Engineering carbonyl reductase for one-pot chemobiocatalytic enantioselective synthesis of a value-added N-containing chiral alcohol from N-acetyl-d-glucosamine

Abstract

Direct chemobiocatalytic conversion of biomass into value-added chemicals is promising yet challenging, because of the combined advantages of the two technologies and their incompatible issues. In this work, structure-guided engineering of carbonyl reductase from Streptomyces coelicolor (ScCR) was performed to improve its catalytic activity and stability, thus allowing for direct conversion of chitin-derived N-acetyl-D-glucosamine (NAG) into chiral 3-acetamido-5-(1-hydroxyethyl)furan (3A5HEF) by integrating its variants with chemical catalysts, with no isolation of intermediates. Upon three rounds of site-directed mutagenesis, two robust variants M3 (S167F/P168S) and M4 (S167Y/P168S) were identified, with 6-fold higher catalytic efficiency (k_cat/K_m) toward 3-acetamido-5-acetylfuran (3A5AF) than the parent enzyme and improved stability. Mechanistic insights into the improved activities of variant M3 were provided on the basis of the molecular docking study and molecular dynamics simulations. Chemobiocatalytic conversion of NAG into (R)-3A5HEF was performed via sequential chemical dehydration by tyrosine hydrochloride/CaCl₂ and biocatalytic asymmetric reduction by variant M4, with 53% yield and >99% ee. In the scale-up two-step synthesis, (R)-3A5HEF was obtained with 42% isolated yield. In addition to 3A5AF, greatly improved catalytic activities of the two variants toward other carbonyl compounds, particularly acetophenones, were observed (up to 41-fold higher activity than the parent enzyme). Various carbonyl compounds were reduced to the target chiral alcohols by the variants, with 46–99% yields and >99% ee. This work demonstrates that protein engineering is a powerful strategy to address the incompatibility between chemo- and biocatalysts in chemobiocatalysis.