Experimental Section, Materials and Methods

General Information

All solvents, reactants and starting materials were received from commercial suppliers (Sigma-Aldrich, ChemPur, Thermo Fischer Scientific (Acros Organic)) and used as received. Ultrapure water (UPW, 0.06 µS⋅cm-1) was produced with an Ultra Clear Reinstwassersystem by SG Water (now Evoqua, Guenzburg, Germany) and used throughout this study. All experiments were carried out at atmospheric conditions. An IKA HS 260c shaker (IKA-Werke, Staufen, Germany) tempered with a Huber CC-K6 thermostat (Peter Huber Kältemaschinenbau AG, Offenburg, Germany) was used throughout all experiments for tempering and shaking of the reaction vessels at 180 rpm if not mentioned otherwise. Recombinant (S )-selective IRED fromPaenibacillus elgii B69 was transformed, cloned and overproduced in E. coli BL21 (DE3) as previously described .

Gaschromatography

The Conversion of all reactions was measured by gas chromatography with a Trace 1310 gas chromatograph from Thermo Scientific (Dreieich, Germany) with a flame ionization detector equipped with a HP-5 column from Agilent Technologies (30 m x 0.250 mm, 0.25 µm, 19091J-433, SN: USF724723H). Helium was used as the carrier gas (purity: 99.999%) with a flow rate of 0.75 mL⋅min‑1 was used for all measurements. Temperatures of the injector and detector were set to 250 °C. Temperature program: 40 °C for 10 min, followed by a heating rate of 125.0 °C⋅min-1 to 250 °C and a hold time for 5 min. The sample was injected within split-mode with a split flow of 7.5 mL⋅min-1 and a purge flow of 3 mL⋅min‑1. Authentic reference material was used for external quantification (see Figure S 1 to 4 for further information). All measurements were carried out in triplicate and the calculated standard deviations are shown.

General Procedure for Biotransformations

50 mg lyophilized whole E. coli cells containing the corresponding overproduced enzyme were rehydrated in 400 µL of a 100 mmol⋅L-1 sodium phosphate buffer (NaPi) pH 7.5 and shaken at 30 °C and 180 rpm in a TS-100 thermos shaker from Biosan. After 30 min, 100 µL of a 2‑MP substrate stock solution (250 mmol·L-1 in NaPi) and 50 mg of d‑glucose were added successively to the cell suspension (100 mmol⋅L-1final concentration of the substrate and 500 mmol·L-1of d‑glucose). A blank sample without the whole cell biocatalyst was prepared to ensure the absence of undesired reactions. The vials were tightly sealed and shaken at 30 °C and 180 rpm for 24 h.
After 24 h the reaction vessels were centrifuged for 5 min at 14000 rpm and 100 µL of the supernatant was transferred to 900 µL DCM together with 30 µL of a 10 m NaOH. The basified extraction was promoted by vigorous shaking, 250 µL of the DCM phase were transferred to GC Vials together with 25 µL of a solution containing n -heptane as a standard and afterwards analyzed by gas chromatography.
For the evaluation of the maximal substrate concentration, desired volumes of the substrate either from stock solution or the pure substance were mixed together with buffer and d‑glucose in a 500 µL-scale as mentioned above. Samples for GC were taken to investigate the reaction system at definite time intervals (conversion vs. time) and the final concentration of the substrate was set to 150 mmol·L-1.

Adsorption- and Desorption-Studies

The specific characteristics of the commercially available resins are shown in Table S1 (see SI). The exchange resins were used as received in their native ionic form.
To evaluate the capability for a DSP of the case studied biocatalytic IRED-catalyzed reaction, the capacity of the resins with respect to substrate and product were determined. A 100 mmol·L-1test solution of product and substrate in 100 mmol·L-1NaPi buffer solution together with 500 mmol·L-1 of d‑glucose was prepared for the adsorption experiments. 4 mL of this test solution were then added to 0.1 g, 0.5 g and 1.0 g of each resin, respectively. The shaking‑flask assays were performed at 30 °C and 180 rpm for 30 min. The adsorption of the substrate and the product were determined by analyzing the residual in the solution before and after adding the adsorbent resins via GC. 100 µL of the supernatant were transferred to 900 µL DCM together with 30 µL of a 10 m NaOH and further processed for the GC as mentioned above. For calculations of the adsorption, the measured concentration of the prepared test solution was set at 100% and compared to the amount of substrate after the adsorption.
After the successful adsorption of 2-MP and 2-MPN desorption experiments were executed with Amberlite IR-120(H), Dowex Monosphere 650C and Marathon MSC(H) cation exchange resins. Therefore, 0.5 g of each resin was loaded with the substrate and the product as mentioned above and after a filtration-step rinsed in 4 mL of a two phase system composed of 5% NaOH and cyclopentyl methyl ether (CPME) at a volumetric ratio of 1:1 at 30 °C and 180 rpm for 30 min. Afterwards, the concentrations of the substrate and the product in the ether phase were determined via GC as mentioned above.

Preparative Scale Experiment

For the preparative scale experiment, 15 g lyophilized whole E. coli cells containing the corresponding overproduced enzyme were rehydrated in 150 mL of a 100 mmol·L-1 sodium phosphate buffer (NaPi) pH 7.5 and stirred at 30 °C. After 30 min, 2.13 mL of 2‑MP and 15 g of d‑glucose were added successively to the cell suspension (150 mmol·L-1final concentration of the substrate and 500 mmol·L‑1of d‑glucose). The reaction mixture was stirred at 30 °C for 24 h. The conversion of the reaction was tracked after {0; 6; 17 and 24} h by removing 500 µL from the reaction broth and centrifuging it at 14000 rpm for five minutes. 100 µL of the supernatant were then proceeded and analyzed via GC as mentioned above.
After reaching a maximum conversion, 30 g Dowex Monosphere 650C cation exchange resin (0.2 g·mL-1) were added to the reaction broth and stirred for 30 min at 30 °C. A 500 µL-sample was taken and analyzed as mentioned before to check the full adsorption onto the resin. The cation exchange resin beads were filtered out of the reaction broth, and the remaining resins were desorbed in 75 mL of 5% NaOH and 75 mL of CPME at 30 °C for 30 min. To promote the phase separation, the reaction mixture was centrifuged for 5 min at 4000 rpm. The ether phase was dried with Na2SO4 for 2 h. Afterwards, the amine was oiled out with in situ generated HCl gas as amine hydrochloride and the remaining solvent CPME was removed by evaporation at 80 mbar. The purity of the viscous oil was checked by NMR (see SI).