Project description:Dimethyl sulfate (DMS) is a methylating reagent that has long been used to detect footprints of DNA-bound proteins in vitro as well as in vivo. Here we describe DMS-seq for in vivo genome-wide mapping of protein-DNA interactions. DMS-seq exploits the cell-permeable nature of DMS to obviate the need for nuclear isolation, thereby simplifying the process to detect binding sites of transcription factors. Furthermore, we found that DMS preferentially attacks nucleosome centers in vivo, evidencing for DMS-seq as a first method that locates them without using genetically-modified histones and is hence applicable to any eukaryote. DMS-seq should be a simple and unique method in epigenomics.
Project description:H3 ChIP and input DNA were hybridized to Affymetrix GeneChip S. cerevisiae Tiling 1.0R Array Genome-wide mapping of nucleosomes generated by micrococcal nuclease (MNase) suggests that yeast promoter and terminator regions are very depleted of nucleosomes, predominantly because their DNA sequences intrinsically disfavor nucleosome formation. However, MNase has strong DNA sequence specificity that favors cleavage at promoters and terminators and accounts for some of the correlation between occupancy patterns of nucleosomes assembled in vivo and in vitro. Using an improved method for measuring nucleosome occupancy in vivo that does not involve MNase, we confirm that promoter regions are strongly depleted of nucleosomes, but find that terminator regions are much less depleted than expected. Unlike at promoter regions, nucleosome occupancy at terminators is strongly correlated with the orientation of and distance to adjacent genes. In addition, nucleosome occupancy at terminators is strongly affected by growth conditions, indicating that it is not primarily determined by intrinsic histone-DNA interactions. Rapid removal of RNA polymerase II (Pol II) causes increased nucleosome occupancy at terminators, strongly suggesting a transcription-based mechanism of nucleosome depletion. However, the distinct behavior of terminator regions and their corresponding coding regions suggests that nucleosome depletion at terminators is not simply associated with passage of Pol II, but rather involves a distinct mechanism linked to 3’ end formation.
