The lesser known story of X-chromosome reactivation

Abstract

X-chromosome inactivation (XCI) is an important mechanism employed by mammalian XX female cells to level X-linked gene expression with that of male XY cells. XCI occurs early in development as the pluripotent cells of the inner cell mass (ICM) in blastocysts successively differentiate into cells of all three germ layers. X-chromosome reactivation (XCR), the reversal of XCI, is critical for germ cell formation as a mechanism to diversify the X-chromosome gene pool. Here we review the characterization of XCR, and further explore its natural occurrence during development and the in vitro models of cellular reprogramming. We also review the key regulators involved in XCI for their role in suppressing the active histone marks and the genes in the active chromosome for their inhibition of X inactivation signals.

X-chromosome inactivation (XCI) is an essential process occurring in female XX cells as a dosage compensation measure during development.¹ It ensures balanced X-chromosome-encoded proteins in male and female cells, and occurs randomly during early development, thus accounting for the mosaicism observed in female somatic cells. Once the cell has inactivated one of the X chromosomes, the pattern is maintained throughout the subsequent series of cell divisions. In mice, the paternal inactive X chromosome (Xi) is maintained throughout the early cleavage until the blastocyst stage, where cells of the inner cell mass (ICM) reactivate the inactive X chromosome.² At subsequent phases of early development, humans and mice share the pattern of XCI. Epiblast cells randomly inactivate one X chromosome, while the primordial germ cells (PGCs) reactivate the Xi during their migration to the genital ridges.^3–6 Interestingly, murine extra-embryonic trophoblast cells show non-random inactivation of the paternal X chromosome maintained in trophectoderm.^6,7 This pattern is, however, not conserved, as human trophectoderm cells randomly inactivate the paternal or maternal X chromosome. In addition to the PGCs and early developing embryo, cells cultured under defined conditions or undergoing reprogramming show X-chromosome reactivation (XCR).⁸ XCI has been extensively studied, while XCR is not well-understood, mainly due to the lack of easily accessible models. Here, we will review the developmental process of XCR and molecular mechanism involved in XCI and XCR.

X-Chromosome Reactivation During Development and in Pluripotent Stem Cells

For many years, XCI during female development was the focus of studying the X-chromosome status, and the process was seen as unidirectional toward later developmental stages and cell maturation. When the question of whether the two X chromosomes are both active before one undergoes XCI, or whether they are both inactive before one switches on was first raised,⁹ the concept of X reactivation was starting to emerge. In that study, the activity of X-linked enzymes in female pluripotent embryonal carcinoma cells (ECCs) before and after differentiation was used as an indicator of X-chromosome state. This method was previously employed to determine the X-chromosome state from the XO/XX ratio of X-linked gene product activity in early stage embryos.¹⁰ XX cells displayed twice as much activity of glucose 6 phosphate dehydrogenase (G6PD) and hypoxanthine-guanine phosphoribosyl transferase (HPRT) as XO cells but no difference in autosomal enzymes. During differentiation of these XX EC cells, the enzymatic activity was reduced by a factor of two, an indication of XCI.

Kratzer and Chapman used electrophoretic profiles of the G6PD enzyme dimer to determine the state of the X chromosome during germ cell development in the mouse species Mus caroli.⁴ When the germ cells enter the meiotic cycle at day 11, they reactivate Xi, as shown by the appearance of the G6PD heterodimer. This also explained the observation that mature oocytes have two active X chromosomes (XaXa).¹¹

In addition to in vitro differentiation assays and developmental studies, cell hybrid experiments were emerging as a useful model to elucidate the state of the X chromosome. Interspecific somatic hybrids of mouse and female human cells, both with mutations allowing growth of hybrids in selective media, showed some local derepression at the HPRT but not the flanking loci of the Xi.¹² Notwithstanding the very low occurrence rate and the previous failures to grow such hybrids, the first instance of reactivation in vivo was reported. Later, staining hybrids of mouse XO ECCs with somatic cells revealed that all three chromosomes could be replicated in early S phase, an indication of Xa status. Furthermore, the normal X-linked enzyme PGK1A coming exclusively from the somatic Xi was detected after fusion.¹³ After the successful derivation of embryonic stem cells (ESCs) from the ICM of the blastocyst and their subsequent in vitro propagation, both X chromosomes of female cells were observed to be active.^9,14–16 In fact, this characteristic is a measure of the true nature of pluripotency of ESCs. Subsequent fusion experiments of ESCs and somatic cells confirmed reactivation of the somatic Xi, recapitulating the finding with EC and ES cells.^13,17,18 The discovery of the reactivation of Xi paved the way for further research into identifying factors existing in the pluripotency environment that reverse the deposited epigenetic marks as well as their mode of interaction during self-renewal and differentiation.

Another important discovery regarding X reactivation was the reprogramming of PGCs into ES-like cells.¹⁹ Thus far, this reversal was known to occur only in vivo during development and in in vitro fusion experiments. Murine PGCs at 7 d postcoitum were harvested and cultured on feeder cells in the presence of leukemia inhibitory factor (LIF) and bFGF (removed during subsequent culture). After 9 d, colonies resembling ES cells emerged and stained positive for alkaline phosphatase and SSEA-1. Moreover, they were able to differentiate in vitro, form teratomas when injected into mice and give rise to chimeras when injected into blastocysts. This new population of pluripotent cells was named embryonic germ cells (EGCs) and found to reactivate the Xi present in the original PGCs.^19,20

X-Chromosome Reactivation During Reprogramming

A milestone in stem cell research was the reprogramming of mouse²¹ and human²² fibroblast cells to ES-like cells named induced pluripotent stem cells (iPSCs). As this newly discovered cell type was characterized for similarities and differences to true ESCs, the question of the fate of Xi present in the starting population began to be addressed. Mouse iPSCs, like mouse ESCs, were shown to have a distributive character of H3K27me3 silencing histone mark and the Ezh2 component of the poly-comb repressive complex (PcG) responsible for this epigenetic change.⁸ Upon differentiation of ES and iPSCs with retinoic acid, enrichment of both H3K27me3 and Ezh2 was seen in the form of foci in the nuclei. Thus, reprogramming was able to successfully reactivate Xi present in the starting fibroblast population. This event was shown to occur at the final stages of de-differentiation, when iPSCs rely on endogenous rather than exogenous pluripotency gene expression.²³ When iPSCs were generated using an X-GFP reporter located in Xi of fibroblasts, all iPS colonies fluoresced green, a test of reactivation.⁸ Furthermore, upon differentiation of these GFP⁺ colonies, both GFP⁺ and GFP^- populations were recovered. This indicated that inactivation is random and not a consequence of the previous Xi state of the starting fibroblast culture.

XCI in human female ESCs and iPSCs is, however, more complex and, to date, still subject of debate. While it is well-established that miPSCs reactivate the Xi and maintain a state of cytogenetically stable XaXa in good culture conditions, there is no definite answer to XCI in hiPSCs. One reason for the difficulty in establishing a model of X state in hES- and ES-like cells is the primed nature of pluripotency characterizing them.²⁴ Mouse ESCs and iPSCs are thought to represent a naïve state of stable pluripotency. However, the human counterparts are conditioned to differentiate. The flattened morphology and reliance on bFGF/activin signaling is in clear contrast to the dome-shaped colonies and reliance on the LIF/Bmp4/Stat3 pathway that characterize mESCs and miPSCs but very similar to epiblast stem cells (EpiSCs). More differentiated mouse and human post-implantation EpiSCs colonies also appear flat, rely on bFGF/activin pathway and form teratomas when injected into blastocysts.^25,26 However, in female cells, one X chromosome is inactivated. Interestingly, primed hESCs were converted to round-shaped colonies akin to mESCs by forced expression of OCT4 and KLF4, or KLF2 and KLF4, and subsequent culture in serum-free media containing LIF, GSK3 and ERK1/2 inhibitors.²⁷ Furthermore, inducible hiPSCs were generated using OCT4/SOX2/KLF4 transgenes while cultured in serum-free medium containing LIF and the inhibitor compounds. Propagation in these compounds and forskolin (an activator of KLF2/KLF4 signaling) enabled renewal up to 20 passages in the absence of forced transgene expression. These modified culture conditions converted hiPSCs to cells morphologically and epigenetically similar mESCs.

Thus, culture conditions and environment contribute to hESC behavior. Indefinite stem cell maintenance in vitro is theoretically possible but physiologically irrelevant, as these cells do not spend unlimited time in a pluripotent environment. As a consequence, the XaXa pattern characteristic of pluripotent hESCs is only seen in limited lines. In support of this, fluorescence in situ hybridization (FISH) targeting the Xist non-coding RNA responsible for XCI, showed a positive signal in the H9 but not H7 female hESCs, independently of the passage number range tested.²⁸ Loss of the XCI markers is also seen as clonal selection occurs in long-term culture.²⁸ In a different study, genotyping array showed non-random XCI and in H9 and HSF6 and instability in H7 female hESCs under suboptimal culture conditions.²⁹

A high degree of variation between hESCs themselves³⁰ is attributed not only to culture techniques and environment but also to the derivation conditions. Cells of the ICM grow in a hypoxic environment. Most culturing in the labs occurs under atmospheric levels of oxygen. Lengner et al.³¹ set out to test the effect of oxygen levels during derivation and maintenance of several female hESCs. Even though ESCs derived in 5% vs. 20% O₂ levels passed the usual pluripotency assay tests, such as immunostaining for pluripotency markers and teratoma formation, XCI was detected via FISH and SNP analysis in cells derived under hyperoxic conditions. XCI was irreversible when XaXi cells were switched to 5% O₂ levels but occurred readily in XaXa cells switched to 20% O₂ culture environment. XCI was shown to occur randomly through inactivation of either paternal or maternal X and was exacerbated by frequent freeze-thawing cycles and low recovery rates.

Overexpression of Yamanaka's factors (Oct4, Sox2, Klf4, cMyc) or other variations of reprogramming methods reactivate Xi in miPSCs. The case of Xi during reprogramming of human somatic cells is less conclusive, as conflicting reports have been published to date. Tchieu et al. reported female fibroblast reprogramming with four-factor polycistronic lentivirus or retrovirus containing a puromycin selection gene in which all generated iPSC colonies had undergone XCI. Cell source, age of fibroblast donors or iPSC passage number did not affect the outcome of Xi. SNP analysis for allelic expression, X-linked enzyme activity and polycomb repressive complex (PRC2) component enrichment at Xi confirmed gene dosage preservation, even though the Xist transcript was lost in some lines during cell passaging. Cytogenetic confirmation of the presence of both chromosomes ruled out chromosomal loss. In addition, despite the heterogenous starting population, each iPSC clone carried the same Xi and conserved the same pattern during differentiation into several lineages. This study strongly supported the view that reprogramming of human fibroblasts to iPSCs resulted in retention of the same Xi found in the cell of origin, and it appears that four-factor reprogramming is not a strong force in reversing the epigenetic hurdles placed on a mature cell. Another study from Colman et al.³³ supported this finding. Furthermore, they noticed that the Rett fibroblast cultures were skewed toward expressing only one X chromosome, which they called X dominant. Very few cultures with an active X unfavored were detected in highly passaged cells. As a result, iPSCs also displayed the same XCI skewing reflected in the parental fibroblasts. Overexpressing human telomerase reverse transcriptase (hTERT) was found to restore the Xi mosaicism in derived iPSC lines; however, all the colonies displayed XaXi pattern. Reactivation of Xi only occurred when cultured in mouse ES medium supplemented with hLIF, GSK3 and ERK1/2 inhibitors and forskolin, as previously discussed in reference ²⁷, and resulted in expression of both wild-type and mutant MeCP2 gene. iPSCs with XaXa pattern are, however, observed and published by our group.³⁴ Four-factor reprogramming of Rett patient fibroblast as well as an unrelated female fibroblast line Detroit 551 gave rise to both XaXa and XaXi clones. This study did not look at long-term fibroblast changes that could result in skewing toward a particular Xi pattern or whether XaXa/XaXi population ratios differed between wt and Rett fibroblasts. However, our observations during multiple reprogramming events are consistently reproducible. The argument that the XaXi iPSCs are only partially reprogrammed is at odds with the pluripotent character of these cells as tested by all the available pluripotency assays as well as their distinct gene expression profile and teratoma formation potential compared with more differentiated EpiSCs. Xi reactivation during reprogramming is possible, although the mechanism remains unclear. Culture conditions appear to play a role in long-term clonal expansion of cells that could account for the skewing of Xi patterns. Four-factor polycistronic vectors are used to deliver the same copy number of each factor to each cell, aiming to reduce some variability in the highly inefficient reprogramming events. But the viral particle number delivered to each cell differs, and this could give a slight advantage to a particular fibroblast subpopulation. It is also interesting to determine how amenable to reactivation the starting population is and whether testing individual fibroblast clones, rather than mixture populations, could provide invaluable answers to the observed mosaicism. Xi retention and reactivation during de-differentiation have both been reported, but they may not be mutually exclusive.

X reactivation was also shown to occur as EpiSCs reprogram to PGCs and subsequently to EGCs.³⁵ EpiSCs were then tested and shown to give rise to a continuous Blimp1⁺ PGC precursor population, as Blimp1 is the earliest detected PGC marker.^35,36 These progenitor cells then go on to become a Stella⁺ population of PGCs and, under appropriate culture conditions, reprogram to EGCs. Thus, EpiSCs then PGCs, when cultured on feeder cells and LIF/BMP4, give rise to ES-like cells as previously discussed. This final population failed to enrich for H3K27me3 histone mark like the EpiSC and PGC parental cells, indicating reactivation of Xi. Another report on the amenability of EpiSCs to reprogramming presented epiblast ES-like cells derived from the epiblast tissue of mouse embryos.³⁷ For the ES colonies to emerge, the tissue was separated from the PGC and visceral endoderm sections and cultured in medium promoting the LIF/Stat3 signaling pathway. H3K27me3 staining of the nuclei revealed absence of foci and indication of X reactivation. With longer in vitro culturing, these cells downregulated EpiSC markers and upregulated the pluripotency genes Nanog, Oct4 and Sox2. In principle, these two findings report two populations that are most likely the same derivation product as a result of culturing in mES conditions. The first report seems to emphasize the step-by-step nature of the reprogramming process, while the second report continuously cultures the EpiSCs to ESCs without testing for intermediates.

X-Chromosome Reactivation in Breast Cancer Cell Lines

Some breast and ovarian cancer lines, like HCC and MCF7, do not stain positive for Xi,^38,39 prompting questions about their X-chromosome state. Because BRCA1 was shown to co-localize and co-immunoprecipitate with Xist, and BRCA1^-/- cells lack Xist foci, a role of BRCA1 in maintenance of Xi in female somatic cells is suggested.^38,40 Thus, if BRCA1 is absent, reactivation of Xi could be facilitated. Recent reports dispute the finding that BRCA1 interacts with Xist, as they find no co-localization between them.^39,41 Furthermore, MCF7 cell populations were found to consist of mostly Xistnegative cells, and this was suggested to be a result of loss of Xi and duplication of Xa. SNP analysis confirmed homogeneity of X alleles.³⁹ Thus, instability of chromosomes in cancer lines, rather than Xi conversion to Xa and its involvement of BRCA1, seems a likely occurrence as well. However, because various cancer and ovarian cells behave differently, neither explanation could be ruled out. In fact, destabilization of Xi could result in its reactivation or loss (i.e., aneuploidy), thus accounting for the absence of Xist. This is an interesting example of X reactivation as it occurs in more epigenetically stringent cells, rather than more malleable multipotent cells, like PGCs and EpiSCs.

Possible Players in X-Chromosome Reactivation

In order to conduct in vitro experiments on X reactivation, we must take into account the differentiation stage of the cells, their growth environment, and the changing sensitivity to stimuli during progression to maturation. It is also of great importance to look at the major players in XCI and how their disruption could generate fertile ground for its reversal. XCI is a concerted mechanism, whereby transcription and epigenetic changes at the X inactivation center (Xic) enable silencing of one X chromosome.⁴² Understanding XCI is important in considering the changes that can take place during the reverse process. The first steps appear to involve a “pairing” of X chromosomes, and “counting”^43,44 to determine the number of Xa present, and the initiation of silencing as a dosage compensation measure. In fact, tetraploid male and female ES cells initiate XCI and silence 1 and 0–4 Xa, respectively. Thus, X ploidy ratio increases the likelihood of XCI.⁴⁵ ChIP with H3K9me at Xic showed enrichment of this mark,⁴⁶ as the Xist long non-coding RNA coats the Xi in cis from center to flanking loci of Xic.^42,47,48 Xist is transcribed at low levels in male and female ES cells but is upregulated and stabilized upon differentiation.⁴⁹ This Xist cloud is a hallmark of the presence of Xi.

In addition, other epigenetic marks, like hypermethylation and hypoacetylation of Xi gene promoters as well as recruitment of additional repressors, like PcG proteins, contribute to spreading of the inactivation signal, and maintenance of transcription inhibition.^42,50 PRC2 component Ezh2 recruits histone methyltransferases that are responsible for the H3K27me3 his-tone modification, a highly utilized stain of the heterochromatic and focal nature of Xi. The condensation of chromatin due to these marks excludes RNA Polymerase II and, therefore, shuts down transcription in these regions. Another PcG component, Eed, is also shown to be important in maintenance of XCI in extraembryonic cells.⁵¹ Trophoblast cells differ from ES cells in that they maintain XCI. In the aforementioned study, Eed^-/- trophoblasts had undergone XCI; however, upon differentiation, the XCI signal was not maintained, and Xi became reactivated. But despite the strong data in extraembryonic cells, PcG components seem to also play an important role in XCI maintenance in embryonic tissue.

Histone Variant Accumulation Denotes Xi Presence

MacroH2A1 histone variants were shown to accumulate and form macrochromatin bodies (MCB) at the site of Xi.^52–54 MacroH2A knockdown in mES cells did not interfere with XCI before and after differentiation,⁵⁵ and even though a full gene knockout could better resolve their function, it appears they play a supporting but dispensable role in Xi maintenance. In addition to Xist transcript, which enables silencing, the reverse transcript encoding Tsix overlaps and extends just downstream of the Xist region and correlates with Xa.⁵⁶ Tsix expression was shown to be biallelic before XCI, then its presence was observed only on the Xa during XCI, and finally, transcription was suppressed after Xi was established. Xite transcript, the product of an enhancer element located downstream of Xist and Tsix, seems to regulate the distribution of Tsix along the X chromosome that will retain activation.⁵⁷ Because it acts in cis, its deletion increases the probability of the XCI occurring on that chromosome. However, Xist-Tsix-Xite deletion of one X on female mES cells did not abrogate XCI, and the presence of an X activator working in trans was hypothesized.⁴⁵ It is important to note that once XCI has been completed, the presence of the RNA transcripts seems dispensable for maintenance.

Rnf12, an Ubiquitin Ligase Involved in X-Chromosome Inactivation

The hypothesized XCI activator could likely be the newly discovered X-linked gene RLIM/RNF12. This ubiquitin ligase was shown to induce inactivation of both female X and the single male chromosome of mES cells in a dose-dependent manner.⁵⁸ Furthermore, Rnf12 seems essential for XCI, because the majority of Rnf12^-/- female mES did not undergo XCI during differentiation.⁵⁹ Rnf12 was shown to achieve this regulation by binding enrichment at the Xist but not Tsix promoter. Another report supported the finding that Rnf12 is essential for XCI, as the maternally transmitted allele with an Rnf12 deletion caused embryonic lethality.⁶⁰ However, this role appeared to be confined to the in vivo model, where the maternal Rnf12 dose is required for XCI imprinting, since the Rnf12^-/- ES cells still retained the ability to undergo XCI upon differentiation. Moreover, Xist was detected in the knockout cells, suggesting normal Xist-dependent XCI. Thus, high levels of Rnf12 cause spreading of XCI to all X chromosomes, but its absence does not interfere with XCI initiation. This suggests the complex mechanism of XCI regulation and other possible factors involved in this process.

Role of Methylationin X-Chromosome State

Another critical process in the XCI regulation is DNA methylation. Panning et al.⁶¹ reported that DNA methyltransferase (Dnmt)-knockout female mES cells express biallelic Xist in female, while male mESCs upregulate their only Xist transcript. This suggests that methylation of the Xist promoter in the ES cell stage is required for suppression of transcription and, therefore, Xa maintenance. Allele-specific methylation analysis of single human cells revealed that there is approximately twice as much methylation in the Xa compared with Xi, with the majority of methylation marks found in gene bodies.⁶² This provides more insight into a complex regulation of transcription, especially in determining which marks more accurately define an active vs. a silenced state.

X-Chromosome Reactivation and RNAi

More recently, testing began for a hypothesized but not very well-characterized RNAi in XCI. One report shows that ablation of RNase III domain of Dicer interferes with the processing of Xist/Tsix into small RNAs and, consequently, with XCI. Dicer could be involved in downregulating Xist in the Xa,⁶³ but another report showed that, while Xist is not detected in Dicerknockout female mESCs, this is most likely due to aneuploidy caused by X instability rather than an effect of Dicer.⁶⁴ The claim is substantiated by the lack of small RNA remnants of Xist/Tsix duplex and the normal XCI found in hybrid ES cells. More studies on the role of RNAi processing triggered by the complementarity of Xist/Tsix are needed, particularly because this pathway is a logically relevant player in XCI.

Pluripotency Genes and X Reactivation

Because ground state pluripotency precludes XaXa chromosomal pattern in female ES cells, it is no surprise that the majority of pluripotency factors are involved in maintaining this active state. Nanog is an essential pluripotency factor, which, however, can be replaced as a reprogramming factor. But a doxycycline-inducible deletion of Nanog during reprogramming revealed that this gene is essential for pre-iPSCs to achieve ground state, even though its presence is dispensable once pluripotency has been achieved.⁶⁵ Nanog is also enriched at the Rnf12 locus, suggesting a repressive force on the transcription of this gene. Furthermore, ChIP in ES cells revealed that Nanog binds to intron 1 of Xist and, in conjunction with Oct3/4 and Sox2, represses expression of Xist.⁶⁶ Another region that composes Xic, namely, DXPas34, was also shown to bind reprogramming factors cMyc and Klf4 in addition to the pluripotency factor Rex1. Rex1 seems to control expression of X-reactivation element Tsix, which is consequently downregulated.⁶⁷ Finally, Oct4, in addition to being highly enriched at intron 1 of Xist, is also sufficient to repress Rnf12, as conditional Oct4 down-regulation correlates with increased Rnf12 levels.⁶⁸

Conclusion

While a lot of work has been focused on XCI at the stem cell level, not much is being done on differentiated cells. Overexpression of these involved genes during reprogramming or after differentiation could shed light on the most important factors responsible for overcoming epigenetic changes. Furthermore, since hESCs and iPSCs, in contrast to mESCs, are primed for differentiation and display XCI patterns even while pluripotent, it is of importance to fully investigate Xi retention or reactivation in this setting. In vitro reprogramming is not always sufficient to reactivate Xi, and finding why it is the case, and whether genetic manipulation could allow full reversal, is currently research in progress and, to date, still unclear. Investigation of the niches that enable X reactivation in vivo, such as the reactivation of Xi in PGCs during migration to the genital ridge, could be necessary to identify novel factors and signaling pathways involved.

With the therapeutic potential of iPS-derived, patient-specific cells, any skewing or inactivation of the wild-type allele could affect the outcome for patients carrying X-linked diseases, such as Rett syndrome or Duchenne muscular dystrophy. Moreover, finding the right derivation and culture conditions that eliminate or reduce XCI initiation is invaluable in propagating true pluripotent stem cells. Silva et al.⁶⁹ categorized female hESCs into three groups: class I, or the pre-XCI, which inactivate one X after differentiation; class II cells, which retain an Xi and Xi markers and class III cells, which contain an Xi but no longer express H3K27me3 or Xist. Class II were shown to convert to class III upon further passaging, suggesting a role for other factors maintaining XCI once it is completed. Even though this study used a variety of ESCs that were derived and passaged differently, and grouping into classes was made on chromatin mark observation without passage number normalization, it suggests that long-term culturing could employ a range of other important factors. This study also highlighted the drawbacks of exclusively using chromatin staining for XCI presence, as loss of Xist and H3K27me3 histone mark did not correlate with the absence of Xi. Absence of Xist in newly migrated PGCs was also not an indication of full reactivation, as some X-linked genes were biallelically expressed, but some were not.⁷⁰ Even in the instances when the Xist signal is present, biallelic expression of certain X-linked loci is observed.⁵⁴

These findings suggest partial activation or inactivation of certain X regions without direct accompaniment of changes in epigenetic marks. Many markers used to probe the presence or absence of Xi may not be a cause, but rather a consequence, of XCI, thus encouraging further optimization of detection tests. Like XCI, X reactivation is a complex process involving many players, interactions and environmental influences, some explored, and some yet to be identified. How cells are able to control the X state during development, switching one on and off in cycles and preserving the activation status of the other is important in understanding and applicability to disease models. The XCI process is continuously investigated; however, looking at the changes taking place at the active chromosome could be also insightful in developing a working model for retention of activation and inhibition of silencing signals. Evaluation of available data alludes to an X reactivation mechanism that, like reprogramming, is not necessarily the direct reversal of X inactivation.