GTC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Genes to Cells (2006) 11, 1225-1237. doi:10.1111/j.1365-2443.2006.01012.x
© 2006 Blackwell Publishing or its licensors
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nimura, K.
Right arrow Articles by Kaneda, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nimura, K.
Right arrow Articles by Kaneda, Y.

Dnmt3a2 targets endogenous Dnmt3L to ES cell chromatin and induces regional DNA methylation

Keisuke Nimura1, Chisaki Ishida1, Hiroshi Koriyama1, Kenichiro Hata2, Shinya Yamanaka3, En Li4,a, Kiyoe Ura1,* and Yasufumi Kaneda1

1 Division of Gene Therapy Science, Osaka University Graduate School of Medicine, 2-2 Yamada-oka, Suita, Osaka 565-0871, Japan
2 Division of Human Genetics, Department of Integrated Genetics, National Institute of Genetics, Research Organization of Information and Systems (ROIS), 1111 Yata, Mishima, Shizuoka 411-8540, Japan
3 Department of Stem Cell Biology, Institute for Frontier Medical Science, Kyoto University, Kyoto 606-8507, Japan
4 Cardiovascular Research Center, Massachusetts General Hospital, Department of Medicine, Harvard Medical School, Charlestown, MA 02129, USA


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 References
 
DNA methylation is involved in fundamental cellular processes such as silencing of genes and transposable elements, but the underlying mechanism of regulation of DNA methylation is largely unknown. DNA methyltransferase 3-like protein (Dnmt3L), a member of the Dnmt3 family of proteins, is required during the establishment of DNA methylation patterns in germ cells. Dnmt3L does not possess enzymatic activity. Rather, in vitro analysis indicates that Dnmt3L stimulates DNA methylation by both Dnmt3a and Dnmt3b through direct binding to these proteins. In the current study, we demonstrated that in vivo, Dnmt3L physically and functionally interacted with the Dnmt3 isoform Dnmt3a2. In wild-type embryonic stem (ES) cells, but not in cells lacking Dnmt3a, endogenous Dnmt3L was concentrated in chromatin foci. In ES cells deficient in both Dnmt3a and Dnmt3b, Dnmt3L was distributed diffusely throughout the nucleus and cytoplasm, and ectopic expression of Dnmt3a2, but not Dnmt3a or Dnmt3b, restored wild-type Dnmt3L localization. We showed that endogenous Dnmt3L physically interacted with Dnmt3a2, but not Dnmt3a or Dnmt3b, in ES cells and embryonic testes. We also found that specific CpG sites were demethylated upon depletion of either Dnmt3a or Dnmt3L, but not Dnmt3b, in ES cells. These results provide evidence for a physical and functional interaction between Dnmt3L and Dnmt3a2 in the nucleus. We propose that Dnmt3a2 recruits Dnmt3L to chromatin, and induces regional DNA methylation in germ cells.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 References
 
Methylation of CpG dinucleotides is a fundamental DNA modification in all vertebrates and flowering plants and is essential for normal development (Li et al. 1992; Kakutani et al. 1996; Okano et al. 1999). DNA methylation is also critical for host genome defense, and is central to such epigenetic phenomena as the monoallelic expression of imprinted genes and dosage compensation. In mammals, DNA methylation patterns change dramatically during early development and play an important role in gene regulation, along with other epigenetic phenomena such as histone modification and ATP-dependent chromatin remodeling (Bird & Wolffe 1999; Li 2002). Considerable progress has been made recently in our understanding of the molecular pathways involved in targeting the latter two processes to particular loci in the genome. Remarkably, despite considerable effort by a large number of research groups, very little is known about the mechanisms by which specific patterns of DNA methylation are established. The problem was made significantly more complex by the identification of three distinct dimethyltransferases (Dnmts), Dnmt1, Dnmt3a and Dnmt3b, in humans and mice. The C-terminal regions of all Dnmts contain conserved methyltransferase motifs, similar to those of bacterial restriction methyltransferases. Based on the observations that Dmnt1 modifies hemi-methylated DNA more efficiently than unmethylated DNA in vitro, and is associated with nuclear replication foci, it is considered the major maintenance methyltransferase, and appears to act in concert with DNA replication. Dnmt3a and Dnmt3b are closely related proteins, and contain a highly conserved proline-tryptophan-tryptophan-proline motif (PWWP domain), a cysteine-rich zinc-binding region and a catalytic domain. They exhibit comparable activity towards unmethylated and hemi-methylated DNA substrates, and are essential for de novo methylation in postimplantation embryos and embryonic stem (ES) cells (Freitag & Selker 2005; Grace Goll & Bestor 2005).

A third member of the Dnmt3 family, Dnmt3L, lacks several conserved residues known to be involved in Dnmt enzymatic activity, and is enzymatically inactive in purified form (Aapola et al. 2000; Chedin et al. 2002). Recombinant Dnmt3L protein stimulates the DNA methylation activity of both Dnmt3a and Dnmt3b in vitro by directly binding to these enzymes (Suetake et al. 2004; Gowher et al. 2005). Dnmt3L is specifically expressed in germ cells and ES cells. A recent study showed that Dnmt3L is required for the establishment of maternal methylation-specific imprinting (Bourc’his et al. 2001; Hata et al. 2002) and for targeted de novo methylation-associated silencing of retrotransposons in male germ cells (Bourc’his & Bestor 2004; Hata et al. 2005; Webster et al. 2005). Dnmt3a, which encodes at least two isoforms, Dnmt3a and Dnmt3a2, but not Dnmt3b, is required for methylation of imprinted genes in germ cells (Kaneda et al. 2004), suggesting that there is a functional interaction between Dnmt3L and Dnmt3a. On the other hand, over-expressed Dnmt3L can interact physically and functionally with co-expressed active isoforms of both Dnmt3a and Dnmt3b (Chen et al. 2005).

To determine the mechanisms by which Dnmt3L regulates DNA methylation, we analyzed the distribution of endogenous Dnmt3L in Dnmt-deficient mouse ES cells. Here we show that Dnmt3L localized to chromatin foci which partially overlapped foci of heterochromatin, by specifically associating with a Dnmt3a isoform termed Dnmt3a2. We found that endogenous Dnmt3L physically interacted with Dnmt3a2, but not Dnmt3a or Dnmt3b, in ES cells and embryonic testes. Importantly, we observed region-specific genome demethylation upon depletion of Dnmt3a or Dnmt3L, suggesting that the specific interaction between Dnmt3L and Dnm3a2 on chromatin acts to maintain genome stability and establish sex-specific DNA methylation patterns in germ cells.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 References
 
Dnmt3L is expressed at high levels in undifferentiated ES cells and localizes to heterochromatic foci

Dnmt3L is homologous to Dnmt3a and Dnmt3b in both its N- and C-terminal domains but lacks the PWWP domain and several key residues in the catalytic domain (Fig. 1a). The Dnmt3L gene is specifically expressed in germ cells and undifferentiated ES cells (Hata et al. 2002). To examine the expression and cellular distribution of endogenous Dnmt3L, we generated an antibody directed against full-length recombinant Dnmt3L. This antibody detected a 50 kDa protein in extracts from undifferentiated ES cells, but not from differentiated ES cells (Fig. 1b). We used the same antibody to examine the localization of Dnmt3L in undifferentiated ES cells by immunofluorescence histochemistry. Dnmt3L-positive signals were clearly absent from the cytoplasm, and localized to several distinct foci in the nucleus (Fig. 1c, upper). In undifferentiated ES cells, heterochromatic foci that stained strongly with DAPI were large and diffuse compared to those in differentiated ES cells (Meshorer et al. 2006). We found that Dnmt3L-positive foci partially overlapped the diffuse DAPI-stained heterochromatic foci present in undifferentiated ES cells. The presence of diffuse DAPI-stained loci that did not also contain Dnmt3L suggested that endogenous Dnmt3L is not a general heterochromatin-binding protein. The absence of detectable antibody staining in Dnmt3L–/– ES cells confirmed the specificity of our antibody (Fig. 1c, lower). These results clearly demonstrated that endogenous Dnmt3L localizes to chromatin in ES cell nuclei, and forms Dnmt3L-specific foci that partially overlap DAPI-positive heterochromatin.


Figure 1
View larger version (28K):
[in this window]
[in a new window]
  		Figure 1  Dnmt3L expression in ES cells. (A) Schematic representation of the Dnmt family of methyltransferases in mouse. NLS, nuclear localization signal; PWWP, the PWWP domain; PHD, the cystein-rich PHD domain. Motifs I, IV, VI, IX and X correspond to conserved methyltransferase catalytic motifs. Dnmt3L lacks motif X. (B) Dnmt3L is specifically expressed in undifferentiated ES cells. Total protein prepared from 2 x 105 cells was resolved by SDS-PAGE and subjected to Western blot analysis with a polyclonal rabbit antiserum raised against recombinant His-tagged full-length Dnmt3L. Undiff, undifferentiated wild-type ES cells; Diff, differentiated ES cells; Dnmt3L–/–, undifferentiated Dnmt3L–/–(C) Cellular localization of endogenous Dnmt3L in ES cells. Wild-type and Dnmt3L–/– ES cells were immunostained using anti-Dnmt3L antibody. DNA was counterstained with DAPI. In the merged images, intense Dnmt3L-labeled foci in the nucleus partially overlap diffusely stained DAPI-positive foci. ES cells. The anti-Dnmt3L antibody recognized a protein of the expected molecular weight, approximately 50 kDa, in extracts from undifferentiated ES cells. The membrane was probed with an antibody against histone H3 as an internal control.

 
Dnmt3L nuclear localization depends on Dnmt3a and Dnmt3b

As Dnmt3L does not have methyltransferase activity, we examined whether it contributed to methylation of specific loci indirectly, i.e. by recruiting an enzymatically active isoform of Dnmt to chromatin. To investigate the putative in vivo interaction between Dnmt3L and other Dnmts, we examined the localization of Dnmt3L in Dnmt1–/–, Dnmt3a–/– Dnmt3b–/–, Dnmt3a–/– and Dnmt3b–/– ES cells by immunofluorescence using our anti-Dnmt3L antibody. Nuclear localization of Dnmt3L was unaffected in Dnmt1-deficient cells. Similar to wild-type ES cells, Dnmt3L localized to the nucleus, forming several strongly staining Dnmt3L-specific foci in Dnmt1–/– ES cells (Figs 1C and 2A). In striking contrast, Dnmt3L was diffusely distributed in Dnmt3a–/– Dnmt3b–/– ES cells, with near-homogeneous distribution both in the cytoplasm and in the nucleus (Fig. 2A). The altered pattern of cytoplasmic localization Dnmt3L in these cells suggested that Dnmt3L failed to enter the nucleus efficiently, or was rapidly exported from the nucleus. When we analyzed the localization of Dnmt3L in Dnmt3a- or Dnmt3b-null mutant ES cells, we found that endogenous Dnmt3L localized predominantly to the nucleus when either gene was expressed, in agreement with previous results using ectopically expressed, epitope-tagged Dnmt3L (Hata et al. 2002). Interestingly, we found that the distribution of Dnmt3L in Dnmt3a–/– ES cells was different from that in Dnmt3b–/– ES cells. The accumulation of Dnmt3L in specific foci was readily observed in Dnmt3b–/– ES cells. In contrast, the pattern of Dnmt3L-positive foci was impaired by disruption of Dnmt3a. We observed an average of 6.5–8.0 Dnmt3L-positive foci in wild-type, Dnmt1–/–, and Dnmt3b–/– ES cells. In contrast, few Dnmt3L-specific foci were observed in Dnmt3a–/– ES cells (Fig. 2B). These results suggested that endogenous Dnmt3L interacted preferentially with Dnmt3a at specific foci in the nucleus of ES cells.


Figure 2
View larger version (27K):
[in this window]
[in a new window]
  		Figure 2  Dnmt3a and Dnmt3b affect the nuclear localization of Dnmt3L. (A) Indirect immunofluorescence analysis of Dnmt1–/– (c/c) (Li et al. 1992), Dnmt3a–/–, and Dnmt3b–/– single mutant ES cells, and Dnmt3a/Dnmt3b double mutant ES cells (Okano et al. 1999) using anti-Dnmt3L antibody. Endogenous Dnmt3L was detected using a polyclonal Dnmt3L antibody (red), and cells were counterstained with DAPI to visualize the DNA (blue). Merge represents a composite image of the two staining patterns. (B) Average number of Dnmt3L-positive foci per nucleus in wild-type and single Dnmt3 mutant ES cells. At least 30 nuclei from each set of cells were analyzed. The average number of foci decreased significantly upon depletion of Dnmt3a, but not Dnmt3b (P < 0.005).

 
Dnmt3a2, but not Dnmt3a or Dnmt3b, forms Dnmt3L specific foci

In undifferentiated ES cells, the level of expression of Dnmt3a2, which lacks the N-terminal 219 amino acids of Dnmt3a, is higher than Dnmt3a (Chen et al. 2002). To confirm that nuclear localization of Dnmt3L was mediated specifically by Dnmt3 proteins, we expressed EGFP-tagged Dnmt3a, Dnmt3a2 and Dnmt3b in Dnmt3a–/– Dnmt3b–/– ES cells, and monitored Dnmt3L localization by immunofluorescence. We found intense Dnmt3L-positive multiple foci in the nucleus of ES cells only in the presence of ectopic Dnmt3a2 (Fig. 3A). Expression of Dnmt3a or Dnmt3b resulted in nuclear localization of Dnmt3L; however, intense Dnmt3L-labeled multiple foci, such as those seen in wild-type ES cells, were not detected. We observed an average of five intense Dnmt3L-positive foci in Dnmt3a–/– Dnmt3b–/– ES cells expressing Dnmt3a2 (Fig. 3B), while expression of Dnmt3a or Dnmt3b failed to result in the formation of a significant number of Dnmt3L-positive foci in the nucleus. The distribution of all exogenous EGFP-tagged Dnmt3 enzymes was similar to the nuclear localization pattern of endogenous Dnmt3L in Dnmt3a–/– Dnmt3b–/– ES cells (Fig. 3A). These results suggested that all forms of Dnmt3 are involved in nuclear localization of Dnmt3L, but localization of endogenous Dnmt3L to specific foci in the nuclei of undifferentiated ES cells is mediated by Dnmt3a2.


Figure 3
View larger version (36K):
[in this window]
[in a new window]
  		Figure 3  Dnmt3a2 targets Dnmt3L to chromatin foci. (A) The indicated EGFP-Dnmt3 fusion proteins or EGFP alone were expressed in Dnmt3a/Dnmt3b double mutant ES cells by transient transfection. ES cells were immunostained with anti-Dnmt3L antibody (red), counterstained with DAPI (blue), and Dmnt3 enzymes were visualized directly (green). Merge represents a composite image of the three staining patterns. (B) Average number of Dnmt3L-positive foci per nucleus in Dnmt3a/Dnmt3b double mutant ES cells expressing the indicated EGFP-Dnmt3 fusion proteins. A minimum of 50 nuclei from each set of cells was analyzed. The average number of foci increased significantly upon Dnmt3a2 expression, compared to Dnmt3a, Dnmt3b or control GFP expression (P < 0.0001). (C) EGFP-Dnmt3a2 was expressed in wild-type and Dnmt3L deficient ES cells by transient transfection. Average number of Dnmt3a2-labeled foci per nucleus is shown as a bar graph. A minimum of 50 nuclei from each set of cells was analyzed.

 
It should be noted that epitope-tagged Dnmt3 proteins localized differently depending on the cell type. In NIH3T3 cells, which normally do not express Dnmt3L, exogenous Dnmt3a and Dnmt3b localized to heterochromatin, and Dnmt3a2 was present in a diffuse pattern distinct from heterochromatin (Hata et al. 2002; Chen et al. 2004). We observed similar patterns of localization for each of our Dnmt3 constructs in NIH3T3 cells, with the exception of EGFP-Dnmt3a2 (Supplementary Fig. S1). When EGFP-Dnmt3a2 was ectopically expressed in NIH3T3 cells, we observed heterochromatic localization, as well as a diffuse pattern of nuclear staining. We found no evidence of this diffuse pattern of nuclear Dnmt3a2 staining in undifferentiated ES cells.

We next examined the effect of Dnmt3L on the localization of Dnmt3a2, by expressing EGFP-tagged Dnmt3a2 in Dnmt3L–/– ES cells. We observed intense, GFP-positive nuclear foci that partially overlapped heterochromatic foci when EGFP-Dnmt3a2 was ectopically expressed in Dnmt3L-deficient cells, or in wild-type ES cells (Fig. 3C). This result suggested that there are chromatin-targeting factors for Dnmt3a2 in ES cells, but not in NIH3T3 cells, and that Dnmt3L and Dnmt3a2 are part of a hierarchy of intranuclear interactions, the function of which is, in part, to recruit Dnmt3L to chromatin.

Specific interaction between Dnmt3L and Dnmt3a2 in ES cells

Dnmt3L can bind directly to both Dnmt3a and Dnmt3b in vitro (Suetake et al. 2004; Gowher et al. 2005). One mechanism for localization of Dnmt3L to specific foci in vivo in ES cells may involve preferential interaction of Dnmt3L with Dnmt3a2. To examine whether endogenous Dnmt3L physically interacted with any of the Dnmt3 enzymes, we fractionated ES cells and performed immunoprecipitation analysis using our anti-Dnmt3L antibody. Consistent with results from immunofluorescence cell staining, we detected a very low level of Dnmt3L in the soluble cytoplasmic fraction of wild-type ES cells (Figs 1C and 4A, upper panel). Dnmt3L proteins were extracted from nuclear pellets after digestion of chromosomal DNA with DNase I in the presence of increasing concentrations of NaCl; Dnmt3a2, Dnmt3b and core histone H3 were similarly extracted. We were unable to detect Dnmt3a using an antibody that was specific for both Dnmt3a and Dnmt3a2. In contrast to wild-type ES cells, Dnmt3L was present in the cytoplasmic fraction from Dnmt3a–/– Dnmt3b–/– ES cells (Fig. 4A, middle panel). The level of Dnmt3a2 or Dnmt3b in the nucleus was unaffected in Dnmt3L deficient cells (Fig. 4A, lower panel). These results provided evidence that stable association of Dnmt3L with chromatin was dependent on the presence of Dnmt3a2 and Dnmt3b. To gain further insight into the nature of the interaction between Dnmt3L with the different Dnmt3 enzymes, we examined the binding of Dnmt3L to Dnmt3a2 and Dnmt3b in vivo. High salt nuclear extracts were prepared from wild-type and Dnmt3 single mutant ES cells, and subjected to immunoprecipitation using anti-Dnmt3L antibody. The immune complexes were then analyzed by Western blot using anti-Dnmt3a and anti-Dnmt3b antibodies. We found that Dnmt3a2 was present in Dnmt3L immune complexes isolated from wild-type ES cells (Fig. 4B, upper), and Dnmt3b–/– cells, but not from Dnmt3a–/– or Dnmt3L–/– cells. In contrast, we were unable to detect a significant level of interaction between Dnmt3L and Dnmt3b in wild-type ES cells, even though Dnmt3b was highly expressed (Fig. 4B, lower), nor was there a significant level of interaction between Dnmt3L and Dnmt3b in nuclear extracts prepared from Dnmt3a–/– ES cells. The very faint signal corresponding to Dnmt3b in anti-Dnmt3L immune complexes from Dnmt3a–/– ES cells was also observed in Dnmt3L–/– ES cells, indicating that it was probably a background signal (Fig. 4B, lower). Although we cannot rule out a weak interaction between Dnmt3L and Dnmt3b in the nucleus, our biochemical data, together with the results from immunocytochemistry, suggested that Dnmt3L associates specifically with Dnmt3a2 in the nucleus of ES cells. These results strongly implicate Dnmt3a2 as a targeting factor for Dnmt3L chromatin localization, enabling subsequent activation of regional DNA methylation.


Figure 4
View larger version (57K):
[in this window]
[in a new window]
  		Figure 4  Dnmt3L interacts with Dnmt3a2, but not Dnmt3b, in ES cells. (A) Dnmt3L physically associated with chromatin in the presence of Dnmt3a and Dnmt3b. Soluble cytoplasmic proteins and nuclei from wild-type, Dnmt3a/Dnmt3b double mutant, or Dnmt3L-null ES cells were isolated using 0.15% NP-40 extraction. Isolated nuclei were treated with DNase I, then with increasing concentrations of NaCl (100, 200, 300, 400 and 500 mM) to separate nuclear extract from nuclear debris. Equivalent aliquots were as separated by SDS-PAGE and immunoblotted with the indicated antibodies. (B) Co-immunoprecipitation analysis revealed a specific interaction between Dnmt3L and Dnmt3a2. Nuclear extracts from the indicated ES cells were treated with 500 mM NaCl and subjected to immunoprecipitation using anti-Dnmt3L antibody. Immune complexes were collected using Protein A Sepharose, separated by SDS-PAGE, and analyzed by Western blot using monoclonal antibodies against Dnmt3a and Dnmt3b.

 
DNA methylation at specific CpG sites in ES cells requires both Dnmt3a2 and Dnmt3L but not Dnmt3b

To understand the functional significance of the interaction of Dnmt3L and Dnmt3a2, we examined DNA methylation status in various Dnmt3-deficient ES cells, focusing on the Dnmt3b gene locus. Dnmt3b consists of 24 exons spanning 38 kb (Ishida et al. 2003). Genomic DNA from wild-type, Dnmt3a- and Dnmt3b-deficient ES cells was digested to completion with XbaI and the methylation-sensitive restriction enzyme, PmlI, then analyzed by Southern blot hybridization. Four probes were used to investigate DNA methylation status at nine CpG sites over the length of the Dnmt3b locus (Fig. 5A). As shown in Fig. 5B, methylation of most of the CpG sites in Dnmt3b gene locus was drastically reduced in Dnmt3a–/– Dnmt3b–/– ES compared to wild-type, as indicated by the increased sensitivity to PmlI. In contrast, DNA methylation status was preserved in Dnmt3a- or Dnmt3b-null ES cells, suggesting that Dnmt3a and Dnmt3b are largely functionally redundant at the Dnmt3b gene locus in ES cells. Curiously, one CpG site, at nucleotide (nt) –811 relative to the transcriptional start site of Dnmt3b (Ishida et al. 2003) was almost completely demethylated in Dnmt3a–/– ES cells, as in Dnmt3a–/– Dnmt3b–/– ES cells, whereas no significant demethylation was observed at this site in Dnmt3b–/– ES cells (Fig. 5b). Dnmt3a2 is the major isoform expressed in undifferentiated ES cells. These results strongly suggest that Dnmt3a2 is required for stable inheritance of DNA methylation at this particular CpG site at the Dnmt3b gene locus.


Figure 5
View larger version (54K):
[in this window]
[in a new window]
  		Figure 5  Inactivation of Dnmt3a or Dnmt3L but not Dnmt3b results in demethylation of specific CpG sites in the Dnmt3b gene locus in ES cells. (A) Genomic structure of Dnmt3b and the probes used for Southern hybridization. Exons are shown as filled boxes; the 5' CpG island is represented by an open box; Restriction enzyme sites are abbreviated as follows: X, XbaI; P, PmlI. (B) Genomic DNA from Dnmt3a–/–, Dnmt3b–/–, and Dnmt3a/Dnmt3b double mutant ES cells, as well as wild-type ES cells, was digested with the methylation-insensitive enzyme XhaI and the methylation-sensitive enzyme PmlI. The same blot was hybridized successively with probes 1–4. The location of each probe is indicated in (A). Methylation of the 5' flanking region of the CpG island of Dnmt3b (probe 2) was lost in Dnmt3a–/– but not in Dnmt3b–/– ES cells. (C) Analysis of the DNA methylation status of the 5' flanking region of the CpG island of Dnmt3b (nt –875–736) by bisulfite sequencing. Genomic DNA from wild-type, Dnmt3a–/–, Dnmt3b–/–, and Dnmt3L–/– undifferentiated ES cells was analyzed. Circles represent individual CpG dinucleotides. • methylated; {circ} unmethylated CpG site.

 
We next analyzed the DNA methylation status of the region upstream of Dnmt3b (nt –875 to –736 relative to the transcriptional start site) in each Dnmt3b-deficient ES cell line using bisulfite genomic sequencing. In Dnmt3a–/– Dnmt3b–/– ES cells, DNA methylation was completely depleted at this upstream region of Dnmt3b. We found that methylation of two CpG sites, including the PmlI site, was considerably decreased in Dnmt3a-, but not Dnmt3b-null cells (Fig. 5c). We found these two CpG sites were also demethylated in Dnmt3L-deficient ES cells (Fig. 5c). Dnmt3L itself has no DNA methyltransferase activity but stimulates the DNA methylation activity of Dnmt3 enzymes in vitro. In light of our earlier data on the physical interaction between Dnmt3L and Dnm3a2 (Fig. 4B), these results suggest that the preferential interaction of Dnmt3L with Dnmt3a2 in the nucleus induces regional DNA methylation in ES cells.

Dnmt3L interacts with Dnmt3a2 but not Dnmt3a or Dnmt3b in embryonic testes

Dnmt3L is specifically expressed in germ cells (Sakai et al. 2004). To investigate the localization of Dnmt3L in germ cells, cytological sections from embryonic day 16.5 (E16.5) testis were stained with our anti-Dnmt3L antibody. Chromatin was counterstained with DAPI. As shown in Fig. 6A, the size the nucleus in gonocytes that expressed Dnmt3L was larger than in surrounding gonadal somatic cells. Weak, homogeneous DAPI staining was observed, but no heterochromatic foci in the nuclei of gonocytes. We observed intense Dnmt3L-positive staining in the nucleus of gonocytes. To examine whether there was a specific molecular interaction between Dnmt3L and any of the Dnmt3 enzymes in testes, we performed immunoprecipitation analysis of the whole cell extracts prepared from E16.5 testes using our anti-Dnmt3L antibody. Consistent with what was observed for ES cells (Fig. 4B), immunoblot analysis revealed a specific interaction between Dnmt3L and Dnmt3a2 in the testis, although both Dnmt3a and Dnmt3b were expressed (Fig. 6B). In gonocytes, the level of expression of Dnmt3a2 is higher than Dnmt3a, and there are significant levels of Dnmt3b at this stage as well (Sakai et al. 2004). Presumably, the lack of a detectable interaction between Dnmt3L and Dnmt3a reflects the exclusive expression patterns of these two Dnmt3 proteins. Our results strongly suggested that Dnmt3L selectively interacts with Dnmt3a2 but not Dnmt3b in gonocytes to induce regional DNA methylation.


Figure 6
View larger version (43K):
[in this window]
[in a new window]
  		Figure 6  Dnmt3L specifically interacts with Dnmt3a2 in embryonic gonocytes. (A) Localization of endogenous Dnmt3L in testes. Cross-section of E16.5 wild-type testis stained with hematoxylin and eosin, and with the anti-Dnmt3L antibody. Sections were counterstained with DAPI to visualize the DNA. In gonocytes, entire nuclei were weakly and homogeneously positive for DAPI, compared to surrounding somatic cells. Merged images indicated that Dnmt3L localization did not precisely overlap DAPI-positive staining. Scale bars, 10 µm. (B) Dnmt3L associates with Dnmt3a2, but not with Dnmt3a or Dnmt3b, in embryonic testes. Whole cell extracts from E16.5 embryonic testes were subjected to immunoprecipitation using anti-Dnmt3L antibody, and analyzed as for Fig. 4B. (C) Model for the induction of regional DNA methylation through Dnmt3a2-mediated recruitment of Dnmt3L to chromatin. Dnmt3a and Dnmt3b also interact with Dnmt3L in the cytoplasm and mediate nuclear transport of Dnmt3L. In the nucleus, Dnmt3b dissociates from Dmnt3L during stable association with chromatin. Dnmt3a2 is targeted to particular chromatic regions by as yet unidentified tissue specific factor(s).

 

   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 References
 
In this study, we provide evidence for a physical and functional interaction between endogenous Dnmt3L and Dnmt3a2 in the nucleus of ES cells. Dnmt3L, which is expressed specifically in germ cells and undifferentiated ES cells, is essential for appropriate gene regulation in germ cells (Bourc’his et al. 2001; Hata et al. 2002). Recent studies suggest that Dnmt3L functions to target DNA methylation to specific genomic regions, i.e. imprinted gene loci (Bourc’his & Bestor 2004; Hata et al. 2005). Since Dnmt3L itself lacks enzymatic activity, it is believed to function by interacting with other Dnmt enzymes (Hata et al. 2002). We showed that endogenous Dnmt3L localizes to the nucleus and forms a specific pattern of foci that partially overlap DAPI-dense heterochromatin in ES cells (Fig. 1C). This pattern of foci consisted of a number (6.5 per nucleus) of clearly delineated Dnmt3L-specific foci in undifferentiated ES cells. The focal localization of Dnmt3L was impaired when Dnmt3a was disrupted, and in Dnmt3a–/– Dnmt3b–/– ES cells, expression of the short isoform of Dnmt3a, Dnmt3a2, restored focal staining of Dnmt3L (Figs 2 and 3). In the reciprocal analysis, nuclear localization of Dnmt3a2 was unaffected in Dnmt3L-deficient cells (Fig. 3C). In ES cells, we detected Dnmt3a2 and Dnmt3b but not Dnmt3a by Western blotting (Fig. 4). These findings suggest that endogenous Dnmt3L is recruited to specific nuclear regions via an interaction with Dnmt3a2. We showed that Dnmt3L preferentially associated with Dnmt3a2, but not Dnmt3a or Dnmt3b in ES cells and testes (Figs 4B and 6B). Furthermore, our analysis of the methylation status of the Dnmt3b gene locus showed that methylation of at least two specific CpG sites were almost completely abolished in Dnmt3a- or Dnmt3L-null ES cells, but not in Dnmt3b-null cells (Fig. 5), suggesting that the specific interaction we observed between Dnmt3a2 and Dnmt3L is functionally relevant for region-specific DNA methylation.

Dnmt3L associates in vitro with the C-terminal domains of both Dnmt3a and Dnmt3b through its C-terminal domain, and stimulates the DNA methylation activity of both enzymes (Suetake et al. 2004; Gowher et al. 2005). Over-expression of Dnmt3L stimulates de novo methylation in the presence of active isoforms of human DNMT3A and DNMT3B (Chen et al. 2005). Our data, which showed that endogenous Dnmt3L localized to the nucleus in a Dmnt3-dependent manner, and co-localized with exogenously expressed Dnmt3 enzymes in Dnmt3a–/– Dnmt3b–/– ES cells (Fig. 3a), is important evidence that the direct interaction seen in vitro also occurs in vivo, and that the function of Dnmt3L in regulating focal methylation involves a direct interaction between these proteins. We found that within the nucleus, Dnmt3L interacted with Dnmt3a2, but not Dnmt3a or Dnmt3b in wild-type ES cells and embryonic testes, as determined by co-immunoprecipitation analysis. This indicates that although Dnmt3a, Dnmt3a2 and Dnmt3b have similar enzymatic properties in vitro, only Dnmt3a2 interacts stably with Dnmt3L in vivo. One of the reasons that we did not observe an interaction between Dnmt3L and Dnmt3a may be due to the exclusive expression patterns of these two Dnmt3 proteins (Fig. 4) (Sakai et al. 2004). In contrast, Dnmt3b was expressed in ES nuclei, similar to Dnmt3L and Dnmt3a2. It should be noted that the localization of Dnmt3L differed between Dnmt3a–/– ES cells and Dnmt3a–/– Dnmt3b–/– ES cells, suggesting that Dnmt3b is involved in nuclear localization of Dnmt3L. Based on our results, we propose the following model. Dnmt3L is actively transferred into the nucleus via an interaction with several Dnmt3 enzymes. Once in the nucleus, the various Dnmt3 enzymes recognize distinct chromatin structures or factors, thereby targeting different genomic regions. During stable incorporation of Dnmt3b into chromatin, Dnmt3L is transferred from endogenous Dnmt3b to other chromatin proteins. In this model, over-expression of epitope-tagged Dnmt3b would be able to maintain its binding with Dnmt3L (Chen et al. 2005). When Dnmt3a2 binds to chromatin, its interaction with Dnmt3L is maintained, inducing regional active DNA methylation. Differences in the nuclear localization of Dnmt3a2 in ES cells and in NIH3T3 cells (Fig. 3a and Supplementary Fig. S1) suggests the existence of ES cell-specific factors that target Dnmt3a2 to designated genomic regions. In our model, the interaction between Dnmt3L and other Dnmt3 isoforms in the nucleus is somewhat more complex than what is suggested by observations of the behavior of these proteins in vitro (Fig. 6c).

Recent studies demonstrated that Dnmt3a and Dnmt3b have distinct functions on chromatin (Takeshima et al. 2006). Dnmt3b is specifically required for DNA methylation of centromeric minor satellite repeats (Li et al. 1992; Kakutani et al. 1996; Okano et al. 1999). We showed that methylation of specifc CpG sites at the Dnmt3b gene locus decreased in ES cells depleted of either Dnmt3a or Dnmt3L (Fig. 5). These results demonstrate that each Dnmt3 enzyme is targeted to different genomic regions in the nucleus. During murine development, the Dnmt3L gene is specifically expressed during the time that new DNA methylation patterns are acquired in the male and female germ lines. Recently, Dnmt gene disruption studies suggested a functional link between Dnmt3L and Dnmt3a during germ cell development and de novo DNA methylation (Hata et al. 2002; Kaneda et al. 2004). Consistent with these in vivo studies, we demonstrated a specific nuclear interaction between endogenous Dnmt3L and Dnmt3a2 that correlated with the co-localized spatial and temporal expression of the two proteins (Hata et al. 2002; La Salle et al. 2004; Sakai et al. 2004). While the molecular mechanisms by which Dnmt3L specifically recognizes Dnmt3a2 in chromatin remain to be elucidated, our findings provide strong evidence that stable binding of Dnmt3L to Dnmt3a2 induces de novo DNA methylation at specific genomic loci, and that this interaction may be required for genomic stability and the establishment of methylation-dependent imprinting in germ cells.


   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
References
 
Plasmid construction

Full-length Dnmt3L and Dnmt3b cDNAs were amplified by PCR from ES cells. The full-length Dnmt3A1 cDNA (GENBANK: BF122396 [GenBank] ) was purchased from Invitrogen. The Dnmt3L coding region was amplified and cloned into pDEST17 (Invitrogen) to generate recombinant His-tagged Dnmt3L. To express Dnmt-EGFP fusion proteins in ES cells, the coding regions of Dnmt3a1, Dnmt3a2, Dnmt3b and Dnmt3L were amplified and subcloned into pCAGIP-EGFP-gw using Gateway Technology (Invitrogen). A control expression vector was generated containing the EGFP coding region from pEGFP-N1 (Clontech). The following primers were used for PCR:

Dnmt3L-gw-s: 5'-AAAAAGCAGGCTCAATGGGTTCCCGGGAGACACC-3', Dnmt3L-gw-as: 5'-AGAAAGCTGGGTCCTGGTTCTAGGAAAGACTT-3', Dnmt3b-gw-s: 5'-AAAAAAGCAGGCTTTCAGGAAACAATGAAGGGA-3', Dnmt3b-gw-as: 5'-AGAAAGCTGG GTAG AACTATTCACAGGCAAA-3', Dnmt3a2-gw-s: 5'-CACCATGAATGCTGTGGAAGAGAACCAG-3', Dnmt3a2-gw-as: 5'-AGAAAGCTGGGTTTAACTTTTTTATCATCAC-3'.

Antibodies

Recombinant His-tagged Dnmt3L was expressed in E. coli, and purified according to the manufacturer's instructions (Qiagen). Purified, recombinant Dnmt3L was then used to immunized two rabbits and raise anti-Dnmt3L polyclonal antiserum. Anti-Dnmt3A and anti-Dnmt3B monoclonal antibodies were obtained from Imgenex. Anti-H3 polyclonal antibody was obtained from Abcam.

Cell culture

Wild-type J1 and mutant ES cells were maintained without feeder cells in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum, 0.1 mM nonessential amino acids (Invitrogen), 0.1 mM ß-mercaptoethanol and LIF (Chen et al. 2003). ES cells were induced to differentiate by treatment with retinoic acid for 5 days.

Western blot analysis

Proteins were separated on a 4–20% by SDS polyacrylamide gradient electrophoresis and transferred to a PVDF membrane (Millipore). Membranes were incubated in 5% skim milk (for anti-Dnmt3L) or 3% skim milk (for anti-Dnmt3A and anti-Dnmt3B) in wash buffer (2 mM Tris-Cl, pH 8.0, 0.02% NaCl, and 0.05% Tween20) for 1 h. Membranes were then incubated with the indicated antibodies at dilutions of 1: 2000 (anti-Dnmt3L) or 1 : 250 (anti-Dnmt3A and anti-Dnmt3B) in blocking buffer for 1 h at room temperature. Membranes were rinsed in wash buffer, and primary antibodies were detected with horseradish peroxidase-conjugated anti-rabbit, anti-mouse (Amersham Biosciences) or anti-sheep (Jackson ImmunoResearch) secondary antibodies.

Immunofluorescence analysis

ES or NIH3T3 cells were cultured on collagen type I-coated glass coverslips, then incubated with 4% paraformaldehyde in PBS for 5 min, permeabilized in PBS containing 0.5% Triton X-100 for 10 min, and rinsed in PBS. Slides were incubated with 5% skim milk in PBS for 30 min at room temperature, then incubated with anti-Dnmt3L antibody (diluted 1 : 100) overnight at 4 °C. To visualize the primary antibody, Alexa 546-conjugated goat anti-rabbit IgG antibody (diluted 1 : 300, Molecular Probes) was incubated with the cells for 30 min at room temperature. All antibodies were diluted in 5% skim milk/PBS. Preparations were rinsed in PBS, and 0.5 mg/mL DAPI (Sigma) was added for DNA staining.

Testes were dissected from embryos (E16.5) and incubated in 4% paraformaldehyde. After dehydration, the pellets were embedded in Tissue-Tec OCT Compound (Sakura Finetek Japan), frozen on dry ice, and stored at –80 °C. For immunofluorescence staining, 5-µm-thick sections were incubated with anti-Dnmt3L antibody 1 : 100 dilution) overnight at 4 °C. Subsequently, sections were incubated with Alexa 546-conjugated goat anti-rabbit IgG antibody (1 : 250, Molecular Probes) and 2.5 mg/mL DAPI. All antibodies were diluted in PBS.

Samples were mounted in Vectorshield (Vector Laboratories Inc.). A confocal laser scanning microscope, Radiance 2100 (Bio-Rad), was used for image acquisition.

Protein expression and analysis

Plasmids encoding EGFP-Dnmt3 proteins were transiently transfected into Dnmt3a–/– Dnmt3b–/– double knockout ES cells (Chen et al. 2003) using Lipofectamine 2000 reagent (Invitrogen). After 24 h, the cells were fixed and observed by fluorescence microscopy, as described above.

Cell fractionation

Cells were fractionated as previously described (Ge et al. 2004) with minor modifications. ES cells (3.4 x 106) were extracted in 500 µL nuclear isolation buffer (10 mM Tric-Cl, pH 7.5, 60 mM KCl, 15 mM NaCl, 1.5 mM MgCl2, 1 mM CaCl2, 0.25 M Sucrose, 10% glycerol, 1 mM DTT) containing 0.15% NP-40 and EDTA-free "complete" protease inhibitor cocktail (Roche) on ice. Cytoplasmic soluble proteins were separated from nuclei by centrifugation at 20 000 g for 5 min. Nuclear pellets were further treated with 0.28 units/µL DNase I (Takara) in nuclear isolation buffer with increasing amounts of NaCl (100, 200, 300, 400 and 500 mM) at 25 °C for 15 min. After incubation on ice for 10 min, 10 mM EDTA was added, and the sample was incubated on ice for 10 min. Nuclear extracts were separated from pellets by centrifugation.

Co-immunoprecipitation analysis

Nuclear extracts were prepared from ES cells with nuclear isolation buffer containing 500 mM NaCl as described above. Wild-type embryos (E16.5) were removed from pregnant females and the testes were removed. Isolated testes were homogenized in nuclear isolation buffer containing 0.15% NP-40 and 500 mM NaCl. Nuclear extracts were pretreated with mouse IgG AC (Santa Cruz Biotechnology), then incubated with the anti-Dnmt3L antibody at 4 °C overnight. Protein A Sepharose beads (Amersham Bioscience) were added to the sample to purify immune complexes. The beads were washed 3 times with nuclear isolation buffer containing 500 mM NaCl and 0.15% NP-40. Purified proteins were resolved by 4–20% SDS-PAGE and subjected to Western blot analysis using anti-Dnmt3a and -Dnmt3b antibodies.

DNA methylation analysis by Southern hybridization

Genomic DNA isolated from various ES cell lines was digested with methylation-sensitive PmlI and methylation-insensitive XbaI (100 units each enzyme per 10 µg DNA). A second aliquot of each enzyme (equal to the first) was added after 2 h to each sample. Following phenol/chloroform/isoamyl alcohol (25 : 24 : 1) extraction and ethanol precipitation, genomic DNA was resuspended in TE buffer. For complete digestion, the DNA subjected to digestion again using the same enzymes. Purified DNA swas separated by electrophoresis (10 µg DNA per lane) on a 1.25% agarose gel and transferred to a Hybond-XL nylon membrane. The membrane was incubated with a random-primed radiolabelled genomic DNA probe in hybridization solution (Toyobo) at 70 °C for 12–16 h, washed in 2XSSC and 0.1% SDS at 70 °C, then subjected to autoradiograpy at –80 °C. The probes used in Fig. 5A were as follows: probe 1, BamHI-NheI fragment of the 5' upstream region of genomic Dnmt3b (accession AF068626 [GenBank] ); probe 2, XbaI-PmlI fragment of the region flanking the 5' CpG island of genomic Dnmt3b; probe 3, PvuII-PvuII fragment of the middle portion of the Dnmt3b cDNA; probe 4, ScaI-ScaI fragment of the 3' downstream region of the Dnmt3b cDNA.

Bisulfite genomic sequencing

Bisulfite genomic sequencing was carried out as previously described (Herman et al. 1996). Genomic DNA was digested using XbaI and BamHI prior to sodium bisulfite treatment. Purified DNA was subjected to gene-specific PCR amplification using Taq polymerase (Roche). The primer set used to amplify the Dnmt3benhancer was F1 (5'-AGTGAAGTCAGGGAGCCAACAAAARAAAAG-3') and R1 (5'-ACTAAATAAATATATGTTAAAGGGATAAAC-3'), and the top strand of the 140 bp fragment (nt –875~ –736) containing five CpG sites was analyzed (Ishida et al. 2003). The PCR conditions were 95 °C for 9 min, followed by 40 cycles of 94 °C for 20 s, 52 °C for 20 s, 72 °C for 30 s, and 72 °C for 5min. The amplified PCR products were subcloned into pGEM (Promega) and sequenced.


   Acknowledgements
 
We thank Dr F. Urnov for valuable comments on the manuscript, Drs P. Vertino and S. Miyagawa for technical advice and members of our laboratory for support and discussion. This work was supported by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan and the Inamori Foundation.


   Footnotes
 
Communicated by: Fuyuki Ishikawa

aPresent address: Novartis Institute for Biomedical Research, Cambridge, MA 02139, USA Back

* Correspondence: E-mail: kiyoeura{at}gts.med.osaka-u.ac.jp


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Aapola, U., Kawasaki, K., Scott, H.S., Ollila, J., Vihinen, M., Heino, M., Shintani, A., Kawasaki, K., Minoshima, S., Krohn, K., Antonarakis, S.E., Shimizu, N., Kudoh, J. & Peterson, P. (2000) Isolation and initial characterization of a novel zinc finger gene, DNMT3L, on 21q22.3, related to the cytosine-5-methyltransferase 3 gene family. Genomics 65, 293–298.[CrossRef][Medline]

Bird, A.P. & Wolffe, A.P. (1999) Methylation-induced repression —belts, braces, and chromatin. Cell 99, 451–454.[CrossRef][Medline]

Bourc’his, D. & Bestor, T.H. (2004) Meiotic catastrophe and retrotransposon reactivation in male germ cells lacking Dnmt3L. Nature 431, 96–99.[CrossRef][Medline]

Bourc’his, D., Xu, G.L., Lin, C.S., Bollman, B. & Bestor, T.H. (2001) Dnmt3L and the establishment of maternal genomic imprints. Science 294, 2536–2539.[Abstract/Free Full Text]

Chedin, F., Lieber, M.R. & Hsieh, C.L. (2002) The DNA methyltransferase-like protein DNMT3L stimulates de novo methylation by Dnmt3a. Proc. Natl. Acad. Sci. USA 99, 16916–16921.[Abstract/Free Full Text]

Chen, Z.X., Mann, J.R., Hsieh, C.L., Riggs, A.D. & Chedin, F. (2005) Physical and functional interactions between the human DNMT3L protein and members of the de novo methyltransferase family. J. Cell. Biochem. 95, 902–917.[CrossRef][Medline]

Chen, T., Tsujimoto, N. & Li, E. (2004) The PWWP domain of Dnmt3a and Dnmt3b is required for directing DNA methylation to the major satellite repeats at pericentric heterochromatin. Mol. Cell. Biol. 24, 9048–9058.[Abstract/Free Full Text]

Chen, T., Ueda, Y., Dodge, J.E., Wang, Z. & Li, E. (2003) Establishment and maintenance of genomic methylation patterns in mouse embryonic stem cells by Dnmt3a and Dnmt3b. Mol. Cell. Biol. 23, 5594–5605.[Abstract/Free Full Text]

Chen, T., Ueda, Y., Xie, S. & Li, E. (2002) A novel Dnmt3a isoform produced from an alternative promoter localizes to euchromatin and its expression correlates with active de novo methylation. J. Biol. Chem. 277, 38746–38754.[Abstract/Free Full Text]

Freitag, M. & Selker, E.U. (2005) Controlling DNA methylation: many roads to one modification. Curr. Opin. Genet. Dev. 15, 191–199.[CrossRef][Medline]

Ge, Y.Z., Pu, M.T., Gowher, H., Wu, H.P., Ding, J.P., Jeltsch, A. & Xu, G.L. (2004) Chromatin targeting of de novo DNA methyltransferases by the PWWP domain. J. Biol. Chem. 279, 25447–25454.[Abstract/Free Full Text]

Gowher, H., Liebert, K., Hermann, A., Xu, G. & Jeltsch, A. (2005) Mechanism of stimulation of catalytic activity of Dnmt3A and Dnmt3B DNA-(cytosine-C5)-methyltransferases by Dnmt3L. J. Biol. Chem. 280, 13341–13348.[Abstract/Free Full Text]

Grace Goll, M. & Bestor, T.H. (2005) Eukaryotic cytosine methyltransferases. Annu. Rev. Biochem. 74, 481–514.[CrossRef][Medline]

Hata, K., Kusumi, M., Yokomine, T., Li, E. & Sasaki, H. (2005) Meiotic and epigenetic aberrations in Dnmt3L-deficient male germ cells. Mol. Reprod. Dev. 73, 116–122.

Hata, K., Okano, M., Lei, H. & Li, E. (2002) Dnmt3L cooperates with the Dnmt3 family of de novo DNA methyltransferases to establish maternal imprints in mice. Development 129, 1983–1993.[Medline]

Herman, J.G., Graff, J.R., Myohanen, S., Nelkin, B.D. & Baylin, S.B. (1996) Methylation-specific PCR: a novel PCR assay for methylation status of CpG islands. Proc. Natl. Acad. Sci. USA 93, 9821–9826.[Abstract/Free Full Text]

Ishida, C., Ura, K., Hirao, A., Sasaki, H., Toyoda, A., Sakaki, Y., Niwa, H., Li, E. & Kaneda, Y. (2003) Genomic organization and promoter analysis of the Dnmt3b gene. Gene 310, 151–159.[CrossRef][Medline]

Kakutani, T., Jeddeloh, J.A., Flowers, S.K., Munakata, K. & Richards, E.J. (1996) Developmental abnormalities and epimutations associated with DNA hypomethylation mutations. Proc. Natl. Acad. Sci. USA 93, 12406–12411.[Abstract/Free Full Text]

Kaneda, M., Okano, M., Hata, K., Sado, T., Tsujimoto, N., Li, E. & Sasaki, H. (2004) Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting. Nature 429, 900–903.[CrossRef][Medline]

La Salle, S., Mertineit, C., Taketo, T., Moens, P.B., Bestor, T.H. & Trasler, J.M. (2004) Windows for sex-specific methylation marked by DNA methyltransferase expression profiles in mouse germ cells. Dev. Biol. 268, 403–415.[CrossRef][Medline]

Li, E. (2002) Chromatin modification and epigenetic reprogramming in mammalian development. Nat. Rev. Genet. 3, 662–673.[CrossRef][Medline]

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]

Meshorer, E., Yellajoshula, 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]

Okano, M., Bell, D.W., Haber, D.A. & Li, E. (1999) DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development. Cell 99, 247–257.[CrossRef][Medline]

Sakai, Y., Suetake, I., Shinozaki, F., Yamashina, S. & Tajima, S. (2004) Co-expression of de novo DNA methyltransferases Dnmt3a2 and Dnmt3L in gonocytes of mouse embryos. Gene Expr. Patterns 5, 231–237.[CrossRef][Medline]

Suetake, I., Shinozaki, F., Miyagawa, J., Takeshima, H. & Tajima, S. (2004) DNMT3L stimulates the DNA methylation activity of Dnmt3a and Dnmt3b through a direct interaction. J. Biol. Chem. 279, 27816–27823.[Abstract/Free Full Text]

Takeshima, H., Suetake, I., Shimahara, H., Ura, K., Tate, S. & Tajima, S. (2006) Distinct DNA methylation activity of Dnmt3a and Dnmt3b towards Naked and Nucleosomal DNA. J. Biochem. 139, 503–515.[Abstract/Free Full Text]

Webster, K.E., O’Bryan, M.K., Fletcher, S., et al. (2005) Meiotic and epigenetic defects in Dnmt3L-knockout mouse spermatogenesis. Proc. Natl. Acad. Sci. USA 102, 4068–4073.[Abstract/Free Full Text]

Received: 8 May 2006
Accepted: 28 June 2006




This article has been cited by other articles:


Home page
 Hum Mol GenetHome page
M. Yamada, T. Hamatani, H. Akutsu, N. Chikazawa, N. Kuji, Y. Yoshimura, and A. Umezawa
Involvement of a novel preimplantation-specific gene encoding the high mobility group box protein Hmgpi in early embryonic development
Hum. Mol. Genet., February 1, 2010; 19(3): 480 - 493.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nimura, K.
Right arrow Articles by Kaneda, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nimura, K.
Right arrow Articles by Kaneda, Y.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS