High Throughput Determination of RNA 3D Structures and Dynamics in Vivo

In addition to coding proteins, RNA plays fundamental roles in virtually every aspect of biology. The extreme functional diversity of RNA stems from its ability to fold into complex structures and, like machines, dynamically take input, transmit signal and force, and execute genetic instructions. RNA structures regulate every step of gene expression in cells and control the life cycle of RNA viruses. As a result, physiological and abnormal activities underlie a variety of human diseases. In recent years, targeting RNA has transitioned from an interesting academic idea to a reality in the clinic, with the development of oligonucleotides and small molecules that bind specific RNA sequences and structures, ushering in a new era in RNA medicine. Despite decades of technology development, RNA structure analysis remains a major challenge, especially compared to proteins. Traditional physical methods such as crystallography, NMR and cryo-EM has only been applied to purified “well-behaving” samples in vitro, leaving the vast majority of cellular and viral RNAs beyond reach. Recent chemical probing methods provided experimental constraints that improved de novo modeling but has so far been limited to small and simple RNAs. This RNA structure analysis bottleneck has significantly limited functional studies and therapeutic development. In this MIRA application, I outline a research program to tackle the ultimate challenge in RNA structure biology: in vivo determination of structures and dynamics for any RNA in any biological sample at high resolution. This proposal is based on the simple mathematical theory that the 3D structure of any object is equivalent to the spatial distances among its components. Therefore, RNA 3D structure determination can be transformed into a problem of measuring spatial distances among the nucleotides. To achieve this goal, we will develop ic3D (in vivo crosslinking of 3D structures, or “I see 3D”), a technology that uses 3 new classes of “molecular rulers” - reversible chemical crosslinkers with defined lengths - to precisely measure inter-nucleotide distances at the atomic level. Coupled with proximity ligation, high throughput sequencing and Rosetta-based 3D modeling, ic3D enables in vivo global analysis of RNA structures and ensembles of conformations. We will perform rigorous benchmarking against a wide selection of simple and complex models that represent the full diversity of possible RNA structures in vivo. We will use ic3D to discover and model 3D structures across the transcriptome. The completion of this project will have broad impact in understanding the structural basis of RNA functions, mechanisms of RNA-mediated diseases, and revealing new structure targets for therapeutic interventions.