北京化工大学最新Science论文：吴边/崔颖璐团队利用AI挖掘出聚氨酯塑料降解酶

Core Viewpoint - The article discusses a significant advancement in the recycling of polyurethane plastics through the development of a highly active urethanase enzyme, Ab PURase, which can nearly completely degrade commercial polyurethane materials in just 8 hours, highlighting the potential of AI in identifying effective biocatalysts for industrial applications [2][6]. Group 1: Research Development - The research team from Beijing University of Chemical Technology and the Institute of Microbiology, Chinese Academy of Sciences, published a paper in Science on October 30, 2025, focusing on the development of a framework called GRASE (GNN-based Recommendation of Active and Stable Enzyme) for screening enzymes with potential activity [2][4]. - GRASE combines self-supervised and supervised learning to identify efficient and glycolysis-compatible urethanases, addressing the challenges posed by the difficult-to-degrade chemical bonds in polyurethanes [4][6]. Group 2: Enzyme Characteristics - Ab PURase, derived from Alicyclobacillus sp., exhibits an activity level 465 times higher than known urethanases in a 6 molar diethylene glycol solution, enabling nearly complete depolymerization of kilogram-scale commercial polyurethane within 8 hours [5]. - Structural analysis indicates that the enzyme's stability and efficiency in harsh solvents are likely due to its tightly packed hydrophobic core and a Lid Loop structure stabilized by proline [5]. Group 3: Industrial Implications - This research marks the first successful large-scale biological depolymerization of polyurethane under industrial conditions, providing a new, efficient, and sustainable pathway for the green recycling of polyurethane plastics [6]. - The findings underscore the significant potential of artificial intelligence in accelerating the discovery of biocatalysts with industrial application potential [6].