2-Keto-d-Gluconate-Yielding Membrane-Bound d-Glucose Dehydrogenase from Arthrobacter globiformis C224: Purification and Characterization

Glucose dehydrogenase (GlcDH) is the rate-limiting catalyst for microbial conversion of glucose to the important organic acid 2-ketogluconic acid (2KGlcA). In this study, a d-glucose dehydrogenase was purified from the industrial 2KGlcA producer Arthrobacter globiformis C224. After four purification steps, the GlcDH was successfully purified over 180 folds and specific activity of 88.1 U/mg. A single protein band of 87 kDa was detected by SDS-PAGE. The purified GlcDH had the broad substrate specificity with the Km values for d-glucose, d-xylose, d-galactose and maltose of 0.21 mM, 0.34 mM, 0.46 mM and 0.59 mM, respectively. The kinetic studies proved that A. globiformis GlcDH followed the ping-pong kinetic mechanism. The GlcDH showed an optimum catalytic activity at pH 5.0 and 45 °C with the stable activity at temperature of 20–40 °C and pH of 6.0–7.0. Organic solvents, metal ions or EDTA could significantly influence the GlcDH activity to different degrees.

performance and develop the A. globiformis strains, the present study will aim to: (1) first purify the GlcDH from the membrane fraction of 2KGlcA industrial strain A. globiformis C224; and (2) characterize the biochemical and catalytic properties of the purified GlcDH.

Extraction and Purification of GlcDH from A. globiformis C224
Membrane proteins are embedded in the mosaic lipid bilayer with low contents and hydrophobic characteristics, which will hinder the extraction of the interested membrane proteins [16]. Some kinds of detergents including sodium dodecyl sulfate (SDS), bile acid salts, nonionic detergents and zwitterionic detergents can soluble the membrane proteins without affecting their structural features [17,18]. Triton X-114 is a nonionic detergent used for extracting the membrane-associated proteins [19][20][21] including the cell envelope associated proteins from M. tuberculosis [22] or membrane-associated proteins from M. bovis [23]. In this study, approximate 9758.1 mg of crude membrane-protein with the GlcDH activity 4730.6 U was extracted from 8.910 g of A. globiformis C224 cells at 0 °C with the ultrasonic and 1% Triton X-114 assistance (Table 1). With the increase of storage temperature to 37 °C, the extracted solution was partitioned into Triton X-114 and aqueous phases. After centrifugation, 6004.5 mg of membrane-protein was recovered from Triton X-114 phase with the increased GlcDH specific activity of 0.7 U/mg and 1.5-fold purification. As shown in Table 1, the consecutive precipitation steps with acetone, polyethylene glycol 6000 and ethanol yielded 33.6 mg of purified GlcDH with the increased specific activity of 7.9 U/mg and 16.6-fold purification. Figure 1 presents the hydroxyapatite column chromatography profile. The fraction eluted with buffer D had the highest GlcDH activity of 10.62 U/mL. The column purification yielded 1.8 mg of purified GlcDH with a maximum specific activity of 159.4 U/mg and 183.5-fold purification, which were higher than GlcDH from G. suboxydans with a specific activity 35.0 U/mg and 23.3-fold purification [12].

Type Identification of GlcDH from A. globiformis C224
The purified GlcDH showed a single band on SDS-PAGE which indicated that it had the high purity without subunits (Figure 2A). The molecular weight of GlcDH in SDS-PAGE was estimated to be 87 kDa. The GlcDHs purified from E. coli [11] and Pseudomonas sp. [14]on the SDS-PAGE had one band, while others from B. thuringiensis [13] and A. calcoaceticus [15] had two subunits.
Partial sequences of purified GlcDH from A. globiformis C224 were further analyzed by applying the protein band in the SDS-PAGE for MALDI-TOF-MS and NCBI BLAST search. The mass spectrum of GlcDH peptides is shown in Figure 2B. The similar sequences in the database of the NCBI blast, predicted molecular mass and score of the GlcDH are shown in Table 2. The molecular mass of GlcDH was determined as approximately 87 kDa, which agreed with that obtained from SDS-PAGE. From Table 2, the GlcDH from A. globiformis C224 had the similar amino acid sequence to GlcDHs from Ps. entomophila L48 and Ps. putida GB-1 with the high scores of 142 and 114, respectively. The substrate specificity of the crude membrane protein and purified GlcDH from A. globiformis C224 was summarized in Table 3. Purified GlcDH showed a broad substrate specificity with different monosaccharide including D-glucose, maltose, D-galactose, D-xylose or D-arabinose as the substrates. However, compared with the catalytic capacity using glucose as the substrate, other substrates found to be catalytic active with the A. globiformis GlcDH were maltose (17%), D-galactose (20%), D-xylose (22%) and D-arabinose (8.0%), thus indicating that D-glucose is the preferred substrate of A. globiformis GlcDH. Similar results also were found that the GlcDH from G. suboxydans could catalyze D-glucose or maltose [12].

Kinetic Studies of GlcDH from A. globiformis C224
Single Substrate Kinetics Figure 3 shows the Lineweaver-Burk double reciprocal plotting profiles of D-glucose, D-xylose, maltose or D-galactose as substrates. Based on the Lineweaver-Burk double reciprocal plots, the Km and Vmax values for D-glucose, D-xylose, D-galactose and maltose were showed in Table 4, which verified the conclusion that the optimum substrate for A. globiformis GlcDH was D-glucose. Some GlcDHs purified from A. calcoaceticus and B. cereus had the Km values of 22 mM and 7 mM for glucose [15,24], which were significantly higher than those obtained in the present study.

Dual Substrates Kinetics
In the present study, N-methylphenazonium methyl sulfate (PMS) was used as the artificial electron acceptor to investigate the dual substrates kinetics mechanism of GlcDH. By changing the concentrations of glucose and PMS, the kinetic profiles by plotting the reciprocal of the initial velocity (1/v) against the reciprocal of one substrate (1/[S]) were obtained as shown as the straight lines in Figure 4. GlcDH showed ping-pong behavior due to the parallel plotted lines of initial reaction rates. Similar ping-pong kinetic mechanisms were also observed in the immobilized lipase [25] and tRNA-guanine transglycosylase from E. coli [26]. By calculation, the Km values for glucose and PMS were estimated as 9.2 mM and 0.4 mM and Vmax was 140.11 μmol/mg·min, respectively. The affinity of GlcDH towards PMS seems to be higher than that towards glucose since Km(PMS) of 0.4 mM was significantly lower than Km(glucose) of 9.2 mM. Hence, the catalytic reaction began from the bind of PMS to A. globiformis GlcDH, and then transferred the O2 to the bound glucose to produce gluconic acid. and 11.4 mM for PMS, respectively. According to the lower Km(PMS) value, it could be conformed that the higher affinity of the GlcDH towards PMS than that to glucose. Generally, the catalytic product can affect the activity of enzymes by competitive, noncompetitive or uncompetitive ways. The competitive model had the lines converging at the same point on the y-axis on the Lineweaver-Burk double reciprocal plots, noncompetitive model had the lines converging at the same point but not on the y-axis, while typical uncompetitive model had a series of parallel lines. As shown in Figure 5, the Lineweaver-Burk double reciprocal plots of A. globiformis GlcDH were the typical parallel with glucose as the substrate lines and converging lines with PMS as the substrate when gluconic acid was used as the inhibitor at the concentration from 0 to 200 mM. Hence, it could be concluded that the gluconic acid was anon-competitive inhibitor to glucose and a competitive inhibitor to PMS.  Figure 6A,B show the influence of pH on the GlcDH activity and stability. From Figure 6A, the highest GlcDH activity appeared at pH 5.0. The GlcDH relative activity decreased to approximately 51% or 36% when the pH dropped to 4.0 or increased to 6.0. Other GlcDHs from B. thuringiensis M15 [13], G. suboxydans [12], Pseudomonas sp. [14] and E. coli [11] had the different optimal pHs with the values of 8.0, 3.0, 4.5 and 7.0 (Table 5), respectively. Figure 6B also demonstrated that the purified A. globiformis C224 GlcDH had the stable activity at pH values between 6.0 and 7.0 after two days storage at 4 °C, and decreased to near zero at pH 4.0 and 9.0.

Effect of Temperature on GlcDH Activity and Stability
Eight reaction temperatures ranging from 15 °C to 85 °C were used for investigating its effect on A. globiformis GlcDH activity. In Figure 6C, GlcDH reached to the maximum activity (relative activity of 100%) with the increase of reaction temperature from 15 to 45 °C, and followed the sharp decline of activity to 30% at 55 °C, and even no activity at 85 °C. Hence, a temperature of 45 °C could be selected for maximize the A. globiformis C224 GlcDH-catalyzing reaction efficiency. Other GlcDHs such as B. thuringiensis GlcDH had the higher reaction temperature of 55 °C to catalyze glucose to produce gluconic acid [13].
The purified A. globiformis GlcDH was kept at temperatures from 20 to 70 °C for 10-120 min. As shown in Figure 6D, low temperature of below 40 °C had the stable and highest GlcDH activity after kept for 120 min. Increase of storage temperature to over 50 °C inhibited the GlcDH activity of above 58% after even 10 min treatment. No GlcDH activity was detected after incubation of GlcDH for 60 min at 60 or 70 °C.

Effect of Organic Solvents on GlcDH Activity
Organic solvents including methanol, ethanol, acetone or n-hexane at the concentrations of 20%, 40%, 60% and 80% (v/v) were used to investigate their influence on the purified GlcDH from A. globiformis C224. As shown in Figure 7A, all the tested organic solvents showed various degrees of inhibition of GlcDH activity. At 20% (v/v) concentration methanol had a significant inhibitory effect with 80.9% of GlcDH relative activity remaining after treatment. With the increase of concentration to 80% (v/v), ethanol showed the highest inhibition on the GlcDH with 3.6% of relative activity remaining. Compared to other solvents, n-hexane had the lowest influence on the GlcDH activity, and even stimulated the GlcDH activity from 83.3% to 92.1% at the concentrations from 40% (v/v) to 80% (v/v). The 3D conformation changes and the blocked catalytic activity center in GlcDH are possible causes for this catalytic activity inhibition in these organic solvents. Figure 7B presents the effect of metal ions such as Mg 2+ , Zn 2+ , Mn 2+ , Cu 2+ , Fe 3+ , Ca 2+ or the ligand EDTA on the activity of purified A. globiformis C224 GlcDH. Mg 2+ had an enhancing effect on the purified GlcDH activity with an increase of about 20%. Zn 2+ , Mn 2+ or Ca 2+ at low concentrations of 0.5, 1 and 10 mM had a slight negative effect on the GlcDH activity with inhibition rates of about 3% and 10%, while Cu 2+ or Fe 3+ at concentrations of over 10 mM almost completely inhibited the GlcDH activity. EDTA at a concentration of 0.5 mM also completely inhibited the A. globiformis GlcDH activity. Therefore, Mg 2+ is possibly a cofactor on A. globiformis GlcDH activity.

Preparation of Cell Culture
The slant stock culture was diluted with sterilized water, inoculated into 50 mL of seed medium in a 500 mL Erlenmeyer flask and incubated at a 30 °C for 20 h. This culture at the concentration of 2% (v/v) was inoculated into a 5-L fermenter (GRCB, Green Bio-engineering Co., Ltd, Zhenjiang, China) with working volume of 3 L to prepare the A. globiformis cells. Cells were harvested when the OD650 value of broth reached to 0.65-0.70 (diluted 20-fold with 1 M HCl). The broth was centrifuged at 10,000× g, 4 °C for 20 min to collect A. globiformis cells. The collected cells was washed with buffer A (10 mM potassium phosphate buffer containing 5 mM MgCl2·6H2O, pH 6.0), and then frozen at −20 °C until use.

Extraction and Purification of GlcDH from A. globiformis C224
Frozen cells were suspended in buffer B (10 mM Tris-HCl, containing 1% (w/v) Triton X-114 and 150 mM NaCl, pH 7.4), and extracted with ultrasonic disrupter (Bilon, Shanghai, China) for 60 min. The supernatant containing GlcDH was collected after centrifugation at 16,000× g, 4 °C for 30 min, and divided into the hydrophobic and hydrophilic fractions by Triton X-114 phase separation (incubating the supernatant for 10 min at 37 °C). The Triton X-114 phase (hydrophilic fractions) was obtained by centrifuging at 13,000× g, 25 °C for 5 min [17].
The obtained Triton X-114 phase was precipitated with five volumes of ice-cold acetone. The precipitate was dissolved in the buffer C (10 mM potassium phosphate buffer containing 5 mM MgCl2, 1% (w/v) of Triton X-100 and 1 M of potassium chloride, pH 6.0) [28], mixed with polyethylene glycol 6000 and centrifuged at 10,000× g for 20 min to collect the supernatant. The ethanol precipitate fraction was obtained by adding 40% (v/v) of ethanol, dissolved in the buffer D (10 mM sodium phosphate buffer containing 5 mM MgCl2 and 1% (w/v) of Triton X-100, pH 7.0), and then applied to a hydroxylapatite column (5 mL, BIO-RAD, Hercules, CA, USA) with the stepwise elution of the buffer D, buffer E (50 mM sodium phosphate buffer containing 5 mM MgCl2 and 1% (w/v) of Triton X-100, pH 7.0) and buffer F (100 mM sodium phosphate buffer containing 5 mM MgCl2 and 1% (w/v) of Triton X-100, pH 7.0). The elution fractions having GlcDH activity were collected, concentrated by ultrafiltration (Millipore, Hercules, MA, USA), and identified with SDS-PAGE as the purified sample.

Polyacrylamide Gel Electrophoresis
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out according to the method of Laemmli [29]. Briefly, purified GlcDH was dissolved in the distilled water and mixing with SDS-PAGE sample loading buffer containing DTT to reduce the protein. Then the GlcDH molecular weight was obtained by comparing and calculating its electrophoretic mobility with those of standard molecular weight markers (

MALDI-TOF-MS Analysis
The bands of GlcDH in SDS-PAGE were excised, rinsed three times with ultrapure water, and destained with 200 μL 25 mM NH4HCO3 solution containing 50% CH3CN (destaining solution). The gel slices were vacuum-dried, and digested at 37 °C overnight with trypsin. The trypsin hydrolysate containing peptide mixture was combined and freeze-dried for MALDI-TOF-MS analysis with AB 5800 (AB Sciex, Framingham, MA, USA). Spectra mass peaks with the range from 1000 to 2500 m/z were singled out for MS/MS analysis. The peptide mass finger-printings of GlcDH and MS/MS date were obtained and processed using Mass-Lynx V4.1 software (Waters, Milford, MA, USA), and subsequently were converted to PKL files by the Protein-Lynx 2.2.5 software (Waters). The PKL files were analyzed using the MASCOT search engine. The search parameters were confirmed as follows: database, NCBInr; enzyme, Trypsin; taxonomy, Homo sapiens; and allowance of one missed cleavage.

Assay of GlcDH Activity and Protein Determination
The GlcDH activity was measured with ferricyanide as electron acceptor by the method of Wood et al. [30]. The reaction mixture contained 10 µmol of potassium ferricyanide, 0.7 mL of Mcllvaine buffer (pH 3.0), 100 µmol of D-glucose and enzyme solution in a total volume of 1.0 mL. The reaction was ended by adding 0.5 mL of the ferric-Dupanol reagent and 3.5 mL of water, and measured OD660 nm value. One unit of GlcDH activity was defined as the amount of GlcDH catalyzing the oxidation of 1 µmol of D-glucose per min, and 4.0 of OD660 nm value was equal to 1 µmol of oxidized D-glucose. Protein content was determined by the Bradford Protein Assay Kit (Sangon Biotech) with bovine serum albumin as standard.

Substrate Specificity
To determine the substrate specificity of GlcDH, various substrates including D-glucose, D-gluconate, maltose, D-sorbose, D-galactose, D-mannose, D-fructose, D-arabinose, malic acid, sucrose, citric acid or D-xylose (all obtained from Sinopharm Chemical Reagent CO., Ltd.), were separately added to the reaction solution to the concentration of 33 mM.

Effect of pH, Temperature, Organic Solvents or Metal Ions on GlcDH Activity
In order to investigate the effect of pH on the activity of purified GlcDH, eight pHs in the range of 3.0-10.0 were used with the Mcllvaine buffer. The pH stability of GlcDH was determined by incubating the enzyme in the pH range from 3.0 to 10.0 for 48 h at 4 °C and measuring the residual activity.
The optimum catalytic temperature of GlcDH was selected by determining the enzyme activity in the range of 15-85 °C. The temperature stability was determined by measuring the residual activity at various temperatures (20-70 °C) in buffer A for 120 min without adding substrate.
In order to determine the effect of organic solvents on GlcDH activity, methanol, ethanol, acetone or n-hexane was used, organic solvents were added to the reaction solution to the concentration of 20%, 40%, 60% and 80% (v/v), respectively. For assaying the effect of metal ions or metal chelating reagent on GlcDH activity, MgC12, CuCl2, ZnCl2, Fe2(SO4)3, MnSO4 or CaC12 at the concentrations of 0.5, 1, 10 and 50 mM were used. The metal chelating reagent EDTA at the concentrations of 0.5, 1, 10 and 50 mM were added and incubated for 60 min at 4 °C to investigate its influence on GlcDH activity.

Statistical Analysis
Each experiment was repeated three times using duplicate samples. The results were expressed as means ± standard deviations. Statistical comparisons were made by one-way analysis of variance (ANOVA), followed by Duncan's multiple-comparison test. Differences were considered significant when the p-values were <0.05.

Conclusions
This is the first report on the purification and characterization of a GlcDH from the 2KGlcA industrial producer A. globiformis C224. The purified GlcDH showed a molecular weight of 87 kDa and specific activity of 88.1 U/mg. The GlcDH followed a ping-pong mechanism and maintained the highest activity at pH 5.0 and 45 °C. Organic solvents, some metal ions or EDTA can adversely affect GlcDH activity. Our findings to elucidate the structural/enzymic characteristics of A. globiformis GlcDH provided us with new insights into improving 2KGlcA industrial stains used in the erythorbic acid industry. Further studies are underway in our laboratory to improve GlcDH properties such as transformation rate, reduce the production period and to find a wider range of production temperatures.