Computational Design of Synthetic Co-Transcriptional RNA Elements for the Dynamic Regulation of Gene Expression

David Sparkman-Yager and James Carothers

Recent developments in the fields of synthetic biology and metabolic engineering have opened the doors for the microbial production of valuable organic compounds. However, these developments have also highlighted the need for genetic controllers able to respond to the concentrations of targeted metabolites. Ligand-responsive self-cleaving RNA elements (aptazymes) provide one potential solution to this need, as well as an ideal tool for the study of the engineering of ligand-dependent controllers. As RNA elements they are genetically-compact and, because they rely on an autocatalytic backbone-cleavage reaction, can theoretically be employed in any host organism using the native biochemical mechanisms. Conventional aptazymes identified to date, generated almost-entirely through selection methods, suffer from limited dynamic ranges (< 100), and cannot be tuned to respond to targeted metabolite concentration. To overcome these limitations we have developed a kinetic computational design methodology to create aptazymes with unprecedented in vitro dynamic ranges (up to 240) that utilizes the directionality of RNA transcription to dictate device function (SCoR-Rs). Screening up to 10⁷ sequences per construct in silico, we have designed, synthesized, and characterized >50 SCoR-Rs responsive to either the small-molecule theophylline or p-aminophenylalanine (p-AF). Kinetic design metrics allow the prediction of background cleavage rates over 5 orders of magnitude. Both Theophylline and p-AF-responsive SCoR-Rs have been successfully incorporated into the 5’-UTR of E. coli transcripts for the ligand-dependent regulation of gene expression levels. We are currently interrogating the ability of “Timer” domains inserted within the SCoR-R devices to increase the time available for co-transcriptional ligand-binding, and therefore device sensitivity, to allow greater utility of these genetic controllers in non-ideal conditions. We are additionally applying these same design principles to design other RNA elements such as ligand-inducible transcriptional terminators (SCoR-Ts) and selectively-occluded RBSs (SCoR-BSs).