Automated Physics-Based Design of Synthetic Riboswitches from Diverse Aptamers

Riboswitches are RNA-based sensors that use an aptamer domain to bind ligand, change shape, and alter gene expression levels; typically, by modulating the translation or transcriptional termination of mRNAs. While natural riboswitches have evolved as exquisite sensors, it remains difficult to engineer non-natural riboswitches that utilize different aptamers to detect and respond to ligands of interest, even though their application as biosensors, medical diagnostics, and metabolic controllers would be transformative. Previously, we developed a statistical thermodynamic model of ribosome-mRNA interactions that predicts translation rate from mRNA sequence. By incorporating the thermodynamics of ligand-binding, RNA folding, structural switching, and macromolecular crowding interactions, we developed a sequence-structure-function relationship for translation-regulating riboswitches formulated entirely using physical chemistry calculations. The model is combined with an optimization algorithm to enable the automated design of synthetic riboswitches using any selected aptamer domain.

Overall, we applied automated design to generate over 60 synthetic riboswitches using several different aptamers, which were characterized inside cells and within in vitro conditions. Using this data, we validated model predictions under equilibrium, ligand-limited, and non-equilibrium conditions; and in dilute or crowded environments. The sequence-structure-function model explained how changing aptamer structure and affinity, co-transcriptional ligand binding, ligand and mRNA concentrations, and the surrounding mRNA sequences controlled riboswitch activation. Our algorithm reliably generated riboswitches with high activation ratios, up to 383-fold, enabling non-experts to develop chemical sensors for their own applications. Finally, using a so-called "perfect riboswitch", we quantify the outer limits of biological sensing and highlight the potential applications of riboswitch-based sensing.