Analysis of chromatin dynamics during glucocorticoid receptor activation

Abstract

Steroid hormone receptors initiate a genetic program tightly regulated by the chromatin environment of the responsive regions. Using the glucocorticoid receptor (GR) as a model factor for transcriptional initiation, we classified chromatin structure through formaldehyde-assisted isolation of regulatory elements (FAIRE). We looked at dynamic changes in FAIRE signals during GR activation specifically at regions of receptor interaction. We found a distribution of GR-responsive regions with diverse responses to activation and chromatin modulation. The majority of GR binding regions demonstrate increases in FAIRE signal in response to ligand. However, the majority GR-responsive regions shared a similar FAIRE signal in the basal chromatin state, suggesting a common chromatin structure for GR recruitment. Supporting this notion, global FAIRE sequencing (seq) data indicated an enrichment of signal surrounding the GR binding site prior to activation. Brg-1 knockdown showed response element-specific effects of ATPase-dependent chromatin remodeling. FAIRE induction was universally decreased by Brg-1 depletion, but to varying degrees in a target specific manner. Taken together, these data suggest classes of nuclear receptor response regions that react to activation through different chromatin regulatory events and identify a chromatin structure that classifies the majority of response elements tested.

Fig 1

GR-responsive regions have varied chromatin-remodeling responses to receptor recruitment. (A) A total of 54 identified GR-responsive regions were tested for GR recruitment in A1-2 cells by ChIP analysis following 1 h of DEX treatment. Samples were analyzed by real-time PCR, and the graph shows the ChIP signal in ethanol vehicle- and DEX-treated samples after 1 h. Samples that did not demonstrate a minimum of 30% induction after GR activation were classified as not responsive (gray bars). The data are from one representative experiment of two replicates. (B) FAIRE analysis was performed on all 54 regions following 1 h of DEX treatment followed by real-time PCR analysis. The graph depicts the relative FAIRE signal at each locus in the presence of ethanol vehicle or DEX. Data are averages for a minimum of 4 biological replicates with standard errors; and regions determined for panel A not to have significant GR binding under these conditions are represented by gray bars. *, P < 0.05; **, P < 0.01. (C) Strength of GR binding, as determined for panel A, correlates with the induction of FAIRE signal, as determined for panel B. (D) Chromatin-remodeling events measured by FAIRE demonstrate rapid kinetics that are maintained for several hours after induction. A1-2 cells were treated with DEX, and samples were harvested at the indicated time points for FAIRE analysis. Samples were analyzed by real-time PCR using the indicated primer sets at genes that demonstrated significant increases in FAIRE signal at 1 h.

Fig 2

FAIRE induction does not indicate a loss of nucleosome occupancy at GR binding regions. A1-2 cells were treated with DEX or ethanol vehicle for 1 h, followed by ChIP analysis using antibodies specific for histone H3, histone H2A, and histone H3 lysine 4 trimethylation. Real-time PCR analysis was performed at a TSS of a GR-induced gene (A), regions that demonstrated a typical basal FAIRE signal that was significantly increase upon GR binding (B), a region that demonstrated no change in FAIRE upon GR binding (C), a region that showed high FAIRE signal that was also induced upon GR binding (D), and a region that showed very low FAIRE signal that was induced upon GR binding (E). Data are averages of 5 biological replicates with standard errors (*, P < 0.05). (F) ChIP assays were also performed with antibodies specific for histone H2B and H2A.Z at these regions.

Fig 3

FAIRE signal and induction correlate with nuclease sensitivity. A1-2 cells were treated with DEX or ethanol vehicle for 1 h, followed by DNase treatment of isolated nuclei for the indicated time points. DNase sensitivity was measured by the ability to amplify regions surrounding GR binding locations described for Fig. 2 using real-time PCR. The data are from a representative experiment, and standard deviations of the technical replicates are shown.

Fig 4

GR recruitment can initiate chromatin-remodeling events several nucleosomes from binding regions. (A) FAIRE experiments described for Fig. 1 were used to analyze regions adjacent to GR recruitment at the FKBP5 6000 (∼6-fold FAIRE increase) and PNMT (∼4-fold FAIRE increase) response elements. Data are averages for a minimum of 4 biological replicates with standard errors. *, P < 0.05; **, P < 0.01. (B) The regions surrounding the FKBP5 6000 response element were then tested for changes in DNase sensitivity as described for Fig. 3.

Fig 5

The majority of GR-responsive regions lie within a narrow window of chromatin openness, as measured by FAIRE. (A) The 40 GR-responsive regions were grouped based upon their distance from the nearest transcriptional start sites. Box-and-whisker plots of each group's fold FAIRE changes upon GR activation are shown. (B) Box-and-whisker plots of each group's basal FAIRE signal prior to GR activation. (C) Scatter plot of GR-responsive regions correlating (P < 0.05) the basal chromatin FAIRE signature to the ability of GR to induce changes in FAIRE signal. (D) Scatter plot of GR-responsive regions comparing the overall FAIRE signal after DEX induction to the amount of FAIRE signal induction.

Fig 6

GR-responsive regions show an increased FAIRE signal across the whole genome. (A) Average FAIRE signals, determined from FAIRE-seq data, surrounding 15,816 GR binding regions and 15,000 random loci. A total of 15,000 random GREs absent from the ChIP-seq data show no FAIRE signal. Average input sequencing data from T47D cells are also included to ensure no sequencing bias. Average signal on the y axis is represented as RPM (reads per million sequences), where 0.1 RPM corresponds to 3 reads. (B) The average FAIRE-seq signal of the 40 GR-responsive sites from Fig. 1 shows a profile similar to that of the larger 15,000-data-point set. (C) FAIRE-seq data were placed in 200-bp bins surrounding 15,671 GR binding sites, and the heat map represents the FAIRE signal surrounding those GR binding peaks. Bins are marked by the distance (in bp) of the bin to the peak of the GR binding site. The additional strip (on left) represents the distance of the GR binding site to the closest TSS.

Fig 7

SWI/SNF function is critical for chromatin reorganization at most GR-responsive sites. (A) A1-2 cells were transduced with lentivirus containing an inducible shRNA targeting the ATPase subunit of the SWI/SNF complex, Brg-1. A stable clone, termed A1-A3, was generated and tested for Brg-1 knockdown following 72 h of shRNA induction by treatment with 10 μg/ml doxycycline. Knockdown was analyzed by immunoblotting cell lysates with antibodies specific for Brg-1. Expression of GR and β-actin was also analyzed. (B) A1-A3 cells were treated with 10 μg/ml DOX for 72 h followed by treatment with DEX or ethanol vehicle for 8 h. Cells were harvested, and lysates were analyzed for luciferase assay and total protein. Data are average relative luciferase activities (luciferase activity/total protein) for three biological replicates with standard deviations. (C to E) A1-A3 cells were treated for 72 h with 10 μg/ml doxycycline followed by treatment with DEX or ethanol vehicle for 1 h. Cells were then subjected to FAIRE analysis by real-time PCR using the indicated primer sets. Data are averages for four biological replicates with standard errors. *, P < 0.05. (F) Average FAIRE signals in A1-A3 cells of 21 GR-responsive regions in the presence and absence of DEX and Brg-1 (**, P < 0.0001).