Interrogating and Elucidating Drivers of Selective RNA-Ligand Interactions

Abstract

Recent efforts to better understand the human genome have uncovered several roles of non-coding RNAs in biology. We have also seen the number of RNA viruses causing epidemics and pandemics rising in recent years, with the most recent being the COVID-19 pandemic. With the growing number of RNA targets came increased efforts to target RNAs for therapeutic purposes. Numerous studies have now been published on RNA-targeting with small molecules, but in the context of all drug-targeting efforts, RNA-targeting is still developing. To date, there is only one FDA-approved drug whose mode of action is known to be caused by binding to non-ribosomal RNA. Despite the limited number of successful drugs, there are growing resources for RNA-small interactions. Many databases have been published detailing RNA-binding small molecules for better understanding the existing ligands that bind RNA. These databases, including the RNA-targeted BIoactive ligaNd Database (R-BIND) curated in the Hargrove lab, have led to a consensus of the bulk properties of small molecules that bind RNA. And while this is a critical first step, there is still much to learn about the types of small molecules that can bind to different RNA structures. RNAs contain a range of structures that fold into varied 3-dimensional shapes that can be targeted by small molecules. We hypothesized that small molecules that bind RNAs with similar structures, containing similar pockets, may have unique properties from molecules that bind to other structures. To date, the concept of small molecule differentiation of RNA structures is underexplored, and thus in this dissertation it will be investigated from a number of approaches. First, we have updated and expanded the scope of the R-BIND database. Specifically, with new ligands being added to the database, we have been able to implement a secondary structure search that identifies lead molecules based on the structures they bind. Further, through linear discriminant analysis we have seen that the small molecules in R-BIND that target unique structural classes have some distinct properties. Additionally, we have shown that small molecules from different libraries (e.g. commercial or synthetic) are enriched in different properties which may also influence the types of targets they are suited for. While understanding the existing literature supported the idea that there are unique properties, we were limited in terms of structural classes to those with several published small molecule binders. Therefore, to continue identifying novel trends in small molecule differentiation of RNA structures, we turned to screening methods to identify small molecules that bind a set RNA secondary and tertiary structures. Initial work was focused on developing a novel peptide-based fluorescent indicator with global RNA binding to use in high throughput screen. In this work nanomolar affinity peptides were identified, however fluorescent-labeling resulted in significant aggregation that could not be overcome for use in high-throughput screens. Ultimately, this study was performed with a small molecule dye as the indicator. A set of >15,000 molecules was screened for binding to nine RNA targets. This screening resulted in a set of 181 hits, 82 of which were unique to one of the nine targets. With these unique hits properties enriched in molecules selective for one RNA structure over another were identified. Of particular interest, we found that molecules that bound to stem loops, which may contain more traditional pockets, were more sphere-like than molecules that bound to g-quadruplexes where stacking is thought to be a binding mechanism. Finally, we investigated the specific recognition of small molecules with an RNA tertiary structure – triple helices. In this work we identified a novel interaction between a small molecule-metal-RNA complex, where small molecule chelation of iron or copper is necessary for binding to RNA. Together this work represents significant efforts to better understand how small molecules recognize unique RNA structures, which can inform future therapeutic design.