QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: mol.105.018622v1 70/1/277    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Romero, L. Articles by Kirby, G. M. Search for Related Content PubMed PubMed Citation Articles by Romero, L. Articles by Kirby, G. M.

Chemical Inducers of Rodent Glutathione S-Transferases Down-Regulate Human GSTA1 Transcription through a Mechanism Involving Variant Hepatic Nuclear Factor 1-C

Department of Biomedical Sciences, University of Guelph, Guelph, Ontario, Canada

Received September 7, 2005; accepted April 11, 2006

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
The regulation of human GSTA1 by chemical inducers of rodent glutathione S-transferases (GSTs) and the regulatory role of hepatic nuclear factor (HNF) 1 was investigated in Caco-2 cells. Treatment of preconfluent and confluent cells with 12-O-tetra-decanoyl phorbol-13-acetate (TPA), 3-methylcholanthrene (3-MC), 2-tert-butyl-4-hydroxy-anisol (BHA), and phenobarbital (PB) reduced GSTA1 mRNA levels in preconfluent and confluent cells. Constitutive levels of GSTA1 and HNF1 mRNA were elevated 6.25- and 50-fold, respectively, in postconfluent cells compared with preconfluent cells. Overexpression of HNF1 in cells transfected with a GSTA1 promoter-luciferase construct (pGSTA1-1591-luc) resulted in dose-related increases in reporter activity not observed when an HNF1 response element (HRE) in the proximal promoter was mutated (pGSTA1-HNF1-luc). TPA, 3-MC, BHA, and PB reduced HNF1 mRNA levels in preconfluent and confluent cells and caused marked reductions in luciferase activity in pGSTA1-1591-luc transfectants. Transcriptional repression was abrogated with pGSTA1-HNF1-luc and with truncated constructs that eliminated a functional HRE. Moreover, cotransfection of pHNF1 with pGSTA1-1591-luc partially prevented the reduction in luciferase activity by rodent GST inducers. Immunoblot analysis of DNA binding studies indicate that variant (v)HNF1-C binding to HRE is increased in preconfluent cells treated with 3-MC, BHA, and PB. In addition, overexpression of vHNF1-C repressed GSTA1 transcriptional activity in luciferase reporter assays. Finally, treatment with 3-MC, BHA, and PB increased vHNF1-C mRNA levels in preconfluent cells. These data demonstrate that repression of human GSTA1 transcription by chemical inducers of rodent GSTs occurs, in part, through a mechanism involving the repressive action of vHNF1-C.

Only a few studies have addressed the regulation of human GST genes. Although the intron and exon structure of the two major human GSTA genes (A1 and A2) is highly homologous with those of the rat and mouse, there are significant differences in the regulation of transcription between the rodent and the human GSTA genes (Morel et al., 1994; Suzuki et al., 1994). For example, the human GSTA 5'-flanking region lacks the ARE and the xenobiotic-responsive element through which the rodent genes are regulated (Klone et al., 1992). Potential regulatory elements have been identified in the human GSTA1 promoter, including AP-1 and AP-2 consensus sequence and a glucocorticoid response element (Whalen and Boyer, 1998). In addition, a hepatic nuclear factor (HNF) 1 site localized to –182 and –170 of the GSTA1 promoter region functions as an enhancer (Klone et al., 1992; Clairmont et al., 1994). Moreover, promoter activity was not elevated in HepG2 cells transfected with CAT reporter plasmids driven by the 5'-flanking sequence of the human GSTA1 gene and subsequently treated with three archetypal rodent GSTA inducers: -naphthoflavone, 3-MC, or tert-butylhydroquinone (Suzuki et al., 1994). Likewise, treatment of HepG2 cells with the phorbol ester 12-O-tetradecanoyl phorbol-13-acetate (TPA) caused a decrease of GSTA mRNA levels to below 5% of controls (Eickelmann et al., 1995). Thus, despite strong similarities in protein primary structure, tissue distribution, and cDNA sequence, lack of sequence homology between the promoter regions of the human and rodent GSTA genes suggests that the transcriptional regulation may be dramatically different.

The Caco-2 intestinal epithelial cell line is derived from a human colonic adenocarcinoma. Caco-2 cells are widely used as an in vitro model of the intestinal epithelium because they spontaneously differentiate, in the postconfluent state, into mature enterocyte-like cells that form polarized monolayers sealed by tight junctions and display a well developed apical brush-border membrane (Anderberg et al., 1993). Upon differentiation, there is an increase in specialized enzymatic and structural features detectable as differentiation markers, including expression of dipeptidyl peptidase IV, sucrase isomaltase, intestinal alkaline phosphatase, and villin (Vecchini et al., 1997). Moreover, GSTA1 activity and protein expression is significantly increased as Caco-2 cells differentiate (Peters and Roelofs, 1989), making this cell line particularly appropriate for studying GSTA1 regulation and expression. Previous studies in our laboratory demonstrated that treatment with interleukin-1 down-regulates the expression of GSTA in Caco2 cells by a transcriptional mechanism involving an HRE (Romero et al., 2002). Therefore, we hypothesized that archetypical inducers of rodent GSTs reduced GSTA1 transcriptional activity through a mechanism involving HNF1. We postulated that increases in HNF1 expression were responsible for the elevated levels of GSTA1 in differentiated Caco-2 cells and that higher levels of HNF1 might compensate for the repressive effects of rodent GST inducers. We found that GSTA1 mRNA was markedly reduced in Caco2 cells incubated for 24 h with TPA, 3-MC, BHA, and PB. Therefore, reporter activity also was reduced in similarly treated Caco2 cells transfected with GSTA1 luciferase reporter plasmids. Using truncated constructs, this down-regulatory effect was mapped to a region between –355 and –165 bp upstream to the coding region and mutation of an HNF1 site within this region abrogated the effect. We demonstrate that treatment with xenobiotics reduced expression of the transcriptional activator HNF1 and decreased its binding to HRE in preconfluent Caco-2 cells. Moreover, 3-MC, BHA, and PB increased mRNA levels of the dominant repressive isoform vHNF1-C and increased vHNF1-C-HRE complex formation in preconfluent cells. These results demonstrate clear differences in the transcriptional regulation of GSTA1 by chemicals that normally induce GSTs in rodents and establish pivotal roles for HNF1 and vHNF1-C in the down-regulation of this gene.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Chemicals and Antibodies. 12-O-Tetradecanoyl phorbol-13-acetate, 3-methylcholanthrene, 2-tert-butyl-4-hydroxy-anisol, and dimethyl sulfoxide as well as monoclonal -tubulin antibody and horseradish peroxidase-conjugated goat anti-rabbit IgG were purchased from Sigma (Oakville, ON, Canada). Phenobarbital was obtained from BDH Inc. (Toronto, ON, Canada). Human recombinant GSTA1-1 antibody was obtained from Biotrin International (Dublin, Ireland), and HNF1-specific antibody was from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). [-³²P]ATP (3000 mCi/mmol) was purchased from GE Healthcare (Piscataway, NJ).

Cell Culture and Treatments. Caco-2 cells, obtained from American Type Culture Collection (Manassas, VA), were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum and 100 µg/ml penicillin and streptomycin under 5% CO₂ at 37°C. The cells were seeded in 60-mm dishes for mRNA isolation and cell extract preparation and in 75-cm² flasks for nuclear protein extraction. Cells were treated with 50 nM TPA, 5 µM 3-MC, 100 µM BHA, and 5 mM PB for 24 h at approximately 80% confluence. Test compounds were dissolved in dimethyl sulfoxide such that all cell cultures contained a final concentration of 0.1% dimethyl sulfoxide in the media.

Isolation of RNA and Real-Time RT-PCR. Total RNA was prepared using TRIzol reagent (Invitrogen, Burlington, ON, Canada) according to the manufacturer's instructions, and concentrations were determined by absorbance at 260 nm. Isolated RNA (1 µg/20-µl reaction volume) was used for first-strand cDNA synthesis using 0.1 µg of random primers, 20 units of RNase inhibitor, and 200 units of Moloney murine leukemia virus reverse transcriptase (Invitrogen). Real-time PCR was performed on a Roche light cycler using a DNA Master SYBR Green I kit (Roche Diagnostics, Mississauga, ON, Canada). The PCR reaction was done in a volume of 10 µl, containing 1 µl of SYBR Green I, 5 µM each primer, and 2 mM Mg⁺. The PCR parameters were 1 cycle of 95°C for 1 min and 35 cycles of 95°C for 15 s, 70°C for 5 s, and 92°C for 15 s. The mRNA levels were normalized against glyceraldehyde-3-phosphate dehydrogenase mRNA. Primer sequences used for real-time RT-PCR are shown in Table 1.

A standard curve was generated for the absolute quantification of vHNF1A/B and vHNF1-C mRNA levels by real-time RT-PCR. Complementary DNA for vHNF1A/B and vHNF1-C was produced by PCR amplification from total cDNA using primers specific for vHNF1A/B and vHNF1-C. Ten-fold dilutions of the PCR products were then used to construct the standard curve.

Plasmid Construction and Reporter Assays. The proximal –1591-bp promoter of the hGSTA1 gene was cloned into the pGL3 basic vector (Promega, Madison, WI) upstream of the firefly luciferase gene, by using the XhoI and KpnI sites. In brief, the –1591-bp fragment was amplified with specific oligonucleotide primers (designed from the 5'-flanking region of the hGSTA1 gene; accession no. X67663 [GenBank] ) using human genomic DNA as template. The amplified fragment was cloned into the pCR-TOPO cloning vector (Invitrogen) following product instructions. The GSTA1 promoter was released by incubating with KpnI and XhoI and inserted into the luciferase vector to create pGSTA1-1591-Luc. A series of reporter constructs with promoter truncations were also designed including pGSTA1-850-Luc, pGSTA1-355-Luc, pGSTA1-165-Luc, and pGSTA1-84-Luc. An HNF1 site at –182 to –170 bp was mutated by site-directed mutagenesis from the pGSTA1-1591 plasmid, using a QuikChange site-directed mutagenesis kit (QIAGEN, Mississauga, ON, Canada) and designated as pGSTA1-HNF1-Luc. Promoter sequences and the presence of the mutation were confirmed by direct sequencing. Primer sequences for the luciferase reporter gene constructs are shown in Table 2.

Promoter activity of the pGSTA1-Luc constructs was determined by the dual luciferase reporter assay system (Promega). Caco2 cells (1.2 x 10⁵ cells/well in 24-well plates) were transfected for 18 h in suspension in fetal bovine serum-DMEM that included each of the various hGSTA1 promoter-luciferase constructs (1 µg/well) as well as the Renilla reniformis luciferase plasmid pRL-TK (10 ng/well) in the presence of LipofectAMINE reagent (3 µl/well) (Invitrogen). Transfected cells were then incubated in supplemented DMEM containing 50 nM TPA, 5 µM 3-MC, 100 µM BHA, or 5 mM PB for 24 h at 37°C. Cells were harvested and dual luciferase assays were performed according to the manufacturer's instructions. The relative luciferase activity was calculated by normalizing firefly luciferase activity to that of R. reniformis luciferase. Similar control experiments were performed with a basic promoterless pGL3-luciferase plasmid. The expression plasmids pHNF1 and pvHNF1-C were kind gifts from Dr. Marie-Josée Vilarem (Institut National de la Santé et de la Recherche Médicale, Montpellier, France) and Dr. Moshe Yaniv (Institut Pasteur, Paris, France), respectively.

Nuclear Extract Preparation and Electrophoretic Mobility Shift Assay. Nuclear extracts were isolated from confluent Caco2 cells grown on 10-cm dishes after treatment with chemicals for 24 h, by a modification of the Dignam method (Dignam et al., 1983). In brief, cells were washed in ice-cold phosphate-buffered saline, harvested by scraping, and resuspended in 100 µl of ice-cold buffer A (10 mM HEPES, 1.5 mM MgCl₂, 10 mM KCl, pH 7.9, 0.5 M DTT, and 0.5 mM phenylmethylsulfonyl fluoride, containing 2 µg/ml pepstatin, 10 µg/ml aprotinin, and 2 µg/ml leupeptin) in a prechilled 1.5-ml microfuge tube. The cells were lysed by passage through a 25-gauge needle several times. Nuclei were pelleted by centrifugation at 14,000g at 4°C, extracted in 60 µl of buffer C (20 mM HEPES, 25% glycerol, 420 mM KCl, 1.5 mM MgCl₂, and 2 mM EDTA, pH 7.9, supplemented with protease inhibitors as described in buffer A), and incubated for 15 min on ice. Thereafter, 60 µl of buffer C (20 mM HEPES, 25% glycerol, and 0.2 mM EDTA, pH 7.9, with added protease inhibitors) were added. Cellular debris was then precipitated by centrifugation, and the supernatant was stored as nuclear extracts at –80°C. Protein concentrations of nuclear extracts were determined by the Bradford dye-binding technique using the Bio-Rad protein assay kit (Bio-Rad, Mississauga, ON, Canada). A double-stranded oligonucleotide of the HNF1 response element in the hG-STA1 was end-labeled with [-³²P]ATP and T₄ polynucleotide kinase (GE Healthcare) according to manufacturer's instructions, except that reactions were allowed to continue for 1 h at 37°C. The sequence of the HNF1-containing oligonucleotide was 5'-GAAAAGGAACACATTAACCAGTTT-3', and the mutant was 5'-GAAAAGGAACACATcgcCCAGTTT-3'. DNA binding reactions were performed in a 20-µl volume containing 0.5 ng of radiolabeled probe, 10 µg of nuclear extract, and 1x gel shift binding buffer using a Gel Shift Assay kit (Promega) according to manufacturer's protocol. Binding reactions were incubated at room temperature for 20 min. For competition assays, a 100-fold molar excess of unlabeled oligonucleotide was incubated with the samples. In supershift experiments, the nuclear extracts were preincubated with 2 µg of HNF1 antibody for 1 h. Protein-bound and free probes were separated by electrophoresis in 6% acrylamide gels in 0.5x Tris borate-EDTA buffer at 200 V for 15 min. The gels were dried and visualized by filmless autoradiographic analysis using the Typhoon 9410 gel and blot imager (GE Healthcare).

To identify the specific nuclear proteins bound to the HNF1 response element (HRE), biotinylated probes and streptavidin-agarose were used in an affinity purification procedure as described previously (Scotto et al., 1999). Binding reactions included 50 µg of nuclear extract and 5 µg of biotinylated HRE oligonucleotides identical in sequence to those used in EMSA. Streptavidin-agarose 50% (Novagen. Madison, WI) was treated with 1 mg/ml bovine serum albumin and 0.2 mg/ml salmon sperm DNA to block nonspecific binding and equilibrated for 1 h in binding buffer. Pretreated streptavidin-agarose was combined with DNA-protein complexes and incubated for 1 h at room temperature. Streptavidin-agarose was washed three times with phosphate-buffered saline, and proteins bound to biotinylated target DNA were eluted after a 1-h incubation in 1% SDS, 10 mM DTT.

SDS-Polyacrylamide Gel Electrophoresis, Western Blot Analysis, and Two-Dimensional Gel Electrophoresis. HNF1 protein was identified by Western blot analysis as described previously (Romero et al., 2002). In brief, equal volumes of affinity-purified nuclear protein were separated by SDS-PAGE on a 12% polyacrylamide gel and either stained with SYPRO stain or transferred to nitrocellulose and incubated for 1 h with either rabbit anti-human HNF1 or HNF1 (vHNF1) polyclonal antibody (Santa Cruz Biotechnology, Inc.) diluted to 1:250. After incubation with goat anti-rabbit peroxidase secondary antibody (Vector Laboratories, Burlington, ON, Canada), bands were detected by chemiluminescence (ECL Plus; GE Healthcare) and visualized using a Typhoon 9410 scanner (GE Healthcare).

Statistical Analysis. One-way analysis of variance was used to assess statistically significant differences among treatment groups. For each statistically significant effect, Fisher's least significant test was used for comparison between multiple group means. Significance was established at P < 0.05.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Effect of Rodent GST Inducers on Human GSTA1 and HNF1 mRNA Levels. There is limited sequence homology between the 5'-flanking regions of human, mouse, and rat class GSTs, suggesting that transcription may be regulated differently between these species. Indeed, previous studies have shown that GSTA1/2 mRNA levels in HepG2 and Caco-2 cells decrease when treated with TPA. To determine whether inducers of rodent GSTs induce mRNA levels of human GSTA1, we treated preconfluent, confluent, or postconfluent Caco-2 cells with TPA, 3-MC, BHA, or PB. GSTA1 mRNA levels in preconfluent cells were significantly reduced to 7% by TPA (P < 0.001), 39% by 3-MC (P < 0.01), 51% by BHA (P < 0.01), and 30% by PB (P < 0.001) relative to control levels (Fig. 1A). GSTA1 mRNA levels were also markedly reduced in confluent cells with reductions to 5% for TPA (P < 0.05), 12% for 3-MC (P < 0.05), 87% for BHA, and 46% for PB (P < 0.05) compared with control cells (Fig. 1B). There were no statistically significant reductions in GSTA1 mRNA levels in postconfluent cells; however, there was a 1.4-fold increase (P < 0.05) in GSTA1 mRNA levels in PB-treated cells (Fig. 1C). Compared with preconfluent control cells, there was a 6.25-fold (P < 0.001) increase in constitutive GSTA1 mRNA levels in postconfluent cells.

View larger version (15K): [in this window] [in a new window]   Fig. 1. Rodent GST inducers reduce human GSTA1 mRNA levels in preconfluent and confluent Caco-2 cells. Caco-2 cells at different stages of confluence: preconfluent (A), confluent (B), and postconfluent (C) were exposed for 24 h to TPA, 3-MC, BHA, and PB. Total RNA was isolated and GSTA1 mRNA levels were quantified by real-time PCR as described under Materials and Methods. Results are the mean of three independent experiments. Significant differences from control are designated as *, P < 0.05; **, P < 0.01; or ***, P < 0.001.

We have previously shown that reductions of GSTA1/2 mRNA by interleukin-1 were mediated by HNF1, so we examined HNF1 expression in Caco-2 cells treated with rodent GST inducers. Similar to GSTA1 repression observed in preconfluent cells, HNF1 mRNA levels were reduced to 8% by TPA (P < 0.001), 43% by 3-MC (P < 0.05), 50% by BHA (P < 0.05), and 19% by PB (P < 0.001) relative to control levels (Fig. 2A). In confluent cells, there were less marked reductions to 50% (P < 0.05) for TPA, 68% for 3-MC (P < 0.05), and 58% for BHA relative to controls (Fig. 2B). PB did not alter HNF1 mRNA levels. There were no statistically significant differences in treated and control postconfluent cells; however, there was a 50-fold increase in HNF1 levels in control cells compared with preconfluent control cells (Fig. 2C).

View larger version (15K): [in this window] [in a new window]   Fig. 2. HNF1 mRNA levels are reduced in preconfluent and confluent Caco-2 cells exposed to inducers of rodent GSTs. Caco-2 cells at different stages of confluence: preconfluent (A), confluent (B), and postconfluent (C) were exposed for 24 h to TPA, 3-MC, BHA, and PB. Total RNA was isolated and HNF1 mRNA levels were quantified by real-time PCR as described under Materials and Methods. Results are the mean of three independent experiments. Significant differences from control values are designated as *, P < 0.05; **, P < 0.01; or ***, P < 0.001.

Role of HNF1 in GSTA1 Transcriptional Activity. To determine the role of HNF1 in transcriptional regulation of GSTA1 and the involvement of HNF1 in down-regulation of GSTA1 by rodent GST inducers reporter assays were preformed using various luciferase reporter plasmid constructs containing truncations of the 5'-flanking region of GSTA1 (Fig. 3). Three of these constructs (pGSTA1-1591-Luc, pG-STA1-850-Luc, and pGSTA1-355-Luc) contained an HRE located between –182 and –170 bp upstream from the transcription start site. The HRE was excluded in the smallest truncation construct (pGSTA1-165-Luc) and was mutated in the largest construct (pGSTA1-HRE-Luc).

View larger version (12K): [in this window] [in a new window]   Fig. 3. Diagrammatic representation of GSTA1-LUC constructs.

View larger version (15K): [in this window] [in a new window]   Fig. 4. Overexpression of HNF1 increases GSTA1 transcriptional activity. Caco-2 cells were transiently transfected with pHNF1 and cotransfected with either pGSTA1-1591-luc (black columns) or pGSTA1-HRE-luc (white columns). Cells were harvested and luciferase activities were measured and corrected for differences in transfection efficiency based on R. reniformis activity. Luciferase activities (mean ± S.E.) are expressed relative to 100% control values from cell transfected with pGSTA1-1591-luc only. Results are derived from at least three separate experiments. Significant differences from control values are designated as *, P < 0.05 or ***, P < 0.001.

View larger version (17K): [in this window] [in a new window]   Fig. 5. Inducers of rodent GSTs reduce GSTA1 transcriptional activity. Caco-2 cells were transiently transfected with pGSTA1-1591-luc and then cultured for 24 h in the presence of TPA, 3-MC, BHA, and PB. Cells were harvested, and luciferase activities were measured and corrected for differences in transfection efficiency based on R. reniformis activity. Luciferase activities (mean ± S.E.) are expressed as a percentage relative to the 100% control values. Results are derived from at least three separate experiments. Statistically significant differences in luciferase activity are indicated as ***, P < 0.001.

View larger version (15K): [in this window] [in a new window]   Fig. 6. Reduced transcriptional activity of GSTA1 by rodent GST inducers is mediated by HRE. Caco-2 cells were transiently cotransfected with either pGSTA1-850-luc, pGSTA1-355-luc, pGSTA1-165-luc, or pGSTA1-HRE-luc and then cultured for 24 h in the presence of TPA (A), 3-MC (B), BHA (C), and PB (D). Cells were harvested, and luciferase activities were measured and corrected for differences in transfection efficiency based on R. reniformis activity. Luciferase activities (mean ± S.E.) are expressed as a percentage relative to the 100% control values. Results are derived from at least three separate experiments. Statistically significant differences in luciferase activity are indicated as * P < 0.05.

View larger version (17K): [in this window] [in a new window]   Fig. 7. Overexpression of HNF1 partially reverses GSTA1 transcriptional repression by rodent GST inducers. A, Caco-2 cells were transiently transfected with pGSTA1-1591-luc alone (white columns) and cotransfected with either pHNF1 (black columns) or empty expression plasmid (pcDNA3.1) and then cultured for 24 h in the presence of TPA, 3-MC, BHA, or PB. Cells were harvested and luciferase activities were measured and corrected for differences in transfection efficiency based on R. reniformis activity. Luciferase activities (mean ± S.E.) are expressed as a percentage relative to the 100% untreated control values. B, Caco-2 cells were transiently transfected with either pHNF1 (black columns) or an empty expression plasmid (pcDNA3.1) and then cultured for 24 h in the presence of TPA, 3-MC, BHA, and PB. Total RNA was isolated and GSTA1 mRNA levels were quantified by real-time PCR as described under Materials and Methods. Messenger RNA values are expressed as -fold differences relative to untreated controls. Results represent the mean of three independent experiments. Significantly different from untreated control; *, P < 0.05). Significantly different from cells transfected with pcDNA3.1; , P < 0.05.

Influence of Rodent GST Inducers on HNF1 Binding to GSTA1 HRE. To determine whether treatment of Caco-2 cells with TPA, 3-MC, BHA, and PB reduced binding of HNF1 to the HRE, gel shift assays were performed by incubating nuclear extracts from cells treated with the various rodent GST inducers and radiolabeled HRE probes (Fig. 8). The electrophoretic mobility shift observed with nuclear extracts incubated with the wild-type HRE probe and nuclear extract from control Caco-2 cells was abolished in the presence of 100-fold excess of nonradioactive HRE probe as well as in incubates in which the HRE probe was used. Inclusion of anti-HNF1 antibodies in the binding reaction resulted in a supershift of the HRE-protein complex (data not shown). Densitometry did not reveal any statistically significant differences in the intensity of the shifted bands from nuclear extracts derived from cells treated with TPA, 3-MC, BHA, and PB compared with controls.

View larger version (62K): [in this window] [in a new window]   Fig. 8. Gel shift analysis of protein binding to HRE. Nuclear extracts were prepared from cells that were incubated with either TPA, 3-MC, BHA, or PB for 24 h. Twenty micrograms of nuclear extracts was incubated for 20 min with 20 µg of radiolabeled HNF1-containing oligonucleotide. The arrowhead indicates DNA bound with HNF. For competition assays, a 100-fold molar excess of unlabeled HNF oligonucleotide was added to the incubation reaction mixture. Nuclear extracts incubated with the mutant HNF1-containing oligonucleotide (HNF1) are also shown.

View larger version (14K): [in this window] [in a new window]   Fig. 9. Expression of vHNF1C in Caco-2 cells. A, overexpression of vHNF1-C represses GSTA1 transcriptional activity. Preconfluent Caco-2 cells were transiently transfected with pGSTA1-1591-luc alone (white columns) and cotransfected with either pvHNF1-C (black columns) or an empty expression plasmid (pcDNA3.1). Cells were harvested, and luciferase activities were measured and corrected for differences in transfection efficiency based on R. reniformis activity. Luciferase activities (mean ± S.E.) are expressed as a percentage relative to the 100% untreated control values. B, vHNF1-C is expressed at high levels in preconfluent Caco-2 cells. Total RNA was isolated and vHNF1-C mRNA levels were quantified by real-time PCR as described under Materials and Methods. C, 3-MC, BHA, and PB increase vHNF1-C mRNA levels. Pre-confluent Caco-2 cells were exposed for 24 h to TPA, 3-MC, BHA, and PB. Total RNA was isolated and vHNF1-C mRNA levels were quantified by real-time PCR as described under Materials and Methods. Results are the mean of three independent experiments. Significant differences from control values are designated as *, P < 0.05.

View larger version (39K): [in this window] [in a new window]   Fig. 10. Immunodetection of HNF1 and vHNF1C binding to HRE. Binding reactions were carried out using 50 µg of nuclear extract from cells incubated with either TPA, 3-MC, BHA, or PB for 24 h and a biotinylated HRE probe identical to that used for the EMSA. The binding reactions were then incubated with streptavidin-agarose, washed, and HNF1 protein bound to biotin-target DNA was collected in 1% SDS, 10 mM DTT and analyzed by Western blot analysis using polyclonal anti-bodies to HNF1 or HNF1. The immunoblot shown represents one of two experiments with similar results.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
The transcriptional regulation of rodent GSTs by various xenobiotic inducers, including phenobarbital, planar aromatic compounds (Rushmore et al., 1990), antioxidants (Pinkus et al., 1996), indoles (Renwick et al., 1999), isothiocyanates (Hecht, 2000), oltipraz (Kang et al., 2003), and other dithiol-3-thiones (Kensler et al., 2000), has been extensively studied. Indeed, specific cis-acting regulatory elements such as the electrophile responsive element (ERE) and the ARE that are responsible for inducible expression of class GSTs in mice and rats have been identified and well characterized (Rushmore et al., 1991; Friling et al., 1992). In view of the protection against potent chemical carcinogens provided by GST inducers in rodents, there has been considerable interest in extrapolating the chemopreventive properties to humans. However, lack of sequence homology in the 5'-flanking regions of rodent and human -class GST genes and the absence of functional ARE and ERE regulatory regions suggests that control of human GSTA1 expression may be different, putting into question the rationale of extrapolating results from rodent GST regulation studies to humans. Indeed, the results of this study indicate that human GSTA1 responds in a negative manner to several rodent GST inducers via a mechanism mediated by differential expression of both transactivator and repressor isoforms of HNF1 as well as variable binding of these isoforms to the HRE in the proximal GSTA1 promoter.

Transcriptional repression of GSTA1 by TPA, 3-MC, BHA, and PB was manifested in Caco-2 cells as reduced GSTA1 mRNA levels and transcriptional activity. TPA produced the most profound repression with a reduction of GSTA1 mRNA levels to 7% of controls. A similar reduction of GST mRNA levels was observed previously in TPA-treated Caco-2 cells and HepG2 cells (Eickelmann et al., 1995). In that study, reduction was attributed to a post-transcriptional event (GST mRNA degradation); however, the results of our gene reporter assays clearly indicate that TPA represses GSTA1 transcription in Caco-2 cells. In mice, induction of GSTA1 by TPA follows overexpression of c-jun and c-fos and increased binding to the ERE by the Fos/Jun heterodimeric complex (AP-1) (Bergelson et al., 1994). Our reporter assays have eliminated the involvement of AP-1 in transcriptional repression of GSTA1 by TPA and have implicated HNF1 in the process. Although TPA is known to suppress the expression of other genes (Lim et al., 2005), the involvement of HNF1 has not previously been demonstrated.

Apart from the above-mentioned TPA study, repression of GSTA1 expression by chemical inducers of rodent GSTs has not previously been demonstrated in cultured human cells. In a study of human hepatocytes in primary culture, GSTA1 mRNA levels were increased by 3-MC, PB, and dithiolethiones; however, induction by 3-MC and PB was extremely variable, occurring at low levels in only one-third of hepatocyte cultures. Interindividual variability in responsiveness of P450 enzymes to inducers has been previously shown in human hepatocytes (Pichard et al., 1991). Lack of induction of GSTA1 by chemical inducers in Caco-2 cells may be due to differences in regulation between hepatocytes and enterocytes, because we also observed GSTA1 repression in HT-29 cells (data not shown). In Caco-2 cells, expression of the drug-metabolizing enzyme CYP1A1 is increased by various chemical inducers, including 3-MC and phenobarbital; however, other enzymes, including -glutamyltranspeptidase, and various membrane proteins associated with glucose utilization such as sucrase isomaltase and hexose transporters, are down-regulated (Boulenc et al., 1992; Carriere et al., 1998).

The HNF1 family of trans-acting transcription factors consists of isoforms of both HNF1 and vHNF1 (also called HNF1) generated by alternative splicing events that result in varying degrees of transactivating properties (Cereghini, 1996). The three isoforms of HNF1 enhance transcription, with isoforms B and C being more potent transactivators than the A isoform. Of the three isoforms of vHNF1 protein, vHNF1A and B enhance transcriptional activity, whereas the vHNF1-C isoform acts as a transdominant repressor when dimerized with other HNF1 proteins (Bach and Yaniv, 1993). The finding that HNF1 overexpression increased GSTA1 reporter activity in luciferase constructs with an intact HRE indicates that constitutive expression of GSTA1 is regulated by HNF1. Moreover, marked overexpression of GSTA1 in postconfluent Caco-2 cells correlated with elevated levels of HNF1 at that stage of confluence. These results support the previous association of reduced human GSTA2 expression with lower levels of HNF1 expression in kidney tissue and renal cell carcinoma (Clairmont et al., 1994). In that study, a functional HRE was identified within the GSTA2 promoter (GGAACATATTAAC) at the same location and with similar sequence homology to the HRE in the GSTA1 promoter (GGAACACATTAAC) identified in the present study (single base differences are underlined). The finding that overexpression of vHNF1-C significantly repressed GSTA1 transcriptional activity emphasizes the repressive role of this isoform. Moreover, the differential expression of vHNF1-C associated with degree of confluence may explain the low and high constitutive GSTA1 expression in preconfluent and postconfluent Caco-2 cells, respectively. Indeed, dedifferentiated hepatoma cells express high levels of vHNF1, whereas HNF1 is expressed in highly differentiated cells reflecting the different functional roles of these isoforms during differentiation (Baumhueter et al., 1988; Cereghini et al., 1988).

The repressive effect of chemical inducers on GSTA1 transcription was observed in preconfluent but not in postconfluent cells. The reason for this discrepancy may be related to the fact that treatment with xenobiotics reduced HNF1 and increased vHNF1-C mRNA levels (with the exception of TPA treatment) in preconfluent cells but not in postconfluent cells. These findings were also reflected in DNA binding assays in which the degree of HRE binding of HNF1 and vHNF1-C was decreased and increased, respectively, in pre-confluent cells treated with the xenobiotics. This suggests that HNF1 down-regulation and vHNF1-C up-regulation results in GSTA1 repression. It remains to be determined whether the mechanism underlying altered HNF1 expression involves altered transcription or mRNA degradation. The fact that overexpression of HNF-1 prevented transcriptional repression of GSTA1 by rodent GST inducers in luciferase reporter assays further implicates HNF-1 in the mechanism of transcriptional repression. Ceramide, a key regulator of apoptosis, represses rat GSTA2 transcription in H4IIE cells by reducing nuclear HNF1 levels via ubiquitination (Park et al., 2004). Ceramide also prevents induction of GSTA2 by oltipraz and tert-butylhydroquinone by inhibiting HNF1 DNA binding.

That transcriptional repression by rodent GST inducers does not occur when the HRE is deleted or mutated in luciferase reporter constructs indicates that down-regulation of GSTA1 is mediated through interaction of HNF1 proteins, and perhaps other nuclear proteins, and the HRE. This supports our previous demonstration of HRE involvement in GSTA1 repression by interleukin-1 (Romero et al., 2002). Although results of gel shift assays suggest that nuclear protein binding to the HRE is not altered by rodent GST inducers, the identity and composition of the proteins in these complexes were not resolved using this approach. However, Western blot analysis of HRE-bound proteins in DNA binding assays revealed that the relative amount of HNF1 and vHNF1 present in complexes varies considerably in untreated and xenobiotic-treated preconfluent and postconfluent cells. In view of the different transactivator and repressor functions of HNF isoforms, the composition of HNF1-HRE complexes probably has a profound impact on GSTA1 gene regulation. Thus, increased binding of vHNF1-C to HRE may explain repression of GSTA1 in preconfluent cells treated with 3-MC, BHA, and PB. Moreover, absence of vHNF1-C binding to HRE may explain the lack of repression by xenobiotics observed in postconfluent cells. Our results do not explain the mechanism underlying TPA-mediated repression of GSTA1 expression because TPA treatment reduced both vHNF1-C mRNA levels and vHNF1-C-HRE binding in pre-confluent cells. Moreover, preliminary experiments comparing HRE-bound proteins in TPA-treated and control cells by two-dimensional gel electrophoresis, did not (apart from a reduction in HNF1-C bound to HRE) reveal any treatment-related differences in the profile of HRE-bound proteins. The possibility of TPA-mediated modification of HNF1 function in GSTA1 transcription merits further investigation.

In conclusion, we have demonstrated that 1) constitutive expression of GSTA1 is regulated by interactions of HNF1 proteins and an HRE in the proximal promoter of the GSTA1 gene; 2) GSTA1 transcription is repressed by chemical inducers of rodent GSTs, including TPA, 3-MC, BHA, and PB; and 3) the mechanism of repression of GSTA1 expression by 3-MC, BHA, and PB in preconfluent Caco-2 cells is associated with increased binding of vHNF1-C to the HRE.

	Footnotes

 
This research was funded by the Canadian Institutes for Health Research. Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org.

doi:10.1124/mol.105.018622.

ABBREVIATIONS: GST, glutathione S-transferase; BHA, 2-tert-butyl-4-hydroxy-anisol; PB, phenobarbital; TPA, 12-O-tetradecanoyl phorbol-13-acetate; 3-MC, 3-methylcholanthrene; ARE, antioxidant-responsive element; AP, activator protein; HNF, hepatic nuclear factor; v, variant; luc, luciferase; DMEM, Dulbecco's modified Eagle's medium; PCR, polymerase chain reaction; RT-PCR, reverse transcriptase-polymerase chain reaction; bp, base pair(s); h, human; HRE, hepatic response element; DTT, dithiothreitol; EMSA, electrophoretic mobility shift assay; ERE, electrophile responsive element.

Address correspondence to: Dr. Gordon M. Kirby, Department of Biomedical Sciences, University of Guelph, Guelph, ON, Canada N1G 2W1. E-mail: gkirby{at}uoguelph.ca

	References

Top Abstract Materials and Methods Results Discussion References

 
Anderberg EK, Lindmark T, and Artursson P (1993) Sodium caprate elicits dilatations in human intestinal tight junctions and enhances drug absorption by the paracellular route. Pharm Res 10: 857–864.[CrossRef][Medline]

Bach I and Yaniv M (1993) More potent transcriptional activators or a transdominant inhibitor of the HNF1 homeoprotein family are generated by alternative RNA processing. EMBO (Eur Mol Biol Organ) J 12: 4229–4242.[Medline]

Baumhueter S, Courtois G, and Crabtree GR (1988) A variant nuclear protein in dedifferentiated hepatoma cells binds to the same functional sequences in the beta fibrinogen gene promoter as HNF-1. EMBO (Eur Mol Biol Organ) J 7: 2485–2493.[Medline]

Bergelson S, Pinkus R, and Daniel V (1994) Induction of AP-1 (Fos/Jun) by chemical agents mediates activation of glutathione S-transferase and quinone reductase gene expression. Oncogene 9: 565–571.[Medline]

Boulenc X, Bourrie M, Fabre I, Roque C, Joyeux H, Berger Y, and Fabre G (1992) Regulation of cytochrome P450IA1 gene expression in a human intestinal cell line, Caco-2. J Pharmacol Exp Ther 263: 1471–1478.[Abstract/Free Full Text]

Carriere V, Dussaulx E, Rousset M, Zweibaum A, and Cambier D (1998) Decreased expression of gamma-glutamyltranspeptidase in the intestinal cell line Caco-2 by inducers of cytochrome P450 1A1. Biochem Pharmacol 56: 817–823.[CrossRef][Medline]

Cereghini S (1996) Liver-enriched transcription factors and hepatocyte differentiation. FASEB J 10: 267–282.[Abstract]

Cereghini S, Blumenfeld M, and Yaniv M (1988) A liver-specific factor essential for albumin transcription differs between differentiated and dedifferentiated rat hepatoma cells. Genes Dev 2: 957–974.[Abstract/Free Full Text]

Clairmont A, Ebert T, Weber H, Zoidl C, Eickelmann P, Schulz WA, Sies H, and Ryffel GU (1994) Lowered amounts of the tissue-specific transcription factor LFB1 (HNF1) correlate with decreased levels of glutathione S-transferase alpha messenger RNA in human renal cell carcinoma. Cancer Res 54: 1319–1323.[Abstract/Free Full Text]

Dignam JD, Lebovitz RM, and Roeder RG (1983) Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res 11: 1475–1489.[Abstract/Free Full Text]

Ding VD and Pickett CB (1985) Transcriptional regulation of rat liver glutathione S-transferase genes by phenobarbital and 3-methylcholanthrene. Arch Biochem Biophys 240: 553–559.[CrossRef][Medline]

Eaton DL and Bammler TK (1999) Concise review of the glutathione S-transferases and their significance to toxicology. Toxicol Sci 49: 156–164.[Free Full Text]

Eickelmann P, Morel F, Schulz WA, and Sies H (1995) Turnover of glutathione S-transferase alpha mRNAs is accelerated by 12-O-tetradecanoyl phorbol-13-acetate in human hepatoma and colon carcinoma cell lines. Eur J Biochem 229: 21–26.[Medline]

Friling RS, Bergelson S, and Daniel V (1992) Two adjacent AP-1-like binding sites form the electrophile-responsive element of the murine glutathione S-transferase Ya subunit gene. Proc Natl Acad Sci USA 89: 668–672.[Abstract/Free Full Text]

Hayes JD, Judah DJ, McLellan LI, Kerr LA, Peacock SD, and Neal GE (1991) Ethoxyquin-induced resistance to aflatoxin B1 in the rat is associated with the expression of a novel alpha-class glutathione S-transferase subunit, Yc2, which possesses high catalytic activity for aflatoxin B1–8,9-epoxide. Biochem J 279: 385–398.[Medline]

Hayes JD and Pulford DJ (1995) The glutathione S-transferase supergene family: regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance. Crit Rev Biochem Mol Biol 30: 445–600.[Medline]

Hecht SS (2000) Inhibition of carcinogenesis by isothiocyanates. Drug Metab Rev 32: 395–411.[CrossRef][Medline]

Kang KW, Cho IJ, Lee CH, and Kim SG (2003) Essential role of phosphatidylinositol 3-kinase-dependent CCAAT/enhancer binding protein beta activation in the induction of glutathione S-transferase by oltipraz. J Natl Cancer Inst 95: 53–66.[Abstract/Free Full Text]

Kensler TW, Curphey TJ, Maxiutenko Y, and Roebuck BD (2000) Chemoprotection by organosulfur inducers of phase 2 enzymes: dithiolethiones and dithiins. Drug Metab Drug Interact 17: 3–22.[Medline]

Klone A, Hussnatter R, and Sies H (1992) Cloning, sequencing and characterization of the human alpha glutathione S-transferase gene corresponding to the cDNA clone pGTH2. Biochem J 285: 925–928.[Medline]

Lim K, Kang YS, Son MY, Park C, Yun EJ, Song KS, Kim JS, Kim YR, Park JI, Yoon WH, et al. (2005) Phosphorylation of octamer-binding transcriptional factor may be correlated with H2B histone gene repression during 12-O-tetradecanoylphorbol 13-acetate-dependent differentiation of HL-60 cells. Oncol Rep 14: 727–731.[Medline]

Morel F, Schulz WA, and Sies H (1994) Gene structure and regulation of expression of human glutathione S-transferases alpha. Biol Chem Hoppe Seyler 375: 641–649.[Medline]

Park IN, Cho IJ, and Kim SG (2004) Ceramide negatively regulates glutathione S-transferase gene transactivation via repression of hepatic nuclear factor-1 that is degraded by the ubiquitin proteasome system. Mol Pharmacol 65: 1475–1484.[Abstract/Free Full Text]

Peters WH and Roelofs HM (1989) Time-dependent activity and expression of glutathione S-transferases in the human colon adenocarcinoma cell line Caco-2. Biochem J 264: 613–616.[Medline]

Pichard L, Fabre I, Domergue J, Joyeux H, and Maurel P (1991) Effect of FK 506 on human hepatic cytochromes P-450: interaction with CyA. Transplant Proc 23: 2791–2793.[Medline]

Pinkus R, Weiner LM, and Daniel V (1996) Role of oxidants and antioxidants in the induction of AP-1, NF-B and glutathione S-transferase gene expression. J Biol Chem 271: 13422–13429.[Abstract/Free Full Text]

Renwick AB, Mistry H, Barton PT, Mallet F, Price RJ, Beamand JA, and Lake BG (1999) Effect of some indole derivatives on xenobiotic metabolism and xenobiotic-induced toxicity in cultured rat liver slices. Food Chem Toxicol 37: 609–618.[CrossRef][Medline]

Romero L, Higgins MA, Gilmore J, Boudreau K, Maslen A, Barker HJ, and Kirby GM (2002) Down-regulation of alpha class glutathione S-transferase by interleukin-1 in human intestinal epithelial cells (Caco-2) in culture. Drug Metab Dispos 30: 1186–1193.[Abstract/Free Full Text]

Rushmore TH, King RG, Paulson KE, and Pickett CB (1990) Regulation of glutathione S-transferase Ya subunit gene expression: identification of a unique xenobiotic-responsive element controlling inducible expression by planar aromatic compounds. Proc Natl Acad Sci USA 87: 3826–3830.[Abstract/Free Full Text]

Rushmore TH, Morton MR, and Pickett CB (1991) The antioxidant responsive element. Activation by oxidative stress and identification of the DNA consensus sequence required for functional activity. J Biol Chem 266: 11632–11639.[Abstract/Free Full Text]

Scotto C, Delphin C, Deloulme JC, and Baudier J (1999) Concerted regulation of wild-type p53 nuclear accumulation and activation by S100B and calcium-dependent protein kinase C. Mol Cell Biol 19: 7168–7180.[Abstract/Free Full Text]

Suzuki T, Smith S, and Board PG (1994) Structure and function of the 5' flanking sequences of the human alpha class glutathione S-transferase genes. Biochem Biophys Res Commun 200: 1665–1671.[CrossRef][Medline]

Vecchini F, Pringault E, Billiar TR, Geller DA, Hausel P, and Felley-Bosco E (1997) Decreased activity of inducible nitric oxide synthase type 2 and modulation of the expression of glutathione S-transferase alpha, bcl-2 and metallothioneins during the differentiation of CaCo-2 cells. Cell Growth Differ 8: 261–268.[Abstract]

Whalen R and Boyer TD (1998) Human glutathione S-transferases. Semin Liver Dis 18: 345–358.[Medline]

This article has been cited by other articles:

				L. Ng, K. Nichols, K. O'Rourke, A. Maslen, and G. M. Kirby Repression of Human GSTA1 by Interleukin-1beta Is Mediated by Variant Hepatic Nuclear Factor-1C Mol. Pharmacol., January 1, 2007; 71(1): 201 - 208. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: mol.105.018622v1 70/1/277    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Romero, L. Articles by Kirby, G. M. Search for Related Content PubMed PubMed Citation Articles by Romero, L. Articles by Kirby, G. M.