Enzymatic Enantioselective Decarboxylative Protonation of Heteroaryl Malonates

The enzyme aryl/alkenyl malonate decarboxylase (AMDase) catalyses the enantioselective decarboxylative protonation (EDP) of a range of disubstituted malonic acids to give homochiral carboxylic acids that are valuable synthetic intermediates. AMDase exhibits a number of advantages over the non-enzymatic EDP methods developed to date including higher enantioselectivity and more environmentally benign reaction conditions. In this report, AMDase and engineered variants have been used to produce a range of enantioenriched heteroaromatic α-hydroxycarboxylic acids, including pharmaceutical precursors, from readily accessible α-hydroxymalonates. The enzymatic method described here represents an improvement upon existing synthetic chemistry methods that have been used to produce similar compounds. The relationship between the structural features of these new substrates and the kinetics associated with their enzymatic decarboxylation is explored, which offers further insight into the mechanism of AMDase.

2-Bromobenzofuran (0.275 g, 1.36 mmol) and anhydrous diethyl ether (7.0 mL) were added to a flame-dried flask, under nitrogen, cooled to -78 °C before the dropwise addition of nBuLi (1.6 M in hexanes, 0.851 mL, 1.36 mmol). The resulting mixture was stirred for 20 minutes at -78 °C and then a solution of diethyl ketomalonate 6 (0.287 g, 1.64 mmol) in S4 anhydrous diethyl ether (3.0 mL) was added dropwise at -78 °C. The reaction mixture was allowed to warm to room temperature, then stirred overnight, before being quenched with saturated ammonium chloride solution (5.0 mL). The mixture was then extracted with ethyl acetate (3 x 30 mL) and the combined organic extracts were then dried over anhydrous MgSO 4 , before being evaporated under reduced pressure, to give the crude product as an oil.
The product was purified by flash chromatography using a 10:1 mixture of hexane and ethyl acetate as the eluent to give 21 (0.15 g, 38%) as a pale yellow oil.
10% of this solution was then added to a flask containing diethyl ether (2 mL), pre-heated magnesium turnings (0.427 g, 17 mmol) and a few iodine crystals. This mixture was heated and allowed to reflux until the iodine colour faded, after which the rest of the 5bromobenzothiophene solution was added in 10% portions. When the addition was complete, the mixture was heated under reflux for 2 hours, after which time the solution was cooled to -78 °C. Diethyl ketomalonate 6 (0.342 g, 1.966 mmol) was then added dropwise to the solution, which immediately turned orange in colour. The solution was allowed to warm to room temperature and then stirred overnight, before being quenched with a saturated ammonium chloride solution (10 mL). The resulting mixture was extracted with diethyl ether (3 x 30 mL) and the combined organic layers were then washed with brine (10 mL), dried over anhydrous magnesium sulphate, and solvent subsequently removed under reduced pressure. The crude product was purified by flash chromatography using a 5:1 mixture of hexane and ethyl acetate as the eluent to give malonate 27 (0.12 g, 20%) as a yellow oil α-Furyl-α-hydroxy malonic acids (8a-e). The general procedure used to synthesise the αfuryl-α-hydroxy malonic acids 8a-e via hydrolysis of their respective diethyl malonates will be illustrated by the synthesis of α-(furan-2-yl)-α-hydroxymalonic acid 8a.
After this, a solution of α-acetoxy-α-(furan-2-yl) diethyl malonate 10 (0.10 g, 0.35 mmol) in dry toluene (10 mL) was added. The reaction mixture was stirred for a further 60 minutes before being filtered to remove any remaining sodium metal. The filtrate was diluted with aqueous HCl (30 mL 10% v/v) before being extracted with diethyl ether (3 x 30 mL). The combined organic extracts were dried over MgSO 4 and then evaporated to yield the crude α- α-(benzofuran-2-yl)-α-hydroxy diethyl malonate 21 (0.100 g, 0.342 mmol) in dichloromethane (4.5 mL) was added to a solution of sodium hydroxide (0.030 g, 0.753 mmol) in methanol (0.5 mL) in a centrifuge tube. The mixture was allowed to stand for 3 hours, after which a precipitate had formed. The tube was centrifuged (5000 RPM, 3 minutes), the supernatant was poured off, and the solid pellet that had formed was washed with ethyl acetate (4 mL) and diethyl ether (8 mL). The remaining solvent was removed in vacuo to yield the disodium salt of α-(benzofuran-2-yl)-α-hydroxy malonic acid 22 (0.085 g, 90%) as a white powder. Screening malonic acid derivatives as substrates for AMDase: New AMDase substrates were initially identified using the BTB colorimetric assay previously reported. [1,2] In a 96well plate, 10 µL of the candidate substrate (0.5 M in 25 mM TRIS buffer at pH 7) was added to 185 µL of a BTB-containing buffer solution (0.01% BTB in 10 mM MOPS buffer at pH 7.2). The 96-well plate was then incubated at 37 °C for 15 minutes before the addition of AMDase (ca. 1 µM). The absorbance at 620 nm was recorded over the course of 6 hours using a UV-Vis photospectrometer, with an increase in absorbance being indicative of a positive result. In order to verify positive hits from the colorimetric assay, a second assay was performed using 1 H NMR. In an Eppendorf tube, 50 µL of the candidate substrate (0.5 M in 25 mM TRIS buffer at pH 7) was added to 925 µL of TRIS buffer (25 mM at pH 7). The Eppendorf tube was then incubated at 37 °C for 15 minutes before the addition of AMDase (ca. 10 µM). The solution was incubated overnight at 37 °C before being frozen and lyophilised, with the resultant solids being dissolved in an appropriate deuterated solvent and submitted for 1 H NMR analysis.

Calculation of kinetic parameters (K m and k cat ):
In order to calculate the kinetic parameters associated with a particular substrate, varied concentrations of the substrate were analysed using the colorimetric assay conditions described previously. [1,2] The change in absorbance (ΔA) at 620 nm for each substrate concentration was tracked over a specific time period (Δt) using a UV-Vis photospectrometer. The rate of decarboxylation ( ) at each substrate concentration was then determined, allowing for the calculation of K m and k cat using standard Michaelis-Menten kinetics. Confirming the absolute configuration of AMDase produced α-hydroxy carboxylic acids using Mosher esters: The configuration of the enzymatically produced α-hydroxy carboxylic acids is based on the stereochemical course of AMDase catalysed decarboxylation reaction as determined previously by detailed labelling experiments and high resolution X-ray structures of AMDase (Fig. 1) [1][2] along with the configuration of many AMDase products as determined previously. [1][2][3][4][5][6][7] Comparison of optical rotations of the chiral products with literature [α] D values where these are available (compounds 23, 26 & 29) are also consistent with the stated absolute (see page S12-S13). In addition the absolute configuration α-hydroxy carboxylic acids can be confirmed via the preparation of their respective Mosher esters [25] , followed by a comparison of their TLC retention time with Mosher esters formed using racemic α-hydroxy carboxylic acids (Fig. S1). The general procedure used will be illustrated by the example of α-hydroxy-α-(thiophen-2-yl) acetic acid 19.
This resulted in the resolution of two distinct bands after approximately 4 hours. These bands were cut out from the TLC plate, and each band was seperately submerged in methanol to recover the purified diastereoisomer. This solution was then filtered to remove any excess silica, before the methanol was removed in vacuo. Each sample was then characterised using 1 H NMR, with the difference in the position of the heterocyclic signals between the diastereoisomers being used to determine the stereochemistry of the C α -O bond. This 1 H NMR analysis showed that the top band isolated via TLC was (R)-(S) configured and the bottom band was (S)-(S) configured (Table S1).