The SWI/SNF complex : Roles in transcription and pre-mRNA processing

Sammanfattning: The regulation of gene expression is fundamental to the development of complex organisms and an important driving force in this process. When and where the genes are expressed decide the fate of a cell and its physiological context in the organism. It is well established that the packaging of the DNA into a more compact but dynamic chromatin structure affects the basal regulation of gene expression. In this thesis, we will discuss how the chromatin-remodeling SWI/SNF complex influences the regulation of genes, and we will focus on the roles of SWI/SNF in transcription and pre-mRNA processing. In Paper I, we show through a genome-wide approach that the levels of the different SWI/SNF subunits affect the alternative processing of a subset of Drosophila melanogaster pre-mRNAs in S2 cells. It was previously not known whether the effects on pre-mRNA processing were attributed exclusively to the ATPase subunit Brahma or if other subunits of the SWI/SNF complex were also involved in the regulation of pre-mRNA processing. Analysis of microarray data and RT-qPCR showed that depletion of the SWI/SNF subunits Moira and SNR1 mimic to a large extent the effects of Brahma, which suggests a role for SWI/SNF in pre-mRNA processing. Moreover, RNAi experiments in larvae also provide evidence for an effect of SWI/SNF on pre-mRNA processing in vivo. In Paper II, we show that Brahma modulates the abundance of a specific trans-spliced transcript derived from the mod(mdg4) locus of D. melanogaster. We have characterized the relative expression of anti-sense mod(mdg4) transcripts in S2 cells, mapped transcription start sites and cleavage sites, identified and quantified cis-spliced and trans-spliced transcripts, and obtained insight into the regulation of the mod(mdg4) trans-splicing. Using RNA interference and over-expression of recombinant Brahma proteins, we show that the levels of Brahma affect the levels of the mod(mdg4)-RX trans-spliced mRNA isoform in S2 cells. Interestingly, the trans-splicing effect is independent of the ATPase activity of Brahma, which suggests that the mechanism by which Brahma modulates trans-splicing is independent of its chromatin‑remodeling activity. In Paper III, we show that the one of the SWI/SNF complexes, PBAP, specifically regulates the transcription of the CG44250 and CG44251 genes in S2 cells. Depletion of BRM reduced the levels of CG44250/51 transcripts, whereas BRM overexpression had the opposite effect. These changes in transcript levels were accompanied by changes in the density of Pol-II at the CG44250/51 locus. Intriguingly, the effect of BRM on the expression of the CG44250/51 genes was independent of the ATPase function of BRM, as shown by over-expression of a mutant form of BRM that lacks ATPase activity. Altogether, the results presented in this thesis confirm that SWI/SNF can regulate not only transcription but also pre-mRNA processing, and they reveal that some of the regulatory functions of SWI/SNF are independent of BRM’s nucleosome‑remodeling activity.