Expanding the biotechnological scope of metabolic sensors through computation-aided designs

Abstract

Metabolic sensors are microbial strains modified so that biomass formation correlates with the availability of specific metabolites. These sensors are essential for bioengineering (e.g. in growth-coupled designs) but creating them is often a time-consuming and low-throughput process that can potentially be streamlined by in silico analysis. Here, we present the systematic workflow of designing, implementing, and testing versatile Escherichia coli metabolic sensor strains. Glyoxylate, a key metabolite in (synthetic) CO2 fixation and carbon-conserving pathways, served as the test molecule. Through iterative screening of a compact metabolic model, we identified non-trivial growth-coupled designs that resulted in six metabolic sensors with a wide sensitivity range for glyoxylate, spanning three orders of magnitude in detected concentrations. We further adapted these E. coli strains for sensing glycolate and demonstrated their utility in both pathway engineering (testing a key metabolic module via glyoxylate) and applications in environmental monitoring (quantifying glycolate produced by photosynthetic microalgae). The versatility and ease of implementation of this workflow make it suitable for designing and building multiple metabolic sensors for diverse biotechnological applications.