Chromatin remodeling factors and histone chaperones were previously shown to cooperatively affect nucleosome assembly and disassembly processes in vitro. Here we show that S. pombe CHD remodellers, Hrp1 and Hrp3 physically interact with the histone chaperone Nap1. Genome wide analysis of Hrp1, Hrp3 and Nap1 occupancy, combined with nucleosome density measurements in respective mutants revealed that the CHD factors and Nap1 co-localized in particular to promoter regions where they remove nucleosomes near the transcriptional start site. Hrp1 and Hrp3 also regulate nucleosome density in coding regions where they have redundant roles to stimulate transcription. Previously, DNA replication dependent and independent nucleosome disassembly processes have been described. We found that nucleosome density increased in the hrp1 mutant in the absence of DNA replication. Finally, regions where nucleosome density increased in hrp1, hrp3 and nap1 mutants also showed nucleosome density and histone modification changes in HDAC and HAT mutants. Thus, this study revealed an important in vivo role for CHD remodellers and Nap1 in nucleosome disassembly at promoters and coding regions, which are linked to changes in histone acetylation.
Keywords: ChIP on CHIP and expression profiling
Overall design: Expression profiling experiments were performed and quantified according to (Xue et al., 2004). Data was normalized using Lowess per spot per chip intensity-dependent Gene Spring v 7.2 (Silicon Genetics).
IGR and ORF DNA microarrays (Eurogentec) and GeneChip® S. pombe Tiling Array (Affymetrix) were used to create ChIP on CHIP protein binding and histone H3 density maps. Cells cultures for both microarray platforms were grown to mid logarithmic phase, harvested, fixated lysed and ChIPed as described in (Sinha et al., 2006). Antibodies specific for a-myc (9E10, Sigma) and anti-H3 C-terminal antibody (AB1791, Abcam) were used for the ChIP.
For the Eurogentec ChIP on CHIP DNA microarrays, 1.0 mg of input DNA and ChIP DNA was labelled with either Cy3 or Cy5 fluorescent dyes respectively and both labelled samples were together hybridized and quantified according to (Wiren et al., 2005). All binding maps and the histone H3 density experiments were performed in at least duplicate experiments generating 6 respective 4 measurement points since each microarray slide contain duplicate spots for each fragment. In addition, we performed cy3 and cy5 dye swap for every experiment to correct for dye bias. HDAC ChIP on CHIP data published in (Wiren et al., 2005), containing acetylation levels and H3 density were quality filtered, normalisednormalized and combined to generate direct effects in histone density changes, as described above for the Hrp1, Hrp3 and Nap1.
The Affymetrix Tiling arrays ChIP DNA was extracted as described above and amplified to 5.0 mg using protocol published in (Robyr and Grunstein, 2003). Next, DNA was fragmentated to ~100 bp using DNAse I, followed by biotin labelling and hybridized according to (Lengronne et al., 2004). Affymetrix Tiling arrays to determine H3 density throughout genes at direct binding for Hrp1 and Hrp3 were performed once.
Less...