Cells regulate gene expression in a particular spatio-temporal manner by executing specific transcriptional programs. The different transcriptional programs allow the generation of the different cell types in multicellular organisms and mediate particular cellular responses to external cues such as viral infections. The regulation of how, when and which information is accessed in the DNA is vital for the normal biological function of cells and organisms.

The classic approach for studying gene expression has focused on the proteins involved in chromatin regulation, transcription and post-transcriptional gene expression events. During the years, it has become clear that not only proteins but also RNAs play important roles in the regulation of specific gene expression programs. However, how the RNA component of the chromatin modulates gene expression is to this date not fully understood. The main objective of this thesis has been to expand the knowledge about different mechanisms by which RNA impacts gene expression. In particular, at the transcriptional level, we have studied how RNA influences transcription and chromatin compaction by characterizing the chromatin-associated RNAs in Drosophila melanogaster. At the post-transcriptional level, the thesis aims at understanding the functions of ADAR3, a member of the Adenosine Deaminase Acting on RNA (ADAR) protein family that is catalytically inactive. 

In paper I, we have investigated the role of the RNA exosome in the degradation of RNA on chromatin, and we have tried to understand the importance of this phenomenon for the regulation of chromatin accessibility. We hypothesized that deregulation of RNA abundance on chromatin can directly impact chromatin compaction and transcriptional outcome. To address this question, we profiled the chromatin-associated transcriptome of Drosophila melanogaster S2 cells in control cells and in cells depleted of RRP6 and DIS3, the two RNA exosome catalytic subunits. Moreover, to measure chromatin compaction, we performed ATAC-seq and asked whether the deregulation of the chromatin-associated transcriptome had any impact on chromatin compaction. We have identified a subset of genes involved in development and morphogenesis the regulation of which is dependent on the RNA exosome. These genes show alterations of both RNA levels and chromatin compaction in exosome depleted cells, and are characterized by the presence of Polycomb and Trithorax factors, suggesting that RNA degradation by the RNA exosome is necessary to maintain the homeostasis of balanced chromatin states. The RNA exosome has also been linked to the homeostatic control of RNAs derived from repetitive regions of the genome. Thus, we have analyzed the repetitive transcriptome and show that the RNA exosome is required to maintain the structural integrity of the centromeric regions, where we have observed that the RNA exosome is required for the repression of specific transposon families, mainly LTR retrotransposons. 

Another finding of paper I is that snoRNAs are enriched on chromatin. In paper II, we have characterized a novel function for a specific snoRNA gene, snoRNA:U3:9B. We hypothesized that the snoRNA:U3:9B can impact gene transcriptional regulation by modulating the chromatin compaction of target genes, and that this regulation is biologically relevant in specific scenarios. We have used an inducible Sindbis viral replicon as a model of viral infection, and we show that the snoRNA:U3:9B is induced upon induction of the Sindbis replicon. We have observed attenuated transcriptional response of immune response genes in flies depleted of snoRNA:U3:9B, accompanied by decreased chromatin accessibility in these genes, which suggests that the snoRNA:U3:9B modulates the chromatin accessibility of genes implicated in D. melanogaster immune response.

In paper III, we have characterized the function of ADAR3 with focus on brain development. We observed that ADAR3 is broadly expressed in cortical neurons during early development but only expressed in a small subpopulation of in vitro differentiated cortical neurons, which suggests a specialized function in the mature brain. We generated a transgenic cell line by ectopically introducing ADAR3 in a neuroblastoma cell line (N2a) that did not express ADAR3 endogenously. To characterize ADAR3 function, we performed co-immunoprecipitation experiments followed by mass spectrometry to identify the ADAR3 interactome, which suggested links to mRNA stability and translation. Studies of mRNA stability based on high-throughput RNA-sequencing revealed that ADAR3 affects the stability of a large number of mRNAs. However, the corresponding protein levels were in general not affected. Moreover, we observed that ADAR3 inhibits translation and is associated with polysomes. Taken together, our results suggest that ADAR3 binds and stabilizes specific mRNAs in non-productive polysome complexes. Some of the affected transcripts are related to neurogenesis and we propose a model in which ADAR3 inhibits neuronal differentiation during early brain development.