Development of a vanillate biosensor for the vanillin biosynthesis pathway in E. coli

Abstract

Genetically encoded small molecule sensors can facilitate metabolic engineering by enabling high-throughput detection of metabolite concentrations, directed evolution of host and pathway enzymes, and dynamic regulation. The engineered de novo vanillin biosynthesis pathway assembled in Escherichia coli is industrially relevant and ideal for biosensor deployment given that the pathway requires only three heterologous enzyme-catalyzed reactions, generates naturally occurring metabolites, and may benefit from dynamic regulation. However, pathway flux is stalled and diverted by the activity of the Homo sapiens catechol O-methyltransferase, which is intended to catalyze the conversion of protocatechuate to vanillate. To confront this challenge, we constructed and applied a vanillate sensor based on the Caulobacter crescentus VanR-VanO system. Using components from a previously characterized E. coli promoter library, we achieved greater than 14-fold dynamic range in our best rationally constructed sensor. We characterized sensor substrate specificity and found that this construct and an evolved variant are remarkably selective, exhibiting no detectable response to the regioisomer byproduct isovanillate. We then harnessed the evolved biosensor to conduct rapid bioprospecting of natural catechol O-methyltransferases. We identified eight that appear to have greater desired activity than the originally used variant, including three previously uncharacterized O-methyltransferases. Collectively, these efforts enrich our knowledge of how biosensing can aid metabolic engineering and constitute the foundation for future improvements in vanillin pathway productivity.