Signal Transduction-mediated Activation of the Aryl Hydrocarbon Receptor in Rat Hepatoma H4IIE Cells*

We have investigated mechanisms of omeprazole (OME)-mediated induction of CYP1A1 and CYP3A, using the rat hepatoma H4IIE cell line, in comparison with mechanisms exerted by traditional aryl hydrocarbon receptor (AhR) ligands such as benso(a)pyrene (B(a)P) and 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). OME did not bind specifically to AhR, and it could not activate the AhR complex in rat cytosol to a xenobiotic-responsive element (XRE)-binding formin vitro. Genistein, a tyrosine kinase inhibitor, and daidzein, an inhibitor of casein kinase II, efficiently inhibited OME-mediated but not B(a)P- or TCDD-mediated induction of CYP1A1, as monitored at the transcriptional, mRNA, and protein levels as well as by analysis of activation of XRE-luciferase reporter constructs transfected into H4IIE cells. The protease inhibitorN α-p-tosyl-l-lysine chloromethyl ketone (TLCK) and lavendustin A also had similar OME-specific effects. In addition, insulin pretreatment caused an almost complete inhibition of OME-dependent CYP1A1 induction but only partially affected TCDD and B(a)P-mediated induction of CYP1A1. Staurosporine, an inhibitor of protein kinase C, impaired the induction by both B(a)P and OME. OME caused an approximately 2-fold increase in the level of CYP3A expression, but all inhibitors used were ineffective in preventing this induction. Gel shift analysis with radiolabeled XRE and specific peptide antibodies toward AhR and aryl hydrocarbon receptor nuclear translocator protein (Arnt) revealed an OME-mediated translocation of the AhR·Arnt complex into the nuclei. Genistein inhibited the specific nuclear XRE binding caused by OME, but it potentiated the formation of the TCDD-induced XRE·AhR complex. Although daidzein was able to effectively inhibit the OME-stimulatedCYP1A1 gene transcription, it did not influence the OME-dependent AhR·XRE complex formation. The data are consistent with a mechanism for OME-mediated induction of CYP1A1 that involves activation of the AhR complex via intracellular signal transduction systems and that is distinct from induction mediated by AhR ligands.

variety of compounds, including polycyclic aromatic hydrocarbons and halogenated aromatic hydrocarbons (reviewed by Gonzalez (1)). The transcriptional activation of the CYP1A1 gene is mediated by the aryl hydrocarbon receptor (AhR), 1 which has been described as a ligand-dependent transcription factor and is a member of the basic helix-loop-helix superfamily of DNA-binding proteins. Unliganded AhR is part of a cytosolic protein complex containing heat shock protein 90 (Hsp90) (2,3) and possibly another 46-kDa protein (4). Binding of the ligand to the AhR results in release of Hsp90 and dimerization to the aryl hydrocarbon receptor nuclear translocator protein (Arnt), followed by translocation into the nucleus (5)(6)(7). Once inside the nucleus, the AhR⅐Arnt heterodimer associates with specific cis-acting enhancers, designated xenobiotic-or dioxin-responsive elements (XREs or DREs), that promote the activation of the CYP1A1 gene (8). The physiological role of the AhR remains unknown, but it is likely to be important for the development of the liver and immune system, since AhR-deficient mice have been shown to have reduced liver size and decreased accumulation of lymphocytes in spleen and lymph nodes (9).
The substituted benzimidazole, omeprazole (OME), a potent suppressor of gastric acid secretion (10), has achieved a prominent position in therapy for individuals suffering from gastroesophageal reflux disease. OME, as well as other benzimidazoles, has been shown to induce CYP1A in human and rodents. Thus, OME increases the expression of CYP1A1 and CYP1A2 in human hepatocytes (11) and in the human alimentary tract (12). The anthelmintic drugs albendazole (13), and oxfendazole (14), as well as thiabendazole (15,16), induce the CYP1A subfamily in rat and rabbit. The benzimidazoles also induce the expression of human CYP3A4 (17).
The mechanism of induction of CYP1A by the benzimidazoles has been under debate. Originally, it was found that OME causes an increased expression of CYP1A mRNA in primary hepatocytes (11,18) as well as in different human cell lines (19,20), indicating a transcriptional activation mediated via the AhR. Traditional ligands for AhR are planar hydrophobic aromatic compounds. Thus, OME and other benzimidazoles do not conform to the structural features of AhR ligands. Accordingly, Daujat et al. (18) did not register any specific binding of OME to the hepatic cytosolic AhR as revealed by competition experiments using [ 3 H]TCDD as an AhR ligand. Similar results were obtained using thiabendazole (15) and lanzoprazole (17) as competitors. In contrast, Quattrochi and Tukey (19) showed that OME triggered the translocation of proteins to the nuclei, which subsequently bound to the XRE in the regulatory region of the human CYP1A1 gene.
In the present investigation, we have evaluated the cellular events associated with induction of the CYP1A1 gene by benzimidazoles, using rat hepatoma H4IIE cells. The results indicate that OME activates the AhR complex by a novel intracellular mechanism distinct from those exerted by AhR ligands and which may have components in common with an endogenous activation mechanism for the AhR.

EXPERIMENTAL PROCEDURES
Materials-Genistein was obtained from Research Biochemicals International. Daidzein and lavendustin A were obtained from Calbiochem. N ␣ -p-Tosyl-L-lysine chloromethyl ketone (TLCK), B(a)P, and indole-3-carbinol were purchased from Sigma. Omeprazole was a gift from Astra Hä ssle AB (Mölndal, Sweden). 2,3,7,8-Tetrachlorodibenzop-dioxin (TCDD) was from Larondan Fine Chemicals (Malmö, Sweden), and [ 3 H]TCDD (specific activity, 34.7 Ci/mmol) (Chemsyn) was purchased from Campro Scientific (Veenendaal, The Netherlands). All cell medium and supplements, including insulin, were purchased from Life Technologies, Inc. H4IIE cells were obtained from the American Type Culture Collection (Rockville, MD). Anti-rat CYP1A1 serum was purchased from Gentest Corporation (St. Woburn, MA), and an anti-rat CYP3A1/3A2 serum was a kind gift from Dr. James Halpert (Tucson, AZ). An anti-AhR antibody was purchased from Affinity Bioreagents, Inc. (Golden, CO). The specificity of this antibody was determined using an antiserum against the rat AhR prepared in rabbits by immunization of conjugates between ovalbumin and a peptide corresponding to amino acids 12-31 of the rat AhR as described and characterized previously (51). Antiserum against rat Arnt was prepared in rabbits by immunization of the conjugate between ovalbumin and a peptide corresponding to amino acids 39 -58 of human Arnt (21). Coupling was carried out via tyrosine linked to the C terminus of the peptide. Immunization and serum production was accomplished as outlined in Ref. 22. The antiserum reacted with a 95-kDa protein as evident from Western blotting analysis. The identity of the band recognized was verified by using a reference Arnt antiserum kindly provided by Dr. Oliver Hankinson (UCLA) and by control experiments in which the antibodies had been preadsorbed by incubation with 1 M of the peptide used for immunization for 8 h at room temperature. When this serum was used in Western blot analysis, the reactivity of the 95-kDa band was abolished.
Cell Culture Conditions and Treatments-The rat hepatoma H4IIE cells were grown in minimum essential medium containing 10% fetal bovine serum and 100 g/ml streptomycin, 100 IU/ml penicillin, nonessential amino acids, and sodium pyruvate in humidified 5% CO 2 at 37°C. When the cells had reached 90% confluence, the medium was exchanged, and cells were cultured for an additional 1 h before treatment with the compounds investigated, which were dissolved in Me 2 SO. The final concentration of solvent never exceeded 0.2%. Insulin was dissolved in aqueous solution. Control cells were treated with Me 2 SO.
Western Blotting Analysis-After treatment, the cells were washed twice in ice-cold PBS (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na 2 HPO 4 , 1.4 mM KH 2 PO 4 ) and scraped from the plate into PBS. Cells were pelleted by brief centrifugation (1,000 ϫ g), the supernatant was aspirated, and the pellet was resuspended in phosphate buffer (50 mM sodium phosphate, pH 7.4, 0.1 mM EDTA, 10% glycerol). The cells were disrupted by sonication for 20 s and centrifuged at 10,000 ϫ g for 15 min at 4°C. The 10,000 ϫ g supernatant was collected, and protein concentration was determined according to Lowry et al. (23). Solubilized proteins were separated by SDS-polyacrylamide gel electrophoresis (24) using a Mini-Protean II apparatus (Bio-Rad). The proteins were transferred to nitrocellulose membranes (Hybond C, Amersham Corp.) for 17 h at 20 mA, and the membranes were blocked in 5% nonfat milk in TTBS (50 mM Tris-HCl, 200 mM NaCl, 0.05% Tween 20). The filters were incubated with anti-rat CYP1A1 or CYP3A1/3A2 serum, and the primary antibodies were detected by using peroxidase-conjugated anti-goat immunoglobulin (Dako AS, Glostrup, Denmark) or peroxidase-conjugated protein A (Bio-Rad), respectively. The horseradish peroxidase-linked antibodies were detected by enhanced chemiluminescence (ECL) reagent (Amersham).
Preparation of Cytosol-Rat livers were perfused in PBS through v. porta, cut into small pieces, and homogenized using a Dounce homogenizer, in 3 volumes of EPGM buffer (1 mM EDTA, 20 mM potassium phosphate buffer, pH 7.2, 10% glycerol, 2 mM 2-mercaptoethanol). The resulting homogenate was centrifuged for 60 min at 105,000 ϫ g. Excessive lipids were removed from the ensuing supernatant, and the cytosol was frozen in small aliquots at Ϫ70°C. H4IIE cells were rinsed twice in phosphate-buffered saline and harvested by scraping. The cells were suspended in cold TEG buffer (20 mM Tris, pH 7.4, 1 mM EDTA, 10% glycerol, 0.2 mM PMSF, 0.5 mM DTT) and homogenized in a Dounce homogenizer with 20 up-and-down strokes, using a B-type pestle. The homogenate was centrifuged at 105,000 ϫ g for 60 min, and the resulting supernatant was immediately frozen and stored at Ϫ70°C until use. For electrophoretic mobility shift assay, 20 l of H4IIE cell cytosol was incubated in the presence of ligand at 28°C for 3-4 h. An aliquot of the activated cytosol was used in the EMSA as described below.
Analysis of Specific Ligand Binding to the AhR-Rat liver cytosol (diluted to 5-8 mg/ml in EPGM buffer) corresponding to 1.5 mg of protein, was incubated with 10 nM of [ 3 H]TCDD for 3 h at 28°C in the absence or in the presence of competitor. The samples were treated with 5 l of a charcoal-dextran solution (0.1 mg of dextran/mg of charcoal) in HEDG buffer (25 mM Hepes, pH 7.4, 1 mM EDTA, 1 mM DTT, 8.7% glycerol) at a concentration of 40 mg/ml and centrifuged for 10 min at 5,000 ϫ g. An 250-l aliquot was layered onto a 5-ml 5-30% sucrose gradient prepared in HEDG buffer and centrifuged for 18 h at 100,000 ϫ g, upon which 300-l fractions were collected from the bottom.
Northern Blot Analysis-Total RNA was isolated according to the method of Chomczynski and Sacchi (25), and 15-20 g of total RNA was used in Northern blot analysis using standard procedures (26). Probes for CYP1A1 and ␤-actin, corresponding to nucleotides 1062-1362 and 566 -1043 in the cDNA, respectively, were obtained by RT-PCR amplification of RNA from H4IIE cells. The probes were labeled with [␣-32 P]dCTP (3000 Ci/mmol, Amersham) using the Megaprime DNA labeling system (Amersham).
Quantitative RT-PCR on hnRNA-Analysis of the relative amount of CYP1A1 hnRNA (cf. Ref. 27) in variously treated H4IIE cells was performed by competitive PCR according to the PCR MIMIC 228 kit (CLONTECH, Paolo Alto, CA). Total RNA was isolated as described for Northern blot analysis. The reverse transcriptase reaction was carried out with an intron 1-specific primer, RT-primer (5Ј-GGT CTA TAG AGT GAG ATC CAA GTC AG-3Ј), using the 1st Strand TM cDNA synthesis kit (CLONTECH) and conditions as recommended by the manufacturer. The forward primer was specific for exon 1 (5Ј-GGT CCT AGA GAA CAC TCT TCA GTT CAG TC-3Ј). Three l of cDNA and 2 l of MIMIC standard (four or five different dilutions) was amplified in a 25-l reaction containing 1.7 mM MgCl 2 , 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 0.01% gelatin (w/v), 200 M dNTP, 0.25 M concentration of each primer, and 0.6 units of Taq polymerase. The PCR amplification was carried out in a Perkin-Elmer amplifier (model 2400) with an initial denaturing at 95°C for 1 min, followed by 30 cycles of denaturing at 95°C for 15 s, annealing at 54°C for 20 s, and extension at 72°C for 1 min. The program ended with a final extension at 72°C for 7 min. The PCR fragment corresponding to CYP1A1 hnRNA was 953 bp long, and the fragment containing the MIMIC standard was 561 bp. Half of the total PCR product was analyzed by electrophoresis on 1.5% agarose gels. Negative polaroid photographs of the gels were analyzed with a laser densitometer (Personal Densitometer with ImageQuant version 3.2 software, Molecular Dynamics).
Preparation of Nuclear Extracts-Nuclear extracts were prepared essentially as described by Struhl (26). In brief, the cells were washed and harvested in ice-cold phosphate-buffered saline and pelleted by brief centrifugation. The supernatant was discarded, and the cell pellet was washed once in hypotonic buffer (10 mM Hepes, pH 7.9, 1.5 mM MgCl 2 , 10 mM KCl, 0.2 mM PMSF, 0.5 mM DTT). The cells were resuspended in hypotonic buffer and allowed to swell on ice for 10 min before homogenization in a glass Dounce homogenizer with 14 up-and-down strokes, using a B-type pestle. The nuclei were collected by centrifugation at 3,300 ϫ g for 15 min and resuspended in 180 -200 l of low salt buffer (20 mM Hepes, pH 7.9, 25% glycerol, 1.5 mM MgCl 2 , 10 mM KCl, 0.2 mM EDTA, 0.2 mM PMSF, 0.5 mM DTT). KCl (2.5 M) was added dropwise to a final concentration of 0.4 M, and the nuclei were incubated for 30 min with continuous gentle mixing. The extracted nuclei were pelleted by centrifugation at 25,000 ϫ g for 30 min at 4°C. The resulting supernatant (nuclear extract) was aliquoted and immediately frozen at Ϫ70°C. Protein concentration was determined by the Bradford method (28).
Electrophoretic Mobility Shift Assay-EMSA was carried out in a total volume of 20 l with 8 -10 g of nuclear protein and a final concentration of 30 mM Hepes, pH 7.9, 20% glycerol, 2.25 mM MgCl 2 , 200 mM KCl, 0.2 mM EDTA, 0.12 mM PMSF, 0.5 mM DTT, 0.1 g/l poly[d(I-C) d(I-C)] (Pharmacia Biotech Inc.). The double-stranded probe (referred to as XRE) corresponding to the mouse DRE3 sequence (5Ј-GAGCTCGGAGTTGCGTGAGAAGAGCC-3Ј) was labeled with T4 polynucleotide kinase (Pharmacia) using [␥-32 P]ATP (ϳ3000 Ci/mmol, Amersham). For competition controls, a 10-fold molar excess of the double-stranded oligonucleotide was used. The DNA-protein complexes were separated on a 4% nondenaturing polyacrylamide gel, in Tris/ glycine buffer (50 mM Tris, 380 mM glycine, 2 mM EDTA) at 4°C. The protein components of the retarded [ 32 P]XRE complex were identified by using specific antiserum against the AhR and Arnt proteins in the incubation reaction.
Reporter Gene Expression-A fragment containing two XRE elements (29) was amplified from rat genomic DNA using primers specific for the rat CYP1A1 5Ј-flanking region (forward primer, 5Ј-CCG AGC ATT GCA CGA AAC C-3Ј; reverse primer, 5Ј-CAT ACT GAA GCA GGC GAC AC-3Ј). The PCR product was digested with SmaI and BamHI and cloned into the pGL3-prom vector (Promega) containing the firefly luciferase gene as reporter. H4IIE cells were trypsinized and subcultivated at a ratio of 1:4 on six-well plates. Four days after seeding, the cells were transfected using DMRIE-C reagent (Life Technologies). Three g of the plasmid; 0.2 g of the control plasmid pRL-SV40 (Promega), containing the renilla luciferase gene as reporter; and 4 g of DMRIE-C reagent were used for the transfections, which were performed according to the manufacturer's recommendations in a total volume of 0.75 ml and in presence of 10% fetal bovine serum. Five hours later, the medium was changed to obtain normal culture conditions (see above). The next day, the medium was changed before treatment with inducers in the absence or presence of inhibitors. After 20 h, the cells were washed twice with phosphate-buffered saline and harvested in 100 l of 1 ϫ Passive Lysis™ buffer (Promega). The luciferase activities were analyzed for 1 min in 10-l cell extracts with the Dual Luciferase Assay™ kit (Promega) on a TD-20/20 luminometer (Turner Design). The induction in the system is documented as the ratio between the activity of luciferase ligated to the reporter plasmid and that of the Renilla luciferase constituting the control for the transfection efficiency.

RESULTS
Induction of CYP1A1-Incubation of the rat H4IIE hepatoma cells with OME, B(a)P, and indole-3-carbinol caused a time-dependent induction of CYP1A1. Significant induction after OME treatment was registered after 12 h, and the level of CYP1A1 after 24 h was extensive. Using indole-3-carbinol, maximal induction was achieved already in 6 h, after which the CYP1A1 level declined, probably because of rapid clearance of the carbinol from the medium (Fig. 1). The half-maximal dose, registered under our conditions, was 135 Ϯ 20 M (n ϭ 4) for OME and 450 Ϯ 60 M (n ϭ 4) for indole-3-carbinol. CYP1A1 induction following B(a)P treatment was evident after 3 h and increased up to 24 h. Thus, the OME-mediated induction was slower as compared with the effect by other inducers. OME did not exhibit any synergistic effect on the induction of CYP1A1 caused by TCDD (data not shown).

Effect of Omeprazole on TCDD Binding to the AhR and Activation of the Cytosolic Receptor to an XRE-binding Form-
The ability of OME to compete with the specific TCDD binding to the AhR was evaluated. Cytosol, corresponding to 1.5 mg of protein, was incubated with 10 nM [ 3 H]TCDD in the presence or absence of competitor for 3 h at 28°C before being subjected to sucrose gradient centrifugation. OME at a concentration of 200 M was unable to significantly inhibit binding of TCDD (88 Ϯ 5% of maximum TCDD binding, n ϭ 4), whereas B(a)P at a 40 M concentration completely abolished specific TCDD binding.
Activation of the AhR by TCDD can be monitored in vitro by the ligand-dependent transformation of the cytosolic receptor complex to an XRE-binding form. The ability of OME to cause a similar activation of H4IIE cytosol was evaluated using EMSA. As shown in Fig. 2, OME at 250 and 500 M (not shown) was incapable of causing such a transition, whereas B(a)P (40 M) and TCDD (10 nM) caused a pronounced binding of cytosolic proteins to the 32 P-labeled XRE probe.
Effect of Protein Kinase Inhibitors on OME-, B(a)P-, and TCDD-mediated Induction of CYP1A1-Because of the inability of OME to activate the cytosolic AhR to an XRE-binding form in vitro and its inability to compete for specific [ 3 H]TCDD binding to the AhR, it was hypothesized that OME causes activation of the AhR by mechanisms separate from those of traditional AhR ligands. Such an activation could be mediated by an intracellular signal transduction chain that covalently modifies components of the AhR complex, e.g. by phosphorylation. The effect of several different protein kinase inhibitors on OME-, B(a)P-, or TCDD-mediated induction of CYP1A1 was evaluated using Western blotting (Fig. 3). As a control, the OME dependent induction of CYP3A was monitored (cf. Ref. 17).
The tyrosine kinase inhibitor genistein has been shown to inhibit the dioxin-dependent activation of the CYP1A1 gene in keratinocytes (30). In contrast, the effect of genistein (100 M) was the opposite in rat hepatoma H4IIE cells, and the compound potentiated the TCDD-mediated induction of CYP1A1 (Fig. 3, A and B). Genistein did not inhibit CYP1A1 induction mediated by B(a)P (40 M) but almost completely abolished induction of CYP1A1 caused by OME (Fig. 3B). Another tyrosine kinase inhibitor, lavendustin A, did not influence the B(a)P-or TCDD-mediated induction of CYP1A1, but it inhibited the OME response by about 40% (Fig. 3, A and B).
Daidzein is an inactive analogue of genistein regarding its ability to inhibit tyrosine kinase activity. However, it has been shown to inhibit casein kinase II as well as insulin-mediated signaling in Swiss 3T3 cells (31). Daidzein almost completely inhibited OME-dependent induction of CYP1A1, but it neither influenced B(a)P-or TCDD-mediated CYP1A1 induction nor affected the OME-dependent increase in the expression of CYP3A in the cells (Fig. 3, A, B, and C).
The extensive effect of daidzein on OME-mediated CYP1A1 induction suggested a cross-talk of an insulin-regulated signal transduction chain and the CYP1A1 induction mechanism. Insulin by itself did not influence the amount of CYP1A1 protein, but it almost totally abolished the OME-mediated induction of the enzyme, whereas it had no significant effect on CYP1A1 induced by B(a)P or TCDD (Fig. 3B).
It has previously been reported that long term treatment with phorbol esters, which down-regulate the level of protein kinase C, or protein kinase C inhibitors such as staurosporine inhibits the dioxin-dependent transcriptional activation of the mouse Cyp1a1 and Cyp1a2 genes, as well as the CYP1A1 gene in human keratinocytes (30,32,33). It was proposed by Gradin et al. (30) that at least one of the components in the AhR complex, possibly the AhR itself, is phosphorylated by protein kinase C, causing transcriptional activation of the CYP1A1 gene. In rat H4IIE cells, staurosporine (100 M) inhibited both the OME-and B(a)P-dependent induction of CYP1A1 by approximately 50% (Fig. 3B). However, staurosporine had no effect on OME-mediated CYP3A induction, indicating the absence of a general effect of the inhibitor on cytochrome P450 expression.
To find other components that might be specifically involved in the OME-dependent induction of CYP1A1, several protease inhibitors were tested. Many of them inhibited both TCDD-and OME-dependent induction, with the effect of TLCK being larger on OME-mediated induction compared with B(a)P-or TCDD-mediated CYP1A1 induction.
None of the inhibitors used in this study, either alone or in combination with OME, had any significant inhibitory effect on CYP3A induction (Fig. 3C), indicating that they are all specific inhibitors for the OME-dependent induction of CYP1A1.
The Effects of Inhibitors on Omeprazole-mediated Induction of CYP1A1 mRNA Levels-The effects of the various inhibitors used were also monitored on an mRNA level. The H4IIE cells were treated for 12 h with OME or TCDD, and total RNA was isolated and subjected to Northern blot analysis. The filters were hybridized with CYP1A1 and ␤-actin probes, and the level of mRNA was estimated by densitometric scanning of autoradiograms. As shown in Fig. 4A (left part), constitutive expression of CYP1A1 mRNA was not detected in H4IIE cells, whereas OME caused induction of mRNA levels to approximately 35% of the magnitude seen after TCDD treatment. This is in agreement with previously reported results showing an approximate 30% response on CYP1A1 mRNA induction in HepG2 cells caused by OME compared with that of TCDD (34). Genistein, daidzein, insulin, and TLCK were very efficient inhibitors of OME-induced CYP1A1 mRNA expression, with genistein being the most potent (Fig. 4B (upper part)). On the other hand, genistein potentiated the TCDD-mediated induction of CYP1A1 mRNA by approximately 50%, whereas daidzein and TLCK had no effect. By contrast, insulin caused a partial inhibition of the TCDD-mediated induction of CYP1A1 mRNA.
Effect of Omeprazole and Inhibitors on the Rate of CYP1A1 Gene Transcription-To monitor transcriptional activation of the CYP1A1 gene, an hnRNA RT-PCR assay was used as described by Elferink et al. (27). In this method, hnRNA (i.e. nascent, unspliced transcripts) was quantified and shown to mirror the rate of transcription of the CYP1A1 gene. Total RNA isolated from H4IIE cells treated under various conditions was subjected to reverse transcription using an RT primer, specific for intron 1 of the CYP1A1 gene, followed by quantitative PCR. As shown in Fig. 5A, both OME and TCDD caused an increase in the level of CYP1A1 hnRNA. Genistein, daidzein, and insulin inhibited the OME-dependent induction of CYP1A1 hnRNA to an extent comparable with the effect seen at the mRNA level, whereas genistein and daidzein exhibited no significant effect on the TCDD-mediated increase of CYP1A1 hnRNA.
The ability of the inhibitors to inhibit transcriptional activa-tion was also examined in H4IIE cells transfected with the pGL3-prom vector containing two XRE elements and firefly luciferase as reporter gene. As a control for transfection efficiency, the cells were cotransfected with pRL-SV40 containing Renilla luciferase as reporter. As shown in Fig. 5B, similar results were here obtained as reached using detection of CYP1A1 hnRNA in nontransfected cells. Thus, both genistein and daidzein caused a slight induction of reporter product formation, whereas the OME-dependent induction (about 16fold as compared with control) was effectively inhibited in the presence of either genistein or daidzein. By contrast, these compounds stimulated the TCDD-mediated induction.
The Effect of Omeprazole on Binding of Nuclear Components to XRE-Quattrochi and Tukey (19) have previously shown that nuclear extracts from OME-treated human 101L cells contain a protein complex that binds specifically to XRE and causes increased expression of reporter gene activity. We investigated the effect of the inhibitors on the formation of similar complexes in H4IIE cells. To analyze the nature of the OME-induced nuclear proteins bound to DNA, EMSA was carried out using nuclear extracts, and 32 P-labeled XRE in the presence of specific antibodies against AhR or Arnt. As shown in Fig. 6A (lanes 4 -6), there was a time-dependent binding of nuclear proteins to XRE appearing 2 h after stimulation of the cells with OME. However, TCDD-induced protein-XRE binding reached a maximum level earlier (lane 7). The gel shift band with the lowest mobility obtained after OME or TCDD stimulation was a result of specific interaction of nuclear proteins with XRE, as judged by the complete competition using a 10fold molar excess of unlabeled XRE (Fig. 6, R). Nuclear extracts from OME-treated H4IIE cells, but not TCDD-treated cells, caused the formation of another band shift with higher mobility that was also specifically competed by unlabeled XRE. (Fig. 6C,  lane 6, X). The presence of AhR antibody in the EMSA inhibited the formation of the OME-or TCDD-induced complexes (Fig.  6B, lanes 4 and 6), whereas the addition of the anti-Arnt antiserum to the binding reaction resulted in a supershift (Fig.  6B, lanes 9 and 11). The addition of control or preimmune serum did not have any effect on the bandshifts (data not shown). These results show that the nuclear protein complex recruited by OME contains both AhR and Arnt.
Genistein effectively inhibited the OME-induced binding of the AhR complex to XRE (Fig. 6C, lane 7), whereas it potentiated the binding induced by TCDD (lane 11). This is in agreement with the results obtained by Northern blotting analysis, where genistein completely blocked the OME-induced CYP1A1 mRNA but potentiated the action of TCDD. Genistein by itself slightly induced the formation of the protein-XRE complex (lane 3). By contrast, daidzein reproducibly had no significant effect on OME-or TCDD-induced binding of nuclear factors to XRE (Fig. 6C, compare lanes 8 and 12 with lanes 6 and 10), whereas it did block the increased CYP1A1 transcription caused by OME. This indicates that daidzein interferes with the transcriptional activation of the CYP1A1 gene rather than with the binding of the AhR⅐Arnt complex to XRE. Insulin alone induced the lower band of the OME-specific DNA complexes (Fig. 6C, lane 5). This complex was also induced in cells co-treated with insulin and TCDD (lane 13). These results indicate the involvement of other proteins in addition to the AhR complex in the XRE-dependent regulation of CYP1A1 transcription.

DISCUSSION
Based on the results of the present investigation, it is possible to discriminate between the mechanisms by which OME causes induction of CYP1A1 in the rat H4IIE hepatoma cell line in contrast to B(a)P and TCDD. It is evident that the OME-mediated induction involves translocation and activation of AhR. However, OME was unable to activate the cytosolic form of AhR in vitro, a property exerted by traditional AhR ligands. Furthermore, OME-dependent induction of CYP1A1 was inhibited by genistein, daidzein, and TLCK in a specific manner and exhibited a slower response than induction caused by B(a)P. Both genistein and daidzein were unable to significantly inhibit either B(a)P or TCDD-mediated induction of CYP1A1 at the transcriptional, mRNA, and protein levels under conditions otherwise identical to those used for OME-dependent induction. These compounds also did not influence the OME-dependent induction of CYP3A protein in the cells. The data indicate that genistein inhibits the formation of the AhR⅐XRE complex caused by OME, whereas daidzein causes inhibition of the OME-induced transcriptional activation. Taken together, these results support the existence of mechanisms for a ligand-independent manner of activation of the AhR, which involve OME-activated intracellular events. OME was shown to be a poor ligand for the AhR, as monitored by competition for specific [ 3 H]TCDD binding to AhR, in agreement with previous results (18). The benzimidazole could not activate the cytosolic form of the receptor to an XREbinding form in vitro, indicating that it is unable to induce dissociation of Hsp90 from the AhR complex. However, EMSA showed that OME induced the formation of an XRE-binding complex in the nuclei of H4IIE cells containing both the AhR and Arnt proteins. This is in agreement with a recent report by Daujat et al. (35), which showed the presence of activated AhR in the nuclei of Caco-2 cells after exposure to OME. The importance of AhR in OME-mediated induction of CYP1A1 has also been evident from experiments in which selective induction of the enzyme by OME was obtained in the centrilobular part of the liver acinus, a region where the AhR is also specifically located (51).
It has been reported that signals generated through release of human keratinocytes from cell substratum cause activation of CYP1A1 expression in the absence of AhR ligands (36). Also, suspension of mouse Hepa 1c1c7 cells for 4 h causes the induction of CYP1A1 mRNA and activation of the murine AhR to a DNA-binding form in the absence of any AhR ligands added (37). Furthermore, in the absence of exogenous ligands, studies with AhR-deficient mice have revealed that AhR controls basal expression of Cyp1a2 and Ugt*06 (9), suggesting alternate mechanisms for activation of the AhR. Our results fully support the idea of a ligand-independent activation mechanism of AhR.
Genistein has been shown to inhibit the dioxin-dependent expression of CYP1A1 in human keratinocytes, acting at the ligand binding domain of the receptor, through its capacity to inhibit tyrosine kinase, whereas in this system, daidzein was without effect (30). By contrast, genistein was unable to influ-ence the extent of dioxin-dependent CYP1A1 expression in murine hepatoma cells (32). Genistein is relatively specific in its action on protein-tyrosine kinases compared with other types of kinases and exerts in vitro effects at about 2 M and in vivo at approximately 100 M (38). The compound has been shown to inhibit the effects of tyrosine kinases including those of the platelet-derived growth factor and insulin receptor as well as pp60 src , pp110 gag-fes , and pp56 lck . From the present investigation, it is evident that treatment of the H4IIE cells with genistein, in combination with either TCDD or OME, results in completely opposite effects. Genistein potentiated the induction caused by TCDD but inhibited OME-dependent CYP1A1 induction. Similar responses were obtained on all cellular levels investigated, i.e. expression of CYP1A1 enzyme, CYP1A1 mRNA, CYP1A1 gene transcription, and binding of the AhR complex to XRE. This indicates that genistein prevents the omeprazole-dependent activation of the AhR complex. The mechanisms behind this prevention are unknown, and this area requires further investigation.
Another mode of action was registered using daidzein. This compound has been described as an efficient inhibitor of casein kinase II and inhibits insulin-dependent cell cycle progression in 3T3 fibroblasts (31). Casein kinase II has been shown to be involved in phosphorylation of several different transcriptional factors such as the estrogen receptor (39), c-Myc, Max, c-Myb (40), the androgen receptor (41), the human vitamin D receptor (42), and c-Jun (43). Daidzein very effectively inhibited OMEmediated but not TCDD-mediated CYP1A1 induction, as monitored on enzyme, mRNA, and transcriptional levels, but it did not prevent the formation of an AhR⅐XRE complex. This might indicate a modulatory action of a putative casein kinase II in the activation of the transcriptional complex. Such a selective effect of casein kinase II phosphorylation on the transcriptional activation but not binding of the response element has recently been described for the human vitamin D receptor (44). The rat AhR contains several consensus sequences both for protein kinase C and casein kinase II, and it has been shown by phosphatase treatment experiments that constitutive phosphorylation is important for DNA binding activity (45). In favor of a multicomponent signal transduction system involved in the OME-dependent CYP1A1 induction pathway is the finding that insulin, known to activate the mitogen-activated protein kinase cascade, potently inhibits the OME-dependent induction, as monitored at the transcriptional, mRNA, and protein levels. In contrast, TCDD-mediated induction of CYP1A1 was only partially inhibited as registered at the transcriptional and mRNA levels. Insulin did not significantly influence the extent of AhR binding to the XRE, but it had the effect (similar to that of daidzein) of inducing binding of the higher mobility protein complex on XRE (Fig. 6) and inhibiting CYP1A1 gene expression. Further investigations are necessary to identify the proteins responsible for these effects.
Another compound that allowed discrimination between OME-and TCDD-mediated CYP1A1 induction was TLCK, a trypsin-like serine protease inhibitor. This protease inhibitor did not influence the TCDD-or B(a)P-mediated CYP1A1 induction, nor did it influence the TCDD-dependent increase in the CYP1A1 mRNA; but it effectively inhibited OME-dependent CYP1A1 induction at both mRNA and protein levels. Perhaps a proteolytic step is involved in the OME-dependent induction of CYP1A1, although the specificity of TLCK is not adequate to allow any conclusive statements before the system has been identified in more detail.
In a recent paper, Dzeletovic et al. (46) have shown that OME does not bind to the AhR, in agreement with our data. However, they show that OME induces transcription of the CYP1A1  pGL3 reporter plasmid (B). A, total RNA was isolated from H4IIE cells 12 h after treatment with indicated inducers and inhibitors (cf. Fig. 4). The relative amount of CYP1A1 hnRNA was determined by competitive PCR according to PCR MIMIC TM . Aliquots from the competitive PCR products were analyzed by agarose gel electrophoresis, and quantifications were performed by densitometric scanning of negative photographs of the gels. Results from four experiments are presented. The amount of CYP1A1 hnRNA in TCDD-induced cells was set to 1.0 in each experiment. The inset shows representative results of agarose gel electrophoresis of PCR products from the amplification of samples from controls, and OME-and TCDD-treated cells. For each sample, four different concentrations of MIMIC standard (3-fold serial dilutions) were used. C, control; G, genistein; D, daidzein; I, insulin. B, 4 days after seeding, H4IIE cells were transfected using DMRIE-C with the XRE-pGL3 and the pRL-SV40 plasmids. After medium change, the cells were stimulated the next day for 20 h using the indicated combination of inducers and inhibitors: 200 M OME, 10 nM TCDD, and 95 M genistein (G) or 100 M daidzein. The luciferase activities were analyzed and are given as firefly over Renilla luciferase. 100% activity is defined as the mean activity after TCDD induction. One representative experiment out of three with similar results is given for OME (mean Ϯ S.D. of three plates), and results shown for TCDD are mean Ϯ S.D. of three different experiments using three plates in each. The mean value for induction with OME in all three experiments was 16-fold versus control. The mean value for inhibition by genistein in all three experiments was 86 Ϯ 13%, whereas the overall effect of daidzein was inhibition to control levels.
gene in primary human hepatocytes, but not in mouse cells. The induction response by OME in reconstituted yeast cells required Hsp90, worked with both the mouse and the human forms of the receptor, and was lost when a mutant form of the human receptor (Val 381 3 Asp) that abolishes ligand binding was used. These results clearly show the essential role of the AhR for omeprazole-dependent activation of the CYP1A1 gene and indicate an important function of the ligand binding do-main for the ability of OME to activate the receptor. The answer as to whether the consequence of the mutation is caused by a direct effect on the ligand binding capacity of the receptor or is indirect, resulting in a conformational change preventing functional interactions with other essential cellular components necessary for the OME-mediated activation, has to await further research. In support of the former possibility, Dzeletovic et al. (46) propose that a degradation product of FIG. 6. Electrophoretic mobility shift assay of nuclear extracts. A, nuclear extracts were prepared from H4IIE cells treated with Me 2 SO, 250 M OME, or 10 nM TCDD for the time indicated. 8 g of nuclear protein was used in each binding reaction with the 32 P-labeled double-stranded XRE oligonucleotide. B, antibodies against the AhR (␣AhR) and Arnt (␣arnt) proteins were added into the gel shift reaction, using nuclear proteins extracted from H4IIE cells stimulated for 12 h with OME or TCDD. C, H4IIE cells were treated with Me 2 SO (C), genistein (G), daidzein (D), and insulin (I) alone or in combination with OME or TCDD for 12 h. Nuclear extracts were prepared and analyzed by EMSA. Competition experiments were performed by using a 10-fold molar excess of unlabeled XRE (DRE3 oligonucleotide). The mobilities of the free probe, the AhR⅐Arnt complex, and an unknown protein-DNA complex are indicated with F, R, and X, respectively. The supershift produced by anti-Arnt antiserum is indicated with S. OME, in particular a planar sulfenamide, constitutes a ligand of the AhR and that species differences in metabolism of OME should explain the inability of OME-dependent activation in mouse cells. Our data here presented are in favor of the latter possibility. The definite answer to this question has to await isolation of the proposed active degradation compound of OME and examination of its AhR-activating properties or the identification of any cellular components active in the activation of the receptor.
It is evident that a variety of agents have been shown to activate CYP1A1, as monitored by the aryl hydrocarbon hydroxylase response, including vitamins E and K (47), various metabolic inhibitors including amino nucleoside and actinomycin D (48), several inhibitors and substrates for P450 (49), and various hormones and biogenic amines (50). Although their mechanisms of activation have not been thoroughly investigated, it is reasonable to assume that many of these agents are not ligands of the AhR, despite being inducers of CYP1A1. It may be hypothesized that some of those compounds activate the AhR through signal transduction systems similar to the one observed in this instance.
In conclusion, the present data indicate that the action of OME for induction of CYP1A1 involves the activation of intracellular signal transduction systems that result in binding of the AhR complex to XRE, subsequent gene activation, and elevation of the corresponding message and product, which is different than the mechanism of AhR ligands. It is conceivable that this effect is shared by other classes of chemical compounds that can interact in similar signal transduction systems and that the mechanism is not limited only to benzimidazole compounds.