COF-derived photocatalytic artificial enzymes-assisted precision analysis of active ingredients in natural product extracts.

Discovery of active ingredients from complex natural products remains a critical challenging task in drug development, primarily due to the lack of rapid and precise analytical tools for bioactive compound screening. To address this challenge, we developed a biosensing-assisted computer-aided screening strategy. A covalent organic framework (COF) derived photocatalytic artificial oxidase was synthesized with robust catalytic capability toward TMB under natural light excitation. Leveraging arbutin as the chromogenic substrate, a sensitive biosensing assay was developed for α-glucosidase (α-Glu) activity detection with a linear range of 2-30 μg mL-1, enabling precise screening of active inhibitors within complex natural extract. This approach facilitated the successful identification of bioactive extracts from 25 natural product candidates. Subsequently, combined with computer-aided virtual screening, two active monomers (Honokiol and Magnolol) were rapidly resolved and characterized, with IC50 values of 175 ± 5.62 μM and 991.9 ± 36.5 μM against α-Glu, respectively. The screened active monomers exhibited significant inhibitory activity against α-Glu. Compared to conventional methods, the biosensing-assisted strategy effectively constrains the analytical scope and minimizes false-positive results, whereas the computational approach obviously accelerates the identification and analysis efficiency of active monomers. This study provides a new strategy for efficiently and accurately discovering bioactive compounds in complex systems, with promising prospects for high-throughput drug screening.