A crystal structure of a complex between the DNA methyltransferase regulatory factor Dnmt3L and the catalytic domain of Dnmt3a leads to a model being proposed for the preferential methylation of DNA on maternally imprinted genes. Genetic imprinting, found in flowering plants and placental mammals, uses DNA methylation to yield gene expression that is dependent on the parent of origin1. DNA methyltransferase 3a (Dnmt3a) and its regulatory factor, DNA methyltransferase 3-like protein (Dnmt3L), are both required for the de novo DNA methylation of imprinted genes in mammalian germ cells. Dnmt3L interacts specifically with unmethylated lysine 4 of histone H3 through its amino-terminal PHD (plant homeodomain)-like domain2. Here we show, with the use of crystallography, that the carboxy-terminal domain of human Dnmt3L interacts with the catalytic domain of Dnmt3a, demonstrating that Dnmt3L has dual functions of binding the unmethylated histone tail and activating DNA methyltransferase. The complexed C-terminal domains of Dnmt3a and Dnmt3L showed further dimerization through Dnmt3a–Dnmt3a interaction, forming a tetrameric complex with two active sites. Substitution of key non-catalytic residues at the Dnmt3a–Dnmt3L interface or the Dnmt3a–Dnmt3a interface eliminated enzymatic activity. Molecular modelling of a DNA–Dnmt3a dimer indicated that the two active sites are separated by about one DNA helical turn. The C-terminal domain of Dnmt3a oligomerizes on DNA to form a nucleoprotein filament. A periodicity in the activity of Dnmt3a on long DNA revealed a correlation of methylated CpG sites at distances of eight to ten base pairs, indicating that oligomerization leads Dnmt3a to methylate DNA in a periodic pattern. A similar periodicity is observed for the frequency of CpG sites in the differentially methylated regions of 12 maternally imprinted mouse genes. These results suggest a basis for the recognition and methylation of differentially methylated regions in imprinted genes, involving the detection of both nucleosome modification and CpG spacing.

The Dnmt3a DNA methyltransferase contains in its N-terminal part a PWWP domain that is involved in chromatin targeting. Here, we have investigated the interaction of the PWWP domain with modified histone tails using peptide arrays and show that it specifically recognizes the histone 3 lysine 36 trimethylation mark. H3K36me3 is known to be a repressive modification correlated with DNA methylation in mammals and heterochromatin in Schizosaccharomyces pombe. These results were confirmed by equilibrium peptide binding studies and pulldown experiments with native histones and purified native nucleosomes. The PWWP-H3K36me3 interaction is important for the subnuclear localization of enhanced yellow fluorescent protein-fused Dnmt3a. Furthermore, the PWWP-H3K36me3 interaction increases the activity of Dnmt3a for methylation of nucleosomal DNA as observed using native nucleosomes isolated from human cells after demethylation of the DNA with 5-aza-2′-deoxycytidine as substrate for methylation with Dnmt3a. These data suggest that the interaction of the PWWP domain with H3K36me3 is involved in targeting of Dnmt3a to chromatin carrying that mark, a model that is in agreement with several studies on the genome-wide distribution of DNA methylation and H3K36me3.

Using peptide arrays and binding to native histone proteins, we show that the ADD domain of Dnmt3a specifically interacts with the H3 histone 1-19 tail. Binding is disrupted by di- and trimethylation of K4, phosphorylation of T3, S10 or T11 and acetylation of K4. We did not observe binding to the H4 1-19 tail. The ADD domain of Dnmt3b shows the same binding specificity, suggesting that the distinct biological functions of both enzymes are not related to their ADD domains. To establish a functional role of the ADD domain binding to unmodified H3 tails, we analyzed the DNA methylation of in vitro reconstituted chromatin with Dnmt3a2, the Dnmt3a2/Dnmt3L complex, and the catalytic domain of Dnmt3a. All Dnmt3a complexes preferentially methylated linker DNA regions. Chromatin substrates with unmodified H3 tail or with H3K9me3 modification were methylated more efficiently by full-length Dnmt3a and full-length Dnmt3a/3L complexes than chromatin trimethylated at H3K4. In contrast, the catalytic domain of Dnmt3a was not affected by the H3K4me3 modification. These results demonstrate that the binding of the ADD domain to H3 tails unmethylated at K4 leads to the preferential methylation of DNA bound to chromatin with this modification state. Our in vitro results recapitulate DNA methylation patterns observed in genome-wide DNA methylation studies.