Programmable Detection of Target RNA Sequences In Vivo

The mRNA profile of a cell provides critical information, including metabolic state, cellular stress, and foreign gene acquisition. While there are methods to determine what mRNAs are present, they require cell lysis, insertion of potentially disruptive sequences, or massive overexpression of the detector. Here, we propose to detect a target RNA in vivo using two orthogonal dCas13s that use gRNAs to bind to two adjacent locations, thus inducing the proximity-dependent release of a prokaryotic sigma factor via intein trans-splicing. To this end, we evaluated the ability for four dCas13s, from Ruminococcus flavefaciens, Prevotella sp. P5-125, Porphyromonas gulae, and Riemerella anatipestifer, to target a ribosome binding site (RBS) controlling red fluorescent protein (RFP) in Escherichia coli and confirmed RNA binding by measuring a reduction in fluorescence. One gRNA was found to be toxic, suggesting off-target binding to an endogenous mRNA. Separately, we have designed a system where a protein co-localization event causes two domains of a Pseudomonas fluorescens sigma factor to covalently fuse via an intein. As a proof-of-principle, a rapamycin-driven protein-protein interaction is used to induce expression of a fluorescent reporter from the sigma factor-specific promoter. Finally, we are in the process of using pairs of dCas13:gRNAs to drive the same proximity-dependent sigma factor release by binding to adjacent sites on a target RNA. Current iterations of this system have not been responsive to the presence of the target mRNA: we are currently trying to identify whether the problem may be caused by steric hindrances that disrupt RNA-binding or sigma factor release. If successful, this system would enable the real-time in vivo detection of specific mRNAs without external perturbation, for the purposes of both studying endogenous gene expression and engineering systems to respond to intracellular changes such as stress and pathogen infection.