Melissa, Kyle, Tim and Paul’s paper “Using in-cell SHAPE-Seq and simulations to probe structure-function design principles of RNA transcriptional regulators” is out in RNA! This paper marks an important milestone for us – its our first manuscript on using information gleaned from in-cell SHAPE-Seq experiments to inform the design of new RNA regulators!
This project started off as a desire to more deeply understand the pT181 transcriptional attenuation system. You can think of this as an RNA-based transcriptional repressor – basically an antisense RNA molecule targets the 5′ UTR of a specific mRNA and governs the transcription of the mRNA. It’s great because it works only on the RNA level – an RNA input governs an RNA output meaning that you can make ‘RNA-only’ genetic circuits out of it. We have spent quite some time building off of this mechanism – making orthogonal antisense/target pairs, making logic gates with attenuators, quantitatively modeling their behavior, and making more advanced networks that control gene expression timing. Despite this, most of the engineering of this mechanism was pretty difficult to achieve and we admittedly did not completely understand how it works!
Hence our desire to dive into the mechanism of the attenuator to better understand (and ultimately engineer) its function. The attenuator basically acts like an RNA-responsive riboswitch – specific interactions between the antisense RNA and the mRNA are the key to triggering the regulatory event (see the figure). These interactions aren’t simply an RNA binding to another RNA – since the attenuator controls transcription the interaction needs to be fast, and nature has evolved it to work via a kissing-hairpin mechanism. The thing is almost anything you do to these hairpins will break the system. In a previous paper, Melissa figured out that you can change the attenuator as long as you include inner loops in the hairpins. The goal of this new paper was to figure out why.
Enter the power of in-cell SHAPE-Seq. In-cell SHAPE-Seq allowed us to characterize the structures of the attenuator RNA hairpins inside the cell. By performing this on a number of different versions of the mechanism we observed a specific reactivity signature of more functional versions that confirmed that the inner loops were indeed present in the cell, and that they were necessary for function. To figure out why, we teamed up with Paul Gaspar and Alan Chen at the RNA Institute at the University at Albany who are experts at MD simulations of nucleic acids. Besides generating awesome movies of our RNAs, Paul and Alan’s simulations taught us that the high reactivity signatures we were seeing from our functional attenuators were really signatures of structural flexibility – the functional attenuator hairpins all had high degrees of flexibility. This is likely so the RNA molecules can quickly find their binding interactions and undergo additional structural transitions needed for the regulatory event.
Finally we wanted to put our new knowledge to the test and see if we could use this to design new versions of attenuators. To do so, we turned to our favorite nucleic acid design suite – NUPACK – and had it design us RNA hairpins that had inner loops in them. After sifting through a few designs we in fact found highly functional attenuator versions that also showed high in-cell SHAPE-Seq reactivities in key regions!
Overall this project was an exciting success and the first example of our vision for a new future of RNA engineering.