iRNA Webinar 05.2023 with Dr. Matvei Khoroshkin, UCSF: A systematic search for RNA structural switches across the human transcriptome

On May 2, 2023, Dr. Matvei Khoroshkin from Dr. Hani Goodarzi’s group at UCSF presented their recent preprint:
The manuscript describes a new approach the authors called “SwitchSeeker”, which allows for the identification of RNA structural switches in eukaryotes. SwitchSeeker generates secondary structures from sequence and draws on their corresponding folding energy profile to identify potential switches in cases where a sequence has two possible conformations with similarly low energy. The model underlying SwitchSeeker was also informed by DMS-MaPseq data from an in vivo RNA structure screen on an initial set of 3,750 candidate switches.
Matvei explained how they also used a functional screen to assess the impact of RNA switches predicted by SwitchSeeker on gene expression, focusing on switches located in the 3′UTR. The screen resulted in the identification of 1,454 functional switches.
They then further characterized the RNA switch located in the 3′UTR of the RORC transcript. They first confirm the existence of two stable RNA structures via single-particle cryo-EM before showing that mutations that disrupt one or the other conformation have opposite effects on RORC transcript levels in the 293T cell line, a result that could be reproduced using antisense oligonucleotides designed to specifically disrupt each conformation.
The manuscript concludes with an investigation of the mechanism underlying RNA switch-dependent gene regulation. A genome-wide CRISPRi screen for factors that alter RORC levels hinted at the implication of nonsense-mediated decay. Further biochemical assays, including crosslinking immunoprecipitation showed that core nonsense-mediated decay factor UPF1 differentially binds to each of the two RORC RNA conformations.
Given that SwitchSeeker revealed a large number of functional RNA switches when applied under a single condition in one cell line, it is expected that future analyses will reveal numerous additional switches with biological functions. It will also be interesting to assess whether this new tool performs well when predicting RNA structures beyond the 3′UTR.

A recording of the presentation is available here: