Decoding the Epitranscriptome: Detection of and Functional Insights into RNA Modifications

RNA modifications are emerging as key regulators of gene expression, but their functional roles remain poorly understood due to the lack of accurate, transcriptome-wide detection methods. This dissertation presents a suite of high-accuracy and single-resolution next-generation sequencing (NGS) techniques to map four modifications: pseudouridine (Ψ), N1-methyladenosine (m1A), 5-formylcytosine (f5C), and N6-methyladenosine (m6A), and explores their biological significance. To map Ψ, a bisulfite-induced deletion sequencing (BID-seq) method was developed, enabling base-resolution, quantitative detection of Ψ sites from low-input RNA. BID-seq allows precise assignment of Ψ writers, uncovers tissue-specific pseudouridylation patterns, and reveals that Ψ can stabilize mRNA, in contrast to the destabilizing effect of m⁶A. For m1A detection in small RNAs, an antibody-free approach based on sodium borohydride reduction and commercial reverse transcriptases was established. This method improves mutation and read-through rates, allowing the identification of previously undetectable m1A sites in mitochondrial tRNA, and rRNA, including low-stoichiometry sites. To profile f5C, a chemical reduction and sequencing approach (f5C-seq) was optimized using pic-borane to induce C-to-T transitions during reverse transcription. This technique confirms known f5C sites and uncovers novel ones in chromatin-associated RNAs, including those sensitive to ALKBH1. Finally, a single-assay method (m6A-Ψ-seq) was developed to simultaneously detect m6A and Ψ in the same RNA sample. By integrating sequential allyl-labeling and bisulfite treatment with optimized reverse transcription chemistry this method enables the study of potential interplay between multiple modifications. Finally, we identified a candidate Ψ-binding protein and investigated its role in regulating Ψ-modified RNAs and RNA metabolism. This protein selectively binds to Ψ-modified transcripts, and its knockdown leads to preferential downregulation of its bound and pseudouridylated RNAs, particularly those involved in cell cycle regulation. Collectively, methods developed in this thesis provide powerful tools to decode the epitranscriptome with high resolution and accuracy, offering new insights into the regulatory functions of RNA modifications across biological systems.