Corticosterone pattern-dependent glucocorticoid receptor binding and transcriptional regualtion

Ultradian glucocorticoid rhythms are highly conserved across mammalian species, however, their functional significance is not yet fully understood. Here we demonstrate that pulsatile corticosterone replacement in adrenalectomised rats induces a dynamic pattern of glucocorticoid receptor (GR) binding at ~3,000 genomic sites in liver at the pulse peak, subsequently not found during the pulse nadir. In contrast, constant corticosterone replacement induced prolonged binding at the majority of these sites. Additionally, each pattern further induced markedly different transcriptional responses. During pulsatile treatment, intragenic occupancy by active RNA polymerase II exhibited pulsatile dynamics with transient changes in enrichment, either decreased or increased depending on the gene, which mostly returned to baseline during the inter-pulse interval. In contrast, constant corticosterone exposure induced prolonged effects on RNA polymerase II occupancy at the majority of gene targets, thus acting as a sustained regulatory signal for both transactivation and repression of glucocorticoid target genes. The nett effect of these differences were consequently seen in the liver transcriptome as RNA-seq analysis indicated that despite the same overall amount of corticosterone infused, twice the number of transcripts were regulated by constant corticosterone infusion, when compared to pulsatile. Target genes that were found to be differentially regulated in a pattern-dependent manner were enriched in functional pathways including carbohydrate, cholesterol, glucose and fat metabolism as well as inflammation, suggesting a functional role for dysregulated glucocorticoid rhythms in the development of metabolic dysfunction. Overall design: 3 hour corticosterone infusions (0.0035 g CORT-HBC Sigma-UK in 1mL sterile saline) delivered intravenously to adrenalectomised Sprague Dawley (male) rats, in either a pulsatile or constant pattern. Pulsatile: 20 min infusions (at a rate of 1ml/hr) every hour (at 0-20 min, 60-80 min and 120-140 min). Constant: constant infusion (at a rate of 0.333 ml/hr). Dose matched hourly by rate. ChIP-Seq: Liver samples collected at 140 min and 180 min for both infusion patterns; corresponding to the 3rd pulse peak and nadir of the pulsatile infusion respectivly. Corticosterone treated samples were compared to pulsatile vehicle infusions (control tissue) collected at 140 min and 180 min. Separate ChIP assays conducted with antibodies for the glucocorticoid receptor or pSer2 RNA polymerase II (phosphorylated serine 2 residue within the carboxy terminal domain) and sequenced. For each ChIP-seq replicate/condition/timepoint, liver samples from 6 rats were collected (72 rats in total). From each replicate/condition/timepoint, 10 x 100ug chromatin aliquots were prepared and immunoprecipitated with antibodies to either GR or pSer2 Pol2 before being pooled to prepare a separate library for each replicate/condition/timepoint/antibody. Note: Two separate ChIP-seq experiments were performed independently to allow for concordant peak calling analysis. RNA-Seq: Liver samples collected at 180 min for both infusion patterns. Pulsatile corticosterone infusion compared to pulsatile vehicle infusion samples, and constant corticosterone infusion compared to constant vehicle infusion liver samples (N=3).