Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 10, 214
eCollection

A Model of Glucocorticoid Receptor Interaction With Coregulators Predicts Transcriptional Regulation of Target Genes

Affiliations

A Model of Glucocorticoid Receptor Interaction With Coregulators Predicts Transcriptional Regulation of Target Genes

Federico Monczor et al. Front Pharmacol.

Abstract

Regulatory factors that control gene transcription in multicellular organisms are assembled in multicomponent complexes by combinatorial interactions. In this context, nuclear receptors provide well-characterized and physiologically relevant systems to study ligand-induced transcription resulting from the integration of cellular and genomic information in a cell- and gene-specific manner. Here, we developed a mathematical model describing the interactions between the glucocorticoid receptor (GR) and other components of a multifactorial regulatory complex controlling the transcription of GR-target genes, such as coregulator peptides. We support the validity of the model in relation to gene-specific GR transactivation with gene transcription data from A549 cells and in vitro real time quantification of coregulator-GR interactions. The model accurately describes and helps to interpret ligand-specific and gene-specific transcriptional regulation by the GR. The comprehensive character of the model allows future insight into the function and relative contribution of the molecular species proposed in ligand- and gene-specific transcriptional regulation.

Keywords: cofactor interaction; glucocorticoid receptor; nuclear receptor; receptor allosterism; transcriptional activity.

Figures

FIGURE 1
FIGURE 1
DEX and RU486, but not CYP, induce a dose-dependent and gene-specific response in A549 cells. (A) Gene expression response of three responsive genes, GILZ, SLC19A2, and THBD, to increasing concentrations of DEX measured by qRTPCR. Results are mean ± SEM of three independent experiments performed in triplicates. Fitted parameters are detailed in Table 1. (B) Binding affinity predictions corresponding to GREs present in proximal 5′UTR regions to transcription start sites (TSS) corresponding to GILZ, SLC19A2, and THBD. Results are expressed as GR binding scores (GBS), as previously described (Datson et al., 2011). (C) Gene expression response of three responsive genes, GILZ, SLC19A2, and THBD, to increasing concentrations of RU486 measured by qRTPCR. Results are mean ± SEM of three independent experiments performed in triplicates. Fitted parameters are detailed in Table 2. (D) Gene expression response of three responsive genes, GILZ, SLC19A2, and THBD, to increasing concentrations of CYP measured by qRTPCR. Results are mean ± SEM of three independent experiments performed in triplicates.
FIGURE 2
FIGURE 2
A multifactor complex model for ligand–nuclear receptor–DNA–coregulator interactions. (A) The simplest cubic representation of our model, based on the CTC model, describing the interactions between the GR (R), a ligand (L), a coregulator (C) and DNA (D). The equilibria indicated by arrows are governed by their corresponding equilibrium constants Ka, Kc, and Kd, which are modified by specific parameters α, β, γ, and δ described in Supplementary Information S1 and Supplementary Table S1. The mathematical depiction of the model is shown below (A). (B) Simulation of the effect of variations in α values (representing the effect of ligand binding on the binding of the coregulator or vice versa) on ligand-dependent response, as indicated by the model. (C) Simulation of the effect of variations in β values (representing the effect of coregulator binding on the DNA binding, or vice versa) on ligand-dependent response, as indicated by the model. (D) Simulation of the effect of variations in γ values (representing the effect of ligand binding on the DNA binding, or vice versa) on ligand-dependent response, as indicated by the model. (E) Simulation of the effect of variations in δ values (representing how the binding of any two partners affects the binding of the third) on ligand-dependent response, as indicated by the model. Note that only when the δ parameter is taken into consideration, simultaneous variation in both EC50 and Rmax can be simulated by the model.
FIGURE 3
FIGURE 3
GR knockdown affects GILZ response to DEX. (A) Western blot depicting the effect of varying concentrations of a previously described specific siRNA targeting the GR (Fitzsimons et al., 2008) on GR protein levels in A549 cell lysates. The image shown is representative of five independent blots. (B) Quantification of the effect of varying concentrations of the specific siRNA used in (A) on GR expression. Results are expressed mean ± SEM of five independent blots. Statistically significant changes were identified using Student’s t-test. p < 0.05; ∗∗p < 0.01. (C) DEX dose-dependent effect on GILZ expression at decreasing GR expression levels induced by siRNA-induced GR knockdown. The calculated EC50 values were not affected by GR knockdown, while GILZ maximal response to DEX was significantly attenuated by increasing GR knockdown (Table 4). DEX-induced GILZ expression was detectable even at maximal GR knockdown (100 pmol siRNA), indicating GILZ induction is robust even at low levels of GR expression.
FIGURE 4
FIGURE 4
Quantification of GR LBD binding to coregulator-derived NR box peptides using a customized MARCoNI array. (A) Example image of the GR LBD binding signal detected from two MARCoNI arrays in the presence of vehicle (DMSO) or DEX. The heatmap indicates relative binding intensity. (B) Peptide dose-dependent binding isotherms detected for the 12 NR box peptides (Supplementary Table S3) in the absence of any GR ligand (vehicle = DMSO). (•) NRIP1_LxxLL185_173_195, (x) NRIP1_LxxLL21_8_30, (formula image) NRIP1_LxxLL266_253_275_C263S, (formula image) NRIP1_LxxLL380_368_390, (formula image) NRIP1_LxxLL500_488_510, (ο) NRIP1_LxxLL713_700_722, (formula image) NRIP1_LxxLL819_805_831, (Δ) NRIP1_LxxLL936_924_946, (∇) NRIP1_LxxML1068_1055_1077, (formula image) PRGC1_LxxLL144_130_155, (•) PPRB_LxxLL645_632_655, (∗) ZNHI3_LxxLL101_89_111. Results are expressed mean ± SEM of three independent experiments. The molar annotation as concentration refers to the molar concentration of the peptides in the spot solution.
FIGURE 5
FIGURE 5
DEX, RU486 and CYP induce specific GR LBD-to-NR box peptide binding profiles. (A) Characteristic binding profiles induced by 1 × 10-7 DEX (red), RU486 (blue) or CYP (green) obtained using the quantitative in vitro assay, MARCoNI. Modulation Index (MI) > 0 suggests ligand-favored binding, while MI < 0 suggests ligand-disfavored binding of a peptide compared to DMSO. (B) Heatmap depiction of details of ligand-induced binding of coregulator peptides using MARCoNI. (C) Venn diagrams showing the number of peptides whose binding was favored (left), unfavored (center) or unchanged (right) by GR ligands. In all cases, statistically significant changes relative to DMSO were identified by Student’s t-test. p < 0.05, ∗∗p < 0.01 or ∗∗∗p < 0.001.
FIGURE 6
FIGURE 6
RU486 and CYP differentially affect DEX-induced gene expression. Transcriptional response of three GR-responsive genes, GILZ, SLC19A2, and THBD, to increasing concentrations of DEX measured by qRTPCR. Cells were preincubated with 10-7M RU486 (A,C,E) or CYP (B,D,F). Results are expressed as mean ± SEM of three independent experiments performed in triplicates.
FIGURE 7
FIGURE 7
NR0B1 knockdown affects SLPI response to DEX but not to RU486. (A) DEX but not RU486 induces SLPI expression in A549 cells, as expected, coincubation with RU486 blocked DEX-mediated effects, ∗∗p < 0.01. (B) Effect of cell transfection with siRNA targeting NR0B1 on mRNA and (C,D) protein levels in A549 cells measured by qRTPCR and Western-blot, respectively. The expression of the glyceraldehyde 3-phosphate dehydrogenase (GADPH) was used as internal control and normalization in (C,D), ∗∗∗p < 0.001. (E) Transcriptional response of SLPI to increasing concentrations of DEX measured by qRTPCR in cell transfected with specific siRNAs against NR0B1 or a scramble siRNA control. The fitted parameters are detailed in Table 5. (F) Transcriptional response of SLPI to RU486 measured by qRTPCR. Results are expressed as mean ± SEM of three independent experiments performed in triplicates.

Similar articles

See all similar articles

Cited by 2 PubMed Central articles

References

    1. Adams M., Meijer O. C., Wang J., Bhargava A., Pearce D. (2003). Homodimerization of the glucocorticoid receptor is not essential for response element binding: activation of the phenylethanolamine N-methyltransferase gene by dimerization-defective mutants. Mol. Endocrinol. Baltim. 17 2583–2592. 10.1210/me.2002-0305 - DOI - PubMed
    1. Aoyagi S., Archer T. K. (2011). Differential glucocorticoid receptor-mediated transcription mechanisms. J. Biol. Chem. 286 4610–4619. 10.1074/jbc.M110.195040 - DOI - PMC - PubMed
    1. Atucha E., Zalachoras I., van den Heuvel J. K., van Weert L. T., Melchers D., Mol I. M., et al. (2015). A mixed glucocorticoid/mineralocorticoid selective modulator with dominant antagonism in the male rat brain. Endocrinology 156 4105–4114. 10.1210/en.2015-1390 - DOI - PubMed
    1. Bain D. L., Connaghan K. D., Maluf N. K., Yang Q., Miura M. T., De Angelis R. W., et al. (2014). Steroid receptor-DNA interactions: toward a quantitative connection between energetics and transcriptional regulation. Nucleic Acids Res. 42 691–700. 10.1093/nar/gkt859 - DOI - PMC - PubMed
    1. Bolton E. C., So A. Y., Chaivorapol C., Haqq C. M., Li H., Yamamoto K. R. (2007). Cell- and gene-specific regulation of primary target genes by the androgen receptor. Genes Dev. 21 2005–2017. 10.1101/gad.1564207 - DOI - PMC - PubMed

LinkOut - more resources

Feedback