![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Pharmacology and Toxicology, Medical Sciences Building, University of Toronto, Toronto, Ontario, Canada
(Received November 9, 2007; accepted December 14, 2007)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
-[1-propynyl]estra-4,9-dien-3-one (RU486), suggesting GR involvement. This small magnitude increase in AHR levels was functionally significant; pretreatment of Hepa-1 cells with DEX caused a 75% increase in the maximum induction of an AHR-activated luciferase reporter plasmid by TCDD. A luciferase reporter under control of the proximal 2.5 kilobases of the mouse Ahr 5'-flanking region and promoter was induced approximately 2.5-fold by DEX when cotransfected with a mouse GR expression plasmid. This is the first demonstration that glucocorticoids increase AHR levels in hepatoma cells via a GR-dependent transcriptional mechanism, suggesting a novel aspect of cross-talk between the AHR and the GR.
). Although the role of the AHR in control of drug metabolism and endocrine disruption is partly understood (Safe, 1995), we know little about the regulation of the AHR itself by endocrine and other factors (Harper et al., 2006).
Our interest in endocrine control of AHR expression and function was stimulated by our observation that hypophysectomy results in a significant loss of hepatic AHR protein and TCDD-binding capacity in male rats without decreasing CYP1A1 induction in response to MC treatment (Timsit et al., 2002). This finding suggested that pituitary hormones and/or pituitary-dependent factors may act as positive regulators of hepatic AHR levels and aromatic hydrocarbon responsiveness. We are interested in identifying the pituitary-dependent factors that modulate hepatic AHR expression and function and determining the molecular mechanisms by which they act.
Our initial focus has been on adrenal glucocorticoids as potential candidates for the pituitary-dependent endocrine factors involved in modulation of AHR levels and activity. Several lines of evidence suggest that glucocorticoids can augment aromatic hydrocarbon responsiveness. Elevated levels of endogenous corticosterone (CORT) induced by stress result in enhanced induction of hepatic CYP1A1-catalyzed 7-ethoxyresorufin O-deethylation (EROD) in response to PAH exposure in rats (Konstandi et al., 2000). The in vivo induction of rat hepatic CYP1A1 and/or associated catalytic activities by MC is potentiated by the synthetic glucocorticoid dexamethasone (DEX) (Sherratt et al., 1989) and diminished by adrenalectomy (Nebert and Gelboin, 1969). Cell culture studies in PLHC-1 fish hepatocellular carcinoma cells (Celander et al., 1996) and H4IIE rat hepatoma cells (Lai et al., 2004) confirm that glucocorticoids enhance HAH- and/or PAH-dependent CYP1A induction. The ability of glucocorticoids to enhance the induction of CYP1A1 by PAHs is mediated, at least in part, by the presence of functional glucocorticoid-responsive elements (GREs) in the first intron of the rat CYP1A1 gene (Mathis et al., 1989). The enhancement of CYP1A inducibility by glucocorticoids may also be mediated in part by alterations in AHR protein levels. Treatment of H4IIE rat hepatoma cells with DEX resulted in increased binding of TCDD to the cytosolic AHR (Wiebel and Cikryt, 1990). Treatment of pregnant mice with cortisol yielded offspring with elevated levels of AHR mRNA and protein in craniofacial tissue (Abbott et al., 1994).
However, other lines of evidence do not support this positive influence of glucocorticoids on the AHR system and aromatic hydrocarbon responsiveness. Treatment of rat mammary fibroblasts with DEX resulted in a decrease in AHR protein levels (Brake et al., 1998). Adrenalectomy in male rats was reported to have no effect on the TCDD-binding capacity of the hepatic AHR as measured by isoelectric focusing (Carlstedt-Duke et al., 1979). Finally, adrenalectomized rats are highly sensitive to TCDD-induced lethality, and this toxic response can be ameliorated by CORT treatment (Gorski et al., 1988); however, modulation of lethality by adrenal steroids could occur via events downstream of the AHR. Although there are certainly effects of glucocorticoids on the AHR pathway, these effects appear to be complex and the molecular mechanisms and functional impacts remain poorly understood.
The goals of the present study were to determine whether glucocorticoids modulate AHR expression and function in Hepa-1 mouse hepatoma cells and to explore the molecular mechanisms involved. Hepa-1 cells express abundant levels of AHR protein and are highly responsive to TCDD and MC treatment (Riddick et al., 1994). These cells also possess a functional GR signaling pathway (Cuthill et al., 1987; Prokipcak and Okey, 1988), making them an excellent cell culture model to examine the influences of glucocorticoids on AHR expression and function.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
-minimum essential medium supplemented with 10% fetal bovine serum (Invitrogen, Carlsbad, CA) and maintained in an atmosphere of 5% CO2 and 95% air at 37°C. According to technical data provided by Invitrogen, our culture conditions resulted in cells being exposed to a background level of approximately 0.1 µM cortisol in the medium. For experiments involving cytosol and RNA isolation, cells were plated in 55-cm2 dishes and cultured for approximately 66 h to 65 to 80% confluence. Cells were then exposed to vehicle alone [0.1% dimethylsulfoxide (DMSO)], CORT (0.1, 1, and 10 µM), DEX (0.01, 0.1, and 1 µM), RU486 (1 µM), or a combination of DEX (0.1 µM) and RU486 (1 µM). Cells were harvested after a treatment period of 24 h. CORT, DEX, and RU486 were purchased from Sigma-Aldrich (St. Louis, MO).
Analysis of mRNA Levels by Semiquantitative RT-PCR. Total RNA was isolated from Hepa-1 cells by the acid guanidinium thiocyanate-phenol-chloroform extraction method (Chomczynski and Sacchi, 1987) using Tri Reagent (Sigma-Aldrich). RNA samples were then treated with 20 U DNase I (GE Healthcare Bio-Sciences, Baie d'Urfé, QC, Canada) at 37°C for 20 min to remove genomic DNA contamination. RNA yield and purity were assessed by determining the A260/A280 ratio (
1.7 for all samples), and RNA integrity was assessed by comparing the relative intensities of the 28S and 18S rRNA bands as visualized on ethidium bromide-stained agarose gels. For the reverse transcription step, RNA (1 µg) was incubated with oligo d(T)15 (2 µg; Roche Diagnostics, Laval, QC, Canada) at 60°C for 5 min. Primer-annealed samples were then incubated in a final volume of 40 µl with Moloney murine leukemia virus-reverse transcriptase (400 U; Invitrogen), RNA Guard (60 U; GE Healthcare Bio-Sciences), a 1 mM concentration of each 2'-deoxynucleoside 5'-triphosphate (Invitrogen), 10 mM dithiothreitol, and 1x RT buffer containing 50 mM Tris-75 mM KCl-3 mM MgCl2. Reactions were allowed to proceed for 60 min at 37°C, followed by incubation at 70°C for 10 min.
PCR primers were synthesized by ACGT Corporation (Toronto, ON, Canada) or Integrated DNA Technologies, Inc. (Coralville, IA). The specificity of primers against the mouse genome was confirmed by BLAST search (www.ncbi.nlm.nih.gov/BLAST/) and by Primer-UniGene Selectivity analysis (Boutros and Okey, 2004). PCR primer sequences were as follows: mouse AHR, 5'-GGTGCCTGCTGGATAATTCATCTG-3' (forward primer) and 5'-TCGTCCTTCTTCATCCGTCAGTG-3' (reverse primer) (Giannone et al., 1998); mouse β-actin, 5'-CTACAAGAGCTGCGTGTGG-3' (forward primer) and 5'-TAGCTCTTCTCCAGGGAGGA-3' (reverse primer) (Giannone et al., 1998); and mouse tyrosine aminotransferase (TAT), 5'-GCCAATCCTGGACAGAACAT-3' (forward primer) and 5'-TTCTGAAGGTGCCGCTTACT-3' (reverse primer) (Sakuma et al., 2004).
All PCR assays began with a hot start phase, typically 3 or 5 min at 94 or 95°C and ended with a final extension phase, typically 7 min at 72°C. Each 50-µl PCR sample contained input cDNA derived from 50 or 75 ng of RNA, Taq polymerase (10 U; Invitrogen), an appropriate concentration of each primer (0.08 or 0.4 µM), a 1.6 mM concentration of each 2'-deoxynucleoside 5'-triphosphate, and 1x PCR buffer containing 20 mM Tris-50 mM KCl-3 mM MgCl2. Cycling conditions were as follows: AHR and β-actin duplex reaction (95°C for 20 s, 58°C for 20 s, and 72°C for 40 s) for 19 to 21 cycles; TAT (94°C for 30 s, 53°C for 30 s, and 72°C for 40 s) for 28 to 35 cycles. Amplified PCR products (AHR, 503 bp; β-actin, 450 bp; and TAT, 230 bp) were separated on 6% polyacrylamide gels, stained with Vistra Green (GE Healthcare Bio-Sciences), and quantitated by PhosphorImager analysis (Molecular Dynamics, Sunnyvale, CA) using IPLabGel software (Signal Analytics Corporation, Vienna, VA). AHR signal intensity was normalized to the internal reference standard, β-actin, for semiquantitative analysis. TAT mRNA levels were assessed as a qualitative positive control response. PCR conditions (input cDNA and cycle number) were optimized to yield product within the exponential range of amplification.
Immunoblot Analysis. Cytosol was isolated from Hepa-1 cells in HEGD buffer (25 mM Hepes, 1.5 mM EDTA, 10% glycerol, and 1 mM dithiothreitol, pH 7.4) by differential centrifugation. Protein concentration was determined by the method of Bradford (1976). Cytosolic protein (10 or 50 µg) isolated from Hepa-1 cells was resolved by SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose (Hybond ECL; GE Healthcare Bio-Sciences). For detection of AHR protein, a rabbit polyclonal antibody directed against the N-terminal fragment of the AHR encoded by the b-1 allele (amino acids 1–402; BIOMOL Research Laboratories, Plymouth Meeting, PA) (Pollenz et al., 1994) was used at a 1:20,000 dilution, followed by a donkey anti-rabbit Ig-horseradish peroxidase conjugate (GE Healthcare Bio-Sciences) at a dilution of 1:10,000. For detection of GR protein, a rabbit polyclonal antibody directed against a human GR peptide (amino acids 346–367; ABR-Affinity BioReagents, Golden, CO) was used at a concentration of 5 µg/ml, followed by a donkey anti-rabbit Ig-horseradish peroxidase conjugate (GE Healthcare Bio-Sciences) at a dilution of 1:5,000. An enhanced chemiluminescence system (GE Healthcare Bio-Sciences) was used for protein detection, films were scanned on a HP Scanjet 3970 scanner (Hewlett Packard, Palo Alto, CA), and relative quantitation was performed using IPLabGel software. Immunoblot quantitative analyses were performed under conditions that yielded a linear relationship between the amount of cytosolic protein and immunoreactive signal intensity.
AHR Radioligand Binding by Sucrose Density Gradient Analysis. Hepa-1 cytosol (0.5 ml, 
2 mg of protein/ml) was incubated with 10 nM [3H]TCDD (26.2–26.7 Ci/mmol at the time of use; Chemsyn, Lenexa, KS) in the absence or presence of a 100-fold molar excess of nonradioactive 2,3,7,8-tetrachlorodibenzofuran (Dr. Stephen Safe, Texas A&M University, College Station, TX) for 75 min at 4°C. TCDD and 2,3,7,8-tetrachlorodibenzofuran were added to cytosol in DMSO in 5-µl volumes. After incubation, unbound radioligand was removed by treating samples with dextran-coated charcoal (1 mg/mg cytosolic protein), and samples were analyzed by sucrose density gradient centrifugation as described previously (Riddick et al., 1994). Gradients were fractionated using an ISCO model 640 Fractionator (Instrumentation Specialties Co., Lincoln, NE), and radioactivity in each fraction was measured by liquid scintillation spectrometry.
Electrophoretic Mobility Shift Assay. The following complementary synthetic DNA oligonucleotides were synthesized and purified by Integrated DNA Technologies, Inc.: 5'-GATCTGGCTCTTCTCACGCAACTCCG-3' and 5'-GATCCGGAGTTGCGTGAGAAGAGCCA-3'. The core nucleotides of a well characterized DRE from the mouse Cyp1a1 5'-flanking region are italicized. These oligonucleotides were annealed and radiolabeled with [
-32P]ATP (3000 Ci/mmol; GE Healthcare Bio-Sciences) using T4 polynucleotide kinase (GE Healthcare Bio-Sciences). Hepa-1 cytosol (25 µl; 2.75 mg of protein/ml) was incubated with 20 nM nonradioactive TCDD (Wellington Laboratories Inc., Guelph, ON, Canada) or the vehicle DMSO for 2 h at room temperature. TCDD was added to cytosol in DMSO in 1-µl volumes. Binding reactions were performed in a 25-µl reaction volume containing 16.7 µl of transformed cytosol (46 µg of protein), poly[d(I-C)] (212.5 ng; Roche Diagnostics), and 32P-labeled DRE probe (200,000 cpm, 0.5 ng) for 15 min at room temperature. Binding reactions were performed in HEGD buffer containing 50 mM NaCl. Protein-DNA complexes were analyzed by electrophoresis on nondenaturing 4% polyacrylamide gels, and radioactivity was detected and quantitated by PhosphorImager analysis using IPLabGel software. The difference in signal between TCDD- and DMSO-treated samples was used to calculate the amount of TCDD-inducible DRE binding.
|
). The glucocorticoid-inducible luciferase plasmid, pGRE-luc, was provided by Dr. Gordon Kirby (University of Guelph, Guelph, ON, Canada) (Gerbal-Chaloin et al., 2002).
A new luciferase plasmid driven by the proximal 5'-flanking region and promoter of the mouse Ahr gene, mAHR-pGL3, was generated as follows. C57BL/6 mouse genomic DNA was used as a template for PCR amplification of a fragment of the mouse Ahr gene spanning from position -2451 to +150 [relative to +1 denoting the G residue of the major transcription initiation site (Mimura et al., 1994)]. All numbering of nucleotide positions is according to Build 36 of the February 2006 (mm8) mouse genome assembly found on the University of California, Santa Cruz Genome Browser Web site (http://genome.ucsc.edu/cgi-bin/hgGateway). PCR primers were obtained from Integrated DNA Technologies, Inc. and had the following sequences: 5'-AACACTCGAGGTAGACTCCTTCCTAACTCAGCACACT-3' (forward primer) and 5'-CACCGCTCGAGGAGTCCGTCCACCAGTTCGTCCT-3' (reverse primer). Each primer contained a XhoI restriction site at the 5'-end to facilitate cloning of the PCR product into the XhoI site of the promoterless pGL3-Basic plasmid. The identity of the mAHR-pGL3 reporter plasmid was confirmed by restriction analysis and DNA sequencing.
Transient Transfection and Luciferase Assay. For all transfection studies, Hepa-1 cells were seeded in 12-well plates and cultured for approximately 24 h to 50% confluence, followed by the manipulations described below.
To assess induction of pGudluc1.1 activity by TCDD and MC, cells were initially exposed to vehicle (0.1% DMSO) or DEX (0.1 µM) for 24 h. After a change of medium, cells were cotransfected with pGudluc1.1 (1.4 µg) and pRL-TK (0.1 µg) using SuperFect reagent (QIAGEN, Valencia, CA). Immediately after transfection, cells were treated with vehicle (0.1% DMSO), TCDD (1 pM–1 nM), or MC (Sigma-Aldrich) (0.1 nM–1 µM). Cells were harvested after a treatment period of 24 h, cell extracts were prepared in 1x passive lysis buffer, and dual luciferase measurements (Promega) were performed using a TD-20/20 luminometer (Turner Designs, Sunnyvale, CA). Firefly luciferase activity was normalized to Renilla luciferase activity.
To assess the response of mAHR-pGL3 to histone deacetylase inhibitors, cells were transfected with pGL3-Basic (1.5 µg) or mAHR-pGL3 (1.5 µg) using SuperFect reagent. After culturing for 24 h, cells were treated with vehicle (0.1% DMSO or ethanol), n-butyrate (Sigma-Aldrich) (0.1–5 mM), or trichostatin A (TSA) (0.5–50 nM; Sigma-Aldrich). Cells were harvested after a treatment period of 24 h, and firefly luciferase activity was normalized to cellular protein concentration as determined by the method of Bradford (1976). The effects of n-butyrate and TSA on endogenous levels of AHR mRNA and protein were assessed according to the RT-PCR and immunoblot procedures described above.
To assess the response of mAHR-pGL3 to DEX, cells were cotransfected with pGL3-Basic (0.7 µg), pGRE-luc (0.7 µg), or mAHR-pGL3 (0.7 µg) and pRL-TK (0.1 µg) using SuperFect reagent. At the same time, cells were also cotransfected with 2.0 µg of a mouse GR expression plasmid, pSVmGR (Dr. John A. Cidlowski, National Institute of Environmental Health Sciences, Research Triangle Park, NC) (Webster et al., 1997), or the empty vector, pSV2 (American Type Culture Collection). After culturing for 24 h, cells were treated with vehicle (0.1% DMSO) or DEX (1 nM–1 µM). Cells were harvested after a treatment period of 24 h, and firefly luciferase activity was normalized to Renilla luciferase activity. Levels of GR protein were assessed according to the immunoblot procedure described above.
|
|
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
The role of the GR in the induction of AHR expression by DEX was tested using the GR antagonist RU486. At a concentration of 1 µM, RU486 blocked the GR-mediated induction of TAT mRNA caused by DEX (0.1 µM) without causing TAT induction on its own (Fig. 4). DEX alone at a concentration of 0.1 µM increased AHR mRNA, protein, and [3H]TCDD binding by 26, 33, and 35%, respectively, and these small magnitude increases were not observed when Hepa-1 cells were cotreated with DEX and RU486 (Fig. 4).
|
To examine whether small changes in the AHR protein level caused by DEX resulted in functionally important alterations in the AHR signaling pathway, we first determined the ability of TCCD to transform the cytosolic AHR in vitro to a DRE-binding form. By electrophoretic mobility shift assay, the amount of AHR · DRE binding elicited by a maximally effective TCDD concentration did not differ for cytosol isolated from Hepa-1 cells that were previously treated in culture with vehicle or DEX at concentrations of 0.01, 0.1, or 1 µM (Fig. 5).
|
|
; FitzGerald et al., 1996; Garrison and Denison, 2000). Before studying the DEX responsiveness of this novel reporter construct, we first confirmed the functionality of mAHR-pGL3 by checking for a strong induction in response to histone deacetylase inhibitors, as observed previously with other mouse Ahr-based luciferase constructs in Hepa-1 cells (Garrison and Denison, 2000; Garrison et al., 2000). In comparison with the promoterless pGL3-Basic construct, mAHR-pGL3 luciferase activity was increased approximately 7-fold by n-butyrate and TSA (Fig. 7A). Consistent with previous studies in wild-type Hepa-1 cells (Zhang et al., 1996; Garrison et al., 2000), the marked effects of histone deacetylase inhibitors on mAHR-pGL3 luciferase activity were accompanied by only small magnitude increases (<45%) in endogenous AHR mRNA (Fig. 7B) and protein (Fig. 7C) levels. Our initial characterization of the mAHR-pGL3 construct revealed no significant response to the following chemicals in Hepa-1 cells in comparison with the promoterless pGL3-Basic construct (data not shown): TCDD (1 pM–10 nM), MC (0.1 nM–1 µM), phenobarbital (10 µM–1 mM), dehydroepiandrosterone (1–100 µM), hydrogen peroxide (0.1–1 mM), cumene hydroperoxide (0.1 mM), butylated hydroxytoluene (1–100 µM), trans-stilbene oxide (1–100 µM), progesterone (0.1–10 µM), testosterone (0.1–100 µM), cyclic AMP (0.01–2 mM), retinoic acid (0.1–100 nM), phorbol 12-myristate 13-acetate (1–500 nM), and 5-aza-2'-deoxycytidine (0.01–10 µM).
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Glucocorticoids potentiate HAH- and/or PAH-dependent CYP1A induction (Nebert and Gelboin, 1969; Sherratt et al., 1989; Celander et al., 1996; Konstandi et al., 2000; Lai et al., 2004). Induction of other AHR target genes is augmented or diminished by glucocorticoids, suggesting multiple molecular mechanisms (Linder et al., 1999). Potentiation of CYP1A1 induction by glucocorticoids is GR-mediated and involves binding of the GR to multiple GREs, initially identified in the first intron of the rat CYP1A1 gene (Mathis et al., 1989). Conserved GREs are found in the first intron of the CYP1A1 gene from rat, mouse, and human (Linder et al., 1999). However, a dramatic species difference in the CYP1A1 response exists: DEX potentiates the induction of rat CYP1A1 by MC at the transcriptional level via a GR · GRE interaction in the first intron, whereas DEX inhibits the induction of human CYP1A1 by MC at the protein level via a GR-independent mechanism (Monostory et al., 2005).
Our study focused on a distinct mechanism that may contribute to potentiated induction of AHR target genes by glucocorticoids: alteration of AHR expression and function. The first major finding of our study is that DEX increases AHR mRNA, protein, and TCDD binding by approximately 50% in Hepa-1 cells. While the present study was nearing completion, two highly relevant reports appeared in the literature (Sonneveld et al., 2007; Dvo
ák et al., 2008). Dvoák et al. (2008) found that DEX decreases AHR mRNA levels in HepG2 human hepatoma cells without affecting AHR protein levels. DEX decreased the induction of an AHR-activated luciferase reporter by TCDD, and TCDD-induced EROD activity was also inhibited. It remains unclear how these effects are produced in the absence of a decrease in AHR protein levels. Sonneveld et al. (2007) found that DEX acts via the GR to augment the induction by TCDD of AHR-activated luciferase reporters, EROD activity, and several endogenous AHR target genes in rodent cells (rat H4IIE and mouse Hepa-1) but not human cells (HepG2 and T47D breast carcinoma cells). DEX increased AHR mRNA levels in rat H4IIE cells but not in human cells. Together with our current results, these reports solidify the importance of species difference in AHR · GR interactions: glucocorticoids have a positive impact on AHR expression and function in rodent hepatoma cells but a negative impact on AHR expression and function in human hepatoma cells.
The magnitude of the increase in AHR levels elicited by DEX in our study was relatively small, in the 50% range. Hepa-1 cells express very high basal levels of AHR protein, with concentrations that are at least 1 order of magnitude higher than rodent liver levels (Timsit et al., 2002). An upper ceiling may be placed on the glucocorticoid response in Hepa-1 cells by the high basal AHR levels. Also, background levels of cortisol, a lower affinity GR agonist, in the culture medium may have limited the DEX response to a small degree.
The second major finding of our study is that the increase in AHR levels caused by glucocorticoids is mediated by the GR via a transcriptional mechanism. Our evidence for GR involvement in AHR induction by glucocorticoids includes the following: 1) AHR induction occurred at DEX concentrations effective in elevating TAT mRNA, a prototypical GR target; 2) AHR induction by DEX was blocked by RU486, a GR antagonist; and 3) induction of the mAHR-pGL3 luciferase reporter by DEX was augmented in the presence of exogenous GR. These findings are consistent with a direct mechanism whereby DEX increases the rate of transcription of the Ahr gene via a GR · GRE interaction; however, it is important to consider other potential indirect mechanisms. We suspect that the pregnane X receptor (PXR) does not play a significant role in the induction of AHR by DEX. Although supramicromolar concentrations of DEX activate murine PXR (Kliewer et al., 1998), we used submicromolar concentrations of DEX that selectively activate the GR. RU486 is a murine PXR agonist at high micromolar concentrations (Kliewer et al., 1998), whereas the concentration of RU486 used in the present study (1 µM) was previously shown to have minimal effect on murine PXR activation (Kliewer et al., 1998) and a clear ability to inhibit DEX-induced TAT activity in rat hepatoma cells (Gagne et al., 1985). Submicromolar concentrations of DEX act via the GR to stimulate expression of PXR via a transcriptional mechanism (Pascussi et al., 2000). Although PXR may be a potential player in AHR induction by DEX, any role would seem to be minor compared with that for the GR.
We made use of our novel mAHR-pGL3 reporter construct to demonstrate for the first time that DEX increases Ahr promoter activity in a GR-dependent manner. The GRE consensus sequence is defined as 5'-GGTACANNNTGTTCT-3' (Lu et al., 2006), with the italicized positions known to tolerate substitution. An imperfect GRE half-site of uncharacterized function (5'-TGATCT-3') located at positions -1009 to -1004 of the mouse Ahr gene was noted previously (Mimura et al., 1994; FitzGerald et al., 1996; Garrison and Denison, 2000). Our bioinformatic analysis identified four potential GREs (with three or fewer mismatches from the consensus) within the cloned region of the mouse Ahr 5'-flank contained in the mAHR-pGL3 construct: positions -2432 to -2418, -1073 to -1059, -712 to -698, and -237 to -223. The GRE sequence located at positions -2432 to -2418 (5'-GGCACATAGTGTGCT-3') is particularly attractive as the two mismatches from the consensus are located in the italicized positions known to tolerate substitution. We do not yet know if these potential GRE sequences mediate the induction of the mAHR-pGL3 luciferase activity by DEX in Hepa-1 cells, but these sites are strong candidates for further study.
We and others (FitzGerald et al., 1996; Garrison and Denison, 2000) found that luciferase reporters driven by the mouse Ahr proximal 5'-flank and promoter are refractory to modulation by a wide range of chemicals. Of note, our mAHR-pGL3 construct showed no induction by TCDD or MC although there are putative DREs containing the conserved core (5'-GCGTG-3') at positions +54 to +58, +89 to +93, and +92 to +96. Although the mechanism is uncertain, we confirmed that two histone deacetylase inhibitors caused strong induction of mAHR-pGL3 luciferase activity in Hepa-1 cells (Garrison and Denison, 2000; Garrison et al., 2000).
The third major finding of our study is that a small increase in AHR levels caused by glucocorticoids has a functionally significant impact on aromatic hydrocarbon responsiveness. Pretreatment of Hepa-1 cells with DEX for 24 h increased AHR levels by approximately 50%, and the subsequent induction of an AHR-activated luciferase reporter (pGudluc1.1) by TCDD and MC was significantly augmented. The pGudluc1.1 reporter construct is devoid of functional GREs and, importantly, basal pGudluc1.1 luciferase activity was not elevated in Hepa-1 cells pretreated for 24 h with DEX relative to vehicle control. Thus, the enhanced responsiveness to TCDD and MC exposure can be attributed to a cellular effect of DEX during the 24-h pretreatment period and not to any direct effect of DEX on the reporter construct. The increase in AHR levels produced by DEX is a likely mechanistic explanation.
It is puzzling how DEX produces an increase in AHR levels and an enhanced transcriptional response to ligands without an apparent change in AHR · DRE binding. Differences in analytical sensitivity in experimental endpoints may be important, but we also note that cells contain distinct pools of AHR proteins that bind ligand but differ in transformation to DNA-binding status (Denison, 1992).
Do changes in AHR levels have a significant impact on the responsiveness of cells to ligands? As addressed in detail in our recent review (Harper et al., 2006), both receptor theory and experimental manipulations demonstrate that changes in AHR levels affect the CYP1A1 induction response. Our findings support the idea that relatively small increases in AHR protein levels result in significant augmentation of cellular responsiveness to HAHs and PAHs.
In conclusion, glucocorticoids increase AHR levels in mouse hepatoma cells via a GR-dependent transcriptional mechanism. This small increase in AHR levels is associated with an augmented transcriptional response to HAHs and PAHs. This work reveals that glucocorticoid potentiation of CYP1A1 induction is mediated not only by the presence of GREs in the first intron of the CYP1A1 gene but also by increasing AHR expression at the transcriptional level. The present work is important and novel as it provides the first direct evidence for the transcriptional induction of AHR expression by DEX. Alterations in glucocorticoid levels have the potential to modulate the responsiveness of organisms to the toxic and/or adaptive effects of aromatic hydrocarbons. Future studies will address additional details of the molecular mechanisms involved, and the in vivo relevance of this aspect of cross-talk between the AHR and the GR will be examined in adrenalectomized rodent models with glucocorticoid replacement.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
ABBREVIATIONS: AHR, aryl hydrocarbon receptor; HAH, halogenated aromatic hydrocarbon; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxin; PAH, polycyclic aromatic hydrocarbon; MC, 3-methylcholanthrene; DRE, dioxin-responsive element; CORT, corticosterone; EROD, 7-ethoxyresorufin O-deethylation; DEX, dexamethasone; GRE, glucocorticoid-responsive element; GR, glucocorticoid receptor; DMSO, dimethylsulfoxide; RU486, 17β-hydroxy-11β-[4-dimethylamino phenyl]-17-[1-propynyl]estra-4,9-dien-3-one; RT, reverse transcriptase; PCR, polymerase chain reaction; TAT, tyrosine aminotransferase; bp, base pairs; TSA, trichostatin A; ANOVA, analysis of variance; PXR, pregnane X receptor.
Address correspondence to: Dr. David S. Riddick, Department of Pharmacology and Toxicology, Medical Sciences Building, University of Toronto, Toronto, ON, Canada M5S 1A8. E-mail: david.riddick{at}utoronto.ca
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
ák Z, Vrzal R, Pávek P, and Ulrichová J (2008) An evidence for regulatory cross-talk between aryl hydrocarbon receptor and glucocorticoid receptor in HepG2 cells. Physiol Res, in press. expression in human hepatocytes: synergistic increase of CYP3A4 induction by pregnane X receptor activators. Mol Pharmacol 58: 361-372.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
, significantly different (p < 0.05) from all other treatment groups; 
