Print version | Recommend article | RSS-Feed

Methylation of the Arginine 2 in Histone H3 Controls Deposition of Lysine 4 Tri-Methylation

In a recent study, Kirmizis et al. demonstrated that H3R2 is methylated not only in mammals but also in Saccharomyces cerevisiae. By using a specific H3R2me2a antibody in a ChIP-on-Chip analysis the authors determined the distribution of the methylation throughout the yeast genome, finding an enrichment of H3R2me2a at all heterochromatic genes, at inactive euchromatic genes, and at the 3“-end of moderately transcribed genes. The presence of H3R2 methylation inversely correlates with the presence of H3K4 tri-methylation (H3K4me3) in all cases, reflecting the fact that H3R2 methylation disrupts the ability of the Set1-complex to methyl­ate H3K4 by preventing the Set1-methyltransferase subunit Spp1 from binding to to histone H3. These results show that H3R2 methylation controls the distribution of H3K4me3 and provide the first mechanistic insight into the function of arginine methylation on chromatin.

Introduction

Modifications of histones have an effect on chromatin structure and thereby control important biological processes. For example, methylation at lysine and arginine residues within histones is linked to gene expression. Methylation at histone H3 lysine 4 (H3K4) can be detected at the 5“-end of active genes. It also contributes to transcriptional activation by recruiting chromatin remodeling enzymes. It is known that the neighboring arginine residue H3R2 is asymmetrically dimethylated (H3R2me2a) in mammalian cells, but its location within genes and its function are unknown.

Materials and Methods

ChIP-on-Chip

Formaldehyde cross-linking and chromatin immunoprecipitations (ChIP) were done as described earlier except for the following changes: the immunocomplexes were eluted from the beads using 200 µl of elution buffer, and RNase treatment was performed while the cross-links were being reversed at 65°C for 5 hours. After that, each ChIP sample was purified using a commercially available PCR purification kit and DNA was eluted in 50 µl of EB buffer. A Linker-Mediated PCR was used to generate amplicons from the ChIP samples. Sample labeling, hybridization, and data extraction were performed by NimbleGen Systems.

Downstream analysis was carried out using the statistical package R and associated array analysis modules were performed in Bioconductor.

Results and Discussion

H3R2 methylation in heterochromatic regions

Kirmizis et al. used a high-resolution ChIP-on-Chip assay to investigate the distribution of H3R2 methylation in the yeast genome. H3R2me2a was found to be present at all four heterochromatic regions in yeast: the two silent mating type loci (HMR and HML), the rRNA-encoding DNA (rDNA repeat), and the telomeres (Figure 1). All regions enriched with H3R2 methylation were devoid of the active methyl-mark H3K4me3. In yeast strains expressing H3R2 mutants (H3R2A and H3R2Q) the silencing at the HMR locus, telomere, rDNA, and to a moderate extent at the HML locus is lost (data not shown), indicating that the arginine 2 on H3 is needed for heterochromatic silencing and that H3R2 methylation might play a role in this process.

To investigate the mechanism by which H3R2me2a may function to regulate heterochromatin the authors compared the occupancy of key heterochromatic factors, such as Rap1p and Sir2p, and found that it coincides with H3R2me2a enrichment at telomeric sites. However, a comparision of wild type and H3R2A mutant strains revealed that the binding levels of these factors are not changed by the mutation (data not shown). This indicates that H3R2 methylation does not depend on Rap1p and Sir2p recruitment.

H3R2 methylation within euchromatin

To analyze the function of H3R2me2a within euchromatin, 5,065 genes were devided into five groups according to their transcriptional rate and the average enrichment of H3R2me2a was assessed for each group. Composite profiles show that H3R2me2a occurs near the middle of the coding region and peaks towards the 3“-end of genes (Figure 2a). H3R2me2a enrichment is most abundant on the least active genes, indicating an inverse correlation with transcriptional activity. Figure 2b shows that H3K4me3 is present at the 5“-end of genes and that H3K4me3 enrichment is most abundant on the most active genes. These results suggest an antagonistic relationship between H3R2 and H3K4 methyl­ations.

A comparison of the distribution of H3R3me2a with the distribution of the three H3K4 methylation states (H3K4me1, H3K4me2, and H3K4me3) revealed that in all three analyzed groups of genes (inactive, moderately transcribed, and highly transcribed) the pattern of H3R3me2a is mutually exclusive specifically with H3K4me3: at inactive genes, promoter and coding regions H3R2me2a is enriched and H3K4me3 is absent; at moderately transcribed genes, H3R2me2a is present at the 3“-end while H3K4me3 is present at the 5“-end; and at highly transcribed genes, H3R2me2a is absent and HrK4me3 is present at promoter and coding regions (data not shown). This inverse correlation of H3R2me2a and H3K4me3 is not observed with the tri-methylation of the two other known methylated lysines in S. cerevisae, H3K36me3 and H3K79me3 (data not shown). These results show that H3R2me2a covers the promoter and coding regions of inactive genes and recedes, as the transcriptional activity increases, from the 5“-end and is replaced by H3K4 tri-methylation.

Functional analysis

Further analysis using cells grown under repressive conditions (glucose) and then shifted to activating conditions (galactose) showed that the inverse correlation between H3R2me2a and H3K4me3 is dynamically co-regulated in the sense that when one mark is removed from the nucleosomes the other one is incorporated. In addition, mutation of H3R2 abolishes or greatly reduces tri-methylation of H3K4 while the mono- and di-methylation of this residue are unaffected. Furthermore, the activation of GAL genes is delayed in H3R2A and H3K4A mutant strains, indicating that these two residues might be involved in a common regulatory mechanism.

The mechanism responsible for the inverse correlation between H3R2me2a and H3K4me3 was investigated using a methylase assay (data not shown). The results suggest that H3R2 is a recognition site of the Set1p methyl­ase complex and that the methylation of H3R2 inhibits the Set1p enzyme from methylating H3K4. The Set1p subunit Spp1 was shown to bind to regions of genes that show H3K4 tri-methylation but was absent in areas of genes where H3R2 methylation was present even though H3K4me1 or H3K4me2 was also abundant in these regions.

Conclusions

In conclusion, the findings prove the existence and point to a function of H3R2 methylation in S. cerevisae. H3R2me2a and H3K4me3 correlate inversely in heterochromatin and euchromatin. Global analysis showed that in inactive genes, H3R2me2a is abundant in the promoter and coding regions, while no or very little methylation of H3K4 takes place. During activation, the presence of H3R2me2a does not interfere with mono- or di-methylation of H3K4 by Set1p. For tri-methylation of H3R4, the methylation at H3R2 has to be removed. Once the region has been cleared of H3R2me2a, the Spp1 protein can recognize H3K4me2 and thereby might promote the tri-methylation of H3K4 by Set1p. Due to the similarities between the yeast (set1p) and the human (MLL) H3K4 methylation complexes, the authors expect the mechanism by which H3R2 methylation functions in yeast to be conserved in higher organisms.

This article was summarized for BIOCHEMICA using the following original publication:

Kirmizis A et al. (2007) Nature 449:928–932

This article was originally published in Biochemica 1/2008, pages 6-7. ©Springer Medizin Verlag 2008