The “lnc” between 3D chromatin structure and X chromosome inactivation

https://doi.org/10.1016/j.semcdb.2016.04.002Get rights and content

Abstract

The long non-coding RNA Xist directs a remarkable instance of developmentally regulated, epigenetic change known as X Chromosome Inactivation (XCI). By spreading in cis across the X chromosome from which it is expressed, Xist RNA facilitates the creation of a heritably silent, heterochromatic nuclear territory that displays a three-dimensional structure distinct from that of the active X chromosome. How Xist RNA attaches to and propagates across a chromosome and its influence over the three-dimensional (3D) structure of the inactive X are aspects of XCI that have remained largely unclear. Here, we discuss studies that have made significant contributions towards answering these open questions.

Introduction

X Chromosome Inactivation (XCI) is a critical step in the development of eutherian female organisms. In the mouse, XCI is implemented in two distinct waves during embryogenesis. In a process known as imprinted XCI, the cells of early 2–4 cell stage female conceptuses transcriptionally silence the paternal X chromosome (Xp) such that X-linked gene expression is derived almost exclusively from the maternal X (Xm) chromosome [1], [2], [3]. The inactive Xp is then maintained in almost all cells of the developing pre-implantation blastocyst, except for the epiblast cells. These cells, which give rise to the embryo proper, reverse silencing and sustain two transcriptionally active sex chromosomes (XaXa) [4]. Upon implantation, the differentiating epiblast cells transition through a second round of XCI, in which one of the two X chromosomes, this time chosen at random, undergoes transcriptional silencing and facultative heterochromatinization that is heritable throughout all subsequent cell divisions [5], [6], [7], [8], [9], [10], [11], [12]. Studies in a variety of model systems have revealed important distinctions in the implementation of both imprinted and random XCI across species [13], making it pertinent to emphasize that in this review, we limit our discussion to that of random XCI.

XCI is attributed with the alignment of X-linked gene dosage in differentiated female cells with that of their male (XY) counterparts [14], [15], [16]. Inhibition of XCI in the developing blastocyst affects survival of female embryos and appears to be caused by a failure to instate X-linked gene silencing in the future placenta [17], [18], [19]. While this phenotype demonstrates a requirement for imprinted XCI in female embryogenesis, it has hampered efforts to understand whether random XCI is essential for development of the embryo proper, which would require inhibition of XCI specifically in the epiblast but not the trophectoderm cells of an implanting blastocyst. This arrangement of tissue specific control over XCI can be achieved through tetraploid complementation of XCI-deficient epiblast cells, but has not yet been directly tested and as such, the requirement for random XCI in development remains unresolved [20].

The master regulator of XCI is the long non-coding (lnc) RNA Xist [21], [22], [23]. The Xist locus resides upon the X chromosome and is thought to remain inactive throughout the development of male mouse embryos and within the adult male organism [4], [17], [24], [25], [26], [27]. During female development however, Xist expression is tightly and dynamically regulated. Xist activity becomes detectable in female embryos at the 2–4-cell stage, at which time Xist RNA is transcribed solely from the Xp chromosome [13], [28], [29]. This expression pattern is maintained in cells of the trophectoderm and underlies the imprinted XCI that is observed in these cells and in the extra-embryonic tissues derived from them [30], [31]. In later stage pre-implantation embryos, the erasure of imprinted XCI in epiblast cells coincides with a resetting of Xist expression, such that the gene appears inactive or is maintained at very low levels of expression [3], [13], [27], [29], [31]. As these cells exit from the ‘naïve’ pluripotent state, one of the two Xist alleles is upregulated to produce Xist transcripts that spread in cis along the X chromosome from which they are expressed, thereby forming a ‘coat’ that remains stable throughout the interphase portion of the cell cycle [22], [32]. How the upregulation of Xist is ensured on only one of the two X chromosomes, chosen at random, is a matter of intense debate. Most models posit that Xist activation is promoted by the release of multiple inhibitory pathways acting in the naïve pluripotent state as well as by an increase in Xist activators due to induction of differentiation, the details of which have been reviewed elsewhere (see Goodrich et al. in this issue, page 3) [33], [34]. Through mechanisms that are still largely unclear, Xist RNA initiates the eviction of RNA polymerase II from within the Xist RNA coated territory [35], [36]. The onset of X-linked gene silencing is accompanied by a temporally sequential deposition of repressive H3K27me3 and H3K9me2 marks, enrichment of the repressive macroH2A histone variant, and DNA methylation on the forming Xi [6], [7], [8], [9], [10], [11], [12], [37], [38]. Together these and other epigenetic modifications enable the formation and maintenance of a transcriptionally silent, heterochromatic chromosome, known as the inactive X (Xi) [39].

The Xist locus remains active specifically on the Xi throughout the somatic lifetime of a cell. Investigations into the requirement of Xist post initiation of XCI defined two distinct phases of the process during differentiation. The initiation phase captures the onset of XCI and is characterized by Xist RNA-dependent X-linked gene silencing. During this window of time, which generally correlates with the first 72 h of differentiation, depletion of Xist RNA allows for reactivation of silenced X-linked genes and reverses the XCI program. The maintenance phase follows and is characterized by X-linked gene silencing that is largely Xist-independent. The loss of Xist RNA during this period correlates with the depletion of some enriched heterochromatic marks on the Xi, such as macroH2A and H3K27me3, but extensive reactivation of silenced X-linked genes is not observed [11], [39], [40], [41], [42], [43].

The Xist RNA ‘coat’ that forms across the Xi has been largely defined by signals observed upon the application RNA Fluorescence In Situ Hybridization (FISH) [44]. Some of the earliest insight into how Xist RNA associates with the X chromosome was obtained by treating fixed somatic cells with RNase H, an enzyme that degrades RNA chains involved in RNA:DNA heteroduplexes [32]. RNase H does not act on double- or single-stranded RNAs. When the application of RNase H preceded the hybridization of fluorescent DNA FISH probes to Xist RNA, the Xist RNA coat over the Xi territory was observed to be intact. This finding suggested that the Xist transcripts were not engaged in extensive, if any, RNA:DNA heteroduplexes with genomic DNA. However, to ensure that the Xist RNA can indeed bind DNA, the authors also applied RNase H treatment after the DNA probe hybridization step. This reversed order of experimental steps resulted in loss of the FISH signal due to degradation of Xist RNA by RNase H and confirmed that, although the Xist transcripts are capable of hybridizing to DNA, they do not directly base pair with DNA in the cell. An extension of this conclusion is that Xist RNA instead spreads via associations with chromatin and/or the nuclear matrix. In support of this, it has been shown that Xist transcripts undergo extensive cis-medicated chromosome spread when expressed from cDNA transgenes integrated at ectopic sites on the X or autosome, or in the context of X:autosomal translocations [41], [45], [46], [47], [48]. This property of Xist was recently exploited in a human pluripotent cell model of Down’s Syndrome (trisomy of chromosome 21) [49]. XIST expressed from a transgene integrated on one of three chromosome 21’s in these cells effectively implemented heritable autosomal inactivation in cis. This work not only confirmed that Xist/XIST RNA spreads independently of X-linked genomic sequences, but also demonstrated the competence of chromosome 21 to be silenced, while also uncovering a potential therapeutic approach towards the treatment of trisomic disorders that catapulted Xist RNA biology into the realm of translational medicine.

In the first half of this review, we will focus on the data that suggests that 3D chromatin conformation determines the spread of Xist RNA across the X chromosome. We will also cover the identification of trans-activating factors and the putative roles(s) they play in mediating this process. It has long been appreciated that, once formed, the Xi is structurally distinct from that of the Xa [50]. In the second part of this review, we will address differences in the 3D structure of the Xi in relation to the Xa from cytological, biochemical and genomic perspectives. Our aim is to highlight how the Xist RNA, by exploiting the 3D architecture of the genome to exert its function, also critically impacts genome organization.

Section snippets

Modeling X chromosome inactivation in vitro

The study of XCI in implanting blastocysts has been limited, at least in part, by their availability, difficulty of genetic manipulation, and low cell number for biochemical analyses. Embryonic stem cells (ESC), derived from epiblast cells of the mouse pre-implantation blastocyst, expand indefinitely in culture and provide an essentially limitless supply of starting material [51], [52]. These cells grow clonally, which is convenient for the isolation of targeted genetic mutants. Like the

3D chromosome conformation directs the initial pattern of Xist RNA spreading

As mentioned above, initial experiments probing the nature of the Xist RNA coat implied that it associates with and spreads across chromatin without directly interacting with DNA [32], [41], [49]. How this is achieved has remained a long-standing question in the field. Efforts to biochemically purify the Xist RNA, which is thought to function as a large ribonucleoprotein complex (RNP), have been stymied due to its large size and its tendency to purify with the insoluble components of the

Association of Xist RNA with chromatin is mediated by protein factors

The molecular parameters that regulate the chromatin association of Xist RNA can be assessed via distinct experimental avenues. One approach defines regions of the Xist RNA that mediate spreading by experimentally identifying Xist mutants that display defects in Xist RNA cloud formation or cloud stability. Alternatively, if one assumes that a protein, or set of factors, binds both Xist RNA and/or chromatin to tether the two, then it follows that depletion of the factor(s) would similarly

Multiple domains within Xist RNA confer binding to chromatin

The requirement of the Xist RNA repeats for initiation of XCI was investigated in an ambitious study, in which inducible Xist transgenes, deleted of single or multiple repeats, were integrated upon the X chromosome in male ESCs [57]. As described earlier, induction of wild-type Xist resulted in cell death. In screening mutants for survival, the authors uncovered a role for the A-repeat in mediating silencing of X-linked genes. Remarkably, chromosomal coating and Xist RNA cloud formation was not

Does the 3D structure of the Xi differ from that of the Xa?

In 1949, Barr and Bertram documented the presence of a nucleolar-associated focus, which they termed a ‘nucleolar satellite’. The focus stained strongly with nucleic acid dyes, such as cresyl violet or DAPI, and importantly, was observed in female, but not male, feline neurons [86]. A series of papers published during the following decade reported similar findings in somatic cells from a variety of mammalian tissues, many of which posited that the nucleolar satellite was composed of two female

Shape and volume distinguish the Xi from the Xa

Compacted chromatin is thought to adopt a more densely packaged arrangement of nucleosomes than open chromatin and, therefore, would be expected to fill a smaller volume. This logic has been used to predict a smaller nuclear volume for the Xi over the Xa. DNA FISH probes that paint chromosomes have generally been employed to estimate chromosomal volume and through application of this technique, multiple studies using RNA/DNA FISH to identify the Xist RNA-coated Xi, have reported that the volume

The basic chromatin structure of the Xi is similar that of the Xa

In the nucleus, chromatin is built upon the basic 11 nM ‘bead on a string’ fiber, composed of linear DNA wrapped twice around nucleosomes. It is thought that the 11 nM fiber then folds into a 30 nM fiber which forms the basic structure of chromosomes [100]. The structure, indeed, even the existence of the 30 nM fiber in vivo is highly debated [101]. In vitro however, isolation of chromatin under low salt conditions (5nmM NaCl) yields a 10 nM fiber that transitions into a 30 nM one with increasing

The Xi forms a bipartite structure with 3D contacts that differ from the Xa

The advent of high throughput sequencing coupled with the development of Chromosome Conformation Capture (3C) techniques has made it possible to probe the frequency with which any one DNA sequence contacts all other genomic sites across the genome of a cell [109]. These novel applications provide an independent method for probing genome organization, and have revealed that regions of open, or more transcriptionally active, chromatin self-associate to form what has been termed the ‘A’

Towards a comprehensive understanding of Xist RNA function

A quarter century of research on Xist RNA biology has detailed how upregulated expression of Xist during embryogenesis results in coating of the presumptive Xi by the Xist RNA and heterochromatinization of the chromosome, to ultimately lock in a heritable, transcriptionally silenced state. To achieve this remarkable feat, the Xist RNA exploits and also shapes chromosome conformation by directly preventing contacts between sites on the X or through indirect alteration of the linear chromatin

Acknowledgements

We are grateful to Dr. Yolanda Markaki for providing 3D-SIM images. Our work is supported through funds from the UCLA Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research, the UCLA David Geffen School of Medicine, CIRM, and NIH P01 GM099134 (to KP), and NIH F32 NRSA and Helen Hay Whitney postdoctoral fellowships (to AP-J). We sincerely apologize to colleagues whose work we were unable to discuss due to space limitations.

References (115)

  • A. Wutz et al.

    A shift from reversible to irreversible X inactivation is triggered during ES cell differentiation

    Mol. Cell.

    (2000)
  • E. Yildirim et al.

    Xist RNA is a potent suppressor of hematologic cancer in mice

    Cell

    (2013)
  • L.-F. Zhang et al.

    Perinucleolar targeting of the inactive X during S phase: evidence for a role in the maintenance of silencing

    Cell

    (2007)
  • C.J. Brown 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)
  • J.T. Lee et al.

    A 450 kb transgene displays properties of the mammalian X-inactivation center

    Cell

    (1996)
  • S. Ohno et al.

    Formation of the sex chromatin by a single X-chromosome in liver cells of Rattus norvegicus

    Exp. Cell Res.

    (1959)
  • B. Panning et al.

    X chromosome inactivation is mediated by Xist RNA stabilization

    Cell

    (1997)
  • C. Naughton et al.

    Analysis of active and inactive X chromosome architecture reveals the independent organization of 30 nm and large-scale chromatin structures

    Mol. Cell.

    (2010)
  • A. Monfort et al.

    Identification of spen as a crucial factor for Xist function through forward genetic screening in haploid embryonic stem cells

    Cell Rep.

    (2015)
  • C. Chu et al.

    Genomic maps of long noncoding RNA occupancy reveal principles of RNA-chromatin interactions

    Mol. Cell.

    (2011)
  • C. Chu et al.

    Systematic discovery of Xist RNA binding proteins

    Cell

    (2015)
  • B. Moindrot et al.

    A pooled shRNA screen identifies rbm15, spen, and wtap as factors required for Xist RNA-mediated silencing

    Cell Rep.

    (2015)
  • Y. Hasegawa et al.

    The matrix protein hnRNP U is required for chromosomal localization of Xist RNA

    Dev. Cell.

    (2010)
  • Y. Hasegawa et al.

    The matrix protein hnRNP U is required for chromosomal localization of Xist RNA

    Dev. Cell.

    (2010)
  • L.L. Hall et al.

    Stable C0T-1 repeat RNA is abundant and is associated with euchromatic interphase chromosomes

    Cell

    (2014)
  • Y. Jeon et al.

    YY1 tethers Xist RNA to the inactive X nucleation center

    Cell

    (2011)
  • M. Petersen et al.

    LNA: a versatile tool for therapeutics and genomics

    Trends Biotechnol.

    (2003)
  • S. Ohno et al.

    Somatic association of the positively heteropycnotic x-chromosomes in female mice (Mus musculus)

    Exp. Cell Res.

    (1958)
  • S. Ohno et al.

    On the X-chromosomes of mouse mammary carcinoma cells

    Exp. Cell Res.

    (1959)
  • S. Ohno et al.

    On the end-to-end association of the X and Y chromosomes of Mus musculus

    Exp. Cell Res.

    (1959)
  • P. Zhu et al.

    Structural insights of nucleosome and the 30-nm chromatin fiber

    Curr. Opin. Struct. Biol.

    (2016)
  • M. Monk et al.

    Sequential X chromosome inactivation coupled with cellular differentiation in early mouse embryos

    Nature

    (1979)
  • N. Takagi et al.

    Preferential inactivation of the paternally derived X chromosome in the extraembryonic membranes of the mouse

    Nature

    (1975)
  • L.H. Williams et al.

    Transcription precedes loss of Xist coating and depletion of H3K27me3 during X-chromosome reprogramming in the mouse inner cell mass

    Dev. Camb. Engl.

    (2011)
  • E. Beutler et al.

    The normal human female as a mosaic of X-chromosome activity: studies using the gene for C-6-PD-deficiency as a marker

    Proc. Natl. Acad. Sci. U. S. A.

    (1962)
  • C. Costanzi et al.

    Histone macroH2A1 is concentrated in the inactive X chromosome of female mammals

    Nature

    (1998)
  • J.E. Mermoud et al.

    Histone macroH2A1.2 relocates to the inactive X chromosome after initiation and propagation of X-inactivation

    J. Cell Biol.

    (1999)
  • K. Plath et al.

    Role of histone H3 lysine 27 methylation in X inactivation

    Science

    (2003)
  • I. Okamoto et al.

    Eutherian mammals use diverse strategies to initiate X-chromosome inactivation during development

    Nature

    (2011)
  • O. Frota-Pessoa et al.

    Christmas factor: dosage compensation and the production of blood coagulation factor IX

    Science

    (1963)
  • R.L. Gardner et al.

    X chromosome inactivation studied by injection of a single cell into the mouse blastocyst

    Nature

    (1971)
  • M. Monk et al.

    Dosage compensation for an X-linked gene in pre-implantation mouse embryos

    Nature

    (1977)
  • Y. Marahrens et al.

    Xist-deficient mice are defective in dosage compensation but not spermatogenesis

    Genes Dev.

    (1997)
  • J.W. Mugford et al.

    Failure of extra-embryonic progenitor maintenance in the absence of dosage compensation

    Dev. Camb. Engl.

    (2012)
  • N. Takagi

    Primary and secondary nonrandom X chromosome inactivation in early female mouse embryos carrying Searle’s translocation T(X; 16)16H

    Chromosoma

    (1980)
  • P.P.L. Tam

    Mouse embryonic chimeras: tools for studying mammalian development

    Development

    (2003)
  • G.D. Penny et al.

    Requirement for Xist in X chromosome inactivation

    Nature

    (1996)
  • G. Borsani et al.

    Characterization of a murine gene expressed from the inactive X chromosome

    Nature

    (1991)
  • N. Brockdorff et al.

    Conservation of position and exclusive expression of mouse Xist from the inactive X chromosome

    Nature

    (1991)
  • C.J. Brown et al.

    A gene from the region of the human X inactivation centre is expressed exclusively from the inactive X chromosome

    Nature

    (1991)
  • Cited by (14)

    • Attrition of X Chromosome Inactivation in Aged Hematopoietic Stem Cells

      2021, Stem Cell Reports
      Citation Excerpt :

      XCI is a process that involves the transcriptional silencing of one Chr X in cells of female mammals to balance gene expression between XX females and XY males by embryonic day 6.5 (E6.5) (Brockdorff et al., 2020; Fang et al., 2019; Galupa and Heard, 2018; Lebon et al., 1995; Loda and Heard, 2019). The silencing of the Xi is regulated by the long non-coding RNA (lncRNA) Xist (Brown et al., 1992; Marahrens et al., 1998; Penny et al., 1996; Sahakyan et al., 2018), which coats the future Xi and then recruits repressive complexes (Chaumeil et al., 2006; da Rocha and Heard, 2017; Engreitz et al., 2013; McHugh et al., 2015; Minajigi et al., 2015; Pandya-Jones and Plath, 2016; Pinheiro and Heard, 2017). In addition to the essential role of Xist in embryonic development, Xist-mediated gene silencing in hematopoietic cells is critical for postnatal survival in mice (Savarese et al., 2006; Yildirim et al., 2013).

    • The Role of Xist in X-Chromosome Dosage Compensation

      2018, Trends in Cell Biology
      Citation Excerpt :

      In the XX hermaphrodite C. elegans, the 3D conformation of the X chromosomes is remodeled to reduce chromosome-wide gene expression by half to achieve gene-expression balance between XX hermaphrodites and XO males [16]. 3D chromosome conformation also differs between the active and inactive X chromosomes in mammals [17], suggesting that chromosome confirmation remodeling might also be at play in human XCD. However, unlike the mammalian inactive X chromosome (Xi), lncRNAs have not been reported to regulate X-chromosome dose in the C. elegans.

    • SMCHD1 Merges Chromosome Compartments and Assists Formation of Super-Structures on the Inactive X

      2018, Cell
      Citation Excerpt :

      The mammalian inactive X chromosome (Xi) is unique in its reconfiguration into an unusual chromosome conformation. Whereas the active X (Xa) is similar to autosomes in being organized into A/B compartments and TADs, the ∼60 A/B compartments and ∼110 TADs originally on the Xa are thought to be eliminated globally during XCI (Minajigi et al., 2015; Giorgetti et al., 2016; Pandya-Jones and Plath, 2016). The Xi is instead folded into two “megadomains” of 73 and 93 Mb in mice, separated by a boundary near the macrosatellite repeat, Dxz4 (Deng et al., 2015; Giorgetti et al., 2016; Minajigi et al., 2015; Rao et al., 2014).

    • Choosing the Active X: The Human Version of X Inactivation

      2017, Trends in Genetics
      Citation Excerpt :

      Instead, our knowledge is primarily based on studies of experimentally manipulated ES cells, which do not replicate the timing of the in vivo process [17]. Male ES cells have been manipulated to have an inducible Xist locus on their sole X. Not surprising, the endpoint of such studies is rapid death of the cells being studied when XI is induced [33]. Most importantly, mammals differ in the details of the silencing process because of evolutionary tinkering with the genes that lie within the Xic, and with variation in the species-specific staging of embryogenesis [27,34].

    • X chromosome inactivation: The importance of being inactive

      2016, Seminars in Cell and Developmental Biology
    View all citing articles on Scopus
    View full text