Automated electrochemistry platforms provide a pathway for speeding up the discovery and optimization of electrochemical systems, especially when a broad range of molecules and reaction conditions must be explored.1-3 Additionally, incorporation of machine learning enables closed-loop optimization campaigns, where the obtained experimental data guides the selection of the next set of experimental parameters to test. This type of autonomous exploration is ideally suited for the study of enzyme electrocatalysis, as the combinatorics of different enzymes, substrates, and redox mediators creates a prohibitively large parameter space. Even within a single enzyme class, mutations to the amino acid sequence can significantly impact rate, function, and selectivity of the catalyzed reaction. Thus, an automated electrochemical platform that enables rapid sampling both reaction parameters and amino acid sequence space would enable discovery of new, non-native enzymatic reactions.In this work we use an open source and modular automated electrochemistry platform to study enzyme electrocatalysis in several common bioelectrocatalytic systems, such as glucose oxidase, glucose dehydrogenase, and aldehyde dehydrogenase. We use automated cyclic voltammetric substrate titration experiments (Figure 1) to gain insight into the kinetics of the redox-mediated enzymatic catalysis and demonstrate the reaction conditions in which traditional mechanistic simplifications, like Michaelis-Menten, break down. Additionally, the automated electrochemistry platform allows for the rapid screening of different substrate molecules, enabling the quantification of substrate promiscuity in these enzymatic catalysts. We show how Bayesian Optimization can be incorporated into our workflow to autonomously tune solution parameters to promote increased reaction rates. These studies demonstrate the power of autonomous electrochemistry for understanding and screening biocatalytic systems and represent a significant step towards totally machine-guided biocatalyst evolution.[1] Oh, I. ; Pence, M. A.; Lukhanin, N.; Rodriguez, O.; Schroeder, C. M.*; Rodriguez-Lopez, J.*, “The Electrolab: An open-source, modular platform for automated characterization of redox-active electrolytes” Device, 2023, 1, 5, 100103[2] Rodriguez, O.; Pence, M. A.; Rodriguez-Lopez, J.*, “Hard Potato: An Open Source Python Library to Control Commercially Available Potentiostats and Automate Electrochemical Experiments” Anal. Chem., 2023, 95, 4840–4845[3]Pence, M. A.; Rodriguez, O.; Lukhanin, N.; Schroeder, C. M.*; Rodriguez-Lopez, J.*, “Automated Measurement of Electrogenerated Redox Species Degradation Using Multiplexed Interdigitated Electrode Arrays”, ACS Meas. Sci. Au, 2023, 3, 62–72 Figure 1
Read full abstract