DNA methyltransferases Dnmt3a and Dnmt3b are essential for de novo methylation and mammalian development.

The establishment of DNA methylation patterns requires de novo methylation that occurs predominantly during early development and gametogenesis in mice. Here we demonstrate that two recently identified DNA methyltransferases, Dnmt3a and Dnmt3b, are essential for de novo methylation and for mouse development. Inactivation of both genes by gene targeting blocks de novo methylation in ES cells and early embryos, but it has no effect on maintenance of imprinted methylation patterns. Dnmt3a and Dnmt3b also exhibit nonoverlapping functions in development, with Dnmt3b specifically required for methylation of centromeric minor satellite repeats. Mutations of human DNMT3B are found in ICF syndrome, a developmental defect characterized by hypomethylation of pericentromeric repeats. Our results indicate that both Dnmt3a and Dnmt3b function as de novo methyltransferases that play important roles in normal development and disease.     

Molecular enzymology of the catalytic domains of the Dnmt3a and Dnmt3b DNA methyltransferases

The C-terminal domains of the mammalian DNA methyltransferases Dnmt1, Dnmt3a, and Dnmt3b harbor all the conserved motifs characteristic for cytosine-C5 methyltransferases. Whereas the isolated catalytic domain of Dnmt1 is inactive, we show here that the C-terminal domains of Dnmt3a and Dnmt3b are catalytically active. Neither Dnmt3a nor Dnmt3b shows a significant preference for the satellite 2 sequence, although Dnmt3b is required for methylation of these regions in vivo. However, the catalytic domain of Dnmt3a methylates DNA in a distributive reaction, whereas Dnmt3b is processive, which accelerates methylation of macromolecular DNA in vitro. This property could make Dnmt3b a preferred enzyme for methylation at satellite 2 repeats, since they are highly CG-rich. We have also analyzed the catalytic activities of six different mutations found in ICF (immunodeficiency, centromeric instability, and facial abnormalities) patients in the catalytic domain of Dnmt3b. Five of them display catalytic activities reduced by 10-50-fold; one mutant was inactive in our assay (residual activity <1%). These results confirm that a reduced catalytic activity of Dnm3b causes ICF. However, the mutations in general do not completely abrogate catalytic activity. This finding may explain why ICF patients are viable, whereas nmt3b knock-out mice die during embryogenesis.     

Distinct enzymatic properties of recombinant mouse DNA methyltransferases Dnmt3a and Dnmt3b

Recombinant mouse Dnmt3a and Dnmt3b were expressed in sf9 cells and purified to near homogeneity. The purified Dnmt3a and Dnmt3b gave specific activities of 1.8 +/- 0.3 and 1.3 +/- 0.1 mol/h/mol enzyme towards poly(dGdC)-poly(dGdC), respectively, which were the highest among those reported. Dnmt3a or Dnmt3b showed similar K(m) values towards poly(dIdC)-poly(dIdC) and poly(dGdC)-poly(dGdC). The K(m) values for S-adenosyl-L-methionine were not affected by the methyl-group acceptors, poly(dI-dC)-poly(dIdC) and poly(dG-dC)-poly(dGdC). The results indicate that the enzymes are de novo-type DNA methyltransferases. Dnmt3a and Dnmt3b activities were inhibited by Mn(2+) and Ni(2+) and showed broad pH optima around neutral pH. Both enzymes were susceptible to sodium ions, which inhibited their activity at around physiological ionic strength. However, Dnmt3a was fully active at physiological potassium concentration, but Dnmt3b was not. Using designed oligonucleotides for the analysis of cytosine methylation, we demonstrated that, in addition to CpG, Dnmt3a methylated CpA but not CpT and CpC, and that Dnmt3b methylated CpA and CpT but scarcely CpC. The relative activity of Dnmt3b towards nonCpG sequences was higher than that of Dnmt3a. These differences in enzymatic properties of Dnmt3a and Dnmt3b may contribute to the distinct functions of these enzymes in vivo.     

De novo methylation of nucleosomal DNA by the mammalian Dnmt1 and Dnmt3A DNA methyltransferases

In the cell, DNA is wrapped on histone octamers, which reduces its accessibility for DNA interacting enzymes. We investigated de novo methylation of nucleosomal DNA in vitro and show that the Dnmt3a and Dnmt1 DNA methyltransferases efficiently methylate nucleosomal DNA without dissociation of the histone octamer from the DNA. In contrast, the prokaryotic SssI DNA methyltransferase and the catalytic domain of Dnmt3a are strongly inhibited by nucleosomes. We also found that full-length Dnmt1 and Dnmt3a bind to nucleosomes much stronger than their isolated catalytic domains, demonstrating that the N-terminal parts of the MTases are required for the interaction with nucleosomes. Variations of the DNA sequence or the histone tails did not significantly influence the methylation activity of Dnmt3a. The observation that mammalian methyltransferases directly modify nucleosomal DNA provides an insight into the mechanisms by which histone tail and DNA methylation patterns can influence each other because the DNA methylation pattern can be established while histones remain associated to the DNA.     

Distinct DNA methylation activity of Dnmt3a and Dnmt3b towards naked and nucleosomal DNA

In mammals, the resetting of DNA methylation patterns in early embryos and germ cells is crucial for development. De novo type DNA methyltransferases Dnmt3a and Dnmt3b are responsible for creating DNA methylation patterns during embryogenesis and in germ cells. Although their in vitro DNA methylation properties are similar, Dnmt3a and Dnmt3b methylate different genomic DNA regions in vivo. In the present study, we have examined the DNA methylation activity of Dnmt3a and Dnmt3b towards nucleosomes reconstituted from recombinant histones and DNAs, and compared it to that of the corresponding naked DNAs. Dnmt3a showed higher DNA methylation activity than Dnmt3b towards naked DNA and the naked part of nucleosomal DNA. On the other hand, Dnmt3a scarcely methylated the DNA within the nucleosome core region, while Dnmt3b significantly did, although the activity was low. We propose that the preferential DNA methylation activity of Dnmt3a towards the naked part of nucleosomal DNA and the significant methylation activity of Dnmt3b towards the nucleosome core region contribute to their distinct methylation of genomic DNA in vivo.     

Profound flanking sequence preference of Dnmt3a and Dnmt3b mammalian DNA methyltransferases shape the human epigenome

Mammalian DNA methyltransferases methylate cytosine residues within CG dinucleotides. By statistical analysis of published data of the Human Epigenome Project we have determined flanking sequences of up to +/-four base-pairs surrounding the central CG site that are characteristic of high (5'-CTTGCGCAAG-3') and low (5'-TGTTCGGTGG-3') levels of methylation in human genomic DNA. We have investigated the influence of flanking sequence on the catalytic activity of the Dnmt3a and Dnmt3b de novo DNA methyltransferases using a set of synthetic oligonucleotide substrates that covers all possible +/-1 flanks in quantitative terms. Methylation kinetics experiments revealed a >13-fold difference between the preferred (RCGY) and disfavored +/-1 flanking base-pairs (YCGR). In addition, AT-rich flanks are preferred over GC-rich ones. These experimental preferences coincide with the genomic methylation patterns. Therefore, we have expanded our experimental analysis and found a >500-fold difference in the methylation rates of the consensus sequences for high and low levels of methylation in the genome. This result demonstrates a very pronounced flanking sequence preference of Dnmt3a and Dnmt3b. It suggests that the methylation pattern of human DNA is due, in part, to the flanking sequence preferences of the de novo DNA MTases and that flanking sequence preferences could be involved in the origin of CG islands. Furthermore, similar flanking sequence preferences have been found for the stimulation of the immune system by unmethylated CGs, suggesting a co-evolution of DNA MTases and the immune system.     

DNMT3L stimulates the DNA methylation activity of Dnmt3a and Dnmt3b through a direct interaction

In mammals, the resetting of DNA methylation patterns in early embryos and germ cells is crucial for development. Two DNA methyltransferases, Dnmt3a and Dnmt3b, are responsible for the creation of DNA methylation patterns. Dnmt3L, a member of the Dnmt3 family, has been reported to be necessary for maternal methylation imprinting, possibly by interacting with Dnmt3a and/or Dnmt3b (Hata, K., Okano, M., Lei, H., and Li, E. (2002) Development 129, 1983-1993). In the present study, the effect of DNMT3L, a human homologue of Dnmt3L, on the DNA methylation activity of mouse Dnmt3a and Dnmt3b was examined in vitro. DNMT3L enhanced the DNA methylation activity of Dnmt3a and Dnmt3b about 1.5-3-fold in a dose-dependent manner but did not enhance the DNA methylation activity of Dnmt1. Although the extents of stimulation were different, a stimulatory effect on the DNA methylation activity was observed for all of the substrate DNA sequences examined, such as those of the maternally methylated SNRPN and Lit-1 imprinting genes, the paternally methylated H19 imprinting gene, the CpG island of the myoD gene, the 5 S ribosomal RNA gene, an artificial 28-bp DNA, poly(dG-dC)-poly(dG-dC), and poly(dI-dC)-poly(dI-dC). DNMT3L could not bind to DNA but could bind to Dnmt3a and Dnmt3b, indicating that the stimulatory effect of DNMT3L on the DNA methylation activity may not be due to the guiding of Dnmt3a and Dnmt3b to the targeting DNA sequence but may comprise a direct effect on their catalytic activity. The carboxyl-terminal half of DNMT3L was found to be responsible for the enhancement of the enzyme activity.     

In vivo activity of murine de novo methyltransferases, Dnmt3a and Dnmt3b

The putative de novo methyltransferases, Dnmt3a and Dnmt3b, were reported to have weak methyltransferase activity in methylating the 3' long terminal repeat of Moloney murine leukemia virus in vitro. The activity of these enzymes was evaluated in vivo, using a stable episomal system that employs plasmids as targets for DNA methylation in human cells. De novo methylation of a subset of the CpG sites on the stable episomes is detected in human cells overexpressing the murine Dnmt3a or Dnmt3b1 protein. This de novo methylation activity is abolished when the cysteine in the P-C motif, which is the catalytic site of cytosine methyltransferases, is replaced by a serine. The pattern of methylation on the episome is nonrandom, and different regions of the episome are methylated to different extents. Furthermore, Dnmt3a also methylates the sequence methylated by Dnmt3a on the stable episome in the corresponding chromosomal target. Overexpression of human DNMT1 or murine Dnmt3b does not lead to the same pattern or degree of de novo methylation on the episome as overexpression of murine Dnmt3a. This finding suggests that these three enzymes may have different targets or requirements, despite the fact that weak de novo methyltransferase activity has been demonstrated in vitro for all three enzymes. It is also noteworthy that both Dnmt3a and Dnmt3b proteins coat the metaphase chromosomes while displaying a more uniform pattern in the nucleus. This is the first evidence that Dnmt3a and Dnmt3b have de novo methyltransferase function in vivo and the first indication that the Dnmt3a and Dnmt3b proteins may have preferred target sites.     

Stimulation effect of Dnmt3L on the DNA methylation activity of Dnmt3a2

Quantification of DNA methyltransferases Dnmt3a and Dnmt3a2, and Dnmt3L in isolated male gonocytes in day 16.5 embryos confirmed that not Dnmt3a but Dnmt3a2 and Dnmt3L were the major Dnmt3s. The expression level of Dnmt3L constituted 5- to 10-fold molar excess compared to that of Dnmt3a2. The stimulation property of the DNA methylation activity of Dnmt3a2 with Dnmt3L towards substrate DNA in naked or nucleosomes was similar to that of Dnmt3a. However, the DNA methylation activity of not Dnmt3a but Dnmt3a2 was severely inhibited at the physiological salt concentration. Interestingly, the activity of Dnmt3a2 was significantly detected in the presence of Dnmt3L even at the physiological salt concentration. This indicates that Dnmt3a2 functions only in the presence of Dnmt3L in male gonocytes, and may explain why Dnmt3L is required specifically in mouse gonocytes for DNA methylation.     

The PWWP domain of Dnmt3a and Dnmt3b is required for directing DNA methylation to the major satellite repeats at pericentric heterochromatin

Dnmt3a and Dnmt3b are responsible for the establishment of DNA methylation patterns during development. These proteins contain, in addition to a C-terminal catalytic domain, a unique N-terminal regulatory region that harbors conserved domains, including a PWWP domain. The PWWP domain, characterized by the presence of a highly conserved proline-tryptophan-tryptophan-proline motif, is a module of 100 to 150 amino acids found in many chromatin-associated proteins. However, the function of the PWWP domain remains largely unknown. In this study, we provide evidence that the PWWP domains of Dnmt3a and Dnmt3b are involved in functional specialization of these enzymes. We show that both endogenous and green fluorescent protein-tagged Dnmt3a and Dnmt3b are particularly concentrated in pericentric heterochromatin. Mutagenesis analysis indicates that their PWWP domains are required for their association with pericentric heterochromatin. Disruption of the PWWP domain abolishes the ability of Dnmt3a and Dnmt3b to methylate the major satellite repeats at pericentric heterochromatin. Furthermore, we demonstrate that the Dnmt3a PWWP domain has little DNA-binding ability, in contrast to the Dnmt3b PWWP domain, which binds DNA nonspecifically. Collectively, our results suggest that the PWWP domains of Dnmt3a and Dnmt3b are essential for targeting these enzymes to pericentric heterochromatin, probably via a mechanism other than protein-DNA interactions.     

Dnmt3a and Dnmt1 functionally cooperate during de novo methylation of DNA.

Dnmt3a is a de novo DNA methyltransferase that modifies unmethylated DNA. In contrast Dnmt1 shows high preference for hemimethylated DNA. However, Dnmt1 can be activated for the methylation of unmodified DNA. We show here that the Dnmt3a and Dnmt1 DNA methyltransferases functionally cooperate in de novo methylation of DNA, because a fivefold stimulation of methylation activity is observed if both enzymes are present. Stimulation is observed if Dnmt3a is used before Dnmt1, but not if incubation with Dnmt1 precedes Dnmt3a, demonstrating that methylation of the DNA by Dnmt3a stimulates Dnmt1 and that no physical interaction of Dnmt1 and Dnmt3a is required. If Dnmt1 and Dnmt3a were incubated together a slightly increased stimulation is observed that could be due to a direct interaction of these enzymes. In addition, we show that Dnmt1 is stimulated for methylation of unmodified DNA if the DNA already carries some methyl groups. We conclude that after initiation of de novo methylation of DNA by Dnmt3a, Dnmt1 becomes activated by the pre-existing methyl groups and further methylates the DNA. Our data suggest that Dnmt1 also has a role in de novo methylation of DNA. This model agrees with the biochemical properties of these enzymes and provides a mechanistic basis for the functional cooperation of different DNA MTases in de novo methylation of DNA that has also been observed in vivo.     

Establishment and maintenance of genomic methylation patterns in mouse embryonic stem cells by Dnmt3a and Dnmt3b

We have previously shown that the DNA methyltransferases Dnmt3a and Dnmt3b carry out de novo methylation of the mouse genome during early postimplantation development and of maternally imprinted genes in the oocyte. In the present study, we demonstrate that Dnmt3a and Dnmt3b are also essential for the stable inheritance, or “maintenance,” of DNA methylation patterns. Inactivation of both Dnmt3a and Dnmt3b in embryonic stem (ES) cells results in progressive loss of methylation in various repeats and single-copy genes. Interestingly, introduction of the Dnmt3a, Dnmt3a2, and Dnmt3b1 isoforms back into highly demethylated mutant ES cells restores genomic methylation patterns; these isoforms appear to have both common and distinct DNA targets, but they all fail to restore the maternal methylation imprints. In contrast, overexpression of Dnmt1 and Dnmt3b3 failed to restore DNA methylation patterns due to their inability to catalyze de novo methylation in vivo. We also show that hypermethylation of genomic DNA by Dnmt3a and Dnmt3b is necessary for ES cells to form teratomas in nude mice. These results indicate that genomic methylation patterns are determined partly through differential expression of different Dnmt3a and Dnmt3b isoforms.     

Np95 interacts with de novo DNA methyltransferases,Dnmt3a and Dnmt3b,and mediates epigenetic silencing of the viral CMV promoter in embryonic stem cells

Recent studies have indicated that nuclear protein of 95 kDa (Np95) is essential for maintaining genomic methylation by recruiting DNA methyltransferase (Dnmt) 1 to hemi‐methylated sites. Here, we show that Np95 interacts more strongly with regulatory domains of the de novo methyltransferases Dnmt3a and Dnmt3b. To investigate possible functions, we developed an epigenetic silencing assay using fluorescent reporters in embryonic stem cells (ESCs). Interestingly, silencing of the cytomegalovirus promoter in ESCs preceded DNA methylation and was strictly dependent on the presence of either Np95, histone H3 methyltransferase G9a or Dnmt3a and Dnmt3b. Our results indicate a regulatory role for Np95, Dnmt3a and Dnmt3b in mediating epigenetic silencing through histone modification followed by DNA methylation.     

Dnmt3L cooperates with the Dnmt3 family of de novo DNA methyltransferases to establish maternal imprints in mice

Genomic imprinting is regulated by differential methylation of the paternal and maternal genome. However, it remains unknown how parental imprinting is established during gametogenesis. In this study, we demonstrate that Dnmt3L, a protein sharing homology with DNA methyltransferases, Dnmt3a and Dnmt3b, but lacking enzymatic activity, is essential for the establishment of maternal methylation imprints and appropriate expression of maternally imprinted genes. We also show that Dnmt3L interacts with Dnmt3a and Dnmt3b and co-localizes with these enzymes in the nuclei of transfected cells, suggesting that Dnmt3L may regulate genomic imprinting via the Dnmt3 family enzymes. Consistent with this model, we show that [Dnmt3a(-/-), Dnmt3b(+/-)] mice also fail to establish maternal methylation imprints. In addition, both Dnmt3a and Dnmt3L are required for spermatogenesis. Together, our findings suggest that Dnmt3L may cooperate with Dnmt3 family methyltransferases to carry out de novo methylation of maternally imprinted genes in oocytes.     

On the evolutionary origin of eukaryotic DNA methyltransferases and Dnmt2

The Dnmt2 enzymes show strong amino acid sequence similarity with eukaryotic and prokaryotic DNA-(cytosine C5)-methyltransferases. Yet, Dnmt2 enzymes from several species were shown to methylate tRNA-Asp and had been proposed that eukaryotic DNA methyltransferases evolved from a Dnmt2-like tRNA methyltransferase ancestor [Goll et al., 2006, Science, 311, 395-8]. It was the aim of this study to investigate if this hypothesis could be supported by evidence from sequence alignments. We present phylogenetic analyses based on sequence alignments of the methyltransferase catalytic domains of more than 2300 eukaryotic and prokaryotic DNA-(cytosine C5)-methyltransferases and analyzed the distribution of DNA methyltransferases in eukaryotic species. The Dnmt2 homologues were reliably identified by an additional conserved CFT motif next to motif IX. All DNA methyltransferases and Dnmt2 enzymes were clearly separated from other RNA-(cytosine-C5)-methyltransferases. Our sequence alignments and phylogenetic analyses indicate that the last universal eukaryotic ancestor contained at least one member of the Dnmt1, Dnmt2 and Dnmt3 families of enzymes and additional RNA methyltransferases. The similarity of Dnmt2 enzymes with DNA methyltransferases and absence of similarity with RNA methyltransferases combined with their strong RNA methylation activity suggest that the ancestor of Dnmt2 was a DNA methyltransferase and an early Dnmt2 enzyme changed its substrate preference to tRNA. There is no phylogenetic evidence that Dnmt2 was the precursor of eukaryotic Dnmts. Most likely, the eukaryotic Dnmt1 and Dnmt3 families of DNA methyltransferases had an independent origin in the prokaryotic DNA methyltransferase sequence space.