Using Choline Chloride ‐ Based DESs as Co ‐ Solvent for 3,5 ‐ Bis(trifluoromethyl) Acetophenone Bioreduction with Rhodococcus erythropolis XS1012

: ( S ) ‐ 3,5 ‐ Bistrifluoromethylphenyl ethanol(( S ) ‐ BTPE) is a key pharmaceutical intermediate of the NK ‐ 1 receptor antagonist. The asymmetric bioreduction of 3,5 ‐ bis(trifluoromethyl) acetophenone (BTAP) to ( S ) ‐ BTPE using Rhodococcus erythropolis XS1012 has been established in a phosphate buffer system. To overcome the problem of unsatisfactory yields at high substrate concentration, deep eutectic solvents (DESs) have been introduced to the buffer system. After screening 13 kinds of choline chloride ‐ based DESs, [ choline chloride][urea] ([ChCl][U]) showed great influence on the cell activity and significantly increased the cell membrane permeability. Subsequently, some major parameters for this reaction were determined. A remarkable ( S ) ‐ BTPE yield of 91.9% was gained at 150 mM substrate concentration under optimized reaction conditions with >99.9% product enantioselectivity. Compared to reduction in a buffer system, the developed [ChCl][U] ‐ containing system increased the yield from 82.6% to 91.9%. It maintains a yield of 80.7% with the substrate concentration up to 300 mM, compared to only 63.0% in buffer system. This study demonstrated that [ChCl][U] is a feasible co ‐ solvent to improve the bioreduction process. 4 ∙ 7H 2 0.75 g/L, pH 6.5) at a ratio of and cultured at 30 and 200 for another 48 h. The incubated cells were harvested after centrifugation at 9000 rpm and 4 °C for 10 min, and subjected to biocatalytic reduction.


Introduction
Chirality is a significant characteristic of drugs and drug candidates that mainly makes a difference in pharmacokinetics, pharmacodynamics, and toxicity [1,2]. Both optical pure 3,5-bistrifluoromethylphenyl ethanol ((R)-and (S)-BTPE) is an intermediate for the synthesis of NK-1 receptor antagonist, while (S)-BTPE is included in the antagonists currently under clinical evaluation [3]. The chemical method for BTPE isomer preparation always employs transition metals as catalysts, which are expensive and highly toxic. Furthermore, they also have strict requirements for reaction conditions [4]. The biocatalytic process for chiral alcohol synthesis uses whole cells or isolated enzymes from the cells for implementation [5]. Compared to traditional chemical routes, the preparation of chemicals by biocatalysis has many merits, such as high enantioselectivity, energy effectiveness, environment protection, mild reaction conditions, and so on [6,7]. Immobilized Saccharomyces rhodotorula cells were used for asymmetric reduction of BTAP to (S)-BTPE in the aqueous-organic system with 1.42 g/L substrate, but only achieved the conversion rate at 93.3% with e.e. value of 95.9% [8]. In our previous report, the asymmetric production of (S)-BTPE by Candida tropicalis 104 cells gained a high yield (70.3%) and excellent enantioselectivity at 50 mM substrate concentration [9]. The above literature indicates that the biocatalytic reduction of (S)-BTPE in the aqueous phase is generally unsatisfied when at high substrate concentration. In a previous study, Li et al. introduced the ionic liquid [N1, 1,1,1][Cys] to the bioreduction of (S)-BTPE by Candida tropicalis 104, which increased BTAP concentration from 50 mM to 90 mM, and the yield from 70.3% to 80.4% [10].
Ionic liquids (ILs) display many unique characteristics, such as almost negligible vapor pressure, excellent chemical and thermal stability, low melting points, low toxicity, and good biocompatibility. Because of their superior performance, they have drawn extensive interest and are considered to be replacements for organic solvents [11,12]. Recently, another kind of interesting solvent for biotransformation and organic synthesis has drawn attention, namely deep eutectic solvents (DESs) [13][14][15][16]. Due to their similar properties to ionic liquids, they are also regarded as analogs of ILs, but with the advantages of lower cost of synthesis [17] and easier preparation [18]. DESs generally consist of two or three components, such as choline chloride with a hydrogen donor (such as urea and amino acids). For example, a study using [ChCl][U] as co-solvent has improved the reaction efficiency of steroid 1-en-dehydrogenation biotransformation by Arthrobacter simplex [19].
However, little attention has been paid to asymmetric reduction using DESs as the reaction medium by whole-cell biocatalysis. In our previous study, Rhodococcus erythropolis XS1012 was found to have the ability for the asymmetric reduction of BTAP to (S)-BTPE with high selectivity (>99.9%) but provided poor catalytic efficiency [20]. Choosing a co-solvent appropriately can provide a higher reactant concentration [21], so it is of great interest to carry out the investigation of the influences of choline chloride-based DESs on biocatalysis reduction.

The Bioreduction Carried out in Phosphate Buffer System
Rhodococcus erythropolis XS1012 was selected as a biocatalyst for the asymmetric reduction of BTAP to (S)-BTPE with superior enantioselectivity (>99.9% e.e.). Figure 1 shows the effect of substrate concentration on the bioreduction in the phosphate buffer system after the optimization of various parameters of the reaction (data not shown); the yield was 82.6% at 150 mM BTAP, and only 63.0% under 300 mM BTAP. To further enhance the product yield under high substrate concentration, DESs were introduced to the reaction system. phosphate buffer (Na2HPO4-NaH2PO4, 100 mM, pH 7.0), 45 g (DCW)/L R. erythropolis XS1012, 50 g/L glucose, 15% (v/v) isopropanol content, various BTAP concentration, and 30 °C, 200 rpm for 24 h.

Screening of Choline Chloride-Based Deep Eutectic Solvents (DES)
To improve the reduction efficiency of BTAP to (S)-BTPE under high substrate concentration, 13 varieties of DESs (in Table S1) were assessed for their behaviors in bioreduction at a weight-to-volume ratio of 1%. The reaction mixture without DES addition was set as the control group. As illustrated in Table 1 [Tyr] also brought increasing effects, with a yield of 57.7% and 57.6%, respectively. However, the other 9 types of DESs affected the reduction only slightly, and some DES addition in the reaction system even reduced the yield.

Influence of Various DESs on Cell Membrane Permeability
Although the product yield rose after introducing DESs, the drawbacks of DESs such as high viscosity, unknown corrosivity, and instability of mixture [22] may influence the mass transfer of the substrate and product in the DES-containing buffer systems, therefore influencing the efficiency of the reduction. Cell membrane permeability can be represented by leakage of nucleic acids and proteins, and they exhibited maximum ultraviolet absorption peak at 260 and 280 nm, respectively. The cells suspended in phosphate buffer without DES were set as control. The results in Table 2 demonstrate that the OD260nm values all increased in the DES-containing buffer systems.
[ChCl][GSH] and [ChCl][U] gave the lowest and highest values, respectively, of the OD280nm compared to the control. We speculate that DESs may cause cell permeability alteration or membrane expansion [23], which increases the frequency of the substrate transformation with the enzyme in cells, and promotes the substrate and product that comes in and out of the cells [24].

Effect of Various DESs on Cell Activity
To further understand the effect of DESs on R. erythropolis XS1012 cells, cell activity was estimated by the value of sugar metabolic retention (MAR), which can be altered with cell tolerance to the DESs and substrate. The cells suspended in phosphate buffer without DESs were set as the control group.
As shown in Figure 2, not all the DESs in the buffer system (with substrate) had a negative influence on cell activity, but the MAR values in all tested DES-containing systems (without substrate) were lower than the control, suggesting that DESs exhibited their toxicity in cells to varying degrees. , and urea (U)) had a benign effect on the tested bacteria (including two Gram-positive bacteria, i.e., Bacillus subtilis and Staphylococus aureus, and two Gram-negative bacteria, i.e., E. coli and Pseudomonas aeruginosa) and no inhibition from them [25]. That had similar results to our study to some extent. Obviously

Effects of [ChCl][U] and Its Components on the Asymmetric Reduction
To examine the effect of [ChCl][U] and its components on the reduction, bioreductions were compared in the presence of ChCl, urea, ChCl and urea, or in a [ChCl][U]-containing system. As shown in Figure 3, all the evaluated systems exhibited improved product yields compared to the control in the phosphate buffer system except for the ChCl-containing system. The [ChCl][U]-containing system gave the highest yield in R. erythropolis XS1012 catalyzed bioreduction. However, ChCl showed a negative influence on bioreduction compared to the phosphate buffer system. The addition of ChCl, urea, ChCl and urea, or [ChCl][U] did not affect the product e.e. values, which were all above 99%.

Control
ChCl

Effects of [ChCl][U](1:1) Content, Diverse Buffer Systems, Buffer pH on Asymmetric Reduction
As shown in Figure 4a, there was an increase in [ChCl][U](1:1) content up to 1% (w/v) with a clear progress in yield, while a further rise [ChCl][U](1:1) content led to a decline in yield. This may have been caused by the high viscosity of the DESs, which led to some problems in transporting; moreover, this could have reduced the mass transfer efficiency [26,27].  The results in Table 3 imply the influence of diverse buffer systems on bioreduction. After screening five types of buffer systems, the reaction in K2HPO4-KH2PO4 buffer gave the highest yield of 70.2%. The reaction also gained a yield of 61.3% in the distilled water system. The optimal pH of K2HPO4-KH2PO4 buffer was also investigated; it revealed that the maximum yield appeared at pH 6.5 (Figure 4b). When there was a continued increase in pH, there was an obvious decrease in the product yield; the product e.e. values were above 99.9% across the range.

Effects of the Type and Content of Co-Substrates, Reaction Temperature on the Asymmetric Reduction of BTAP in [ChCl][U]-Containing Phosphate Buffer System
Although bioreduction has many advantages for chiral asymmetric reduction, there still exists a problem regarding the reduction of equivalents in the form of NAD(P)H. The addition of NAD(P)H is neither economically nor technically feasible. Therefore, it is necessary to seek cheap co-substrates to replace the NAD(P)H [28]. Several sugars and alcohols were tested in the reaction as co-substrates. As illustrated in Table 4, after the co-substrate was added, all employed varieties of co-substrates except for methanol and ethanol could promote the reaction to varying degrees. In particular, isopropanol brought the yield markedly up from 0.2% to 76.1%. It is possible that isopropanol acted as the co-substrate, or that isopropanol can enhance the cell membrane permeability and increase the solubility of the substrate that promotes the reaction [29]. However, the joint addition of glucose and isopropanol did not make an obvious difference when compared with the addition of isopropanol individually. It is known that organic solvents have toxicity. Isopropanol is less toxic than methanol and ethanol, but can cause impaired gluconeogenesis [30]. In our investigations, the appropriate content of isopropanol (10-30%, v/v) (Figure 5a) was measured; the highest yield of 91.9% was obtained with the addition of 20% (v/v) isopropanol. Additionally, the impact of the reaction temperature (Figure 5b) on the reaction was also investigated, and 30 °C was determined to be optimal, beyond which there will be a reduction in the product yield. It seems that high reaction temperature is beneficial to the formation of enzyme-substrate complexes and improves the colliding probability between the substrate and enzyme [31]. However, too high a temperature may damage the enzyme structure and lower the enzyme catalytic efficiency.

Effects of Cell Concentration, Substrate Concentration and Reaction Time on the Asymmetric Reduction of BTAP in [ChCl][U](1:1)-Containing Phosphate Buffer System
Cell concentration is significant for biocatalysis because excessive biomass may make the reaction transformation difficult, while a lack of cells led to a low amount of enzyme. Based on the cell activity experiment, the substrate showed toxicity to the cells, and it is necessary to determine the proper concentration of the substrate. Figure 6a shows the effect of cell concentration on the asymmetric reduction. The yield rose with increasing cell concentration up to 45 g (DCW)/L, further increasing the cell concentration, which accounts for an appreciable loss of product yield. The results in Figure 6b show the influence of substrate concentration on reduction, and the maximum yield of 91.9% was obtained at 150 mM substrate concentration. Additionally, the results in Figure  6c show the impact of the reaction time on the reaction; the product yield rose rapidly from 0 to 24 h, and the best yield of 80.7% for (S)-BTPE was achieved at 300 mM of BTAP after a 24 h reaction in the [ChCl][U](1:1)-containing phosphate buffer system, further increasing the reaction time, at which point the product yield started to decrease. During the examinations above, there was no impact on the product e.e. value (>99.9%).

Chemicals
The substrate BTAP (purity >99%) was purchased from Beijing Golden Olive Company, China. Product (S)-BTPE and (R)-BTPE (purity >99%) were provided by Hangzhou Xinhai Biotechnology Co., Ltd., Hangzhou 310052 China. DESs (purity >98%) used in this research were synthesized by Shanghai Fujie Chemical Co., Ltd., Shanghai 201203 China. All other chemicals were from commercial sources and were of analytical grade.

Culture of the Strain
The strain R. erythropolis XS1012 (CCTCC M 2013650) stored in our laboratory can catalyze the reduction of substrate BTAP to (S)-BTPE with high enantioselectivity. The single colony of R. erythropolis XS1012 was selected from the agar plate after being cultivated for 2-3 days at 30 °C, then inoculated into a 250 mL flask containing 100 mL medium (glucose 15 g/L, yeast extract 6 g/L, peptone 7.5 g/L, (NH4)2SO4 3 g/L, KH2PO4 1.5 g/L, NaCl 0.75 g/L, MgSO4•7H2O 0.75 g/L, pH 6.5). After that, it was incubated at 30 °C, 200 rpm for 24 h as the seed culture. Subsequently, seed culture was transferred into another 100 mL of fermentation medium (glucose 15 g/L, yeast extract 6 g/L, peptone 7.5 g/L, KH2PO4 1.5 g/L, (NH4)2SO4 3 g/L, NaCl 0.75 g/L, MgSO4•7H2O 0.75 g/L, pH 6.5) at a volume ratio of 6% and cultured at 30 °C and 200 rpm for another 48 h. The incubated cells were harvested after centrifugation at 9000 rpm and 4 °C for 10 min, and subjected to biocatalytic reduction.

Bioreduction Process and Screening of DESs
13 kinds of choline chloride-based DESs (listed in Table 1) were assessed for their behaviors in the reaction. The bioreduction was carried out in 50 mL flasks containing the reaction mixture (total volume of 10 mL) including Na2HPO4-NaH2PO4 (100 mM, pH 7.0), 45 g (DCW)/L R. erythropolis XS1012, 1% (w/v) DESs, 150 mM BTAP and used 50 g/L glucose and isopropanol (10%, v/v) as co-substrates. The reaction mixture was incubated at 30 °C and 200 rpm for 24 h, and the product yield and e.e. values were determined by chiral gas chromatography analysis. Triplicate runs were performed for each trial.

GC Analysis Methods
Agilent Technologies 7820A GC system was used to determine the product yield and e.e. values, using a flame ionization detector and Varian CP-Chirasil-Dex CB column (25 m × 0.25 mm × 0.25 μm) with the split ratio at 15:1. The injector and detector temperature were both kept at 250 °C. The column temperature was maintained at 80 °C and kept constant for 2 min, then raised up to 180 °C with a heating rate of 5 °C/min. Nitrogen was used as a carrier gas at a flow rate of 2.0 mL/min. An internal standard method was adopted for the calculations. The results of chiral analysis of the bioconverted sample and standard reference were shown in Figure S1. The retention times for BTAP, dodecane, (S)-BTPE, and (R)-BTPE were 4.1 min, 8.8 min, 10.1 min, and 10.6 min, respectively.

Cell Membrane Permeability Assay
The cell membrane permeability of R. erythropolis XS1012 cells was assessed in various DESs-containing buffer systems. Conditions: 45 g (DCW)/L R. erythropolis XS1012 cells were suspended in 10 mL of Na2HPO4-NaH2PO4 (100 mM, pH 7.0) in the presence of various choline chloride-based DESs at a content of 1% (w/v) or Na2HPO4-NaH2PO4 buffer (100 mM, pH 7.0), 30 °C and 200 rpm for 24 h.

Metabolic Activity Retention (MAR) Assay
R. erythropolis XS1012 cells were pre-treated by 13 DESs (1%, w/v) respectively, with the addition of 150 mM BTAP or not, substrate or without for 3 h in 10 mL of Na2HPO4-NaH2PO4 (100 mM, pH 7.0), respectively. Cells (DESs-treatment with the substrate or without) were resuspended in 10 g/L glucose solution to start the sugar metabolism at 30 °C and 200 rpm for another 3 h. Then centrifugation at 9000 rpm and 4 °C for 10 min, the rest of the glucose content was analyzed by the biological sensing analyzer.

Impact of Vital Parameters on the Bioreduction
To determine the suitable conditions for the bioreduction of BTAP to (S)-BTPE in the [ChCl][U]-containing system, some key parameters, such as the content of the [ChCl][U], the suitable sort and addition of co-substrates, the type of buffer system, buffer pH, the concentration of substrate, and cell and reaction temperature were investigated. The reaction mixtures consisted of buffer (100 mM), the appropriate co-substrates, certain amount of R. erythropolis XS1012 cells,  Table 3 and (1:1)(1%, w/v)-containing system and Na2HPO4-NaH2PO4 (100 mM, pH 7.0), 50 g/L glucose and 15% isopropanol (v/v) as dual co-substrates in phosphate buffer system.

Conclusions
In the study, we screened 13 kinds of choline chloride-based DESs to examine whether they can enhance the product yield of the reduction, and investigated the effect of DESs on cell membrane permeability via the value of MAR to represent the cell activity. The [ChCl][U](1:1) was regarded as the most suitable DES. Subsequently, we investigated the key variables of the asymmetric reduction to make the reaction system more efficient. Isopropanol was agreed to be the co-substrate with the highest concentration, of 20% (v/v), which indicates that R. erythropolis XS1012 has benign tolerance to isopropanol. The optimal conditions were determined: 45 g (DCW)/L R. erythropolis XS1012, 150 mM BTAP, K2HPO4-KH2PO4 buffer (100 mM, 6.5), 20% (v/v) isopropanol as co-substrate, the addition of [ChCl][U](1:1) was 1% (w/v). The reaction was conducted at 30 °C, 200 rpm for 24 h. A yield of 80.7% was achieved under 300 mM BTAP. Through the research, it was proved that by introducing the DESs to the reaction medium could make a difference to the bioreduction. According to the research into the effect of the DESs on cells, this is probably due to DESs improving cell membranes or weakening the substrates' toxicity to cells. This research may enrich the investigation regarding the toxicity of DESs to cells and provide a valuable reference regarding the performance of DESs in bioreduction.