Designing a DeepPCD Strategy for Efficient Engineering Nitrilases toward Industrial Bioconversion

Abstract

The integration of artificial intelligence with protein engineering has created unprecedented opportunities for enzyme discovery and the enhancement of overall catalytic performance. In this study, a novel nitrilase, BhNIT, was identified from Bradyrhizobiumhuanghuaihaiense through a computationally assisted screening strategy. Using BhNIT as the target protein for engineering, we established DeepPCD as an integrated computationally guided workflow that organizes sequence-design, structure-filtering, stability-assessment, evolutionary-compatibility, and molecular dynamics analyses to prioritize experimentally testable mutations. Compared with approaches that require extensive experimental screening of large mutant libraries, the DeepPCD workflow facilitates efficient enzyme engineering modifications. By modifying only seven candidate residues of BhNIT, we obtained a mutant exhibiting a 1.3 × 105-fold improvement in hydrolytic efficiency. In addition, this mutant showed a 20.4°C increase in melting temperature and a broader operational pH range extending toward more alkaline conditions, substantially enhancing its potential for industrial applications. Molecular dynamics simulations and structural analysis revealed that the enhanced performance stems from effective reshaping of the substrate-binding pocket and access channel, along with a strengthened interaction network. Furthermore, preliminary computational application to additional nitrilases suggested the potential transferability of this workflow within the nitrilase family. Overall, this work provides an experimentally validated case study for computationally guided nitrilase engineering and offers a practical workflow for prioritizing mutations that may improve activity and stability, although broader generalization will require further validation across more diverse enzyme systems.