High-Throughput Design of Regulatory Protein-Based Biosensors for Screening Biosynthesis Libraries

In vivo high-throughput screening systems for detecting select metabolites are useful for engineering enzymes and flux through biosynthesis pathways. Transcriptional regulator proteins serve as gene switches in microorganisms and can be applied as highly specific and sensitive endogenous molecular biosensors. For cases where there is no known regulator that responds to a metabolite of interest, protein engineering may be used to alter inducer specificity of an existing biosensor system. We combine ligand binding pocket saturation mutagenesis with Fluorescence Activated Cell Sorting based on fluorescent reporter protein expression to generate novel molecular biosensors based on the AraC regulatory protein from E. coli. We have isolated AraC variants with altered specificity toward a variety of small molecules such as mevalonate, p-coumaric acid, triacetic acid lactone (TAL), and other 2-pyrones. These novel biosensors and their use in screening for improved biosynthesis of the corresponding effector will be described.

Through the course of biosensor engineering we have optimized parameters to improve the design process and streamline high-throughput screening efforts.  Examples include frequency of positive vs. negative sorts, time to sort induced cells, method of recovering post-sort clones, and half-life of the reporter protein.  Next-Generation-Sequencing (NGS) was also used to monitor AraC library sequence evolution toward responsiveness to various ligands.  Key results from these studies will be presented.  Finally, a recently solved crystal structure of one AraC ligand binding domain variant (responding to TAL) reveals determinants of molecular recognition, and this will be discussed.