(87a) ­Engineering Novel Modular Biosensors to Confer Metabolite-Responsive Regulation of Transcription

Efforts to engineer microbial factories have benefitted from increasingly facile technologies for synthesizing novel enzymatic ensembles and mutating entire organisms, yet screening and optimizing metabolic pathways remain rate-limiting steps. Metabolite-responsive biosensors may help to address these persistent challenges by enabling the monitoring of metabolite levels in individual cells and the implementation of metabolite-responsive feedback control. To date, however, we are currently limited to the use of naturally-evolved biosensors, which are insufficient for monitoring many metabolites of interest.

To help address this need for novel biosensors, we have developed a modular and general approach for engineering novel transcriptional regulators that are actuated by the presence or absence of metabolites of interest â?? the Small molecule Responsive Transcriptional Regulator (SmaRTR) biosensor platform. The central goal of this approach is to develop a systematic strategy for converting proteins that bind small molecules into proteins that regulate gene expression in a manner that is regulated by the presence of those small molecule analytes. In our initial evaluation of this strategy, we utilized the BCR-ABL1 zinc finger protein as a model DNA-binding domain. We built and characterized a library of promoters including BCR-ABL1 binding sites at various positions to elucidate which promoter design features are most important for achieving effective zinc finger mediated-transcriptional repression. Using this rich data set, computational feature identification analysis was applied to quantitatively assess the contribution of each design feature to promoter repressibility, ultimately generating new insights into promoter output tuning.

We next evaluated several strategies for converting the well-characterized maltose binding protein (MBP), a model ligand-binding protein, into a novel maltose-actuated transcription factor that regulates gene expression in a ligand-dependent manner. Through this approach, several robust and functional novel biosensors were identified, establishing the fundamental feasibility of the SmaRTR approach. Moreover, we evaluated a series of strategies for design-driven tuning of biosensor and promoter performance and ultimately identified biosensor systems that substantially outperformed our initial designs. Finally, we developed methods for evaluating the extent to which our biosensor design strategy is generalizable to the conversion of ligand-binding proteins into biosensors. Ultimately, this technology may provide new tools for high throughput screening and improved biomanufacturing through implementation of metabolic feedback control.