Figure 1.
LXRs repress GR-induced transcriptional activity in HCT116 cells.
A. Over-expression of LXRs strongly suppresses dexamethasone-stimulated transcriptional activity of the MMTV promoter in HCT116 cells. HCT116 cells were transfected with pRShGRα together with pMMTV-Luc and pGL4.73[hRluc/SV40], and increasing amounts of pCMX-hLXRα and pCMX-hRXRα (0.03–1 µg), and were incubated in the presence or absence of 10−6 M dexamethasone. Upper panel: Bars represent mean ± S.E. values of the firefly luciferase activity normalized for the renilla luciferase activity in the presence or absence of 10−6 M dexamethasone. *: p<0.01, n.s.: not significant, compared to the condition treated with dexamethasone in the absence of LXRα/RXRα. Lower panel: Total RNA from transfected HCT116 cells was harvested and relative mRNA expression of LXRα and GR were measured with the SYBR-Green real-time PCR. Bars represent mean ± S.E. values of the relative ratios of LXRα mRNA vs. GR mRNA using RPLP0 as an internal control. Dex: dexamethasone, RLA: relative luciferase activity. B. Over-expression of GR reverses the negative effect of LXRα/RXRα on dexamethasone-stimulated transcriptional activity of the MMTV promoter. HCT116 cells were transfected with increasing amounts of pRShGRα (0.25–1.5 µg) together with pMMTV-Luc, pGL4.73[hRluc/SV40], pCMX-hLXRα and pCMX-hRXRα, and were incubated in the presence or absence of 10−6 M dexamethasone. Upper panel: Bars represent mean ± S.E. values of the firefly luciferase activity normalized for the renilla luciferase activity in the presence or absence of 10−6 M dexamethasone. *: p<0.01, n.s.: not significant, compared to the condition treated with dexamethasone in the absence of LXRα/RXRα. Lower panel: Total RNA from transfected HCT116 cells was harvested and relative mRNA expression of LXRα and GR were measured with the SYBR-Green real-time PCR. Bars represent mean ± S.E. values of the relative ratios of LXRα vs. GR mRNA using RPLP0 as an internal control. Dex: dexamethasone, RLA: relative luciferase activity. C. Over-expression of LXRα (left panel) or LXRβ (right panel) in the presence of indicated LXR ligand suppresses dexamethasone-stimulated GR transcriptional activity on the MMTV promoter. HCT116 cells were transfected with pRShGRα, pMMTV-Luc and pGL4.73[hRluc/SV40] together with pCMX-hLXRα or -mLXRβ and pCMX-mRXRα, and were incubated in the presence or absence of 10−6 M dexamethasone and/or indicated LXR ligands. Bars represent mean ± S.E. values of the firefly luciferase activity normalized for the renilla luciferase activity. *: p<0.01 compared to the condition treated with dexamethasone in the absence of LXR ligand. Dex: dexamethasone, RLA: relative luciferase activity, 22-R-HC: 22-R-hydroxycholesterol.
Figure 2.
Over-expression of LXRα regulates dexamethasone-stimulated GR transcriptional activity in a gene promoter-specific way.
A and B. Over-expression of LXRα/RXRα suppresses dexamethasone-stimulated transcriptional activity of the PEPCK (A) but not the GILZ (B) promoter. HCT116 cells were transfected with pPEPCK-luc (A) or pGILZ-Luc (B), together with pGL4.73[hRluc/SV40] and pRShGRα in the presence or absence of pCMX-hLXRα and pCMX-hRXRα, and were incubated with or without 10−6 M dexamethasone and/or 10−6 M GW3965. Bars represent mean ± S.E. values of the firefly luciferase activity normalized for the renilla luciferase activity. *: p<0.01, n.s.: not significant between the two conditions indicated. Dex: dexamethasone, RLA: relative luciferase activity. C. Over-expression of LXRα/RXRα does not affect dexamethasone-induced repression of the transcriptional activity of NF-κB. HCT116 cells were transfected with (κB)3-Luc and pGL4.73[hRluc/SV40] in the presence or absence of pRSV-RelA (p65)/pRSV-NF-κBI (p50), pRShGRα, and pCMX-hLXRα and pCMX-hRXRα, and were incubated with or without 10−6 M dexamethasone. Bars represent mean ± S.E. values of the firefly luciferase activity normalized for the renilla luciferase activity. *: p<0.01, n.s.: not significant, compared to the condition obtained in the absence of p65/p50 and dexamethasone treatment, or between the two conditions indicated. Dex: dexamethasone, RLA: relative luciferase activity.
Figure 3.
Ligand activation of endogenous LXRs differentially regulates dexamethasone-induced mRNA expression of glucocorticoid-responsive genes in a gene-specific fashion.
A, B, C, D, E, F and G. HepG2 cells were transfected with negative control siRNA for luciferase GL2 (white bars) or siRNAs for LXRα/β (black bars), and were treated with or without 10−6 M dexamethasone and/or 10−6 M GW3965. Total RNA was harvested and mRNA levels of G6Pase (A), PEPCK (B), GILZ (C), ABCG1 (D), LXRα (E), LXRβ (F) and GR (G) were measured with the SYBR-Green real-time PCR. Bars represent mean ± S.E. values of the fold induction of mRNA expression. *: p<0.01, n.s.: not significant compared to the condition transfected with the control siRNA in the presence of dexamethasone or between the two conditions indicated. Dex: dexamethasone. H. siRNAs for LXRα suppress expression of LXRα protein in HepG2 lysates. HepG2 cells were transfected with negative control siRNA for luciferase GL2 or siRNAs for LXRα, and were treated with or without 10−6 M dexamethasone and/or 10−6 M GW3965 for 24 hours, and cells lysates were prepared. hLXRα (upper panel), hGR (middle panel) and hβ-actin (lower panel) were visualized with their specific antibodies in Western lots. I. Ligand activation of endogenous LXRs does not influence dexamethasone-induced translocation of GR from the cytoplasm into the nucleus. HepG2 were treated for 2 hours with or without 10−6 M dexamethasone and/or 10−6 M GW3965. Whole cell lystates and nuclear extracts were run on 4–20% SDS-PAGE gels, blotted to nitrocellulose membranes and hGR (upper panel: nuclear GR, middle panel: whole cell GR) and hHSP4 (lower panel) were visualized with their specific antibodies in Western blots.
Figure 4.
GW3965 attenuates dexamethasone-induced hyperglycemia in rats and represses dexamethasone-induced G6Pase mRNA expression in mouse and rat livers and PEPCK mRNA expression in mouse livers.
A. GW3965 attenuates dexamethasone-induced increase of blood glucose levels in rats. Blood glucose levels were measured in rats treated with the compounds indicated. Bars represent mean ± S.E. values of the blood glucose concentration. *: p<0.01, compared to the animals treated with dexamethasone or between the two conditions indicated. Dex: dexamethasone. B, C and D. GW3965 suppresses dexamethasone-induced mRNA expression of glucocorticoid-responsive G6Pase in the livers of rats (B), G6Pase and PEPCK in wild type, but not LXRα/β knockout mice (C and D). Rats, wild type (WT) and LXRα/β knockout mice were orally gavaged for three days with GW3965, and were injected with dexamethasone or physiologic saline intramuscularly. Twenty-four hours after the injection, Total RNA was harvested from their livers and mRNA levels of rat G6Pase (B), mouse G6Pase (C) and mouse PEPCK (D) were measured with the SYBR-Green real-time PCR. Bars represent mean ± S.E. values of the fold induction of mRNA expression against vehicle. *: p<0.01, n.s.: not significant, compared to the animals treated with dexamethasone or between the two conditions indicated. Dex: dexamethasone, KO: knockout.
Figure 5.
GW3965 alters dexamethasone-induced mRNA expression of its responsive genes in rat liver in microarray analysis.
A, B and C: GW3965 modulates the transcriptional activity of a fraction of dexamethasone-responsive genes. Venn diagrams demonstrating the number of genes regulated by dexamethasone (Dex) and/or GW3965 (A), Dex and/or Dex+GW3965 (B) and GW3965 and/or Dex+GW3965 (C). The common biologic pathways regulated independently by Dex or GW3965 are also demonstrated in the right side of panel A. D. The 72-hour GW3965 treatment does not alter GR protein levels in the rat liver. Tissue lysates were prepared from rat livers treated with indicated compounds. GR (upper panel) and β-actin (lower panel) were visualized with their specific antibodies in Western blots. E. The number of genes regulated by Dex and GW3965 in respective biologic pathways is shown. Nomenclatures of the demonstrated biologic activities are based on the ontology of the Gene Ontology: http://www.geneontology.org. F. GW3965 does not influence mRNA expression of dexamethasone-suppressive genes in rat livers. Total RNA was harvested from rat livers and mRNA levels of Ccl9 (left upper panel), Cxcl12 (right upper panel) and H6pd (lower panel) were measured with the SYBR-Green real-time PCR. Bars represent mean ± S.E. values of the fold induction of mRNA expression. *: p<0.01, n.s.: not significant between the two conditions indicated. Dex: dexamethasone.
Table 1.
mRNA Expression of Selective Glucocorticoid-responsive Genes in the Liver of Rats Treated with Dexamethasone in the Absence or Presence of GW3965.
Figure 6.
LXRα/RXRα attenuates the association of GR to GREs in vitro and in vivo.
A. Upper panel: LXRα/RXRα suppresses association of the GR to its classic consensus GRE in vitro. HeLa cells were transfected with pCMX-hLXRα and pCMX-hRXRα, or their carrying plasmid pCMX, and were treated with or without 10−6 M dexamethasone for 2 hours. Nuclear extracts were harvested from the cells and the binding activity of GR to GREs was evaluated in the presence or absence of wild type (wt) or mutant (mut) GRE oligonucleotides (Oligos) using the TRansAM™ GR kit (Active Motif). Bars represent mean ± S.E. values of the absorbance at 450 nm in the presence or absence of 10−6 M of dexamethasone. *: p<0.01, compared to the condition transfected with the control plasmid in the presence of dexamethasone and in the absence of GRE Oligos or between the two conditions indicated. Dex: dexamethasone. Lower panel: Overexpression of LXRα/RXRα does not influence dexamethasone-induced translocation of GR from the cytoplasm into the nucleus. Nuclear extracts obtained from HeLa cells were run on 4–20% SDS-PAGE gels, blotted to nitrocellulose membranes and GR (upper panel: nuclear GR, middle panel: whole cell GR) and hHSP4 (lower panel) were visualized with their specific antibodies in Western blots. B. LXRα/RXRα overexpression differentially affects the association of GR to its GREs in ChIP assays. HeLa cells were transfected with the empty control pCMX vector, or pCMX-hLXRα and -hRXRα, treated with or without 10−6 M dexamethasone, and ChIP assays were performed using anti-hGRα, anti-hLXRα and anti-hRXRα antibodies. Bars represent mean ± S.E. values of the fold precipitation of G6Pase GREs (left panel) or GILZ GREs (right panel) determined in the SYBR Green-based real-time PCR. *: p<0.01, n.s.: not significant, compared to the condition transfected with the control plasmid in the presence of dexamethasone or between the two conditions indicated. Dex: dexamethasone, IP: immunoprecipitation. C. DBD of LXRα confers LXRα-mediated repression of GR transcriptional activity and binding to G6Pase GREs. HCT116 cells were transfected with pRShGRα together with pMMTV-Luc and pGL4.73[hRluc/SV40], pCMX-hRXRα and the pCDNA3-6myc plasmid expressing wild type LXRα (1–447) or indicated LXRα mutants. Cells were incubated in the presence (black columns) or absence (white columns) of 10−6 M dexamethasone. Using aliquots of cell lysates, luciferase assays were performed. ChIP assays were also performed by treating cells with 4% formaldehyde, and by precipitating LXRα/G6Pase GREs with anti-Myc-antibody. Bars represent mean ± S.E. values of the firefly luciferase activity normalized for the renilla luciferase activity (left panel) or of the fold precipitation of G6Pase GREs (right panel) determined in the SYBR Green-based real-time PCR in the presence or absence of 10−6 M dexamethasone. *: p<0.01, n.s.: not significant, compared to the condition treated with dexamethasone in the absence of LXRα/RXRα or between the two conditions indicated. NTD: N-terminal domain, DBD: DNA-binding domain, LBD: ligand-binding domain. D, E, F, G. LXRα/RXRα heterodimer binds to classic consensus and G6Pase GREs in gel mobility shift assays and reduces GR binding to its GREs. D. Nuclear extracts from HeLa cells or LXRα/RXRα recombinant proteins were incubated with 32P-labeled classic consensus GRE (left panel) or G6Pase GRE (right panel) with dexamethasone (Dex) in the presence or absence of an excess amount (50-fold molar excess) of unlabeled classic consensus GRE wild type (wt), GRE mutated (mut), G6Pase GRE wt, or G6Pase GRE mut. E. Nuclear extracts from HeLa cells were incubated with 32P-labeled classic consensus GRE in the presence or absence of dexamethasone (Dex) and in the presence or absence of anti-GR antibody (Anti-GR Ab). F. Nuclear extracts from HeLa cells were incubated with 32P-labeled classic consensus GRE in the presence or absence of dexamethasone (Dex) and/or GW3965 under the presence of increased amounts of LXRα/RXRα recombinant proteins (upper panel). The intensities of band A from 3 independent experiments including that shown in the upper panel were measured with the image J software and fold GRE binding was calculated by comparing to control (lane 1) (lower panel). Bars represent mean ± S.E. values of fold GRE binding from 3 independent experiments. *: p<0.01, compared to the conditions indicated. G. LXRα/RXRα recombinant proteins were incubated with 32P-labeled classic consensus GRE in the presence or absence of GW3965 and anti-LXRα antibody (Anti-LXRα Ab).
Table 2.
Oligonucleotides Used in SYBR Green real-Time PCR Analysis and Gel Mobility Shift Assays.