PLOS I genetics REVIEW Xist RNA in action: Past, present, and future Agnese Loda \ Edith Heard 1.2* 1 Directors' research, European Molecular Biology Laboratory (EMBL), Heidelberg, Germany, 2 College de France, Paris, France * edith.heard@embl.org Abstract Check for updates tl OPEN ACCESS Citation: Loda A, Heard E (2019) Xist RNA in action: Past, present, and future. PLoS Genet 15(9): e1008333. https://doi.org/10.1371/iournal. pgen.1008333 Editor: Marisa S Bartolomei, University of Pennsylvania, UNITED STATES Published: September 19,2019 Copyright: © 2019 Loda, Heard. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: E.H. is supported by a European Research Council Advanced Investigator award (AdG250367): European Union Seventh Framework Programme grants SYB0SS (GA242129) and M0DHEP (GA259743). A.L. is supported by a European Marie Sktodowska-Curie Actions (MSCA) Individual fellowship (IF-838408). The f unders had no role in the preparation of the article. Competing interests: The authors have declared that no competing interests exist. In mammals, dosage compensation of sex chromosomal genes between females (XX) and males (XY) is achieved through X-chromosome inactivation (XCI). The X-linked X-inactive-specific transcript (Xist) long noncoding RNA is indispensable for XCI and initiates the process early during development by spreading in cis across the X chromosome from which it is transcribed. During XCI, Xist RNA triggers gene silencing, recruits a plethora of chromatin modifying factors, and drives a major structural reorganization of the X chromosome. Here, we review our knowledge of the multitude of epigenetic events orchestrated by Xist RNA to allow female mammals to survive through embryonic development by establishing and maintaining proper dosage compensation. In particular, we focus on recent studies characterizing the interaction partners of Xist RNA, and we discuss how they have affected the field by addressing long-standing controversies or by giving rise to new research perspectives that are currently being explored. This review is dedicated to the memory of Denise Barlow, pioneer of genomic imprinting and functional long noncoding RNAs (IncRNAs), whose work has revolutionized the epigenetics field and continues to inspire generations of scientists. Introduction X-inactive-specific transcript (Xist) RNA was one of the first long noncoding RNAs (IncRNAs) to be discovered in the early 1990s [1-5], a decade before the Human Genome Project (HGP) revealed that the large majority of our genome accounts for noncoding sequences [6]. Two other famous IncRNAs were discovered around that time: H19, at the Igf2 imprinted cluster [7,8], and Aim, antisense to the imprinted Igf2r gene. Aim was uncovered by Denise Barlow as the first example of IncRNA playing a direct role in controlling the imprinted expression of neighboring genes [9,10]. As all three IncRNAs were implicated in epigenetic processes, their discovery raised expectations that epigenetic regulation might be a common feature of IncRNAs. Since then, thousands of intergenic, intronic, and antisense IncRNAs have been identified [11-14], and their widespread transcriptional activity has been increasingly recognized not only in humans but in many different organisms including mouse, zebrafish, and yeast [15-18]. Although their identification has become easier, defining the biological relevance of IncRNAs has remained challenging. Indeed, like Aim, it took nearly 30 years of research to characterize the function of Xist, and several important questions still need to be answered. PLOS Genetics | https://doi.orq/10.1371/iournal.pqen.1008333 September 19, 2019 1/17 '(S>-PLOS I GENETICS The X-inactivation centre: A paradigm for the study of IncRNAs Xist RNA is the master regulator of X-chromosome inactivation (XCI), the epigenetic process that equalizes the dosage of X-linked genes between female (XX) and male (XY) mammals. At the onset of XCI, Xist is up-regulated from one of the two X chromosomes, and its RNA spreads in cis along the entire X and triggers the inactivation of most of its > 1,000 genes. Ultimately, Xist spreading results in the conversion of one of the two active X chromosomes into a uniquely organized heterochromatic entity known as the "Barr body" [191. Although XCI is a chromosome-wide process, some X-linked genes escape silencing and remain expressed from both the active (Xa) and the inactive (Xi) chromosome. This concerns 12%-20% of human X-linked genes and 3%-7% in mouse and may play an important role in female development and disease susceptibility [20]. For example, only 1% of human female embryos carrying a single X (XO) survive to term, and all of them are affected by Turner syndrome [21,22]. On the other hand, XX women are more susceptible than XY men to autoimmune diseases, and women with supernumerary X chromosomes (XXX) are even more susceptible [23,24]. As the up-reg-ulation of both Xist alleles would lead to silencing of both X chromosomes and presumably consequent cell death, the spatio-temporal expression of Xist during development needs to be accurately regulated. Historically, the X-linked minimal genetic region that is necessary and sufficient to initiate XCI in female cells has been defined as the X-inactivation centre (Xic) (Fig 1). The Xic guarantees the monoallelic expression of Xist when it is present in two copies. Although the exact extent of the functional Xic is still to be fully determined, it encompasses several IncRNAs that act as regulators of Xist. For example, one major repressor of Xist in the mouse is Tsix, a IncRNA that completely overlaps with the Xist transcriptional unit [251. Tsix transcription through the promoter of Xist works as a break for Xist expression, and accordingly, deleting Tsix or terminating its transcription prematurely results in nonrandom inactivation of the mutated X chromosome [26-32]. Other loci that produce IncRNAs at the Xic are Xite, Tsx, Linx, Jpx, and Ftx, all of which have been proposed to work as either negative or positive regulators of Xist [33-40]. Thus, throughout the years, unraveling the mechanisms that orchestrate appropriate Xist regulation within the Xic has represented a powerful system to more generally understand how IncRNAs contribute to gene regulation, providing meaningful insights into the roles of antisense transcription and RNA-mediated silencing. This has been covered in several reviews [41-44]. Here we focus on the function of Xist RNA itself, from its ability to trigger the formation of facultative heterochromatin by recruiting a multitude of different factors to its impact on nuclear localization and 3D chromosome architecture. Xistis a multi-tasking RNA molecule essential for XCI Xist was initially proven to be indispensable for XCI by targeted mutagenesis and transgenesis in mouse embryonic stem cells (ESCs) and in mice. Deletion of the Xist gene results in skewed inactivation of the wild type X chromosome, indicating that this locus is essential for gene silencing [45,46]. Subsequently, the generation of inducible Xist transgenes integrated on autosomes demonstrated that Xist RNA alone is sufficient to trigger gene silencing [47-49], although the efficiency of autosomal silencing is lower when compared with the inactivation of X-linked genes [50]. The early transgenic studies also unveiled two key features of Xisfs function. First, the ability of Xist RNA to trigger gene silencing is strictly dependent on the developmental context. If Xist is ectopically induced outside of a critical temporal window during early ESCs differentiation, it will be no longer capable of initiating gene silencing [47]. Second, Xist has different tasks, such as cis-localization to the chromosome from which it is expressed and the ability to trigger gene silencing, and these tasks are mediated by genetically independent domains of the RNA [51]. The various functional domains of Xist have now been defined PLOS Genetics | https://doi.orq/10.1371/iournal.pqen.1008333 September 19, 2019 2/17 PLOS I genetics Mouse X Chromosome ----Xic -800 Kb (Nora et al. 2012) Xite Cdx4 Chid Tsx | Tsix Jpx -^ Linx V XistRNA ,'"A Before XCI During XCI • HDAC3 T H4 acetylation T H3K27acetylation r> X-linked gene Xist RNA Silencing PRC1 recruitment hnRNPK Localization RBM15 Cpcgf 5) irybp! CIZl\ hnRNPU (SAF-A) . (WTApV (metti^pbmisb) Y m6A RNA methylation Before XCI During XCI -PLOS I GENETICS by genetic and biochemical approaches and overlap at least partially with blocks of tandem repeats named A-F repeats, which are conserved between human and mouse and were therefore suspected to play a role in the function of Xist [2,5,51,521 (Fig 1). By deleting each of these repeat regions or a combination of them in mouse Xist-inducible transgenes, the highly conserved repeat A element, consisting of 7.5 copies of a 26 nt core sequence at the 5' end of Xist RNA, was characterized as the necessary element to trigger gene silencing [511. Intriguingly, deletion of the A-repeat region at the endogenous Xist gene seems to affect Xist RNA's capacity to be expressed from its endogenous promoter [53], suggesting that this region maybe important both for regulating transcription and/or processing of Xist and for mediating its silencing function. On the other hand, Xist RNA localization to the X chromosome appeared to be mediated by a combination of different domains, including repeats C, E, and F, via redundant mechanisms [51,54-61]. Finally, Repeats B and C have been found to be required for the recruitment of the Polycomb repressive complex 1 (PRC1) to the inactive X chromosome (Xi), and PRC1 activity has been proposed to subsequently recruit PRC2 (see the section "hnRNP K") [60,62-64]. Several studies aimed to characterize putative RNA-binding proteins that Xist might recruit to fulfil its multiple and genetically uncoupled functions. In this context, the transcription factor YY1 was reported to mediate Xist localization by bridging the repeat C region of the RNA with a "nucleation centre" at the repeat F of the DNA [59]. The heterogeneous nuclear ribonucleoprotein U (hnRNP U, also known as SAF-A) was also shown to interact with Xist RNA by binding to regions previously reported to mediate Xist localization, including repeat C and a region downstream of repeat E within exon 7 [56,61]. Finally, a 1.2-kb transcript derived from the Xist locus known as Rep A and overlapping with Xist repeat A, was previously reported to recruit PRC2 to the Xi by directly binding its catalytic subunit: EZH2 [65]. EZH2 activity is responsible for the deposition of H3K27me3, one of the silent chromatin marks acquired along the Xi during XCI [66—71]. Similarly, the chromatin remo-deller ATRX was reported to bind to Rep A RNA to reinforce PRC2 recruitment [72]. A major breakthrough in the characterization of the interactome of Xist was achieved when three independent proteomic approaches purified Xist RNA and its binding proteins in mouse ESCs and fully differentiated cells [73-75]. Amongst the factors that were previously reported to bind Xist RNA, only hnRNP U was confirmed by all proteomic studies. On the contrary, neither EZH2 nor any other PRC2 components were identified, but PRC1 proteins were. YY1 was also not pulled down by Xist RNA, whereas ATRX was only identified as a weak interactor of Xist in one of the three studies [75]. In parallel, the functional relevance of several of the newly identified interactors including SPEN, RBM15, and WTAP was confirmed by two genetic screens aiming to define the key factors accounting for the ability of Xist to trigger gene silencing [76,77]. Altogether, these unbiased approaches provided the field with an unprecedented opportunity to dissect the different functions of Xist and to identify its key protein partners. The research to understand their mechanisms of action in XCI is currently ongoing and will greatly enhance our general understanding of facultative heterochromatin formation, nuclear localization, and chromosome 3D spatial organization beyond the XCI field. Interactors of Xist contributing to chromosome-wide gene silencing SPEN The only binding factor of Xist RNA that was identified by all proteomic and genetic approaches is SPEN (also known as SHARP, SMRT/HDACl-associated repressor protein), a 400-kDa protein harbouring four N-terminal RNA recognition motifs (RRMs) and a highly conserved C-terminal Spen paralog and ortholog domain (SPOC). The functional role of PLOS Genetics | https://doi.orq/10.1371/iournal.pqen.1008333 September 19, 2019 4/17 '(S>-PLOS I GENETICS SPEN in XCI was initially confirmed by knockdown and knockout experiments in which loss of SPEN resulted in defective silencing of the few X-linked genes that were tested [73,74,76,771. More recently, the requirement of SPEN to achieve effective XCI was further proven by chromosome-wide analysis of X-linked expression upon differentiation of SPEN null ESCs [78]. In other contexts, SPEN's SPOC domain is able to recruit SMRT, a component of the transcriptional corepressor complex that activates histone deacetylases and leads to transcriptional repression [79-82]. As the functional depletion of SMRT and HDAC3 in male ESCs carrying an inducible Xist allele leads to impaired gene silencing, resembling what is observed after SPEN knockdown, a model was put forward according to which SPEN might initiate transcriptional silencing by actively recruiting HD AC3 via SMRT binding [74]. Indeed, in a recent study exploring the choreography of chromatin changes associated with XCI, histone deacetylation was found to be one of the earliest chromatin alterations induced by Xist RNA coating, and HDAC3 activity was specifically shown to promote silencing of most X-linked genes [83]. Nevertheless, HDAC3 does not appear to be robustly recruited to the Xi at the onset of XCI, nor does it seem to be a direct partner of Xist RNA, indicating that Xist RNA might initiate gene silencing through its interaction with SPEN by activating the prebound HDAC3 rather than by recruiting it de novo along the future Xi [83]. A key indication of SPEN's function in XCI was the demonstration that it binds to Xist RNA via the A repeat [73], originally identified as the functional domain required for gene silencing [51]. Indeed, depletion of SPEN only affects gene silencing without compromising the ability of Xist to localize to the Xi in cis [77]. However, global hypoacetylation of histone H4 along the Xi is not affected in SPEN knockout ESCs [77], and loss of HDAC3 does not fully prevent XCI to occur but rather delays the process [83], suggesting that other redundant mechanisms are likely to have evolved to guarantee proper XCI, at least during mouse development. Furthermore, as SPEN depletion results in reduced accumulation of the Polycomb-dependent chromatin marks H3K27me3 and H2AK119ub along the Xi, SPEN was suggested to also play a role in Polycomb recruitment [74,77]. However, Xist RNA lacking the A repeat is still able to recruit both PRC1 and PRC2 and to create a repressive compartment from which most hallmarks of active transcription, including RNA Pol II and general transcription factors, are excluded [83-86]. Accordingly, three subunits of the PRC1 complex, RING1B, RYBP, and PCGF5, were pulled down by Xist RNA lacking the A repeat [73], suggesting that PRC1 is recruited to the Xi independently of the silencing domain of Xist. Interestingly, none of the core PRC2 components were pulled down by the full length Xist RNA [73-75]. The relative timing of H3K27me3 and H2AK119Ub deposition and their dependency on the A repeat region of Xist have been precisely addressed by following the X-chromosome-wide enrichment of these histone marks during XCI [83]. This study demonstrated that PRC1-dependent H2AK119Ub accumulation precedes H3K27me3 deposition and also showed that Xist RNA lacking the A repeat is capable of recruiting PRC1 and PRC2 activity to regions of the Xi without active transcription, whereas spreading of both H3K27me3 and H2AK119Ub into domains of active genes is impaired [83]. Thus, active transcription seems to prevent the efficient spreading of Polycomb along the Xi, suggesting that the reduced accumulation of Polycomb-dependent histone marks observed in the absence of SPEN is likely related to impaired gene silencing. Although SPEN has been proven to act as a key player of Xist-mediated silencing, the exact molecular mechanism by which it contributes to the establishment and/or maintenance of gene silencing remains elusive. For example, whether SPEN's role in XCI is restricted to the SPOC domain or whether other domains may recruit additional factors to the Xi is not yet clear nor is the impact of SPEN on the function of Xist in vivo. Indeed, SPEN knockout mice do not survive through embryonic development but reach E12.5, a time point that is never reached by female mice carrying Xist deletions, which fail to undergo PLOS Genetics | https://doi.orq/10.1371/iournal.pqen.1008333 September 19, 2019 5/17 '(S>-PLOS I GENETICS XCI [46,871. However, a knockout model to deplete the maternal pool of SPEN that would presumably enable early initiation of XCI in preimplantation embryos has never been performed. RBM15 and WTAP Another Xist RNA interactor that was consistently identified is RNA-binding motif protein 15 (RBM15), a component of the N6-adenosine (m6A) RNA methylation machinery [73-75]. Interestingly, RBM15 interacts with Wilms tumor 1-associated protein (WTAP) [881, another subunit of the m6A methyltransferase complex, which was found to interact with Xist in an A-repeat-dependent manner [73]. m6A is the most abundant mRNA modification, and its impact on post-transcriptional gene regulation is being increasingly recognized [89-911. Once established, this modification is recognized by a set of "m6A-readers", such as the YT521-B homology (YTH) domain-containing proteins, which specifically binds to m6A and trigger downstream alterations affecting several aspects of mRNA metabolism, including alternative pre-mRNA splicing, 3'-end processing, nuclear export, translation, and decay [89]. The identification of RMB15 and WTAP as Xist RNA interactors pointed to a potential role of the m6A RNA methylation pathways in mediating the function of Xist. Indeed, human XIST RNA is methylated at 78 m6A residues, some of which encompass the repeat A element, and these modifications appeared to be required for proper gene silencing, although only a few loci along the Xi were tested [92]. Furthermore, XIST RNA methylation was reported to rely not only on the activity of RBM15 but also on its paralog RBM15B [92], also identified as a. Xist interactor [75]. Both proteins are able to recruit the catalytic subunit of the (m6A) RNA methylation machinery METTL3 to Xist RNA in a WTAP-dependent manner, and their functional redundancy is likely to explain previously contradictory results obtained upon RBM15 depletion [74,78,92]. Indeed, RBM15 knockdown alone did not affect X/st-mediated silencing of the Gpc4 gene as assessed by single molecule RNA FISH [74], and chromosome-wide analysis of X-linked gene activity revealed only a minor reduction in silencing efficiency in RBM15 knockout cells [78]. Importantly, double knockout of RBM15 and RBM15B seems to be essential for ESC viability which has meant that the exact role of these factors in XCI cannot be fully assessed [78]. Nevertheless, this study showed that deleting specific m6A sites in the 5' region of Xist RNA results in a limited silencing defect, similarly to RBM15 and WTAP knockouts, indicating that the m6A pathway might not play a central role in Xist-mediated silencing, and highlighting the need to further assess the exact impact of m6A methylation on XCI [78]. For example, one of the open questions concerns the recognition of m6A residues along Xist by the YTHDC1 reader. Depletion of YTHDC1 was shown to result in defective XCI, whereas tethering YTHDC1 to Xist rescues the pheno-type in the absence of a functional m6A methylation complex [92]. How YTHDC1 would execute this function and contribute to gene silencing remains to be defined. Another unexplored possibility is that m6A methylation recruited via Xist RNA to the X chromosome might affect the stability of X-linked mRNAs. In fact, although the exact impact of m6A on gene regulation is not yet fully understood, specific functions have been reported for at least some readers, resulting in opposite effects on gene expression. For example, binding of YTHDF1 improves the efficiency of translation of m6A-methylated mRNAs [93], whereas the YTHDF2 reader destabilizes its target mRNAs and promotes their degradation [941. This different impact on mRNA stability might work as an effective strategy to potentially regulate XCI at a post-transcriptional level, leading to destabilization of mRNAs of the majority of X-linked gene, as well as increased stability of mRNAs transcribed from escaping genes that remain active along the otherwise silent Xi. PLOS Genetics | https://doi.orq/10.1371/iournal.pqen.1008333 September 19, 2019 6/17 hnRNP K The heterogeneous nuclear ribonucleoprotein K (hnRNP K) was identified as an interactor of Xist by two proteomic approaches, although not in an A-repeat dependent manner [73-751. However, none of the genetic screens confirmed its potential role as a key factor for Xist-medi-ated silencing [76,77]. Chu and colleagues initially showed that hnRNP K is recruited by Xist RNA independently of the A-repeat domain, and more recent studies have shown this interaction to be dependent on Xist RNA repeats B and C [62,631. Functionally, Chu and colleagues validated the impact of hnRNP K on Xist-mediated silencing by knockdown experiments and were able to show that its depletion significantly reduces the accumulation of H3K27me3 and H2AK119ub along Xi [73]. Subsequently, hnRNP K was shown to recruit PRC1 to the Xi by binding the noncanonical PCGF3/5-PRC1 complex [63]. In particular, a 600-bp element encompassing Xist repeat B and a small part of repeat C (i.e., the Xist RNA Polycomb Interaction Domain, XR-PID) was identified as the required element to mediate hnRNP K/Xist interaction, and deletions of XR-PID abolished Polycomb recruitment and resulted in reduced gene silencing [63]. However, as this strong silencing defect was observed upon deletion of the XR-PID element but in the presence of a functional repeat A, this observation is difficult to interpret. One possible explanation is the nature of the Xist transgene itself, as this study employed a short form of Xist RNA encompassing repeats A, F, B, and part of C, but lacking the elements downstream the first 3.9 kb of Xist exon 1. Indeed, the lack of these sequences might somehow enhance the role of Polycomb recruitment in silencing establishment. More recently, three independent studies further explored the contribution of Polycomb recruitment to the establishment and maintenance of XCI by generating a series of Xist mutants in which repeats B and C have been deleted from the Xist endogenous locus [60,62,78]. The effect on gene silencing along the Xi varies in these studies, most likely reflecting the different time points at which X-linked gene expression was assessed during XCI. Nevertheless, none of the studies reported complete abrogation of XCI initiation. Thus, as gene silencing can be established upon induction of a defective Xist RNA that lacks repeats B and C and is unable to recruit PRC1/PRC2, although slightly less efficiently, Polycomb recruitment and transcriptional silencing appear to be largely uncoupled [62] Accordingly, the hnRNP K-dependent accumulation of H3K27me3 and H2AK119ub along the Xi seems to be necessary to stabilize silencing during XCI, rather than initiating it [621. This model is in line with the observation that the subset of X-linked genes that are not silenced upon deletion of HDAC3 don't accumulate Polycomb marks upon Xist RNA spreading [831. Regardless of the interplay between Polycomb recruitment and gene silencing establishment, the initial recruitment of noncanonical PRC1 activity to the Xi leads to PRC2 accumulation, indicating that PRC1 is responsible for the Xist-dependent recruitment of PRC2 to the Xi [64]. Accordingly, deleting the core catalytic subunit of PRC1, RING1A/B, or the non-canonical PRC1 components PCGF3 and PCGF5 strongly reduced the Xist-dependent deposition of H2AK119ub and H3K27me3 during XCI [641. The recruitment of PRC2 downstream of PRC1 along the Xi is supported by several other lines of evidence, including the fact that none of the core PRC2 components were identified in the proteomic studies, the identification of PRC1 components as Xist interactors [73], and the relative dynamics of H2AK119ub accumulation along the Xi early during XCI, occurring upstream of H3K27me3 [83]. LBR Another factor that has been proposed to associate with Xist RNA is the Lamin B receptor (LBR). LBR is a transmembrane protein anchored to the inner nuclear membrane where it binds Lamin B and works as a structural scaffold for proteins involved in chromatin silencing PLOS Genetics | https://doi.orq/10.1371/iournal.pqen.1008333 September 19, 2019 7/17 '(S>-PLOS I GENETICS at the nuclear lamina [951. McHugh and colleagues initially showed impaired silencing of two X-linked genes upon LBR knockdown [74]. Subsequently, Chen and colleagues identified the LBR element required to mediate the interaction with Xist RNA and also showed that losing this interaction results in defective gene silencing [96]. The same study proposed LBR to play a key role in recruiting the Xi to the nuclear lamina, suggesting that this nuclear repositioning might be required for XCI [96]. However, the Xist-coated chromosome can still be recruited to the nuclear lamina in case of a major silencing defect, for example, upon SPEN depletion [96], and the actively transcribed Xa has been shown to localize to the nuclear lamina in both male and female nuclei [97], indicating that the recruitment of the Xi to the nuclear lamina is not sufficient to initiate gene silencing. Furthermore, LBR mutant mice do not show a clear sex bias in embryonic lethality as would be expected in case of impaired XCI [98,99]. More recently, Nesterova and colleagues performed an X-chromosome-wide analysis of Xist-medi-ated silencing upon differentiation of LBR knockout ESCs and were able to show that lack of LBR leads to a minor silencing defect [78]- Deleting the element of Xist RNA mediating the interaction with LBR resulted in slightly more affected gene silencing, but the overall effect on XCI remained weak. Thus, although the LBR-Xist interaction might be necessary to maintain and/or stabilize gene repression, the exact role of LBR during XCI still remains unclear. Interactors of Xist implicated in Xist RNA chromatin association One of the most fascinating and yet open questions about the function of Xist concerns how it remains associated only with the chromosome in cis rather than diffusing to neighboring chromosomes in trans. Based on RNA fluorescent in situ hybridization analysis (FISH), the Xist RNA domain appears to be confined to the same nuclear territory of Xi throughout the entire cell cycle [100], although human XIST was reported to dissociate from the Xi during mitosis [2,101], and this observation was confirmed by live-cell imaging in mouse cells [102]. Regardless of the fact that Xist RNA might be temporarily displaced from the Xi through cell division, its localization does not seem to rely on a sequence-specific mechanism. For example, neither RNase H nor DNase treatment affects the Xist RNA domain within the nuclear space, indicating that Xist does not bind the Xi simply via formation of RNA/DNA hybrids [101]. Rather, as Xist RNA remains with the nuclear matrix fraction after removal of chromosomal DNA, its localization to the Xi chromatin might be directed by proteins of the nuclear matrix [101]. hnRNP U (SAF-A) The matrix attachment protein hnRNP U has been proposed to play a key role in Xist RNA localization to chromatin. hnRNP U is enriched along the Xi [103,104], and its interaction with Xist was confirmed in proteomic studies [73-75]. In particular, hnRNP U seems to directly interact with exons 1 and 7 of both human and mouse XIST/Xist RNA [56,105]. Several knockdown studies validated the role of hnRNP U on Xist localization by reporting diffusion of Xist RNA from the Xi territory upon hnRNP U depletion [61,73,74]. Nevertheless, hnRNP U is dispensable for the localization of human XIST [106], and its requirement to localize Xist RNA seems to be at least partially cell type specific, indicating that Xist localization is likely to rely on a combination of anchoring factors rather than on hnRNP U alone [106,107]. Furthermore, the impact of hnRNP U on XCI establishment and/or maintenance remains to be fully understood. Indeed, hnRNP U was initially suggested to be necessary for the establishment of gene silencing [61], but given its recruitment to the Xi at a late time point during XCI, its functional relevance is more likely related to XCI maintenance [104]. PLOS Genetics | https://doi.orq/10.1371/iournal.pqen.1008333 September 19, 2019 8/17 '©'PLOS I GENETICS CIZÍ Cipl-interacting zinc finger protein 1 (CIZ1) is another nuclear matrix protein identified as a Xist interactor [73] and recently found to contribute to Xist localization [57,58]. CIZ1 has been originally characterized as a binding factor of key regulators of DNA replication and contrary to hnRNP U, is recruited by Xist RNA during the earliest stages of XCI [57,58]. Functionally, it comprises several domains, including three DNA binding zinc finger motifs [108] and a C-ter-minal nuclear matrix-anchoring MH3 domain (matrin 3-homologous domain 3) [109]. Thus, one tempting hypothesis is that CIZ1 might act as a bifunctional protein scaffold able to bind the Xi DNA via its zinc finger motifs and Xist RNA via the C-terminal nuclear matrix-anchoring domain. Indeed, the C-terminal domain is required to recruit CIZ1 to Xi, and deleting the E repeat within exon 7 of Xist abolishes this recruitment [58]. Nevertheless, whether CIZ1 directly interacts with Xist RNA or rather anchors the Xi via binding to other Xist RNA inter-actors is not yet clear. It should be noted that in hnRNP U knockout cells, CIZ1 remains localized with Xist RNA, although Xist RNA is dispersed throughout the nucleoplasm. Thus, CIZ1 and hnRNP U interact with Xist independently of each other, highlighting the complexity of the mechanisms directing Xist RNA coating and the formation of the Xi territory [57]. Finally, CIZ1 depletion results in dispersed Xist localization in mouse embryonic fibroblast (MEFs) and activated B and T lymphocytes but is certainly dispensable for XCI initiation as knockout mice survive embryonic development [58]. Nevertheless, CIZ1 null mice develop 100% penetrant female-specific lymphoproliferative disorder, indicating that loss of proper Xist localization in lymphocytes and impaired XCI might be the cause of the sex-specific phenotype. Consistent with this, deletion of Xist RNA in the mouse blood compartment was previously shown to lead to hematologic cancer [110]. However, none of these studies precisely assessed the allele-specific reactivation of X-linked genes along the Xi, so that the interplay between Xist RNA derealization, X-linked gene dosage, and cancer development is yet to be clearly defined. Factors involved in chromosome-wide 3D structural reorganization of theXi Another remarkable feature of Xist RNA is its ability to direct a major reorganization of the 3D spatial architecture of the Xi. Evidence of such structural reshaping was originally provided by assessing the local organization of X-linked genes during XCI [86]- In particular, this study revealed that upon gene silencing X-inactivated genes are relocalised from the periphery of the Xist RNA domain to a more internal compartment, which is depleted of RNA Pol II and general transcription factors, whereas genes that escape XCI remain external of the silent compartment [86]. Similarly, X-linked transcribed long interspersed nuclear elements (LINE) are spatially separated from the Xist RNA domain early upon XCI and only start to intermingle with it at a later stage, when silencing has occurred [111]. More recently, our understanding of the Xi's structural changes occurring during XCI has been greatly enhanced by the development of the Chromosome Conformation Capture (3C) technologies, which measure the physical arrangement of DNA in the 3D nuclear space. 3C methods unveiled several levels of higher order chromatin folding, including sub-megabase-scale topologically associating domains (TAD), defined as regions of DNA that preferentially contact each other along the chromosomes [40,112]. Comparative Hi-C analysis of the Xi 3D structure in mouse, human, and macaque has revealed that in contrast to the Xa and autosomes, the Xi is relatively depleted of TADs and appear to be folded into an unusual bipartite structure in which two large megadomains are separated by a boundary region encompassing the DXZ4/Dxz4 macrosatel-lite [75,113-116]. However, how Xist RNA coating results in partitioning of the Xi into PLOS Genetics | https://doi.orq/10.1371/iournal.pqen.1008333 September 19, 2019 9/17 '(S>-PLOS I GENETICS megadomains via DXZ4 is not yet clear. Similarly, the exact causality and temporal dynamics between the collapse of TADs along the Xi and the establishment of gene silencing also remains to be defined. For example, Xist RNA was proposed to direct the 3D organization of Xi by repelling architectural factors, thus possibly avoiding the establishment of a transcription-favorable chromatin status (Minajigi 2015). However, how Xist RNA can repulse a subset of factors involved in chromatin 3D organization remains elusive, as do the sequence elements within Xist RNA that would account for this function. SmcHDl Recently, the structural maintenance of chromosomes flexible hinge domain containing 1 (SmcHDl) protein was hypothesized to participate in reshaping the 3D conformation of Xi [117-1191. SmcHDl interacts with Xist RNA [75] and was previously found to be required for XCI maintenance, as its deletion in vivo does not affect XCI initiation but results in reactivation of approximately 10% of X-linked genes along Xi [120,121]. Mechanistically, SmcHDl has previously been shown to allow for CpG island methylation along the Xi; however, this does not account for its function in maintaining X-linked gene silencing [121]. As SmcHDl carries an SMC hinge domain normally found in proteins involved in chromosome condensation and compaction, its role in XCI maintenance might rather be structural. Indeed, knockdown of SmcHDl leads to decompaction of the human Xi, similar to XIST depletion [122], and strengthening of TAD-like structures along Xi in several cell types [117-119], although whether this structural reorganization is accompanied by transcriptional changes remains unclear [117,119]. Polycomb factors and Xi 3D organisation Another unexplored possibility is that the distribution of chromatin features along the X chromosome before XCI initiation might direct its structural reorganization. For example, the involvement of the Polycomb complexes in directing the 3D folding of chromosomes is increasingly recognized [123-125]. In particular, genomic loci that are repressed by PRC1 have been shown to form self-interacting domains of compacted chromatin that are thought to exclude the transcriptional machinery and to transmit the silent state of PRC1 targets through cell division [124]. In the context of XCI, X-linked regions that preferentially accumulate Polycomb-dependent histone marks have been reported to be pre-marked by Polycomb prior to Xist spreading [83], indicating that indeed the distribution of Polycomb along the X before XCI might direct its spatial reorganization during XCI. In line with this hypothesis, the X-linked regions pre-marked by Polycomb are spatially located in close 3D proximity of the Xist locus and correspond to the regions that become first targeted by Xist RNA at the beginning of XCI [83,126]. Finally, Polycomb complexes might contribute to the 3D reorganization of the Xi by supporting the formation of membrane-less organelles similar to paraspeckles, defined as protein-rich nuclear condensates built around a specific IncRNA scaffold and able to influence gene regulation by sequestrating a subset of specific proteins by RNA-protein interactions (for review [127]). Indeed, PRCl-bound chromatin has been recently shown to undergo liquid-liquid phase separation (LLPS) to form nuclear aggregates that colocalize with H3K27me3-dense chromatin regions [1281. Interestingly, the potential involvement of phase-separated aggregates in shaping the 3D structure of Xi is not limited to PRC1 but is likely to concern other interactors of Xist RNA. For example, FUS, hnRNPA2Bl, and RBM14 all carry low-complexity domains that are prone to mediate phase separation and were previously reported to be implicated in the formation of paraspeckles [127,129-131]. Given that phase-separated aggregates PLOS Genetics | https://doi.orq/10.1371/iournal.pqen.1008333 September 19, 2019 10/17 '(S>-PLOS I GENETICS have been proposed to play a role in the formation of heterochromatin [132,1331, an exciting hypothesis is that Xist RNA might direct the reorganization of Xi by forcing a local high concentration of specific proteins via RNA-protein interactions, thus creating a phase-separated silent compartment in which factors important for the establishment of facultative heterochromatin are sequestered. However, whether or not such a structure is assembled and its potential impact on the initiation and/or maintenance of XCI remains to be proven. Conclusions Almost thirty years after its discovery, Xist RNA continues to provide a powerful model system for exploring a multitude of epigenetic mechanisms, including the developmentally regulated formation of facultative heterochromatin and the 3D organization of the genome in nuclear space. Future studies are likely to unveil the molecular mechanisms through which noncoding RNAs induce or change the 3D structure of the genome, as well as whether these changes in structure are a cause or a consequence of changes in gene activity and how they play a role in maintaining differences between active and inactive compartments of the mammalian genomes. Acknowledgments The authors thank Jan Zylicz for critical and thoughtful reading of the manuscript. References 1. Brown SDM. XIST and the mapping of the X chromosome inactivation centre [Internet]. BioEssays. 1991. pp. 607-612. https://doi.org/10.1002/bies.950131112 PMID: 1772416 2. Brown CJ, Hendrich BD, Rupert JL, Lafreniere RG, Xing Y, Lawrence J, et al. The human XIST gene: analysis of a 17 kb inactive X-specific RNA that contains conserved repeats and is highly localized within the nucleus. Cell. 1992; 71:527-542. https://doi.orq/10.1016/0092-8674(92)90520-m PMID: 1423611 3. Borsani G, Tonlorenzi R, Simmler MC, Dandolo L, Arnaud D, Capra V, et al. Characterization of a murine gene expressed from the inactive X chromosome. Nature. 1991; 351: 325-329. https://doi.org/ 10.1038/351325a0 PMID: 2034278 4. Brockdorff N, Ashworth A, Kay GF, Cooper P, Smith S, McCabe VM, et al. Conservation of position and exclusive expression of mouse Xist from the inactive X chromosome. Nature. 1991; 351: 329-331. https://doi.org/10.1038/351329a0 PMID: 2034279 5. Brockdorff N, Ashworth A, Kay GF, McCabe VM, Norris DP, Cooper PJ, et al. The product of the mouse Xist gene is a 15 kb inactive X-specific transcript containing no conserved ORF and located in the nucleus. Cell. 1992; 71: 515-526. https://doi.orq/10.1016/0092-8674(92)90519-i PMID: 1423610 6. Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, Baldwin J, et al. 2001. Initial sequencing and analysis of the human genome. Nature. 409: 860. https://doi.org/10.1038/35057062 PMID: 11237011 7. Brannan CI, Dees EC, Ingram RS. The product of the H19 gene may function as an RNA. Molecular and cellular. Am Soc Microbiol; 1990; Available: https://mcb.asm.Org/content/10/1/28.short 8. Bartolomei MS, Zemel S, Tilghman SM. Parental imprinting of the mouse H19gene. Nature. 1991; 351:153-155. https://doi.org/10.1038/351153a0 PMID: 1709450 9. Wutz A, Smrzka OW, Schweifer N, Schellander K, Wagner EF, Barlow DP. Imprinted expression of the Igf2r gene depends on an intronic CpG island. Nature. 1997. pp. 745-749. https://doi.org/10.1038/ 39631 PMID: 9338788 10. Sleutels F, Zwart R, Barlow DP. The non-coding Air RNA is required for silencing autosomal imprinted genes. Nature. 2002. pp. 810-813. https://doi.orq/10.1038/41581 Oa PMID: 11845212 11. Okazaki Y, Furuno M, KasukawaT, Adachi J, Bono H, Kondo S, et al. RIKEN Genome Exploration Research Group Phase I & II Team Analysis of the mouse transcriptome based on functional annotation of 60,770 full-length cDNAs. Nature. 2002; 420: 563-573. https://doi.org/10.1038/nature01266 PMID: 12466851 12. Katayama S, Tomaru Y, KasukawaT, Waki K, Nakanishi M, Nakamura M, et al. RIKEN Genome Exploration Research Group; Genome Science Group (Genome Network Project Core Group); PLOS Genetics | https://doi.org/10.1371/iournal.pqen.1008333 September 19, 2019 11/17 '(S>-PLOS I GENETICS FANTOM Consortium. Antisense transcription in the mammalian transcriptome. Science. 2005; 309: 1564-1566. https://doi.Org/10.1126/science. 1112009 PMID: 16141073 13. ENCODE Project Consortium, Birney E, Stamatoyannopoulos JA, Dutta A, Guigö R, GingerasTR, et al. Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project. Nature. 2007; 447: 799-816. https://doi.org/10.1038/nature05874 PMID: 17571346 14. Guttman M, Amit I, Garber M, French C, Lin MF, Feldser D, et al. Chromatin signature reveals overa thousand highly conserved large non-coding RNAs in mammals. Nature. 2009; 458: 223-227. https:// doi.org/10.1038/nature07672 PMID: 19182780 15. PontingCP, Oliver PL, ReikW. Evolution and functions of long noncoding RNAs. Cell. 2009; 136: 629-641. https://doi.Org/10.1016/i.cell.2009.02.006 PMID: 19239885 16. Deläs MJ, Sabin LR, Dolzhenko E, Knott SR, Munera Maravilla E, Jackson BT, et al. IncRNA requirements for mouse acute myeloid leukemia and normal differentiation. Elife. 2017; 6. https://doi.org/10. 7554/eLife.25607 PMID: 28875933 17. Pauli A, Valen E, Lin MF, Garber M, Vastenhouw NL, Levin JZ, et al. Systematic identification of long noncoding RNAs expressed during zebrafish embryogenesis. Genome Res. 2012; 22: 577-591. https://doi.Org/10.1101/qr.133009.111 PMID: 22110045 18. Nam J-W, Bartel DP. Long noncoding RNAs in C. elegans. Genome Res. 2012; 22: 2529-2540. https://doi.Org/10.1101/qr.140475.112 PMID: 22707570 19. Barr ML, Bertram EG. A morphological distinction between neurones of the male and female, and the behaviour of the nucleolar satellite during accelerated nucleoprotein synthesis. Nature. 1949; 163: 676. https://doi.org/10.1038/163676a0 PMID: 18120749 20. Balaton BP, Brown CJ. Escape Artists of the X Chromosome. Trends Genet. 2016; 32: 348-359. https://doi.orq/10.1016/i.tiq.2016.03.007 PMID: 27103486 21. Hook EB, Warburton D. The distribution of chromosomal genotypes associated with Turner's syndrome: livebirth prevalence rates and evidence for diminished fetal mortality and severity in genotypes associated with structural X abnormalities or mosaicism. Hum Genet. Springer; 1983; 64:24-27. 22. Cockwell A, MacKenzie M, Youings S, Jacobs P. A cytogenetic and molecular study of a series of 45, X fetuses and their parents [Internet]. Journal of Medical Genetics. 1991. pp. 151-155. https://doi.org/ 10.1136/imq.28.3.151 PMID: 1675683 23. Schurz H, Salie M, Tromp G, Hoal EG, Kinnear CJ, Möller M. The X chromosome and sex-specific effects in infectious disease susceptibility. Hum Genomics. 2019; 13: 2. https://doi.Org/10.1186/ S40246-018-0185-Z PMID: 30621780 24. Libert C, Dejager L, Pinheiro I. The X chromosome in immune functions: when a chromosome makes the difference. Nat Rev Immunol. 2010; 10: 594-604. https://doi.Org/10.1038/nri2815 PMID: 20651746 25. Lee JT, Davidow LS, Warshawsky D. Tsix, a gene antisense to Xist at the X-inactivation centre. Nat Genet. 1999; 21: 400-404. https://doi.org/10.1038/7734 PMID: 10192391 26. LeeJT. Disruption of imprinted X inactivation by parent-of-origin effects at Tsix. Cell. 2000; 103:17-27. https://doi.Org/10.1016/s0092-8674(00)00101 -x PMID: 11051544 27. SadoT, Wang Z, Sasaki H, Li E. Regulation of imprinted X-chromosome inactivation in mice by Tsix. Development. 2001; 128:1275-1286. PMID: 11262229 28. Luikenhuis S, Wutz A, Jaenisch R. Antisense Transcription through theXist Locus Mediates Tsix Function in Embryonic Stem Cells. Mol Cell Biol. American Society for Microbiology Journals; 2001; 21: 8512-8520. 29. Shibata S, Lee JT. Tsix transcription- versus RNA-based mechanisms in Xist repression and epige-netic choice. Curr Biol. 2004; 14:1747-1754. https://doi.Org/10.1016/i.cub.2004.09.053 PMID: 15458646 30. OhhataT, Hoki Y, Sasaki H, SadoT. Crucial role of antisense transcription across the Xist Promoterin Tsix-mediated Xist chromatin modification [Internet]. Development. 2007. pp. 227-235. https://doi.org/ 10.1242/dev.008490 PMID: 18057104 31. Sado T, Hoki Y, Sasaki H. Tsix silences Xist through modification of chromatin structure. Dev Cell. 2005; 9:159-165. https://doi.Org/10.1016/i.devcel.2005.05.015 PMID: 15992549 32. Debrand E, Chureau C, Arnaud D. Functional Analysis of the DXPas34Locus, a 3' Regulator of Xist Expression. Molecular and cellular. Am Soc Microbiol; 1999; Available: https://mcb.asm.org/content/ 19/12/8513.short 33. Anguera MC, Ma W, Clift D, Namekawa S, Kelleher RJ 3rd, Lee JT. Tsx produces a long noncoding RNA and has general functions in the germline, stem cells, and brain. PLoS Genet. 2011; 7: e1002248. https://doi.orq/10.1371 /journal.pqen. 1002248 PMID: 21912526 PLOS Genetics | https://doi.org/10.1371/iournal.pqen.1008333 September 19, 2019 12/17 '(S>-PLOS I GENETICS 34. Ogawa Y, Lee JT. Xite, X-inactivation intergenic transcription elements that regulate the probability of choice. Mol Cell. 2003; 11: 731-743. PMID: 12667455 35. Stavropoulos N, Rowntree RK, Lee JT. Identification of developmental^ specific enhancers for Tsix in the regulation of X chromosome inactivation. Mol Cell Biol. 2005; 25: 2757-2769. https://doi.org/10. 1128/MCB.25.7.2757-2769.2005 PMID: 15767680 36. Chureau C, Chantalat S, Romito A, Galvani A, Duret L, Avner P, et al. Ftx is a non-coding RNA which affects Xist expression and chromatin structure within the X-inactivation center region [Internet]. Human Molecular Genetics. 2011. pp. 705-718. https://doi.orq/10.1093/hmq/ddq516 PMID: 21118898 37. Sun S, Del Rosario BC, Szanto A, Ogawa Y, Jeon Y, Lee JT. Jpx RNA activates Xist by evicting CTCF. Cell. 2013; 153:1537-1551. https://doi.Org/10.1016/i.cell.2013.05.028 PMID: 23791181 38. Tian D, Sun S, Lee JT. The long noncoding RNA, Jpx, is a molecular switch for X chromosome inactivation. Cell. 2010; 143: 390-403. https://doi.Org/10.1016/i.cell.2010.09.049 PMID: 21029862 39. Furlan G, Gutierrez Hernandez N, Huret C, Galupa R, van Bemmel JG, Romito A, et al. The Ftx Non-coding Locus Controls X Chromosome Inactivation Independently of Its RNA Products. Mol Cell. 2018; 70: 462-472.e8. https://doi.Org/10.1016/i.molcel.2018.03.024 PMID: 29706539 40. Nora EP, Lajoie BR, Schulz EG, Giorgetti L, Okamoto I, Servant N, et al. Spatial partitioning of the regulatory landscape of the X-inactivation centre. Nature. 2012; 485: 381-385. https://doi.org/10.1038/ naturel 1049 PMID: 22495304 41. Gendrel A-V, Heard E. Noncoding RNAs and epigenetic mechanisms during X-chromosome inactivation. Annu Rev Cell Dev Biol. 2014; 30: 561-580. https://doi.org/10.1146/annurev-cellbio-101512-122415 PMID: 25000994 42. Galupa R, Heard E. X-Chromosome Inactivation: A Crossroads Between Chromosome Architecture and Gene Regulation. Annu Rev Genet. 2018; 52: 535-566. https://doi.org/10.1146/annurev-qenet-120116-024611 PMID: 30256677 43. Furlan G, Rougeulle C. Function and evolution of the long noncoding RNA circuitry orchestrating X-chromosome inactivation in mammals. Wiley Interdiscip Rev RNA. Wiley Online Library; 2016; 7:702-722. 44. Galupa R, Heard E. X-chromosome inactivation: new insights into cis and trans regulation. CurrOpin Genet Dev. 2015; 31: 57-66. https://doi.orq/10.1016/i.qde.2015.04.002 PMID: 26004255 45. PennyGD, KayGF, SheardownSA, RastanS, BrockdorffN. Requirement for Xist in X chromosome inactivation. Nature. 1996; 379:131-137. https://doi.org/10.1038/379131a0 PMID: 8538762 46. Marahrens Y, Panning B, Dausman J, Strauss W, Jaenisch R. Xist-deficient mice are defective in dosage compensation but not spermatogenesis. Genes Dev. 1997; 11:156-166. https://doi.org/10.1101/ gad.11.2.156 PMID: 9009199 47. Wutz A, Jaenisch R. A shift from reversible to irreversible X inactivation is triggered during ES cell differentiation. Mol Cell. 2000; 5: 695-705. PMID: 10882105 48. Chow JC, Hall LL, Baldry SEL, Thorogood NP, Lawrence JB, Brown CJ. Inducible XIST-dependent X-chromosome inactivation in human somatic cells is reversible. Proc Natl Acad Sci USA. 2007; 104: 10104-10109. https://doi.orq/10.1073/pnas.0610946104 PMID: 17537922 49. TangYA, Huntley D, Montana G, Cerase A, NesterovaTB, BrockdorffN. Efficiency of Xist-mediated silencing on autosomes is linked to chromosomal domain organisation. Epigenetics Chromatin. 2010; 3:10. https://doi.org/10.1186/1756-8935-3-10 PMID: 20459652 50. Loda A, Brandsma JH, Vassilev I, Servant N, Loos F, Amirnasr A, et al. Genetic and epigenetic features direct differential efficiency of Xist-mediated silencing at X-chromosomal and autosomal locations. Nat Commun. 2017; 8: 690. https://doi.Org/10.1038/s41467-017-00528-1 PMID: 28947736 51. Wutz A, Rasmussen TP, Jaenisch R. Chromosomal silencing and localization are mediated by different domains of Xist RNA. Nat Genet. 2002; 30:167-174. https://doi.Org/10.1038/nq820 PMID: 11780141 52. Nesterova TB, Slobodyanyuk SY, Elisaphenko EA, Shevchenko Al, Johnston C, Pavlova ME, et al. Characterization of the genomic Xist locus in rodents reveals conservation of overall gene structure and tandem repeats but rapid evolution of unique sequence. Genome Res. 2001; 11: 833-849. https:// doi.org/10.1101/qr.174901 PMID: 11337478 53. Hoki Y, Kimura N, Kanbayashi M, Amakawa Y, OhhataT, Sasaki H, et al. A proximal conserved repeat in the Xist gene is essential as a genomic element for X-inactivation in mouse. Development. 2009; 136:139-146. https://doi.org/10.1242/dev.026427 PMID: 19036803 54. Beletskii A-K. Hong Y, Pehrson J, Egholm M, Strauss WM. PNA interference mapping demonstrates functional domains in the noncoding RNA Xist [Internet]. Proceedings of the National Academy of Sciences. 2001. pp. 9215-9220. https://doi.Org/10.1073/pnas. 161173098 PMID: 11481485 55. Sarma K, Levasseur P, Aristarkhov A, Lee JT. Locked nucleic acids (LNAs) reveal sequence requirements and kinetics of Xist RNA localization to the X chromosome. Proc Natl Acad Sci USA. 2010; 107: 22196-22201. https://doi.Org/10.1073/pnas. 1009785107 PMID: 21135235 PLOS Genetics | https://doi.org/10.1371/iournal.pqen.1008333 September 19, 2019 13/17 '(S>-PLOS I GENETICS 56. Yamada N, Hasegawa Y, Yue M, Hamada T, Nakagawa S, Ogawa Y. Xist Exon 7 Contributes to the Stable Localization of Xist RNA on the Inactive X-Chromosome [Internet]. PLoS Genetics. 2015. p. e1005430. https://doi.orq/10.1371 /journal.pqen. 1005430 PMID: 26244333 57. Sunwoo H, Colognori D, Froberg JE, Jeon Y, Lee JT. Repeat E anchors Xist RNA to the inactive X chromosomal compartment through CDKN1 A-interacting protein (CIZ1). Proc Natl Acad Sci USA. 2017; 114:10654-10659. https://doi.orq/10.1073/pnas. 1711206114 PMID: 28923964 58. Ridings-Figueroa R, Stewart ER, NesterovaTB, Coker H, Pintacuda G, Godwin J, et al. The nuclear matrix protein CIZ1 facilitates localization of Xist RNA to the inactive X-chromosome territory. Genes Dev. 2017; 31: 876-888. https://doi.org/10.1101/qad.295907.117 PMID: 28546514 59. Jeon Y, Lee JT. YY1 tethers Xist RNA to the inactive X nucleation center. Cell. 2011; 146:119-133. https://doi.Org/10.1016/i.cell.2011.06.026 PMID: 21729784 60. Colognori D, Sunwoo H, Kriz AJ, Wang C-Y, Lee JT. Xist Deletional Analysis Reveals an Interdepen-dency between Xist RNA and Polycomb Complexes for Spreading along the Inactive X [Internet]. Molecular Cell. 2019. pp. 101-117.e10. https://doi.orq/10.1016/i.molcel.2019.01.015 PMID: 30827740 61. Hasegawa Y, Brockdorff N, Kawano S, Tsutui K, Tsutui K, Nakagawa S. The matrix protein hnRNP U is required for chromosomal localization of Xist RNA. DevCell. 2010; 19: 469-476. https://doi.org/10. 1016/i.devcel.2010.08.006 PMID: 20833368 62. Bousard A, Raposo AC, Zylicz JJ, Picard C, Pires VB, Qi Y, et al. Exploring the role of Polycomb recruitment in Xist-mediated silencing of the X chromosome in ES cells [Internet]. 63. Pintacuda G, Wei G, Roustan C, Kirmizitas BA, Solcan N, Cerase A, et al. hnRNPK Recruits PCGF3/ 5-PRC1 to the Xist RNA B-Repeat to Establish Polycomb-Mediated Chromosomal Silencing. Mol Cell. 2017; 68: 955-969.e10. https://doi.orq/10.1016/i.molcel.2017.11.013 PMID: 29220657 64. Almeida M, Pintacuda G, Masui O, Koseki Y, Gdula M, Cerase A, et al. PCGF3/5-PRC1 initiates Polycomb recruitment in X chromosome inactivation [Internet]. Science. 2017. pp. 1081-1084. https://doi. org/10.1126/science.aal2512 PMID: 28596365 65. Zhao J, Sun BK, Erwin JA, Song J-J, Lee JT. Polycomb proteins targeted by a short repeat RNA to the mouse X chromosome. Science. 2008; 322: 750-756. https://doi.org/10.1126/science.1163045 PMID: 18974356 66. de Napoles M, Nesterova T, Brockdorff N. Early loss of Xist RNA expression and inactive X chromosome associated chromatin modification in developing primordial germ cells. PLoS One. 2007; 2: e860. https://doi.orq/10.1371 /iournal.pone.0000860 PMID: 17848991 67. Mak W, Baxter J, Silva J, Newall AE, Otte AP, Brockdorff N. Mitotically stable association of polycomb group proteins eed and enxl with the inactive x chromosome in trophoblast stem cells. Curr Biol. 2002; 12:1016-1020. https://doi.orq/10.1016/s0960-9822(02)00892-8 PMID: 12123576 68. Plath K, Fang J, Mlynarczyk-Evans SK, Cao R, Worringer KA, Wang H, et al. Role of histone H3 lysine 27 methylation in X inactivation. Science. 2003; 300:131-135. https://doi.Org/10.1126/science. 1084274 PMID: 12649488 69. Silva J, Mak W, Zvetkova I, Appanah R, Nesterova TB, Webster Z, et al. Establishment of histone h3 methylation on the inactive X chromosome requires transient recruitment of Eed-Enx1 polycomb group complexes. DevCell. 2003; 4: 481^95. PMID: 12689588 70. Wang L, Brown JL, Cao R, Zhang Y, Kassis JA, Jones RS. Hierarchical recruitment of polycomb group silencing complexes. Mol Cell. 2004; 14: 637-646. https://doi.Org/10.1016/i.molcel.2004.05.009 PMID: 15175158 71. Fang J, Chen T, Chadwick B, Li E, Zhang Y. Ringl b-mediated H2A ubiquitination associates with inactive X chromosomes and is involved in initiation of X inactivation. J Biol Chem. 2004; 279: 52812-52815. https://doi.Org/10.1074/ibc.C400493200 PMID: 15509584 72. Sarma K, Cifuentes-Rojas C, Ergun A, Del Rosario A, Jeon Y, White F, et al. ATRX directs binding of PRC2 to Xist RNA and Polycomb targets. Cell. 2014; 159: 869-883. https://doi.Org/10.1016/i.cell. 2014.10.019 PMID: 25417162 73. Chu C, Zhang QC, da Rocha ST, Flynn RA, Bharadwaj M, Calabrese JM, et al. Systematic discovery of Xist RNA binding proteins. Cell. 2015; 161: 404-416. https://doi.Org/10.1016/i.cell.2015.03.025 PMID: 25843628 74. McHugh CA, Chen C-K, Chow A, Surka CF, Tran C, McDonel P, et al. The Xist IncRNA interacts directly with SHARP to silence transcription through HDAC3. Nature. 2015; 521: 232-236. https://doi. orq/10.1038/naturel4443 PMID: 25915022 75. Minajigi A, Froberg J, Wei C, Sunwoo H, Kesner B, Colognori D, et al. Chromosomes. A comprehensive Xist interactome reveals cohesin repulsion and an RNA-directed chromosome conformation. Science. 2015; 349. https://doi.org/10.1126/science.aab2276 PMID: 26089354 PLOS Genetics | https://doi.org/10.1371/iournal.pqen.1008333 September 19, 2019 14/17 '(S>-PLOS I GENETICS 76. Moindrot B, Cerase A, Coker H, Masui O, Grijzenhout A, Pintacuda G, et al. A Pooled shRNA Screen Identifies Rbm15, Spen, and Wtapas Factors Required for Xist RNA-Mediated Silencing. Cell Rep. 2015; 12: 562-572. https://doi.Org/10.1016/i.celrep.2015.06.053 PMID: 26190105 77. Monfort A, Di Minin G, Postlmayr A, Freimann R, Arieti F, Thore S, et al. Identification of Spen as a Crucial FactorforXist Function through Forward Genetic Screening in Haploid Embryonic Stem Cells. Cell Rep. 2015; 12: 554-561. https://doi.Org/10.1016/i.celrep.2015.06.067 PMID: 26190100 78. Nesterova TB, Wei G, Coker H, Pintacuda G. Systematic Allelic Analysis Defines the Interplay of Key Pathways in X Chromosome Inactivation. bioRxiv. biorxiv.org; 2018; https://www.biorxiv.org/content/ 10.1101 /477232v1 .abstract 79. Shi Y, Downes M, Xie W, Kao HY, Ordentlich P, Tsai CC, et al. Sharp, an inducible cofactor that integrates nuclear receptor repression and activation. Genes Dev. 2001; 15:1140-1151. https://doi.org/ 10.1101/gad.871201 PMID: 11331609 80. You S-H, Lim H-W, Sun Z, Broache M, Won K-J, Lazar MA. Nuclear receptor co-repressors are required forthe histone-deacetylase activity of HDAC3 in vivo. Nat Struct Mol Biol. 2013; 20:182-187. https://doi.org/10.1038/nsmb.2476 PMID: 23292142 81. GuentherMG, Barak O, Lazar MA. The SMRT and N-CoR corepressors are activating cofactors for histone deacetylase 3. Mol Cell Biol. 2001; 21: 6091-6101. https://doi.org/10.1128/MCB.21.18.6091-6101.2001 PMID: 11509652 82. Mikami S, KanabaT, Mishima M. Structural insights into the recruitment of SMRT by the co-repressor SHARP under phosphorylative regulation [Internet]. 2013. 83. Zylicz JJ, Bousard A, Zumer K, Dossin F, Mohammad E, da Rocha ST, et al. The Implication of Early Chromatin Changes in X Chromosome Inactivation. Cell. 2019; 176:182-197.e23. https://doi.org/10. 1016/i.cell.2018.11.041 PMID: 30595450 84. Kohlmaier A, Savarese F, Lachner M, Martens J, Jenuwein T, Wutz A. A chromosomal memory triggered by Xist regulates histone methylation in X inactivation. PLoS Biol. 2004; 2: E171. https://doi.org/ 10.1371/journal.pbio.0020171 PMID: 15252442 85. Schoeftner S, Sengupta AK, Kubicek S, Mechtler K, Spahn L, Koseki H, et al. Recruitment of PRC1 function at the initiation of X inactivation independent of PRC2 and silencing. EMBO J. 2006; 25: 3110-3122. https://doi.org/10.1038/si.emboi.7601187 PMID: 16763550 86. Chaumeil J. A novel role for Xist RNA in the formation of a repressive nuclear compartment into which genes are recruited when silenced [Internet]. Genes & Development. 2006. pp. 2223-2237. https:// doi.org/10.1101/gad.380906 PMID: 16912274 87. Kuroda K, Han H, Tani S, Tanigaki K, Tun T, Furukawa T, et al. Regulation of marginal zone B cell development by MINT, a suppressor of Notch/RBP-J signaling pathway. Immunity. 2003; 18: 301-312. PMID: 12594956 88. Horiuchi K, KawamuraT, Iwanari H, Ohashi R, Naito M, KodamaT, et al. Identification of Wilms'tumor 1 -associating protein complex and its role in alternative splicing and the cell cycle. J Biol Chem. 2013; 288: 33292-33302. https://doi.org/10.1074/ibc.M113.500397 PMID: 24100041 89. Roignant J-Y, Soller M. m 6 A in mRNA: An Ancient Mechanism for Fine-Tuning Gene Expression [Internet]. Trends in Genetics. 2017. pp. 380-390. https://doi.Org/10.1016/i.tig.2017.04.003 PMID: 28499622 90. Zhao BS, Roundtree IA, He C. Post-transcriptional gene regulation by mRNA modifications. Nat Rev Mol Cell Biol. 2017; 18: 31-42. https://doi.org/10.1038/nrm.2016.132 PMID: 27808276 91. Liao S, Sun H, Xu C. YTH Domain: A Family of N6-methyladenosine (m6A) Readers. Genomics Prote-omics Bioinformatics. 2018; 16: 99-107. https://doi.Org/10.1016/i.gpb.2018.04.002 PMID: 29715522 92. Patil DP, Chen C-K, Pickering BF, Chow A, Jackson C, Guttman M, et al. m6A RNA methylation promotes XIST-mediated transcriptional repression [Internet]. Nature. 2016. pp. 369-373. https://doi.org/ 10.1038/nature19342 PMID: 27602518 93. Wang X, Zhao BS, Roundtree IA, Lu Z, Han D, Ma H, et al. N(6)-methyladenosine Modulates Messenger RNA Translation Efficiency. Cell. 2015; 161:1388-1399. https://doi.Org/10.1016/i.cell.2015.05. 014 PMID: 26046440 94. Wang X, Lu Z, Gomez A, Hon GC, Yue Y, Han D, et al. N6-methyladenosine-dependent regulation of messenger RNA stability. Nature. 2014; 505:117-120. https://doi.org/10.1038/nature12730 PMID: 24284625 95. G ruenbaum Y, Margalit A, Goldman RD, Shumaker DK, Wilson KL. The nuclear lamina comes of age. Nat Rev Mol Cell Biol. 2005; 6: 21-31. https://doi.org/10.1038/nrm1550 PMID: 15688064 96. Chen C-K, Blanco M, Jackson C, Aznauryan E, Ollikainen N, Surka C, et al. Xist recruits the X chromosome to the nuclear lamina to enable chromosome-wide silencing. Science. 2016; 354: 468-472. https://doi.org/10.1126/science.aae0047 PMID: 27492478 PLOS Genetics | https://doi.org/10.1371/iournal.pgen.1008333 September 19, 2019 15/17 '(S>-PLOS I GENETICS 97. Dyer KA, Canfield TK, Gartier SM. Molecular cytological differentiation of active from inactive X domains in interphase: implications for X chromosome inactivation. Cytogenet Cell Genet. 1989; 50: 116-120. https://doi.orq/10.1159/000132736 PMID: 2776476 98. Shultz LD, Lyons BL, Burzenski LM, Gott B, Samuels R, Schweitzer PA, et al. Mutations at the mouse ichthyosis locus are within the lamin B receptor gene: a single gene model for human Pelger—Huet anomaly. Hum Mol Genet. Oxford University Press; 2003; 12: 61-69. 99. Cohen TV, Klarmann KD, Sakchaisri K, Cooper JP, Kuhns D, Anver M, et al. The lamin B receptor under transcriptional control of C/EBPe is required for morphological but not functional maturation of neutrophils. Hum Mol Genet. Narnia; 2008; 17: 2921-2933. 100. Jonkers I, Monkhorst K, Rentmeester E, Grootegoed JA, Grosveld F, Gribnau J. Xist RNA is confined to the nuclear territory of the silenced X chromosome throughout the cell cycle. Mol Cell Biol. 2008; 28: 5583-5594. https://doi.Org/10.1128/MCB.02269-07 PMID: 18625719 101. ClemsonCM, McNeil JA, Willard HF, Lawrence JB. XIST RNA paints the inactive X chromosome at interphase: evidence for a novel RNA involved in nuclear/chromosome structure. J Cell Biol. 1996; 132: 259-275. https://doi.Org/10.1083/jcb. 132.3.259 PMID: 8636206 102. Ng K, Daigle N, Bancaud A, OhhataT, Humphreys P, Walker R, et al. A system for imaging the regulatory noncoding Xist RNA in living mouse embryonic stem cells. Mol Biol Cell. 2011; 22: 2634-2645. https://doi.Org/10.1091 /mbc.E11 -02-0146 PMID: 21613549 103. Heibig R, Fackelmayer FO. Scaffold attachment factor A (SAF-A) is concentrated in inactive X chromosome territories through its RGG domain. Chromosoma. 2003; 112:173-182. https://doi.org/10. 1007/S00412-003-0258-0 PMID: 14608463 104. Pullirsch D, Härtel R, Kishimoto H, Leeb M, Steiner G, Wutz A. The Trithoraxgroup protein Ash2l and Saf-A are recruited to the inactive X chromosome at the onset of stable X inactivation [Internet]. Development. 2010. pp. 935-943. https://doi.Org/10.1242/dev.035956 PMID: 20150277 105. Smeets D, Markaki Y, Schmid VJ, Kraus F, Tattermusch A, Cerase A, et al. Three-dimensional super-resolution microscopy of the inactive X chromosome territory reveals a collapse of its active nuclear compartment harboring distinct Xist RNA foci [Internet]. EpigeneticsS Chromatin. 2014. p. 8. https:// doi.orq/10.1186/1756-8935-7-8 PMID: 25057298 106. Kolpa HJ, Fackelmayer FO, Lawrence JB. SAF-A Requirement in Anchoring XIST RNA to Chromatin Varies in Transformed and Primary Cells. DevCell. 2016; 39: 9-10. https://doi.Org/10.1016/i.devcel. 2016.09.021 PMID: 27728783 107. Sakaguchi T, Hasegawa Y, Brockdorff N, Tsutsui K, Tsutsui KM, SadoT, et al. Control of Chromosomal Localization of Xist by hnRNP U Family Molecules. DevCell. 2016; 39:11-12. https://doi.org/ 10.1016/i.devcel.2016.09.022 PMID: 27728779 108. Warder DE, Keherly MJ. Ciz1, Cip1 interacting zinc finger protein 1 binds the consensus DNA sequence ARYSR(0-2)YYAC. J Biomed Sei. 2003; 10: 406-417. https://doi.Org/10.1007/bf02256432 PMID: 12824700 109. Ainscough JF-X, Ainscough JF, Rahman FA, Sercombe H, Sedo A, Gerlach B, et al. C-terminal domains deliver the DNA replication factor Ciz1 to the nuclear matrix [Internet]. Journal of Cell Science. 2006. pp. 115-124. 110. Yildirim E, Kirby JE, Brown DE, Mercier FE, Sadreyev Rl, Scadden DT, et al. Xist RNA is a potent suppressor of hematologic cancer in mice. Cell. 2013; 152:727-742. https://doi.Org/10.1016/i.cell.2013. 01.034 PMID: 23415223 111. Chow JC, CiaudoC, Fazzari MJ, Mise N, Servant N, Glass JL, et al. LINE-1 Activity in Facultative Het-erochromatin Formation during X Chromosome Inactivation [Internet]. Cell. 2010. pp. 956-969. https://doi.Org/10.1016/i.cell.2010.04.042 PMID: 20550932 112. Dixon JR, Selvaraj S, Yue F, Kim A, Li Y, Shen Y, et al. Topological domains in mammalian genomes identified by analysis of chromatin interactions [Internet]. Nature. 2012. pp. 376-380. https://doi.org/ 10.1038/nature11082 PMID: 22495300 113. Deng X, Ma W, Ramani V, Hill A, Yang F, Ay F, et al. Bipartite structure of the inactive mouse X chromosome. Genome Biol. 2015; 16:152. https://doi.Org/10.1186/s13059-015-0728-8 PMI D: 26248554 114. RaoSSP, Huntley MH, Durand NC, StamenovaEK, BochkovID, Robinson JT, et al. A 3D map of the human genome at kilobase resolution reveals principles of chromatin looping. Cell. 2014; 159:1665-1680. https://doi.Org/10.1016/i.cell.2014.11.021 PMID: 25497547 115. Giorgetti L, Lajoie BR, Carter AC, Attia M, Zhan Y, Xu J, et al. Structural organization of the inactive X chromosome in the mouse. Nature. 2016; 535:575-579. https://doi.org/10.1038/nature18589 PMID: 27437574 PLOS Genetics | https://doi.org/10.1371/iournal.pqen.1008333 September 19, 2019 16/17 '(S>-PLOS I GENETICS 116. DarrowEM, Huntley MH, DudchenkoO, Stamenova EK, Durand NC, SunZ, etal. Deletion of DXZ4 on the human inactive X chromosome alters higher-order genome architecture. Proc Natl Acad Sei U SA. 2016; 113: E4504-12. https://doi.org/10.1073/pnas.1609643113 PMID: 27432957 117. Jansz N, Keniry A, Trussart M, Bildsoe H, BeckT, Tonks ID, et al. Smchdl regulates long-range chromatin interactions on the inactive X chromosome and at Hox clusters. Nat Struct Mol Biol. 2018; 25: 766-777. https://doi.orq/10.1038/s41594-018-0111 -z PMID: 30127357 118. Wang C-Y, JeguT, Chu H-P, Oh HJ, Lee JT. SMCHD1 Merges Chromosome Compartments and Assists Formation of Super-Structures on the Inactive X. Cell. 2018; 174: 406-421 .e25. https://doi. org/10.1016/i.cell.2018.05.007 PMID: 29887375 119. Gdula MR, NesterovaTB, Pintacuda G, Godwin J, Zhan Y, Ozadam H, etal. The non-canonical SMC protein SmcHDl antagonises TAD formation and compartmentalisation on the inactive X chromosome. Nat Commun. 2019; 10: 30. https://doi.Org/10.1038/s41467-018-07907-2 PMID: 30604745 120. Blewitt ME, Gendrei A-V, Pang Z, Sparrow DB, Whitelaw N, Craig JM, et al. SmcHDl, containing a structural-maintenance-of-chromosomes hinge domain, has a critical role in X inactivation. Nat Genet. 2008; 40: 663-669. https://doi.org/10.1038/ng.142 PMID: 18425126 121. Gendrel A-V, Apedaile A, Coker H, Termanis A, Zvetkova I, Godwin J, et al. Smchdl-dependent and -independent pathways determine developmental dynamics of CpG island methylation on the inactive X chromosome. DevCell. 2012; 23: 265-279. https://doi.Org/10.1016/i.devcel.2012.06.011 PMID: 22841499 122. Nozawa R-S, NagaoK, Igami K-T, ShibataS, Shirai N, Nozaki N, etal. Human inactiveXchromosome is compacted through a PRC2-independent SMCHD1-HBiX1 pathway. Nat Struct Mol Biol. 2013; 20: 566-573. https://doi.Org/10.1038/nsmb.2532 PMID: 23542155 123. Denholtz M, Bonora G, Chronis C, Splinter E, de Laat W, Ernst J, et al. Long-Range Chromatin Contacts in Embryonic Stem Cells Reveal a Role for Pluripotency Factors and Polycomb Proteins in Genome Organization [Internet]. Cell Stem Cell. 2013. pp. 602-616. https://doi.Org/10.1016/i.stem. 2013.08.013 PMID: 24035354 124. Kundu S, Ji F, Sunwoo H, Jain G, Lee JT, Sadreyev Rl, et al. Polycomb Repressive Complex 1 Generates Discrete Compacted Domains that Change during Differentiation. Mol Cell. 2018; 71:191. https:// doi.org/10.1016/i.molcel.2018.06.022 PMID: 29979966 125. Schoenfelder S, Sugar R, Dimond A, Javierre B-M, Armstrong H, Mifsud B, et al. Polycomb repressive complex PRC1 spatially constrains the mouse embryonic stem cell genome. Nat Genet. 2015; 47: 1179-1186. https://doi.org/10.1038/ng.3393 PMID: 26323060 126. Engreitz JM, Pandya-Jones A, McDonel P, Shishkin A, Sirokman K, SurkaC, etal. The Xist IncRNA exploits three-dimensional genome architecture to spread across the X chromosome. Science. 2013; 341:1237973. https://doi.org/10.1126/science.1237973 PMID: 23828888 127. Fox AH, Nakagawa S, HiroseT, Bond CS. Paraspeckles: Where Long Noncoding RNA Meets Phase Separation [Internet]. Trends in Biochemical Sciences. 2018. pp. 124-135. https://doi.Org/10.1016/i. tibs.2017.12.001 PMID: 29289458 128. Tatavosian R, Kent S, Brown K, YaoT, Due HN, Huynh TN, et al. Nuclear condensates of the Polycomb protein chromobox2 (CBX2) assemble through phase separation. J Biol Chem. 2019; 294: 1451-1463. https://doi.org/10.1074/jbc.RA118.006620 PMID: 30514760 129. NaganumaT, NakagawaS.Tanigawa A, Sasaki YF, GoshimaN, HiroseT. Alternative3'-end processing of long noncoding RNA initiates construction of nuclear paraspeckles [Internet]. The EMBO Journal. 2012. pp. 4020-4034. https://doi.org/10.1038/emboi.2012.251 PMID: 22960638 130. West JA, Mito M, Kurosaka S, TakumiT, TanegashimaC, ChujoT, et al. Structural, super-resolution microscopy analysis of paraspeckle nuclear body organization [Internet]. The Journal of Cell Biology. 2016. pp. 817-830. https://doi.Org/10.1083/icb.201601071 PMID: 27646274 131. Vance C, Rogelj B, Hortobägyi T, De Vos KJ, Nishimura AL, Sreedharan J, et al. Mutations in FUS, an RNA processing protein, cause familial amyotrophic lateral sclerosis type 6. Science. 2009; 323: 1208-1211. https://doi.org/10.1126/science.1165942 PMID: 19251628 132. Strom AR, Emelyanov AV, Mir M, Fyodorov DV, Darzacq X, Karpen GH. Phase separation drives het-erochromatin domain formation. Nature. 2017; 547: 241-245. https://doi.Org/10.1038/nature22989 PMID: 28636597 133. Larson AG, Elnatan D, Keenen MM, Trnka MJ, Johnston JB, Burlingame AL, et al. Liquid droplet formation by HP1 a suggests a role for phase separation in heterochromatin. Nature. 2017; 547: 236-240. https://doi.org/10.1038/nature22822 PMID: 28636604 134. Loda A, GribnauJH. X chromosome inactivation: Spreading of silencing. Rotterdam: Erasmus University of Rotterdam; 2016 PLOS Genetics | https://doi.org/10.1371/iournal.pgen.1008333 September 19, 2019 17/17