Optimization of a bacterial three-hybrid assay through in vivo titration of an RNA–DNA adapter protein

Noncoding RNAs regulate gene expression in every domain of life. In bacteria, small RNAs (sRNAs) regulate gene expression in response to stress and are often assisted by RNA–chaperone proteins, such as Hfq. We have recently developed a bacterial three-hybrid (B3H) assay that detects the strong binding interactions of certain E. coli sRNAs with proteins Hfq and ProQ. Despite the promise of this system, the signal-to-noise has made it challenging to detect weaker interactions. In this work, we use Hfq–sRNA interactions as a model system to optimize the B3H assay, so that weaker RNA–protein interactions can be more reliably detected. We find that the concentration of the RNA–DNA adapter is an important parameter in determining the signal in the system and have modified the plasmid expressing this component to tune its concentration to optimal levels. In addition, we have systematically perturbed the binding affinity of Hfq–RNA interactions to define, for the first time, the relationship between B3H signal and in vitro binding energetics. The new pAdapter construct presented here substantially expands the range of detectable interactions in the B3H assay, broadening its utility. This improved assay will increase the likelihood of identifying novel protein–RNA interactions with the B3H system and will facilitate exploration of the binding mechanisms of these interactions.