Regulation of adenosine deamination type RNA-editing by mRNA splicing

For the longest time it was believed that genetic information stored in DNA is transcribed into RNA, and translated into proteins without changes to the information. The view on this central dogma of molecular biology has been challenged by several observations, demonstrating that genetic information is changed at the RNA level and that RNA can have a regulatory role itself. The major factors leading to post transcriptional changes of genetic information are alternative splicing and RNA-editing. In metazoa, the most prominent type of RNA-editing is adenosine deamination, mediated by the ADAR class of enzymes. Both, RNA-editing by ADARs and (alternative) splicing occur in the nucleus, and are most likely cotranscriptionally coupled. Moreover, mRNA splicing may control editing: On the one hand, several editing sites depend on the presence of intronic sequences that aid in guiding ADARs to their substrate sites. Thus, the speed of splicing may control the availability of binding sites for ADARs. On the other hand, the speed of splicing can control nuclear retention time. Extended nuclear retention, in turn, may also increase editing by exposing mRNAs to the editing-competent nuclear environment. Interestingly, systematic studies on the coupling of these two processes are still in their infancy. Here we propose to study the impact of mRNA splicing on RNA-editing at a transcriptome- wide level but also on specific model substrates. This will be done by applying RNA-seq to RNAs isolated from cells treated with splicing inhibitors and by evaluating editing levels at known and new editing sites. The global approach will be complemented by studies on model substrates where splicing efficiencies have been artificially altered. Our study will show to which extent the two most important processes leading to post transcriptional changes in genetic information are coupled, both quantitatively and qualitatively.

In eukaryotes, gene products are diversified by various processes. At the transcript level, adenosine to inosine editing (A-to-I editing) is a major diversifying factor. Inosine is generated by the deamination of adenosine. A pre-requisite for editing is a double-stranded RNA. During translation, inosine is typically interpreted as guanosine. Therefore, editing frequently causes the incorporation of amino acids that differ from those originally encoded in the DNA. Notably, editing levels are highly variable. A transcript can be edited to 10% but also to 90%. The regulation of editing is of major biological relevance as de-regulation is frequently associated with disease including schizophrenia, epilepsy, or depression. In this project we could show that editing levels are largely determined by the efficiency of pre-mRNA splicing, the removal of non-coding regions (introns) from the pre-mRNA. The most significant results of our work are as follows: 1) The efficiency of mRNA-splicing determines the level of editing. Reduced splicing increases the level of editing. Especially when the double-stranded RNA is formed between exonic and intronic parts of the transcript. 2) Consequently, alternative splicing factors proteins that modulate the efficiency of splicing for particular exons or introns can influence the level of editing. 3) The level of editing is different between different human tissues. We found evidence that that the efficiency of splicing also determines the differences of editing levels for different tissues. 4) Moreover, we identified over 90,000 novel A-to-I editing sites in the mouse transcriptome. As the mouse is a well-established model system for biomedical research, this collection of editing sites is an important resource for future research. In sum, we could show that splicing is a major factor regulating editing levels. Thereby we have created the basis for future research in the field of A-to-I editing. Our research suggests that a significant number of editing regulatory proteins will also be splicing factors. This narrows the search for factors controlling A-to-I editing levels.