The epitranscriptome represents an additional layer of gene regulation by modifying RNA molecules, influencing processes such as RNA stability, translation, and localization. The first part of this thesis focuses on the importance of messenger RNA (mRNA) modifications, especially those occurring at the 5′ end of the molecule. Eukaryotic mRNAs are capped at their 5' end with a modified guanine nucleotide usually referred to as the m7G canonical cap structure, which is essential for mRNA stability and translation. While many organisms incorporate a ribose methylation (Nm) on the first transcribed nucleotide, mammals and other vertebrates introduce an additional modification: N6-methyladenosine (m6Am) catalyzed by the cap-specific methyltransferase PCIF1. In this first project we showed that despite the prevalence of m6Am, Pcif1-depleted mice exhibited normal development and fertility, although they displayed reduced body weight compared to the wild-type controls. Gene expression analysis on several tissues of Pcif1 mutant mice supported a positive correlation between m6Am and transcript stability. We then performed a study across different organisms having a homologue of PCIF1. Interestingly, Drosophila Pcif1 lacks methyltransferase activity but maintains the ability to interact with RNA polymerase II, as does the mammalian counterpart. In contrast, in trypanosomes PCIF1 functions as a methyltransferase, contributing to a more complex cap modification N6,N6,2′-O-trimethyladenosine (m62Am) characteristic of these parasites, highlighting the evolutionary adaptability of PCIF1 and its diverse roles across species.
Beyond the canonical cap structure, RNAs can acquire non-canonical caps (NCCs) at their 5' end, often incorporated by RNA polymerase using nucleotide-like precursors. The second part of this thesis explores the role of one of these RNAs in vivo and the enzymes responsible for their turnover. The innate immune system employs a surveillance mechanism to distinguish between self and non-self nucleic acids, activating the interferon pathway in response to foreign genetic material. Cellular mRNAs, marked by a specific methylation on the first nucleotide, evade this immune detection. Our study identified a class of NNC-capped RNAs, Ap4A-capped RNAs, which accumulate in mice lacking the Nudt2 decapping enzyme. Moreover, we showed that NUDT2 is essential for their turnover, as it functions both as a hydrolase for Ap4A metabolite and as a decapping enzyme for Ap4A-capped RNAs. These RNAs lacks the immune-evading methylation on the first nucleotide and trigger the interferon response through RIG-I activation, mimicking pathogen infection. However, when we artificially installed this methylation, these Ap4A-RNAs were not immunogenic. Consequently, Nudt2-deficient mice phenocopy a rare human disease characterized by mutations on NUDT2; these phenotypes are similar to those observed in the interferonopathies, a condition characterized by excessive and uncontrolled interferon activation. Furthermore, we showed that suppressing the interferon pathway in Nudt2 mutant mice can rescue some of these phenotypes, suggesting a potential therapeutic approach for this disease.
