Skip to main content

The mutation-buffering capacity of RNA chaperones

Periodic Reporting for period 1 - MuRChap (The mutation-buffering capacity of RNA chaperones)

Reporting period: 2017-09-01 to 2019-08-31

Due to their intrinsic thermodynamic properties, RNA can misfold easily in cells. One way to mitigate RNA misfolding is through the actions of RNA chaperones, which bind and unwind structured RNA molecules and thereby offer opportunities for these misfolded species to refold properly. Such rescue activity has implications for the fitness effects of individual mutations - at least mutations that compromise RNA folding or structure might be buffered by RNA chaperones. However, little is known about the rules governing such mutation buffering. Here, we describe how a model RNA chaperone, the DEAD-box RNA helicase CYT-19, affects the fitness effects of mutations in a model structured RNA, the Tetrahymena group I intron, whose self-splicing activity is dependent on its structure. The goal was to comprehensively catalogue mutational effects on this self-splicing intron in the absence and presence of CYT-19. Conclusions: To date, this work has yielded considerable information about the mutational effects on the Tetrahymena group I intron. My results highlighted the overall complexities in delineating such mutational effects on a model RNA. More importantly, understanding the in vivo mutational effects on a model RNA will have considerable implications for understanding the robustness in RNA folding. Through this project, two-way transfer of knowledge and skills has been achieved.
I have performed deep mutational scanning on the P1ex helix of the 5’ splice site, which is critical for its self-splicing activity, and experimentally assayed differential splicing activity of all possible P1ex mutants (~65,000 variants in total) in the presence and absence of an interacting RNA chaperone CYT-19 at either 37C or 30C. Mutant introns were then recovered from various conditions and sequenced on an Illumina platform. To deconvolute the genotype-structure-phenotype relationships, I made correlations between the abundance of sequencing reads, mutational features on the P1ex helix, and the computationally predicted RNA structures. In particular, the abundance of sequencing reads and the predicted RNA structures were used as input for training a gradient boosted model in order to predict nucleotides that are important for the splicing activity of the intron. Two of the top three predicted nucleotides that could reduce the splicing activity were consistent with what was shown in the literature. This work is now being written up for a publication.
To my knowledge, this is the first quantitative assessment of mutational effects on the Tetrahymena group I self-splicing intron, in the absence and presence of an RNA chaperone. The expected outcome will (and has) improve our understanding of RNA robustness, and may reveal insights into making better RNA-based tools. My results suggest that, to understand in vivo RNA structure and function, it is important to consider how mutational fitness effects can be moderated by RNA chaperones, and how the overall in vivo fitness landscapes of structured RNA may be less rugged than in vitro fitness landscapes.
Experimental Overview_Soo