Efficient (3S)-Acetoin and (2S,3S)-2,3-Butanediol Production from meso-2,3-Butanediol Using Whole-Cell Biocatalysis

(3S)-Acetoin and (2S,3S)-2,3-butanediol are important platform chemicals widely applied in the asymmetric synthesis of valuable chiral chemicals. However, their production by fermentative methods is difficult to perform. This study aimed to develop a whole-cell biocatalysis strategy for the production of (3S)-acetoin and (2S,3S)-2,3-butanediol from meso-2,3-butanediol. First, E. coli co-expressing (2R,3R)-2,3-butanediol dehydrogenase, NADH oxidase and Vitreoscilla hemoglobin was developed for (3S)-acetoin production from meso-2,3-butanediol. Maximum (3S)-acetoin concentration of 72.38 g/L with the stereoisomeric purity of 94.65% was achieved at 24 h under optimal conditions. Subsequently, we developed another biocatalyst co-expressing (2S,3S)-2,3-butanediol dehydrogenase and formate dehydrogenase for (2S,3S)-2,3-butanediol production from (3S)-acetoin. Synchronous catalysis together with two biocatalysts afforded 38.41 g/L of (2S,3S)-butanediol with stereoisomeric purity of 98.03% from 40 g/L meso-2,3-butanediol. These results exhibited the potential for (3S)-acetoin and (2S,3S)-butanediol production from meso-2,3-butanediol as a substrate via whole-cell biocatalysis.


Feasibility of (3S)-Acetoin Production from meso-2,3-Butanediol by Whole-Cell Catalysis
The whole-cell biocatalysis for (3S)-acetoin production from meso-2,3-butanediol was firstly conducted by using E. coli (pET-rrbdh), E. coli (pET-rrbdh-nox), and E. coli (pET-rrbdh-nox-vgb) in a 10 mL reaction mixture (pH 7.0) containing 20 g/L 2,3-butanediol and 40 g WCW/L at 30 • C for 12 h, and E. coli (pET28a) was used as the control. The configuration and yield of acetoin produced by whole-cell biocatalysis were analyzed and quantified using a GC system equipped with a chiral column. The results were shown in Figure 3 and Table 3. As shown in Table 3, 7.86 g/L of (3S)-acetoin from 20 g/L 2,3-butanediol as substrate was achieved by E. coli (pET-rrbdh). Meanwhile, a small amount of (3R)-acetoin (0.71 g/L) could be detected in the reaction system partially due to the existence of (2R,3R)-2,3-butanediol (4.04%) in the 2,3-butanediol substrate ( Figure 3). As reported in previous studies, the (2R,3R)-2,3-BDH enzyme could convert (2R,3R)-2,3-butanediol into (3R)-acetoin, which resulted in a small amount of (3R)-acetoin production [13,22]. The (2R,3R)-2,3-BDH coupled with NOX enzyme in E. coli (pET-rrbdh-nox) resulted in the rapid increase of (3S)-acetoin yield (13.88 g/L), indicating that NAD + regeneration could remarkably improve the whole-cell biocatalytic efficiency. Considering NOX enzyme required oxygen as substrate, the vgb gene encoding VHB enzyme introduced into E. coli (pET-rrbdh-nox) was employed to improve oxygen transfer during the whole-cell biocatalytic process [7,24]. As shown in Table 3, the expression of VHB led to an increase of (3S)-acetoin yield (16.79 g/L) by 20.97% compared with that by E. coli (pET-rrbdh-nox). The result was consistent with those of previous studies showing that VHB protein could be used to improve biocatalytic efficiency in biotransformation system with NOX-dependent NAD(P) + regeneration [7,25]. Table 3. The feasibility of (3S)-AC production from meso-2,3-BD by the recombinant strains.   Table 3. The feasibility of (3S)-AC production from meso-2,3-BD by the recombinant strains.   Interestingly, a small amount of acetoin was also produced from 2,3-butanediol by the control strain E. coli (pET28a), which suggested that some non-specific dehydrogenases in E. coli might catalyze conversion of 2,3-butanediol into acetoin as observed in our previous study [13]. Based on the above results, the recombinant strain of E. coli (pET-rrbdh-nox-vgb) was chosen to perform the optimization of bioconversion conditions. Interestingly, a small amount of acetoin was also produced from 2,3-butanediol by the control strain E. coli (pET28a), which suggested that some non-specific dehydrogenases in E. coli might catalyze conversion of 2,3-butanediol into acetoin as observed in our previous study [13]. Based on the above results, the recombinant strain of E. coli (pET-rrbdh-nox-vgb) was chosen to perform the optimization of bioconversion conditions. 2.2.2. Effects of pH, Temperature, and WCW on (3S)-Acetoin Production by E. coli (pET-rrbdh-nox-vgb) To obtain high (3S)-acetoin concentrations, the effects of pH, temperature, and WCW on whole-cell biocatalytic efficiency were systematically investigated by using induced E. coli (pET-rrbdh-nox-vgb) cells. As shown in Figure 4A, the whole-cell biocatalytic reactions containing 20 g/L 2,3-butanediol and 40 g/L wet cells were conducted under different pH-values for 12 h at 30 • C. The results showed that the yield of (3S)-acetoin was gradually increased with the pH increase from 6.0 to 8.5. A maximum (3S)-acetoin concentration of 18.31 g/L could be obtained at pH 8.5. Thus, pH 8.5 as the optimum pH-value was used for the following experiments. The effect of temperature on (3S)-acetoin production from meso-2,3-butanediol by induced E. coli (pET-rrbdh-nox-vgb) cells was studied in the range of 25-42 • C. Figure 4B indicates that 18.45 g/L of (3S)-acetoin was achieved at 12 h when the whole-cell biocatalytic reaction was performed at 30 • C, which was used as the optimum temperature for further experiments. Among the parameters affecting whole-cell biocatalytic reaction, the concentration of the biocatalyst also plays an important role on the bioconversion efficiency [7,26], so WCWs from 10 to 60 g/L were used to evaluate the effect on the whole-cell biocatalysis process. As shown in Figure 4C, the (3S)-acetoin concentration was improved rapidly with the increase of WCW in the whole-cell biocatalytic reaction. The (3S)-acetoin concentration of 18.20 and 18.64 g/L could be produced from 20 g/L 2,3-butanediol by induced E. coli (pET-rrbdh-nox-vgb) cells at 12 h when the cell concentrations were 40 g and 60 g WCW/L. The WCW of 40 g/L was chosen as an optimum cell concentration after considering the cost. To obtain high (3S)-acetoin concentrations, the effects of pH, temperature, and WCW on wholecell biocatalytic efficiency were systematically investigated by using induced E. coli (pET-rrbdh-nox-vgb) cells. As shown in Figure 4A, the whole-cell biocatalytic reactions containing 20 g/L 2,3-butanediol and 40 g/L wet cells were conducted under different pH-values for 12 h at 30 °C. The results showed that the yield of (3S)-acetoin was gradually increased with the pH increase from 6.0 to 8.5. A maximum (3S)acetoin concentration of 18.31 g/L could be obtained at pH 8.5. Thus, pH 8.5 as the optimum pH-value was used for the following experiments. The effect of temperature on (3S)-acetoin production from meso-2,3-butanediol by induced E. coli (pET-rrbdh-nox-vgb) cells was studied in the range of 25-42 °C. Figure 4B indicates that 18.45 g/L of (3S)-acetoin was achieved at 12 h when the whole-cell biocatalytic reaction was performed at 30 °C, which was used as the optimum temperature for further experiments. Among the parameters affecting whole-cell biocatalytic reaction, the concentration of the biocatalyst also plays an important role on the bioconversion efficiency [7,26], so WCWs from 10 to 60 g/L were used to evaluate the effect on the whole-cell biocatalysis process. As shown in Figure 4C   Substrate concentration is another limiting factor in the whole-cell biocatalytic reaction. Low substrate concentration resulted in low product yield, whereas the bioconversion reaction might be inhibited by high substrate concentration. The effect of the initial 2,3-butanediol concentration on (3S)-acetoin production from meso-2,3-butanediol by E. coli (pET-rrbdh-nox-vgb) was tested in the range of 20-140 g/L for 24 h. The results indicated that the product (3S)-acetoin concentration was significantly increased by increasing the initial substrate concentration from 20 to 100 g/L ( Figure 4D). When the initial 2,3-butanediol concentration continued to increase to 140 g/L, the (3S)-acetoin concentration appeared the decrease trend, implying that 100 g/L 2,3-butanediol as initial substrate concentration was suitable for (3S)-acetoin production. Ultimately, maximum (3S)-acetoin concentration of 60.29 g/L from 100 g/L 2,3-butanediol was obtained at 24 h ( Figure 4D). Previous studies showed that the use of some metal ions such as Mn 2+ , Ca 2+ , Fe 2+ , Fe 3+ , and Mg 2+ as chemical stimulators could accelerate the conversion from 2,3-butanediol to acetoin by improving the activity of (2R,3R)-2,3-BDH [22,27]. However, these metal ions showed stimulation or inhibition for NADH oxidase, another enzyme in the biocatalyst [7]. To evaluate their effect on (3S)-acetoin production, five metal ions including Mn 2+ , Ca 2+ , Fe 2+ , Fe 3+ , and Mg 2+ at 0, 2.5, and 5.0 mM were chosen to conduct the whole-cell biocatalysis by using induced E. coli (pET-rrbdh-nox-vgb) cells for 24 h. The results showed that all the listed metal ions except Fe 2+ could promote (3S)-acetoin production at different concentrations ( Figure 4E). Especially Mn 2+ exhibited obvious stimulation effect on (3S)-acetoin production in the bioconversion reaction. Maximum (3S)-acetoin concentration of 65.77 g/L was achieved by Mn 2+ at 5 mM at 24 h.

Optimization of Synchronous Catalysis Conditions
Furthermore, four factors including pH, temperature, the WCW ratio of E. coli (pET-rrbdh-nox-vgb) cells to E. coli (pET-ssbdh-fdh) cells and metal ions were optimized using single factor experiments. The results were given in Figure 6. Figure 6A indicated that the pH value in the synchronous catalysis system had an important effect on (2S,3S)-2,3-butanediol production from meso-2,3-butanediol. The optimum pH was observed at 7.0, which was not consistent with that of (3S)-acetoin from meso-2,3-butanediol by E. coli (pET-rrbdh-nox-vgb). A probable explanation was that all the reported BDHs were reversible enzymes catalyzing the interconversion between acetoin and 2,3-butanediol. Alkaline environment favors the oxidation of 2,3-butanediol to acetoin by BDH enzymes, which readily reduced acetoin as a substrate to 2,3-butanediol in acid solution [5,13,22,27]. The effect of temperature on synchronous catalysis was investigated by using the optimum pH of 7.0. Two biocatalysts of induced E. coli (pET-rrbdh-nox-vgb) and E. coli (pET-ssbdh-fdh) cells exhibited high biocatalytic stability at 30 • C ( Figure 6B). The ratio of E. coli (pET-rrbdh-nox-vgb) cells to E. coli (pET-ssbdh-fdh) cells was optimized at pH 7.0 and 30 • C as shown in Figure 6C. The results indicated that 15.17 and 15.43 g/L of (2S,3S)-2,3-butanediol production from 20 g/L meso-2,3-butanediol could be achieved at 6 h when the ratios of E. coli (pET-rrbdh-nox-vgb) to E. coli (pET-ssbdh-fdh) were 40:30 and 40:40 respectively. To consider the cost, the ratio (40:30) of two biocatalysts was chosen for following experiments. The effect of metal ions with different concentrations on synchronous catalysis was shown in Figure 6D. The results showed that Mn 2+ among all the listed meal ions played an obvious stimulation effect on (2S,3S)-2,3-butanediol production from meso-2,3-butanediol in the synchronous catalysis reaction. Maximum (2S,3S)-2,3-butanediol concentration of 16.71 g/L was obtained in the presence of 5.0 mM Mn 2+ for 6 h.

Construction of the Recombinant Strains as Biocatalysts
The bacterial strains, plasmids, and primers used in this study are listed in Table 4. The genomic DNA of P. polymyxa ATCC12321, L. brevis 20054, Serratia sp. T241 and C. boidinii NCYC 1513 were extracted the OMEGA Bacterial Genomic DNA Kit (OMEGA, Shanghai, China). The vgb gene (GenBank Accession number AAA75506) source was from the pBR322-vgb plasmid previously constructed in our lab. For construction of the recombinant pET-rrbdh plasmid, the rrbdh gene (GenBank Accession number ADV15558) encoding (2R,3R)-2,3-BDH from P. polymyxa ATCC12321 were amplified by PCR using the primers P1/P2 with BamHI and HindIII sites respectively. The obtained PCR products were inserted into the expression plasmid pET28a at the BamHI and HindIII sites. ClonExpress MultiS One Step Cloning Kit (Vazyme, Nanjing, China) was used to develop the recombinant pET-rrbdh-nox, pET-rrbdh-nox-vgb and pET-ssbdh-fdh plasmids according to the protocol. In brief, a series of primers (P3-P6) were designed with adjacent oligos overlapped by 15-20 bp at each end of the assembly. The PCR-amplification products of the rrbdh and nox (GenBank Accession number AAN04047) genes using the primers P3/P4 and P5/P6 were assembled into the linearized pET28a vector at BamHI and HindIII sites, generating the recombinant pET-rrbdh-nox plasmid. Similarly, the rrbdh, nox, vgb, ssbdh (GenBank Accession number AEF51363) and fdh (GenBank Accession number O13437) genes were amplified using the primers (P7-P16), respectively. The PCR-amplification products of the rrbdh, nox and vgb genes or

Construction of the Recombinant Strains as Biocatalysts
The bacterial strains, plasmids, and primers used in this study are listed in Table 4. The genomic DNA of P. polymyxa ATCC12321, L. brevis 20054, Serratia sp. T241 and C. boidinii NCYC 1513 were extracted the OMEGA Bacterial Genomic DNA Kit (OMEGA, Shanghai, China). The vgb gene (GenBank Accession number AAA75506) source was from the pBR322-vgb plasmid previously constructed in our lab. For construction of the recombinant pET-rrbdh plasmid, the rrbdh gene (GenBank Accession number ADV15558) encoding (2R,3R)-2,3-BDH from P. polymyxa ATCC12321 were amplified by PCR using the primers P1/P2 with BamHI and HindIII sites respectively. The obtained PCR products were inserted into the expression plasmid pET28a at the BamHI and HindIII sites. ClonExpress MultiS One Step Cloning Kit (Vazyme, Nanjing, China) was used to develop the recombinant pET-rrbdh-nox, pET-rrbdh-nox-vgb and pET-ssbdh-fdh plasmids according to the protocol. In brief, a series of primers (P3-P6) were designed with adjacent oligos overlapped by 15-20 bp at each end of the assembly.

Biocatalyst Preparation
The recombinant strains harboring the plasmids pET-rrbdh, pET-rrbdh-nox, pET-rrbdh-nox-vgb, and pET-ssbdh-fdh were cultured in LB medium containing 50 µg/mL kanamycin at 37 • C until the OD 600 reach up to 0.6, and then 0.5 mM IPTG was added into the culture for induction expression. After 8 h induction at 30 • C, the cells were harvested by centrifugation at 8000× g for 10 min at 4 • C and then washed twice with 0.85% NaCl. The cell pellets were resuspended in 50 mM potassium phosphate buffer for bioconversion experiments.

Enzyme Assays
The induced cells after harvest by centrifugation were resuspended in 50 mM potassium phosphate buffer (pH 7.4) and disrupted by sonication in an ice bath for 10 min. The disrupted cells were centrifuged at 10,000× g for 10 min at 4 • C to remove the cell debris, and the obtained supernatant was used as the crude enzyme extract for enzyme activity assays. The enzyme activities of (2R,3R)-2,3-BDH and FDH were assayed by measuring the changes in absorbance at 340 nm corresponding to the reduction of NAD + using a UV/visible spectrophotometer (UV-1800, Mapada, Shanghai, China) [22,33]. The reaction mixtures containing 50 mM potassium phosphate buffer (pH 8.5), 0.2 mM NAD + and 50 mM 2,3-butanediol ((2S,3S)-2,3-butanediol or (2R,3R)-2,3-butanediol or meso-2,3-butanediol) were used to determine (2R,3R)-2,3-BDH activity. The assay of FDH activity was carried out in 50 mM potassium phosphate buffer (pH 7.0) with 0.2 mM NAD + and 100 mM formate as substrate. The enzyme activities of (2S,3S)-2,3-BDH and NOX were assayed by measuring the changes in absorbance at 340 nm corresponding to the oxidation of NADH using a UV/visible spectrophotometer (UV-1800, Mapada, Shanghai, China) [13,34]. The reaction mixtures for (2S,3S)-2,3-BDH activity assay contained 50 mM potassium phosphate buffer (pH 7.0), 0.2 mM NADH and 50 mM (3R/3S)-acetoin as substrate. The NOX activity assay was carried out in 50 mM potassium phosphate buffer with 0.2 mM NADH as the substrate. One unit of (2R,3R)-2,3-BDH and FDH activity was defined as the amount of enzyme required to reduce 1 µmol of NAD + in one minute. One unit of (2S, 3S)-2,3-BDH and NOX activity was defined as the amount of enzyme required to oxidize 1 µmol of NADH in one minute. All enzyme activities were determined in triplicate, and standard deviations of the biological replicates were represented by error bars. The protein concentration was determined by the Bradford method using bovine serum albumin as the standard.

Product Analysis
The samples in the bioconversion reactions were centrifuged at 10,000× g for 5 min. The concentration of acetoin and 2,3-butanediol in the supernatant was analyzed by a GC system (Agilent 7820A, Santa Clara, CA, USA), but before GC analysis, the supernatant was extracted by ethyl acetate with the addition of n-butanol as the internal standard. The GC system consisted of a FID detector and a chiral column (Supelco β-DE 120, 30-m length, and 0.25-mm inner diameter; Sigma-Aldrich). The operation conditions were as follows: N 2 was used as the carrier gas at a flow rate of 1.2 mL/min; the injector temperature and the detector temperature were 215 • C and 245 • C, respectively; and the column temperature was maintained at 50 • C for 1.5 min, then raised to 180 • C at a rate of 15 • C/min. The concentration of acetoin and 2,3-butanediol in the supernatant was determined using standard curves [5,35].