Selective Modulation of the Human Glucocorticoid Receptor Compromises GR Chromatin Occupancy and Recruitment of p300/CBP and the Mediator Complex

Abstract

Exogenous glucocorticoids are frequently used to treat inflammatory disorders and as adjuncts for the treatment of solid cancers. However, their use is associated with severe side effects and therapy resistance. Novel glucocorticoid receptor (GR) ligands with a patient-validated reduced side effect profile have not yet reached the clinic. GR is a member of the nuclear receptor family of transcription factors and heavily relies on interactions with coregulator proteins for its transcriptional activity. To elucidate the role of the GR interactome in the differential transcriptional activity of GR following treatment with the selective GR agonist and modulator dagrocorat compared to classic (ant)agonists, we generated comprehensive interactome maps by high-confidence proximity proteomics in lung epithelial carcinoma cells. We found that dagrocorat and the antagonist RU486 both reduced GR interaction with CREB-binding protein/p300 and the mediator complex compared to the full GR agonist dexamethasone. Chromatin immunoprecipitation assays revealed that these changes in GR interactome were accompanied by reduced GR chromatin occupancy with dagrocorat and RU486. Our data offer new insights into the role of differential coregulator recruitment in shaping ligand-specific GR-mediated transcriptional responses.

Keywords: chromatin immunoprecipitation; glucocorticoid receptor; glucocorticoids; proximity labeling; selective GR modulation.

Graphical abstract

Fig. 1

Overview of TurboID experiments using the T2A split/link design.A, concept of the split/link design. In the control cell line, TurboID and GR are separated by a T2A tag, leading to their expression as two separate proteins. In the target cell line, TurboID-mutTA-GR is expressed as a single protein due to T2A inactivation. Direct comparison of proteins identified in the target versus control cell line allows distinction between aspecific interactors and GR interactome components. Created with Biorender.com. B, hierarchical clustering of all TurboID conditions (Euclidian distance, 300 clusters, maximal 10 iterations). C, venn diagram demonstrating overlap in significantly enriched proteins in target versus control cell line (1% FDR, S₀ = 0.58), for each ligand. D–G, volcano plots representing changes in log₂(LFQ intensity) (x-axis) and -log₁₀(p values) (y-axis) between target and control cell lines for each condition. The two cut-offs represent S₀ = 0.58 in combination with 5% or 1% FDR. Full circles indicate significantly enriched proteins at the 5% FDR level. Bait is annotated in black, known GR interactors in purple, and components of the MLL3/4 complex in orange. BL, biotin ligase; FDR, false discovery rate; GR, glucocorticoid receptor; TbID, TurboID.

Fig. 2

Identified protein complexes associated with Dex-induced GR.Small circles represent significant interactors at a 5% FDR; larger circles represent significant interaction partners at a 1% FDR. The blue color scale indicates the log₂(fold change) between target and control cell line. The purple-yellow color scale corresponds to the confidence of the interaction according to the STRING database. FDR, false discovery rate; GR, glucocorticoid receptor.

Fig. 3

Changes in the enrichment of GR coregulators and mediator subunits between Dex, RU486, and Dagr.A, comparison of GR coregulator enrichments between Dex and Dagr. B, comparison of mediator subunit enrichments between Dex and Dagr. C, comparison of GR coregulator enrichments between Dex and RU486. D, comparison of mediator subunit enrichments between Dex and RU486. Proteins indicated in blue are at least two-fold more enriched in the Dex condition, proteins in gray are at least two-fold more enriched in the RU486 condition; proteins in yellow are at least two-fold more enriched in the Dagr condition. GR, glucocorticoid receptor.

Fig. 4

Evaluation of coregulator function in ligand-dependent GR signaling. A549 cells were transfected with 20 nM siRNA or aspecific control (siMOCK) for 72 h, followed by 6 h treatment with GR ligands A, evaluation of knockdown efficiency at mRNA level using RT-qPCR. Changes in mRNA levels between siRNA and siMOCK were statistically significant (p < 0.01) unless indicated otherwise (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001). B, validation at protein level using immunoblot analysis. Images are representative of three biological replicates. C, evaluation of functional effects of coregulator knockdowns via RT-qPCR. Data are normalized versus solvent condition for each siRNA target. A and C, graphs represent mean ± SEM (≥3 biological replicates, each in three technical replicates), normalized versus HPRT1 and RPLP0 mRNA levels. Statistically significant changes versus siMOCK were evaluated via 2-way ANOVA with Dunnett’s post hoc testing on log₂-transformed data and are indicated with black frames (p < 0.05). GR, glucocorticoid receptor; RT-qPCR, reverse transcription quantitative PCR.

Fig. 5

Assessment of ligand-induced changes in GR and p-PolII(S2) chromatin occupancy and target gene regulation.A and B, A549 cells were treated with ligands or solvent control for 1 h and crosslinked using 2 mM disuccinimidyl-glutarate and 1% formaldehyde. Nuclear extracts were subjected to pulldowns using GR (A) or p-PolII(S2) (B) antibodies or normal mouse IgG overnight, followed by DNA purification and RT-qPCR of input and IP samples. Cp values of standard curves were used for relative quantification, and data were expressed as percent versus total input. Graphs represent mean ± SEM of the target genes (three biological replicates, each in at least two technical replicates). Statistically significant changes between IgG and GR or p-PolII(S2) conditions were evaluated using unpaired t-tests; statistically significant changes between ligands were evaluated via one-way ANOVA with Sidak’s post hoc testing (ligand effects). (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001). C, evaluation of ligand effects on mRNA levels of GR target genes via RT-qPCR. A549 cells were transfected with siMOCK for 72 h, followed by 6 h treatment with GR ligands. Graphs represent mean ± SEM (≥3 biological replicates, each in three technical replicates), normalized versus HPRT1 and RPLP0 mRNA levels. Statistically significant changes between ligands were evaluated via one-way ANOVA with Sidak’s post hoc testing on log₂-transformed data (∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001). GR, glucocorticoid receptor; RT-qPCR, reverse transcription quantitative PCR.