Structure-guided engineering of ChKRED20 from Chryseobacterium sp. CA49 for asymmetric reduction of aryl ketoesters.

ChKRED20 is a robust NADH-dependent ketoreductase identified from the genome of Chryseobacterium sp. CA49 that can use 2-propanol as the ultimate reducing agent. The wild-type can reduce over 100 g/l ketones for some pharmaceutical relevant substrates, exhibiting a remarkable potential for industrial application. In this work, to overcome the limitation of ChKRED20 to aryl ketoesters, we first refined the X-ray crystal structure of ChKRED20/NAD+ complex at a resolution of 1.6 Å, and then performed three rounds of iterative saturation mutagenesis at critical amino acid sites to reshape the active cavity of the enzyme. For methyl 2-oxo-2-phenylacetate and ethyl 3-oxo-3-phenylpropanoate, several gain-of-activity mutants were achieved, and for ethyl 2-oxo-4-phenylbutanoate, improved mutants were achieved with kcat/Km increasing to 196-fold of the wild-type. All three substrates were completely reduced at 100 g/l loading catalyzed with selected ChKRED20 mutants, and deliver the corresponding chiral alcohols with >90% isolated yield and 97 - >99%ee.

Crystal structure and iterative saturation mutagenesis of ChKRED20 for expanded catalytic scope.

ChKRED20 is an efficient and robust anti-Prelog ketoreductase that can catalyze the reduction of ketones to chiral alcohols as pharmaceutical intermediates with great industrial potential. To overcome its limitation on the bioreduction of ortho-substituted acetophenone derivatives, the X-ray crystal structure of the apo-enzyme of ChKRED20 was determined at a resolution of 1.85 Å and applied to the molecular modeling and reshaping of the catalytic cavity via three rounds of iterative saturation mutagenesis together with alanine scanning and recombination. The mutant Mut3B was achieved with expanded catalytic scope that covered all the nine substrates tested as compared with two substrates for the wild type. It exhibited 13-20-fold elevated k cat/K m values relative to the wild type or to the first gain-of-activity mutant, while retaining excellent stereoselectivity toward seven of the substrates (98-> 99% ee). Another mutant 29G10 displayed complementary selectivity for eight of the ortho-substituted acetophenone derivatives, with six of them delivering excellent stereoselectivity (90-99% ee). Its k cat/K m value toward 1-(2-fluorophenyl)ethanone was 5.6-fold of the wild type. The application of Mut3B in elevated substrate concentrations of 50-100 g/l was demonstrated in 50-ml reactions, achieving 75-> 99% conversion and > 99% ee.

Exploration of a N-terminal disulfide bridge to improve the thermostability of a GH11 xylanase from Aspergillus niger.

To improve the thermostability of xylanase XynZF-2 from Aspergillus niger XZ-3S, a disulfide bridge was introduced in the N-terminal domains by site-directed mutagenesis (V1C and E27C). Simultaneously, the active sites of XynZF-2 were predicted by bioinformatics software and verified by site-directed mutagenesis (E103D and E194D). The mutated active sites xynED- and the mutated disulfide bridge xynDC-encoding genes were constructed and expressed in Escherichia coli BL21 (DE3). Compared to the native xylanase, it was found that the residual activity of the mutated XynED was 0.17%. The optimum temperature of the variant XynDC was increased to 45°C from 40°C of XynZF-2. After treatment at 40°C for 60 min, the variant XynDC retained 66.77% of their original activity, while the XynZF-2 retained about 44.36% residual activity. t1/2(45°C) of the variant XynDC also increased from 7 min to 14 min. The results of the mutated xylanases indicated that the active center of XynZF-2 mainly consisted of two catalytic residues (Glu103 and Glu194), and the introduction of a disulfide bridge in the N-terminal domains can improve the thermostability of XynZF-2.