Project description:The nucleosome plays a central role in genome regulation. Traditional methods for mapping nucleosomes depend on the resistance of the nucleosome core to micrococcal nuclease (MNase). However, the lengths of the protected DNA fragments are heterogeneous, limiting the accuracy of nucleosome position information. To resolve this problem, we removed residual linker DNA by simultaneous digestion of yeast chromatin with MNase and exonuclease III (ExoIII). Paired-end sequencing of mono-nucleosomes revealed not only core particles (145-147 bp), but also intermediate particles in which ~8 bp project from one side (154 bp) or both sides (161 bp) of the nucleosome core. We term these particles "pseudo-chromatosomes" because they are present in yeast lacking linker histone. They are also observed after MNase-ExoIII digestion of chromatin reconstituted using recombinant core histones. We propose that the pseudo-chromatosome provides a DNA framework to facilitate H1 binding. Comparison of budding yeast nucleosome sequences obtained using micrococcal nuclease (MNase-seq) and MNase + exonuclease III (ExoIII) (MNase-ExoIII-seq): wild type cells and hho1-null cells. Nucleosome sequences from native chromatin and H1-depleted chromatin from mouse liver. Nucleosome sequences from a plasmid reconstituted into nucleosomes using recombinant yeast histones or native chicken erythrocyte histones.
Project description:In humans, there are eleven subtypes of linker histones that exhibit cell- and tissue-specific expression. Linker histone H1 proteins bind to both the core histones and linker DNA of chromatin fibers; and not only participate in control of gene activity but also serve to stabilize higher order chromatin structure. To determine the potential roles of linker histones in differentiation, we examined the global distribution of linker histone subtype H1.5 in human IMR90 fibroblasts and H1 embryonic stem cells (hESCs). Surprisingly, H1.5 binds to and represses a large fraction of gene family clusters in fully differentiated cell types representing all three embryonic germ layers. Little or no H1.5 enrichment at gene family clusters was detected in undifferentiated hESCs or partially differentiated somatic cells. We also found that SIRT1 histone deacetylase and H3K9me2, a repressive histone modification, are also enriched at gene family cluster in IMR90 cells, likely generating a stably repressive chromatin domain. To find out the mechanism of H1.5 targeting, H1.5 or SIRT1 was depleted in IMR90 cells by siRNA, and the binding patterns of SIRT1 and H1.5 were examined. In H1.5 knockdown cells, SIRT1 binding pattern was changed dramatically, and this changed pattern highly correlates to SIRT1 distribution in hESC. However, depletion of SIRT1 could not change the global binding pattern of H1.5. Depletion of H1.5 or SIRT1 leads to up-regulation of ~50% gene family clusters. However, the sets of gene family clusters that are affected by these two factors are different, suggesting H1.5 and SIRT1 may regulate gene transcription via different pathways. One-color array. Two replicates for each sample.
