Mechanisms of gene silencing induced by long noncoding RNAs

摘要

ABSTRACT Long noncoding RNAs (lncRNAs) play essential roles in human health by repressing transcription of genetically- linked genes. Altered expression of the genes that lncRNAs target for silencing drives devastating genetic disorders such as Angelman, Beckwith-Wiedemann, and Rett Syndromes, as well as cancers and immune syndromes. The most potent and most studied repressive lncRNA, Xist, silences transcription across one entire X chromosome to achieve X-inactivation in females. Still, the mechanisms by which Xist and other lncRNAs silence their target genes are insufficiently understood. Indeed, at the level of RNA sequence, it is unclear what distinguishes repressive lncRNAs from those that lack repressive activity, severely limiting our understanding of mechanism and our ability to identify repressive function in other lncRNAs. The long-term goals of this research are to determine how RNA sequence specifies repressive function in lncRNAs and determine the mechanisms by which the repression is carried out. Our studies over the past five years have brought us closer to these goals. Most notably, we developed a method of non-linear sequence comparison which demonstrated that despite lacking linear homology, repressive lncRNAs are enriched in similar sequence motifs, suggesting a general model for how they encode repression. We discovered that repressive lncRNAs exhibit enriched associations with specific proteins, supporting the general model. We developed new approaches to map, quantify, and manipulate lncRNA-protein interactions. Lastly, our data provide an explanation for how a single domain in the 5¢ end of Xist is required both for the transcription of Xist and for the earliest stages of Xist-induced gene silencing. On the basis of our findings, we hypothesize that lncRNAs with shared sequence properties repress transcription by related mechanisms. During the next funding period, we will test this hypothesis with three Specific Aims: (1) determine how a single domain orchestrates both Xist production and target-gene silencing; (2) determine how RNA-protein interactions and RNA structure specify repressive function in three evolutionarily unrelated lncRNAs; and (3) develop and rigorously validate a new approach to scan transcriptomes for lncRNAs and lncRNA domains that harbor analogous functions. These advances will accelerate the discovery of RNA- mediated events that are essential for development and disease.