Efficient chemo-enzymatic route for enantiopure S-Pregabalin by desymmetrization of 3-isobutyl glutarimide using an evolved D-hydantoinase

10

INTRODUCTION
Biocatalysis has become an enabling synthetic technology due to its potential to provide exquisite selectivity, evolvability, and inherent sustainability. Breakthroughs in bioinformatics, molecular biology, protein engineering and adjacent technology fields have enabled fast and reliable discovery and engineering of enzymes (1, 2) tailored towards specific process needs. As a result, an exploding number of new enzyme classes as well as chemoenzymatic processes with increasing sophistication have made their way to large-scale manufacture of chemicals and pharmaceuticals. An outstanding example is the production of (S)-3-isobutyl-γ-aminobutyric acid (S-Pregabalin, branded as Lyrica® by Pfizer), which has been a popular synthetic target for the biocatalysis community (3) since its launch in 2004. Pregabalin is a lipophilic γ-aminobutyric acid (GABA) analogue for the treatment of several nervous system disorders including epilepsy, neuralgia, fibromyalgia and anxiety disorders. The first commercial synthesis featured a classic resolution of racemic Pregabalin using (S)-mandelic acid in a three-step crystallization process (4), in which around 70% of the materials were lost. Given the broad clinical applications and significant market demand for S-Pregabalin, several chemoenzymatic pathways have been developed with the aim of enhancing efficiency, cost-effectiveness, and environmental sustainability. For example, a team at Pfizer reported (5) a biocatalytic kinetic resolution process in which the (S)-cyanodiester was selectively hydrolyzed by the commercial Thermomyces lanuginosus lipase (TLL) and the unreacted (S)-cyanodiester was recycled by base-catalyzed epimerization. This route could be even improved in an academic study using an engineered TLL variant (6) with improved catalytic efficiency. Following a similar concept, Zheng et al. (7, 8) devised an impressive regio- and enantioselective hydrolysis of isobutylsuccinonitrile catalyzed by Escherichia coli cells harboring nitrilase BrNIT from Brassica rapa to provide (S)-3-cyano-5-methylhexanoic acid ((S)-CMHA), a critical chiral intermediate of S-Pregabalin. In particular, the above cells were immobilized to simplify the downstream process and reused for 12 consecutive batches while maintaining performance (>41.1% conversion, >98%ee); the E-factor for the whole process was calculated to be 3.72, much lower than that of the previous processes. Despite these enzymatic resolution processes significantly reduced chemical usage and synthesis cost, chemical re-racemization of the undesired enantiomer is still needed to achieve high atom economy, which requires harsh conditions and is therefore not environmentally friendly. Alternatively, enzymatic desymmetrization (9) using amidase for enantioselective ring opening of a symmetrical prochiral cyclic imine has proven to be a powerful strategy to produce chiral pharmaceutical compounds with a theoretical yield of 100%. Nojiri et al. (10) discovered that amidase from Comamonas sp. KNK24-9 hydrolyzed 3-isobutyl glutaric acid diamide (IBD) to produce (R)-3-isobutyl glutaric acid monoamide (R-IBM), a direct precursor of S-Pregabalin, with 94% yield, >99% ee. Variants of the acting amidase CoAM with single amino acid exchanges were later designed, which enabled complete conversion of IBD at 107.4 mmol l-1 to R-IBM with 98.9% ee. In an effort to access β-substituted GABA derivatives, Yitao et al. (11) screened a panel of nitrilases and found BjNIT6402 catalyzed desymmetric hydrolysis of 3-isobutylglutaronitrile to yield optically active 3-(cyanomethyl)-5-methylhexanoic acid with quantitative conversion and 90% ee; the subsequent Curtius rearrangement of the cyanocarboxylic acid intermediate, followed by acidic hydrolysis afforded S-Pregabalin.