Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Jun 20:7:11.
doi: 10.1186/1756-8935-7-11. eCollection 2014.

Lineage-specific regulation of imprinted X inactivation in extraembryonic endoderm stem cells

Affiliations

Lineage-specific regulation of imprinted X inactivation in extraembryonic endoderm stem cells

Sarra Merzouk et al. Epigenetics Chromatin. .

Abstract

Background: Silencing of the paternal X chromosome (Xp), a phenomenon known as imprinted X-chromosome inactivation (I-XCI), characterises, amongst mouse extraembryonic lineages, the primitive endoderm and the extraembryonic endoderm (XEN) stem cells derived from it.

Results: Using a combination of chromatin immunoprecipitation characterisation of histone modifications and single-cell expression studies, we show that whilst the Xp in XEN cells, like the inactive X chromosome in other cell types, globally accumulates the repressive histone mark H3K27me3, a large number of Xp genes locally lack H3K27me3 and escape from I-XCI. In most cases this escape is specific to the XEN cell lineage. Importantly, the degree of escape and the genes concerned remain unchanged upon XEN conversion into visceral endoderm, suggesting stringent control of I-XCI in XEN derivatives. Surprisingly, chemical inhibition of EZH2, a member of the Polycomb repressive complex 2 (PRC2), and subsequent loss of H3K27me3 on the Xp, do not drastically perturb the pattern of silencing of Xp genes in XEN cells.

Conclusions: The observations that we report here suggest that the maintenance of gene expression profiles of the inactive Xp in XEN cells involves a tissue-specific mechanism that acts partly independently of PRC2 catalytic activity.

Keywords: Epigenetic; Single-cell analyses; Stem cells; X-inactivation.

PubMed Disclaimer

Figures

Figure 1
Figure 1
H3K27me3 accumulates on the inactive X chromosome in female XEN cells. (A) Heatmap of single-cell, steady-state RNA levels for the ten most stringent epigenetic discriminants for trophoblast stem (TS) cells and extraendoderm stem (XEN) cells (P < 5 × 10−4 by F-test). The complete data set is available in Additional file 2B. R, Regulator preferentially associated with transcription repression; D, Regulator associated with a dual function; it can be involved in transcription repression or activation. n = 72 female XEN cells (GHP7/9). (B) Representative image of immunofluorescence followed by fluorescence in situ hybridisation on RNA for H3K27me3 (green) and Xist (red) on female XEN cells (GHP7/3 cell line). Similar results were obtained with the GHP7/9 cell line (data not shown). Quantification of fluorescence intensities for Xist and H3K27me3 across the inactive X domain show that the two domains do not strictly overlap. Maximal projections after deconvolution are shown. Scale bar = 5 μm. (C) Pie charts showing the percentage of nuclei exhibiting an accumulation of Xist RNA only (red), coaccumulation of Xist RNA and H3K27me3 (yellow) and accumulation of H3K27me3 only (green) in female XEN cells (GHP7/3 cell line). (D) Boxplots showing the distribution of volumes occupied by Xist RNA and H3K27me3 on the inactive X in XEN cells (GHP7/3 cell line). The two distributions are significantly different. P < 0.05 by Kolmogorov–Smirnov test (n > 50). Vertical bars below and above the box-plots show the minimal and maximal values in the cell population respectively.
Figure 2
Figure 2
Distribution of H3K27me3 on the X chromosome in XEN cells. (A) Boxplots of the distribution of H3K27me3 along X-linked intergenic regions and X-linked gene bodies of expressed X-linked genes (Exp.) or along genes that are not significantly expressed (Not exp.) in male and female extraendoderm stem (XEN) cells (GHP7/7 and GHP7/9 cell lines). Expression data were extracted from the Gene Expression Omnibus database [GSE:15519] [29]. n = 1,054 intergenic regions and n = 642 X-linked genes. *P < 0.05 by Kolmogorov–Smirnov test. (B) Representative examples of H3K27me3 distribution along expressed genes enriched in H3K27me3 in female compared to male XEN cells ([K27-high]) and along expressed genes showing low levels of H3K27me3 in female and in male XEN cells ([K27-low]) (GHP7/7 and GHP7/9 cell lines). mm9 University of California Santa Clara (UCSC) Genome Browser screenshots are shown. IP, immunoprecipitated DNA; IN, input DNA. (C) Scatterplot of H3K27me3 percentages along the body of expressed X-linked genes in female (y-axis) relative to male (x-axis) XEN cells (GHP7/7 and GHP7/9 cell lines (see Additional file 4A for results in the GHP7/3 cell line and Additional file 2C for the complete data set). Each dot represents a single gene and its respective percentage of H3K27me3 in male and female XEN cells. k-means clustering was applied, which led to the identification of three classes of genes. K27-high genes (red dots) are concentrated in the upper left quadrant consistently, with them being depleted in H3K27me3 in male cells and enriched in H3K27me3 in female cells. K27-low genes (blue dots) are significantly depleted in H3K27me3 in both male and female cells. In agreement with H3K27me3 marking preferentially the silent state, very few expressed genes are enriched in H3K27me3 in both male and female cells ([0] genes; indicated by maroon dots). These [0] genes may have resulted from the fact that gene expression data [29] and our chromatin immunoprecipitation followed by chip hybridisation experiment data were obtained from different XEN cell lines. The pie chart underneath the scatterplot represents the percentage of expressed genes in each H3K27me3 class.
Figure 3
Figure 3
RT-qPCR analysis of X-linked gene expression in single female XEN cells. (A) Cumulative histograms showing allelic gene expression levels of Xist, Atrx, Jpx, Ftx and Pbdc1 in the 72 individual extraendoderm stem (XEN) cells (GHP7/9 cell line). AU, Arbitrary unit. Logarithmic scale is used for Xist. (B) Cumulative histograms of the percentages of female XEN cells showing expression from the inactive paternal X (Xp) (dark blue), from the active maternal X (Xm) (red) or from both X chromosomes (light blue) for each indicated gene. X-linked genes are grouped according to their H3K27me3 enrichment in female XEN cells. The asterisks mark significant differences found upon analysis of the expression profile of a given gene showing low levels of H3K27me3 [K27-low] compared to any of the genes enriched in H3K27me3 [K27-high] analysed (P < 0.05 by Χ2 test). On the left, the diagram depicts the position of analysed genes along the X chromosome. The Pgk1a polymorphic region is shown in grey.
Figure 4
Figure 4
Fluorescence in situ hybridisation on RNA analysis of X-linked gene expression in XEN compared to TS female cells. (A) Representative images of fluorescence in situ hybridisation on RNA (RNA-FISH) analysis in extraendoderm stem (XEN) female cells (GHP7/9 cell line). The inactive X chromosome is identified through Xist RNA accumulation (green). Primary transcription at the indicated X-linked gene is codetected in red. Scale bar = 5 μm. (B) Cumulative histograms of the percentages of nuclei with the depicted expression pattern are shown. X-linked genes are grouped according to their H3K27me3 class. [K27-high], Enriched in H3K27me3 in female compared to male XEN cells; [K27-low], Low levels of H3K27me3 in female and in male XEN cells. On the left of the histogram, the diagram shows the localisation of RNA-FISH probes along the X chromosome. n > 200. (C) The same types of data shown in (A) are given for female trophoblast stem (TS) cells (F3 cell line). (D) The same types of data shown in (B) are given for female TS cells. *P < 0.05 by Χ2 test for significant differences with expression profiles in XEN cells (n > 200).
Figure 5
Figure 5
X-linked gene expression in female XEN cells treated with the EZH2 inhibitor GSK126. (A) Representative images produced by immunostaining followed by fluorescence in situ hybridisation on RNA for H3K27me3 (green) and Xist (red) on female extraendoderm stem (XEN) cells (GHP7/9 cell line) treated or not with 2 μM GSK126 for 5 days. Arrowheads indicate the nuclear position of the inactive X chromosome coated with Xist RNA. The percentage of visible H3K27me3 accumulation on the Xist-coated X chromosome in each condition is indicated. n > 50. Scale bar = 5 μm. (B) Cumulative histograms showing allelic gene expression levels of Xist, Atp7a, Chic1, Hmgn5, Jpx, Ftx and Pbdc1 in female XEN cells (GHP7/9 cell line) treated (+) or not (−) with 2 μM GSK126 for 5 days. Mean values and standard deviations of two independent experiments are shown for each gene. Values have been standardised by the ubiquitously expressed Rplp0 gene. On the left, the diagram shows the position of the genes analysed by reverse transcription followed by quantitative polymerase chain reaction along the X chromosome. No significant difference in the relative levels of paternal X (Xp) chromosome expression could be detected between treated and untreated cells (P > 0.05 by Χ2 test). AU, Arbitrary unit; [K27-high], Enriched in H3K27me3 in female compared to male XEN cells; [K27-low], Low levels of H3K27me3 in female and in male XEN cells; Xm, Maternal X chromosome.
Figure 6
Figure 6
X-linked gene expression upon conversion of XEN cells into visceral endoderm. (A) Representative photographs illustrating the epithelium-like cell morphology observed after 5 days of bone morphogenetic protein 4 (BMP4) induction of extraendoderm stem (XEN) female cells (GHP7/9 cell line) are shown. (B) Heatmap of single-cell, steady-state RNA levels for the 18 most stringent discriminants of cells converted into visceral endoderm and undifferentiated XEN cells is shown (P < 10−3 by F-test). Hierarchical clustering has been applied. The complete data set is available in Additional file 2E. n = 72 female XEN cells (red) and n = 82 XEN cells treated with BMP4 (25 μM) for 5 days (yellow). The expected tissue specificity of each gene is indicated: EPI, Epiblast; ICM, Inner cell mass; PE, Parietal endoderm; Pl. Lab, Placental labyrinth; PrE, Primitive endoderm; TE, Trophectoderm; TGC, Trophoblast giant cell; VE, Visceral endoderm. The colour scale is the same as the one shown in Figure 1A. (C) Three-dimensional projections of principal components (PCs) 1, 2 and 3 of single-cell expression profiles of genes shown in (B). (D) Two-colour fluorescence in situ hybridisation on RNA analysis for the indicated genes in XEN cells treated with BMP4 for 5 days. Arrowheads point to transcription signals associated with the Xist-coated inactive X chromosome. Scale bar = 5 μm. (E) Cumulative histogram showing the percentage of nuclei with the depicted expression pattern in XEN cells treated with BMP4 for 5 days (D5). Results obtained in untreated XEN cells (D0) are also shown to facilitate the comparison. No significant differences between X-linked gene expression profiles in BMP4 treated compared to untreated XEN cells could be detected (Χ2 test). n > 100. [K27-high], Enriched in H3K27me3 in female compared to male XEN cells; [K27-low], Low levels of H3K27me3 in female and in male XEN cells; Not exp., genes not significantly expressed; Xa, Active X; Xi, Inactive X.

References

    1. Okamoto I, Otte AP, Allis CD, Reinberg D, Heard E. Epigenetic dynamics of imprinted X inactivation during early mouse development. Science. 2004;303:644–649. - PubMed
    1. Okamoto I, Arnaud D, Le Baccon P, Otte AP, Disteche CM, Avner P, Heard E. Evidence for de novo imprinted X-chromosome inactivation independent of meiotic inactivation in mice. Nature. 2005;438:369–373. - PubMed
    1. Patrat C, Okamoto I, Diabangouaya P, Vialon V, Le Baccon P, Chow J, Heard E. Dynamic changes in paternal X-chromosome activity during imprinted X-chromosome inactivation in mice. Proc Natl Acad Sci U S A. 2009;106:5198–5203. - PMC - PubMed
    1. Kay GF, Barton SC, Surani MA, Rastan S. Imprinting and X chromosome counting mechanisms determine Xist expression in early mouse development. Cell. 1994;77:639–650. - PubMed
    1. Nesterova TB, Barton SC, Surani MA, Brockdorff N. Loss of Xist imprinting in diploid parthenogenetic preimplantation embryos. Dev Biol. 2001;235:343–350. - PubMed