(622e) High-Throughput Promoter Optimization for Improved Biobutanol In Vivo Biosensor

Bacterial transcription factor-promoter pairs exist in nature to allow dynamic regulation of gene expression based on the cellular environment, including carbon and energy sources. This capability can be useful in the development of in vivo biosensors and synthetic regulons. These transcription factor-promoter interactions, however, are not optimized for biotechnological applications and require further investigation. Protein−DNA binding interactions can be elucidated and ultimately manipulated by quantifying the sequence−function relationship of promoter DNA. A method for elucidating sequence−function relationships employs fluorescence-activated cell sorting (FACS) and high-throughput sequencing (called “sort-seq”). There is industrial interest in butanol as a renewable fuel source. Biobutanol production from renewable feedstocks has been demonstrated, but further improvements to productivity are required for economic viability. High-throughput screens on non-growth-related phenotypes and dynamic butanol-dependent regulation represent powerful metabolic engineering strategies that are largely unavailable to these efforts. This capability gap is due to a lack of inducible transcription factor/promoter pairs with user-defined controls. A butanol-responsive transcription factor, BmoR, and its cognate promoter PBMO have been previously described in the native form, but PBMO remains relatively uncharacterized and optimizing its function by sequence modification has not been attempted.

In this work, we demonstrate the engineering of the PBMO promoter at the nucleotide level to improve biosensor characteristics, specifically an improved dynamic range, and to generate synthetic promoters. To this end, we use massively parallel reporter assays to study the sequence-function relationship of PBMO using the ‘sort-seq’ method. A mutagenized PBMO library cloned upstream of gfp in E. coli was induced with butanol and sorted into activity-based (i.e., fluorescence-based) populations. These populations were deep sequenced and mutations affecting fluorescence were identified. We confirmed the efficacy of identified mutations through testing PBMO variants generated through site-directed mutagenesis of PBMO. These variants contained single, multiple, and structural mutations. Transcription factor-promoter binding interaction was confirmed via surface plasmon resonance assay. The best performing engineered PBMO promoter improved the dynamic range 4-fold over the original. We also discuss the application of this method to additional transcription factor-promoter systems.