Untersuchungen zu den Interaktionspartnern modifizierter Nukleosidein messenger RNA

The central dogma of molecular biology describes the transfer of genetic information through the three essential layers DNA, messenger RNA and protein. Regulation of gene expression takes place during each individual step and is considered epigenetic when it includes inheritable modulation of genetic information by reversible chemical modification of the information carriers. The existence of this form of regulation is established on the levels of DNA and protein. In stark contrast, epigenetic regulation on the level of messenger RNA is an underdeveloped field of research that lately experienced a strong increase in scientific interest and number of publications. Recent work addressing the biological function and sequence context of N6-methyladenosine, N1-methyladenosine and pseudouridine hints towards actively modulated mRNA modification patterns that represent a formerly unknown network of epigenetic regulation. In addition to a detailed analysis of the sequence context of such nucleoside derivatives, the knowledge of their protein interaction partners is an essential step towards a deeper understanding of their biological function. In this work, interactome studies for a number of RNA modifications were performed, to better understand the described regulatory networks. For this purpose N6-methyladenosine (m6A), N6,N6-dimethyladenosine (m62A), pseudouridine (Ψ), N6-isopentenyl-2-methyl- thioadenosine (ms2i6A), N6-isopentenyladenosine and the four 2’-O-methylated nucleosides Am, Gm, Cm and Um (Figure 1) were incorporated into oligoribonucleotides, using solid phase synthesis. In the case of i6A, ms2i6A, m6A and m62A, this required synthesis of the phosporamdites 15, 22, 29 and 31. In a cooperative effort with the group of Michiel Vermeulen, the synthesized strands were used for interactome studies utilizing stable isotope labeling with amino acids in cell culture (SILAC). In these studies, groups of proteins that specifically interact with the modifications of interest (reader proteins), but also networks of repelled proteins that show preferential binding to the unmodified control strands, could be identified. For m6A, undoubtedly the “epicenter” of mRNA epigenetics, a number of formerly unknown readers were observed. Among them are FMR1, FXR1 and FXR2, proteins that are functionally related to the fragile-X-syndrome, a major cause of inheritable mental retardation. The characteristic phenotype results from a loss in RNA-binding activity of FMR1 and the newly discovered connection between the RNA modification m6A and the binding behavior of the central proteins illuminates a highly interesting novel mechanistic detail of these critical RNA-protein interactions. In addition to these readers, the negative interactome of repelled proteins was characterized for the first time. It shows a comparably high degree of variability, depending on the sequence context of the modification and is a rich source for future studies. For the G3BP1 and G3BP2 pair of repelled proteins, a selective interaction with unmodified sequences was validated. Furthermore, their stabilizing effect on bound transcripts was demonstrated which shows that they functionally antagonize m6A-binders of the YTH-family that destabilize mRNA. Figure 2 is a short graphical summary of the findings regarding this modification. The group of ribose-methylated nucleosides (Am, Cm, Um and Gm) was found to interact with a highly complex set of proteins that appears to be specific for the individual nucleobases. An involvement of the Nm group in an interferon response to viral infections has been published previously and the results of the investigations presented herein suggest that Um is the modification that mediates this involvement. Aside from this, the repellence of nucleolytically active complexes and interactions with splicing regulators appear to be features of the group. The fact that Gm strongly repels the YTHDF family of proteins implies an antagonism between m6A and Gm that warrants further investigation, since this would enable a previously unknown mechanism of fine-tuning the m6A interactome. The results are summarized in figure 3. The mRNA hypermodification ms2i6A was another key focus of this thesis, being an unusually complex mRNA modification. To further investigate the biological function of the modification, interactome studies were conducted and a knockout cell line of the biosynthetic enzyme, CDK5RAP1, was characterized. The analysis, combining deep transcriptomics with an analysis of the whole cellular proteome (figure 4), shows that the absence of ms2i6A leads to a general defect in mitochondrial protein biosynthesis, which is in accordance with published data. In addition to this, a pronounced lack of ribosomal proteins was apparent that hints towards a formerly unknown connection between the modification and their biosynthesis. Preliminary studies for the remaining three modifications i6A, m62A and Ψ presented herein represent a good foundation for subsequent investigations. In conclusion, the data presented in this dissertation analyses the complex interactome of a number of mRNA modifications in an unprecedented depth and quality, among them all four 2’-O-methylated nucleosides. Especially for the adenosine derivative m6A a highly detailed network of readers and repelled proteins in relation to its sequence context could be obtained, which will serve as a valuable resource for future studies. Investigations on the biological role of ms2i6A revealed its interactome and include an in-depth characterization of a knockout cell line lacking this hypermodified mRNA modification.