International Journal of Biological Macromolecules 76 (2015) 224–229
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Mutations at calcium binding site III in cyclodextrin glycosyltransferase improve -cyclodextrin specificity Xiaofeng Ban b , Zhengbiao Gu a,b,c , Caiming Li b , Min Huang b , Li Cheng a,b , Yan Hong a,b,c , Zhaofeng Li a,b,c,∗ a
State Key Laboratory of Food Science and Technology, Jiangnan University, Wuxi 214122, China School of Food Science and Technology, Jiangnan University, Wuxi 214122, China c Synergetic Innovation Center of Food Safety and Nutrition, Jiangnan University, Wuxi, Jiangsu 214122, China b
a r t i c l e
i n f o
Article history: Received 9 September 2014 Received in revised form 18 February 2015 Accepted 19 February 2015 Available online 4 March 2015 Keywords: Cyclodextrin glycosyltransferase -Cyclodextrin Residue 315 Calcium binding site Product specificity
a b s t r a c t Cyclodextrin glycosyltransferases (CGTases, EC 2.4.1.19) are industrially important enzymes that produce cyclodextrins from starch by intramolecular transglycosylation. In this study, the effects of amino acid residue at position 315 in calcium binding site III (CaIII) on product specificity of CGTase were investigated by replacing Ala315 in the CGTase from Bacillus circulans STB01 with arginine, aspartic acid, threonine, leucine and valine. The cgt gene, which encodes this enzyme, was expressed in B. subtilis WB600 alongside site-directed mutants A315R, A315D, A315T, A315L and A315V. The results showed that CaIII plays an important role in cyclodextrin product specificity. Replacement of Ala315 by charged amino acid residues enhanced -cyclodextrin specificity, compared with the wild-type CGTase. Mutations A315R and A315D resulted in an approximately 10% increase in -cyclodextrin activity. Furthermore, under conditions resembling the industrial production processes, the mutants A315R and A315D displayed obvious increases in the production of -cyclodextrin, indicating they were much more suitable for the industrial production of -cyclodextrin than the wild-type enzyme. The enhancement of -cyclodextrin specificity for the mutants might be due to the stability of CaIII by charged amino acid substitutions. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Cyclodextrin glycosyltransferase (CGTase, EC 2.4.1.19) belongs to glycosyl hydrolases family 13, which includes many enzymes that catalyze transglycosylation using similar ␣-retaining mechanisms [1–4]. CGTases act on oligomers or polymers of glucose, catalyzing four reactions: cyclization, disproportionation, coupling and hydrolysis [5]. The primary industrial application of CGTases is the production of cyclodextrins from starch via the cyclization reaction [6,7]. Cyclodextrins are cyclic ␣(1–4) linked oligomers of glucose that normally consist of 6, 7, or 8 glucose residues (␣-, , ␥-cyclodextrin, respectively). A nonpolar central cavity enables cyclodextrins to form inclusion complexes with hydrophobic guest molecules [8–10], leading to a variety of applications in the food, pharmaceutical, and cosmetics industries, as well as many others [11–14].
∗ Corresponding author at: School of Food Science and Technology, Jiangnan University, Wuxi 214122, China. Tel.: +86 510 85329237; fax: +86 510 85329237. E-mail address: zfl[email protected] (Z. Li). http://dx.doi.org/10.1016/j.ijbiomac.2015.02.036 0141-8130/© 2015 Elsevier B.V. All rights reserved.
A major disadvantage of CGTases for the industrial production of cyclodextrins is that the reaction produces a mixture of ␣-, - and ␥-cyclodextrins [15,16]. Two types of cyclodextrin production process can be used. The non-solvent process produces cyclodextrin mixtures that can be further separated by chromatographic procedures or selective precipitation with organic complexing agents, while the solvent process requires complexing agents to selectively extract one type of cyclodextrin and thus directs the enzymatic reaction to produce the cyclodextrin of interest [8,14]. Therefore, mutants with improved cyclodextrin product specificity are of high industrial interest, which could reduce the use of organic solvents for selective complexation of the cyclodextrin of interest, lowering the costs of cyclodextrin production [8,17,18]. Most CGTases, like other ␣-amylases, have two or more calcium binding sites [15,19,20]. Studies have suggested that the calcium binding sites in CGTases contribute to their thermostability and product specificity [15,21–23]. The enzyme used in our studies, the -CGTase from Bacillus circulans STB01, possesses three calcium binding sites called CaI, CaII and CaIII [15]. Our earlier work demonstrated that mutations at Ala31, which is near CaI, enhanced the -cyclodextrin specificity of this CGTase, which may be a result of stabilizing CaI [22]. Since Ala315, which is near CaIII, might play an important role
X. Ban et al. / International Journal of Biological Macromolecules 76 (2015) 224–229
in stabilizing CaIII, we speculated that the identity of the amino acid residue at 315 at CaIII might also be related to the product specificity of this -CGTase. In the present study, site-directed mutagenesis was performed on Ala315 of the -CGTase from B. circulans STB01 to investigate the effects of different amino acid side chains at this position on product specificity of CGTase. 2. Materials and methods 2.1. Bacterial strains and plasmids Escherichia coli JM109 [F (traD36, proAB+ lacIq , (lacZ)M15) endA1 recA1 hsdR17(rk − , mk + ) mcrA supE44 − gyrA96 relA1 (lacproAB) thi-1] [24] was used for recombinant DNA manipulations. The cgt gene (GenBank accession number: KJ660983) encoding the wild-type -CGTase from B. circulans STB01 was previously used to construct plasmid cgt/pST [15]. This plasmid was used for site-directed mutagenesis, sequencing, and expression of CGTase proteins. The (mutant) CGTases were produced with Bacillus subtilis WB600 [trpC2 nprE aprE epr bpr mpr nprB; Emr ] [25] harboring (mutant) plasmid cgt/pST. 2.2. Site-directed mutagenesis Prime STAR HS DNA polymerase and PCR reagents were purchased from TaKaRa Shuzo (Otsu, Japan) and used according to the manufacturer’s instructions. Site-directed mutagenesis was performed using a one-step PCR method with plasmid cgt/pST as the template. The primers used for mutagenesis are shown in Table 1. DNA sequences were confirmed by sequencing with an ABI PRISM BigDye primer cycle sequencing kit and AmpliTaq DNA polymerase (Perkin-Elmer, Foster City, CA, USA). The resulting (mutant) plasmids were transformed into B. subtilis WB600. 2.3. Production and purification of CGTase proteins A single colony of B. subtilis WB600 harboring (mutant) plasmid cgt/pST was inoculated into 50 mL of Luria-Bertani medium containing 5 g/mL of kanamycin and 10 g/mL of erythromycin and grown at 37 ◦ C overnight. The overnight culture (2 mL) was then diluted into 50 mL of terrific broth contained 5 g/mL of kanamycin and 10 g/mL of erythromycin and incubated on a rotary shaker (200 rpm/min) at 37 ◦ C for 48 h. For CGTase purification, cells were removed by centrifugation at 10,000 × g for 20 min. The cell-free supernatant was loaded onto a Q-Sepharose anionexchange column (Amersham Biosciences, Piscataway, NJ, USA) and then the eluted CGTase was further purified using a phenylSuperose column (HR10/10) (Amersham Biosciences, Piscataway, NJ, USA). The active fractions were pooled and concentrated by ultrafiltration (Amicon, Millipore, Billerica, MA, USA), and then dialyzed against 10 mM phosphate buffer (pH 6.5) at 4 ◦ C for 48 h [15]. Table 1 Mutant primers used for site directed mutagenesis. Primer
Sequence from 5 to 3 directiona
A315R-For A315R-Rev A315D-For A315D-Rev A315T-For A315T-Rev A315V-For A315V-Rev A315I-For A315I-Rev
GCAGCCGATTACCGCCAGGTGGATG GTCATCCACCTGGCGGTAATCGGCTG GCAGCCGATTACGACCAGGTGGATG GTCATCCACCTGGTCGTAATCGGCTG GCAGCCGATTACACCCAGGTGGATG GTCATCCACCTGGGTGTAATCGGCTG GCAGCCGATTACGTCCAGGTGGATG GTCATCCACCTGGACGTAATCGGCTG GCAGCCGATTACATCCAGGTGGATG GTCATCCACCTGGATGTAATCGGCTG
a
Underlined sequences present introduction of mutants.
225
The purified enzymes were confirmed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE, 10% gel) and stored at −80 ◦ C. 2.4. CGTase assays All assays were conducted by incubating 0.1 mL of appropriately diluted CGTase with 0.9 mL of 1% (w/v) maltodextrin (DE = 5, ROQUETTE frères, Lestrem, FR) in 10 mM phosphate buffer (pH 6.5) at 50 ◦ C for 10 min. The ␣-, -, and ␥-cyclodextrin formation activities were determined using the methyl orange [26], phenolphthalein [27], and bromocresol green methods [28], respectively. One unit of each activity was defined as the amount of enzyme that produced 1 mol of the corresponding cyclodextrin per minute. 2.5. HPLC product analysis The formation of cyclodextrins under conditions resembling an industrial production process was measured by incubating 1 unit/mL CGTase (total cyclization activity) with 5% (wet basis, w/v) maltodextrin solution in 10 mM phosphate buffer (pH 6.5) at 50 ◦ C for 10 min. At regular time intervals, samples were taken and boiled for 10 min to terminate the reaction. Glucoamylase was added to the boiled sample at a final concentration of 1 unit/mL, and then the mixture was incubated at 30 ◦ C for 2 h, followed by boiling for 10 min. The concentrations of ␣-, -, and ␥-cyclodextrin in the final sample were determined by HPLC on a Lichrosorb NH2 column (Merck, Darmstadt, FRG) eluted with acetonitrile/water (65:45) at a flow rate of 1 mL/min. 2.6. Structure modeling The 3D homology model of the CGTases was generated using SWISS-MODEL protein-modeling server (http://www.expasy.ch/ swissmod/SWISS-MODEL.html) [29]. The proposed complex structures of the wild-type and mutant CGTases were performed using the crystal structure of CGTase from B. circulans strain 251 (PDB accession code 1EO5) [30]. The graphical presentations and illustrative figures presented in this report were prepared using SPDBV 4.04 software. 2.7. Thermostability determination The thermostabilities of the wild-type and mutant CGTases were described by half-lives of their activities. The purified enzymes in 10 mM phosphate buffer (pH 6.5) were incubated at 60 ◦ C and taken at several time intervals. The residual -cyclodextrin formation activity was determined using the phenolphthalein method [27]. 3. Results 3.1. Site-directed mutagenesis at Ala315 To test the influence of mutations at Ala315 of the CGTase from B. circulans STB01 on its cyclodextrin product specificity, mutants A315R, A315D, A315T, A315V and A315I were prepared using one-step PCR. The sequences of all of the mutant plasmids were confirmed by DNA sequencing. The wild-type and mutant CGTase proteins were expressed in B. subtilis WB600 and secreted into the culture medium. There were no significant differences in expression level between the wild-type and mutant CGTases. The CGTases were purified using a combination of anion exchange and hydrophobic chromatographies. The purity and molecular weights of the wild-type and mutant CGTase proteins were verified using
226
X. Ban et al. / International Journal of Biological Macromolecules 76 (2015) 224–229
Table 2 Cyclization activities of the wild-type and mutant CGTases from B. circulans STB01.A Mutant (protein) Wild-type A315R A315D A315T A315V A315I
␣-Cyclization (U/mg) 69.4 9.9 20.3 69.3 62.4 64.6
± ± ± ± ± ±
e
1.1 0.3a 0.4b 0.6e 0.7c 0.9d
-Cyclization (U/mg) 159.9 178.9 176.0 152.9 150.9 154.5
± ± ± ± ± ±
b
2.5 2.7c 2.2c 2.1a 2.0a 2.4a
␥-Cyclization (U/mg) 38.1 78.7 65.4 35.2 34.0 36.8
± ± ± ± ± ±
c
1.0 1.3e 1.1d 0.7ab 0.4a 0.9bc
Total (U/mg) c
267.4 267.5c 261.7bc 257.4b 247.3a 255.9b
Half-life t1/2 (min, 60 ◦ C) 6.4 7.8 7.3 6.7 5.5 5.2
± ± ± ± ± ±
0.2b 0.1d 0.2c 0.1b 0.2a 0.2a
A Each number represents the mean of three independent measurements. Means with different superscript letters within the same column are significantly different (p < 0.05).
SDS-PAGE. The purified enzymes migrated as a single band on SDSPAGE with a molecular mass of approximately 77 kDa (Fig. 1).
production processes, are presented in Fig. 2. The data for the final time point are shown in Table 3. At the final time point, each of the enzymes had converted between 29.6 and 35.2% of the starch to cyclodextrins (Table 3). The wild-type, A315R and A315D displayed essentially the same conversion, approximately 35%, while the other mutants converted less of the starch to cyclodextrins. For all CGTases, cyclodextrin was the main product while ␣- and ␥-cyclodextrins were present at relatively low levels (Fig. 2). At the final time point, the cyclodextrin content in the reaction catalyzed by wild-type CGTase comprised 24.2% ␣-cyclodextrin, 18% ␥-cyclodextrin, and 57.8% -cyclodextrin. The A315R and A315D mutants produced a greater proportion of -cyclodextrin, compared with the wild-type enzyme, at the expense of ␣-cyclodextrin. After 9 h of incubation, the proportion of -cyclodextrin in the reactions catalyzed by mutants A315R and A315D were 15.1% and 10.8% greater than that in the reaction catalyzed by the wild-type enzyme, respectively. The proportion of ␣-cyclodextrin in the reactions catalyzed by mutants A315R and A315D decreased by 56.4% and 38.5%, respectively, compared with that the reaction catalyzed by the wild-type enzyme. In these same reactions, the proportion of ␥-cyclodextrin in the total cyclodextrins did not vary significantly (Table 3).
3.2. Cyclization activities of the wild-type and mutant CGTases
3.3. Structural modeling
The cyclization activities of the wild-type and mutants enzyme are shown in Table 2. The total cyclization activities of the wildtype and mutant CGTases fell within a narrow range, with the least active mutant (A315V) displaying approximately 92% of the total cyclization activity of the most active enzymes (wild-type and A315R). The ␣- and ␥-cyclodextrin forming activities of the wildtype CGTase constituted approximately 40% of the total cyclization activity, while the -cyclodextrin forming activity constituted 60% of the total activity. Although the -cyclodextrin forming activity and its proportion in the total activity were higher than those of many CGTases from other sources as shown in the previous reports [31,32], mutants of this CGTase with higher -cyclodextrin forming activity are of high interest for the production of -cyclodextrin. Compared with the wild-type enzyme, mutants A315R and A315D displayed 85.8% and 70.7% reductions in ␣-cyclodextrinforming activity, respectively. However, they also displayed 107% and 76.1% increases in ␥-cyclodextrin forming activity. The sum of ␣- and ␥-cyclization activities of A315R and A315D were 33.1% and 32.7% of their total cyclization activities, respectively. Thus, the mutants A315R and A315D displayed 12% and 10% increases in -cyclization activity, compared with the wild-type enzyme, respectively. In contrast, the ␣-, -, and ␥-cyclization activities of the mutants A315T, A315V and A315I were almost indistinguishable from those of the wild-type CGTase under these conditions.
Theoretical crystal structure of the wild-type CGTase with three calcium binding sites is shown in Fig. 3. Depictions of the areas in the vicinity of the mutations, taken from 3D homology models of the structures of the wild-type and mutant CGTases, are presented in Fig. 4. The ionic replacement A315R gained interactions between the positively charged side chain at 315 and both Asn274 and Asn578. The ionic replacement A315D and the hydrophilic replacement A315T pick up interactions between the side chain of the residue at 315 and the residue Asn274. In contrast, the hydrophobic replacements A315L and A315I do not gain additional interactions between the side chain of the residue at 315 and neighboring residues.
Fig. 1. SDS-PAGE analysis of purified wild-type and mutant CGTases. (A) Crude wildtype CGTase; MK, Molecular weight markers; (B) purified wild-type CGTase; (C) purified mutant A315R; (D) purified mutant A315D; (E) purified mutant A315T; (F) purified mutant A315V and (G) purified mutant A315I.
3.2.1. Starch conversions of the wild-type and mutant CGTases under conditions resembling industrial production processes The time courses of cyclodextrin production by wild-type and mutant CGTases, under conditions resembling industrial
4. Discussion It was known that the -CGTase from B. circulans STB01 has three calcium binding sites [15]. Calcium binding sites CaI and CaII are located near the active center, while CaIII is relatively distant from the active center (Fig. 3). CaIII is composed of amino acid residues Ala315 and Asp577 (Fig. 4). In recent work, we showed that the mutations A315D and D577K increased the thermostability of the -CGTase from B. circulans STB01 [15]. The present study attempts to evaluate the effects of different side chains of amino acid residue at 315 on cyclodextrin product specificity of CGTase. In this study, substitution of Ala315 by charged residues Arg and Asp increased - and ␥-cyclodextrin-forming activities at the expense of ␣-cyclodextrin-forming activity. However, CaIII is relatively far away from active center of CGTase, suggesting that the mutations might affect CGTase activity indirectly. To the best of our
X. Ban et al. / International Journal of Biological Macromolecules 76 (2015) 224–229
Cyclodextrin prod. (g/L)
8
4
0
0
12 Cyclodextrin prod. (g/L)
12
A
10 2
300 300 20 4
30 6 Time (h)
40 8
6 4 2 0
6 4 2 0
0
2
12
8
0
8
2
4
6
8
6 4 2 0
Cyclodextrin prod. (g/L)
Cyclodextrin prod. (g/L)
6 4 2 0
2
4 6 Time (h)
10
D
0
12
8
0
8
8
10
E
10
6
10
2
4 6 Time (h)
8
10
2
4
8
10
Time (h)
12
4
Time (h)
C
10
B
10
50 10
Cyclodextrin prod. (g/L)
Cyclodextrin prod. (g/L)
12
227
8
F
10 8 6 4 2 0
10
0
6
Time (h)
Fig. 2. Cyclodextrins formed during incubation of a (mutant) CGTase from B. circulans strain STB01 (1 unit/mL total cyclization activity) with 5% (w/v, wet basis) maltodextrin (DE 3) solution at pH 6.0 and 50 ◦ C for 9 h. (A) Wild-type CGTase; (B) mutant A315R; (C) mutant A351D; (D) mutant A315T; (E) mutant A315V and (F) mutant A315I. , ␣-cyclodextrin; 䊉, -cyclodextrin; , ␥-cyclodextrin. Each value represents the mean of three independent measurements.
knowledge, calcium-binding sites are important for the structural integrity and activity of enzymes. In some case, calcium-binding sites may shift the thermostability of the enzyme, while in others they may affect the catalytic activity [33]. The 3D structural models of the wild-type and mutant CGTases suggest that these replacements form additional interactions (compared with alanine) between the side chains and residues Asn274 and Asn578 (Fig. 4). These additional interactions apparently allow a change in conformation that may strengthen the conformational rigidity of CaIII and modify the interaction between domains A and D. The results suggest that this alteration in structure or flexibility shifts the cyclization activity toward the production of larger
cyclodextrins, increasing the preference for - and ␥-cyclodextrins. In contrast, the substitution of Ala315 by valine and isoleucine did not show distinct differences in the ratio of cyclodextrin products, compared with wild-type CGTase. These larger, hydrophobic replacements apparently adopt conformations that are not significantly different than that of the wild-type enzyme. As seen in Fig. 4, the additional hydrophobic atoms of the valine and isoleucine (compared with alanine) are oriented toward Asn578 and away from Asn274. Interestingly, replacement of Ala315 with the hydrophilic residue threonine also failed to alter the cyclodextrin product specificity of this enzyme. These data reinforce the idea that a charged residue at position 315 is necessary to form the
Table 3 Starch conversions of the wild-type and mutant CGTases from B. circulans STB01.A Mutant (protein)
Conversion of starch into cyclodextrins (%)
Wild-type A315R A315D A315T A315V A315I
35.0 35.2 34.8 29.8 29.6 32.2
Products (mg/mL) ␣ 3.9 1.7 2.4 3.2 3.4 3.2
 ± ± ± ± ± ±
0.2d (24.2) 0.1a (10.5) 0.1b (15.0) 0.1c (23.4) 0.2c (25.0) 0.1c (21.6)
9.3 10.7 10.3 8.0 7.8 8.8
␥ ± ± ± ± ± ±
0.1c (57.8) 0.2e (66.0) 0.1d (64.4) 0.1a (58.4) 0.1a (57.4) 0.2b (59.5)
2.9 3.8 3.3 2.5 2.4 2.8
± ± ± ± ± ±
0.1b (18.0) 0.2d (18.4) 0.1c (20.6) 0.1a (18.2) 0.2a (17.6) 0.1b (18.9)
A CGTase proteins (1 units/mL total cyclization activity) were incubated with 5% (w/v, wet basis) maltodextrin (DE 5) solution at pH 6.0 and 50 ◦ C for 9 h. Each value represents the mean of three independent measurements. Means with different superscript letters within the same column are significantly different (p < 0.05). Numbers between brackets indicate the percentage of each cyclodextrin present in the product mixture.
228
X. Ban et al. / International Journal of Biological Macromolecules 76 (2015) 224–229
Fig. 3. Crystal structure modeling of the wild-type CGTase with three calcium binding sites. The calcium binding sites are shown as purple spheres. Main amino acid residues in the active site are rendered as spheres; carbon is rendered in yellow, oxygen in red, and nitrogen in blue. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
alternate conformations more likely to produce - and ␥cyclodextrins. It appears that relatively rigid CGTase conformations promote the production of larger cyclodextrins, rather than small cyclodextrins. These conformations may involve interaction of the side chain at residue 315 with nearby Asn 274 and Asn 578 and influence the geometry or stability of the calcium binding site (Fig. 4). One reasonable explanation these phenomena is that appropriate mutations at CaIII stabilize CGTase and lead to more rigid conformations that promote the formation of -cyclodextrins by this -CGTase. The evidence that appropriate mutation at CaIII stabilize CGTase is that the half-lives of the activities of mutants A315R and A315D increase, compared with that of the wild-type enzyme (Table 2). However, the threonine mutant does not noticeably increase or decrease the ratio of specific cyclodextrin, even though it forms an additional interaction between residue 315 and Asn274. Perhaps the additional hydrogen bond formed by this threonine mutant does not significantly strengthen or rigidify the conformation. This inability to precisely predict the effect of modification on CaIII hints that changing the product specificity of a CGTase may require a mutagenesis strategy that involves a more comprehensive range of residues. Another rational explanation may lie in the comprehensive effects of these replacements on transglycosylation. In the wildtype enzyme, the cyclodextrin-forming reaction is predominant, while other side reactions (disproportionation, coupling and hydrolysis) take place simultaneously and may become more
Fig. 4. Theoretical 3D conformation of the (mutant) CGTases at the calcium binding site III. (A) Wild-type CGTase; (B) mutant A315R; (C) mutant A315D; (D) mutant A315T; (E) mutant A315V and (F) mutant A315I. The calcium ion of calcium binding site III was shown as a large blue sphere. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
X. Ban et al. / International Journal of Biological Macromolecules 76 (2015) 224–229
noticeable with longer incubation times. In this study, the mutations may accelerate side reactions, which might have an impact on production of cyclodextrin. Thus, the changes in proportion of cyclodextrin may also result from alteration of product inhibition or coupling activities. 5. Conclusion In conclusion, residue 315, which belongs to the calcium binding site (CaIII), plays a crucial role in cyclodextrin product specificity. The mutations introducing a charged residue increased - and ␥-cyclodextrin-forming activities, while those with hydrophobic or hydrophilic groups did not noticeably change the proportion of cyclodextrins. Furthermore, when the mutant enzymes were incubated with maltodextrin under conditions resembling industrial production processes, the results revealed that the A315R and A315D display increase in -cyclodextrin specificity. The enhancement of -cyclodextrin specificity might be a result of stabilizing CaIII. Thus, the CaIII, although distant from the active center of CGTase, may play an important role in cyclodextrin product specificity. The mutations A315R and A315D may enhance the industrial utility of the CGTase from B. circulans STB01. Although our understanding of the mechanism by which the calcium binding sites of CGTase influence stability and function is incomplete, the results presented here provide new insight into the function of calcium binding sites. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (no. 31101228), the Natural Science Foundation of Jiangsu Province (BK2011152), and the Fok YingTong Education Foundation, China (no. 131069). References [1] H. Leemhuis, B.W. Dijkstra, L. Dijkhuizen, FEBS Lett. 514 (2002) 189–192. [2] J.D. Schoffer, M.P. Klein, R.C. Rodrigues, P.F. Hertz, Carbohydr. Polym. 98 (2013) 1311–1316.
229
[3] L. Wang, D. Wu, J. Chen, J. Wu, Food Chem. 141 (2013) 3072–3076. [4] Z.F. Li, J.Y. Zhang, M. Wang, Z.B. Gu, G.C. Du, J.K. Li, J. Wu, J. Chen, Appl. Microbiol. Biotechnol. 83 (2009) 483–490. [5] B.A. van der Veen, H. Leemhuis, S. Kralj, J.C. Uitdehaag, B.W. Dijkstra, L. Dijkhuizen, J. Biol. Chem. 276 (2001) 44557–44562. [6] R.Z. Han, L. Liu, H.D. Shin, R.R. Chen, G.C. Du, J. Chen, Appl. Microbiol. Biotechnol. 97 (2013) 5851–5860. [7] B.A. van der Veen, J.C. Uitdehaag, B.W. Dijkstra, L. Dijkhuizen, BBA Protein Struct. Mol. Enzymol. 1543 (2000) 336–360. [8] Z.F. Li, M. Wang, F. Wang, Z.B. Gu, G.C. Du, J. Wu, J. Chen, Appl. Microbiol. Biotechnol. 77 (2007) 245–255. [9] H. Leemhuis, R.M. Kelly, L. Dijkhuizen, Appl. Microbiol. Biotechnol. 85 (2010) 823–835. [10] K. Hirano, T. Ishihara, S. Ogasawara, H. Maeda, K. Abe, T. Nakajima, Y. Yamagata, Appl. Microbiol. Biotechnol. 70 (2006) 193–201. [11] G. Astray, C. Gonzalez-Barreiro, J.C. Mejuto, R. Rial-Otero, J. Simal-Gandara, Food Hydrocoll. 23 (2009) 1631–1640. [12] E.M.M. Del Valle, Process Biochem. 39 (2004) 1033–1046. [13] A.L. Laza-Knoerr, R. Gref, P. Couvreur, J. Drug Target. 18 (2010) 645–656. [14] Z.F. Li, S. Chen, Z.B. Gu, J. Chen, J. Wu, Trends Food Sci. Technol. 35 (2014) 151–160. [15] C.M. Li, X.F. Ban, Z.B. Gu, Z.F. Li, J. Agric. Food Chem. 61 (2013) 8836–8841. [16] N. Szerman, I. Schroh, A.L. Rossi, A.M. Rosso, N. Krymkiewicz, S.A. Ferrarotti, Bioresour. Technol. 98 (2007) 2886–2891. [17] J.C.M. Uitdehaag, G.J.W.M. van Alebeek, B.A. van der Veen, L. Dijkhuizen, B.W. Dijkstra, Biochemistry 39 (2000) 7772–7780. [18] Q.S. Qi, W. Zimmermann, Appl. Microbiol. Biotechnol. 66 (2005) 475–485. [19] J.C.M. Uitdehaag, R. Mosi, K.H. Kalk, B.A. van der Veen, L. Dijkhuizen, S.G. Withers, B.W. Dijkstra, Nat. Struct. Biol. 6 (1999) 432–436. [20] V. Kumar, Carbohydr. Res. 345 (2010) 1564–1569. [21] P.H. Goh, R.M. Illias, K.M. Goh, Int. J. Mol. Sci. 13 (2012) 5307–5323. [22] Z.F. Li, X.F. Ban, Z.B. Gu, C.M. Li, M. Huang, Y. Hong, L. Cheng, Carbohydr. Polym. 108 (2014) 112–117. [23] N. Ishii, K. Haga, K. Yamane, K. Harata, J. Mol. Recognit. 13 (2000) 35–43. [24] C. Yanisch-Perron, J. Vieira, J. Messing, Gene 33 (1985) 103–119. [25] X.C. Wu, W. Lee, L. Tran, S.L. Wong, J. Bacteriol. 173 (1991) 4952–4958. [26] A. Lejeune, K. Sakaguchi, T. Imanaka, Anal. Biochem. 181 (1989) 6–11. [27] M. Makela, T. Korpela, S. Laakso, J. Biochem. Biophys. Methods 14 (1987) 85–92. [28] T. Kato, K. Horikoshi, Anal. Chem. 56 (1984) 1738–1740. [29] T. Schwede, J. Kopp, N. Guex, M.C. Peitsch, Nucleic Acids Res. 31 (2003) 3381–3385. [30] J.C. Uitdehaag, G.J. van Alebeek, B.A. van Der Veen, L. Dijkhuizen, B.W. Dijkstra, Biochemistry 39 (2000) 7772–7780. [31] Z.F. Li, J.Y. Zhang, Q. Sun, M. Wang, Z.B. Gu, G.C. Du, J. Wu, J. Chen, J. Agric. Food Chem. 57 (2009) 8386–8391. [32] R.M. Kelly, L. Dijkhuizen, H. Leemhuis, Appl. Microbiol. Biotechnol. 84 (2009) 119–133. [33] N. Ishii, K. Haga, K. Yamane, K. Harata, J. Biochem. 127 (2000) 383–391.
Contents lists available at ScienceDirect
International Journal of Biological Macromolecules journal homepage: www.elsevier.com/locate/ijbiomac
Mutations at calcium binding site III in cyclodextrin glycosyltransferase improve -cyclodextrin specificity Xiaofeng Ban b , Zhengbiao Gu a,b,c , Caiming Li b , Min Huang b , Li Cheng a,b , Yan Hong a,b,c , Zhaofeng Li a,b,c,∗ a
State Key Laboratory of Food Science and Technology, Jiangnan University, Wuxi 214122, China School of Food Science and Technology, Jiangnan University, Wuxi 214122, China c Synergetic Innovation Center of Food Safety and Nutrition, Jiangnan University, Wuxi, Jiangsu 214122, China b
a r t i c l e
i n f o
Article history: Received 9 September 2014 Received in revised form 18 February 2015 Accepted 19 February 2015 Available online 4 March 2015 Keywords: Cyclodextrin glycosyltransferase -Cyclodextrin Residue 315 Calcium binding site Product specificity
a b s t r a c t Cyclodextrin glycosyltransferases (CGTases, EC 2.4.1.19) are industrially important enzymes that produce cyclodextrins from starch by intramolecular transglycosylation. In this study, the effects of amino acid residue at position 315 in calcium binding site III (CaIII) on product specificity of CGTase were investigated by replacing Ala315 in the CGTase from Bacillus circulans STB01 with arginine, aspartic acid, threonine, leucine and valine. The cgt gene, which encodes this enzyme, was expressed in B. subtilis WB600 alongside site-directed mutants A315R, A315D, A315T, A315L and A315V. The results showed that CaIII plays an important role in cyclodextrin product specificity. Replacement of Ala315 by charged amino acid residues enhanced -cyclodextrin specificity, compared with the wild-type CGTase. Mutations A315R and A315D resulted in an approximately 10% increase in -cyclodextrin activity. Furthermore, under conditions resembling the industrial production processes, the mutants A315R and A315D displayed obvious increases in the production of -cyclodextrin, indicating they were much more suitable for the industrial production of -cyclodextrin than the wild-type enzyme. The enhancement of -cyclodextrin specificity for the mutants might be due to the stability of CaIII by charged amino acid substitutions. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Cyclodextrin glycosyltransferase (CGTase, EC 2.4.1.19) belongs to glycosyl hydrolases family 13, which includes many enzymes that catalyze transglycosylation using similar ␣-retaining mechanisms [1–4]. CGTases act on oligomers or polymers of glucose, catalyzing four reactions: cyclization, disproportionation, coupling and hydrolysis [5]. The primary industrial application of CGTases is the production of cyclodextrins from starch via the cyclization reaction [6,7]. Cyclodextrins are cyclic ␣(1–4) linked oligomers of glucose that normally consist of 6, 7, or 8 glucose residues (␣-, , ␥-cyclodextrin, respectively). A nonpolar central cavity enables cyclodextrins to form inclusion complexes with hydrophobic guest molecules [8–10], leading to a variety of applications in the food, pharmaceutical, and cosmetics industries, as well as many others [11–14].
∗ Corresponding author at: School of Food Science and Technology, Jiangnan University, Wuxi 214122, China. Tel.: +86 510 85329237; fax: +86 510 85329237. E-mail address: zfl[email protected] (Z. Li). http://dx.doi.org/10.1016/j.ijbiomac.2015.02.036 0141-8130/© 2015 Elsevier B.V. All rights reserved.
A major disadvantage of CGTases for the industrial production of cyclodextrins is that the reaction produces a mixture of ␣-, - and ␥-cyclodextrins [15,16]. Two types of cyclodextrin production process can be used. The non-solvent process produces cyclodextrin mixtures that can be further separated by chromatographic procedures or selective precipitation with organic complexing agents, while the solvent process requires complexing agents to selectively extract one type of cyclodextrin and thus directs the enzymatic reaction to produce the cyclodextrin of interest [8,14]. Therefore, mutants with improved cyclodextrin product specificity are of high industrial interest, which could reduce the use of organic solvents for selective complexation of the cyclodextrin of interest, lowering the costs of cyclodextrin production [8,17,18]. Most CGTases, like other ␣-amylases, have two or more calcium binding sites [15,19,20]. Studies have suggested that the calcium binding sites in CGTases contribute to their thermostability and product specificity [15,21–23]. The enzyme used in our studies, the -CGTase from Bacillus circulans STB01, possesses three calcium binding sites called CaI, CaII and CaIII [15]. Our earlier work demonstrated that mutations at Ala31, which is near CaI, enhanced the -cyclodextrin specificity of this CGTase, which may be a result of stabilizing CaI [22]. Since Ala315, which is near CaIII, might play an important role
X. Ban et al. / International Journal of Biological Macromolecules 76 (2015) 224–229
in stabilizing CaIII, we speculated that the identity of the amino acid residue at 315 at CaIII might also be related to the product specificity of this -CGTase. In the present study, site-directed mutagenesis was performed on Ala315 of the -CGTase from B. circulans STB01 to investigate the effects of different amino acid side chains at this position on product specificity of CGTase. 2. Materials and methods 2.1. Bacterial strains and plasmids Escherichia coli JM109 [F (traD36, proAB+ lacIq , (lacZ)M15) endA1 recA1 hsdR17(rk − , mk + ) mcrA supE44 − gyrA96 relA1 (lacproAB) thi-1] [24] was used for recombinant DNA manipulations. The cgt gene (GenBank accession number: KJ660983) encoding the wild-type -CGTase from B. circulans STB01 was previously used to construct plasmid cgt/pST [15]. This plasmid was used for site-directed mutagenesis, sequencing, and expression of CGTase proteins. The (mutant) CGTases were produced with Bacillus subtilis WB600 [trpC2 nprE aprE epr bpr mpr nprB; Emr ] [25] harboring (mutant) plasmid cgt/pST. 2.2. Site-directed mutagenesis Prime STAR HS DNA polymerase and PCR reagents were purchased from TaKaRa Shuzo (Otsu, Japan) and used according to the manufacturer’s instructions. Site-directed mutagenesis was performed using a one-step PCR method with plasmid cgt/pST as the template. The primers used for mutagenesis are shown in Table 1. DNA sequences were confirmed by sequencing with an ABI PRISM BigDye primer cycle sequencing kit and AmpliTaq DNA polymerase (Perkin-Elmer, Foster City, CA, USA). The resulting (mutant) plasmids were transformed into B. subtilis WB600. 2.3. Production and purification of CGTase proteins A single colony of B. subtilis WB600 harboring (mutant) plasmid cgt/pST was inoculated into 50 mL of Luria-Bertani medium containing 5 g/mL of kanamycin and 10 g/mL of erythromycin and grown at 37 ◦ C overnight. The overnight culture (2 mL) was then diluted into 50 mL of terrific broth contained 5 g/mL of kanamycin and 10 g/mL of erythromycin and incubated on a rotary shaker (200 rpm/min) at 37 ◦ C for 48 h. For CGTase purification, cells were removed by centrifugation at 10,000 × g for 20 min. The cell-free supernatant was loaded onto a Q-Sepharose anionexchange column (Amersham Biosciences, Piscataway, NJ, USA) and then the eluted CGTase was further purified using a phenylSuperose column (HR10/10) (Amersham Biosciences, Piscataway, NJ, USA). The active fractions were pooled and concentrated by ultrafiltration (Amicon, Millipore, Billerica, MA, USA), and then dialyzed against 10 mM phosphate buffer (pH 6.5) at 4 ◦ C for 48 h [15]. Table 1 Mutant primers used for site directed mutagenesis. Primer
Sequence from 5 to 3 directiona
A315R-For A315R-Rev A315D-For A315D-Rev A315T-For A315T-Rev A315V-For A315V-Rev A315I-For A315I-Rev
GCAGCCGATTACCGCCAGGTGGATG GTCATCCACCTGGCGGTAATCGGCTG GCAGCCGATTACGACCAGGTGGATG GTCATCCACCTGGTCGTAATCGGCTG GCAGCCGATTACACCCAGGTGGATG GTCATCCACCTGGGTGTAATCGGCTG GCAGCCGATTACGTCCAGGTGGATG GTCATCCACCTGGACGTAATCGGCTG GCAGCCGATTACATCCAGGTGGATG GTCATCCACCTGGATGTAATCGGCTG
a
Underlined sequences present introduction of mutants.
225
The purified enzymes were confirmed by sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE, 10% gel) and stored at −80 ◦ C. 2.4. CGTase assays All assays were conducted by incubating 0.1 mL of appropriately diluted CGTase with 0.9 mL of 1% (w/v) maltodextrin (DE = 5, ROQUETTE frères, Lestrem, FR) in 10 mM phosphate buffer (pH 6.5) at 50 ◦ C for 10 min. The ␣-, -, and ␥-cyclodextrin formation activities were determined using the methyl orange [26], phenolphthalein [27], and bromocresol green methods [28], respectively. One unit of each activity was defined as the amount of enzyme that produced 1 mol of the corresponding cyclodextrin per minute. 2.5. HPLC product analysis The formation of cyclodextrins under conditions resembling an industrial production process was measured by incubating 1 unit/mL CGTase (total cyclization activity) with 5% (wet basis, w/v) maltodextrin solution in 10 mM phosphate buffer (pH 6.5) at 50 ◦ C for 10 min. At regular time intervals, samples were taken and boiled for 10 min to terminate the reaction. Glucoamylase was added to the boiled sample at a final concentration of 1 unit/mL, and then the mixture was incubated at 30 ◦ C for 2 h, followed by boiling for 10 min. The concentrations of ␣-, -, and ␥-cyclodextrin in the final sample were determined by HPLC on a Lichrosorb NH2 column (Merck, Darmstadt, FRG) eluted with acetonitrile/water (65:45) at a flow rate of 1 mL/min. 2.6. Structure modeling The 3D homology model of the CGTases was generated using SWISS-MODEL protein-modeling server (http://www.expasy.ch/ swissmod/SWISS-MODEL.html) [29]. The proposed complex structures of the wild-type and mutant CGTases were performed using the crystal structure of CGTase from B. circulans strain 251 (PDB accession code 1EO5) [30]. The graphical presentations and illustrative figures presented in this report were prepared using SPDBV 4.04 software. 2.7. Thermostability determination The thermostabilities of the wild-type and mutant CGTases were described by half-lives of their activities. The purified enzymes in 10 mM phosphate buffer (pH 6.5) were incubated at 60 ◦ C and taken at several time intervals. The residual -cyclodextrin formation activity was determined using the phenolphthalein method [27]. 3. Results 3.1. Site-directed mutagenesis at Ala315 To test the influence of mutations at Ala315 of the CGTase from B. circulans STB01 on its cyclodextrin product specificity, mutants A315R, A315D, A315T, A315V and A315I were prepared using one-step PCR. The sequences of all of the mutant plasmids were confirmed by DNA sequencing. The wild-type and mutant CGTase proteins were expressed in B. subtilis WB600 and secreted into the culture medium. There were no significant differences in expression level between the wild-type and mutant CGTases. The CGTases were purified using a combination of anion exchange and hydrophobic chromatographies. The purity and molecular weights of the wild-type and mutant CGTase proteins were verified using
226
X. Ban et al. / International Journal of Biological Macromolecules 76 (2015) 224–229
Table 2 Cyclization activities of the wild-type and mutant CGTases from B. circulans STB01.A Mutant (protein) Wild-type A315R A315D A315T A315V A315I
␣-Cyclization (U/mg) 69.4 9.9 20.3 69.3 62.4 64.6
± ± ± ± ± ±
e
1.1 0.3a 0.4b 0.6e 0.7c 0.9d
-Cyclization (U/mg) 159.9 178.9 176.0 152.9 150.9 154.5
± ± ± ± ± ±
b
2.5 2.7c 2.2c 2.1a 2.0a 2.4a
␥-Cyclization (U/mg) 38.1 78.7 65.4 35.2 34.0 36.8
± ± ± ± ± ±
c
1.0 1.3e 1.1d 0.7ab 0.4a 0.9bc
Total (U/mg) c
267.4 267.5c 261.7bc 257.4b 247.3a 255.9b
Half-life t1/2 (min, 60 ◦ C) 6.4 7.8 7.3 6.7 5.5 5.2
± ± ± ± ± ±
0.2b 0.1d 0.2c 0.1b 0.2a 0.2a
A Each number represents the mean of three independent measurements. Means with different superscript letters within the same column are significantly different (p < 0.05).
SDS-PAGE. The purified enzymes migrated as a single band on SDSPAGE with a molecular mass of approximately 77 kDa (Fig. 1).
production processes, are presented in Fig. 2. The data for the final time point are shown in Table 3. At the final time point, each of the enzymes had converted between 29.6 and 35.2% of the starch to cyclodextrins (Table 3). The wild-type, A315R and A315D displayed essentially the same conversion, approximately 35%, while the other mutants converted less of the starch to cyclodextrins. For all CGTases, cyclodextrin was the main product while ␣- and ␥-cyclodextrins were present at relatively low levels (Fig. 2). At the final time point, the cyclodextrin content in the reaction catalyzed by wild-type CGTase comprised 24.2% ␣-cyclodextrin, 18% ␥-cyclodextrin, and 57.8% -cyclodextrin. The A315R and A315D mutants produced a greater proportion of -cyclodextrin, compared with the wild-type enzyme, at the expense of ␣-cyclodextrin. After 9 h of incubation, the proportion of -cyclodextrin in the reactions catalyzed by mutants A315R and A315D were 15.1% and 10.8% greater than that in the reaction catalyzed by the wild-type enzyme, respectively. The proportion of ␣-cyclodextrin in the reactions catalyzed by mutants A315R and A315D decreased by 56.4% and 38.5%, respectively, compared with that the reaction catalyzed by the wild-type enzyme. In these same reactions, the proportion of ␥-cyclodextrin in the total cyclodextrins did not vary significantly (Table 3).
3.2. Cyclization activities of the wild-type and mutant CGTases
3.3. Structural modeling
The cyclization activities of the wild-type and mutants enzyme are shown in Table 2. The total cyclization activities of the wildtype and mutant CGTases fell within a narrow range, with the least active mutant (A315V) displaying approximately 92% of the total cyclization activity of the most active enzymes (wild-type and A315R). The ␣- and ␥-cyclodextrin forming activities of the wildtype CGTase constituted approximately 40% of the total cyclization activity, while the -cyclodextrin forming activity constituted 60% of the total activity. Although the -cyclodextrin forming activity and its proportion in the total activity were higher than those of many CGTases from other sources as shown in the previous reports [31,32], mutants of this CGTase with higher -cyclodextrin forming activity are of high interest for the production of -cyclodextrin. Compared with the wild-type enzyme, mutants A315R and A315D displayed 85.8% and 70.7% reductions in ␣-cyclodextrinforming activity, respectively. However, they also displayed 107% and 76.1% increases in ␥-cyclodextrin forming activity. The sum of ␣- and ␥-cyclization activities of A315R and A315D were 33.1% and 32.7% of their total cyclization activities, respectively. Thus, the mutants A315R and A315D displayed 12% and 10% increases in -cyclization activity, compared with the wild-type enzyme, respectively. In contrast, the ␣-, -, and ␥-cyclization activities of the mutants A315T, A315V and A315I were almost indistinguishable from those of the wild-type CGTase under these conditions.
Theoretical crystal structure of the wild-type CGTase with three calcium binding sites is shown in Fig. 3. Depictions of the areas in the vicinity of the mutations, taken from 3D homology models of the structures of the wild-type and mutant CGTases, are presented in Fig. 4. The ionic replacement A315R gained interactions between the positively charged side chain at 315 and both Asn274 and Asn578. The ionic replacement A315D and the hydrophilic replacement A315T pick up interactions between the side chain of the residue at 315 and the residue Asn274. In contrast, the hydrophobic replacements A315L and A315I do not gain additional interactions between the side chain of the residue at 315 and neighboring residues.
Fig. 1. SDS-PAGE analysis of purified wild-type and mutant CGTases. (A) Crude wildtype CGTase; MK, Molecular weight markers; (B) purified wild-type CGTase; (C) purified mutant A315R; (D) purified mutant A315D; (E) purified mutant A315T; (F) purified mutant A315V and (G) purified mutant A315I.
3.2.1. Starch conversions of the wild-type and mutant CGTases under conditions resembling industrial production processes The time courses of cyclodextrin production by wild-type and mutant CGTases, under conditions resembling industrial
4. Discussion It was known that the -CGTase from B. circulans STB01 has three calcium binding sites [15]. Calcium binding sites CaI and CaII are located near the active center, while CaIII is relatively distant from the active center (Fig. 3). CaIII is composed of amino acid residues Ala315 and Asp577 (Fig. 4). In recent work, we showed that the mutations A315D and D577K increased the thermostability of the -CGTase from B. circulans STB01 [15]. The present study attempts to evaluate the effects of different side chains of amino acid residue at 315 on cyclodextrin product specificity of CGTase. In this study, substitution of Ala315 by charged residues Arg and Asp increased - and ␥-cyclodextrin-forming activities at the expense of ␣-cyclodextrin-forming activity. However, CaIII is relatively far away from active center of CGTase, suggesting that the mutations might affect CGTase activity indirectly. To the best of our
X. Ban et al. / International Journal of Biological Macromolecules 76 (2015) 224–229
Cyclodextrin prod. (g/L)
8
4
0
0
12 Cyclodextrin prod. (g/L)
12
A
10 2
300 300 20 4
30 6 Time (h)
40 8
6 4 2 0
6 4 2 0
0
2
12
8
0
8
2
4
6
8
6 4 2 0
Cyclodextrin prod. (g/L)
Cyclodextrin prod. (g/L)
6 4 2 0
2
4 6 Time (h)
10
D
0
12
8
0
8
8
10
E
10
6
10
2
4 6 Time (h)
8
10
2
4
8
10
Time (h)
12
4
Time (h)
C
10
B
10
50 10
Cyclodextrin prod. (g/L)
Cyclodextrin prod. (g/L)
12
227
8
F
10 8 6 4 2 0
10
0
6
Time (h)
Fig. 2. Cyclodextrins formed during incubation of a (mutant) CGTase from B. circulans strain STB01 (1 unit/mL total cyclization activity) with 5% (w/v, wet basis) maltodextrin (DE 3) solution at pH 6.0 and 50 ◦ C for 9 h. (A) Wild-type CGTase; (B) mutant A315R; (C) mutant A351D; (D) mutant A315T; (E) mutant A315V and (F) mutant A315I. , ␣-cyclodextrin; 䊉, -cyclodextrin; , ␥-cyclodextrin. Each value represents the mean of three independent measurements.
knowledge, calcium-binding sites are important for the structural integrity and activity of enzymes. In some case, calcium-binding sites may shift the thermostability of the enzyme, while in others they may affect the catalytic activity [33]. The 3D structural models of the wild-type and mutant CGTases suggest that these replacements form additional interactions (compared with alanine) between the side chains and residues Asn274 and Asn578 (Fig. 4). These additional interactions apparently allow a change in conformation that may strengthen the conformational rigidity of CaIII and modify the interaction between domains A and D. The results suggest that this alteration in structure or flexibility shifts the cyclization activity toward the production of larger
cyclodextrins, increasing the preference for - and ␥-cyclodextrins. In contrast, the substitution of Ala315 by valine and isoleucine did not show distinct differences in the ratio of cyclodextrin products, compared with wild-type CGTase. These larger, hydrophobic replacements apparently adopt conformations that are not significantly different than that of the wild-type enzyme. As seen in Fig. 4, the additional hydrophobic atoms of the valine and isoleucine (compared with alanine) are oriented toward Asn578 and away from Asn274. Interestingly, replacement of Ala315 with the hydrophilic residue threonine also failed to alter the cyclodextrin product specificity of this enzyme. These data reinforce the idea that a charged residue at position 315 is necessary to form the
Table 3 Starch conversions of the wild-type and mutant CGTases from B. circulans STB01.A Mutant (protein)
Conversion of starch into cyclodextrins (%)
Wild-type A315R A315D A315T A315V A315I
35.0 35.2 34.8 29.8 29.6 32.2
Products (mg/mL) ␣ 3.9 1.7 2.4 3.2 3.4 3.2
 ± ± ± ± ± ±
0.2d (24.2) 0.1a (10.5) 0.1b (15.0) 0.1c (23.4) 0.2c (25.0) 0.1c (21.6)
9.3 10.7 10.3 8.0 7.8 8.8
␥ ± ± ± ± ± ±
0.1c (57.8) 0.2e (66.0) 0.1d (64.4) 0.1a (58.4) 0.1a (57.4) 0.2b (59.5)
2.9 3.8 3.3 2.5 2.4 2.8
± ± ± ± ± ±
0.1b (18.0) 0.2d (18.4) 0.1c (20.6) 0.1a (18.2) 0.2a (17.6) 0.1b (18.9)
A CGTase proteins (1 units/mL total cyclization activity) were incubated with 5% (w/v, wet basis) maltodextrin (DE 5) solution at pH 6.0 and 50 ◦ C for 9 h. Each value represents the mean of three independent measurements. Means with different superscript letters within the same column are significantly different (p < 0.05). Numbers between brackets indicate the percentage of each cyclodextrin present in the product mixture.
228
X. Ban et al. / International Journal of Biological Macromolecules 76 (2015) 224–229
Fig. 3. Crystal structure modeling of the wild-type CGTase with three calcium binding sites. The calcium binding sites are shown as purple spheres. Main amino acid residues in the active site are rendered as spheres; carbon is rendered in yellow, oxygen in red, and nitrogen in blue. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
alternate conformations more likely to produce - and ␥cyclodextrins. It appears that relatively rigid CGTase conformations promote the production of larger cyclodextrins, rather than small cyclodextrins. These conformations may involve interaction of the side chain at residue 315 with nearby Asn 274 and Asn 578 and influence the geometry or stability of the calcium binding site (Fig. 4). One reasonable explanation these phenomena is that appropriate mutations at CaIII stabilize CGTase and lead to more rigid conformations that promote the formation of -cyclodextrins by this -CGTase. The evidence that appropriate mutation at CaIII stabilize CGTase is that the half-lives of the activities of mutants A315R and A315D increase, compared with that of the wild-type enzyme (Table 2). However, the threonine mutant does not noticeably increase or decrease the ratio of specific cyclodextrin, even though it forms an additional interaction between residue 315 and Asn274. Perhaps the additional hydrogen bond formed by this threonine mutant does not significantly strengthen or rigidify the conformation. This inability to precisely predict the effect of modification on CaIII hints that changing the product specificity of a CGTase may require a mutagenesis strategy that involves a more comprehensive range of residues. Another rational explanation may lie in the comprehensive effects of these replacements on transglycosylation. In the wildtype enzyme, the cyclodextrin-forming reaction is predominant, while other side reactions (disproportionation, coupling and hydrolysis) take place simultaneously and may become more
Fig. 4. Theoretical 3D conformation of the (mutant) CGTases at the calcium binding site III. (A) Wild-type CGTase; (B) mutant A315R; (C) mutant A315D; (D) mutant A315T; (E) mutant A315V and (F) mutant A315I. The calcium ion of calcium binding site III was shown as a large blue sphere. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
X. Ban et al. / International Journal of Biological Macromolecules 76 (2015) 224–229
noticeable with longer incubation times. In this study, the mutations may accelerate side reactions, which might have an impact on production of cyclodextrin. Thus, the changes in proportion of cyclodextrin may also result from alteration of product inhibition or coupling activities. 5. Conclusion In conclusion, residue 315, which belongs to the calcium binding site (CaIII), plays a crucial role in cyclodextrin product specificity. The mutations introducing a charged residue increased - and ␥-cyclodextrin-forming activities, while those with hydrophobic or hydrophilic groups did not noticeably change the proportion of cyclodextrins. Furthermore, when the mutant enzymes were incubated with maltodextrin under conditions resembling industrial production processes, the results revealed that the A315R and A315D display increase in -cyclodextrin specificity. The enhancement of -cyclodextrin specificity might be a result of stabilizing CaIII. Thus, the CaIII, although distant from the active center of CGTase, may play an important role in cyclodextrin product specificity. The mutations A315R and A315D may enhance the industrial utility of the CGTase from B. circulans STB01. Although our understanding of the mechanism by which the calcium binding sites of CGTase influence stability and function is incomplete, the results presented here provide new insight into the function of calcium binding sites. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (no. 31101228), the Natural Science Foundation of Jiangsu Province (BK2011152), and the Fok YingTong Education Foundation, China (no. 131069). References [1] H. Leemhuis, B.W. Dijkstra, L. Dijkhuizen, FEBS Lett. 514 (2002) 189–192. [2] J.D. Schoffer, M.P. Klein, R.C. Rodrigues, P.F. Hertz, Carbohydr. Polym. 98 (2013) 1311–1316.
229
[3] L. Wang, D. Wu, J. Chen, J. Wu, Food Chem. 141 (2013) 3072–3076. [4] Z.F. Li, J.Y. Zhang, M. Wang, Z.B. Gu, G.C. Du, J.K. Li, J. Wu, J. Chen, Appl. Microbiol. Biotechnol. 83 (2009) 483–490. [5] B.A. van der Veen, H. Leemhuis, S. Kralj, J.C. Uitdehaag, B.W. Dijkstra, L. Dijkhuizen, J. Biol. Chem. 276 (2001) 44557–44562. [6] R.Z. Han, L. Liu, H.D. Shin, R.R. Chen, G.C. Du, J. Chen, Appl. Microbiol. Biotechnol. 97 (2013) 5851–5860. [7] B.A. van der Veen, J.C. Uitdehaag, B.W. Dijkstra, L. Dijkhuizen, BBA Protein Struct. Mol. Enzymol. 1543 (2000) 336–360. [8] Z.F. Li, M. Wang, F. Wang, Z.B. Gu, G.C. Du, J. Wu, J. Chen, Appl. Microbiol. Biotechnol. 77 (2007) 245–255. [9] H. Leemhuis, R.M. Kelly, L. Dijkhuizen, Appl. Microbiol. Biotechnol. 85 (2010) 823–835. [10] K. Hirano, T. Ishihara, S. Ogasawara, H. Maeda, K. Abe, T. Nakajima, Y. Yamagata, Appl. Microbiol. Biotechnol. 70 (2006) 193–201. [11] G. Astray, C. Gonzalez-Barreiro, J.C. Mejuto, R. Rial-Otero, J. Simal-Gandara, Food Hydrocoll. 23 (2009) 1631–1640. [12] E.M.M. Del Valle, Process Biochem. 39 (2004) 1033–1046. [13] A.L. Laza-Knoerr, R. Gref, P. Couvreur, J. Drug Target. 18 (2010) 645–656. [14] Z.F. Li, S. Chen, Z.B. Gu, J. Chen, J. Wu, Trends Food Sci. Technol. 35 (2014) 151–160. [15] C.M. Li, X.F. Ban, Z.B. Gu, Z.F. Li, J. Agric. Food Chem. 61 (2013) 8836–8841. [16] N. Szerman, I. Schroh, A.L. Rossi, A.M. Rosso, N. Krymkiewicz, S.A. Ferrarotti, Bioresour. Technol. 98 (2007) 2886–2891. [17] J.C.M. Uitdehaag, G.J.W.M. van Alebeek, B.A. van der Veen, L. Dijkhuizen, B.W. Dijkstra, Biochemistry 39 (2000) 7772–7780. [18] Q.S. Qi, W. Zimmermann, Appl. Microbiol. Biotechnol. 66 (2005) 475–485. [19] J.C.M. Uitdehaag, R. Mosi, K.H. Kalk, B.A. van der Veen, L. Dijkhuizen, S.G. Withers, B.W. Dijkstra, Nat. Struct. Biol. 6 (1999) 432–436. [20] V. Kumar, Carbohydr. Res. 345 (2010) 1564–1569. [21] P.H. Goh, R.M. Illias, K.M. Goh, Int. J. Mol. Sci. 13 (2012) 5307–5323. [22] Z.F. Li, X.F. Ban, Z.B. Gu, C.M. Li, M. Huang, Y. Hong, L. Cheng, Carbohydr. Polym. 108 (2014) 112–117. [23] N. Ishii, K. Haga, K. Yamane, K. Harata, J. Mol. Recognit. 13 (2000) 35–43. [24] C. Yanisch-Perron, J. Vieira, J. Messing, Gene 33 (1985) 103–119. [25] X.C. Wu, W. Lee, L. Tran, S.L. Wong, J. Bacteriol. 173 (1991) 4952–4958. [26] A. Lejeune, K. Sakaguchi, T. Imanaka, Anal. Biochem. 181 (1989) 6–11. [27] M. Makela, T. Korpela, S. Laakso, J. Biochem. Biophys. Methods 14 (1987) 85–92. [28] T. Kato, K. Horikoshi, Anal. Chem. 56 (1984) 1738–1740. [29] T. Schwede, J. Kopp, N. Guex, M.C. Peitsch, Nucleic Acids Res. 31 (2003) 3381–3385. [30] J.C. Uitdehaag, G.J. van Alebeek, B.A. van Der Veen, L. Dijkhuizen, B.W. Dijkstra, Biochemistry 39 (2000) 7772–7780. [31] Z.F. Li, J.Y. Zhang, Q. Sun, M. Wang, Z.B. Gu, G.C. Du, J. Wu, J. Chen, J. Agric. Food Chem. 57 (2009) 8386–8391. [32] R.M. Kelly, L. Dijkhuizen, H. Leemhuis, Appl. Microbiol. Biotechnol. 84 (2009) 119–133. [33] N. Ishii, K. Haga, K. Yamane, K. Harata, J. Biochem. 127 (2000) 383–391.
