Functional RNAs as Drug Targets: Towards Manipulating the RNA Frame-Shifting Element in SARS-CoV2

Functional RNAs have been recognised over the last two decades as central regulators of biological function, and such RNAs have been shown to directly impact cellular activity. Moreover, functional RNAs play key roles in many diseases, such as cancer, and, finally, many viruses rely on RNAs as well. This centrality means that these molecules are promising therapeutic targets and present an exciting new avenue for drug development. The pivotal challenge is the structural behaviour of RNAs. In proteins, we generally observe a well-defined structure that can be used for drug development. In contrast, RNAs exhibit multiple structures and shift between them dynamically. Describing RNA structure therefore must consider the entire structural ensemble, which pushes state-of-the-art experimental and computational methods to their limit. Here, we propose a proof-of-principle study to describe the structural ensemble of the RNA frameshifting element (FSE) from SARS-CoV2, then identify binding pockets across the ensemble, and finally find ligands that impact the functional structural changes of FSE. The FSE has multiple conformations that have been identified as functionally important, allowing for functional interference using ligands binding to the RNA. The structural ensembles for the FSE wild type and some of its variants and mutants will be explored using the energy landscape framework, a computational approach that can resolve RNA structural ensembles in detail. One drawback of this methodology currently is the relatively high computational cost. To overcome this shortcoming, a new methodology will be developed based on machine learning to accelerate a crucial step within the framework. After binding sites and potential ligands are identified based on the RNA structural ensemble, computational and experimental validation will be sought for predicted RNA-ligand binding. On the computational side, this will include analysis of the changes to the structural ensemble upon ligand binding. Experimental verification of binding will be obtained by isothermal titration calorimetry (to confirm RNA-ligand binding) and NMR spectrsocopy (to confirm structural changes upon binding). The key objectives of the proposal are therefore as follows: Combine current state-of-the-art approaches with machine-learning to accelerate the study of RNA structural ensembles Map the structural ensemble of the RNA frameshifting element (FSE) of SARS-CoV2 and known mutants Identify small molecule ligands that arrest the frameshifting by: (a) Finding binding pockets across the structural ensemble from (2) (b) Identify candidate small molecules that bind the RNA (c) Verify the impact of the ligand binding on the FSE with simulations and experiments The proposal will establish a new pipeline from describing the structural ensemble of a functional RNA to identifying small molecule ligands for it, including the verification of RNA-ligand binding. This proof-of-principle would open up new avenues for drug development targeting RNAs, which are underutilised in therapeutic interventions. The methodology is transferable to any RNA, and will be openly available for the community to use. In the medium term, this progress will lead to a significant improved understanding of functional RNAs and their biological roles. In the long-term, this my lead to novel antiviral agents, with potentially wide societal impact and improved health outcomes.