Human nuclear pregnane X receptor cross-talk with CREB to repress cAMP activation of the glucose-6-phosphatase gene

Abstract

The nuclear PXR (pregnane X receptor) was originally characterized as a key transcription factor that activated hepatic genes encoding drug-metabolizing enzymes. We have now demonstrated that PXR also represses glucagon-activated transcription of the G6Pase (glucose-6-phosphatase) gene by directly binding to CREB [CRE (cAMP-response element)-binding protein]. Adenoviral-mediated expression of human PXR (hPXR) and its activation by rifampicin strongly repressed cAMP-dependent induction of the endogenous G6Pase gene in Huh7 cells. Using the -259 bp G6Pase promoter construct in cell-based transcription assays, repression by hPXR of PKA (cAMP-dependent protein kinase)-mediated promoter activation was delineated to CRE sites. GST (glutathione transferase) pull-down and immunoprecipitation assays were employed to show that PXR binds directly to CREB, while gel-shift assays were used to demonstrate that this binding prevents CREB interaction with the CRE. These results are consistent with the hypothesis that PXR represses the transcription of the G6Pase gene by inhibiting the DNA-binding ability of CREB. In support of this hypothesis, treatment with the mouse PXR activator PCN (pregnenolone 16alpha-carbonitrile) repressed cAMP-dependent induction of the G6Pase gene in primary hepatocytes prepared from wild-type, but not from PXR-knockout, mice, and also in the liver of fasting wild-type, but not PXR-knockout, mice. Moreover, ChIP (chromatin immunoprecipitation) assays were performed to show a decreased CREB binding to the G6Pase promoter in fasting wild-type mice after PCN treatment. Thus drug activation of PXR can repress the transcriptional activity of CREB, down-regulating gluconeogenesis.

Figure 1. Repression of cAMP-dependent gene expression in Huh7 cells

Human hepatocarcinoma Huh7 cells were infected with adeno-β-gal or adeno-hPXR at an MOI of 15. After 16 h of infection, the cells were washed with PBS and were incubated in serum-free medium in the presence or absence of rifampicin (10 μM) and cAMP (100 μM) for an additional 14 h. (A) Western blot analysis using cell lysates to determine the expression of hPXR. (B) Quantitative real-time PCR was performed as described in the Experimental section. Relative levels of mRNAs were expressed by taking the levels in DMSO-treated Huh7 cells as 1. Results are means±S.D. for three experiments.

Figure 2. Repression of PKA–CRE-mediated activity

The −259 bp G6Pase promoter–luciferase reporter plasmid and its mutant constructs were co-transfected with pCR3/hPXR and/or pCR3/hPKA, and the reporter activity was measured as described in the Experimental section. (A) The −259 bp wild-type promoter, (B) the −259 bp_irs/hnf4 promoter bearing the double mutations of IRS and HNF4 sites, (C) the −259 bp_{irs/cres/hnf4} promoter having the triple mutations at the IRS, two CRE and HNF4 sites. pRL-CMV was always co-transfected for the control. Total amount of DNA was adjusted by adding pcDNA3-V5-His as an empty vector control. Relative fold activities were calculated by taking the activity of the cells that were transfected by the reporter plasmid only in the presence of DMSO as 1. Results are means±S.D. for three experiments.

Figure 3. Repression of the G6Pase promoter through insulin-response signal and HNF4 sites

The −259 bp_hnf4 and −259 bp_irs G6Pase promoters were co-transfected with pCR3/hPXR and/or pcDNA3/mFoxO1, pCR3/hHNF4 or pCR3/hPGC1α, and the reporter activity was measured as described in the Experimental section. (A) The −259 bp_hnf4 promoter bearing the mutation of the HNF4 site, and (B) the −259 bp_irs promoter having the mutation of the IRS site. pRL-CMV was always co-transfected as the control. Relative fold activities were calculated by taking the activity of the cells that were transfected by the reporter plasmid only in the presence of DMSO as 1. Results are means±S.D. for three experiments.

Figure 4. Formation of a PXR complex with CREB

(A) GST pull-down assays were performed using an in vitro-translated ³⁵S-labelled hPXR and CREB with GST–CREB and GST–hPXR fusion proteins respectively in the presence of 0.1% DMSO (DM) or 50 μM rifampicin (Rif). GST was used as negative control as described in the Experimental section. (B) Immunoprecipitation assays: Huh7 cells were transfected with pCR3/FLAGhPXR and pcDNA3/hCREBV5 for 24 h. Subsequently, these cells were treated with rifampicin (Rif; 10 μM) for 2 h, followed by the treatment with cAMP (100 μM) or DMSO (DM) for additional 30 min. Cell extracts were prepared and subjected to Western blotting (WB) of CREB phosphorylation as well as immunoprecipitation (IP) using anti-FLAG M2–agarose as described in the Experimental section. The anti-FLAG precipitants for Western blotting were detected by anti-FLAG or anti-V5 antibodies. The assays were repeated three times.

Figure 5. Inhibiting CREB–CRE interaction via binding to the CREB DBD

GST pull-down assays were performed using a bacterially expressed GST–hPXR (A) or GST–CREB DBD (B) with the in vitro-translated ³⁵S-labelled hPXR, CREB and deleted fragments of CREB (full-length and amino acid residues 1–100, 1–160 and 1–280). (B) Gel-shift assays were performed using the in vitro-translated CREB, hPXR and hRXR and the radiolabelled oligonucleotides of wild-type CRE1 (CRE1_wt) and mutant CRE1 (CRE1_mt) as probes, as described in the Experimental section. Both supershift by anti-CREB antibody and competition with non-labelled oligonucleotides of CRE1_wt and CRE1_mt were carried out to confirm the specificity of the complex formation. The assays were repeated three times.

Figure 6. Repression of the G6Pase and PEPCK1 genes in vivo

(A) Mouse primary hepatocytes were prepared from wild-type (Wt) and PXR-KO mice as described in the Experimental section, and were treated with 0.1% DMSO or 10 μM PCN in the presence or absence of 0.75 mM cAMP for 16 h. Total RNA was extracted and subjected to quantitative real-time PCR analysis with the sets of specific probes. Relative levels were expressed by taking the level in DMSO-treated hepatocytes without cAMP treatment as 1. The assays were repeated three times. (B) Mice were treated with PCN or DMSO as described in the Experimental section. Liver RNAs were prepared from four mice for each of the wild-type (Wt) and PXR-KO groups and were individually subjected to quantitative real-time PCR. Relative mRNA levels were expressed by taking those with DMSO as 1. Results are means±S.D. of replicates (n=4). *P<0.0007, **P<0.0002 and ***P<0.03 for cAMP plus DMSO-treated wild-type hepatocytes compared with cAMP plus PCN-treated wild-type hepatocytes. ^#P<0.01, ^##P<0.007 and ^###P<0.00005 for the DMSO-injected wild-type group compared with the PCN-injected wild-type group.

Figure 7. Dissociation of CREB from the G6Pase promoter in vivo

ChIP assays were performed as described in the Experimental section. From the liver nuclear extracts, DNA fragments were immunoprecipitated with either rabbit normal IgG or an anti-CREB antibody, semi-quantified by PCR for the G6Pase promoter, separated on an agarose gel and visualized by ethidium bromide staining. D and P denote DMSO and PCN respectively; Wt, wild-type. The assays were repeated three times.

Figure 8. Schematic representation of cross-talk

Arrows indicate activation and co-activation, while stop bars indicate repression and co-repression.