The Characterization of a Novel D-allulose 3-Epimerase from Blautia produca and Its Application in D-allulose Production

D-allulose is a natural rare sugar with important physiological properties that is used in food, health care items, and even the pharmaceutical industry. In the current study, a novel D-allulose 3-epimerase gene (Bp-DAE) from the probiotic strain Blautia produca was discovered for the production and characterization of an enzyme known as Bp-DAE that can epimerize D-fructose into D-allulose. Bp-DAE was strictly dependent on metals (Mn2+ and Co2+), and the addition of 1 mM of Mn2+ could enhance the half-life of Bp-DAE at 55 °C from 60 to 180 min. It exhibited optimal activity in a pH of 8 and 55 °C, and the Km values of Bp-DAE for the different substrates D-fructose and D-allulose were 235.7 and 150.7 mM, respectively. Bp-DAE was used for the transformation from 500 g/L D-fructose to 150 g/L D-allulose and exhibited a 30% of conversion yield during biotransformation. Furthermore, it was possible to employ the food-grade microbial species Bacillus subtilis for the production of D-allulose using a technique of whole-cell catalysis to circumvent the laborious process of enzyme purification and to obtain a more stable biocatalyst. This method also yields a 30% conversion yield.


Introduction
Recently, different risk factors such as obesity, hypertension, and diabetes have been rising quickly around the world. The widespread occurrence of these chronic illnesses is strongly correlated with the excessive consumption of meals high in simple carbs and fat [1]. Therefore, in the field of food, nutrition, health care, and pharmacology, caloriefree and low-calorie sweetener alternatives such as rare sugars are gaining increasing attention [2]. Monosaccharides and their derivatives, which are infrequent, are referred to as rare sugars. D-allulose, commonly known as D-psicose belongs to a naturally rare sugar rather than an artificial sweetener, is one of the most significant rare sugars that has been the focus of most research. Several years ago, the US of Food and Drug Administration (FDA) identified it as generally regarded as safe (GRAS) and could be used as a component of food or food additives products [3,4]. The taste, functionality, and sweetness of Dallulose are essentially identical to those of sucrose, and it has little caloric content [5]. Furthermore, D-allulose possesses several physiological functions, such as the prevention of obesity and diabetes [6,7], treatment of hypoglycemia [8,9], metabolism of fat [10,11], and enhancement of antioxidant activity [12]. Additionally, it has great promise in the field of food, agriculture, and medicine as well as health and fitness.
Due to its rarity in nature, its large-scale application presents significant challenges and gains more attention from researchers. The previous studies elaborate that D-allulose was produced from a chemical method in the 1960s but it featured several unexpected Consequently, it is necessary to excavate new enzymes that possess intrinsic properties according to the industrial demands. Additionally, the screening of sequence-based metagenomic library can be used to identify relevant DTE family enzymes [20]. Due to the availability of an abundance of sequence data, it is feasible to discover new DTEs from safe and little investigated sources and even extreme environments. In this work, one unique putative DAE (Bp-DAE) from the Blautia produca genome was discovered using a sequence similarity by blast tool from NCBI. The putative gene of DAE was cloned in the pANY1 vector, expressed in Escherichia coli BL21 (DE3), then purified. The biochemical characterization, including optimal temperature, pH, metal ion, thermostability, molecular mass, and general properties such as substrate specificity and kinetic parameters, was investigated. By conducting molecular docking with different substrates closely linked residues were identified. Furthermore, considering the safety of the strain [38], we used the Bacillus subtilis as the host to express Bp-DAE and the recombinant Bacillus subtilis cells were employed for the biosynthesis of D-allulose from D-fructose. The experimental results elaborate that Bp-DAE could transform D-fructose into D-allulose which could be used to reduce the cost of downstream industries.

Strains, Media, and Materials
The pANY1 vector was presented by the Shenyang Agricultural University, China. Escherichia coli DH5α and the host strain, Escherichia coli BL21 (DE3), were purchased from Stratagene (La Jolla, CA, USA). The shuttle vector, pP43NMK, and the host strain, Bacillus subtilis WB800N, were purchased from Fenghui Biotechnology Co., Ltd., (Hunan, China).
The cloning host, Escherichia coli DH5α, was grown at 37 • C and 220 rpm for 12 h in Luria-Bertani (LB) medium supplemented with Kanamycin (50 µg/mL). And the expression host strain, Escherichia coli BL21 (DE3), was cultivated in LB medium containing 50 µg/mL Kanamycin at 37 • C for 3~4 h and then supplemented with IPTG (1 mmol/mL) at 25 • C for 12 h. Bacillus subtilis WB800N was grown at 37 • C and 220 rpm for 24 h in Terrific Broth (TB) medium containing 50 µg/mL erythromycin.
The Ni 2+ -NTA resin was procured from Sangon Biotech Co., Ltd. (Shanghai, China). The main chemical reagents such as D-allulose, D-fructose, erythromycin, kanamycin, and isopropyl β-D-1-thiogalactopyranoside (IPTG), etc. were purchased from TCI (Tokyo, Japan), Aladdin (Shanghai, China), Macklin (Shanghai, China) and the Sinopharm Chemical Reagent (Beijing, China). Plasmid extraction and gene purification kits, the PCR required high fidelity enzymes, restriction endonuclease, and protein marker etc. were purchased from Vazyme (Nanjing, China). All of these chemicals involved in the reaction were analytical grade unless otherwise specified.

Gene Mining and Phylogenetic Analysis
According to the reported DTEs, Pseudomonas cichorii ST-24 DTE (Genebank ID: BAA24429.1) was used as the probe during the selection of novel DTEs in the NCBI database using the Blast tool. We obtained the potential DAE protein sequence originating from Blautia produca (NCBI accession number: WP_148391986.1). The phylogenetic tree of the different DTEs and DAEs was constructed by MEGA 11. The multiple sequences alignment of Bp-DAE was performed by ClustalX and ESPript.

Gene Cloning and Expression
The gene encoding Bp-DAE was ligated with 6×His-tag and optimized according to the E. coli codon preference, chemically synthesized as well as inserted between the Pst I and BamH I endonuclease sites of the expression vector pANY1. The recombinant plasmid, named as pANY1-Bp-dae and was then transformed into E. coli BL21 (DE3) cells in order to check the expression and its expression level.
The above-mentioned E. coli strains were cultivated in 0.3 L LB medium supplemented with 50 µg/mL Kanamycin at 37 • C, 220 rpm. Add 1 mmol/mL IPTG into the medium as soon as the OD 600 reached 0.6~0.8, which takes about 3 to 4 h, and then the temperature was shifted to 25 • C, the shaking speed was shifted to 150 rpm, the induction was about to start and would last 12 h. After the induction phase, the harvested cultures were collected by centrifugation at 6000 rpm for 10 min at 4 • C. Then the cells collected by centrifugation were washed twice with 0.9% NaCl solution, and about 0.06 g of cells were resuspended in 15 mL of 50 mM Na 2 HPO 4 /NaH 2 PO 4 buffer (pH 7.0), ultrasonic crushing of it on ice for 15 min (500 W, 3 s working, 3 s stopping), centrifuged to collect the crude cell extract, and then filtered through 0.22 µm MCE filter membrane to obtain the crude enzyme that used for the next study and subsequent enzyme purification.

Purification of Enzyme and SDS-PAGE Analysis
First, loading 5 mL binding buffer (pH 7.0) that contained 50 mM Na 2 HPO 4 /NaH 2 PO 4 and 150 mM NaCl into the Chelating Sepharose Fast Flow resin column (1.0 cm × 10.0 cm) that contained Ni 2+ for equilibrating with the column. Then, loading the obtained crude enzyme into the column twice until the target protein is bound to the Ni 2+ column as much as possible. The unbound proteins were eluted away with 5 mL washing buffer (pH 7.0) that contained 50 mM Na 2 HPO 4 /NaH 2 PO 4 , 150 mM NaCl and 50 mM imidazole. Last, the bound protein BP-DAE was eluted by 5 mL elution buffer (pH 7.0) that contained 50 mM Na 2 HPO 4 /NaH 2 PO 4 , 150 mM NaCl and 500 mM imidazole. Then, 1.5 mL centrifuge tube was used to receive the eluate and changed to a new tube for every full 1 mL until finished. In general, the second and third tubes have the highest protein content. The protein purity was determined by a method, sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) analysis, as well as the determination of protein concentration was done by the Bradford method [39]. Thereafter, characterization experiments as well as determination of its kinetic and catalytic properties were performed with pure enzymes.

Enzyme Assay
The enzymatic reaction was conducted in 50 mM Na 2 HPO 4 /NaH 2 PO 4 buffer (pH 8.0), supplemented with 50 g/L D-fructose, a certain amount of enzyme or recombinant wholecell and 1 mM Mn 2+ in a final volume of 1 mL at 55 • C for 10 min. Placing the sample in boiling water (100 • C) for 10 min to terminate the reaction. Then the reaction solution was centrifuged at 12,000× g rpm for 10 min, the supernatant was filtered and taken for analysis by high performance liquid chromatography (HPLC). Where one unit (U) of enzyme activity was defined as the amount of enzyme or recombinant cells required to generate 1 µ mol D-allulose per minute at pH 8.0 and 55 • C [35].

Characteristics of Enzyme
Determining the influence of pH on enzyme activity entailed using different ranges of pH (pH 5 to 11) in different buffer systems, for instance, 50 mM sodium acetate (pH 5, and 5.5), 50 mM phosphate (pH 6.0 to 8.5), and 50 mM glycine-NaOH (pH 9, 10, and 11.0) buffers. The optimum performing enzyme activity was defined as 100%, the pH stability was determined through incubating the enzyme in the above combination of pH buffers at 4 • C for 12 h, with standard enzyme activity assays, and the initial activity was assumed to be 100% at each pH.
Similarly, the effect of temperature was measured using different temperatures from 40 to 80 • C in 50 mM Na 2 HPO 4 /NaH 2 PO 4 buffer (pH 8.0), where the best performing enzyme activity was defined as 100%; the temperature stability of the enzyme was examined through incubating the enzyme in 50 mM Na 2 HPO 4 /NaH 2 PO 4 buffer (pH 8.0) at 30 • C, 40 • C, 55 • C, 55 • C accompanied with Mn 2+ , and 60 • C for several hours followed by standard enzyme activity assays. An unincubated sample was used as the control and its activity was defined as 100%.
The effects were investigated of the presence of different metal ions (1 mM Co 2+ , Mn 2+ , Mg 2+ , Fe 2+ , Ni 2+ , Zn 2+ , Ca 2+, and Cu 2+ ) on the catalytic activity under optimal pH and temperature conditions. Furthermore, 10 mM EDTA was used to chelate metal ions to confirm whether metal ions are required for the catalytic reaction. A sample without added metal ions served as control, and its activity was defined as 100%.

Substrate Specificity and Kinetic Properties of Enzyme
The substrate specificity of Bp-DAE was determined, with standard enzyme activity assays, using a 300 mM concentration of different substrates such as D-fructose, D-allulose, D-tagatose, and D-sorbose; the optimum enzyme activity was defined as 100%.
The kinetic parameters of Bp-DAE were determined by measuring the activity using different range of substrate concentration (5 to 1500 mM), including D-fructose, D-allulose, D-tagatose, and D-sorbose. Kinetic parameters (such as K m , k cat and V max ) were calculated by fitting with Michaelis-Menten plot and the corresponding Lineweaver-Burk plot using GraphPad Prism software (Version 6.0, San Diego, CA, United States).

Protein Homology Modeling and Substrate Molecular Docking
The homology model of Bp-DAE was auto-created in accordance with the reported DAE crystal structure using the SWISS-MODEL server. https://swissmodel.expasy.org/ (accessed on 27 May 2022). The substrate molecules, D-fructose and D-allulose, were used as the ligand to dock the enzyme Bp-DAE with Auto Dock software to perform molecular docking. The PyMOL software was used for the 3-dimensional structural model visualization of the enzyme.

Enzymatic Production of D-allulose
Use of the purified Bp-DAE as biocatalysts for the production of D-allulose from D-fructose. The enzymatic reaction solution was performed in a total volume of 50 mL consisting of 100, 300, and 500 g/L D-fructose and 50 mg purified Bp-DAE (450 U) in 50 mM Na 2 HPO 4 /NaH 2 PO 4 buffer (pH 8.0) supplemented with 1 mM Mn 2+ of the metal ion. The reactions were run at 55 • C and 200 rpm for 14 h in a shaker, samples were taken every one hour and the conversion yield was determined by HPLC.

Transformation of Bacillus subtilis and Whole-Cell Preparation of D-allulose
Based on the synthesized gene encoding Bp-DAE, primers were designed to clone the gene and link it to shuttle plasmid pP43NMK by the SeamLess Cloning technique and named pP43NMK-Bp-dae. Transfer of recombinant plasmids into Bacillus subtilis WB800N and expression of Bp-DAE. The recombinant Bacillus subtilis cells were cultured at 37 • C and 220 rpm for 24 h in Terrific Broth (TB) medium contained 50 µg/mL erythromycin. Firstly, 2 mL of cultured bacterial solution were taken and measured its OD 600 and calculated the weight of the bacterial cells by fitting the curve of OD 600 to the cell dry weight. The remaining bacterial cells were collected by centrifugation at 6000 rpm for 10 min at 4 • C. Then the centrifuged collected cells were washed twice with 0.9% NaCl solution. Finally, resuspend the cells with the appropriate amount of 50 mM Na 2 HPO 4 /NaH 2 PO 4 buffer (pH 8.0) depending on the weight of the cells required for the reaction system and stored them at 4 • C for further use. As with the enzymatic method, the whole-cell catalytic reaction was also carried out in phosphate buffer (50 mM Na 2 HPO 4 /NaH 2 PO 4 buffer, pH 8.0) containing 500 g/L D-fructose, 20 g/L recombinant Bacillus subtilis cells, and 1 mM Mn 2+ in a shake flask with the final volume of 50 mL at 55 • C and 200 rpm.

Analysis of Products
The products involved in the reaction such as D-allulose, D-fructose, D-tagatose, and D-sorbose were analyzed by HPLC equipped with a refractive index detector (RID-20A) (Shimadzu, Japan) [40]. More details were in the filtration of the sample with a 0.22 µm MCE filter membrane prior to the sample analysis and the HPLC system was analyzed through a Ca 2+ -carbohydrate column (Hi-Plex-Ca, Agilent, Church Stretton, Shrops, UK) with its temperature of 84 • C, with deionized water as the mobile phase, and a flow rate of 0.6 mL/min. And each experiment was repeated three times and the date were analyzed and processed using GraphPad Prism software and presented as mean value ± standard deviation (SD).

Screening DAEs and Sequence Analysis of Bp-DAE
The industrial use of D-allulose enzymes did much to pave the way for the mining of novel DTE family enzymes. NCBI database was the unique source for discovering potential contenders of DTE family enzymes. According to the database, a few other enzymes such as transaminase [41], glucose isomerase [42], and dehydrogenase [43] are workable. From the required information of the NCBI database, the putative protein (protein ID: WP_148391986.1) was obtained for the extraction of DAE from Blautia produca with 298 amino acids and 897 base pairs of nucleotides having a theoretical molecular mass of 33.7 kDa. The putative DAE, named Bp-DAE had 41.87% sequence similarity to the Pseudomonas cichorii DTE [23]. It was interesting to note that when Blast was run using Bp-DAE protein as a probe, the sequence shared up to 60% similarity with a Lachnospiraceae DAE [44]. The blast results raised the probability that Bp-DAE belongs to the DAE family enzymes and the safety of the source of strain was not ignorable during the screening process. The human gut bacteria Blautia produca isolated from feces whose species belonging to the Lachnospiraceae family had rapidly attracted more attention for its capabilities to alleviate inflammatory, metabolic disorders and its antibacterial properties against particular microorganisms since its discovery [44,45]. It could be used as a probiotic in food, medicinal and culinary preparations [46].
To further evaluate whether Bp-DAE is a member of the DAE family enzymes, a phylogenetic tree was formed with it using known DTE and DAE protein sequences derived from different strains. The protein sequence-based phylogenetic analysis revealed that the Bp-DAE gene had a close relationship with the previously reported Lachnospiraceae DAE ( Figure 1). The multiple sequence alignment of Bp-DAE with the earlier reported DTE and DAE protein sequences is shown in Figure 2. After comparison, the sequence similarities were all above 60%. It exhibited that the catalytic residues including Glu151, Asp184, His210, and Glu245 are responsible for metal binding sites. The glutamate residues are also liable for the isomerization at the carbon position 3 of D-fructose. Some residues such as Glu157, Arg216, His 187, and Glu 245 were taking part in the substrate binding mechanism while Tyr7, Trp15, Trp113, and Phe247 residues provide a hydrophobic environment in the surroundings. These results were based on the comparison with the already discovered crystal structures of Pseudomonas cichorii DTE [47] and Agrobacterium tumefaciens DAE [48]. All of these results elaborate that Bp-DAE was a member of the DAE family enzymes and ability to epimerize D-fructose to D-allulose.

Heterologous Expression and Purification of Bp-DAE
The recombinant plasmid pANY1-Bp-dae was expressed into the host E. coli BL21 (DE3) as well as cultured overnight on the 50 µg/mL of kanamycin LB agar plates. A positive transformant was then selected for extended cultivation and successful induction of expression. Then the cells of recombinant bacterium E. coli BL21 (DE3)/pANY1-Bpdae were obtained, washed, disrupt with ultrasonication, centrifuged to obtain the crude enzyme, and purified. A single and clear protein band of about 34.5 kDa was obtained by SDS-PAGE ( Figure 3A). In the purified protein, a 6× His-tag tag was ligated with the encoded gene, which possessed a molecular mass of about 0.8 kDa. Therefore, the molecular mass of the purified protein was obtained as 33.7 kDa plus 0.8 kDa by SDS-PAGE analysis which become approximately 34.5 kDa and was consistent with the predicted molecular mass of Bp-DAE. Then the purified enzyme will be used for subsequent enzyme characterization. In addition, HPLC was used for the detection of the conversion from D-fructose to D-allulose to confirm the enzymatic functionality. Figure 3B shows the chromatogram of standard sugars D-fructose and D-allulose displaying 14 and 19.7 min of retention time respectively, and the reaction is also showing the same peaks on the same retention time confirming the capability of Bp-DAE to convert D-fructose to D-allulose.
were all above 60%. It exhibited that the catalytic residues including Glu151, Asp184, His210, and Glu245 are responsible for metal binding sites. The glutamate residues are also liable for the isomerization at the carbon position 3 of D-fructose. Some residues such as Glu157, Arg216, His 187, and Glu 245 were taking part in the substrate binding mechanism while Tyr7, Trp15, Trp113, and Phe247 residues provide a hydrophobic environment in the surroundings. These results were based on the comparison with the already discovered crystal structures of Pseudomonas cichorii DTE [47] and Agrobacterium tumefaciens DAE [48]. All of these results elaborate that Bp-DAE was a member of the DAE family enzymes and ability to epimerize D-fructose to D-allulose.

Characteristics of Bp-DAE
The influence of the presence of various metal ions on the catalytic activity of Bp-DAE presented by Figure 4. The results exhibited that Bp-DAE possessed the maximum epimerization of 202% and 195% with 1 mM Mn 2+ and Co 2+ respectively. Additionally, the presence of 1 mM Mg 2+ , Fe 2+ , and Ni 2+ slightly improved the activity of Bp-DAE but Ca 2+ , Zn 2+ , and Cu 2+ inhibited the enzyme activity. The addition of EDTA (10 mM) entirely reduced enzyme activity, which is interesting because Bp-DAE did not exhibit any activity in the absence of the metal cofactor, demonstrating that the enzyme was strictly a metal-dependent enzyme. The present results were similar to the previously characterized DTE family enzymes such as DaeM [35] and C. minuta DTE [20]. However, some members of DTEs or DAEs family enzyme were not strictly metal-dependent, e.g., Agrobacterium tumefaciens DAE [21], Pseudomonas cichorii DTE [23], and Sinorhizobium DTE [34], but the presence of Mn 2+ and Co 2+ significantly enhanced their enzyme activity. The presence of Mn 2+ and Co 2+ nearly always boosted enzyme activity, while Cu 2+ and Zn 2+ almost invariably decreased the enzyme activity of DTEs and DAEs, as shown in Table 1.
The effect of pH on the enzyme activity and its stability of Bp-DAE was determined at optimum temperature and different ranges of pH (5 to 11) were applied as shown in Figure 5A. Generally, a neutral to alkaline pH environment is more suitable for Bp-DAE activity. It was at optimal activity in 50 mM phosphate buffer (pH 8.0) showing 100% relative activity and maintaining above 80% of its relative activity at pH (7 to 11), suggesting that the enzyme might function in a relatively wide range of pH environments. However, the relative enzyme activity was decreased to 20% while the reaction was carried out in an acidic environment (pH 5 and 5.5) ( Figure 5A). According to Table 1, it could be concluded that DTEs and DAEs generally exhibited the highest activity under neutral and weakly alkaline conditions. As shown in Figure 5B, above 60% of activity was remained at pH 6.5 to 10 after 12 h of incubation at 4 • C. This result indicated that Bp-DAE could be more stable in neutral to alkaline pH conditions.   Amino acid multiple sequence alignment of Bp-DAE with known DAEs and DTEs. Sequence alignment results were performed using the software of Omega, ClustalX, and ESPript. The black wavy lines and arrows represented α-helix and β-strand, respectively. In all the sequences, amino acid residues that are identical are highlighted in red and those that are highly conserved and similar are framed in blue; the residues involved in the metal ligand sites, catalysis sites, and hydrophobic pocket forming were symbolized as red asterisks, blue dots, and black squares, respectively. mass of Bp-DAE. Then the purified enzyme will be used for subsequent enzyme characterization. In addition, HPLC was used for the detection of the conversion from D-fructose to D-allulose to confirm the enzymatic functionality. Figure 3B shows the chromatogram of standard sugars D-fructose and D-allulose displaying 14 and 19.7 min of retention time respectively, and the reaction is also showing the same peaks on the same retention time confirming the capability of Bp-DAE to convert D-fructose to D-allulose.

Characteristics of Bp-DAE
The influence of the presence of various metal ions on the catalytic activity of Bp-DAE presented by Figure 4. The results exhibited that Bp-DAE possessed the maximum epimerization of 202% and 195% with 1 mM Mn 2+ and Co 2+ respectively. Additionally, the presence of 1 mM Mg 2+ , Fe 2+ , and Ni 2+ slightly improved the activity of Bp-DAE but Ca 2+ , Zn 2+ , and Cu 2+ inhibited the enzyme activity. The addition of EDTA (10 mM) entirely reduced enzyme activity, which is interesting because Bp-DAE did not exhibit any activity in the absence of the metal cofactor, demonstrating that the enzyme was strictly a metal-    The effect of pH on the enzyme activity and its stability of Bp-DAE was determined at optimum temperature and different ranges of pH (5 to 11) were applied as shown in Figure 5A. Generally, a neutral to alkaline pH environment is more suitable for Bp-DAE activity. It was at optimal activity in 50 mM phosphate buffer (pH 8.0) showing 100% relative activity and maintaining above 80% of its relative activity at pH (7 to 11), suggesting that the enzyme might function in a relatively wide range of pH environments. However, the relative enzyme activity was decreased to 20% while the reaction was carried out in an acidic environment (pH 5 and 5.5) ( Figure 5A). According to Table 1, it could be concluded that DTEs and DAEs generally exhibited the highest activity under neutral and weakly alkaline conditions. As shown in Figure 5B, above 60% of activity was remained at pH 6.5 to 10 after 12 h of incubation at 4 °C. This result indicated that Bp-DAE could be more stable in neutral to alkaline pH conditions.
The effect of different ranges of temperature on the enzyme activity of Bp-DAE was The effect of different ranges of temperature on the enzyme activity of Bp-DAE was performed under optimum pH conditions and the results are sown in Figure 5C. The results elaborate that the highest Bp-DAE activity was shown at 55 • C, while at 50 • C, the enzyme possessed 80% of its relative activity. Bp-DAE abruptly decreased the activity with gradual increases in temperature; as shown in Figure 5C, at 70 and 80 • C the relative activity decreased to 35%. The temperature stability of Bp-DAE was measured through incubating the enzyme at different ranges of temperature (30 to 60 • C) for a period of time and the variation of activity was examined as shown in Figure 5D.

Relative activity (%)
hours of incubation at 60 °C. One of the key elements affecting the enzyme-catalyzed cess was temperature. In the industrial production of rare sugars, higher tempera allowed the enzymes to keep better activity because higher temperatures increased solubility of the substrate, which sped up the enzyme reaction, thus increasing the version yield and to some extent reducing microbial contamination [49,50]. However high a temperature might trigger the Maillard reaction and lead to the thermal dena tion of the enzyme, resulting in a bad quality of sugar and low yields [49,50].

Substrate Specificity and Kinetic Properties of Bp-DAE
The substrate specificity and kinetic properties of enzymes were examined to in tigate the applicability of Bp-DAE in rare sugar production. Four kinds of the subs such as D-fructose, D-allulose, D-sorbose, and D-tagatose were employed to inspec catalytic activity of Bp-DAE and showed 38.39%, 43.39%, 71.44%, and 100% of rel activities against each substrate respectively ( Figure 6). Interestingly, earlier repo DAEs and DTEs that specifically favor D-allulose or D-tagatose, similarly, Bp-DAE also a high catalytic activity for both substrates. In the previous studies, the epimeriz of DAEs or DTEs had rarely been determined against D-sorbose or its catalytic act was very low while in the current study Bp-DAE had a high relative activity on D-sor compared to earlier reported DAEs or DTEs, which suggest a broad rare sugars speci of this enzyme. The specific activity in descending order were D-allulose (5.27 ± U/mg), D-tagatose (3.76 ± 0.13 U/mg), D-sorbose (2.29 ± 0.1 U/mg), and D-fructose (1 0.099 U/mg). In this case, the epimerization efficiency of D-allulose was about 1.4 t higher than that of D-tagatose, thus, it is the high specificity of this protein for D-allu that led to its precise designation as D-allulose 3-epimerase (DAE). Similar speci For the thermal stability assay, Bp-DAE was incubated at different temperatures (from 30 to 60 • C) for 0, 30, 60, 120, 180, and 240 min, and the variation in activity was measured ( Figure 5D). The results revealed that Bp-DAE showed more than 80% of relative activity at 30 • C after 4 h and sustained 75% activity for 2 h at 40 • C while decrease to 35% after 4 h. The fluctuation in thermal stability at 55 • C was comparable with that at 40 • C, but intriguingly, with the addition of an extra 1 mM Mn 2+ , the enzyme activity was enhanced by around 35%, and the half-life was increased from 60 to 180 min compared with 55 • C. This was not an anomaly, however, since the addition of 2 mM Co 2+ at 50 • C, increased the half-life of DaeM [49] from 297.3 to 539.2 min. Similarly, with the addition of 0.1 mM Co 2+ , the half-life of C. cellulolyticum DAE [25] at 60 • C increased from 10 to 408 min. Moreover, the enzyme activity of BP-DAE was almost totally inactivated after 2 hours of incubation at 60 • C. One of the key elements affecting the enzyme-catalyzed process was temperature. In the industrial production of rare sugars, higher temperatures allowed the enzymes to keep better activity because higher temperatures increased the solubility of the substrate, which sped up the enzyme reaction, thus increasing the conversion yield and to some extent reducing microbial contamination [49,50]. However, too high a temperature might trigger the Maillard reaction and lead to the thermal denaturation of the enzyme, resulting in a bad quality of sugar and low yields [49,50].

Substrate Specificity and Kinetic Properties of Bp-DAE
The substrate specificity and kinetic properties of enzymes were examined to investigate the applicability of Bp-DAE in rare sugar production. Four kinds of the substrate such as D-fructose, D-allulose, D-sorbose, and D-tagatose were employed to inspect the catalytic activity of Bp-DAE and showed 38.39%, 43.39%, 71.44%, and 100% of relative activities against each substrate respectively ( Figure 6). Interestingly, earlier reported DAEs and DTEs that specifically favor D-allulose or D-tagatose, similarly, Bp-DAE had also a high catalytic activity for both substrates. In the previous studies, the epimerization of DAEs or DTEs had rarely been determined against D-sorbose or its catalytic activity was very low while in the current study Bp-DAE had a high relative activity on D-sorbose compared to earlier reported DAEs or DTEs, which suggest a broad rare sugars specificity of this enzyme. The specific activity in descending order were D-allulose (5.27 ± 0.12 U/mg), D-tagatose (3.76 ± 0.13 U/mg), D-sorbose (2.29 ± 0.1 U/mg), and D-fructose (1.76 ± 0.099 U/mg). In this case, the epimerization efficiency of D-allulose was about 1.4 times higher than that of D-tagatose, thus, it is the high specificity of this protein for D-allulose that led to its precise designation as D-allulose 3-epimerase (DAE). Similar specificity preferences were present in Ruminococcus sp.DAE [26], Clostridium sp.DAE [27] and DaeM [35]. preferences were present in Ruminococcus sp.DAE [26], Clostridium sp.DAE [27] and DaeM [35]. The kinetic properties of Bp-DAE with different substrates such as D-fructose, D-allulose, D-tagatose, and D-sorbose according to the Lineweaver-Burk plot equation through the standard enzyme activity assay were given in Table 2 and the schematic diagram of different substrates as shown in Figure 7. The results explain that using different substrates, D-allulose showed the highest catalytic activity by exhibiting a minimum Km value (150.7 mM), the relatively high kcat value (5878 min −1 ), and the highest kcat/Km value (39.1 mM −1 min −1 ) as compared to other substrates. The Km, kcat, and kcat/Km values of D-fructose were 235.7 mM, 3902 min −1 , and 16.57 mM −1 min −1 , respectively. The Bp-DAE to Dfructose Km values from 153 to 323 mM were similar to previously reported DAEs, such as Ruminococcus DAE [26], Clostridium DAE [27], Dorea DAE [29], Treponema primitia DAE [30] and Flavonifractor plautii DAE [31], etc., and interestingly its kcat/Km value (16 mM −1 min −1 ) was extremely similar to that of Ruminococcus DAE [26]. The values of Km, kcat, and kcat/Km for D-tagatose were 238.7 mM, 6415 min −1, and 26.94 mM −1 min −1 , respectively. While the values of Km, kcat, and kcat/Km for D-sorbose were 297.6 mM, 6457 min −1, and 21.74 mM −1 min −1 , respectively. The kinetic properties of Bp-DAE and earlier reported DAEs and DTEs to D-allulose are presented in Table 1.

D -p s i c o s e D -t a g a t o s e D -s o r b o s e D -f r u c t o s e
Relative activity (%) Figure 6. Substrate specificity of Bp-DAE.
The kinetic properties of Bp-DAE with different substrates such as D-fructose, Dallulose, D-tagatose, and D-sorbose according to the Lineweaver-Burk plot equation through the standard enzyme activity assay were given in Table 2 and the schematic diagram of different substrates as shown in Figure 7. The results explain that using different substrates, D-allulose showed the highest catalytic activity by exhibiting a minimum K m value (150.7 mM), the relatively high k cat value (5878 min −1 ), and the highest k cat /K m value (39.1 mM −1 min −1 ) as compared to other substrates. The K m , k cat, and k cat /K m values of D-fructose were 235.7 mM, 3902 min −1 , and 16.57 mM −1 min −1 , respectively. The Bp-DAE to D-fructose K m values from 153 to 323 mM were similar to previously reported DAEs, such as Ruminococcus DAE [26], Clostridium DAE [27], Dorea DAE [29], Treponema primitia DAE [30] and Flavonifractor plautii DAE [31], etc., and interestingly its k cat /K m value (16 mM −1 min −1 ) was extremely similar to that of Ruminococcus DAE [26]. The values of K m , k cat, and k cat /K m for D-tagatose were 238.7 mM, 6415 min −1, and 26.94 mM −1 min −1 , respectively. While the values of K m , k cat , and k cat /K m for D-sorbose were 297.6 mM, 6457 min −1, and 21.74 mM −1 min −1 , respectively. The kinetic properties of Bp-DAE and earlier reported DAEs and DTEs to D-allulose are presented in Table 1. All experiments were repeated in triplicate, and data were presented by mean value ± standard deviation.  All experiments were repeated in triplicate, and data were presented by mean value ± standard deviation.

Structural Analysis of Bp-DAE and Enzyme-Substrate Docking
The structural homology model of Bp-DAE was created in accordance with the previously discovered DTE crystal structure of the engineered DTE PcDTE-IDF8 (SMTL ID: 4pfh.1) derived from the Pseudomonas cichorii DTE with some modifications. The simulated model was docked with molecules D-fructose and D-allulose respectively to investigate the mechanism of enzyme-substrate affinity. Studies had revealed that the residues around the oxygen atoms at position 1, 2, and 3 of the substrates were highly conserved and might precisely control the catalysis [51]. In the protein sequence alignment with known DAE and DTE, we knew that Bp-DAE also had similar catalytic residues and was related to substrate binding, such as Arg 216, Glu 157, His 187. The Bp-DAE-fructose structure diagram ( Figure 8A) and the Bp-DAE-allulose structure diagram ( Figure 8B) showed that the O1, O2, and O3 of D-fructose and D-allulose formed the same hydrogen bonds with BP-DAE protein molecule respectively including O1-Arg216, O1-Glu157, O1-His187, O2-His187, O2-Arg216, O3-Glu151, O3-His210. After molecular docking, we knew that they did combine with oxygen 1, 2, and 3 atoms of the substrate to form hydrogen bonds. Therefore, the binding-site residues at O4, O5, and O6 of the substrate were the focus of the study. According to the Bp-DAE-fructose structural analysis ( Figure 8C), a total of three hydrogen bonds were established with the O4, O5, and O6 atoms of D-fructose, particularly O4-Glu151, O6-Asn37, and O6-Try7. In contrast, the Bp-DAE-allulose structure ( Figure 8D) established four hydrogen bonds of O4-Glu151, O5-Glu151, O6-Asn37, and

Structural Analysis of Bp-DAE and Enzyme-Substrate Docking
The structural homology model of Bp-DAE was created in accordance with the previously discovered DTE crystal structure of the engineered DTE PcDTE-IDF8 (SMTL ID: 4pfh.1) derived from the Pseudomonas cichorii DTE with some modifications. The simulated model was docked with molecules D-fructose and D-allulose respectively to investigate the mechanism of enzyme-substrate affinity. Studies had revealed that the residues around the oxygen atoms at position 1, 2, and 3 of the substrates were highly conserved and might precisely control the catalysis [51]. In the protein sequence alignment with known DAE and DTE, we knew that Bp-DAE also had similar catalytic residues and was related to substrate binding, such as Arg 216, Glu 157, His 187. The Bp-DAE-fructose structure diagram ( Figure 8A) and the Bp-DAE-allulose structure diagram ( Figure 8B) showed that the O1, O2, and O3 of D-fructose and D-allulose formed the same hydrogen bonds with BP-DAE protein molecule respectively including O1-Arg216, O1-Glu157, O1-His187, O2-His187, O2-Arg216, O3-Glu151, O3-His210. After molecular docking, we knew that they did combine with oxygen 1, 2, and 3 atoms of the substrate to form hydrogen bonds. Therefore, the binding-site residues at O4, O5, and O6 of the substrate were the focus of the study. According to the Bp-DAE-fructose structural analysis ( Figure 8C), a total of three hydrogen bonds were established with the O4, O5, and O6 atoms of D-fructose, particularly O4-Glu151, O6-Asn37, and O6-Try7. In contrast, the Bp-DAE-allulose structure ( Figure 8D) established four hydrogen bonds of O4-Glu151, O5-Glu151, O6-Asn37, and O6-Try7. Compared with the fructose substrate, D-allulose had one more hydrogen bond. Based on the biochemical and docking analyses, D-allulose would be the best substrate for Bp-DAE. It was concluded that more hydrogen bonds may have encouraged substrate alignment and restricted substrate mobility, which might be the reason for the higher affinity of BP-DAE for D-allulose [52]. Bp-DAE contains Asn, a neutral polar residue, in place of neutral nonpolar Ala37 (the corresponding residue in the engineered DTE PcDTE-IDF8), which formed one hydrogen bond with the oxygen atom at position 6 of D-fructose. In other DAEs and DTEs, where the position was usually a neutral nonpolar residue (e.g., Leu, Ala, or Ile), followed by a neutral polar (e.g., Tyr) or positive polar (e.g., His) residue [35]. The different type of amino acid residues at this position might be the reason for the higher K m of Bp-DAE and result in lower affinity for the substrate. finity of BP-DAE for D-allulose [52]. Bp-DAE contains Asn, a neutral polar residue, in place of neutral nonpolar Ala37 (the corresponding residue in the engineered DTE PcDTE-IDF8), which formed one hydrogen bond with the oxygen atom at position 6 of D-fructose. In other DAEs and DTEs, where the position was usually a neutral nonpolar residue (e.g., Leu, Ala, or Ile), followed by a neutral polar (e.g., Tyr) or positive polar (e.g., His) residue [35]. The different type of amino acid residues at this position might be the reason for the higher Km of Bp-DAE and result in lower affinity for the substrate.

Enzymatic Production of D-allulose
In a preliminary pre-experiment, 300 mM D-fructose was treated with 5 U/mL Bp-DAE under optimum reaction conditions to explore the ability of Bp-DAE to convert Dfructose to D-allulose. The highest conversion of D-allulose reached around 30% after 4.5 h of reaction. Consequently, different concentrations of substrates such as 100, 300, and

Enzymatic Production of D-allulose
In a preliminary pre-experiment, 300 mM D-fructose was treated with 5 U/mL Bp-DAE under optimum reaction conditions to explore the ability of Bp-DAE to convert D-fructose to D-allulose. The highest conversion of D-allulose reached around 30% after 4.5 h of reaction. Consequently, different concentrations of substrates such as 100, 300, and 500 g/L were used with 9 U/mL of enzyme in a total volume of 50 mL under optimal conditions to further evaluate the high-level yield of D-allulose as well as the reaction equilibrium. The Bp-DAE showed 30, 90, and 150 g/L D-allulose production from 100, 300, and 500 g/L D-fructose after 6, 8, and 10 h respectively, which showed that the D-allulose/D-fructose equilibrium ratio was 30:70 ( Figure 9). In the literature, the conversion yield was typically between 20 to 33% by employing different ketose 3-epimerases (as presented in Table 1). The results of the present findings elaborate that the catalytic performance of Bp-DAE with 500 g/L substrate was at a medium to a higher level but the catalytic rate of Bp-DAE was significantly lower than that of known DAEs or DTEs such as the reaction equilibrium time of Treponema primitia DAE with 500 g/L substrate was 6 h while for Bp-DAE was 10 h which might be due to the low binding efficiency of the substrate. Alike other enzymes for improving some features, either rational modification or random mutation was possible [53]. For example, it was feasible to improve the catalytic rate of Bp-DAE by some techniques like site-directed mutagenesis as well as random and directed evolution [36,37].
presented in Table 1). The results of the present findings elaborate that the catalytic performance of Bp-DAE with 500 g/L substrate was at a medium to a higher level but the catalytic rate of Bp-DAE was significantly lower than that of known DAEs or DTEs such as the reaction equilibrium time of Treponema primitia DAE with 500 g/L substrate was 6 h while for Bp-DAE was 10 h which might be due to the low binding efficiency of the substrate. Alike other enzymes for improving some features, either rational modification or random mutation was possible [53]. For example, it was feasible to improve the catalytic rate of Bp-DAE by some techniques like site-directed mutagenesis as well as random and directed evolution [36,37].

Whole-Cell Catalysis Production of D-allulose
The whole-cell biocatalysts production has the following advantages over enzymatic production: (1) Cell culture is simple, the enzymes present in the cells have greater stability to the biocatalysts, also eliminating the need for complicated and tedious enzyme purification processes, (2) enzyme are protected by a cell wall and membranes from harsh conditions and maintain their full activity [54,55]. These advantages made the bioprocess more cost-effective and more suitable for scale up production. Therefore, the Bp-DAE gene was cloned into the pP43NMK shuttle plasmid, and Bacillus subtilis was used as the expression host, which was considered to be a safe-grade microbial strain. Then, 20 g/L whole cells of recombinant Bacillus subtilis was used with 500 g/L D-fructose substrate for a period of time under optimum enzymatic reaction conditions, and their reaction equilibrium times were recorded. The whole cell catalyst produced 17.43% of D-allulose after 4 h, 25.69% after 8 h, and reached equilibrium after 12 h with a 30.13% conversion yield by producing a yield of about 150.65 g/L ( Figure 10). In contrast, the whole-cell catalytic trial with empty vector (pP43NMK shuttle plasmid without Bp-DAE) recombinant Bacillus subtilis cells used as control, and under a same condition, revealed that this host cell did not transform any D-fructose during the reaction. In the actual production of D-allulose, Bacillus subtilis was a safe-grade microbial strain and more applicable for the food Conversion rate (%)

Whole-Cell Catalysis Production of D-allulose
The whole-cell biocatalysts production has the following advantages over enzymatic production: (1) Cell culture is simple, the enzymes present in the cells have greater stability to the biocatalysts, also eliminating the need for complicated and tedious enzyme purification processes, (2) enzyme are protected by a cell wall and membranes from harsh conditions and maintain their full activity [54,55]. These advantages made the bioprocess more cost-effective and more suitable for scale up production. Therefore, the Bp-DAE gene was cloned into the pP43NMK shuttle plasmid, and Bacillus subtilis was used as the expression host, which was considered to be a safe-grade microbial strain. Then, 20 g/L whole cells of recombinant Bacillus subtilis was used with 500 g/L D-fructose substrate for a period of time under optimum enzymatic reaction conditions, and their reaction equilibrium times were recorded. The whole cell catalyst produced 17.43% of D-allulose after 4 h, 25.69% after 8 h, and reached equilibrium after 12 h with a 30.13% conversion yield by producing a yield of about 150.65 g/L ( Figure 10). In contrast, the whole-cell catalytic trial with empty vector (pP43NMK shuttle plasmid without Bp-DAE) recombinant Bacillus subtilis cells used as control, and under a same condition, revealed that this host cell did not transform any D-fructose during the reaction. In the actual production of D-allulose, Bacillus subtilis was a safe-grade microbial strain and more applicable for the food industry as compared to Escherichia coli. Therefore, the catalysis reaction using the whole cells of Bacillus subtilis for D-allulose production was a desirable method [56]. In this study, Bacillus subtilis cells were successfully used as hosts to express Bp-DAE, which converted D-fructose to D-allulose after 12 h based on a whole-cell catalytic reaction with a conversion yield of about 30%. However, the genomic integration of Bp-DAE had an innate potentiality for higher catalytic efficiency compared with a plasmid-based transformation, which showed an instability limitation during multiple generations of plasmid vectors [56]. It is concluded that a whole-cell catalytic technique combined with the fed-batch strategy is an effective bioprocess approach to achieve a higher yield of D-allulose that could be applied to the production of many high value-added products [35]. nate potentiality for higher catalytic efficiency compared with a plasmid-based transformation, which showed an instability limitation during multiple generations of plasmid vectors [56]. It is concluded that a whole-cell catalytic technique combined with the fedbatch strategy is an effective bioprocess approach to achieve a higher yield of D-allulose that could be applied to the production of many high value-added products [35].

Conclusions
In brief, we found a novel D-allulose 3-epimerase named Bp-DAE based on the NCBI database, and encoded genes were obtained from a potential probiotic Blautia produca. The gene source was safe and conformed to food grade production. Then the encoded genes were expressed in E. coli BL21 (DE3) for the extraction of Bp-DAE and purified the enzyme for characterization. The Bp-DAE exhibited neutral to alkaline pH (optimum 8.0) at optimum temperature 55 °C with good catalytic activity and broad substrate specificity. In addition, it could also maintain activity in slightly acidic environment, which was helpful for industrial production of D-allulose. The current study showed that Bp-DAE enzyme was strictly dependent on metal ions and almost inactivated in the absence of metal ions, its catalytic activity has been greatly increased with the addition of Co 2+ and Mn 2+ ions. Moreover, whole-cell catalysis by employing Bp-DAE-transformed Bacillus subtilis cells and pure enzyme catalysis were used for the conversion of D-fructose to D-allulose, which showed a 30% conversion yield which was intermediate to high. Recently, many studies had been focused on improving the specificity of D-fructose and the thermal stability of enzymes. Conclusively, this research demonstrated that Bp-DAE could be used an effective biocatalyst for the D-allulose production that can be used for industrial purpose to decrease the price of downstream industries.

Conclusions
In brief, we found a novel D-allulose 3-epimerase named Bp-DAE based on the NCBI database, and encoded genes were obtained from a potential probiotic Blautia produca. The gene source was safe and conformed to food grade production. Then the encoded genes were expressed in E. coli BL21 (DE3) for the extraction of Bp-DAE and purified the enzyme for characterization. The Bp-DAE exhibited neutral to alkaline pH (optimum 8.0) at optimum temperature 55 • C with good catalytic activity and broad substrate specificity. In addition, it could also maintain activity in slightly acidic environment, which was helpful for industrial production of D-allulose. The current study showed that Bp-DAE enzyme was strictly dependent on metal ions and almost inactivated in the absence of metal ions, its catalytic activity has been greatly increased with the addition of Co 2+ and Mn 2+ ions. Moreover, whole-cell catalysis by employing Bp-DAE-transformed Bacillus subtilis cells and pure enzyme catalysis were used for the conversion of D-fructose to D-allulose, which showed a 30% conversion yield which was intermediate to high. Recently, many studies had been focused on improving the specificity of D-fructose and the thermal stability of enzymes. Conclusively, this research demonstrated that Bp-DAE could be used an effective biocatalyst for the D-allulose production that can be used for industrial purpose to decrease the price of downstream industries.  Data Availability Statement: All datasets generated for this study are included in the article.