Fusion of Self-Assembling Amphipathic Oligopeptides with Cyclodextrin Glycosyltransferase Improves 2-O-d-Glucopyranosyl-l-Ascorbic Acid Synthesis with Soluble Starch as the Glycosyl Donor

Abstract

In this study, we fused six self-assembling amphipathic peptides (SAPs) with cyclodextrin glycosyltransferase (CGTase) from Paenibacillus macerans to catalyze 2-O-d-glucopyranosyl-l-ascorbic acid (AA-2G) production with cheap substrates, including maltose, maltodextrin, and soluble starch as glycosyl donors. The results showed that two fusion enzymes, SAP5-CGTase and SAP6-CGTase, increased AA-2G yields to 2.33- and 3.36-fold that of wild-type CGTase when soluble starch was used as a substrate. The cyclization activities of these enzymes decreased, while disproportionation activities increased. Enzymatic characterization of the two fusion enzymes was performed, and kinetics analysis of AA-2G synthesis confirmed the enhanced soluble starch specificity of SAP5-CGTase and SAP6-CGTase compared to that in the wild-type CGTase. As revealed by structure modeling of the fusion and wild-type CGTases, enhanced substrate-binding capacity may result from the increased number of hydrogen bonds present after fusion. This study demonstrates an effective protein fusion approach to improving the substrate specificity of CGTase for AA-2G synthesis. Fusion enzymes, especially SAP6-CGTase, are promising starting points for further development through protein engineering.

INTRODUCTION

l-Ascorbic acid (l-AA), also known as vitamin C, is an essential nutrient in humans and animals. It also has wide applications in the food, cosmetic, and pharmaceutical industries owing to its functions in collagen formation, iron absorption, and carnitine synthesis (1–3). However, the extreme instability of l-AA in aqueous solution greatly limits its applications (4). One stable l-AA derivative, 2-O-d-glucopyranosyl-l-ascorbic acid (AA-2G), avoids this instability problem with advantages of nonreducibility, antioxidation, and effortless release of l-AA and glucose. AA-2G currently has wide applications in cosmetics, medicine, and aquaculture (3).

Cyclomaltodextrin glucanotransferase (CGTase) is generally considered the most effective catalyst for AA-2G biosynthesis. Commonly, CGTase can catalyze four reactions. In addition to minor hydrolysis, it mainly catalyzes three transglycosylation reactions: cyclization (cleavage of an α-glycosidic bond in amylase or starch and subsequent formation of a cyclodextrin), coupling (cleavage of an α-glycosidic bond of a cyclodextrin ring and transfer of the resulting malto-oligosaccharide to a acceptor substrate), and disproportionation (cleavage of an α-glycosidic bond of a linear malto-oligosaccharide and transfer of one part to an acceptor substrate). In AA-2G biosynthesis reaction, CGTase transfers a glycosyl residue from glycosyl donors to l-AA (5). The glycosyl donors α- and β-cyclodextrin are commonly used for AA-2G production with CGTase owing to their high transformation efficiency (6, 7). However, neither is suitable for large-scale production of AA-2G owing to the high cost of α-cyclodextrin and the low solubility of β-cyclodextrin in aqueous solution (3). To date, many attempts have been made to find an inexpensive and soluble substrate to replace α- and β-cyclodextrin as glycosyl donors for AA-2G production. For instance, we have enhanced the transformation efficiency of maltodextrin for AA-2G production through site-directed saturation mutagenesis of the −3 subsite (8), +2 subsite (9), and −6 subsite residues (10) in the CGTase from Paenibacillus macerans. Furthermore, we improved the transformation efficiency of soluble starch to AA-2G by fusing the carbohydrate-binding module from Alkalimonas amylolytica α-amylase with CGTase from P. macerans (11).

Self-assembling amphipathic peptides (SAPs) composed of alternating hydrophobic and hydrophilic residues can assemble into ordered nanostructures (12). Therefore, the main applications of SAPs have been in nanotechnology fields such as biomedical nanotechnology (13, 14), molecular electronics (15), and cell culturing. In addition, SAPs can be used to protect membrane proteins (16, 17) and improve the water solubility of plant cytochrome CYP73A1 (18). Wang et al. (19) have immobilized enzymes for high activity and stability with these peptides via hydrogenation.

The six SAPs listed in Table 1 were used in the present study. SAP1 is a well-defined, self-complementary amphipathic β-peptide that forms a β-sheet structure with high hydrophilicity in aqueous solution (20). SAP2 is a well-defined central amphipathic α-helical structure that disorders N and C termini to form a coil structure with high hydrophobicity (21). SAP3, SAP4, and SAP5 are helix-turn-helix peptides that form amphipathic α-helical fibrils after incubation in 10 mM sodium phosphate (pH 7.6) (22). SAP6 is a helix-turn-helix peptide that forms a small amount of nonfibrillar precipitate after the same incubation with SAP3 (22).

TABLE 1.

Self-assembling amphipathic peptides

Peptide	Sequencea	Source	Reference
SAP1	AEAEAKAKAEAEAKAK	From a putative Z-DNA binding protein (Zuotin) in Saccharomyces cerevisiae	20
SAP2	VNYGNGVSCSKTKCSVNWGQAFQERYTAGTNSFVSGVSGVASGAGSIGRR	One amino acid sequence from CbnB2	21
SAP3	DWLKAFYDKVAEKLKEAFKVEPLRADWLKAFYDKVAEKLKEAF	Synthesized by Kristi L. Lazar et al.	22
SAP4	DWLKAFYDKVAEKLKEAFGLLPVLEDWLKAFYDKVAEKLKEAF	Synthesized by Kristi L. Lazar et al.	22
SAP5	DWLKAFYDKVAEKLKEAFKVQPYLDDWLKAFYDKVAEKLKEAF	Synthesized by Kristi L. Lazar et al.	22
SAP6	DWLKAFYDKVAEKLKEAFNGGARLADWLKAFYDKVAEKLKEAF	Synthesized by Kristi L. Lazar et al.	22

^aAll peptides were connected with CGTase with the same linker: PTPPTTPTPPTTPTPT.

The ability of these SAPs to improve enzyme properties has been reported. For instance, the thermal stability and specific activity of lipoxygenase has been enhanced through fusion with these SAPs (23), and the catalytic efficiency, thermal stability, and resistance to alkaline α-amylase oxidation were improved using a similar fusion strategy (24). Therefore, in the present study, we investigated whether the fusion of these six SAPs at the N terminus of CGTase from P. macerans can improve catalytic efficiency, thermal stability, and substrate (maltodextrin, maltose, and soluble starch) specificity. The mechanism responsible for enhanced substrate specificity was further explored through modeling of the three-dimensional structures of these fusion proteins. This study provides a novel strategy for enhancing AA-2G production from soluble starch with CGTase, and we expect our results to expand applications of CGTase in the starch industry.

MATERIALS AND METHODS

Bacterial strains, plasmids, and materials.

Recombinant pET-22b(+)/SAP plasmids, which contained various SAPs (see Table 1), were used. Each SAP was fused with a 5′-terminal histidine tag and a 3′-terminal PT linker (PTPPTTPTPPTTPTPT) synthesized by Sangon (Shanghai, China) (23). The plasmid pET-20b(+)/cgt (eliminating the internal NcoI site within the gene encoding CGTase (cgt; NCBI accession no. AF047363]) was constructed in our previous study (25) and used as the gene source of CGTase from P. macerans. Escherichia coli JM109 was used as the host for plasmid construction, and E. coli BL21(DE3) was used as the expression host. Plasmid pET-22b(+) was used as the cloning and expression vector.

PrimeSTAR HS DNA polymerase, restriction endonucleases, and PCR reagents were purchased from TaKaRa (Dalian, China). DNA sequencing was performed by Sangon. AA-2G was purchased from Wako Pure Chemical (Wako, Japan), and l-AA was purchased from Jiangshan Pharmaceutical (Jiangsu, China). Maltodextrin, maltose, and soluble starch were purchased from Sangon. All other chemicals and reagents were of analytical grade.

Construction of pET-22b(+)/cgt and pET-22b(+)/SAP-cgt.

The recombinant plasmid pET-22b(+)/cgt was constructed using the following procedure. The cgt gene was amplified with PrimeSTAR HS DNA polymerase using the plasmid pET-20b(+)/cgt as the template. The PCR process included 30 cycles of denaturation for 10 s at 98°C, annealing for 15 s at 55°C, and extension for 2 min at 72°C. The primers were designed using Primer Premier 5 with NcoI and XhoI restriction sites (underlined) introduced into the 5′ terminus of the forward and reverse primers, respectively: forward primer, CATGCCATGGATcatcatcatcatcatcatTCACCCGATACGAGCGTGGACA (lowercase letters are the His tag gene); and reverse primer, CCGCTCGAGTTAATTTTGCCAGTCCACCGT. The gel-purified PCR product was digested with NcoI and XhoI and then ligated into the similar restriction-digested expression vector pET-22b(+) to construct recombinant plasmid pET-22b(+)/cgt (Fig. 1a and b).

FIG 1.

Diagrammatic drawing on the construction and purification of wild-type CGTase and SAPs-CGTase. (a and b) Construction of pET-22b(+)/cgt; (c and d) construction of pET-22b(+)/SAP-cgt.

The recombinant plasmid pET-22b(+)/SAP-cgt was constructed according to the following procedure. We amplified cgt with PCR using the plasmid pET-20b(+)/cgt as the template. The forward and reverse primers were CATGCCATGGAT TCACCCGATACGAGCGTGGACA (the underlined letters are NcoI restriction sites) and CCGCTCGAGTTAATTTTGCCAGTCCACCGT (the underlined letters are XhoI restriction sites), respectively. The gel-purified PCR product was digested with NcoI and XhoI and then ligated to the similarly restriction-digested expression vector pET-22b(+)/SAP to construct recombinant plasmid pET-22b(+)/SAP-cgt (Fig. 1c and d).

Preparation and purification of CGTase and SAP-CGTases.

Crude CGTase and SAP-CGTases were prepared using a previously reported method (25). Recombinant Escherichia coli BL21(DE3) cells were inoculated into 20 ml of Luria-Bertani medium containing 100 μg of ampicillin/ml and grown at 37°C overnight. The seed culture was then inoculated into the flask at a concentration of 4% (vol/vol) for enzyme production. The culture medium (initial pH 7.0) contained 8 g of glucose, 0.5 g of lactose, 12 g of peptone, 24 g of yeast extract, 16.43 g of K₂HPO₄, 2.31 g of KH₂PO₄, 0.28 g of CaCl₂, 4 g of glycerol, and 0.1 g of ampicillin/liter. The flask culture was incubated on a rotary shaker (200 rpm) at 25°C for 90 h after being induced with 0.1 mM IPTG (isopropyl-β-d-thiogalactopyranoside) when the optical density at 600 nm reached 0.6. The broth was centrifuged at 10,000 × g and 4°C for 5 min, and the supernatant was purified and used for the subsequent transformation. The purification of the crude enzyme solution was carried out using a nickel-nitrilotriacetic acid agarose column (Qiagen, Chatsworth, CA) as described elsewhere (26).

Biosynthesis and analysis of AA-2G.

The purified wild-type CGTase and SAP-CGTases were diluted with acetic acid-sodium acetate buffer (0.2 M, pH 5.5) to a protein concentration of 1 mg/ml and then mixed with the glycosyl donor (maltodextrin, maltose, or soluble starch) and acceptor (l-AA). The final concentration of both the glycosyl donor and acceptor in the reaction mixture was 1.25% (wt/vol). The mixture was incubated at 37°C for 24 h in darkness under oxygen-free conditions. Finally, glucoamylase (10 U/ml) was added to the reaction mixture, followed by incubation at 65°C and pH 5.5 for 24 h to hydrolyze the reaction intermediate AA-2Gn to AA-2G (“n” indicates the number of glycosyls attached to l-AA). AA-2G was analyzed using a method described previously (6). Briefly, samples were filtered using a 0.45-μm-pore-size membrane and then measured using high-performance liquid chromatography with an SB-Aq column (4.6 mm by 250 mm). The mobile phase was 25 mM KH₂PO₄/H₃PO₄ (pH 2.0) at a flow rate of 0.5 ml/min, and the detection wavelength was 238 nm. Based on the initial transformation conditions (temperature 37°C, pH 5.5), the influence of reaction temperature (20, 28, 36, 44, and 52°C) and pH (0.2 M acetic acid-sodium acetate buffer [pH 4.0, 4.5, 5.0, 5.5, and 6.0]; 0.2 M phosphate buffer [pH 6.0, 6.5, 7.0, and 8.0]) on the biosynthesis of AA-2G by the wild-type and mutant CGTases was also investigated.

The kinetic analysis of the wild-type and SAP-CGTases for AA-2G biosynthesis (using l-AA and soluble starch as glycosyl acceptor and donor, respectively) was performed by measuring the amount of AA-2G produced with various concentrations of l-AA (0.1, 0.5, 1, 2, and 4 g/liter), while the concentration of soluble starch remained fixed (0.1, 0.5, 1, 2, and 4 g/liter). The results were subjected to kinetic analysis using SigmaPlot (Jandel Scientific). The following equations (27) were used to fit the experimental data to determine which kinetic mechanism applied to the transglycosylation reactions catalyzed by CGTase.

The normal ping-pong mechanism is represented by equation 1:

$$v=V_{\max \hspace{0.17em}}\cdot a\cdot b/(K_{m\mathrm{A}}\cdot b+K_{m\mathrm{B}}\cdot a+a\cdot b)$$ (1)

The substrate inhibition mechanism is represented by equation 2:

$$v=V_{\max \hspace{0.17em}}\cdot a\cdot b/(K_{m\mathrm{B}}\cdot a+K_{m\mathrm{A}}\cdot b\cdot (1+b/K_{i\mathrm{B}})+a\cdot b)$$ (2)

where v is the reaction rate (the amount of AA-2G formed by 1 mg of enzyme per h, expressed in mol·liter⁻¹ h⁻¹), V_max is the maximal reaction rate (mol liter⁻¹ h⁻¹), a and b are the donor (soluble starch) and acceptor (l-AA) concentrations (g/liter), respectively, K_mA and K_mB are the affinity constants for the substrates soluble starch and l-AA, respectively, and K_iB is the inhibition constant for the substrate soluble starch. All kinetic reactions were determined within the first 8 h of AA-2G synthesis to obtain accurate experimental data.

Analysis of CGTase activity.

α-Cyclodextrin-forming activity was analyzed using a method described previously (25). Briefly, 0.1 ml of purified enzyme with a concentration of 1 mg/ml diluted with 50 mM phosphate buffer (pH 6.0) was mixed with 0.9 ml of 3% (wt/vol) soluble starch. The mixed broth was incubated for 10 min at 40°C, and the reaction was terminated with the addition of 1.0 M HCl (1.0 ml). Immediately, 1.0 ml of 0.1 mM methyl orange was added, followed by incubation for 20 min at 16°C, after which the absorbance at 505 nm was measured. One unit of α-cyclodextrin-forming activity was defined as the amount of enzyme that produced 1 μM α-cyclodextrin per min.

Hydrolyzing activity was determined using a starch-degrading method (28). Purified enzyme (1 mg/ml) was mixed with 1% (wt/vol) soluble starch dissolved with 50 mM phosphate buffer (pH 6.0) and incubated at 50°C for 10 min. One unit of hydrolyzing activity was defined as the amount of enzyme producing 1 μM reducing sugar per min.

Disproportionation activity was determined using methods described previously (28). 4-Nitrophenyl-α-d-maltoheptaoside-4-6-O-ethylidene and maltose were used as donor and acceptor substrates, respectively. One unit of activity was defined as the amount of enzyme converting 1 μmol of 4-nitrophenyl-α-d-maltoheptaoside-4-6-O-ethylidene per min.

Protein concentrations were determined with the Bradford method using a Bradford protein assay kit (Beyotime, Jiangsu, China) with bovine serum albumin as a standard. The half-life at 40°C of the wild-type CGTase and fusion enzymes was determined by incubating the purified enzyme in phosphate buffer (0.2 M, pH 6.0) at 40°C. Samples were taken at 1-h intervals for analysis of α-cyclodextrin-forming activity. To make sure that all data were determined at the initial reaction rate, all results in every reaction were measured when substrate transformation efficiency was <20%.

Structure modeling of the wild-type CGTase and SAP6-CGTase.

The I-TASSER server (http://zhanglab.ccmb.med.umich.edu/I-TASSER) was used to construct homology models of the wild-type CGTase and SAP6-CGTase based on multiple-threading alignments and iterative template fragment assembly simulations. All graphical molecular representations were generated using Accelrys Discovery Studio Client, version 2.5 (Accelrys, USA). Structural alignment was performed according to the combinatorial extension method by using the jCE/jFATCAT Structure Alignment Server v2.6 (http://source.rcsb.org/jfatcatserver/) (29). The stereochemical quality of the model was examined with PROCHECK (30), Verify3D (31), and ProQ (32). The root mean square deviation (RMSD) between the template and the model alpha carbon backbones was calculated using the combinatorial extension method (29). A maltononaose substrate was transferred from the active site of the Protein Data Bank 1CXK CGTase to the active site of the model. Finally, energy minimization of the enzyme-substrate interaction was carried out with the amber-based energy minimization method provided by Accelrys Discovery Studio Client, version 2.5.

RESULTS AND DISCUSSION

Expression and purification of CGTase and SAP-CGTases.

The recombinant plasmids pET-22b(+)/cgt and pET-22b(+)/SAP-cgt were constructed and verified through DNA sequencing. We then transformed these recombinant plasmids into the host E. coli BL21(DE3) and expressed recombinant proteins. The extracellular expression levels of SAP1-CGTase to SAP6-CGTase were 1.41, 1.62, 2.01, 1.60, 1.37, and 2.81 mg/ml, respectively, values which were lower than that of the wild-type CGTase (2.95 mg/ml).

The crude CGTase and SAP-CGTases were purified with one-step nickel affinity chromatography on nickel-nitrilotriacetic acid resin. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed that the molecular masses of the purified wild-type CGTase and fusion enzymes (purity > 90%) ranged from 75 to 80 kDa (see Fig. S1 in the supplemental material).

Influence of SAP fusion on reaction activities of CGTase.

As shown in Table 2, the influence of fusion enzymes SAP1-CGTase to SAP6-CGTase on cyclization, hydrolysis, and disproportionation activities was investigated. The results showed that these six fusion enzymes retained less cyclization activity than that of the wild-type CGTase (160 ± 5 U/mg). Compared to the hydrolysis (starch-degrading) activity of the wild-type CGTase (8.0 ± 0.1 U/mg), SAP3-CGTase and SAP6-CGTase showed a slight decrease, whereas the other fusion enzymes increased their hydrolysis activities to various degrees. Furthermore, compared to the disproportionation activity of the wild-type CGTase, those of the fusion enzymes SAP5-CGTase and SAP6-CGTase increased by 20.4 and 48.5%, respectively. Nevertheless, the other fusion enzymes showed disproportionation activities lower than that of the wild-type CGTase.

TABLE 2.

Comparison of wild-type CGTase and SAP-CGTase in different reactions and in AA-2G synthesis

Enzyme	Mean ± SDa
	Relative activity (%)			AA-2G titer (10−1 g/liter)b			t1/2 at 40°C (h)
	Cyclization	Hydrolysis	Disproportionation	Maltodextrin	Maltose	Soluble starch
Wild type	100	100	100	12.0 ± 0.6	10.0 ± 0.8	5.5 ± 0.2	7.6 ± 0.4
SAP1-CGTase	10.9 ± 0.6	132.2 ± 8.6	57.5 ± 2.5	6.2 ± 0.3	2.7 ± 0.3	5.1 ± 0.3	6.1 ± 0.4
SAP2-CGTase	8.7 ± 0.4	116.1 ± 9.8	46.7 ± 2.3	6.3 ± 0.2	2.0 ± 0.2	5.2 ± 0.3	6.8 ± 0.3
SAP3-CGTase	34.5 ± 1.5	75.8 ± 7.4	85.3 ± 3.6	6.0 ± 0.4	10.0 ± 0.6	5.9 ± 0.4	7.2 ± 0.4
SAP4-CGTase	29.1 ± 1.7	133.8 ± 9.7	80.4 ± 4.2	7.2 ± 0.5	10.4 ± 0.5	6.1 ± 0.5	6.9 ± 0.5
SAP5-CGTase	60.7 ± 2.6	120.9 ± 8.9	120.4 ± 4.5	16.3 ± 0.6	11.2 ± 0.7	12.8 ± 0.4	6.4 ± 0.4
SAP6-CGTase	34.9 ± 2.3	70.9 ± 6.7	148.5 ± 5.1	11.7 ± 0.3	11.4 ± 0.6	18.5 ± 0.9	5.3 ± 0.9

^aThe cyclization (α-cyclodextrin-forming activity), hydrolysis (starch-degrading activity), and disproportionation reaction activities for the wild-type CGTase were 160 ± 5, 8.0 ± 0.1, and 794 ± 8 U/mg, respectively, and these values were defined as 100% for the relative activity. Each value represents the mean of three independent measurements, and the deviation from the mean was below 5%.

^bThat is, for different glycosyl donors.

We also investigated the AA-2G yields produced by the purified wild-type CGTase and fusion enzymes with different glycosyl donors such as maltodextrin, maltose, and soluble starch. As shown in Table 2, these donors resulted in AA-2G yields of 1.20, 1.00, and 0.55 g/liter, respectively, with the wild-type CGTase. The AA-2G titers produced by SAP5-CGTase were 1.63, 1.12, and 1.28 g/liter with maltodextrin, maltose, and soluble starch as glycosyl donors, which was 1.36-, 1.12-, and 2.33-fold those of the wild-type CGTase, respectively. SAP6-CGTase had a AA-2G yield similar to that of the wild-type CGTase when maltodextrin was used as the glycosyl donor, but the yield increased by 14% and 2.36-fold with maltose and soluble starch as glycosyl donors, respectively. With other fusion enzymes, however, AA-2G yields with these three glycosyl donors were not obviously increased compared to yields with the wild-type CGTase. From these results, we conclude that the AA-2G yields of SAP5-CGTase and SAP6-CGTase increase noticeably when soluble starch is used as the glycosyl donor.

In addition, we analyzed AA-2G titers with soluble starch as the glycosyl donor using our previous CGTase mutants K47L/Y89F/N94P/D196Y (33), Y260R/Q265K/Y195S (9), and Y167S/G179K/N193R/G180R (10). These yields were 0.88, 0.62, and 0.99 g/liter, respectively—clearly much lower than those of SAP5-CGTase and SAP6-CGTase. As shown in Table 2, the half-life of wild-type CGTase at 40°C was 7.6 h, whereas those of the fusion enzymes SAP1-CGTase to SAP6-CGTase decreased comparatively, indicating that the thermostability of all fusion enzymes decreased.

Influence of reaction temperature and pH on AA-2G biosynthesis with the wild-type CGTase and the fusion enzymes SAP5-CGTase and SAP6-CGTase.

We investigated the influence of reaction temperature and pH on AA-2G biosynthesis with soluble starch as the glycosyl donor using the wild-type CGTase, SAP5-CGTase, and SAP6-CGTase. As shown in Fig. 2A, the optimal temperature for AA-2G production with the wild-type CGTase was 36°C, which was also reported in our previous studies (8–10) and is similar to the optimal temperature for α-CGTase-catalyzed AA-2G synthesis with β-cyclodextrin as the glycosyl donor (6). However, SAP5-CGTase and SAP6-CGTase produced the highest AA-2G titer at 28°C, and this optimal temperature is the same as that of production with the chimeric enzymes CGT-CBM_Amy and CGTΔE-CBM_Amy (11).

FIG 2.

Influence of reaction temperature (A) and pH (B) on AA-2G synthesis by the wild-type, SAP5-CGTase, and SAP6-CGTase with soluble starch as the glycosyl donor (pH 4.0 to 6.0 with acetic acid-sodium acetate buffer; pH 6.0 to 8.0 with phosphate buffer.).

Figure 2B shows the influence of pH on AA-2G synthesis with the wild-type CGTase, SAP5-CGTase, and SAP6-CGTase with soluble starch as the glycosyl donor. All enzymes exhibited the best capability for AA-2G synthesis at pH 6.5, which was similar to the optimal pH for production with the chimeric enzymes CGT-CBM_Amy and CGTΔE-CBM_Amy (11) for AA-2G synthesis. Our previous studies indicated that the optimal pH for cyclization and AA-2G biosynthesis with the wild-type CGTase using β-cyclodextrin as the glycosyl donor was 5.5 (6, 25).

Influence of fusion on the reaction kinetics of CGTase.

The reaction kinetics of the purified wild-type CGTase and fusion enzymes SAP5-CGTase and SAP6-CGTase were investigated. The Lineweaver-Burk plot of the kinetics for the wild-type CGTase (inset) and SAP6-CGTase are shown in Fig. 3. The Lineweaver-Burk plot of the wild-type CGTase shows parallel lines indicating a normal ping-pong type of kinetics, and the linear regression of the experimental data corresponded well to the values calculated using equation 1. The experimental data for SAP6-CGTase corresponded well to the values calculated using equation 2, which represents a substrate inhibition mechanism and indicates that too high a concentration of soluble starch inhibited CGTase activity and AA-2G synthesis. In addition, experimental data for SAP5-CGTase were best fit by equation 2 (data not shown); the detailed kinetic parameters are listed in Table 3.

FIG 3.

Lineweaver-Burk plots of the AA-2G synthesis with soluble starch as the glycosyl donor by the wild-type CGTase (inset) and SAP6-CGTase.

TABLE 3.

Kinetic parameters of wild-type CGTase, SAP5-CGTase, and SAP6-CGTase^a

Enzyme	Vmax (10−1 mmol liter−1 h−1)	kcat (h−1)	l-AA		Soluble starch
			Km (g/liter)	kcat/Km (liter/g·h)	Km (g/liter)	kcat/Km (liter/g·h)	Ki (g/liter)
Wild type	0.8 ± 0.1	6.0	0.6 ± 0.1	10.0	1.2 ± 0.2	5.0	ND
SAP5-CGTase	1.6 ± 0.3	13.1	0.5 ± 0.1	26.2	1.0 ± 0.1	13.1	16.3
SAP6-CGTase	2.4 ± 0.4	19.7	0.5 ± 0.1	39.4	0.8 ± 0.1	24.6	19.1

^aAll data were determined under the optimal transformation conditions of each enzyme for AA-2G production, such as wild-type CGTase (36°C, pH 6.5), SAP5-CGTase (28°C, pH 6.5), and SAP6-CGTase (28°C, pH 6.5). ND, not detectable.

The maximal reaction rate (V_max) of SAP5-CGTase and SAP6-CGTase were higher than that of the wild-type CGTase. Meanwhile, compared to that of the wild-type CGTase, the K_m values of SAP5-CGTase and SAP6-CGTase with soluble starch as the substrate decreased by 16.6 and 33.3%, respectively, whereas K_cat/K_m increased by 1.6- and 3.9-fold, respectively. These results indicated that the affinity for and catalytic efficiency of fusion enzymes SAP5-CGTase and SAP6-CGTase with soluble starch increased compared to those of the wild-type CGTase. However, the K_m values of both fusion enzymes with l-AA showed only a small decrease compared to that of the wild-type CGTase, suggesting that the fusion enzymes had insignificant influence on affinity for l-AA. Conversely, the catalytic efficiency of these fusion enzymes toward l-AA increased, as indicated by the higher K_cat/K_m (l-AA) values compared to that of the wild-type CGTase. The wild-type CGTase showed little substrate inhibition, whereas the fusion enzymes showed soluble starch inhibition. Furthermore, the inhibition constant K_i (soluble starch) value of SAP5-CGTase was lower than that of SAP6-CGTase, indicating that substrate inhibition by soluble starch was most pronounced for SAP5-CGTase.

Analysis of the wild-type CGTase and fusion enzyme structure models.

Three-dimensional structure models of the wild-type CGTase from P. macerans and the fusion enzyme SAP6-CGTase are shown in Fig. S2 in the supplemental material. Bacillus circulans strain 251 CGTase (Protein Data Bank accession code 1CXK) was selected as the best template and had ca. 67.9% identity with the models. The sequence alignment showed only one gap at position 249. The overall structures of model and template were very similar, with an RMSD of 0.7 Å. The stereochemical quality of the models was examined with PROCHECK and revealed 92% of residues in the most favored regions and 0.6% in disallowed regions. Verify3D showed that the compatibility of the atomic model had no residues with values less than zero. Values of LG and MaxSub obtained with ProQ showed that the models were very good. These results indicated that the modeled structures of wild-type CGTase and SAP6-CGTase can be used for subsequent analyses.

Compared to the wild-type CGTase, SAP6-CGTase has an additional peptide composed of two α-helixes at the N terminus of the CGTase (see Fig. S2A and B, respectively, in the supplemental material). This possible structure model of SAP6 corresponds to that in a previous report (34). The average RMSD values of SAP5 and SAP6 were 1.394 and 1.647, respectively, as analyzed in our previous work (23, 24). These high RMSD values suggested a relatively flexible structure in these peptides (35, 36). As we knew, reduction of the N- or C-terminal flexibility of enzymes may lead to an increase in their thermostability (37–39). Therefore, the relatively highly flexible structure at the N termini of SAP5-CGTase and SAP6-CGTase may be responsible for reducing their thermostability compared to that of the wild-type CGTase (see Table 2).

Our previous studies suggested that the −3 subsite, the +2 subsite, and the −6 subsite of CGTase from P. macerans play an important role in the binding of linear substrate (8–10, 40). Figure 4 compares the substrate binding subsites (−3, +2, and −6) of the wild-type CGTase and SAP6-CGTase. Compared to the wild-type CGTase, SAP6-CGTase has an additional one and two hydrogen bonds between the substrate glycosyls and the residues D261 and Y154 (corresponding with D202 and Y95 in the wild-type CGTase), respectively, which may improve the binding capacity between the −3 subsite of CGTase and linear substrate. Meanwhile, the number of hydrogen bonds between the −6 subsite residues of SAP6-CGTase and linear substrate is also increased compared to that of the wild-type CGTase—for example, N258 in SAP6-CGTase (corresponding with N199 in wild-type CGTase). At the +2 subsite, an additional hydrogen bond is present between the substrate glycosyl residues and Y325 (SAP6-CGTase) compared to corresponding site Y266 (wild-type CGTase), whereas Y260 (SAP6-CGTase) loses two hydrogen bonds with the +1 glycosyl. These changes at the +2 subsite may be favorable for the binding capacity of SAP6-CGTase with linear substrate, a possibility that has been discussed in a previous study (9). Finally, the total number of hydrogen bonds between SAP6-CGTase and the substrate was higher than that of the wild-type CGTase. These changes may be responsible for the enhanced linear substrate binding capacity of the fusion enzyme.

FIG 4.

Comparison of the substrate-binding area between the wild-type CGTase and SAP6-CGTase by superpositioning of maltononaose structure (PDB accession code 1CXK). (Left) Wild-type CGTase; (right) SAP6-CGTase. Green dotted line, hydrogen bond; blue number, number of substrate glycosyl; red number, number of hydrogen-bond between the residue and the substrate glycosyl.

The AA-2G titers produced by the CGTase mutants K47L/Y89F/N94P/D196Y (mutagenesis in the −3 subsite) (33), Y260R/Q265K/Y195S (mutagenesis in the +2 subsite) (9), and Y167S/G179K/N193R/G180R (mutagenesis in the −6 subsite) with soluble starch as the glycosyl donor were much lower than those of the fusion enzymes SAP5-CGTase and SAP6-CGTase. This result may confirm our speculation above. In addition, the disproportionation activity of SAP5-CGTase and SAP6-CGTase was higher than that of the wild-type CGTase (see Table 2), which was favorable for AA-2G synthesis with linear substrates because disproportionation plays a key role in the AA-2G synthesis reaction. Compared to values for the wild-type CGTase, the decrease in K_m value and increase in K_cat values for the fusion enzymes SAP5-CGTase and SAP6-CGTase (see Table 3) indicate that the affinity and catalytic efficiency of these fusion enzymes with soluble starch substrate are enhanced. These results also further confirm our conclusions drawn from homology modeling.

However, different linear substrates contributed to different AA-2G yields (see Table 2). Compared to that of soluble starch, the enhancement of the transformation efficiency of maltodextrin and maltose appeared unremarkable. Although a detailed explanation for this small difference remains unclear, we speculate that the longer carbohydrate chain makes binding easier with this kind of enzyme. Similar outcomes to those previously reported (5) might confirm this possibility.

In summary, the enhanced soluble starch specificity for AA-2G synthesis was achieved via fusion of self-assembling amphipathic peptides to the N terminus of CGTase from P. macerans. Similarly, soluble starch substrate specificity was improved by fusing the carbohydrate-binding module of A. amylolytica α-amylase with CGTase from P. macerans in a previous study (11). These outcomes suggest that protein fusion technology is an effective approach to the modification of CGTase for AA-2G production. Although the AA-2G yield produced from soluble starch is still much lower than that from α- and β-cyclodextrin, this substrate provides a direction for the enhancement of transformation efficiency of cheap and easily soluble substrates for AA-2G production. Clearly, substrate inhibition was an intractable challenge in this work. Hence, additional studies investigating substrate inhibition should be carried out, and other approaches, such as enzyme immobilization and transformation optimization, should be conducted to improve AA-2G yield further.