X-Inactivation Status Varies in Human Embryonic Stem Cell Lines
http://www.100md.com
《干细胞学杂志》
a Robarts Research Institute, Krembil Centre for Stem Cell Biology, London, Ontario, Canada;
b Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts, USA;
c CyThera, Inc., San Diego, California, USA
Key Words. Human embryonic stem cells ? Epigenesis ? X-chromosome inactivation ? Differentiation
Correspondence: Melissa K. Carpenter, Ph.D., CyThera, Inc., 3550 General Atomics Court, San Diego, California 92121, USA. Telephone: 858-455-2736; Fax: 858-455-3962; e-mail: mcarpenter@cytheraco.com
ABSTRACT
Assays for expression of surface markers and transcription factors as well as microarray analyses have indicated that human embryonic stem cell (hESC) lines exhibit similar expression patterns . In the present study, we examined whether various hESC lines are also similar in their epigenetic states and developmental competence. To address this, we evaluated X-chromosome inactivation (XCI), a process that reflects a major embryonic developmental transition. In mammalian females, dosage compensation of the X-chromosome is achieved through XCI, an epigenetic event that is regulated by a single cis-acting X-inactivation center (Xic/XIC in mouse and humans, respectively). XCI is one of the earliest events in mouse development, occurring during preimplantation development. It is developmentally regulated, with initiation of inactivation occurring at the onset of cellular differentiation via upregulation of Xist/XIST expression and coating of the X chromosome selected for inactivation (Xi) .
The selection of the X chromosome to be inactivated can be either random or nonrandom (imprinted); this selection is tissue specific, with some tissues exhibiting random XCI and others imprinted XCI. Specifically, in mice, X inactivation is imprinted in the extraembryonic trophectoderm and primitive endoderm lineages during preimplantation development ; in these tissues, the maternal allele is expressed whereas the paternal allele is silenced. In contrast, cells of the epiblast randomly inactivate either of the X chromosomes. To date, the developmental regulation of XCI in humans is unclear. Studies have demonstrated that XIST is detectable in oocytes and in both male and female preimplantation embryos up until the blastocyst stage of development . In contrast to the mouse, XIST expression in human extraembryonic trophectoderm is not limited to the paternal allele . However, similar to mouse, epiblast derivatives also exhibit random XCI patterns. In both mouse and humans, once an X chromosome is inactivated, the same X is silenced in all descendent cells; thus, females are mosaic for their X inactivation pattern .
Evaluation of the mechanisms involved in the process of XCI has largely been limited to studies in mouse ESCs. The recent availability, however, of hESCs now provides a unique tool to assess this process in early human development. hESCs are derived from human blastocysts, have an apparently unlimited proliferative capacity, differentiate into ectoderm, mesoderm, and endoderm, and may therefore provide a model system for studying early developmental processes. In the present study, we demonstrate the novel finding that individual hESC lines exhibit distinct patterns of X inactivation. Further analysis of XCI may thus be an important mechanism by which to examine epigenetic states and developmental competence in hESC lines, important considerations for use of hESCs as a model of early human development or in cell replacement therapies.
MATERIALS AND METHODS
Individual hESC Lines Exhibit Distinct Patterns of X-Chromosome Inactivation
Evaluation of the mechanisms involved in the process of XCI have largely been limited to studies in mouse ESCs . These studies demonstrate that before X-inactivation, a small spot or "transcription focus" of Xist is expressed from both X chromosomes in undifferentiated female ESCs and also from the single X chromosome in male ESCs. At the onset of cellular differentiation, however, Xist expression is upregulated on one of the two chromosomes in the female, thereby creating a much larger accumulation that coats the X chromosome that is to be inactivated. Xist expression on the remaining active X chromosome (or the single X in the male) is silenced .
To assess the status of XCI in hESCs, we first analyzed XIST expression in undifferentiated male (H1 and BG02) and female (H7, H9, and CyT25) hESC lines, all of which exhibited a normal karyotype as assessed via g-banding (data not shown). In these experiments, H1, H7, and H9 hESCs were maintained in feeder-free conditions, whereas the BG02 and CyT25 lines were maintained on MEFs, as described in Materials and Methods. Each cell line expressed standard markers of undifferentiated hESCs, including SSEA-4, TRA-1-60, CRIPTO, REX-1, OCT3/4, NANOG, STAT3, and UTF-1 (Figs. 1A, 2Ai, 2Aii). Quantitative PCR (Fig. 1B) and flow cytometry analysis (Table 1) similarly demonstrate expression of characteristic stem cell markers in each of the hESC lines used in this study.
Figure 1. (A): RT-PCR demonstrates the expression of several markers and transcription factors characteristic of H1 (P50), H7 (P34), H9 (P51), CyT25 (P60), and BG02 (P31) hESCs. (B): qPCR demonstrates that upon differentiation, expression of OCT3/4 and Nanog decreases in all hESC cultures assessed (H9, CyT25, H7, H1, and BG02). One sample of H9 undifferentiated cells was used as the calibrator for analysis of OCT3/4 and Nanog expression. Each bar represents duplicate samples. (C): Left panel: RT-PCR demonstrates expression of XIST in control cultures of differentiated female IMR fibroblasts and in H9 (P42) and CyT25 (P60) undifferentiated hESCs and EBs differentiated for 14 and 9 days, respectively. Expression of XIST is not observed in H1 (P63), H7 (P45), and BG02 (P31) undifferentiated hESCs and EBs differentiated for 14 (H1 and H7) or 9 (BG02) days. Right panel: XIST expression is maintained in H9 hESCs after extended passage. Four samples of low- and high-passage cultures (as indicated) from three separate subclones are shown, (D): qPCR demonstrates that in comparison with expression of OCT3/4, XIST expression is detected in cultures of undifferentiated H9 and CyT25 hESCs but not in cultures of undifferentiated H1, H7, and BG02 hESCs. Upon differentiation, XIST expression does not increase in H9 and CyT25 EB cultures and remains undetectable in cultures of H1, H7, and BG02 EBs. One sample of H9-undifferentiated cells was used as the calibrator for analysis of OCT3/4 expression and IMR female fibroblasts for analysis of XIST expression. n = 2 for H1, H7, and H9 hESC and EB cultures, SEM 0.1 (OCT3/4) and < 0.2 (XIST); n = 3 for CyT25 and BG02 cultures, SEM = 0.1; n = 1 for IMR fibroblasts. Abbreviations: Dif, differentiated hESCs/EBs; EB, embryoid body; hESC, human embryonic stem cell; qPCR, quantitative polymerase chain reaction; RT-PCR, reverse transcription–polymerase chain reaction; UD, undifferentiated hESCs.
Figure 2. (A): Immunocytochemical analysis of H9 (P70) cultures shows positive staining for (i) TRA-1-60 and (ii) SSEA-4 in a typical undifferentiated hESC colony. (iii): Fluorescence in situ hybridization analysis demonstrates accumulation of XIST RNA (territory) in H9 (P38) hESC cultures, x 60 magnification, but not in (iv) H7 (P40) hESC cultures, x 60 magnification. (v): Expression of XIST in OCT3/4-positive (red) cells in H9 (P74) cultures. (vi): Detection of two XIST signals in undifferentiated XXX H9 (P69) cultures. X-chromosome paint analysis (in green) demonstrates the presence of two X chromosomes in both (vii) H9 (P45) and (viii) H7 (P48) cultures, x 100 magnification. An inactivated X chromosome is indicated by the box in panel vii, with individual channels shown to the right of the panel: Colocalization of XIST transcripts (in red) with an X chromosome (in green) is demonstrated in H9 cells. Cells are counterstained with 4,6-diamidino-2-phenylindole (in blue). (B): Reverse transcription–polymerase chain reaction demonstrates that in control H7 hESC (UD) cultures, expression of XIST transcripts is undetectable. After treatment with 5-aza-2'-deoxycytidine, (1 and 10 μM), however, XIST expression is detected in H7 hESCs at days 3, 14, and 28 after treatment, as in parallel control IMR female fibroblasts. Abbreviation: hESC, human embryonic stem cell.
Table 1. Flow cytometry analysis demonstrates expression of SSEA-4, TRA-1-60, and TRA-1-81 in undifferentiated H1, H7, and H9 human embryonic stem cells
XIST expression is virtually undetectable in undifferentiated male hESC cultures. The female hESC lines, in comparison, exhibit distinct differences in XIST expression; XIST expression is relatively strong in undifferentiated H9 and CyT25 cultures and is similar to levels of expression found in human female fibroblasts (Figs. 1C, 1D). Expression of XIST transcripts, however, is undetectable by PCR in female H7 hESCs (Figs. 1C, 1D). Chromosome paint analysis confirms that two separate X-chromosome territories are apparent in approximately 80% of cultured H7 and H9 hESCs (Figs. 2Avii, 2Aviii; Table 2). Two separate X-chromosomes are not observed in 100% of the interphase cells due to the proximity of the two X-territories and the limitations of two-dimensional scoring; this is not indicative of aneuploidy because karyotype analysis of mitotic chromosomes reveals two X chromosomes in all cells assessed.
Table 2. Quantitation of the number of cells with detectable XIST accumulation on one X chromosome and the number of X chromosomes (territories) found in H9 hESCs, H7 hESCs and TIG-1 female fibroblasts
FISH analysis clearly demonstrates a large XIST RNA accumulation that paints one X chromosome in approximately 90% of H9 hESCs (Fig. 2Aiii, Table 2), notably in both OCT3/4-positive, Tra1-60–positive, and SSEA-4–positive undifferentiated cells with in the colonies and in differentiated cells at the periphery of the hESC colonies (Figs. 2Ai, 2Aii, 2Aiii, 2Av). These numbers are comparable to those scored for normal female fibroblasts (Table 2), indicating that undifferentiated H9 hESC cultures exhibit XCI patterns similar to those observed in normal female somatic cells. This is further demonstrated by the fact that both undifferentiated (blue and arrows in DAPI DNA image, Fig. 3A) as well as differentiated (Fig. 3B) H9 hESC cells exhibit clear heterochromatic Barr bodies by DAPI DNA. In contrast, there is a clear lack of XIST accumulation on either of the chromosomes in H7 hESCs (Figs. 2Aiv, 2Aviii, 3C; Table 2). In some rare undifferentiated H7 cells, there seems to be a very small spot of XIST RNA that is reminiscent of the small unstable transcription observed to occur in mouse ESCs (not shown); however, this could not be rigorously discriminated from background. Regardless, there is no definitive evidence that XIST expression is ever upregulated in this line or accumulates on either X chromosome during differentiation, suggesting that, in contrast to H9 and CyT25 female hESCs, H7 hESCs fail to initiate classic XCI.
Figure 3. (A): As evidenced in the separated color channels to the right of the full color image, a field of undifferentiated H9 (P45) hESCs clearly shows an inactivated X chromosome (Xi), which is painted by XIST RNA (red and XIST RNA image) and exhibits a DAPI dense heterochromatic Barr body (blue and arrows in DAPI DNA image). Hybridization to hnRNA (green) using the Cot-1 probe demonstrates transcriptional silencing by the apparent hole in the signal (arrows in Cot-1 RNA image) over the Barr body. (B): Differentiated H9 hESCs (14-day embryoid bodies) maintain the inactivation of the Xi, which is still painted by XIST RNA (red and XIST RNA image) and exhibits a heterochromatic Barr body (arrow in DAPI DNA image). The apparent hole in the hnRNA signal (arrow in Cot-1 RNA image) over the Barr body suggests continued silencing. (C): The female H7 (P73) hESC line does not exhibit these unmistakable hallmarks of inactivation in either of its two X chromosomes by day 14 of differentiation. Two X chromosome centromeres (red and X-centromere image) are apparent, but neither one is associated with an obvious Barr body (blue and arrows in DAPI DNA image) or Cot-1 hnRNA hole (green and arrows in Cot-1 RNA image). Abbreviations: DAPI, 4,6-diamidino-2-phenylindole; hESC, human embryonic stem cell; hnRNA, heterogeneous nuclear RNA.
Chromosome-Wide Xi Transcriptional Silencing Is Apparent in H9 hESCs but Not in H7 hESCs
The above data suggest that in H9-undifferentiated hESCs, one of the two X chromosomes is inactivated whereas the H7 hESC line never initiates XCI nor inactivates one of its two X chromosomes. To verify this, we evaluated the transcriptional status of the X chromosomes in both H9 and H7 hESCs by hybridization to heterogeneous nuclear RNA (hnRNA) transcription in the nucleus. Briefly, by using a labeled human Cot-1 DNA probe and RNA FISH, we can visualize hnRNA transcription throughout the nucleus; this assay allows direct assessment of global transcription across the entire X chromosome rather than relying on replication timing or expression of just a few individual genes and has become widely used as a method to evaluate chromosome silencing . H9 hESCs clearly show a lack of hnRNA transcription signal, creating a large "black hole" under the XIST RNA accumulation that is coincident with the heterochromatic Barr body and clearly indicative of transcriptional silencing. This is seen in both undifferentiated and differentiated H9 hESCs (Figs. 3A, 3B).
H7 hESCs, in contrast, never accumulate XIST RNA (Figs. 2Aiv, 3C), and as such, we could not use XIST to localize the Xi in these cells. Rather, an X-specific centromere signal (Fig. 3C) delineates both X chromosomes in the H7 line, allowing us to assess whether these reside within a chromosome-size territory lacking in Cot-1/hnRNA hybridization. In addition, neither of the two X chromosomes in H7 hESCs (Fig. 2Aviii) exhibits a large clear hole in the hnRNA signal (Fig. 3C), suggesting a lack of chromosome-wide silencing in this line, even after differentiation. Although technical limitations do not allow us to conclude that all genes on both X chromosomes remain active, results clearly indicate that the H7 hESCs do not show the chromosome-wide transcriptional silencing that is readily apparent in both H9 hESCs or normal control cells. Consistent with this, the DAPI-dense Barr bodies easily identified in the H9 cells are lacking in the H7 cells (Fig. 3C). Together, these findings demonstrate that H7 hESCs lack clear hallmarks of chromosomal silencing in both the undifferentiated state and in cells 14 days after differentiation.
Differences in XCI Do Not Correlate with Differences in Gross hESC Expression Profiles
Despite differences in XCI between the H7 and H9 and CyT25 hESC lines, expression of several molecular markers that are characteristic of hESCs is appropriate and comparable between cell lines. Specifically, we demonstrate via reverse transcription (RT)-PCR that expression of CRIPTO, REX-1, OCT3/4, NANOG, STAT3, and UTF-1 does not vary between undifferentiated H1, H7, H9, CyT25, and BG02 hESC lines (Fig. 1A). Quantitative PCR and flow cytometry analyses further demonstrate the expression of OCT3/4, NANOG, and SSEA-4, Tra-1-60, and Tra-1-81, respectively, in each of these lines (Figs. 1B, 1D; Table 1), indicating that differences in the pattern of XCI do not correlate with changes in expression of characteristic stem cell markers.
Macrochromatin Body Formation in Differentiated hESCs
We next evaluated whether downstream components of the X-inactivation process are expressed appropriately in various hESC lines. In differentiating mouse ESCs, an accumulation of MacroH2A1 and the formation of macrochromatin bodies (MCBs) occur at 5 days of differentiation, subsequent to the upregulation in Xist expression and coating of the X chromosome at 1 day of differentiation . Therefore, we assessed MCBs using immunocytochemical localization of MacroH2A1.
In undifferentiated H9 hESC cultures maintained in feeder-free conditions, MCBs are evident in nuclei of differentiating cells surrounding hESC colonies but are absent in the nuclei of cells within hESC colonies that express SSEA-4 (Figs. 4Ai, 4Aii, 4Aiii). Therefore, the undifferentiated cells in these cultures exhibit XIST coating of the X chromosome but not MCB formation, demonstrating that the H9 hESCs may be in between these two developmental stages. Our observation that the differentiated cells outside of the colonies exhibit MCBs further indicates that the process of XCI is appropriately temporally regulated in cultures of H9 hESCs (see below). In contrast, MCBs are not evident during differentiation in either male H1 (data not shown) or female H7 (Figs. 4Aiv, 4Av, 4Avi) hESC cultures, confirming a lack of XCI in both cell lines.
Figure 4. (A): (i): SSEA-4-positive (in red) H9 (P37) hESCs counterstained with DAPI (in blue). (ii): Costaining of MacroH2A1(in green) with SSEA-4 in H9 hESCs. (iii): H9 hESCs stained with SSEA-4, MacroH2A1, and DAPI. (iv): SSEA-4-positive (in red) H7 (P54) hESCs counterstained with DAPI (in blue). (v): Lack of MacroH2A1 expression (in green) in SSEA-4–positive H7 hESCs. (vi): H7 hESCs stained with SSEA-4, MacroH2A1, and DAPI; x 40 magnification for (i)-(vi). (vii): Costaining of MacroH2A1 with ?-tubulin in 175-day-differentiated H9 neural progenitors; x 40 magnification. Costaining of MacroH2A1 with -smooth muscle actin, x 40 magnification (viii), and -fetoprotein (AFP), x 40 magnification (ix), in 7- to 14-day H9 (P37) EB cultures. Boxes indicate accumulation of MacroH2A1 and the presence of macro-chromatin bodies. Lack of MacroH2A1 staining with (x) ?-tubulin, x 40 magnification, (xi) -smooth muscle actin, x 40 magnification, and (xii) AFP, x 40 magnification in 7- to 14-day H7 (P71-73) EB cultures. (B): Demonstration that the number of H9 (P37, P42, P45) hESCs expressing MacroH2A1 increases upon differentiation whereas the number of cells expressing XIST remains the same. The values in parentheses indicate the number of samples counted for MacroH2A1 and XIST, respectively, as well as the number of individual experiments conducted. Abbreviations: DAPI, 4,6-diamidino-2-phenylindole; EB, embryoid body; hESC, human embryonic stem cell.
Patterns of X-Inactivation Are Heritable to Differentiated Progeny
To further assess whether XCI in female hESCs is coincident with cellular differentiation, EBs were generated and temporal changes in gene expression were analyzed. We demonstrate that an appropriate decrease in expression of OCT3/4 is observed upon in vitro differentiation of all lines tested (Figs. 1B, 1D). Expression of XIST transcripts, however, persists in EB cultures of H9 and CyT25 hESCs (Figs. 1C, 1D) and in H9-derived neural progenitor cultures (113 days of differentiation; data not shown). XIST expression did not increase with differentiation (Figs. 1C, 1D, 4B), confirming that XCI was complete in the undifferentiated cultures.
The percentage of hESCs exhibiting an accumulation of MCBs, in comparison, increases upon differentiation (Fig. 4B). Moreover, MCBs are detectable in ectoderm, endoderm, and mesoderm derivatives of H9 hESCs; colocalization of MacroH2A1 with ?-tubulin is identifiable in 175-day-differentiated H9 neural progenitors (Fig. 4Avii) and with smooth muscle actin (Fig. 4Aviii) and -fetoprotein (Fig. 4Aix) after 7 to 14 days of hESC differentiation, providing further evidence that these female hESCs are differentiating appropriately and that X-inactivation progresses throughout in vitro differentiation. In contrast, there is a lack of expression of XIST transcripts (Figs. 1C, 1D, 2Aiv, 3C) and MacroH2A1 accumulation (Figs. 4Ax, 4Axi, 4Axii) in H7 hESC and EB cultures and in H7-derived neural progenitor cultures (37 days of differentiation; data not shown), providing strong evidence of a lack of classic XCI in this female hESC line.
XCI Occurs Appropriately in an Aneuploid H9 hESC Line
Supernumerary X chromosomes are tolerated in human cells when the extra X chromosomes are inactivated . For instance, XXY or XXXY cells will exhibit XIST coating of one or two X-chromosomes, respectively. We have generated an H9 hESC subclone in which g-banding reveals that 5% of the cells within the culture have an XXX genotype. In these cultures, we observe patches of cells showing 2 XIST signals (Fig. 2Avi); quantitation reveals that 12% ± 7% (mean ± SEM) of the cells in the culture exhibit two XIST signals. These data indicate that although this H9 subclone has an aneuploid karyotype, mechanisms are maintained for the appropriate dosage compensation of the X chromosome.
XCI Is Stable over Time In Vitro
Numerous hESC lines have now been derived, and despite differences in the derivation process and culture conditions, evaluation of standard markers, telomerase activity, pluripotency, and karyotype indicates that hESC lines are stable over extended periods of culture (see Hoffman and Carpenter for review ). In this study, we demonstrate that after more than 80 passages (20 months) in continuous culture, H9 hESCs also retain appropriate expression of XIST transcripts (Fig. 1C, right panel).
Expression of XIST Is Modulated After Treatment with 5-Aza-2'-Deoxycytidine
5-Aza-2'-deoxycytidine is a well-known demethylating agent that is widely used to demonstrate a correlation between loss of methylation in specific regions of a gene and activation of gene activity. Here, we demonstrate via RT-PCR that after treatment with 5-aza-2'-deoxycytidine, XIST expression is detected in H7 hESCs (Fig. 2B, Table 3). In these experiments, H7 hESCs were treated with 1 or 10 μM 5-aza-2'-deoxycytidine for 24 or 48 hours, and expression of XIST and OCT3/4 was assessed 3, 14, and 28 days after treatment. Treatment with 5-aza-2'-deoxycytidine resulted in differentiation as indicated by the decrease in OCT3/4 expression. This was accompanied by the appearance of XIST expression by 3 days after treatment. The expression of XIST persisted for 28 days, the latest time point analyzed. However, the detection of XIST by PCR was not accompanied by Xist coating of an X chromosome, as assessed via FISH analysis (data not shown), indicating a continued lack of XCI in H7 hESCs.
Table 3. Different subclones of H7 human embryonic stem cells at various passage numbers
DISCUSSION
We thank Meg Byron for her excellent technical assistance with molecular cytology experiments. We thank Dmitri Nusinow and Barbara Panning for technical assistance and insightful discussions. This work was supported in part by the Krembil Foundation and an establishment grant from the Canadian Institutes of Health Research awarded to M.K.C.; an NIH RO1 supplemental grant (GM053234-07S1) awarded to J.L.; and a postdoctoral fellowship from the Ontario Research and Development Challenge Fund awarded to L.M.H. The Lawrence laboratory component of this work involved the H9, H7, and H1 hESC cells approved for NIH funding and no other cell lines (CyT25) that are not currently NIH approved.
DISCLOSURES
The authors indicate no potential conflicts of interest.
REFERENCES
Bhattacharya B, Miura T, Brandenberger R et al. Gene expression in human embryonic stem cell lines: unique molecular signature. Blood 2004;103:2956–2964.
Carpenter MK, Rosler ES, Fisk GJ et al. Properties of four human embryonic stem cell lines maintained in a feeder-free culture system. Dev Dyn 2004;229:243–258.
Monk M, Harper MI. Sequential X chromosome inactivation coupled with cellular differentiation in early mouse embryos. Nature 1979;281: 311–313.
Takagi N, Sasaki M. Preferential inactivation of the paternally derived X chromosome in the extraembryonic membranes of the mouse. Nature 1975;256:640–642.
Daniels R, Zuccotti M, Kinis T et al. XIST expression in human oocytes and preimplantation embryos. Am J Hum Genet 1997;61:33–39.
Ray PF, Winston RM, Handyside AH. XIST expression from the maternal X chromosome in human male preimplantation embryos at the blastocyst stage. Hum Mol Genet 1997;6:1323–1327.
Harrison KB. X-chromosome inactivation in the human cytotrophoblast. Cytogenet Cell Genet 1989;52:37–41.
Ropers HH, Wolff G, Hitzeroth HW. Preferential X inactivation in human placenta membranes: is the paternal X inactive in early embryonic development of female mammals? Hum Genet 1978;43:265–273.
Plath K, Mlynarczyk-Evans S, Nusinow DA et al. Xist RNA and the mechanism of X chromosome inactivation. Annu Rev Genet 2002;36: 233–278.
Heard E, Clerc P, Avner P. X-chromosome inactivation in mammals. Annu Rev Genet 1997;31:571–610.
Thomson JA, Itskovitz-Eldor J, Shapiro SS et al. Embryonic stem cell lines derived from human blastocysts. Science 1998;282:1145–1147.
Xu C, Inokuma MS, Denham J et al. Feeder-free growth of undifferentiated human embryonic stem cells. Nat Biotechnol 2001;19:971–974.
Lawrence JB, Singer RH, Marselle LM. Highly localized tracks of specific transcripts within interphase nuclei visualized by in situ hybridization. Cell 1989;57:493–502.
Clemson CM, Chow JC, Brown CJ et al. Stabilization and localization of XIST RNA are controlled by separate mechanisms and are not sufficient for X inactivation. J Cell Biol 1998;142:13–23.
Hall LL, Byron M, Sakai K et al. An ectopic human XIST gene can induce chromosome inactivation in postdifferentiation human HT-1080 cells. Proc Natl Acad Sci U S A 2002;99:8677–8682.
Clemson CM, McNeil JA, Willard HF et al. 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.
Johnson CV, Singer RH, Lawrence JB. Fluorescent detection of nuclear RNA and DNA: implications for genome organization. Methods Cell Biol 1991;35:73–99.
Lee JT, Strauss WM, Dausman JA et al. A 450 kb transgene displays properties of the mammalian X-inactivation center. Cell 1996;86:83–94.
Panning B, Jaenisch R. DNA hypomethylation can activate XIST expression and silence x-linked genes. Genes Dev 1996;10:1991–2002.
Penny GD, Kay GF, Sheardown SA et al. Requirement for XIST in X chromosome inactivation. Nature 1996;379:131–137.
Clerc P, Avner P. Role of the region 3' to XIST exon 6 in the counting process of X-chromosome inactivation. Nat Genet 1998;19:249–253.
Sheardown SA, Duthie SM, Johnston CM et al. Stabilization of XIST RNA mediates initiation of X chromosome inactivation. Cell 1997;91:99–107.
Panning B, Dausman J, Jaenisch R. X chromosome inactivation is mediated by XIST RNA stabilization. Cell 1997;90:907–916.
Okamoto I, Otte AP, Allis CD et al. Epigenetic dynamics of imprinted X inactivation during early mouse development. Science 2004;303: 644–649.
Huynh KD, Lee JT. Inheritance of a pre-inactivated paternal X chromosome in early mouse embryos. Nature 2003;426:857–862.
Mermoud JE, Costanzi C, Pehrson JR et al. Histone MacroH2A1.2 relocates to the inactive X chromosome after initiation and propagation of x-inactivation. J Cell Biol 1999;147:1399–1408.
Rasmussen TP, Mastrangelo MA, Eden A et al. Dynamic relocalization of histone MacroH2A1 from centrosomes to inactive X chromosomes during X inactivation. J Cell Biol 2000;150:1189–1198.
Costanzi C, Pehrson JR. Histone MacroH2A1 is concentrated in the inactive X chromosome of female mammals. Nature 1998;393:599–601.
MacDonald M, Hassold T, Harvey J et al. The origin of 47, XXY and 47, XXX aneuploidy: heterogeneous mechanisms and role of aberrant recombination. Hum Mol Genet 1994;3:1365–1371.
Hoffman L, Carpenter MK. Characterization and culture of human embryonic stem cells. Nat Biotechnol 2005;23:699–708.
Brown CJ. Equality of the sexes: mammalian dosage compensation. Semin Reprod Med 2001;19:125–132.
Sato N, Sanjuan IM, Heke M et al. Molecular signature of human embryonic stem cells and its comparison with the mouse. Dev Biol 2003;260:404–413.
Ginis I, Luo Y, Miura T et al. Differences between human and mouse embryonic stem cells. Dev Biol 2004;269:360–380.
Dhara SK, Benvenisty N. Gene trap as a tool for genome annotation and analysis of X chromosome inactivation in human embryonic stem cells. Nucleic Acids Res 2004;32:3995–4002.(Lisa M. Hoffmana, Lisa Ha)
b Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts, USA;
c CyThera, Inc., San Diego, California, USA
Key Words. Human embryonic stem cells ? Epigenesis ? X-chromosome inactivation ? Differentiation
Correspondence: Melissa K. Carpenter, Ph.D., CyThera, Inc., 3550 General Atomics Court, San Diego, California 92121, USA. Telephone: 858-455-2736; Fax: 858-455-3962; e-mail: mcarpenter@cytheraco.com
ABSTRACT
Assays for expression of surface markers and transcription factors as well as microarray analyses have indicated that human embryonic stem cell (hESC) lines exhibit similar expression patterns . In the present study, we examined whether various hESC lines are also similar in their epigenetic states and developmental competence. To address this, we evaluated X-chromosome inactivation (XCI), a process that reflects a major embryonic developmental transition. In mammalian females, dosage compensation of the X-chromosome is achieved through XCI, an epigenetic event that is regulated by a single cis-acting X-inactivation center (Xic/XIC in mouse and humans, respectively). XCI is one of the earliest events in mouse development, occurring during preimplantation development. It is developmentally regulated, with initiation of inactivation occurring at the onset of cellular differentiation via upregulation of Xist/XIST expression and coating of the X chromosome selected for inactivation (Xi) .
The selection of the X chromosome to be inactivated can be either random or nonrandom (imprinted); this selection is tissue specific, with some tissues exhibiting random XCI and others imprinted XCI. Specifically, in mice, X inactivation is imprinted in the extraembryonic trophectoderm and primitive endoderm lineages during preimplantation development ; in these tissues, the maternal allele is expressed whereas the paternal allele is silenced. In contrast, cells of the epiblast randomly inactivate either of the X chromosomes. To date, the developmental regulation of XCI in humans is unclear. Studies have demonstrated that XIST is detectable in oocytes and in both male and female preimplantation embryos up until the blastocyst stage of development . In contrast to the mouse, XIST expression in human extraembryonic trophectoderm is not limited to the paternal allele . However, similar to mouse, epiblast derivatives also exhibit random XCI patterns. In both mouse and humans, once an X chromosome is inactivated, the same X is silenced in all descendent cells; thus, females are mosaic for their X inactivation pattern .
Evaluation of the mechanisms involved in the process of XCI has largely been limited to studies in mouse ESCs. The recent availability, however, of hESCs now provides a unique tool to assess this process in early human development. hESCs are derived from human blastocysts, have an apparently unlimited proliferative capacity, differentiate into ectoderm, mesoderm, and endoderm, and may therefore provide a model system for studying early developmental processes. In the present study, we demonstrate the novel finding that individual hESC lines exhibit distinct patterns of X inactivation. Further analysis of XCI may thus be an important mechanism by which to examine epigenetic states and developmental competence in hESC lines, important considerations for use of hESCs as a model of early human development or in cell replacement therapies.
MATERIALS AND METHODS
Individual hESC Lines Exhibit Distinct Patterns of X-Chromosome Inactivation
Evaluation of the mechanisms involved in the process of XCI have largely been limited to studies in mouse ESCs . These studies demonstrate that before X-inactivation, a small spot or "transcription focus" of Xist is expressed from both X chromosomes in undifferentiated female ESCs and also from the single X chromosome in male ESCs. At the onset of cellular differentiation, however, Xist expression is upregulated on one of the two chromosomes in the female, thereby creating a much larger accumulation that coats the X chromosome that is to be inactivated. Xist expression on the remaining active X chromosome (or the single X in the male) is silenced .
To assess the status of XCI in hESCs, we first analyzed XIST expression in undifferentiated male (H1 and BG02) and female (H7, H9, and CyT25) hESC lines, all of which exhibited a normal karyotype as assessed via g-banding (data not shown). In these experiments, H1, H7, and H9 hESCs were maintained in feeder-free conditions, whereas the BG02 and CyT25 lines were maintained on MEFs, as described in Materials and Methods. Each cell line expressed standard markers of undifferentiated hESCs, including SSEA-4, TRA-1-60, CRIPTO, REX-1, OCT3/4, NANOG, STAT3, and UTF-1 (Figs. 1A, 2Ai, 2Aii). Quantitative PCR (Fig. 1B) and flow cytometry analysis (Table 1) similarly demonstrate expression of characteristic stem cell markers in each of the hESC lines used in this study.
Figure 1. (A): RT-PCR demonstrates the expression of several markers and transcription factors characteristic of H1 (P50), H7 (P34), H9 (P51), CyT25 (P60), and BG02 (P31) hESCs. (B): qPCR demonstrates that upon differentiation, expression of OCT3/4 and Nanog decreases in all hESC cultures assessed (H9, CyT25, H7, H1, and BG02). One sample of H9 undifferentiated cells was used as the calibrator for analysis of OCT3/4 and Nanog expression. Each bar represents duplicate samples. (C): Left panel: RT-PCR demonstrates expression of XIST in control cultures of differentiated female IMR fibroblasts and in H9 (P42) and CyT25 (P60) undifferentiated hESCs and EBs differentiated for 14 and 9 days, respectively. Expression of XIST is not observed in H1 (P63), H7 (P45), and BG02 (P31) undifferentiated hESCs and EBs differentiated for 14 (H1 and H7) or 9 (BG02) days. Right panel: XIST expression is maintained in H9 hESCs after extended passage. Four samples of low- and high-passage cultures (as indicated) from three separate subclones are shown, (D): qPCR demonstrates that in comparison with expression of OCT3/4, XIST expression is detected in cultures of undifferentiated H9 and CyT25 hESCs but not in cultures of undifferentiated H1, H7, and BG02 hESCs. Upon differentiation, XIST expression does not increase in H9 and CyT25 EB cultures and remains undetectable in cultures of H1, H7, and BG02 EBs. One sample of H9-undifferentiated cells was used as the calibrator for analysis of OCT3/4 expression and IMR female fibroblasts for analysis of XIST expression. n = 2 for H1, H7, and H9 hESC and EB cultures, SEM 0.1 (OCT3/4) and < 0.2 (XIST); n = 3 for CyT25 and BG02 cultures, SEM = 0.1; n = 1 for IMR fibroblasts. Abbreviations: Dif, differentiated hESCs/EBs; EB, embryoid body; hESC, human embryonic stem cell; qPCR, quantitative polymerase chain reaction; RT-PCR, reverse transcription–polymerase chain reaction; UD, undifferentiated hESCs.
Figure 2. (A): Immunocytochemical analysis of H9 (P70) cultures shows positive staining for (i) TRA-1-60 and (ii) SSEA-4 in a typical undifferentiated hESC colony. (iii): Fluorescence in situ hybridization analysis demonstrates accumulation of XIST RNA (territory) in H9 (P38) hESC cultures, x 60 magnification, but not in (iv) H7 (P40) hESC cultures, x 60 magnification. (v): Expression of XIST in OCT3/4-positive (red) cells in H9 (P74) cultures. (vi): Detection of two XIST signals in undifferentiated XXX H9 (P69) cultures. X-chromosome paint analysis (in green) demonstrates the presence of two X chromosomes in both (vii) H9 (P45) and (viii) H7 (P48) cultures, x 100 magnification. An inactivated X chromosome is indicated by the box in panel vii, with individual channels shown to the right of the panel: Colocalization of XIST transcripts (in red) with an X chromosome (in green) is demonstrated in H9 cells. Cells are counterstained with 4,6-diamidino-2-phenylindole (in blue). (B): Reverse transcription–polymerase chain reaction demonstrates that in control H7 hESC (UD) cultures, expression of XIST transcripts is undetectable. After treatment with 5-aza-2'-deoxycytidine, (1 and 10 μM), however, XIST expression is detected in H7 hESCs at days 3, 14, and 28 after treatment, as in parallel control IMR female fibroblasts. Abbreviation: hESC, human embryonic stem cell.
Table 1. Flow cytometry analysis demonstrates expression of SSEA-4, TRA-1-60, and TRA-1-81 in undifferentiated H1, H7, and H9 human embryonic stem cells
XIST expression is virtually undetectable in undifferentiated male hESC cultures. The female hESC lines, in comparison, exhibit distinct differences in XIST expression; XIST expression is relatively strong in undifferentiated H9 and CyT25 cultures and is similar to levels of expression found in human female fibroblasts (Figs. 1C, 1D). Expression of XIST transcripts, however, is undetectable by PCR in female H7 hESCs (Figs. 1C, 1D). Chromosome paint analysis confirms that two separate X-chromosome territories are apparent in approximately 80% of cultured H7 and H9 hESCs (Figs. 2Avii, 2Aviii; Table 2). Two separate X-chromosomes are not observed in 100% of the interphase cells due to the proximity of the two X-territories and the limitations of two-dimensional scoring; this is not indicative of aneuploidy because karyotype analysis of mitotic chromosomes reveals two X chromosomes in all cells assessed.
Table 2. Quantitation of the number of cells with detectable XIST accumulation on one X chromosome and the number of X chromosomes (territories) found in H9 hESCs, H7 hESCs and TIG-1 female fibroblasts
FISH analysis clearly demonstrates a large XIST RNA accumulation that paints one X chromosome in approximately 90% of H9 hESCs (Fig. 2Aiii, Table 2), notably in both OCT3/4-positive, Tra1-60–positive, and SSEA-4–positive undifferentiated cells with in the colonies and in differentiated cells at the periphery of the hESC colonies (Figs. 2Ai, 2Aii, 2Aiii, 2Av). These numbers are comparable to those scored for normal female fibroblasts (Table 2), indicating that undifferentiated H9 hESC cultures exhibit XCI patterns similar to those observed in normal female somatic cells. This is further demonstrated by the fact that both undifferentiated (blue and arrows in DAPI DNA image, Fig. 3A) as well as differentiated (Fig. 3B) H9 hESC cells exhibit clear heterochromatic Barr bodies by DAPI DNA. In contrast, there is a clear lack of XIST accumulation on either of the chromosomes in H7 hESCs (Figs. 2Aiv, 2Aviii, 3C; Table 2). In some rare undifferentiated H7 cells, there seems to be a very small spot of XIST RNA that is reminiscent of the small unstable transcription observed to occur in mouse ESCs (not shown); however, this could not be rigorously discriminated from background. Regardless, there is no definitive evidence that XIST expression is ever upregulated in this line or accumulates on either X chromosome during differentiation, suggesting that, in contrast to H9 and CyT25 female hESCs, H7 hESCs fail to initiate classic XCI.
Figure 3. (A): As evidenced in the separated color channels to the right of the full color image, a field of undifferentiated H9 (P45) hESCs clearly shows an inactivated X chromosome (Xi), which is painted by XIST RNA (red and XIST RNA image) and exhibits a DAPI dense heterochromatic Barr body (blue and arrows in DAPI DNA image). Hybridization to hnRNA (green) using the Cot-1 probe demonstrates transcriptional silencing by the apparent hole in the signal (arrows in Cot-1 RNA image) over the Barr body. (B): Differentiated H9 hESCs (14-day embryoid bodies) maintain the inactivation of the Xi, which is still painted by XIST RNA (red and XIST RNA image) and exhibits a heterochromatic Barr body (arrow in DAPI DNA image). The apparent hole in the hnRNA signal (arrow in Cot-1 RNA image) over the Barr body suggests continued silencing. (C): The female H7 (P73) hESC line does not exhibit these unmistakable hallmarks of inactivation in either of its two X chromosomes by day 14 of differentiation. Two X chromosome centromeres (red and X-centromere image) are apparent, but neither one is associated with an obvious Barr body (blue and arrows in DAPI DNA image) or Cot-1 hnRNA hole (green and arrows in Cot-1 RNA image). Abbreviations: DAPI, 4,6-diamidino-2-phenylindole; hESC, human embryonic stem cell; hnRNA, heterogeneous nuclear RNA.
Chromosome-Wide Xi Transcriptional Silencing Is Apparent in H9 hESCs but Not in H7 hESCs
The above data suggest that in H9-undifferentiated hESCs, one of the two X chromosomes is inactivated whereas the H7 hESC line never initiates XCI nor inactivates one of its two X chromosomes. To verify this, we evaluated the transcriptional status of the X chromosomes in both H9 and H7 hESCs by hybridization to heterogeneous nuclear RNA (hnRNA) transcription in the nucleus. Briefly, by using a labeled human Cot-1 DNA probe and RNA FISH, we can visualize hnRNA transcription throughout the nucleus; this assay allows direct assessment of global transcription across the entire X chromosome rather than relying on replication timing or expression of just a few individual genes and has become widely used as a method to evaluate chromosome silencing . H9 hESCs clearly show a lack of hnRNA transcription signal, creating a large "black hole" under the XIST RNA accumulation that is coincident with the heterochromatic Barr body and clearly indicative of transcriptional silencing. This is seen in both undifferentiated and differentiated H9 hESCs (Figs. 3A, 3B).
H7 hESCs, in contrast, never accumulate XIST RNA (Figs. 2Aiv, 3C), and as such, we could not use XIST to localize the Xi in these cells. Rather, an X-specific centromere signal (Fig. 3C) delineates both X chromosomes in the H7 line, allowing us to assess whether these reside within a chromosome-size territory lacking in Cot-1/hnRNA hybridization. In addition, neither of the two X chromosomes in H7 hESCs (Fig. 2Aviii) exhibits a large clear hole in the hnRNA signal (Fig. 3C), suggesting a lack of chromosome-wide silencing in this line, even after differentiation. Although technical limitations do not allow us to conclude that all genes on both X chromosomes remain active, results clearly indicate that the H7 hESCs do not show the chromosome-wide transcriptional silencing that is readily apparent in both H9 hESCs or normal control cells. Consistent with this, the DAPI-dense Barr bodies easily identified in the H9 cells are lacking in the H7 cells (Fig. 3C). Together, these findings demonstrate that H7 hESCs lack clear hallmarks of chromosomal silencing in both the undifferentiated state and in cells 14 days after differentiation.
Differences in XCI Do Not Correlate with Differences in Gross hESC Expression Profiles
Despite differences in XCI between the H7 and H9 and CyT25 hESC lines, expression of several molecular markers that are characteristic of hESCs is appropriate and comparable between cell lines. Specifically, we demonstrate via reverse transcription (RT)-PCR that expression of CRIPTO, REX-1, OCT3/4, NANOG, STAT3, and UTF-1 does not vary between undifferentiated H1, H7, H9, CyT25, and BG02 hESC lines (Fig. 1A). Quantitative PCR and flow cytometry analyses further demonstrate the expression of OCT3/4, NANOG, and SSEA-4, Tra-1-60, and Tra-1-81, respectively, in each of these lines (Figs. 1B, 1D; Table 1), indicating that differences in the pattern of XCI do not correlate with changes in expression of characteristic stem cell markers.
Macrochromatin Body Formation in Differentiated hESCs
We next evaluated whether downstream components of the X-inactivation process are expressed appropriately in various hESC lines. In differentiating mouse ESCs, an accumulation of MacroH2A1 and the formation of macrochromatin bodies (MCBs) occur at 5 days of differentiation, subsequent to the upregulation in Xist expression and coating of the X chromosome at 1 day of differentiation . Therefore, we assessed MCBs using immunocytochemical localization of MacroH2A1.
In undifferentiated H9 hESC cultures maintained in feeder-free conditions, MCBs are evident in nuclei of differentiating cells surrounding hESC colonies but are absent in the nuclei of cells within hESC colonies that express SSEA-4 (Figs. 4Ai, 4Aii, 4Aiii). Therefore, the undifferentiated cells in these cultures exhibit XIST coating of the X chromosome but not MCB formation, demonstrating that the H9 hESCs may be in between these two developmental stages. Our observation that the differentiated cells outside of the colonies exhibit MCBs further indicates that the process of XCI is appropriately temporally regulated in cultures of H9 hESCs (see below). In contrast, MCBs are not evident during differentiation in either male H1 (data not shown) or female H7 (Figs. 4Aiv, 4Av, 4Avi) hESC cultures, confirming a lack of XCI in both cell lines.
Figure 4. (A): (i): SSEA-4-positive (in red) H9 (P37) hESCs counterstained with DAPI (in blue). (ii): Costaining of MacroH2A1
Patterns of X-Inactivation Are Heritable to Differentiated Progeny
To further assess whether XCI in female hESCs is coincident with cellular differentiation, EBs were generated and temporal changes in gene expression were analyzed. We demonstrate that an appropriate decrease in expression of OCT3/4 is observed upon in vitro differentiation of all lines tested (Figs. 1B, 1D). Expression of XIST transcripts, however, persists in EB cultures of H9 and CyT25 hESCs (Figs. 1C, 1D) and in H9-derived neural progenitor cultures (113 days of differentiation; data not shown). XIST expression did not increase with differentiation (Figs. 1C, 1D, 4B), confirming that XCI was complete in the undifferentiated cultures.
The percentage of hESCs exhibiting an accumulation of MCBs, in comparison, increases upon differentiation (Fig. 4B). Moreover, MCBs are detectable in ectoderm, endoderm, and mesoderm derivatives of H9 hESCs; colocalization of MacroH2A1 with ?-tubulin is identifiable in 175-day-differentiated H9 neural progenitors (Fig. 4Avii) and with smooth muscle actin (Fig. 4Aviii) and -fetoprotein (Fig. 4Aix) after 7 to 14 days of hESC differentiation, providing further evidence that these female hESCs are differentiating appropriately and that X-inactivation progresses throughout in vitro differentiation. In contrast, there is a lack of expression of XIST transcripts (Figs. 1C, 1D, 2Aiv, 3C) and MacroH2A1 accumulation (Figs. 4Ax, 4Axi, 4Axii) in H7 hESC and EB cultures and in H7-derived neural progenitor cultures (37 days of differentiation; data not shown), providing strong evidence of a lack of classic XCI in this female hESC line.
XCI Occurs Appropriately in an Aneuploid H9 hESC Line
Supernumerary X chromosomes are tolerated in human cells when the extra X chromosomes are inactivated . For instance, XXY or XXXY cells will exhibit XIST coating of one or two X-chromosomes, respectively. We have generated an H9 hESC subclone in which g-banding reveals that 5% of the cells within the culture have an XXX genotype. In these cultures, we observe patches of cells showing 2 XIST signals (Fig. 2Avi); quantitation reveals that 12% ± 7% (mean ± SEM) of the cells in the culture exhibit two XIST signals. These data indicate that although this H9 subclone has an aneuploid karyotype, mechanisms are maintained for the appropriate dosage compensation of the X chromosome.
XCI Is Stable over Time In Vitro
Numerous hESC lines have now been derived, and despite differences in the derivation process and culture conditions, evaluation of standard markers, telomerase activity, pluripotency, and karyotype indicates that hESC lines are stable over extended periods of culture (see Hoffman and Carpenter for review ). In this study, we demonstrate that after more than 80 passages (20 months) in continuous culture, H9 hESCs also retain appropriate expression of XIST transcripts (Fig. 1C, right panel).
Expression of XIST Is Modulated After Treatment with 5-Aza-2'-Deoxycytidine
5-Aza-2'-deoxycytidine is a well-known demethylating agent that is widely used to demonstrate a correlation between loss of methylation in specific regions of a gene and activation of gene activity. Here, we demonstrate via RT-PCR that after treatment with 5-aza-2'-deoxycytidine, XIST expression is detected in H7 hESCs (Fig. 2B, Table 3). In these experiments, H7 hESCs were treated with 1 or 10 μM 5-aza-2'-deoxycytidine for 24 or 48 hours, and expression of XIST and OCT3/4 was assessed 3, 14, and 28 days after treatment. Treatment with 5-aza-2'-deoxycytidine resulted in differentiation as indicated by the decrease in OCT3/4 expression. This was accompanied by the appearance of XIST expression by 3 days after treatment. The expression of XIST persisted for 28 days, the latest time point analyzed. However, the detection of XIST by PCR was not accompanied by Xist coating of an X chromosome, as assessed via FISH analysis (data not shown), indicating a continued lack of XCI in H7 hESCs.
Table 3. Different subclones of H7 human embryonic stem cells at various passage numbers
DISCUSSION
We thank Meg Byron for her excellent technical assistance with molecular cytology experiments. We thank Dmitri Nusinow and Barbara Panning for technical assistance and insightful discussions. This work was supported in part by the Krembil Foundation and an establishment grant from the Canadian Institutes of Health Research awarded to M.K.C.; an NIH RO1 supplemental grant (GM053234-07S1) awarded to J.L.; and a postdoctoral fellowship from the Ontario Research and Development Challenge Fund awarded to L.M.H. The Lawrence laboratory component of this work involved the H9, H7, and H1 hESC cells approved for NIH funding and no other cell lines (CyT25) that are not currently NIH approved.
DISCLOSURES
The authors indicate no potential conflicts of interest.
REFERENCES
Bhattacharya B, Miura T, Brandenberger R et al. Gene expression in human embryonic stem cell lines: unique molecular signature. Blood 2004;103:2956–2964.
Carpenter MK, Rosler ES, Fisk GJ et al. Properties of four human embryonic stem cell lines maintained in a feeder-free culture system. Dev Dyn 2004;229:243–258.
Monk M, Harper MI. Sequential X chromosome inactivation coupled with cellular differentiation in early mouse embryos. Nature 1979;281: 311–313.
Takagi N, Sasaki M. Preferential inactivation of the paternally derived X chromosome in the extraembryonic membranes of the mouse. Nature 1975;256:640–642.
Daniels R, Zuccotti M, Kinis T et al. XIST expression in human oocytes and preimplantation embryos. Am J Hum Genet 1997;61:33–39.
Ray PF, Winston RM, Handyside AH. XIST expression from the maternal X chromosome in human male preimplantation embryos at the blastocyst stage. Hum Mol Genet 1997;6:1323–1327.
Harrison KB. X-chromosome inactivation in the human cytotrophoblast. Cytogenet Cell Genet 1989;52:37–41.
Ropers HH, Wolff G, Hitzeroth HW. Preferential X inactivation in human placenta membranes: is the paternal X inactive in early embryonic development of female mammals? Hum Genet 1978;43:265–273.
Plath K, Mlynarczyk-Evans S, Nusinow DA et al. Xist RNA and the mechanism of X chromosome inactivation. Annu Rev Genet 2002;36: 233–278.
Heard E, Clerc P, Avner P. X-chromosome inactivation in mammals. Annu Rev Genet 1997;31:571–610.
Thomson JA, Itskovitz-Eldor J, Shapiro SS et al. Embryonic stem cell lines derived from human blastocysts. Science 1998;282:1145–1147.
Xu C, Inokuma MS, Denham J et al. Feeder-free growth of undifferentiated human embryonic stem cells. Nat Biotechnol 2001;19:971–974.
Lawrence JB, Singer RH, Marselle LM. Highly localized tracks of specific transcripts within interphase nuclei visualized by in situ hybridization. Cell 1989;57:493–502.
Clemson CM, Chow JC, Brown CJ et al. Stabilization and localization of XIST RNA are controlled by separate mechanisms and are not sufficient for X inactivation. J Cell Biol 1998;142:13–23.
Hall LL, Byron M, Sakai K et al. An ectopic human XIST gene can induce chromosome inactivation in postdifferentiation human HT-1080 cells. Proc Natl Acad Sci U S A 2002;99:8677–8682.
Clemson CM, McNeil JA, Willard HF et al. 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.
Johnson CV, Singer RH, Lawrence JB. Fluorescent detection of nuclear RNA and DNA: implications for genome organization. Methods Cell Biol 1991;35:73–99.
Lee JT, Strauss WM, Dausman JA et al. A 450 kb transgene displays properties of the mammalian X-inactivation center. Cell 1996;86:83–94.
Panning B, Jaenisch R. DNA hypomethylation can activate XIST expression and silence x-linked genes. Genes Dev 1996;10:1991–2002.
Penny GD, Kay GF, Sheardown SA et al. Requirement for XIST in X chromosome inactivation. Nature 1996;379:131–137.
Clerc P, Avner P. Role of the region 3' to XIST exon 6 in the counting process of X-chromosome inactivation. Nat Genet 1998;19:249–253.
Sheardown SA, Duthie SM, Johnston CM et al. Stabilization of XIST RNA mediates initiation of X chromosome inactivation. Cell 1997;91:99–107.
Panning B, Dausman J, Jaenisch R. X chromosome inactivation is mediated by XIST RNA stabilization. Cell 1997;90:907–916.
Okamoto I, Otte AP, Allis CD et al. Epigenetic dynamics of imprinted X inactivation during early mouse development. Science 2004;303: 644–649.
Huynh KD, Lee JT. Inheritance of a pre-inactivated paternal X chromosome in early mouse embryos. Nature 2003;426:857–862.
Mermoud JE, Costanzi C, Pehrson JR et al. Histone MacroH2A1.2 relocates to the inactive X chromosome after initiation and propagation of x-inactivation. J Cell Biol 1999;147:1399–1408.
Rasmussen TP, Mastrangelo MA, Eden A et al. Dynamic relocalization of histone MacroH2A1 from centrosomes to inactive X chromosomes during X inactivation. J Cell Biol 2000;150:1189–1198.
Costanzi C, Pehrson JR. Histone MacroH2A1 is concentrated in the inactive X chromosome of female mammals. Nature 1998;393:599–601.
MacDonald M, Hassold T, Harvey J et al. The origin of 47, XXY and 47, XXX aneuploidy: heterogeneous mechanisms and role of aberrant recombination. Hum Mol Genet 1994;3:1365–1371.
Hoffman L, Carpenter MK. Characterization and culture of human embryonic stem cells. Nat Biotechnol 2005;23:699–708.
Brown CJ. Equality of the sexes: mammalian dosage compensation. Semin Reprod Med 2001;19:125–132.
Sato N, Sanjuan IM, Heke M et al. Molecular signature of human embryonic stem cells and its comparison with the mouse. Dev Biol 2003;260:404–413.
Ginis I, Luo Y, Miura T et al. Differences between human and mouse embryonic stem cells. Dev Biol 2004;269:360–380.
Dhara SK, Benvenisty N. Gene trap as a tool for genome annotation and analysis of X chromosome inactivation in human embryonic stem cells. Nucleic Acids Res 2004;32:3995–4002.(Lisa M. Hoffmana, Lisa Ha)