsite (47, 89, 196, 371), −6 subsite (167), and −7 subsite (145, 146, 147) were experimentally found to influence the cyclizing activity of the enzyme. In our previous studies, both mutant Y195I ␣-CGTase and mutant Y195I/Y167H ␣-CGTase were constructed based on the wild-type ␣-CGTase from Bacillus sp. 602-1. And they were found to be able to drastically alter the cyclodextrin specificity by showing a switching effect from the synthesis of ␣-CD toward both ␤and ␥-CDs, especially ␥-CD (Xie et al., 2013a,b). In this study, we determined the X-ray structure of the mutant Y195I ␣-CGTase and further make the comparison of the three-dimensional structure of the enzyme with the reported ␤-CGTase and ␥-CGTase with the attempt to explore the specificity mechanism of CGTases. 2. Materials and methods

Table 1 Data collection and refinement statistics of the mutant Y195I ␣-CGTase. Data collection Space group Unit cell parameters (Å) Wavelength (Å) Resolution range (Å) Total observations Unique observations Rmerge Mean (I/ I ) Completeness (%) Redundancy Wilson plot B (Å2 ) Refinement Rfree (%) Rwork (%) Model quality RMSD bond length (Å) RMSD bond angles (◦ ) Ramachandran plot (%)

2.1. Strains, plasmids, enzymes and reagents E. coli BL21(DE3) used as the expression host for the CGTase was purchased from TransGen Biotech (Beijing, China). The recombinant plasmid that contains the Y195I mutant cgt gene was constructed in our lab. Polynucleotide kinase, DpNI and Taq DNA ligase were from New England Biolabs (NEB, UK). The plasmid extraction kit was obtained from Tiangen (Valencia, CA, USA). DNA sequencing was performed by Tsingke (Beijing, China). The NiNTA agarose column was purchased from GE Healthcare (USA). The molecular weight cutoff of the ultrafiltration tube was 30 kDa (pall, USA). Other reagents were of analytical or biological grade. 2.2. Expression and purification of the mutant Y195I CGTase Firstly, inoculate a single colony of mutant Y195I CGTase strain into 5 mL LB medium containing 100 ␮g/mL ampicillin in a 250 mL flask. Next, incubate the flask with vigorous shaking at 37 ◦ C for approximately 15 h. Then the cultures with the 1% inoculation quantity were transferred to 5 mL of TB medium with 100 ␮g/mL ampicillin, and kept on cultivating at 37 ◦ C. Till the OD600 reached 0.6–0.8, the medium was cooled on ice and the mutants were then induced with isopropyl ␤-d-1-thiogalactopyranoside (IPTG) at 0.01 mM final concentration at 16 ◦ C. After 24 h of cultivation, 750 ␮L of 1 M glycine and 100 ␮L of 1 M CaCl2 were added, and the whole expression-inducing process lasted 96 h. The purification process was partially referred to Xue’s method (Xue et al., 2011). First, the supernatant obtained by centrifugation was treated by ammonium sulfate fractional precipitation from a saturation level of 25–60%. Next, the precipitant was resolved and loaded onto Ni-NTA affinity chromatography and Gel filtration chromatography Sephacryl-100. Fractions containing enzyme activity were pooled and concentrated by ultrafiltration. The enzyme was further purified to homogeneity by two steps of FPLC purification. The protein concentration was measured with Bradford method at 595 nm (Bradford, 1976).

P65 a = b = 102.59, c = 114.37 ˛ = ˇ = 90,  = 120 1.5418 48.09–2.30 (2.42–2.30) 307,543 (44,129) 30,397 (4400) 0.095 (0.502) 19.0 (4.2) 100.0 (100.0) 10.1(10.0) 38.5 23.21 17.32

0.012 1.103 657 (96.2%) favored 26 (3.8%) allowed 0 (0%) outlier

Average B-factors (Å2 ) Protein Metal ion Water

32.157 32.620 30.982

Model contents Protein residues Metal ions Water molecules

685 2 Ca2+ 272

Values in parentheses are for the data in the highest resolution shell.

recorded in time. Crystals appeared in 10–15 days and continued to grow up to 20–25 days. 2.4. Data collection and structure determination The crystals were soaked in reservoir solution supplemented with 17% (v/v) glycerol, followed by flash-cooling to 100 K. The data were collected using MM-007HF rotating-anode X-ray generator (a Rigaku R-AXIS IV ++ image-plate system mounted on a rotatinganode X-ray source operating at 40 kV and 20 mA) at wavelength ˚ A total of 360 images were collected with 1◦ oscillation per 1.5418 A. image. The data were integrated using the program MOSFLM V7.0.4 and scaled by SCALA v.6.1 from CCP4 package (Winn et al., 2011). The structure was determined using Phaser (McCoy et al., 2007) with the similar protein cyclodextrin glucanotransferase (PDB ID 3BMW) as the search model. Refinement involved iteration of manual model-building in COOT (Emsley and Cowtan, 2004) followed by computational modeling in Phenix.refine (Adams et al., 2011) using standard stereochemical restrains in conjunction with a randomly chosen Rfree set comprising 5% of the reflections. The final model was checked by the CCP4 program PROCHECK (Laskowski et al., 1993). The statistics of data collection and refinement were summarized in Table 1. Superposition of the two structures was done by using PyMOL, and all figures presented here were also prepared by using PyMOL.

2.3. Crystallization 3. Results and discussion The hanging-drop vapor diffusion method was used for the crystallization of purified Y195I CGTase. Crystals were obtained by mixing equal volume of protein (8 mg mL−1 ) and the reservoir solution containing 0.2 M sodium acetate trihydrate, 0.1 M Tris hydrochloride pH 9.0, 30% (w/v) polyethylene glycol 4000 at room temperature. Crystal growth observation was recorded once every 2 days, and each parameter of the crystallization conditions was

The wild-type ␣-CGTase produced cyclodextrins mixture with the ratio of ␣-CD, ␤-CD, and ␥-CD of 68.9, 22.2, and 8.9 respectively. In comparison, the mutant Y195I CGTase resulted in a particularly lower yield of ␣-CD (30), but a higher yield of ␤-CD (33.3) and ␥-CD (36.7) at a different extent (Xie et al., 2013a,b). Most interestingly, the mutant enzyme could improve the production

Please cite this article in press as: Xie, T., et al., Structural basis of a mutant Y195I ␣-cyclodextrin glycosyltransferase with switched product specificity from ␣-cyclodextrin to ␤-/␥-cyclodextrin. J. Biotechnol. (2014), http://dx.doi.org/10.1016/j.jbiotec.2014.03.014

G Model BIOTEC-6633; No. of Pages 5

ARTICLE IN PRESS T. Xie et al. / Journal of Biotechnology xxx (2014) xxx–xxx

3

Fig. 1. (A) An overall view structure of the Y195I mutant ␣-CGTase, (B) the overlap view of the mutant Y195I ␣-CGTase (yellow) and the wild-type ␤-CGTase (blue) and (C) the active sites of the solved mutant Y195I ␣-CGTase. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

of ␥-CD by about four times. It was known that the mutant Y195L ␤-CGTase from Alkalophilic Bacillus sp. 1011 produced considerably more ␥-cyclodextrin than the wild-type enzyme and virtually no ␣-cyclodextrin (Nakamura et al., 1994). To this point of view, it indicated that Tyr195 might play an important role in product specificity of the CGTases. 3.1. Crystallization and preliminary X-ray diffraction data of the mutant Y195I ˛-CGTase The pure mutant Y195I CGTase was successfully obtained at electrophoresis level and with high activity. Typically, the crystals were about 0.40 mm in length, with a width of about 0.10 mm. Finally, a 2.30 A˚ diffraction data set used for solution and refinement of the structure was obtained at cryogenic temperature. The crystal was hexagonal, space group P65 , with unit cell parameters ˚ c = 119.3 A, ˚ ˛ = ˇ = 90◦ , ␥ = 120◦ (Table 1). a = b = 140.2 A, 3.2. Overall three-dimensional structure and active domain of the mutant Y195I ˛-CGTase

these two types of CGTase. Yet minor difference was still observed. Residue Arg146 at −7 subsite in ␣-CGTase was positively charged, instead of a neutrally charged, serine146 in ␤-CGTase. The side chain of arginine is longer than that of the serine which may result in greater space steric hindrance at subsite −7 that determining the length of oligosaccharides binding to the active site. This could be the partial reason why the main product of ␣-CGTase was ␣-CD consisting of six glucosides instead of ␤-CD or ␥-CD. Sequences alignment among ␣-, ␤-, and ␥-CGTase suggested that ␥-CGTase had lower homology with ␣-, and ␤-CGTase in the active domain (Fig. 3). Most residues at active sites in ␣- and ␥-CGTase were also conservative. Four different residues were observed at three subsites (Fig. 4). The first two were residues Ala144 (144, −6) and Tyr260 (260, +3) at −6 and +3 subsites in ␣CGTase, respectively, corresponding to residues Val166 and Phe278 in ␥-CGT. Although the residues were different but their properties were similar which may less affect the catalytic characteristics. The other two different residues were Tyr89 and Lys232, in ␣-CGTase, corresponding to residues glycine (114, −3) and alanine (248, +2) in

We obtained one X-ray structure of the mutant Y195I ␣-CGTase ˚ It consisted of an N-terminal TIM barrel domain and three at 2.3 A. C-terminal IgG-like beta-barrel domains. The overall structure was similar to that of the typical ␤-CGTase from Bacillus circulans 251, with Ca atoms r.m.s.d of 1.5 A˚ (Fig. 1A and B). The electron density diagram of the main active sites in mutant Y195I ␣-CGTase was showed in Fig. 1C, based on the solved structure. It indicated the nine sugar binding subsites, labeled −7 to +2. Although no considerable three-dimensional structure changes were observed, the central site with isoleucine tended to be more flexible than tyrosine. So this made the sugar chain, during the cyclization process, form a larger cyclodextrin like ␤- and ␥-CDs surrounding the central site instead of ␣-CD as the biochemical investigations indicated (Xie et al., 2013a,b). This structure was used to compare the residues at active subsites of ␣-CGTase with that of the typical ␤- and ␥CGTase from Bacillus circulans 251 and Bacillus sp. Strain G-825-6 (Penninga et al., 1995; Kyoko et al., 2006). 3.3. Comparison of the mutant Y195I ˛-CGTase structure with the wild-type, typical ˇ-CGTase and -CGTase Structure superposition between mutant Y195I ␣-CGTase (yellow) and ␤-CGTase (blue) was shown in Fig. 2. The active site’s residues were almost identical and highly conservative between

Fig. 2. The close-up view of the active site showing the difference between mutant Y195I ␣-CGTase (yellow) and ␤-CGTase from Bacillus circulans 251 (blue). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

Please cite this article in press as: Xie, T., et al., Structural basis of a mutant Y195I ␣-cyclodextrin glycosyltransferase with switched product specificity from ␣-cyclodextrin to ␤-/␥-cyclodextrin. J. Biotechnol. (2014), http://dx.doi.org/10.1016/j.jbiotec.2014.03.014

G Model BIOTEC-6633; No. of Pages 5

ARTICLE IN PRESS T. Xie et al. / Journal of Biotechnology xxx (2014) xxx–xxx

4

Fig. 3. Sequences alignment among ␣-, ␤- and ␥-CGTases.

Fig. 4. The close-up view of the active site showing the difference between mutant Y195I ␣-CGTase (green) and ␥-CGTase from Bacillus sp. Strain G-825-6 (pink). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

␥-CGT. Obviously, tyrosine and lysine had long lateral chains than glycine and alanine which may form smaller substance binding cavities and tend to form smaller ␣-CD rather than ␥-CD. In conclusion, a crystal of the mutant Y195I ␣-CGTase was grown and the structure was solved. The overall structure was similar to that of the typical ␤-CGTase from Bacillus circulans 251, with minor difference in flexible domains. The central site with isoleucine tended to be more flexible than with tyrosine thus this made the sugar chain form more ␤- and ␥-CDs instead of ␣-CD. Residues such as Lys232, Lys89 and Arg177 at subsites +2, −3 and −7 in the active domain of the mutant Y195I could form smaller substrate binding cavity. Further study will be focused on the reconstruction alongside the active groove to improve the ␤- and ␥-CD, especially ␥-CD yield by expanding the substrate binding domain.

Acknowledgment This work was supported by Natural Science Foundation of China (Grant no. 31171643).

References Adams, P.D., Grosse-Kunstleve, R.W., Hung, L.-W., 2011. iMOSFLM: a new graphical interface for diffraction-image processing with MOSFLM. Acta Crystallogr. D 67, 271–281. Birgit, E.H., Hans, B., George, E.S., 1989. Three-dimensional structure of cyclodextrin glycosyltransferase from Bacillus circulans at 3.4 A˚ resolution. J. Mol. Biol. 209 (4), 793–800. Bradford, M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal. Biochem. 72, 248–254. Choe, H.W., Park, K.S., Labahn, J., 2003. Crystallization and preliminary X-ray diffraction studies of [alpha]-cyclodextrin glucanotransferase isolated from Bacillus macerans. Acta Crystallogr. D 59 (2), 348–349.

Dauter, Z., Dauter, M., Brzozowski, A.M., 1999. X-ray structure of Novamyl, the fivedomain “maltogenic” alpha-amylase from Bacillus stearothermophilus: maltose and acarbose complexes at 1.7 A˚ resolution. Biochemistry 38 (26), 8385–8392. Emsley, P., Cowtan, K., 2004. Coot: model-building tools for molecular graphics. Acta Crystallogr. D 60, 2126–2132. Kanai, R., Haga, K., Yamane, K., 2001. Crystal structure of cyclodextrin glucanotransferase from alkalophilic Bacillus sp. 1011 complexed with 1-deoxynojirimycin at 2.0 A˚ resolution. J. Biochem. 129 (4), 593–598. Klein, C., Hollender, J., Bender, H., Schulz, G.E., 1992. Catalytic center of cyclodextrin glycosyltransferase derived from X-ray structure analysis combined with sitedirected mutagenesis. Biochemistry 31, 8740–8746. Kyoko, H., Takeo, I., Satoshi, O., Hiroshi, M., 2006. Molecular cloning and characterization of a novel ␥-CGTase from alkalophilic Bacillus sp. Appl. Microbiol. Biotechnol. 70, 193–201. Lawson, C.L., Montfort, R., Strokopytov, B., 1994. Nucleotide sequence and X-ray structure of cyclodextrin glycosyltransferase from Bacillus circulans strain 251 in a maltose-dependent crystal form. J. Mol. Biol. 236, 590–600. Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. McCoy, A.J., Grosse-Kunstleve, R.W., Adams, P.D., 2007. Phaser crystallographic software. J. Appl. Crystallogr. 40, 658–674. Nakamura, A., Haga, K., Yamane, K., 1994. Four aromatic residues in the active center of cyclodextrin glucanotransferase from alkalophilic Bacillus sp. 1011: effects of replacements on substrate binding and cyclization characteristics. Biochemistry 33, 9929–9936. Penninga, D., Strokopytov, B., Rozeboom, H.J., 1995. Site-directed mutations in tyrosine 195 of cyclodextrin glycosyltransferase from Bacillus circulans strain 251 affect activity and product specificity. Biochemistry 34 (10), 3368– 3376. Penninga, D., van der Veen, B.A., Knegtel, R.M.A., 1996. The raw starch binding domain of cyclodextrin glycosyltransferase from Bacillus circulans strain 251. J. Biol. Chem. 271, 32777–32784. Son, Y.J., Rha, C.S., Park, Y.C., 2008. Production of cyclodextrins in ultrafiltration membrane reactor containing cyclodextrin glycosyltransferase from Bacillus macerans. J. Microbiol. Biotechnol. 18, 725–729. Strokopytov, B., Knegtel, R.M.A., Penninga, D., 1996. Structure of cyclodextrin glycosyltransferase complexed with a maltononaose inhibitor at 2.6 A˚ resolution. Implications for product specificity. Biochemistry 35, 4241– 4249. Strokopytov, B., Penninga, D., Henriette, J.R., 1995. X-ray structure of cyclodextrin glycosyltransferase complexed with acarbose. Implications for the catalytic mechanism of glycosidases. Biochemistry 34, 2234–2240.

Please cite this article in press as: Xie, T., et al., Structural basis of a mutant Y195I ␣-cyclodextrin glycosyltransferase with switched product specificity from ␣-cyclodextrin to ␤-/␥-cyclodextrin. J. Biotechnol. (2014), http://dx.doi.org/10.1016/j.jbiotec.2014.03.014

G Model BIOTEC-6633; No. of Pages 5

ARTICLE IN PRESS T. Xie et al. / Journal of Biotechnology xxx (2014) xxx–xxx

Uitdehaag, J.C.M., Kalk, K.H., van der Veen, B.A., 1999a. The cyclization mechanism of cyclodextrin glycosyltransferase (CGTase) as revealed by a ␥-cyclodextrinCGTase complex at 1.8-Å resolution. J. Biol. Chem. 274, 34868–34876. Uitdehaag, J.C.M., Mosi, R., Kalk, K.H., 1999b. X-ray structures along the reaction pathway of cyclodextrin glycosyltransferase elucidate catalysis in the ␣-amylase family. Nat. Struct. Mol. Biol. 6, 432–436. Uitdehaag, J.C., van, A.G.J., van, D.V.B.A., 2000. Structures of maltohexaose and maltoheptaose bound at the donor sites of cyclodextrin glycosyltransferase give insight into the mechanisms of transglycosylation activity and cyclodextrin size specificity. Biochemistry 39 (26), 7772–7780. Winn, M.D., Ballard, C.C., Cowtan, K.D., 2011. Overview of the CCP4 suite and current developments. Acta Crystallogr. D 67, 235–242.

5

Xie, T., Yue, Y., Song, B.H., Chao, Y., Qian, S., 2013a. Increasing of product specificity of ␥-cyclodextrin by mutating the active domain of ␣-cyclodextrin glucanotransferase from Paenibacillus macerans sp. 602-1. Chin. J. Biotechnol. 29 (9), 1–11. Xie, T., Song, B., Yue, Y., Chao, Y., Qian, S., 2013b. Site-saturation mutagenesis of central tyrosine 195 leading to diverse product specificities of an ␣-cyclodextrin glycosyltransferase from Paenibacillus sp. 602-1. J. Biotechnol. (November), http://dx.doi.org/10.1016/j.jbiotec.2013.10.032, pii:S0168-1656(13)00478-1. Xue, Z., Chao, Y., Wang, D., Wang, M., Qian, S., 2011. Overexpression of a recombinant amidase in a complex auto-inducing culture: purification, biochemical characterization, and regio- and stereoselectivity. J. Ind. Microbiol. Biotechnol. 38, 1931–1938.

Please cite this article in press as: Xie, T., et al., Structural basis of a mutant Y195I ␣-cyclodextrin glycosyltransferase with switched product specificity from ␣-cyclodextrin to ␤-/␥-cyclodextrin. J. Biotechnol. (2014), http://dx.doi.org/10.1016/j.jbiotec.2014.03.014