Interfacing a Transcriptional Biosensing with Toehold-Mediated Strand Displacement for Programmable Molecular Diagnostics

There is a great need for programmable diagnostic technologies that can be used to rapidly monitor environmental changes. Despite recent advances in synthetic biology approaches to tackle this challenge, many biosensing platforms still suffer from slow response times, and their imprecise readouts make point-of-use applications challenging. While cell-free biosensing platforms have improved upon whole-cell biosensors by eliminating dependencies on cell growth, they typically rely on protein outputs, causing detection to be rate-limited by translation. Our group has recently developed a novel biosensing platform called ROSALIND that uses regulated in vitro transcription of fluorescence-activating RNA aptamers to visibly report a sensing event within minutes. Here, we report an improvement to further speed up the response time by tracking RNA output in situ using toehold-mediated strand displacement. Toehold-mediated strand displacement is a powerful molecular programming platform that performs basic molecular computations on the timescale of tens of microseconds. We show that when used to monitor the outputs of in vitro transcription reactions, it allows the visible detection of readouts that are up to thirty minutes faster than that of fluorescent aptamers. Furthermore, we show that the use of tunable thresholding computations enables us to program the platform to generate a semi-quantitative readout of target molecule concentration. We demonstrate applications of these capabilities to robustly detect a wide range of molecules, including antibiotics and heavy metals, in water samples. We believe that this work establishes a pathway to advance synthetic biology diagnostic platforms through the implementation of ‘smart’ molecular circuitry that can rapidly process and respond to analyte signals.