Asn336 is involved in the substrate af ﬁ nity of glycine oxidase from Bacillus cereus Biotechnology

Background: Glycine oxidase (GO), a type of D -amino acid oxidase, is of biotechnological interest for its potential inseveral ﬁ elds.Inourpreviousstudy,wehavecharacterizedanewglycineoxidase(BceGO)from Bacilluscereus HYC-7. Here, a variant of N336K with increased the af ﬁ nity against all the tested substrate was obtained by screening a random mutant library of BceGO. It is observed that the residue N336 is invariable between its homogeneous enzymes. This work was aimed to explore the role of the residue N336 in glycine oxidase by site-directed mutagenesis, kinetic assay, structure modeling and substrate docking. Results: The results showed that the af ﬁ nity of N336H, N336K and N336R increased gradually toward all the substrates, with increase in positive charge on side chain, while N336A and N336G have not shown a little signi ﬁ cant effect on substrate af ﬁ nity. The structure modeling studies indicated that the residue Asn336 is located in a random coil between β -18 and α -10. Also, far-UV CD spectra-analysis showed that the mutations at Asn336 do not affect the secondary structure of enzyme. Conclusion: Asn336 site was located in a conserved GHYRNG loop which adjoining to substrate and the isoalloxazine ring of FAD, and involved in the substrate af ﬁ nity of glycine oxidase. This might provide new insight into the structure – function relationship of GO, and valuable clue to redesign its substrate speci ﬁ city for some biotechnological application.

In our previous study, we have reported a new glycine oxidase (BceGO) with glyphosate-oxidative activity from Bacillus cereus and, developed a high through screening method for improving its affinity and activity toward glyphosate [7]. Here, we continued to screen new mutant with higher specificity to glyphosate from a random mutation library of BceGO, and obtained a mutant, N336K, whose K m, app on glyphosate decreased 3.77-fold. Sequence alignment showed that the residue N336 is highly conserved in BceGO and its homogeneous enzymes. Here, we attempted to investigate the role of N336 residue in the catalytic activity of GO by site-directed mutagenesis, three-dimensional structure modeling and ligand docking assay.
Sangon (Shanghai, China), respectively. Escherichia coli DH5α and E. coli BL21 (DE3) were used as for gene cloning and for protein expression, respectively.

Construction of mutant library and site-directed mutagenesis
The BceGO random mutant library was generated by error-prone PCR used pGEX-GO as the template. The amplification mixture, which contained 20 nM primers, 0.2 mM dATP and dCTP, 0.1 mM dTTP and dGTP, 2 U Taq DNA polymerase and Taq buffer containing 5 mM MgCl2 and 0.5 mM MnCl2 in 100 μL volume, was cycled in an Bio-rad thermal cycler (California, USA) for 30 cycles of 94°C for 30 s, 57°C for 30 s, and 72°C for 70 s. PCR products were purified, digested with BamHI and XhoI, cloned into pGEX-6P-1, and transformed into E. coli DH5α to obtain the random mutant library.
PCR-based site-directed mutagenesis was carried out to generate single-mutant [8]. PCR reactions (50 μL) contained 20 ng template (pGEX-GO), 0.2 mM dNTP, 20 nM each primer, 10 μL PCR buffer and 1 unit of Pfu DNA polymerase (Transgen, China). The PCR cycling parameters were: 1 cycle of 2 min at 97°C, 20 cycles of 20 s at 95°C, 30 s at 54°C, and 160 s at 72°C, and incubation of 10 min at 72°C. Then the PCR products were treated with DpnI to digest the parental DNA at 37°C for 8 h. Finally, DpnI digestion mixture was transformed into E. coli DH5α competent cells, and the transformant was selected on ampicillin plates. The primers used were listed in Table 1. The desired mutants were validated by DNA sequencing.

Screening for GO mutants
The mutant library was screened by an enzyme-coupled assay using horseradish peroxidase (5 U/mL) and o-dianisidine dihydrochloride as described previously [9]. Single colony from random mutation library was cultured in deep-well plates containing 0.6 mL LB medium, and induced by IPTG. Then cell extracts containing target protein were prepared by adding the bacteriophage T7. To screen mutants with higher specificity to glyphosate, 100 μL of each cell lysate was incubated with 20 μL of 50 mM glyphosate, 20 μL of 0.32 mg/mL o-dianisidine dihydrochloride, and 1 μL of 5 U/mL horseradish peroxidase in sodium phosphate buffer (50 mM; pH 8.5) followed by measuring the absorbance values at 450 nm. Mutants showed higher absorbance than the wild-type were selected for further activity analysis.

Enzyme expression and purification
The recombinant BceGO and its mutant were purified by affinity chromatography using the methods described previously [7]. Briefly, the recombinant plasmids were transformed into the host E. coli BL21 (DE3). Recombinant cells grew at 37°C in LB medium containing 100 μg/mL ampicillin. Protein expression was induced by adding isopropyl β-D-1-thiogalactopyranoside (IPTG) at a final concentration of 0.2 Mm, when the OD600 reached 0.6. After an overnight induction at 22°C, 1.5 L culture was collected and disrupted by the high pressure homogenizer (NiroSoavi, Italy). Then, the supernatant of the lysate was mixed with 1.5 mL GST•Bind Resin that had been equilibrated with 50 mM disodium pyrophosphate buffer. The resin was washed with disodium pyrophosphate buffer (50 mM, pH 7.5) to elute the unspecific-binding protein. Finally, the GST-free recombinant protein was prepared by on-column cleavage with PreScission protease [10]. The concentration of the wild-type BceGO and mutants was measured by the method of Bradford assay [11]. The purity of the protein was analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE).

Determination of kinetic parameters
The kinetic parameters of wild-type BceGO and mutants were assayed using a fixed amount of enzyme and various concentration of substrates (glycine, 0-300 mM; glyphosate, 0-600 mM; sarcosine, 0-300 mM; D-alanine, 0-600 mM). The absorbance was measured at 450 nm using a microplate reader (Thermo Scientific, Multiscan spectrum). The initial reaction velocities under various concentrations of each substrate were fitted to the Lineweaver-Burk transformation of the Michaelis-Menten equation to figure out apparent kinetic parameters (i.e., K m,app and V max ). Further, the k cat,app was calculated by the equation: k cat,app = V max / [E], in which [E] is the total amount of enzyme in the reaction mixture.

Circular dichroism and secondary structure prediction
Secondary structure of BceGO was predicted by using the program PSIPRED [12]. Circular dichroism (CD) spectra of BceGO and variants were recorded with a Jasco-810 CD spectrometer (Jasco Corp., Japan). The data were collected at room temperature from 190 to 260 nm using 1 mm quartz cuvette (400 μL). The conversion to the Mol CD (Δε) in each spectrum was performed with the Jasco Standard Analysis software. Estimation of the secondary structure content from far-UV circular dichroism (CD) spectra was performed by using the CDPro Table 1 Primers used for gene BceGO mutagenesis. The BamHI and XhoI sites were italic and underlined, and the mutation positions were underlined.

Target sites
Sequence (5′-3′)  [13]. In this study, the percentages of α-helix and β-sheet for each protein sample were averaged by the calculations of results from the CDPro software package. The circular dichroism data were expressed in terms of the mean residue ellipticity (θ mrw ), which were calculated using Equation 1 [14]: where θ obs is the observed ellipticity in degrees, M w is the molecular weight of wild-type and variants proteins, and N is the number of residues in BceGO (369), d is the path length of quartz cuvette (0.1 cm), c is the protein concentration (mg/mL), and the constant number 100 stems from the conversion of the molecular weight to mg/dmoL.

Molecular modeling analysis
To obtain a reasonable model, the structure of BceGO was built with homology modeling in InsightII program (version 2005). The crystal structure of glycine oxidase from B. subtilis (Protein Data Bank code: 1RYI) was used as the template. The binding conformation of the ligands in the BceGO active site was obtained with the docking module in MOE 2009.10, and the result description was prepared using software PyMol 0.99.

Mutagenesis of BceGO
A random mutant library of BceGO was constructed by error-prone PCR to screen new mutants with low affinity and increased activity toward glyphosate by the method of high throughput colorimetric assay. Asn336Lys mutant was selected from 7000 clones, which showed improved specificity toward glyphosate than the wild type. Its apparent K m value decreased 3.77, 1.32, 4.19 and 5.09-fold glyphosate, glycine, sarcosine and D-alanine, respectively (Table 2). However, the turnover numbers (the k cat,app ) were lower than the wild-type BceGO. Protein sequence alignment showed that Asn336 is highly conserved in the GO family, and locates in the loop connecting β-strands 18 and α-helices 10 ( Fig. 1). To elucidate the role of this invariable Asn336, it was substituted with other positively charged amino acids (i.e., His and Arg) and small amino acids (i.e., Ala and Gly) by site-directed mutagenesis.

Purification of BceGO and its variants
In order to characterize the enzyme, the wild-type BceGO and variants with GST tag were produced in E. coli BL21 (DE3) and purified by GSH-agarose affinity chromatography. GST-free recombinant fusion proteins were prepared via on-column cleavage by using PreScission protease. As a result, target proteins with high homogeneity and apparent molecular masses of 41 kDa were obtained (Fig. 2).

Kinetic parameters of BceGO variants
As shown in Table 2, it was observed that K m, app values against all substrates (i.e., glycine, glyphosate, sarcosine and D-Alanine) declined alone with the increase of positive charge on the side chain of residue 336. Especially, the k cat values of N336R toward substrates decreased 28-41-fold as compared to wild-type BceGO. It means that substitution at N366 with positively charged residues is able to improve the affinity Fig. 1. Protein sequence alignment assay. The sequence alignment was according to sites Gly258 and Glu357 of BceGO. The conserved residues were shaded in black by using the BioEdit program, and the site N336 was marked out by a black triangle. The β-strands and α-helixs in this region were indicated with an arrow according to the crystal structure of B. subtilis GO [17]. GO    for the substrates. The both substitutions N336A and N336G did not significantly affect the substrate affinity (the K m,app value) for all the substrates. Additionally, the turnover number (k cat ) of the five mutants toward all tested substrate decreased to some different degrees (Table. 2), suggesting that Asn336 is also involved in the catalytic efficiency of BceGO.

Analysis of protein secondary structure
The program PSIPRED predicted that the residue Asn336 was located in a conservative random coil region. A quantitative analysis of the protein secondary structure for wild-type BceGO and variants has been carried out using SELCON3 program. The data showed that the CD spectra of wild-type GO and mutants (N336H, N336K, N336R, N336A and N336G) were similar (Fig. 3, Table 3). This result suggested that the mutation at Asn336 did not affect the content in secondary structure.

Structure modeling and substrate docking analysis
Protein homology modeling and ligand docking assay revealed the substrates matching the BceGO active site and orientated to the isoalloxazine ring of the flavin cofactor (Fig. 4). The three dimensional structure of GO from B. subtilis showed the active site of GO containing FAD-binding domain and substrate-binding domain including a conserved Rossmann fold βαβ motif. Both theoretical and experimental studies have indicated that the positive charge in the vicinity of the active site could promote the redox potential of the flavin [15]. The carboxylic groups of the substrates through a double bridge to the Arg308 side chain, and the other side of substrates directing toward the active site entrance, might interact with Gly51, Ala54 backbones, and side chains of Arg335 and Asn336. The random coil containing Asn336 formed a bulge at the bottom of active cavity, and the mutation at this site might cause alteration in the loop connecting β-strands 18 and α-helices 10, thereby impacting the charge distribution in vicinity of the flavin. In this work, introduction of basic amino acid to site 336 didn't impair BceGO secondary structure and increased the affinity (the K m, app value) to all substrates, indicating that the positive charge near the flavin contributed to the binding of substrates to BceGO, which was accordant with previous findings [16].

Conclusion
In this work, we investigated the role of Asn336 in the active cavity of BceGO. Together with experimental data and model analysis, it was concluded that the high conserved residue Asn336 played a crucial role in the substrate affinity of BceGO, and positively charged residue could improve its substrate affinity, significantly. This study provides new insights into the structure-function relationship of glycine oxidase and valuable clue to redesign the substrate specificity by protein engineering.