Biochemical properties of L-arabinose isomerase from Clostridium hylemonae to produce D-tagatose as a functional sweetener

d-Tagatose has gained substantial interest due to its potential functionalities as a sucrose substitute. In this study, the gene araA, encoding l-arabinose isomerase (l-AI) from Clostridium hylemonae (DSM 15053), was cloned and expressed in Escherichia coli BL21 (DE3). This gene consists of 1,506 nucleotides and encodes a protein of 501 amino acid residues with a calculated molecular mass of 56,554 Da. Since l-AI was expressed as an intracellular inclusion body, this enzyme was solubilized with guanidine hydrochloride, refolded, and activated with a descending concentration gradient of urea. The purified enzyme exhibited the greatest activity at 50°C, pH 7–7.5, and required 1 mM of Mg2+ as a cofactor. Notably, the catalytic efficiency (3.69 mM-1sec-1) of l-AI from C. hylemonae on galactose was significantly greater than that of other previously reported enzymes. The bioconversion yield of d-tagatose using the C. hylemonae l-arabinose isomerase at 60°C reached approximately 46% from 10 mM of d-galactose after 2 h. From these results, it is suggested that the l-arabinose isomerase from C. hylemonae could be utilized as a potential enzyme for d-tagatose production due to its high conversion yield at an industrially competitive temperature.


Introduction
D-Tagatose, a rare natural sweetener, is an isomer of aldohexose D-galactose and a C-4'-epimer (stereoisomer) of D-fructose. The taste of tagatose is closely equivalent to that of sucrose with no cooling effect or aftertaste, and has a relative sweetness value of 92% when compared to that of 10% solutions. This sugar has a significantly lower caloric value of 1.5 kcal/g, and the consumption of tagatose results in a reduced energy intake as well as promotion of weight loss at medically desirable rates [1,2], which has also been used in the treatment of obesity [3,4]. Owing to its lack of a glycemic effect [5], D-tagatose can be safely consumed by diabetic patients [6]. Several intensive studies have been conducted to investigate the different aspects of D-tagatose in food and pharmaceutical formulations, such as the stability of D-tagatose in food and drinks [7,8], clinical trials for pharmaceutical applications [9], and consumer evaluations [10].
The production of D-tagatose from D-galactose can be achieved through chemical and enzymatic methods. However, the chemical method is limited for industrial applications because of the use of calcium as a catalyst, complicated processes, and the formation of chemical waste and unexpected by-products [11]. The production of D-tagatose by enzymatic methods is regarded as an environmentally friendly protocol, and has thus been studied more intensively in recent years. L-Arabinose isomerase (L-AI; EC 5.3.1.4) is the most effective biocatalyst for the isomerization of D-galactose to D-tagatose, which is an intracellular enzyme catalyzing the reversible isomerization of L-arabinose to L-ribulose and D-galactose to D-tagatose [12]. L-AIs originating from various microorganisms have been identified and biochemically characterized, including Escherichia coli [13,14], Lactobacillus plantarum [15], Bacillus subtilis [16], Lactobacillus reuteri [17], Bacillus licheniformis [18], Geobacillus thermodenitrificans [11], Thermoanaerobacterium saccharolyticum [19], and Thermotoga maritima [20]. The ideal enzyme for production of D-tagatose in the food industry requires high thermostability and a weakly acidic pH optimum [5,21]. The L-AI reaction at high temperature shifts the bioconversion equilibrium to D-tagatose production with an increasing reaction rate between D-galactose and D-tagatose, thereby decreasing the viscosity of the reaction mixtures and reducing the degree of microbial contamination [19,22]. However, the thermophilic and hyperthermophilic L-AIs from various sources mainly present a slightly alkaline pH optimal range around 7-8.5 [18,23], and show low activity at acidic pH conditions [24]. The mesophilic and thermophilic L-AIs also require Mn 2+ and/or Co 2+ as cofactors to enhance their isomerization activity and to maintain thermostability [23]. Moreover, ion-independent and Mn 2+ -dependent L-AIs are more favorable for the food industry owing to the toxicity of Co 2+ [20,23].
The objective of this study was to investigate a novel L-AI from Clostridium hylemonae DSM 1505, and its potential application in the environmentally friendly production of D-tagatose. The gene araA encoding L-AI from C. hylemonae DSM 15053 was cloned and expressed in Escherichia coli BL21 (DE3). The recombinant enzyme was purified and its biochemical characteristics were determined to optimize the production of D-tagatose from D-galactose by L-AI, as a highly valuable sweetener. Cloning of the araA gene from C. hylemonae C. hylemonae (DSM 15053) was cultured in brain heart infusion broth (BD, Sparks, MD, USA) under anaerobic conditions (Hungate tubes; N 2 ) at 37˚C for 24 h. The harvest was used directly as a template DNA for polymerase chain reaction (PCR). The full-length nucleotide sequence of the araA gene was obtained with two oligonucleotides designed as PCR primers, 5 0 -GAGACAGGATCCATGATAAAAAGCAAAGAA-3 0 (Forward) and 5 0 -ACATTCCTCGAGCT AAATCCCCAGCTTATAGGC-0 (Reverse), containing restriction sites of BamHI and XhoI (underlined). The PCR was performed by iQ™5 Multicolor Real-Time PCR Detection Systems (Bio-Rad, Hercules, CA), and the amplified DNA fragment (1.5 kb) was ligated into the pGEM-T vector. The obtained araA sequence was compared with the original araA sequence in the National Center for Biotechnology Information (NCBI) database. The target gene, L-AI from C. hylemonae (ChAI), was then subcloned into the expression vector pET-28a(+) with the NdeI and HindIII restriction sites. The sub-cloned pET28a-ChAI was transformed into E. coli BL21 (DE3) for protein expression.

Refolding and purification of the recombinant L-AI
After isopropyl-β-D-thiogalactoside induction at 16˚C for 20 h, the cells were collected by centrifugation at 5,500 ×g at 4˚C for 10 min, and

Enzyme activity assay
The L-AI activity was measured according to the D-tagatose forming property using 10 mM of D-galactose, 25 mM of Tris-HCl buffer (pH 7.5), 1 mM of MnCl 2 , and 0.1 mg of the purified enzyme. The enzyme reaction was carried out at 35˚C for 30 min, and was stopped by boiling for 10 min. After boiling, the mixture was cooled and centrifuged at 13,000 rpm for 5 min to remove the precipitant. The amount of D-tagatose produced was determined using high-performance liquid chromatography (HPLC) with a refractive index detector (RID) connected to a Sugar-Pak I analytical column (6.5 × 300 mm; Waters Corp., Milford, MA, USA). The column was eluted by de-ionized water at 78˚C at a flow rate of 0.5 mL/min. One unit of L-AI activity was defined as the amount of enzyme producing 1 μmol of D-tagatose per minute at 35˚C and pH 7.5. The concentration of protein was calculated following Bradford's method using bovine serum albumin as the standard [26].

Effect of metal ions on enzyme activity
The effects of metallic ions on L-AI activity were determined by testing various metal ions of MnCl 2 , CoCl 2 , ZnCl 2 , MgCl 2 , CuCl 2 , and CaCl 2 . The purified enzyme was treated with 10 mM of ethylenediaminetetraacetic acid at 4˚C for 24 h to make the enzyme metal-free. The enzyme reactions were then carried out in the absence and presence of 1 mM divalent cations at 35˚C for 30 min. The effect of the metal ion was determined using the same enzyme assay described above.

Enzyme kinetics
The kinetic parameters of L-AI were determined using D-galactose as a substrate at eight different concentrations (30-600 mM) with the purified enzyme (0.1 mg) in 25 mM Tris-HCl buffer (pH 7.5) and 1 mM of Mg 2+ . The enzyme reactions were conducted at 50˚C for 15 min. The amount of keto-sugar formed was determined by HPLC. Based on these reactions, the initial rates of the bioconversion of D-galactose were identified. The kinetic parameters (K m and k cat ) were investigated via nonlinear regression analysis with SigmaPlot 13.0 software (Systat Software Inc.; San Jose, CA, USA). All assays were carried out in duplicate.

Enzymatic conversion of D-galactose to D-tagatose
The production of D-tagatose using 10 mM of D-galactose as a substrate by ChAI was conducted in 25 mM Tris-HCl buffer (pH 7.5) with 1 mM Mg 2+ and 1 mg/mL of purified ChAI at 50-70˚C for 10 h. Samples were drawn at the defined time and the amount of produced D-tagatose was measured by HPLC.

Statistical analysis
All data are expressed as the mean ± standard deviation. Significant differences among the treatments were determined by one-way analysis of variance in IBM SPSS Statistics for Windows (version 21.0, IBM Corp., Armonk, NY). Statistical significance was indicated at a confidence level of 95%.

Cloning and expression of the ChAI gene
Based on the fully sequenced genomic DNA of C. hylemonae (NCBI accession number: NZ_GG657759.1), the araA gene encoding putative L-AI was successfully isolated and amplified by PCR. The amplified DNA fragment of about 1.5 kb was confirmed and compared with the original araA sequence in the NCBI database. The L-AI-encoding sequence was successfully constructed in the recombinant expression vector pET28a-ChAI (Fig 1A). The recombinant L-AI expressed in E. coli formed an insoluble inclusion body, which was solubilized in guanidine HCl (6 M) resulting in~72% dissolution of the aggregated recombinant L-AI ( Table 1). The solubilized recombinant L-AI was refolded and purified through the Ni-NTA affinity chromatography column. Guanidine HCl-treated insoluble fraction of ChAI showed two distinct protein bands on the SDS-PAGE gel, in which the molecular weights were estimated to be 57 kDa for fully denatured soluble ChAI as shown in the lane 2 and 37 kDa for partially denatured enzyme (Fig 1B). Depending the denaturation conditions, it has been known that degree of protein denaturation might be different [27]. Incomplete denaturation by SDS treatment has also been reported for some thermostable enzymes [28]. The purified L-AI exhibited a specific activity of 0.44 U/mg at 35˚C with D-galactose as a substrate.

Effect of temperature on enzyme activity and stability
The temperature profile for ChAI activity is shown in Fig 2. The optimal temperature of ChAI was 50˚C when D-galactose was used as a substrate (Fig 2A). The residual activity of ChAI was approximately 81% and 38% at 65˚C and at 90˚C, respectively. The thermostability of ChAI was investigated by measuring the enzyme activity over a temperature range from 40˚C to 70˚C for 150 min (Fig 2B). ChAI displayed remarkable stability with 90% and 85% of the initial activity preserved at 45˚C and 60˚C, respectively. However, there was a considerable decrease in its activity along with the incubation time at over 65˚C. The enzyme deactivation constant (k d ) value for the thermal inactivation of ChAI was calculated to be 0.486 s -1 and the half-life (t 1/2 ) was 85.6 min at 70˚C (Fig 2C).

Effect of pH on enzyme activity and stability
The stable pH profile for the ChAI activity of D-tagatose production exhibited quite a wide range, although the optimal pH was determined to be 7.5. Interestingly, ChAI could maintain its high-level activity within the acidic pH range (3.0-5.5) as well as within the alkaline pH range (8.0-9.0), with over 80% of its maximum activity retained (Fig 3A). In addition, the pH stability profile for recombinant ChAI showed that the enzyme was most stable at pH 6.5-7. It was also stable at a slightly acidic pH, with about 92% relative activity at pH 6.0 ( Fig 3B).

Effect of metal ions on enzyme activity
The ChAI activity was measured with different types of divalent metal ions (1 mM), and no significant effect on tagatose production was detected by each of Ni + , Zn 2+ , Ca 2+ , and Cu 2+ ions ( Table 2). Several studies have reported that Cu 2+ inhibits L-AI activity; [11,18,24,29] however, this ion significantly enhanced the activity of ChAI. In addition, the enzyme activity    was increased by 20% of the relative activity compared to its initial activity in the presence of Mg 2+ . Moreover, 1 mM of Mg 2+ was found to be the optimal concentration to most effectively enhance the activity of ChAI within the concentration range tested of 0.2 to 2.0 mM (S1 Fig).

Kinetic parameters
ChAI followed a Michaelis-Menten (MM) equation for D-tagatose production from D-galactose (S2 Fig). The apparent MM constant (K m ) was 7.70 mM and the apparent catalytic constant (k cat ) was 28.39 sec -1 . The kinetics analysis showed that ChAI exhibits higher substrate affinity toward D-galactose compared to L-AIs originating from other microorganisms ( Table 3).

Conversion of D-galactose to D-tagatose using ChAI
The bioconversion of D-galactose into D-tagatose using ChAI was conducted at 50-70˚C. The amount of D-tagatose reached equilibrium after 2-h incubation (S3 Fig). Interestingly, the production yields of D-tagatose increased from 36.1% to 45.9% as the reaction temperature increased from 50˚C to 60˚C, and then decreased to 34.4% with a further temperature increase to 70˚C after 10 h of the reaction (Fig 4). Geobacillus thermodenitrificans 408 3.40 0.008 [11] Thermotoga neapolitana 250 13.50 0.054 [32] Discussion A novel L-AI from C. hylemonae was cloned, and its biochemical characteristics were determined focusing on high-value D-tagatose production by isomerization. The newly identified enzyme showed optimal activity at 50˚C, pH 7.5, and required Mg 2+ for enhancing its D-tagatose production property. Moreover, the enzyme displayed remarkable thermostability at 60˚C and pH 6. D-Tagatose production by ChAI resulted in a high conversion yield of approximately 46% at 60˚C. For bio-industrial applications, D-tagatose production must be carried out at temperatures over 60˚C, which can increase the bioconversion yield, reduce the viscosity of the reaction mixture, and decrease the possibility of microbial contamination [19,20,22,30]. However, enzyme reactions above 70˚C may result in undesirable reactions such as the formation of unexpected byproducts and a browning reaction [22]. Although ChAI showed the greatest activity at 50˚C, considerable levels of relative activity (93% at 60 min) and thermostability (85% after 150 min) remained at 60˚C. Interestingly, the MM constant (K m ) of ChAI was lower and the turnover number (k cat ) was higher than any other L-AIs reported previously ( Table 3). The greatest catalytic efficacy (k cat /K m : 3.69 s -1 mM -1 ) of ChAI for tagatose production suggests that ChAI would be a prospective candidate for industrial application at a competitive temperature and production rate.
Although the optimal pH of ChAI activity for D-tagatose production was 7.0, it was still remarkably stable over a wide range of pH 3.0-10.0. The acid tolerance of L-AI is an important factor with respect to isomerization under acidic conditions to prevent the formation of inevitable byproducts and other environmental concerns [31]. L-AI from C. hylemonae maintained high activity and was sufficiently stable at pH 6.0 (~92% activity), which satisfy the industrial requirements of D-tagatose production as a green technology.
The D-tagatose production by ChAI at 50˚C was ca. 35%, which is almost identical to the previous result for a mesophilic L-AI. The reaction toward D-galactose is more dominant in an equilibrium reaction between D-galactose and D-tagatose from a mesophilic L-AI, resulting in a low conversion ratio of D-tagatose [11]. It was previously reported that the bioconversion equilibrium between D-galactose and D-tagatose shifted preferentially toward D-tagatose at higher temperature [20]. In particular, the conversion yield of D-galactose to D-tagatose using the L-AI from E. coli was less than 30% at 37˚C [32,33]. However, in the present study, the production yield of D-tagatose reached approximately 46% at 60˚C. This specific enzyme displays the greatest conversion catalytic property as a mesophilic L-AI for D-tagatose production reported to date. Thermotoga neapolitana L-AI produced D-tagatose from 10 mM of D-galactose with yields of 68% at 80˚C, while only 22% of D-tagatose was obtained at 50˚C after 20 h [34]. It should be pointed out that within a relatively short reaction time of 6 h, this Thermotoga strain showed a bioconversion yield of 56% by the isomerization of D-galactose to D-tagatose at a very high temperature of 80˚C [20]. Thus, hyperthermophilic L-AI produced D-tagatose from D-galactose effectively due to an equilibrium shift toward D-tagatose at high reaction temperature [11]. However, the browning reaction above 70˚C will negatively affect the final products. Therefore, for commercial production, it may be preferable to utilize L-AI s that act well at around 60˚C to limit undesirable color formation. It was previously reported that thermophilic L-AIs exhibited higher conversion yields than hyperthermophilic L-AI s at this temperature as well [34,35]. Thus, considering the inherent biochemical properties, C. hylemonae L-AI could be applied as a potential enzyme for D-tagatose production by improving the protein expression systems (e.g., expression vectors or host) for commercial application.