On the Fly Determination of the Substrate Activation in Hydrolases Using a Neural Network

Active sites of enzymes are able to activate substrates and perform chemical reactions that cannot occur in solutions. We focus here on hydrolysis reactions catalyzed by enzymes and initiated by the nucleophilic attack of the carbonyl carbon atom of the substrate. From the electronic structure standpoint, substrate activation can be characterized in terms of the Laplacian of the electron density. This is a simple and easily visible imaging that allows one to “visualize” electrophilic site on the carbonyl carbon atom, that is presented only in the activated species. The efficiency of the substrate activation by the enzyme can be quantified from the ratio of reactive and nonreactive states from the QM/MM MD trajectories. We propose a neural network that assigns the Laplacian of electron density maps to the reactive and nonreactive species. The neural network is trained on the cysteine protease enzyme-substrate complexes and successfully validated on the zinc-containing hydrolase, thus showing a wide range of applications of the proposed approach.