| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Graduate School of Life and Environmental Sciences, University of Tsukuba, 1-1-1 Ten-noudai, Tsukuba, Ibaraki 305-8572, Japan
2 Technology and Development Team for Mammalian Cellular Dynamics, BioResource Center, RIKEN Tsukuba Institute, 3-1-1 Koyadai, Tsukuba, Ibaraki 305-0074, Japan
3 Division of Developmental Genetics, Institute of Molecular Embryology and Genetics, Kumamoto University, Kumamoto 860-0811, Japan
4 Division of Regeneration Medicine, Institute of Molecular Embryology and Genetics, Kumamoto University, Kumamoto 860-0811, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionExperimental proceduresReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Embryonic stem (ES) cells in mammals are pluripotent and self-renewing; such properties are supported by the gene expression programs specific to these stem cells (Tanaka et al. 2002). Upon differentiation, ES cells undergo dramatic changes in gene expression, which may be accompanied by large-scale nuclear reorganization. To visualize the dynamics of nuclear organization during differentiation and to relate such changes with epigenetic properties, we focused here on DNA methylation patterns.
Methylation of CpG dinucleotide is an important epigenetic system for the regulation of gene expression during mammalian development; such global patterns of DNA methylation are often used as "markers" for nuclear remodeling (Santos et al. 2002). Here we established an experimental strategy to observe DNA methylation at the chromosomal level in individual, living ES cells, and then used this strategy to trace organizational changes in the nucleus during ES cell differentiation.
One class of genes encodes protein factors that contain common methylated DNA binding domains (MBDs; Hendrich & Bird 1998). One such gene, MBD1, has been studied in detail. The MBD1 protein binds specifically to CpG sequences in a methylation-dependent manner and is involved in transcriptional silencing (Fujita et al. 1999, 2003; Ohki et al. 1999). We made a construct in which we fused the GFP reporter gene to MBD and to the nuclear localization signal (nls) sequences of human MBD1, which confer specific binding to methylated DNA. We validated the utility of this construct using an in vitro ES cell differentiation system. This method enabled us to monitor changes in the global DNA methylation pattern in individual cells and the dynamics of chromosomal repositioning during ES cell differentiation. We found that pluripotent ES cells have a unique nuclear organization, which undergoes dramatic organizational changes in a step-wise fashion during differentiation. Conversely, such nuclear properties can be used as markers that define each step of development. The higher-order nuclear structures and epigenetic state appear to be cell type-specific, thus providing evidence for potential linkage between nuclear organization, epigenetic status and cell differentiation.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
To visualize the dynamics of nuclear reorganization and global patterns of DNA methylation in living ES cells, we took advantage of the methylated DNA binding ability of the MBD1 protein (Fujita et al. 1999). MBD1 includes a binding domain for methylated DNA (MBD) and a nuclear localization signal (nls) at its N-terminus. The MBD region of MBD1 is sufficient for specific binding to a single methylated CpG sequence (Fujita et al. 1999). We made the construct CAGGS-EGFP-MBD-nls, in which the MBD-nls sequence was fused to the GFP reporter gene driven by the CAGGS promoter (Niwa et al. 1991). We transfected this construct into mouse E14 TG2a ES cells (Smith & Hooper 1987) and established ES cell lines that constitutively expressed the EGFP-MBD-nls fusion gene and showed normal growth behavior and morphology of undifferentiated cells (Fig. 1A). When these cells were induced to differentiate in vitro, embryoid bodies including highly differentiated, cystic embryoid bodies, were formed at a rate similar to that of wild-type ES cells (Fig. 1B,C). Thus, introduction of the construct did not cause adverse effects on either growth or differentiation of this particular ES cell line.
|
Localization of EGFP-MBD-nls corresponds to sites of methylated DNA
As expected, the EGFP-MBD-nls product localized specifically to nuclei. In metaphase cells, metaphase plates of chromosomes were labeled specifically with EGFP-MBD-nls (Fig. 1D). In mice, the DNA of centromeric satellite sequences that constitute centromeric heterochromatin is highly methylated (Mitchel 1996). The centromeric regions of the acrocentric mouse chromosomes were labeled intensely with EGFP-MBD-nls, whereas the other chromosomal regions were labeled less strongly (Fig. 1D). During interphase, the entire nucleus was labeled with multiple foci showing very strong GFP signals (Fig. 1F,F''). These foci were likely to represent aggregates of centromeric or pericentric heterochromatin or both. Because they overlapped with the fluorescence in situ hybridization (FISH) signals detected by mouse centromere-specific probes (not shown), we hereafter call these aggregates centromeric heterochromatin. Such intensely labeled foci were fewer than the number of chromosomes in mice (i.e., 40/2n), indicating that each cluster comprised several chromosomes. Indeed, FISH analysis with a telomere-specific probe revealed several telomere spots in each of the large clusters (not shown). Thus, these centromeric heterochromatin clusters represent so-called chromocenters (Stephanova et al. 1988). In contrast, subnuclear regions where no GFP signals were also observed (Fig. 1F), the larger regions of which were nucleoli. The numerous smaller areas lacking GFP fluorescence were probably interchromatin compartments (IC) where active transcription and RNA processing take place (Manuelidis 1990). Immunostaining with antibodies against either a splicing factor, SC35 (Fig. 1F',F''), or the active form of RNA polymerase II confirmed this (not shown).
These results suggest that EGFP-MBD-nls localizes specifically to hypermethylated centromeric heterochromatin as well as euchromatic regions. Indeed, the localization of GFP on metaphase chromosomes was essentially the same as the staining pattern of chromosome spreads with an anti-5-MeC(5-methyl cytosine) antibody (Fig. 1E).
We used the fluorescence recovery after photobleaching (FRAP) method to measure the turnover rate of EGFP-MBD-nls (Fig. S1) and found that 50% fluorescence recovery was achieved within 18.2 ± 8.2 s (mean ± 1 SD, n = 12) in undifferentiated cells and 22.2 ± 11.4 s (n = 12) in differentiated cells; this difference was not statistically significant. Thus, the EGFP-MBD-nls fusion protein is turning over rapidly (Fig. S1). This rapid turnover of EGFP-MBD-nls suggests that the localization of this protein traces temporal changes in the status of genomic DNA methylation during cell differentiation.
We next examined the production of endogenous MBD1 and another methylated DNA binding protein, MECP2, in ES cells and compared the localization of these proteins with that of EGFP-MBD-nls. We used an antibody specific to the C terminus of MBD1, which does not react with the MBD domain in the EGFP-MBD-nls fusion protein (not shown). MBD1 protein was produced in undifferentiated ES cells, but the amount was significantly lower than that in differentiating ES cells (Fig. 1H,I). In undifferentiated ES cells, MBD1 localized to DAPI-stained centromeric heterochromatin as well as to euchromatic regions (Fig. 1G–G''). Endogenous MBD1 localization overlapped that of EGFP-MBD-nls, although the fusion protein appeared to be localized more broadly than the endogenous protein (Fig. 1H–H'').
In the nuclei of differentiating ES cells, MBD1 protein was present in both euchromatic and heterochromatic regions but was heavily concentrated on centromeric heterochromatin. Localization of MBD1 on centromeric heterochromatin overlapped that of EGFP-MBD-nls perfectly (Fig. 1I–I''). The intensities of both MBD1 immunostaining signals and EGFP signals increased significantly when cells were more fully differentiated (Fig. 1I and see Figs 2 and 4B). This finding suggests that the fusion protein does not outcompete endogenous MBD1 from methylated DNA sites. Instead, the increases in localized MBD1 protein levels were correlated with increased methylation levels of genomic DNA as revealed by the GFP signals.
|
Changes in nuclear organization and epigenetic state during ES cell differentiation
During the above experiments, we noticed that the localization patterns and signal intensities of the EGFP-MBD-nls protein changed markedly during ES cell differentiation (Fig. 1H,I). To study these processes, we first classified differentiating cells according to the level of OCT3/4 protein, a convenient marker for undifferentiated ES cells (Niwa et al. 2000), and examined the localization of EGFP-MBD-nls at each differentiation stage.
OCT3/4 production was down-regulated as ES cell differentiation proceeded. After withdrawal of leukemia inhibitory factor (LIF), cells started to differentiate and appeared to show different levels of OCT3/4 depending on the differentiation state (Fig. S2). Therefore, we could classify differentiating ES cell populations into at least four classes exhibiting different levels of OCT3/4 (Figs. 2 and S3). Class 1 represents undifferentiated ES cells with the highest level of OCT3/4 (Fig. 2A'), whereas class 2 cells are differentiating and contain approximately half the amount of OCT3/4 compared with class 1 cells (Fig. 2B'). Class 3 cells contain scant OCT3/4, and class 4 cells have almost none (Fig. 2C',D'). The nuclei of class 1 cells contained large chromocenters surrounding the nucleolus (Fig. 2A). In class 2 cells, the large chromocenters found in the undifferentiated cells were dispersed into smaller clusters (Fig. 2B). The numbers of EGFP-MBD-nls positive clusters increased further in class 3 cell nuclei, and the clusters appeared to be decondensed (Fig. 2C). The intensity of the GFP signal decreased dramatically from class 1 to class 3 cells (Fig. 2A–C). In the more fully differentiated class 4 cells (Fig. 2D), the centromeric heterochromatin was heavily labeled with EGFP-MBD-nls and appeared to be more tightly congregated than in class 3 cells. Although we are not certain that these four classes of cells are in the same developmental pathway, class 3 and 4 cells were rarely seen under culture conditions that maintained cells in the undifferentiated state. The 3D reconstructions of confocal images revealed that the typical class 1 nucleus was spherical and that the large chromocenters surrounding nucleoli appeared to be only loosely associated with the nuclear envelope (Fig. 3A–D). Nuclei of class 3 cells were flattened, and the chromocenters were closely associated with the nuclear membrane (Fig. 3E,F).
|
|
Southern blot analysis was performed using DNAs isolated from undifferentiated ES cells and differentiating cells (Fig. 4D). DNAs from Dnmt1–/– ES cells and cultured fibroblast cells were also used as controls. The results demonstrated that the minor satellite DNA of undifferentiated ES cells was highly methylated (Fig. 4D, lane 2), whereas small HpaII digested fragments of the repeat sequences were observed only in a sample from culture without LIF and with RA (Fig. 4D, lane 3). Approximately 70% of cells in the sample corresponded to class 2 and 3, and less than 10% of cells were classified as class 4. Thus the hypomethylated fragments seen in lane 3 were most likely derived from class 2 and 3 cells.
Changes in DNA methylation pattern revealed by EGFP-MBD-nls binding are associated with changes in levels of DNA methyltransferases
As shown by the EGFP-MBD-nls localization, global DNA methylation patterns undergo dramatic changes during ES cell differentiation. Changes in the production and localization of DNA methyltransferases probably underlie this genome-wide epigenetic remodeling. To test this, we examined the production of DNA methyltransferases in ES cells harboring the EGFP-MBD-nls transgene to determine whether the level of DNA methylase production was associated with the localization of EGFP-MBD-nls. Such comparisons of DNA methylation with methylase activities/localizations have not been made at a single-cell level by other groups, because the harsh treatments (HCl treatment or UV irradiation) required for anti-5-methyl cytosine (5MeC) antibody staining (Santos et al. 2002) preclude double staining with other antibodies.
In mammals, at least two active de novo methyltransferases, DNMT3a and DNMT3b, and their isoforms have been identified (Goll & Bestor 2005). We used an anti-DNMT3b antibody (Watanabe et al. 2004) to detect all isoforms of this methyltransferase. We also used the anti-DNMT3a internal antibody, which, in theory, reacts with both DNMT3a2 and DNMT3a1 enzymes (Watanabe et al. 2004). We used the anti-DNMT3a N terminus antibody, which detects the DNMT3a1 protein only, to discriminate between the 3a1 and 3a2 isoforms (Watanabe et al. 2004). DNMT3a2 co-localized with EGFP-MBD-nls on the chromocenters of undifferentiated ES cells (Fig. 5A–A''). As the staining patterns of DNMT3b were essentially similar to that of DNMT3a2, the results for DNMT3b are shown on Fig. S4. Upon induction of differentiation, the levels of DNMT3a2 increased transiently (Fig. 5B–B''). However, the localization of the enzyme changed at this stage; DNMT3a2 was more abundant in euchromatic regions and less so at the centromeric heterochromatin (Fig. 5C–C''). The overall levels of DNMT3a2 then declined gradually. Concomitant with these declines in enzyme concentrations at the centromeric heterochromatin, the intensity of EGFP signal at the centromeric heterochromatin decreased, and the heterochromatin clusters appeared to loosen (Fig. 5D–D''). DNMT3b protein became undetectable in the more fully differentiated cells (Fig. S4), although the GFP signals at the centromeric heterochromatin again increased (Fig. S4 and Fig. 5E). Production of the DNMT3a1 isoform, as detected by the isoform-specific antibody, was up-regulated in the highly differentiated cells, although it was not detected in undifferentiated ES cells and remained low in the class 2 and 3 cells (Figs. 5E' and S4). Thus the dramatic increase in EGFP-MBD-nls fluorescence intensity (and DNA methylation) in the class 4 cells was caused by increased production of the DNMT3a1 isoform.
|
(HP1), which constitutes heterochromatin (Eissenberg & Elgin 2000), remained localized to centromeric heterochromatin in differentiating Dnmt1–/– cells (arrow in Fig. 6E), in which only diffuse GFP signals were detected. Immunostaining with anti-trimethyl histone H3 (Lys9) also showed that the centromeric heterochromatin containing tri-methylated H3 (Lys9) was still present in the cells in which no EGFP-MBD-nls foci were visible (data not shown). These differentiating mutant cells cannot survive after this stage, probably because of the total lack of DNA methylation in the genome (Li et al. 1992; Lei et al. 1996; Jackson-Grusby et al. 2001). We confirmed that the cells died after the EGFP-MBD-nls signal became undetectable.
|
sometimes do not. Thus, EGFP-MBD-nls localization does not reflect the status of heterochromatin organization per se but rather acts as a "pan-indicator" of methylated DNA. Time-lapse analyses of the dynamics of chromosomal repositioning during ES cell differentiation
Because EGFP-MBD-nls localization corresponds to highly methylated centromeric heterochromatin, it should be possible to monitor how chromosomal positions are arranged during cell cycling or cell differentiation using this marker. We observed above that the nuclear disposition of pluripotent ES cells differs from that of their differentiated descendants. To study this phenomenon in detail, we first performed time-lapse analysis of nuclear reorganization processes during the self-renewal cycle of undifferentiated ES cells. Approximately 20 min after formation of the metaphase plate, chromosomes were starting to segregate, but each chromosome was still distinguishable (Fig. S5). In the sister nuclei thus formed, the centromeric heterochromatin began to congregate, and large heterochromatin clusters characteristic of undifferentiated ES cells emerged (Fig. S5). Once the large clusters of centromeric heterochromatin had formed, the relative positioning of the centromere was maintained throughout interphase (Fig. S5). Cell type-specific chromosome positioning was complete as soon as mitosis ended, and no global reorganization of chromosome position occurred during interphase in undifferentiated cells (Supplementary Movie S1).
Because the pattern of nuclear organization proved to be distinct between undifferentiated and differentiated cells, it seemed reasonable to expect transition from the undifferentiated to the differentiated type of nuclear organization. We succeeded in capturing a moment of dramatic change in chromosome positioning and heterochromatin organization during ES cell differentiation (Fig. 7; Supplementary Movie S2). Time-lapse imaging demonstrated that one cell with a nucleus containing about seven large chromocenters underwent mitotic division, and as soon as cell division ceased, the heterochromatin clusters reorganized to yield more than 20 heterochromatin clusters in the nucleus of the differentiating ES cell (Supplementary Movie S2). This result clearly shows that the nuclear reorganization events we described above indeed take place in a consecutive manner during cell differentiation and that such developmental change in chromosomal positioning coincides with the nuclear organizational changes coupled with cell division.
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
The MBD1a isoform binds to non-methylated CpG (Jorgensen et al. 2004), but such DNA binding activity is attributed to the CXXC-3 domain of the protein, which was deleted from the construct used here. Because the MBD-nls fragment that we used binds specifically to methylated DNA (Fujita et al. 1999; Ohki et al. 1999), the GFP localization should be specific to methylated CpG sequences. MBD1 interacts with heterochromatin components such as HP1 (Fujita et al. 2003), so localization of EGFP-MBD-nls to the centromeric heterochromatin may be mediated by HP1 rather than by DNA methylation per se. However, HP1 and EGFP-MBD-nls did not co-localize in differentiating Dnmt1–/– cells. By contrast, HP1ß—a homolog of HP1—behaves differently from HP1 and from EGFP-MBD-nls during differentiation; HP1ß localization in the centromeric heterochromatin decreased gradually as cell differentiation proceeded, whereas EGFP-MBD-nls and HP1 remained in the heterochromatin of wild-type ES cells (Satoru Kobayakawa and Kuniya Abe, unpublished data). These disparities in localization eliminate the possibility that either HP1 or HP1ß can mediate EGFP-MBD-nls localization directly.
The EGFP-MBD-nls fusion protein might have a dominant-negative effect over the endogenous protein. However, cell growth and development of the ES line carrying the construct were normal both in vitro and in vivo. EGFP-MBD-nls co-localized with MBD1, and the amounts of both proteins increased proportionally upon differentiation, suggesting that the fusion product does not outcompete endogenous MBDs from methylated genomic sequences when the fusion protein amount is kept at an appropriate level. Considering these results and its relatively rapid turnover rate (as revealed by the FRAP analysis), EGFP-MBD-nls can be used to follow DNA methylation in situ under physiological conditions. Thus, our experimental system using this transgene presents a useful alternative to the anti-5-MeC antibody for monitoring genomic DNA methylation patterns. This approach has at least two advantages over the existing method. First, the EGFP-MBD-nls reporter can be used to interpret the methylation status of living cells, thus allowing time-lapse observation of the cells and molecular or phenotypic analyses after their observation. Second, because it does not use the harsh cell treatments required for 5-MeC immunostaining, double staining with other antibodies is possible. Indeed, we used multiple antibodies to DNA methyltransferases to relate DNA methylation levels with methylase localization in the same individual cells: to our knowledge, this is the first time this type of study has been done.
Changes in nuclear organization in parallel with alterations in the epigenetic state of the genome are important driving forces for developmental gene expression (Cremer & Cremer 2001; Misteli 2005). However there have been no systematic analyses of the nuclear remodeling events in stem cells. Here we were able for the first time to conduct detailed analyses of developmental changes in DNA methylation and chromosome positioning in living ES cells.
Interestingly, centromeric heterochromatin in the undifferentiated ES cells formed a few large chromocenters associated with nucleoli. Brero et al. (2005) have reported that aggregation of pericentric heterochromatin occurs during differentiation of myoblasts to myotubes and that ectopic expression of MeCP2 induces clustering of the heterochromatin, leading to the formation of large chromocenters. However, we believe that the large chromocenter clusters found in the undifferentiated ES cells were not artifacts caused by introduction of the EGFP-MBD-nls transgene, because ES cells harboring no transgene also showed very similar heterochromatin clustering, as revealed by DAPI staining or HP1 immunostaining. The clustering of the heterochromatin that we noted during later phases of differentiation may be analogous to that observed by Brero et al. (2005). In both cases, heterochromatin clustering coupled with increased DNA methylation occurred as cells progressed toward a more fully differentiated state. It is unclear whether chromocenter aggregation is a general phenomenon common to all terminally differentiated cells. Each differentiated cell type seems to have a specific spatial genomic organization (Mayer et al. 2005), and there may be various ways to achieve nuclear architectures unique to different types of differentiated cells.
Our time-lapse observations of the centromeric heterochromatin clusters unequivocally captured the moment of nuclear reorganization for the first time. This repositioning takes place during mitosis and is coupled with destruction and restructuring of nuclear architectures, probably because chromosomes need to be released from and engage with the various subnuclear structures that tether the chromosomes. We do not know at present what factors or structures within the nucleus are responsible for these dynamics of chromosome repositioning. Congregation or repositioning of centromeric heterochromatin may reflect changes in the epigenetic state of chromatin. Trichostatin A (TSA), a potent inhibitor of histone deacetylase, induces disruption of pericentric heterochromatin and dissociation of HP1 from the chromatin domain, leading to relocation of the heterochromatin to the nuclear periphery (Taddel et al. 2001). Removal of TSA reverses this, suggesting that the level of histone acetylation affects the nuclear compartmentalization of centromeric heterochromatin. We also observed that, in conjunction with chromocenter dissociation, HP1ß was released from heterochromatin aggregates (Satoru Kobayakawa and Kuniya Abe, unpublished data). This implies alterations in the interactions between heterochromatin and HP1ß, but the true importance of this HP1ß release is yet to be determined. Alternatively, the morphology of the nucleus may be a reflection of its transcriptional activities and self-organization (Misteli 2005). Meshorer et al. (2006) have reported that architectural chromatin proteins, including HP1, are hyperdynamic and bind loosely to chromatin in pluripotent ES cells. Thus, central nuclear localization of large chromocenters, a hallmark of undifferentiated ES cells (Wiblin et al. 2005; this study), may be achieved through a gene transcription program regulated by the action of the hyperdynamic chromatin proteins.
Although topological changes in chromatin domains are likely to alter developmental gene expression (Misteli 2005), knowledge of the functional aspects of nuclear positioning is still very limited, partly because of the absence of appropriate experimental models amenable to developmental analysis. The EGFP-MBD-nls can reveal the nuclear architecture specific to a particular developmental stage of living ES cells. Using this system, we have found that, coupled with nuclear reorganization, genomic loci for the pluripotent cell-specific Oct3/4 gene segregate to distinct subnuclear compartments, resulting in repression of the gene upon cell differentiation (Satoru Kobayakawa and Kuniya Abe, unpublished data).
In addition to such specific gene analyses, global analyses of gene expression and epigenetic state of the cells with similar nuclear organizations are now possible with this technique. This would expand the knowledge of the involvement of nuclear architecture in gene expression regulation and should assist the search for molecules or factors that induce changes in DNA methylation and nuclear organization during cell differentiation.
|
Experimental procedures |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
cDNA fragments encoding amino acids 1 through 112 of human MBD1, corresponding to the MBD and nls coding regions, were cloned into the pEGFP-C1 vector (Clontech, Mountain View, CA) (Fujita et al. 1999). The NheI-MluI fragment of the cDNA fragment was blunt-ended and cloned into the pCAGGS vector (Niwa et al. 1991). The construct was co-transfected with the neomycin-resistance gene into E14TG2a ES cells (Smith & Hooper 1987), and stable lines were isolated. E14TG2a cells were cultured on 0.1% gelatin-coated dishes without feeder layers in LIF-supplemented Dulbecco's modified Eagle's medium containing 10% fetal calf serum (Hyclone, Logan, UT) at 37 °C under 5% CO2 in air. ES cells were induced to differentiate by removal of LIF and addition of 100 nM all-trans-retinoic acid as described (Smith & Hooper 1987). Embryoid body formation was carried out as described (Wiles 1993).
Immunostaining
For immunostaining, cells were fixed in 4% paraformaldehyde and permeabilized with 0.2% Triton X-100 for 15 min at room temperature. Samples were incubated with primary antibody in phosphate buffered saline (PBS) overnight at 4 °C and washed four times in PBS. Fluorescent secondary antibody diluted in PBS was applied for 60 min at room temperature and cells were washed again. Coverslips were mounted with DABCO containing 4,6-diamidino-2-phenylindole (DAPI). Immunostaining with monoclonal antibody to 5-methylcytosine (5-MeC) was conducted according to Miniou et al. (1994). Fluorescent images were captured as optical sections by using a confocal laser scanning microscope (LSM-510 META, Carl Zeiss). The 3D reconstructions and intensity measurements were performed using an image processing system (KS 3D-Lite, Carl Zeiss) equipped with LSM software and IMAGE-PRO PLUS software (Media Cybernetics, Silver Spring, MD).
Antibodies
The following primary antibodies were used: anti-MBD1 (M-254; Santa Cruz Biotech, Santa Cruz, CA), anti-MECP2 (#07-013; Upstate, Chicago, IL), anti-OCT3/4 (N-19; Santa Cruz Biotech), anti-GFP (A11122
[GenBank]
; Molecular Probes, Eugene, OR), anti-SC-35 (S4045; Sigma-Aldrich, St Louis, MO), anti-HP1 (MAB3584; Chemicon, Temecula, CA), anti-HP1ß (MAB3448; Chemicon), anti-trimethyl-histoneH3 (Lys9) (#07-442; Upstate, Chicago, IL), anti-B23 (C-19; Santa Cruz Biotech) and Lamin B (C-20; Santa Cruz Biotech). Anti-5MeC antibody was a kind gift of Dr Y. Kagawa of TORAY Research Center. Antibodies for DNMT1, DNMT3a1, DNMT3a internal, DNMT3b antibodies (Watanabe et al. 2004) were kindly provided by Dr Shoji Tajima of Osaka University, Japan. Secondary antibodies used were anti-rabbit Alexa-488, anti-rabbit Alexa-594, anti-mouse Alexa-594 and anti-goat Alexa-594 (Molecular Probes).
Live cell imaging
For live cell imaging, cells were grown on glass-bottomed culture dishes (P35GC-0-10-C; MatTek Corporation, Ashland, MA). Before imaging, the medium was refreshed. Imaging was performed using the LSM510 confocal microscope. A chamber connected to a CO2 incubator (CZI-3, Carl Zeiss) was placed on the stage of the microscope, and the cells were maintained in 5% CO2 in air at 37 °C during imaging.
Fluorescence recovery after photobleaching (FRAP)
For FRAP analysis, ES cells producing the EGFP-MBD-nls protein were identified on the confocal microscope. An area of the selected cell was outlined for bleaching, and the selected area was bleached using 25 bleaching iterations with 75% laser power and 100% transmittance. After photobleaching, scans were taken at routine imaging intensity every 2 s for 2 min.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
* Correspondence: E-mail: abe{at}rtc.riken.jp
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionExperimental proceduresReferences |
|---|
Cremer, T. & Cremer, C. (2001) Chromosome territories, nuclear architecture and gene regulation in mammalian cells. Nat. Rev. Genet. 2, 292–301.[CrossRef][Medline]
van Driel, R., Fransz, P.F. & Verschure, P.J. (2003) The eukaryotic genome: a system regulated at different hierarchical levels. J. Cell Sci. 116, 4067–4075.
Eissenberg, J.C. & Elgin, S.C. (2000) The HP1 protein family: getting a grip on chromatin. Curr. Opin. Genet. Dev. 10, 204–210.[CrossRef][Medline]
Fujita, N., Takebayashi, S., Okumura, K., Kudo, S., Chiba, T., Saya, H. & Nakao, M. (1999) Methylation-mediated transcriptional silencing in euchromatin by methyl-CpG binding protein MBD1 isoforms. Mol. Cell. Biol. 19, 6415–6426.
Fujita, N., Watanabe, S., Ichimura, T., Tsuruzoe, S., Shinkai, Y., Tachibana, M., Chiba, T. & Nakao, M. (2003) Methyl-CpG binding domain 1 (MBD1) interacts with the Suv39h1-HP1 heterochromatic complex for DNA methylation-based transcriptional repression. J. Biol. Chem. 278, 24132–24138.
Goll, M.G. & Bestor, T.H. (2005) Eukaryotic cytosine methyltransferases. Ann. Rev. Biochem. 74, 481–514.[CrossRef][Medline]
Hendrich, B. & Bird, A. (1998) Identification and characterization of a family of mammalian methyl-CpG binding proteins. Mol. Cell. Biol. 18, 6538–6547.
Jackson-Grusby, L., Beard, C., Possemato, R., Tudor, M., Fambrough, D., Csankovszki, G., Dausman, J., Lee, P., Wilson, C., Lander, E. & Jaenisch, R. (2001) Loss of genomic methylation causes p53-dependent apoptosis and epigenetic deregulation. Nature Genet. 27, 31–39.[CrossRef][Medline]
Jorgensen, H.F., Ben-Porath, I. & Bird, A.P. (2004) Mbd1 is recruited to both methylated and nonmethylated CpGs via distinct DNA binding domains. Mol. Cell. Biol. 24, 3387–3395.
Kosak, S.T. & Groudine, M. (2004) Form follows function: the genomic organization of cellular differentiation. Genes Dev. 18, 1371–1384.
Lei, H., Oh, S.P., Okano, M., Juttermann, R., Goss, K.A., Jaenisch, R. & Li, E. (1996) De novo DNA cytosine methyltransferase activities in mouse embryonic stem cells. Development 122, 3195–3205.[Abstract]
Li, E., Bestor, T.H. & Jaenisch, R. (1992) Targeted mutation of the DNA methyltransferase gene results in embryonic lethality. Cell 69, 915–926.[CrossRef][Medline]
Manuelidis, L. (1990) A view of interphase chromosomes. Science 250, 1533–1540.
Mayer, R., Brero, A., von Hase, J., Schroeder, T., Cremer, T. & Dietzel, S. (2005) Common themes and cell type specific variations of higher order chromatin arrangements in the mouse. BMC Cell Biol. 6, 44.[CrossRef][Medline]
Meshorer, E., Yellajoushula, D., George, E., Scambler, P.J., Brown, D.T. & Misteli, T. (2006) Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells. Dev. Cell 10, 105–116.[CrossRef][Medline]
Miniou, P., Jeanpierre, M., Blanquet, V., Sibella, V., Bonneau, D., Herbelin, C., Fischer, A., Niveleau, A. & Viegas-Pequignot, E. (1994) Abnormal methylation pattern in constitutive and facultative (X inactive chromosome) heterochromatin of ICF patients. Hum. Mol. Genet. 3, 2093–2102.
Misteli, T. (2005) Concepts in nuclear architecture. BioEssays 27, 477–487.[CrossRef][Medline]
Mitchell, A.R. (1996) The mammalian centromere: its molecular architecture. Mutat. Res. 372, 153–162.[Medline]
Niwa, H., Miyazaki, J. & Smith, A. (2000) Quantitative expression of Oct-3/4 defines differentiation, dedifferentiation or self-renewal of ES cells. Nat. Genet. 24, 372–376.[CrossRef][Medline]
Niwa, H., Yamamura, K. & Miyazaki, J. (1991) Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108, 193–199.[CrossRef][Medline]
Ohki, I., Shimotake, N., Fujita, N., Nakao, M. & Shirakawa, M. (1999) Solution structure of the methyl-CpG binding domain of the methylation-dependent transcriptional repressor, MBD1. EMBO J. 18, 6653–6661.
Sanford, J., Forrester, L., Chapman, V., Chandley, A. & Hastie, N. (1984) Methylation patterns of repetitive DNA sequences in germ cells of Mus musculus. Nucleic Acids Res. 12, 2823–2836.
Santos, F., Hendrich, B., Reik, W. & Dean, W. (2002) Dynamic reprogramming of DNA methylation in the early mouse embryo. Dev. Biol. 241, 172–182.[CrossRef][Medline]
Smith, A.G. & Hooper, M.L. (1987) Buffalo rat liver cells produce a diffusible activity which inhibits the differentiation of murine embryonal carcinoma and embryonic stem cells. Dev. Biol. 121, 1–9.[CrossRef][Medline]
Stephanova, E., Russanova, V., Chentsov, Y. & Pashev, I. (1988) Mouse centromeric heterochromatin: isolation and some characteristics. Exp. Cell Res. 179, 545–553.[CrossRef][Medline]
Taddel, A., Maison, C., Roche, D. & Almouzni, G. (2001) Reversible disruption of pericentric heterochromatin and centromere function by inhibiting deacetylases. Nat. Cell Biol. 3, 114–120.[CrossRef][Medline]
Tanaka, T.S., Kunath, T., Kimber, W.L., Jaradat, S.A., Stagg, C.A., Usuda, M., Yokota, T., Niwa, H., Rossant, J. & Ko, M.S. (2002) Gene expression profiling of embryo-derived stem cells reveals candidate genes associated with pluripotency and lineage specificity. Genome Res. 12, 1921–1928.
Tsumura, A., Hayakawa, T., Kumaki, Y., Takebayashi, S., Sakaue, M., Matsuoka, C., Shimotohno, K., Ishikawa, F., Li, E., Ueda, H.R., Nakayama, J. & Okano, M. (2006) Maintenance of self-renewal ability of mouse embryonic stem cells in the absence of DNA methyltransferases Dnmt1, Dnmt3a and Dnmt3b. Genes Cells 11, 805–814.
Watanabe, D., Suetake, I., Tajima, S. & Hanaoka, K. (2004) Expression of Dnmt3b in mouse hematopoietic progenitor cells and spermatogonia at specific stages. Gene Expr. Patterns 5, 43–49.[CrossRef][Medline]
Wiblin, A.E., Cui, W., Clark, A.J. & Bickmore, W.A. (2005) Distinctive nuclear organisation of centromeres and regions involved in pluripotency in human embryonic stem cells. J. Cell Sci. 118, 3861–3868.
Wiles, M.V. (1993) Embryonic stem cell differentiation in vitro. Methods Enzymol. 225, 900–918.[Medline]
Yamagata, K., Yamazaki, T., Yamashita, M., Hara, Y., Ogonuki, N. & Ogura, A. (2005) Noninvasive visualization of molecular events in the mammalian zygote. Genesis 43, 71–79.[CrossRef][Medline]
Yusufzai, T.M., Tagami, H., Nakatani, Y. & Feisenfeld, G. (2004) CTCF tethers an insulator to subnuclear sites, suggesting shared insulator mechanisms across species. Mol. Cell 13, 291–298.[CrossRef][Medline]
Received: 30 August 2006
Accepted: 20 December 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | ADVANCED SEARCH | TABLE OF CONTENTS |
5'; Ethidium Bromide staining pattern of digested DNAs in lanes 1–5. For the Southern analysis, equal amounts (5 µg) of genomic DNA were digested and loaded in each lane. The blot was hybridized with pMR150, which detects minor satellite sequences (
D'): Localizations of DNMT3a2 detected by antibody staining. (E') Localization of DNMT3a1. Right (A''-E''): merged image. (A) Undifferentiated ES cell carrying the EGFP-MBD-nls transgene (A) showing strong DNMT3a2 expression at centromeric heterochromatin (A'). (B) Differentiating class 2 cells (see 
