Cooperative control of Polycomb Repressive Complexes by long noncoding RNAs, CpG island DNA, and RNA-binding proteins

点击翻译按钮获取中文摘要

ABSTRACT The goal of this grant is to determine the mechanisms through which long noncoding RNAs (lncRNAs) induce the spread of Polycomb Repressive Complexes (PRCs) over specific regions of the genome. Each of the two PRCs, PRC1 and PRC2, deposits histone modifications that repress transcription and simultaneously stimulate the activity of the other complex. PRCs are essential for the development of most major organ systems and their dysregulation causes a wide range of diseases. PRCs are unique amongst known epigenetic modifiers in that their spread over chromatin is induced by specific lncRNAs. However, despite decades of study, the field has not established how lncRNAs induce the spread of PRCs. Prominent models suggest that lncRNAs rely on 3- dimensional (3D) genome structure to spread PRCs indiscriminately across chromatin, without being influenced by the underlying sequence of DNA. Yet, these models cannot account for the extensive variability of PRC- dependent modifications in lncRNA target domains, and do not consider that outside of lncRNA target domains, DNA elements called CpG islands (CGIs) recruit PRCs to chromatin. Moreover, it remains unclear how many lncRNAs induce the spread of PRCs. While investigators originally proposed that thousands of lncRNAs have this function, the number of validated PRC-inducing lncRNAs remains close to ten. Our findings have begun to resolve these unknowns. For example, we discovered that in mouse trophoblast stem cells, the lncRNA Airn causes PRCs to spread outwards from CGIs that bind PRCs autonomously, prior to expression of Airn. Deletion of just one of these CGIs causes a multi-megabase loss of PRC-deposited modifications in the Airn domain. Moreover, we and others found specific RNA-binding proteins that associate with the PRCs, and at least two of these proteins appear to be required for lncRNAs to induce PRC spread. Thus, rather than spreading PRCs indiscriminately across chromatin, we hypothesize that lncRNAs associate with CGIs to spread PRCs from specific points of contact, and, that this association is mediated by RNA-binding proteins that bind lncRNAs and PRCs. Further, we hypothesize that any chromatin-bound RNA can induce the spread of PRCs, so long as it binds the necessary proteins. We now propose rigorous tests of these hypotheses: Under Aim 1, we will determine the role of CGIs and CGI-bound factors in the spread of PRCs induced by lncRNAs. Under Aim 2, we will determine the role of RNA-binding proteins in the spread of PRCs induced by lncRNAs. Under Aim 3, we will determine the roles of non-canonical RNAs in inducing the spread of PRCs. If successful, our work will demonstrate new roles for DNA regulatory elements, RNA-binding proteins, and RNAs in PRC function. Paradigms we establish should apply to broad areas in nuclear cell biology, and suggest multiple new avenues for controlling gene expression therapeutically.