Hepatocyte Nuclear Factor-4α Mediates Redox Sensitivity of Inducible Nitric-oxide Synthase Gene Transcription*

The underlying redox-sensitive mechanisms that regulate hepatocyte expression of inducible nitric-oxide synthase (iNOS) and its antioxidant functions are largely unknown. We have demonstrated previously that oxidative stress induced by benzenetriol-mediated superoxide production increases interleukin-1β-induced iNOS protein synthesis, steady state iNOS mRNA expression, NO production, iNOS gene transcription, and trans-activation of the iNOS promoter in primary cultures of rat hepatocytes. In this study, we extend these studies by establishing the sequence specificity and binding of nuclear protein to the previously described 15-base cis-regulatory element of the rat hepatocyte iNOS promoter, isolating and identifying thecis-regulatory element transcription factor as hepatocyte nuclear factor-4α (HNF-4α), and confirming the functional role of HNF-4α in mediating redox-sensitive iNOS promoter trans-activation. In addition, we demonstrate that binding of HNF-4α to the transcriptional coactivator, PC4, in the presence of oxidative stress and interleukin-1β stimulation is essential for increased iNOS promoter activity in this setting. Our results indicate that HNF-4α is the transcription factor that mediates redox regulation of hepatocyte iNOS gene transcription.

States of shock or sepsis commonly initiate a complex cellular cascade of interlocking redox modulatory systems that detoxify electrophiles and regulate key cellular functions such as nucleotide synthesis, gene transcription and translation, posttranslational protein modification, enzyme activation, cell cycle regulation, and signal transduction (1)(2)(3)(4)(5)(6). In this regard, hepatocyte expression of inducible nitric-oxide synthase (iNOS) 1 and synthesis of nitric oxide (NO) convey protective antioxidant functions in models of sepsis, shock, and reperfusion injury (7,8). This effect is independent of both the oxidant species and the specific pro-inflammatory cytokine, which characterize these pathophysiologic states. However, the underlying redoxsensitive mechanisms that regulate hepatocyte expression of iNOS and its antioxidant functions are largely unknown. We have demonstrated previously that oxidative stress induced by benzenetriol (BZT)-mediated superoxide production increases IL-1␤-induced iNOS protein synthesis, steady state iNOS mRNA expression, NO production, iNOS gene transcription, and trans-activation of the iNOS promoter in primary cultures of rat hepatocytes (9,10). Using a transient expression assay in IL-1␤-and BZT-stimulated hepatocytes, we had identified a 15-base cis-regulatory element (ARE) of the rat hepatocyte iNOS promoter, confirmed binding of an ARE nuclear protein, and confirmed augmented iNOS expression in the setting of oxidative stress (10). In this study, we extend these findings by establishing the sequence specificity and binding of nuclear protein to the previously described ARE binding site, isolating and identifying the ARE transcription factor as hepatocyte nuclear factor-4␣ (HNF-4␣), and confirming the functional role of HNF-4␣ in mediating redox-sensitive iNOS promoter transactivation. In addition, we demonstrate that binding of HNF-4␣ to the transcriptional coactivator, PC4, in the presence of oxidative stress and IL-1␤ stimulation is critical for increased iNOS promoter activity in this setting.

EXPERIMENTAL PROCEDURES
Materials-The rat hepatocyte iNOS promoter (GenBank accession no. X95629) was a gift from Prof. W. Eberhardt (University of Basel, Basel, Switzerland). The HNF-4␣ expression vector was a gift from Dr. Frances Sladek (University of California, Riverside, CA).
Cell Culture-Male Lewis rats (200 -300 g; Harlan Inc., Indianapolis, IN) fed water and chow ad libitum were used for hepatocyte isolation as described by Schuetz et al. (11). After anesthetization with sodium pentobarbital, the portal vein was cannulated. The liver was perfused with calcium-free Krebs' bicarbonate buffer followed by 150 mg of collagenase D in 200 ml of Krebs' bicarbonate buffer containing 1.2 mmol/liter CaCl 2 and 1.8% bovine serum albumin. All solutions were maintained at 37°C and aerated with 95% O 2 , 5% CO 2 . The partially digested liver was excised, passed over 60-m nylon mesh, and resuspended in Dulbecco's phosphate-buffered saline (DPBS). Hepatocytes were purified by centrifugation through DPBS at 50 ϫ g for 5 min. After a second centrifugation through a 30% Percoll-DPBS gradient, hepatocytes were resuspended in Williams' E medium with 1 mmol/liter Larginine, 1 mol/liter insulin, 15 mmol/liter HEPES, pH 7.4, penicillin/ streptomycin, and 10% heat-inactivated low endotoxin fetal calf serum (FCS). Hepatocyte purity was assessed by leukocyte esterase staining and CD68 immunohistochemistry, whereas viability was assessed by trypan blue exclusion. Preparations were routinely Ͼ90% viable and Ͼ99.5% pure. The cell suspension was incubated in plastic wells for 30 min to remove residual Kupffer cells. Hepatocytes were then plated at a density of 5.0 ϫ 10 5 cells/ml onto collagen-coated wells. After 2 h, the medium was changed to remove unattached cells. After 24 h of incubation at 37°C in 95% O 2 , 5% CO 2 , cells were washed twice and fresh medium was applied for experimental use. ANA-1 macrophages (gift from Dr. George Cox, United States Uniformed Health Services, Bethesda, MD) were maintained in Dulbecco's modified Eagle's medium with 10% heat-inactivated FCS, 100 units/ml penicillin, and 100 g/ml streptomycin.
Induction of NO Synthesis-IL-1␤ (1000 units/ml) was used in the absence of FCS to induce NO synthesis. In selected instances, BZT (10 M), an autocatalytic source of superoxide at pH 7.4, was added to induce oxidative stress. After incubation for 12 h at 37°C in 5% CO 2 , the supernatants and cells were harvested for assays.
Assay of NO Production-NO released from cells in culture was quantified by measurement of the NO metabolite, nitrite. 50 l of cell culture medium were removed from culture dish and centrifuged; the supernatants were mixed with 50 l of sulfanilamide (1%) in 0.5 N HCl. After a 5-min incubation at room temperature, an equal volume of 0.02% N-(1-naphthyl)ethylenediamine was added. Following incubation for 10 min at room temperature, the absorbance of samples at 540 nm was compared with that of an NaNO 2 standard on a MAXLine microplate reader.
Gel Shift Assay-Gel shift assays were performed using 12 g of nuclear cell extract, purified chromatographic fraction, or HNF-4␣ peptide. In competitive binding assays, unlabeled mutant oligonucleotides were added at 200 M excess. Supershift assays were performed by the addition of 2 g of affinity-purified goat polyclonal antibody directed against human HNF-4␣ (Santa Cruz Biochemicals). Probe was prepared by end-labeling the wild-type 28-bp double-stranded ARE with [␥-32 P]ATP (2500 Ci/mmol) using T4 polynucleotide kinase, followed by gel purification on 15% polyacrylamide. Twenty-bp oligonucleotides used as competitors were synthesized to contain double point mutations in relation to the wild-type sequence.
Purification of ARE Transcription Factor-The ARE transcription factor was isolated by reacting the biotinylated DNA-protein complex with streptavidin paramagnetic particles (Dynal Biotech Inc.). Nuclei were isolated from rat hepatocytes treated with IL-1␤ and BZT as described previously (10). Nuclei were resuspended in 10 mM Tris, pH 7.9, 100 mM KCl, 2 mM MgCl 2 , 0.1 mM EDTA, 1 mM dithiothreitol, 1 mM NaMetaBis, and 0.2 mM PMSF, followed by Dounce homogenization on ice. Homogenized nuclei were mixed with 0.06ϫ packed nuclear volume of 4 M ammonium sulfate, pH 7.9, and extracted with gentle mixing at 4°C for 1 h. Extracted nuclei were pelleted by centrifugation at 25,000 ϫ g at 4°C for 20 min. The supernatant was then dialyzed for 6 h against 20 mM HEPES, pH 7.9, 20% glycerol, 100 mM KCl, 2 mM MgCl 2 , 0.2 mM EDTA, 1 mM dithiothreitol, 1 mM NaMetaBis, and 0.2 mM PMSF. Following dialysis, the extract was centrifuged at 16,000 ϫ g at 4°C for 20 min. Protein concentration of the nuclear extract was determined using the Bio-Rad protein assay system and was typically 5-10 mg/ml. ARE transcription factor activity was monitored during the purification by gel shift assay.
IL-1␤and BZT-treated nuclear extract was applied to a Q-Sepharose (Amersham Biosciences, Inc.) anion exchange column at 100 mM NaCl and eluted with one volume of 0.15 M NaCl. The resulting fraction was applied to a heparin-Sepharose (Amersham Biosciences, Inc.) column, washed with 0.35 M NaCl, and eluted with 0.6 M NaCl. This final fraction was diluted to 0.1 M NaCl, applied to a DNA cellulose column, and eluted with 0.4 M NaCl. The resulting fraction was diluted to 0.1 M NaCl and excess nonbinding poly(dI-dC) competitor DNA was added. Following a 10-min incubation at 4°C, the solution was centrifuged at 12,000 ϫ g for 10 min. The resulting supernatant was incubated for 5 min at 25°C with reverse phase HPLC-purified biotinylated 40-mer oligonucleotide containing the ARE binding site (5Ј-CACATGTGGAG-GTCAGGGGACAATTTATGGGA-3Ј) bound to Dynabeads M280 streptavidin in TGED buffer (20 mM Tris-HCl, pH 8.0, 1 mM EDTA, 10% (v/v) glycerol, 1 mM dithiothreitol, 0.01% Triton X-100, and 100 mM NaCl in diethyl pyrocarbonate-treated water). The magnetic beads were then washed three times with TGED buffer in 100 mM NaCl containing excess nonbinding poly(dI-dC) competitor DNA. Serial elutions were then performed using TGED buffer in 1 M NaCl. Fractions were typically stored at Ϫ80°C prior to subsequent use. Protein fractions were concentrated by trichloroacetic acid precipitation, resuspended in SDS loading buffer, boiled, and subjected to electrophoresis on 8% SDS-PAGE.
Protein Sequencing-Protein was separated by SDS-PAGE and stained with Coomassie Brilliant Blue. The individual protein band samples were excised and digested overnight with trypsin (Promega modified). The resulting digest was then injected onto a microbore high performance liquid chromatography (Beckman 32 K Gold) system, and the fractions collected. The 10 best fractions were selected for matrixassisted laser desorption/ionization mass analysis of the intact protein (ABI/Perseptive Voyager DE-Pro); subsequently, the best fractions were selected for Edman sequencing (ABI Procise 470). The resulting data were manually interpreted and searched using Sequest against the NCBI nonredundant data base.
UV Cross-linking-DNA/protein cross-linking of the ARE-nuclear protein complex was performed as described previously (12). Radiolabeled probe was prepared by annealing 1 pmol of an oligonucleotide encompassing the identified binding site (CACATGTGGAGGTCAGGG-GACAATTTATGGGA) with 100 pmol of a complementary oligonucleotide.
Renaturation of ARE Transcription Factor Activity-Following resolution on 10% SDS-PAGE, candidate purified protein was eluted and precipitated with four volumes of cold acetone. Precipitated protein was dissolved in 8 M urea in D-100 buffer and incubated at 4°C for 30 min. The protein was then renatured by dialysis against 1 liter of 1 M urea in D-100 buffer, followed by dialysis against serial changes of D-100 buffer.
Immunoblot Analysis-ANA-1 cells were lysed in buffer (0.8% NaCl, 0.02 KCl, 1% SDS, 10% Triton X-100, 0.5% sodium deoxycholic acid, 0.144% Na 2 HPO 4 and 0.024% KH 2 PO 4 , pH 7.4) and centrifuged at 12000 ϫ g for 10 min at 4°C. Protein concentration was determined by absorbance at 650 nm using protein assay reagent (Bio-Rad). Cell lysate (50 g/lane) were separated by 12% SDS-PAGE, and the products were electrotransferred to polyvinylidene difluoride membrane (Amersham Biosciences, Inc.). The membrane was blocked with 5% skim milk, PBS, 0.05% Tween for 1 h at room temperature. After being washed three times, blocked membranes were incubated with rabbit polyclonal antibody directed against HNF-4␣ (Santa Cruz Biotechnology, Santa Cruz, CA) for 1 h at room temperature, washed three times in PBS plus 0.05% Tween, and incubated with horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. After an additional three washes, bound peroxidase activity were detected by the ECL detection system (Amersham Biosciences, Inc.).
Co-immunoprecipitation Analysis-Cell culture medium was removed and plates rinsed with PBS at room temperature. All the following steps were performed on ice using ice-cold buffers. 0.6 ml of radioimmune precipitation buffer (1ϫ PBS, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 100 g/ml PMSF, and 60 g/ml aprotinin) was added to a 65-mm cell culture plate. Plates were scraped, and the cells lysed. 10 l of 10 mg/ml PMSF stock was added followed by incubation for 30 -60 min on ice. Whole cell lysate was precleared by adding 0.25 g of normal rat control IgG together with protein A-agarose conjugate and incubation at 4°C for 30 min. The beads were pelleted, and the supernatant incubated with primary antibody (polyclonal rabbit HNF-4␣ antibody, Santa Cruz Biotechnology). Resuspended protein A-agarose was added and the tubes incubated at 4°C on a rocker platform overnight. The pellet was collected by centrifugation at 1,000 ϫ g for 5 min at 4°C, and the supernatant discarded. The pellet was washed with radioimmune precipitation buffer multiple times and resuspended in electrophoresis sample buffer. Protein concentration was determined by absorbance at 650 nm using protein assay reagent (Bio-Rad). Cell lysate (50 g/lane) were separated by 12% SDS-PAGE, and the products were electrotransferred to polyvinylidene difluoride membrane (Amersham Biosciences, Inc.). The membrane was blocked with 5% skim milk, PBS, 0.05% Tween for 1 h at room temperature. After being washed three times, blocked membranes were incubated with goat PC4 polyclonal antibody (Santa Cruz Biotechnology) for 1 h at room temperature, washed three times in PBS plus 0.05% Tween, and incubated with horseradish peroxidase conjugated secondary antibody for 1 h at room temperature. After an additional three washes, bound peroxidase activity were detected by the ECL detection system (Amersham Biosciences, Inc.).
Transient Transfection Analysis of the Rat iNOS Promoter-ANA-1 macrophages and rat hepatocytes were transfected using the DEAEdextran technique (13). After cells were washed twice with medium, 10 g of plasmid DNA containing the iNOS promoter construct (1845 bp; GenBank accession no. X95629) coupled to a chloramphenicol acetyltransferase (CAT) reporter gene was added per 10 7 cells in 1 ml of medium without serum prewarmed to 37°C and containing DEAEdextran (250 g/ml) and 50 mM Tris, pH 7.4. In selected instances, an HNF-4␣ expression vector (10 g) or the mutant HNF-4␣ (mHNF-4) was co-transfected with the iNOS promoter plasmid construct. The HNF-4␣ expression vector was constructed by ligation of the BamHI-HindIII HNF-4␣ cDNA fragment from pLEN4 ligated into pcDNA3 (Invitrogen). Using the wild-type HNF-4␣ expression vector, the mutant HNF-4␣ vector in which aspartate was substituted for a Tyr 6 critical to PC4 binding was prepared using PCR-mediated mutagenesis (14,15). The suspension was incubated at 37°C for 45-60 min, followed by a 1-min shock with 10% Me 2 SO at room temperature. The cells were washed, distributed to 100-mm plates, each with about 5 ϫ 10 6 cells in 10 ml of complete medium, and incubated at 37°C in 5% CO 2 . At least 24 h later, the medium was changed, and IL-1␤ or IL-1␤ ϩ BZT was added. Approximately 14 h later, the cells were washed with ice-cold PBS, resuspended in 0.25 mM Tris, pH 7.8, and subjected to three cycles of freezing and thawing. Lysates were centrifuged (11,700 ϫ g for 10 min at 4°C); the supernatant was heated at 65°C for 10 min to inac-tivate CAT inhibitors and then centrifuged as above. The supernatant was assayed for CAT activity using a CAT enzyme-linked immunosorbent assay technique (Roche Molecular Biochemicals). Transfection efficiency was normalized by co-transfection of a ␤-galactosidase reporter gene with a constitutively active early SV40 promoter. All values are expressed as picograms of CAT/mg of protein.
Statistical Analysis-Data are expressed as means Ϯ S.E. Analysis was performed using Student's t test. p values less than 0.05 were considered significant.

Mutagenesis of the ARE Binding
Site-Utilizing nuclear protein isolated from rat hepatocytes treated with IL-1␤ and BZT, gel shift assays with a 32-bp double-stranded DNA probe derived from the iNOS rat hepatocyte promoter (nt Ϫ1353 to nt Ϫ1322) were performed for the identification of the ARE transcription factor. These probes contains the sequence, AGGTCAGGGGACA, previously identified as a high affinity binding site for the ARE transcription factor (9, 10). Competition gel shift studies were performed using a series of mutants in this binding region (Table I). Representative gel shift results for mutants 2-10 are shown in Fig. 1. Mutants 4 -9 disrupt the ARE binding site and do not compete with the wild-type DNA sequence probe for binding of nuclear protein. With nuclear protein isolated from hepatocytes treated with IL-1␤ alone or BZT alone, there is no binding of the wild-type DNA probe in either instance. These data confirm the sequence specificity and binding of nuclear protein to the previously described ARE binding site in the setting of IL-1␤ and BZT stimulation.
Isolation and Characterization of ARE Transcription Factor Protein-Bound ARE complex previously resolved by gel shift analysis was UV cross-linked to a radiolabeled ARE DNA probe. Subtracting the molecular mass of the DNA probe indicates that the molecular mass of the ARE transcription factor protein is ϳ45-50 kDa. Utilizing the biotin-streptavidin DNA affinity technique with the identified ARE DNA binding sequence, ARE transcription factor was then purified and isolated from nuclear extract isolated from rat hepatocytes stimulated with IL-1␤ and BZT. A representative Western blot of purified extract is depicted in Fig. 2. Three major bands were identified. A Southwestern blot was performed using purified nuclear extract and radiolabeled DNA probe containing the ARE binding sequence; this demonstrated binding to band 1 alone (Fig. 2). Bands 1, 2, and 3 were excised, renatured, and analyzed by gel shift analysis (Fig. 3). Only band 1 comigrates with the native ARE complex.
ARE Protein Sequencing and Identification-Band 1 was excised and subjected to protein sequencing. Analysis of two separate trypsin digests of Band 1 yielded two protein sequences: QCVVDKDKRNQ and TMGNDTSPSEGAN. Both of these peptides were identical matches with HNF-4␣ (Gen-Bank accession no. P22449). The molecular weight of HNF-4␣ corresponds to the approximate molecular weight determined from our UV cross-linking studies using the ARE transcription factor and its DNA binding sequence. Gel shift analysis utilizing radiolabeled DNA probe containing the ARE binding sequence was then performed using crude nuclear extract, purified extract (bands 1-3), purified protein (band 1), and a peptide fragment of HNF-4␣ in the presence and absence of HNF-4␣ antibody (Fig. 4). Nuclear extract, purified extract (bands 1-3), and purified protein (band 1) from IL-1␤and  Table I. Utilizing nuclear protein isolated from rat hepatocytes treated with IL-1␤ and BZT, gel shift assays with a 32-bp double-stranded DNA probe derived from the iNOS rat hepatocyte promoter (nt Ϫ1353 to nt Ϫ1322) were performed for the identification of the ARE transcription factor. These probes contains the sequence, AGGTCAGGGGACA, previously identified as a high affinity binding site for the ARE transcription factor. Gel shift assays were performed using 12 g of nuclear cell extract, purified chromatographic fraction, or HNF-4␣ peptide. In competitive binding assays, unlabeled mutant oligonucleotides were added at 200 M excess. Supershift assays were performed by the addition of 2 g of affinity-purified goat polyclonal antibody directed against human HNF-4␣ (Santa Cruz Biochemicals). Probe was prepared by end-labeling the wild-type 28-bp double-stranded ARE with [␥-32 P]ATP (2500 Ci/mmol) using T4 polynucleotide kinase, followed by gel purification on 15% polyacrylamide. Twenty-bp oligonucleotides used as competitors were synthesized to contain double point mutations in relation to the wild-type sequence.  Antibody specificity was confirmed in supershift studies using HNF-4␣ peptide. No shift of HNF-4␣ was noted in the presence of nonspecific sera. Isolated protein band 1 and HNF-4␣ were both specifically recognized by HNF-4␣ antibody. In combination with the protein sequencing data, these data indicate that HNF-4␣ is the ARE transcription factor protein. Sequencing of bands 2 and 3 were also performed. Band 2 is an immunoglobulin component, whereas band 3 corresponds to the transcriptional coactivator, PC4.
Co-immunoprecipitation of HNF-4 and PC4 -To examine the potential interaction between HNF-4␣ and PC4, co-immunoprecipitation experiments were performed using nuclear protein (Fig. 5). In control, IL-1␤-, and BZT-stimulated cells, there was no detectable PC4 protein. In contrast, in the presence of both IL-1␤ and BZT, PC4 was readily detected. Immunoblot analysis of nuclear HNF-4␣ in control, IL-1␤-, BZT-, and IL-1␤ ϩ BZT-treated cells was also performed to normalize for HNF-4␣ expression. Equivalent amounts of HNF-4␣ were noted among the four treatment groups (data not shown). These data suggest that a nuclear HNF-4␣-PC4 protein complex occurs exclusively in the presence of both IL-1␤and BZT-induced oxidative stress.
Transient Transfection Analysis of iNOS Promoter Activity-To corroborate the functional role of HNF-4␣ in the upregulation of iNOS promoter activity in the setting of IL-1␤ and BZT stimulation, a CAT reporter plasmid construct containing the full-length rat hepatocyte iNOS promoter was transfected into rat hepatocytes and ANA-1 murine macrophages. ANA-1 cells were selected because HNF-4␣ is not expressed in control, IL-1␤-, BZT-, and/or IL-1␤ ϩ BZT-treated cells, as determined by immunoblot and Northern blot analysis (data not shown). In rat hepatocytes, NO production, as determined by media levels of nitrite, was 8.8 Ϯ 2.1, 45.3 Ϯ 6.9, 8.7 Ϯ 1.2, and 85.8 Ϯ 6.1 nmol/mg of protein in unstimulated controls, IL-1␤ (1000 units/ ml), BZT (10 M), and IL-1␤ and BZT cells, respectively. In ANA-1 macrophages, NO production was 10.2 Ϯ 1.7, 24.3 Ϯ 3.2, 9.1 Ϯ 1.9, and 28.4 Ϯ 4.3 nmol/mg of protein in unstimulated controls, IL-1␤ (1000 units/ml), BZT (10 M), and IL-1␤ and BZT cells, respectively. Transient transfection analysis was then performed with the iNOS promoter plasmid construct alone (Fig. 6). In rat hepatocytes, IL-1␤ stimulation resulted in a 10-fold increase in CAT expression (p Ͻ 0.01 versus unstimulated control). The combination of IL-1␤ and BZT treatment increased CAT expression by 4-fold over that noted with IL-1␤ alone (p Ͻ 0.01 versus IL-1␤). BZT alone did not alter CAT expression in comparison to that of unstimulated control cells. Similarly, ANA-1 cells also exhibit significantly increased CAT expression in the setting of IL-1␤ stimulation, ϳ8-fold greater than controls (p Ͻ 0.01 versus controls). However, in ANA-1 cells, addition of both IL-1␤ and BZT does not significantly alter CAT expression in comparison to IL-1␤ treatment alone. BZT treatment alone does not induce significant CAT expression. These data suggest that BZTinduced oxidative stress does not augment either IL-1␤-induced iNOS promoter trans-activation or NO production in ANA-1 cells, which do not express HNF-4␣. In contrast, oxidative stress significantly increases IL-1␤-mediated iNOS promoter activation and synthesis of NO in rat hepatocytes expressing HNF-4␣.

FIG. 2. Isolation of ARE protein.
Western blot of crude nuclear extract and purified nuclear extract from rat hepatocytes stimulated with IL-1␤ (1000 units/ml) and BZT (10 M) was performed on 6% SDS-PAGE. Crude nuclear protein and nuclear protein purified utilizing the biotin-streptavidin DNA affinity technique with the identified ARE DNA binding sequence were electrophoresed on 8% SDS-PAGE and stained with Coomassie Brilliant Blue. The Southwestern blot was performed using radiolabeled wild type ARE binding sequence and the purified nuclear protein fraction. Probe was prepared by end-labeling the wild-type 28-bp double-stranded ARE with [␥-32 P]ATP (2500 Ci/ mmol) using T4 polynucleotide kinase, followed by gel purification on 15% polyacrylamide.

FIG. 3. Gel shift analysis of ARE protein.
Bands 1, 2, and 3 were isolated and precipitated with four volumes of cold acetone. Precipitated protein was dissolved in 8 M urea in D-100 buffer and incubated at 4°C for 30 min. The protein was then renatured by dialysis against 1 liter of 1 M urea in D-100 buffer, followed by dialysis against serial changes of D-100 buffer. Gel shift analysis was then performed using renatured proteins from bands 1-3, crude nuclear extract, and purified nuclear extract from rat hepatocytes stimulated with IL-1␤ (1000 units/ ml) and BZT (10 M). The probe was a 32-bp double-stranded DNA sequence derived from the iNOS rat hepatocyte promoter (nt Ϫ1353 to nt Ϫ1322) containing the sequence, AGGTCAGGGGACA, previously identified as a high affinity binding site for the ARE transcription factor.
Co-transfection assays with the iNOS promoter construct and the HNF-4␣ expression vector were also performed in ANA-1 murine macrophages exposed to IL-1␤ and/or BZT (Fig.  6). In this setting, IL-1␤ stimulation of ANA-1 cells again increases CAT expression by over 8-fold (p Ͻ 0.01 versus unstimulated control). In the presence of IL-1␤ ϩ BZT, CAT expression was increased over 3-fold in comparison to that noted in IL-1␤-treated cells (p Ͻ 0.01 versus IL-1␤). In the presence of BZT alone, CAT expression was not significantly different from that of control cells. Interestingly, HNF-4␣ expression in ANA-1 cells treated with only IL-1␤ did not increase CAT expression in comparison to that noted in the absence of HNF-4␣ expression. This result suggests that oxidative stress is a necessary component of the signal transduction pathway by which HNF-4␣ augments cytokine-induced iNOS promoter trans-activation.
In a parallel series of experiments, the mutant HNF-4 expression vector was co-transfected with the iNOS promoterreporter construct. This mutant was selected because an amino acid (Asp for Tyr 6 ) has been substituted in the location critical for PC4 binding to HNF-4␣ (14,15). In this instance, lack of PC4 binding to HNF-4␣ ablates the increased iNOS promoter activity in IL-1␤ ϩ BZT-treated cells, previously noted with co-transfection of wild type HNF-4␣. In IL-1␤ ϩ BZT-treated cells, co-immunoprecipitation experiments for HNF-4 and PC4 demonstrated the presence of an HNF-4␣-PC4 complex in the presence of the wild-type HNF-4 expression vector. In the presence of the mutant HNF-4␣ expression vector (which is recognized by the same antibody), an HNF-4-PC4 complex was not detected (data not shown). These data indicate that formation of an HNF-4␣-PC4 protein complex in ANA-1 cells is required for augmentation of iNOS promoter trans-activation in the setting of IL-1␤ and BZT stimulation. were isolated and precipitated with four volumes of cold acetone. Precipitated protein was dissolved in 8 M urea in D-100 buffer and incubated at 4°C for 30 min. The protein was then renatured by dialysis against 1 liter of 1 M urea in D-100 buffer, followed by dialysis against serial changes of D-100 buffer. Gel shift analysis was then performed using renatured proteins from bands 1-3, crude nuclear extract, and purified nuclear extract from rat hepatocytes stimulated with IL-1␤ (1000 units/ml) and BZT (10 M). The probe was a 32-bp doublestranded DNA sequence derived from the iNOS rat hepatocyte promoter (nt Ϫ1353 to nt Ϫ1322) containing the sequence, AGGTCAGGG-GACA, previously identified as a high affinity binding site for the ARE transcription factor. In selected instances, rabbit HNF-4␣ antibody (Santa Cruz) was preincubated with the nuclear proteins. Supershift analysis was also performed with HNF-4␣ peptide (Santa Cruz Biotechnology, Santa Cruz, CA) to demonstrate specificity of HNF-4␣ antibody.

FIG. 5. Co-immunoprecipitation of HNF-4␣ and PC4.
To examine the potential interaction between HNF-4␣ and PC4, co-immunoprecipitation experiments were performed using nuclear protein from cells treated with control, IL-1␤ (1000 units/ml), and/or BZT (10 M). Whole cell lysate was precleared and the supernatant incubated with primary antibody (polyclonal rabbit HNF-4␣ antibody, Santa Cruz Biotechnology). Protein concentration was determined, separated by 12% SDS-PAGE, and the products were electrotransferred to polyvinylidene difluoride membrane (Amersham Biosciences, Inc.). Blocked membranes were then incubated with goat PC4 polyclonal antibody (Santa Cruz). Following incubation with horseradish peroxidase-conjugated secondary antibody, bound peroxidase activity were detected by the ECL detection system (Amersham Biosciences, Inc.).
FIG. 6. Functional role of HNF-4␣ expression in redox-mediated iNOS promoter activity. Transient transfection assays were performed in rat hepatocytes and ANA-1 macrophages exposed to IL-1␤ (1000 units/ml) and/or BZT (10 M) to assess the functional role of HNF-4␣ in redox augmentation of iNOS promoter activity. 10 g of plasmid DNA containing the iNOS promoter construct (1845 bp; Gen-Bank accession no. X95629) coupled to a CAT reporter gene was transfected using the DEAE-dextran (250 g/ml) technique. In selected instances, an HNF-4␣ expression vector (10 g) or the mutant HNF-4␣ (mHNF-4) was co-transfected with the iNOS promoter plasmid construct. The HNF-4␣ expression vector was constructed by ligation of the BamHI-HindIII HNF-4␣ cDNA fragment from pLEN4 ligated into pcDNA3 (Invitrogen). The mutant HNF-4␣ vector in which aspartate was substituted for a Tyr 6 critical to PC4 binding was prepared using PCR-mediated mutagenesis, as previously described (15). At least 24 h later, the medium was changed, and IL-1␤ or IL-1␤ ϩ BZT was added. CAT activity was assayed using a CAT enzyme-linked immunosorbent assay technique (Roche Molecular Biochemicals). Transfection efficiency was normalized by co-transfection of a ␤-galactosidase reporter gene with a constitutively active early SV40 promoter. All values are expressed as picograms of CAT/mg of protein.

DISCUSSION
In settings of inflammation and oxidative stress, hepatocyte iNOS expression is hepatoprotective and redox-regulated (9, 10, 16 -18). The underlying redox-sensitive mechanism is not well defined. We have previously performed a functional analysis of the rat hepatocyte iNOS promoter in the setting of IL-1␤ and BZT stimulation, identified the ARE transcription factor, and established a functional role for ARE transcription factor in redox-mediated up-regulation of iNOS gene expression (9,10). In this study utilizing rat hepatocytes in primary culture stimulated with IL-1␤ and BZT, we have established the sequence specificity and binding of nuclear protein to the previously described ARE binding site, isolated and identified the ARE transcription factor as HNF-4␣, and confirmed the functional role of HNF-4␣ in mediating redox-sensitive iNOS promoter trans-activation. In addition, we have established the necessity of an association between HNF-4␣ and PC4 for increasing iNOS promoter activity in response to IL-1␤ and BZT.
HNF-4␣ is a member of the steroid hormone receptor superfamily and is critical for development and liver-specific gene expression (15, 19 -22, 24). In adult rodents and humans, HNF-4␣ mRNA is expressed in liver, small intestine, kidney, colon, pancreas, and testis. It is known to act, alone or in combinatorial association with other tissue-specific or basal transcription factors, to promote the transcription of a wide variety of target genes. These include cytochrome P450 CYP2 family members, blood coagulation factors, apolipoproteins, erythropoietin, transthyretin, complement factor B, medium chain acyl-CoA dehydrogenase, HNF-1, ␣ 1 -microglobulin, ornithine transcarbamylase, liver prolactin, and retinol-binding protein. It also activates human immunodeficiency virus-1 long terminal repeat and phosphoenolpyruvate carboxykinase promoters. Mutations in HNF-4␣ are responsible for maturityonset diabetes of the young. Broadly defined, HNF-4␣ homodimers bind to the DR1 sequence, direct repeats of the hexamer AGGTCA separated by a single nucleotide (25,26). Although the identified binding sequence in our studies differs by substitutions at the first and fourth nucleotides of the second repeat, Fraser has demonstrated previously (25) that these substitutions do not markedly alter affinity of HNF-4␣.
The relationship between HNF-4␣ and iNOS gene transcription is not well characterized. HNF-4␣ represses gene transcription of arginase, an enzyme that competes with iNOS for L-arginine (27). This mechanism may serve as a parallel regulatory pathway to up-regulate NO production by increasing iNOS substrate availability. Yoon and colleagues (28) have demonstrated that HNF-4␣ mRNA levels are unaltered by the presence or absence of NO in HepG2 cells. Of greater interest may be the signal transduction pathway by which oxidative stress and HNF-4 combine to increase iNOS promoter activity. In our co-transfection experiments using an HNF-4␣ expression vector in ANA-1 macrophages, iNOS promoter activity was augmented only in the presence of IL-1␤ ϩ BZT, suggesting that HNF-4␣ is necessary but insufficient to increase iNOS promoter trans-activation. Other mechanisms such as alteration in promoter geometry, oxidative modification of HNF-4␣, or the presence of a transcriptional co-activator may play a role. Although band 2 and band 3 did not bind to the ARE binding element in gel shift assays, we hypothesized that these proteins may play a role in facilitating increased iNOS promoter activity in the presence of IL-1␤ϩ BZT. In this regard, PC4, a transcriptional coactivator, was found to bind to HNF-4␣ in the setting of IL-1␤ ϩ BZT stimulation. Functional studies in ANA-1 cells demonstrate that HNF-4␣-PC4 binding is critical for up-regulation of iNOS promoter activity in the presence of oxidative stress.
PC4 is a 15-kDa polypeptide that serves as a potent coactivator in standard reconstituted in vitro transcription systems (23,29,30). It mediates activator-dependent transcription by RNA polymerase II through interactions with the transcriptional activator and basal transcription machinery. PC4 binds double-stranded DNA in a sequence-independent manner. It is subjected to in vivo phosphorylation events that negatively regulate its coactivator functions. The vast majority (95%) of PC4 is phosphorylated and inactive in vivo (23,29,30). Interestingly, the 24 N-terminal residues of HNF-4␣ (AF-1) constitute a critical structural element that has been demonstrated to bind to PC4 (14,15). Our data suggest that HNF-4␣ binds with PC4 under conditions of IL-1␤ and BZT stimulation and that this is essential for redox-mediated increase in iNOS promoter trans-activation. Co-transfection of a mutant HNF-4␣ in which a critical PC4 binding residue has been substituted demonstrates ablation of redox-mediated iNOS promoter activation. It is unknown whether these stimulation conditions alter HNF-4 or PC4 to facilitate this interaction. However, given the dependence of PC4 activity on its phosphorylation status and the participation of various mitogen-activated protein kinase activities in the cellular response to oxidative stress, it is tempting to speculate that PC4 may be the target. Alternatively, IL-1␤ ϩ BZT stimulation may enhance binding of PC4 to HNF-4␣, expose or structurally alter its DNA binding domain, and enhance DNA binding. These are currently the subject of ongoing experiments in our laboratory. These considerations notwithstanding, in this study, we establish the sequence specificity and binding of nuclear protein to the previously described 15-base ARE of the rat hepatocyte iNOS promoter, identify the ARE transcription factor as HNF-4␣, and confirm the functional role of HNF-4␣ in mediating redoxsensitive iNOS promoter trans-activation. In addition, we demonstrate that binding of HNF-4␣ to the transcriptional coactivator, PC4, in the presence of oxidative stress and IL-1␤ stimulation is essential for increased iNOS promoter activity in this setting. Our results indicate that HNF-4␣ is the transcription factor that mediates redox regulation of hepatocyte iNOS gene transcription.