X-chromosome inactivation: a tribute to Mary Lyon

XIST RNA established the precedent for a “chromosomal RNA”, which spreads and stably associates with the nuclear chromosome to induce X-chromosome inactivation.  Although XIST evolved to control a singular chromosome, we find XIST has a remarkably comprehensive ability to inactivate an autosome (Chr21), indicating a mechanism for RNA regulation available genome-wide.  There has been growing interest in the possibility that many large non-coding RNAs function in chromatin, yet for very few specific examples has this been established.   Recent findings from our lab indicate that a diversity of highly repeat-rich RNAs collectively is very abundant across euchromatin.  These repeat-rich RNAs tightly localize in cis across the nuclear chromosome territory, and the RNA territory persists long after transcriptional arrest (by multiple methods). Understanding how RNAs are localized and maintained on the nuclear chromosome territory can illuminate the “black box” of higher-order chromosome architecture.  Our analysis of protein factors, particularly SAF-A, thought to control XIST RNA’s chromosomal association indicates that current concepts of how RNA is tethered to chromatin require major revision.  These and other findings will be presented in the context of a model whereby RNA is not an occasional modifier of chromatin state, but an integral component of nuclear chromosome structure, with different classes of RNA impacting the condensation state of distinct regions.  Sequencing and cytological data points to the prevalence of repetitive sequence RNA, which may have unusual properties to contribute to chromosome structure and function.

Sex-chromosome dosage compensation is essential in most metazoan yet the developmental timing and the underlying strategies are remarkably variable, even amongst placental mammals. X chromosome inactivation (XCI), the mammalian dosage compensation system, is triggered by the accumulation of the long non-coding RNA (lncRNA) XIST on the chromosome, which is responsible for the silencing and heterochromatinization of the inactive X. In the mouse, the up-regulation of Xist on one of the X-chromosomes systematically triggers the XCI process. Strikingly, in human pre-implantation embryos, XIST accumulates on every X chromosomes in both males and females without inducing XCI. This unusual configuration suggests the existence of human-specific mechanisms preventing XIST-mediated silencing.

XACT is a recently identified X-linked lncRNA that coats active X chromosomes in human pluripotent stem cells. XACT is not or weakly conserved across mammals and is absent from the mouse, suggesting that it could fulfill primate-specific function. The Rougeulle lab is now addressing the potential role of XACT during the establishment of the XCI process. Single-cell RNA-sequencing and imaging revealed co-activation and accumulation of XACT and XIST on active X chromosomes in early human pre-implantation embryos. In these contexts, the XIST RNA adopts an unusual, highly-dispersed organization, which may underlie its inability to trigger X chromosome inactivation. Functional studies further demonstrated that XACT influences XIST accumulation in cis. These findings suggest an antagonist action of XIST and XACT in controlling X chromosome activity specifically in human, which highlights the contribution of rapidly evolving lncRNAs to species-specific developmental mechanisms.

The discovery of extensive transcription of long noncoding RNAs (lncRNAs) provide an important new perspective on the centrality of RNA in gene regulation. This talk will discuss genome-scale strategies to discover and characterize lncRNAs, including RNA chemical modifications and RNA structures.  An emerging theme from multiple model systems is that LncRNAs form extensive networks of ribonucleoprotein (RNP) complexes with numerous chromatin regulators, and target these enzymatic activities to appropriate locations in the genome. Consistent with this notion, long noncoding RNAs can function as modular scaffolds to specify higher order organization in RNP complexes and in chromatin states. The importance of these modes of regulation is underscored by the newly recognized roles of long RNAs in human diseases.

I will address the state of the X chromosome in human embryonic stem cells (ESCs). It is well established that conventionally-cultured female human ESCs can carry active X-chromosomes (Xa) or an XIST-RNA-coated inactive X-chromosome (Xi^XIST+). Additionally, many ESC lines are known to have abnormal X-chromosome-inactivation (XCI) states where the Xi no longer expresses XIST RNA and has transcriptionally active regions (eroded Xi=Xe). The fate of each XCI state upon differentiation has remained unclear because individual lines often contain a mixture of XCI-states. We established homogeneous XiXa, XeXa, and XaXa ESC lines and found that they were unable to initiate XIST-expression and X-chromosome-wide silencing upon differentiation such that the ESC XCI state is maintained in differentiated cells. Thus, conventionally-cultured ESCs are not suitable to study the initiation of human XCI. Notably, conventional human ESCs are in a primed pluripotent state. To enable studies of initiation of XCI and overcome the epigenetic instability of the X of primed ESCs, we got interested in naive ESCs. Previously, it had been established that naïve pluripotent cells of the female human blastocyst have two active X-chromosomes that express the XIST RNA but whether this state can be captured in ESCs has remained unclear. We found that naïve ESCs can be established that resemble the unique X-chromosome state of the pre-implantation blastocyst, and thereby identified a cell culture system for studying X-chromosome dosage compensation mechanisms in early human development. Additionally, we demonstrate that this naïve state resets epigenetic abnormalities of the Xi of primed ESCs.

Xist RNA plays a pivotal role in silencing and heterochromatinization of the inactive X chromosomes in female mammals. It has been shown that the A-repeat, one of the conserved repeats present in Xist RNA among many mammalian species, is essential for the silencing function of the RNA in differentiating ES cells. The group had previously attempted to explore the role of the A-repeat in vivo by targeted deletion of the corresponding genomic sequence in the mouse. This unexpectedly abolished transcriptional upregulation of the mutated Xist allele, precluding the further analysis for the behavior of the Xist RNA lacking the A-repeat in vivo. Here, a new Xist allele is introduced lacking the A-repeat under the control of a constitutively active promoter in the mouse. When this allele was paternally transmitted, the mutated Xist RNA was successfully expressed and coated the paternal X chromosome, inducing apparent heterochromatinization albeit defective in silencing in the embryo. A detail analysis of transcriptional state of the mutated X is currently underway. Sado will discuss the behavior of the mutated Xist RNA lacking the A-repeat in vivo and its effects on the X chromosome silencing.

X chromosome inactivation requires the Xist gene, which is exclusively expressed in female cells and encodes a long RNA that associates with the inactive X chromosome (Xi). Xist accumulation  leads to changes in chromatin composition and causes repression of a majority of genes on the Xi. A long standing question relates to the molecular mechanism of the gene repression pathway of Xist. Recently, factors could be identified by different research groups using genetic and biochemical strategies. In order to identify genetically required factors for Xist mediated gene repression screening in haploid mouse embryonic stem cells carrying an engineered Xist expression system was performed. Half a dozen high confidence candidate factors could be identified including genes that have been implicated in promoter regulation, histone modification and chromatin assembly. Ongoing efforts focus on validating these candidates and characterizing their function with the view of defining the underlying pathways in chromatin regulation and transcriptional repression. Although it appears likely that our screen has not reached saturation in a technical meaning, it remains unclear if additional components can be identified through further genetics efforts such as CRISPR nuclease based approaches. Independently, other laboratories have discovered additional and non-overlapping factors using biochemical strategies. It is anticipated that the combined data will contribute to advance our understanding of the molecular mechanisms underlying chromosome wide silencing. The results of our genetic screen and the conclusions that can be drawn at the time will be discussed.

Xist initiates XCI by spreading across the future inactive X-chromosome, excluding RNA Polymerase II (PolII), recruiting the polycomb repressive complex and its associated repressive chromatin modifications, and repositioning active genes into a transcriptionally silenced nuclear compartment. While much is known about the events that occur during XCI, the mechanism by which Xist carries out these various roles remains unclear. We used RAP-MS to identify proteins that directly associate with Xist, and we further show that 3 of these proteins are required for Xist-mediated transcriptional silencing. One of these proteins, SHARP, which is known to interact with the SMRT co-repressor that activates HDAC3, is not only essential for silencing, but is also required for the exclusion of RNA Polymerase II (PolII) from the inactive X. We also show that both SMRT and HDAC3 are required for Xist-mediated silencing, PolII exclusion, and PRC2 recruitment. Another of these proteins, LBR, is required for Xist-mediated silencing but not for PolII exclusion or PRC2 recruitment. We further demonstrate that Xist, through its direct interaction with LBR, recruits the inactive X chromosome to the nuclear lamina and through this leads to changes in the 3-dimensional structure of DNA in the nucleus. This process enables Xist to spread across the entire X chromosome to achieve its essential role in embryonic development. Specifically, by reorganizing nuclear structure, Xist changes the accessibility of DNA and thereby enables the Xist RNA to spread across the entire X chromosome and achieve chromosome-wide silencing. Together, these results present an integrative picture of how Xist can scaffold multiple proteins to orchestrate the complex functions required for the establishment of the inactive X-chromosome.

The Polycomb repressive complexes PRC1 and PRC2 play a key role in developmental gene regulation, functioning primarily by catalysing the histone modifications, H2AK119ub1 and H3K27me3 respectively.  Canonical PRC1 complexes bind to PRC2 mediated H3K27me3, and this interaction has been suggested to account for PRC1 localisation to target loci. In mammals Polycomb target loci include the promoter regions of a large number of developmental regulator loci and also, in female cells, the inactive X chromosome. In the latter model it has been proposed that Xist RNA directly recruits PRC2, with the resultant deposition of H3K27me3 accounting for subsequent recruitment of PRC1.   However, in a previous study we found that variant PRC1 complexes (varPRC1), localise to targets independently of H3K27me3, including on the inactive X chromosome, suggesting Polycomb recruitment mechanisms are more complex than had been thought.  Accordingly, we and others recently demonstrated that PRC1 mediated H2AK119ub1 can recruit PRC2 to target loci, the reverse of the classical recruitment model.   Building on these findings we now show that a varPRC1 complex, PCGF3/5-PRC1, initiates both PRC1 and PRC2 recruitment by Xist RNA.  Specifically, PCGF3/5-PRC1 mediated H2AK119ub1 acts as a signal to recruit other varPRC1 complexes via interaction of the Ranbp2 zinc finger in the core subunit Rybp/YAF2, thus amplifying H2AK119ub1 deposition over the inactive X.  H2AK119ub1 additionally recruits PRC2, in large part via direct interaction with a ubiquitin binding domain that we have identified in the PRC2 cofactor Jarid2.    Our findings thus overturn the prevailing model for Polycomb recruitment by Xist RNA and provide a novel paradigm for chromatin modification by Xist RNA.