Nature Genetics
27, 240 - 241 (2001)
doi:10.1038/85787
Chipping away at chromatinRobert MartienssenCold Spring Harbor Laboratory, 1 Bungtown Road, Cold Spring Harbor, New York 11724, USA.
martiens@cshl.org When DNA-binding proteins are tethered to dam methylase from Escherichia coli, adenine methylation is directed to eukaryotic target sites in vivo. Hybridization of methylated DNA to microarrays allows binding sites to be displayed genome-wide, providing a versatile alternative to chromatin immunoprecipitation.Methylation of adenine, forming N6-methyl adenine (m6A), is widespread in prokaryotic DNA but occurs at very low levels, or not at all, in the nuclear DNA of eukaryotes. Nonetheless, the cells of mammals, Drosophila and yeast can tolerate methylation by the prokaryotic DNA m6A methyl-transferase, dam. On page 304, Bas van Steensel and colleagues1 describe a method that takes advantage of this difference between prokaryotic and eukaryotic DNA to profile chromatin-binding sites in complex genomes.
In mammalian cells2, yeast3 and Drosophila4, dam methylase of Escherichia coli has been used to probe chromatin in much the same way as nucleases and restriction enzymes have been used in the past. In each case, restriction or methylation of a given DNA sequence is interpreted to mean that the DNA is relatively free of chromatin and other hindrances to enzyme access. Using Drosophila DNA as a template, van Steensel and Henikoff5 modified this strategy by fusing the dam methylase gene with the DNA-binding domain of chromatin proteins and transcription factors such as Gal4. They found that, on exposure to these constructs, the binding sites of Gal4 and HP-1, in addition to a few kilobases of surrounding DNA, were methylated—as revealed by restriction digests and Southern blotting. They called the technique 'DamID'.
Now, these same authors describe a method to map dam-methylated sites using microarrays1. DNA methylated by dam fusion proteins becomes sensitive to digestion with DpnI (see figure). Small genomic DpnI fragments were isolated from cells expressing the fusion protein and allowed to hybridize (or not) with microarrays. Probes representing target genes bound fragments methylated by dam fusion proteins, but not control fragments of genomes methylated by dam alone. Specificity of binding was assessed in replicate experiments. Whereas the relative affinities of different sites were also estimated, their accuracy relies on limiting amounts of both tethered and untethered proteins relative to their binding sites, which was not tested directly.
Although the arrays were comprised of a comparatively small set of probes representing genes and transposons, the results indicate the feasibility of the technique for genome-wide mapping. The authors provide several demonstrations. In one, HP1-dam fusions were shown to methylate 12 of 13 transposons on the array. This is consistent with the location of transposons in heterochromatin, which is known to bind HP-1. It also gives credibility to the idea that heterochromatin functions as a genome defense mechanism to prevent transposon spread, although it could also simply reflect its evolutionary origin6,
7. In another demonstration, the GAGA-binding factor (GAF) was shown to bind a variety of euchromatic target genes which had significantly elevated levels of GA dinucleotides. The HP1- and GAF-binding sites were non-overlapping, a result confirmed by immuno-localization using tagged proteins in cultured Drosophila cells.
In a third test, the authors fused one of four Drosophila homologs of the yeast silencing gene SIR2 to dam and expressed the fusion product in cultured Drosophila cells. They found that Sir2-binding sites in Drosophila were different from those of yeast, in that they included genes but not ribosomal DNA. The other three Sir2 proteins have not yet been tested but amino and carboxy terminal fusions between dam and Sir2 gave comparable results in several replicates, indicating significance.
DamID is complementary to, and is perhaps a substantial improvement over, chromatin immunoprecipitation (ChIP), a technique that has enjoyed widespread popularity in recent years. ChIP works by cross linking target sites to their cognate DNA binding protein, and then immunoprecipitating the DNA-protein complex with antibodies raised against it. The target DNA can be labeled, and allowed to hybridize with microarrays8. However, ChIP requires well-behaved antibodies and may be less sensitive than in vivo methods like DamID. In both cases, the power of microarray analysis allows multiple samples to be analyzed at multiple locations in parallel, allowing powerful statistical tests, such as hierarchical clustering, to be applied to the data.
As with ChIP, DamID will be most powerful when using comprehensive genomic microarrays that include promoter and enhancer sequences, in addition to those representing the coding regions of genes. Currently, such arrays are only available for the analysis of yeast; those for larger genomes, including those of Drosophila and Arabidopsis, are being constructed. The power of DamID will be realized in the analysis of chromatin profiles in different cell types during development. The epigenetic changes (including decreased cytosine methylation) that accompany immortalization of cell cultures derived from vertebrates caution that the data from the cultured Drosophila cells be carefully interpreted, with biological context in mind. As the authors suggest, chromatin profiling of cancer cells in culture might document epigenetic changes and help define their impact on tumor progression.
REFERENCES
- van Steensel, B., Delrow, J. & Henikoff, S. Nature Genet. 27, 304−308 (2001). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Kirianov, G.I., Smirnova, T.A., Isaeva, L.V., Vaniushin, B.F. & Bur'ianov, I.I. Biokhimiia 46, 1887−1895 (1981). | PubMed | ChemPort | Add to Connotea (beta) |
- Singh, J. & Klar, A.J. Genes Dev. 6, 186−196 (1992). | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Wines, D.R., Talbert, P.B., Clark, D.V. & Henikoff, S. Chromosoma 104, 332−340 (1996). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- van Steensel, B. & Henikoff, S. Nature Biotechnol. 18, 424−428 (2000). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
- Henikoff, S. Biochim. Biophys. Acta. 1470, 1−8 (1999). | Article | Add to Connotea (beta) |
- The Cold Spring Harbor Laboratory, Washington University & PE Biosystems Arabidopsis Sequencing Consortium. Cell 100, 377−386 (2000). | PubMed | ISI | Add to Connotea (beta) |
- Ren, B. et al. Science 290, 2306−2309 (2000). | Article | PubMed | ISI | ChemPort | Add to Connotea (beta) |
|
