Compound A influences gene regulation of the Dexamethasone-activated glucocorticoid receptor by alternative cofactor recruitment

The glucocorticoid receptor (GR) is a transcription factor of which the underlying gene regulatory mechanisms are complex and incompletely understood. The non-steroidal anti-inflammatory Compound A (CpdA), a selective GR modulating compound in various cell models, has been shown to favour GR-mediated gene repression but not GR-mediated gene activation. Shifting balances towards only a particular subset of GR gene regulatory events may be of benefit in the treatment of inflammatory diseases. We present evidence to support that the combination of CpdA with Dexamethasone (DEX), a classic steroidal GR ligand, can shape GR function towards a unique gene regulatory profile in a cell type-dependent manner. The molecular basis hereof is a changed GR phosphorylation status concomitant with a change in the GR cofactor recruitment profile. We subsequently identified and confirmed the orphan nuclear receptor SHP as a coregulator that is specifically enriched at GR when CpdA and DEX are combined. Combining CpdA with DEX not only leads to stronger suppression of pro-inflammatory gene expression, but also enhanced anti-inflammatory GR target gene expression in epithelial cells, making ligand combination strategies in future a potentially attractive alternative manner of skewing and fine-tuning GR effects towards an improved therapeutic benefit.


Supplementary figure S1. Combination of DEX with CpdA reduces the transactivation activity on a GRE-dependent reporter.
(a) A549 cells with stably integrated p(GRE) 2 -50-luc+, a GRE-dependent promoter construct, were pre-incubated with CpdA (10µM) for 1h, after which DEX (1µM) was added for 5h. Promoter activities are expressed as relative induction factor versus Solvent (+ SEM, n=2). (b) For the analysis of cell viability, a CellTiter-Glo luminescent cell viability kit was used to measure ATP production. Results are reported as fold change in ATP versus Solvent (+ SEM, n=2).

Desmet et al.
Suppl. Figure S1 a b "Prophylactic" set-up "Therapeutic" set-up c b a Desmet et al.
Suppl. Figure S2 Supplementary figure S2. Simultaneously combining CpdA with a non-saturating DEX concentration suppresses particular inflammatory genes at the mRNA level (a) Scheme of the experimental set-ups (b) A549 cells were stimulated with TNFα (2000 units/ml) for 1h, after which CpdA (10µM) and DEX (10nM) were added for 3h. Total RNA was extracted and subjected to RT-qPCR. Expression values were normalized to the reference genes b-Actin and HPRT using qBase+. Three independent replicates were performed. Means + SE, obtained as predictions from the HGLMM fitted to the data, are shown on the original scale for CCL2, CCL5, TNFα and ICAM.
(c) A549 cells were stimulated with CpdA (10µM) and DEX (10nM) for 1h, after which TNFα (2000 units/ml) was added for 4h. Total RNA was extracted and subjected to RT-qPCR. Expression values were normalized to the reference genes b-Actin and HPRT using qBase+. Three independent replicates were performed. Means + SE, obtained as predictions from the HGLMM fitted to the data, are shown on the original scale for CCL2, CCL5, TNFα and ICAM.
The significance of gene-specific ligand combination effects on inflammatory gene expression, estimated as differences (on the transformed scale) to the gene-specific reference level CpdA/DEX, were assessed using a t-test (*: p<0.05; **: p<0.01; ***: p<0.001).

Serum-starved b
Not serum-starved 6h a 3h 5h c CpdA/DEX simultaneously Supplementary figure S3. The effect of CpdA stimulation on DEX-induced gene expression is gene-specific (a) Serum-starved A549 cells were pre-incubated with CpdA (10µM) for 1h, after which DEX (1µM) was added for 5h. Total RNA was extracted and subjected to RT-qPCR. Expression values were normalized to the reference genes Cyclo and HPRT using qBase+. Five independent replicates were performed. Means + SE, obtained as predictions from the HGLMM fitted to the data, are shown on the original scale for FKBP5, GILZ and DUSP1. (b) A549 cells were pre-incubated with CpdA (10µM) for 1h, after which DEX (1µM) was added for 5h. Total RNA was extracted and subjected to RT-qPCR. Expression values were normalized to the reference genes B2M and HPRT using qBase+. Three independent replicates were performed. Means + SE, obtained as predictions from the HGLMM fitted to the data, are shown on the original scale for FKBP5, DUSP1 and GILZ (c) A549 cells were simultaneously stimulated with CpdA (10µM) and DEX (10nM) for 3h or 5h. Total RNA was extracted and subjected to RT-qPCR. Expression values were normalized to the reference genes b-Actin and HPRT using qBase+. Three independent replicates were performed. Means + SE, obtained as predictions from the HGLMM fitted to the data, are shown on the original scale for FKBP5, DUSP1 and GILZ.
Supplementary figure S4. The GR protein levels in A549 and mIC cl2 cells A549 and mIC cl2 cells were pre-incubated with CpdA (10µM) for 1h, after which DEX (10nM) was added for 1h, 2h or 5h. Total cell lysates were prepared and subjected to Western Blot analysis. Actin served as a loading control. The middle lane was used to mark the molecular weight standard (n=2, representative figure).