Mining and Engineering New Biocatalysts for Sustainable Chemoenzymatic Synthesis

Enzymes can catalyze many reactions that produce medicines, agricultural chemicals, and consumer products. They are more efficient and less harmful to the environment than other catalysts. There are three major reasons for this. First, enzymes operate at room temperature, which reduces energy costs and emissions associated with energy generation. Second, enzymes do not require the use of organic solvents. Third, enzymes have a high degree of product specificity. This project will focus on enzymes that form carbon-nitrogen bonds. These bonds are the basis of a class of compounds known as amides. Amide groups are found in many everyday materials and life-saving drugs. The project will discover and characterize classes of enzymes that can form amide bonds efficiently under mild reaction conditions. The resulting biotechnology could improve the understanding of how enzymes work and lead to advanced biomanufacturing of high-value medicines and advanced materials. This project will discover and characterize ATP-grasp and amide bond synthetase enzymes from fungi. These enzymes are particularly well-suited as biocatalysts because they are standalone enzymes that can catalyze both carboxylate activation and amidation with an amine, bypassing the need for multi-enzyme cascades that are more commonly observed in nature. The discovery of these enzymes from sources such as fungi is limited, because the protein sequence-activity relationship is not well understood. Computational and machine learning methods will be developed to fully explore the diversity of these enzymes. Machine learning and AI-based structural prediction tools will be implemented to generate deep insights into the sequence-structural-functional relationships of these enzymes. The primary goal will be to distinguish amide bond forming enzymes from other ATP-dependent enzymes that do not form amide bonds. The proposed computational approaches can be extended to the mining of other families of enzymes and can be generally useful to the biocatalysis and biosynthesis fields. The second objective will involve structural and biochemical analyses on the predicted amide-bond forming enzymes and use the data to refine the computational mining approaches. The substrate scope of enzymes will be investigated and enzyme structural features that are determinant of substrate specificity will be identified. The resulting knowledge will enable high-resolution prediction of the substrate specificity and engineering of the enzymes for desired catalytic scope. In addition, associated natural products from these pathways that encode amide bond forming enzymes will be elucidated. The third objective will involve assembling a library of amide-bond forming enzyme and establishing a public database that can be searched for desired nucleophilic and electrophilic substrates. This database will contribute substantially towards optimization of amide bond forming reactions in drug and materials synthesis. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.