A Chemical Toolbox To Study Modified RNA-protein Interactions

Abstract

Biological outcomes arise from tightly regulated gene expression programs. While a sequence of genetic instructions is essential to define these outcomes, it is not sufficient to convey information at the level of complexity needed to specify the intricacies of life. Epigenetic (DNA) and post-translational (protein) editing are paramount mechanisms that modulate the flow of genetic information without changing the encoded content by altering the way that information is stored, converted, or interpreted. While research in this area has flourished with regards to DNA and proteins, it has lagged significantly for RNA. It is only recently that we grasped the prevalence of post-transcriptional chemical RNA editing in the context of gene expression. The collection of these modifications, known as the “epitranscriptome,” prescribes specific RNA fates by affecting RNA structure and modulating how RNA interacts with biochemical machinery. Studying epitranscriptomics is thus crucial to appreciate gene expression at the RNA level. Here, we describe approaches to more efficiently study the epitranscriptome. We first focus on N6-methyladenosine (m6A), the most abundant and well-studied modification. We present technologies to systematically map novel m6A interactors and to characterize the sequence preferences of known m6A “readers” through chemical proteomics and next-generation sequencing, respectively. We expand the m6A interactome and find that m6A can recruit unknown putative “readers” as well as repel proteins. In addition, we establish differences in the sequence determinants of 3 known m6A “readers,” suggesting a possible mechanism that regulates certain “reading” events. The second half of this thesis shifts focus to 5-formylcytidine (f5C), a less abundant modification that has so far been identified in two tRNA species. We present approaches to map f5C-forming enzymes and to characterize the activity and substrate specificity of ALKBH1, the main f5C “writer” in mammalian cells, showing that f5C might be more widespread than originally thought. Finally, we develop and validate a method to quantitatively sequence f5C across various organisms and tissue types. Taken together, our research provides valuable tools to study both well-known and more cryptic RNA modifications and emphasizes how chemically oriented tactics can serve complex biological problems.