1) Using ChIP-Chip and ChIP-Seq methodologies, we characterized GR regulatory elements throughout the murine genome. While many GR binding sites are found in the immediate vicinity of target promoters, a large set of GR interaction sites are located at great distances from any promoter element. 2) We have characterized the chromatin landscape at local GR interaction events. We find that GR binding to regulatory elements invariably results in local chromatin transitions identified as DNaseI hypersensitive sites (DHSs). While GR can bind to unremodeled nucleosomes in vitro, we report that GR interaction with the chromatin fiber always leads to the local reorganization of nucleosomes, indicating the chromatin remodeling is a universal feature of GR template binding. 3) GR induced chromatin remodeling in model systems has been correlated with Swi/Snf action. While a subset of chromatin transitions induced by GR in vivo also require this remodeling system, we find that many nucleosome remodeling events are Swi/Snf independent. Furthermore, GR frequently interacts at sites already hypersensitive to DNaseI attack, and these sites can be either Swi/Snf dependent or independent. Thus GR can mobilize remodeling systems other than Swi/Snf, and can also interact at pre-existing sites of chromatin remodeling. There are four types of GR chromatin interactions, two GR-dependent or -independent groups each comprising subsets either requiring or independent of Swi-Snf action. 4) The local organization of the GR chromatin interaction sites is highly cell specific, and strongly correlated with GR transcriptional response. GR sites linked to promoters that are non-responsive in a given cell type are refractory to GR directed remodeling, and resistant to receptor binding. This leads to the hypothesis that local chromatin structural organization in a given cell type is a major determinant of tissue specific GR action.
Showing the most recent 10 out of 63 publications