Mechanistic characterization of post-transcriptional gene silencing

Abstract: With the ongoing task of identifying and characterizing genes in the human genome, there is a great demand for tools to study these genes. Today, reverse genetics is the most effective way to explore gene function. This involves suppression of the gene of interest by a set of manipulation techniques. In addition to this, many conditions like viral infections, cancers and cardio-vascular diseases are caused by relative over-expression of certain genes in various cell types. Therefore, there is a great need for development of methods to suppress these unwanted gene expressions. The work presented in this thesis concerns mechanistic studies of two alternative approaches to perform post-transcriptional gene silencing in mammalian cells: DNA-mediated antisense and siRNA-mediated RNA interference (RNAi). Today, both of these applications are used for gene regulation in either cell culture experiments or in vivo studies. Although some parts of the pathways of these two methods have been established, a great deal is still to be revealed about target specificity, knockdown efficiency and avoiding the anti-viral pathways generally occurring in mammalian systems. In paper I, we determined the role of the rat GERp95 homologue to human Ago2. By targeting this protein with siRNA-mediated RNAi, the knockdown efficiency of the endogenous target gene NPY was substantially impaired. This determined the necessity of GERp95 for fully functioning RNAi. In paper II, a stable 3 cleavage product was discovered when targeting the mouse Fas gene by DNA-mediated antisense. This finding puts further light into the complexity of the RNase H-dependent antisense pathway and strengthens the need for careful selection of detection method when analyzing the outcome of post-transcriptional gene silencing in mammalian cells. By a comparative analysis of siRNAs with varying chemical characteristics, we examined the mammalian anti-viral response system in paper III. From our results, we could conclude that LNA-modified siRNAs do not trigger IFN-dependent immune response in human HeLa cells. This in combination with the superior serum-stability and other potential advantages of LNA-modified siRNAs makes them a good candidate for future therapeutic applications. In paper IV we address the issue of siRNA specificity by performing a large-scale study of siRNA target variations. Finally we present some preliminary results regarding a putative novel method for visualization of miRNA action. Collectively, these studies have aimed to increase the general knowledge of post-transcriptional gene silencing in mammalian cells and they have more specifically investigated regulatory pathways, mRNA targeting and specificity for DNA-mediated antisense and RNAi.