分子动力学
合理设计
芳基
化学
基质(水族馆)
酮
组合化学
还原酶
醇脱氢酶
立体化学
酶
计算化学
材料科学
纳米技术
有机化学
生物
烷基
生态学
作者
Bingmei Su,Ze-Hui Shao,Aipeng Li,Muhammad Naeem,Juan Lin,Lidan Ye,Hongwei Yu
出处
期刊:ACS Catalysis
[American Chemical Society]
日期:2019-12-04
卷期号:10 (1): 864-876
被引量:99
标识
DOI:10.1021/acscatal.9b04778
摘要
Inspired by the conformational change of the enzyme–substrate complex in molecular dynamics (MD) simulation with distance restriction, we propose a strategy for identifying the engineering targets based on the comparative analysis of enzyme-/substrate- binding modes in MD simulations with and without distance restriction (prereaction-state simulation and free-state simulation). Taking the short-chain dehydrogenase/reductase (SDR) mutant EbSDR8-G94A/S153L (Mu0) with poor activity toward bulky aryl ketones as an example, H145 and Y188 were identified as the engineering targets due to the distinct conformation difference in the two simulation modes. To break the “beam” structure formed by these residues at the entry of cavity C2 in free-state simulation, the substrate-binding pocket was reconstructed, and meanwhile the relative size of cavities C1 and C2 was modulated to improve the enantioselectivity. In this way, mutants for efficient asymmetric reduction of o-halogenated acetophenones, propiophenones, aromatic ketoesters, and diaryl ketones were designed, delivering chiral alcohols with >99% conversion and >98% ee. The effectiveness of this design strategy was also validated by the successful redesign of PpYSDR, generating a variant for efficient reduction of (4-chlorophenyl) 2-pyridyl ketone into the S -product with >99% conversion and 96% ee. MD simulations suggested a suitable binding pocket with proper energy contribution as the ubiquitous mechanism for the improvement of activity and enantioselectivity toward substrates with varied structures. The success in this study provides hints for the rational design of alcohol dehydrogenases/reductases with both a broad substrate spectrum and high enantioselectivity.
科研通智能强力驱动
Strongly Powered by AbleSci AI