Role of NF-κB in Regulation of PXR-mediated Gene Expression

It is a long-standing observation that inflammatory responses and infections decrease drug metabolism capacity in human and experimental animals. Cytochrome P-450 3A4 cyp304 is responsible for the metabolism of over 50% of current prescription drugs, and cyp3a4 expression is transcriptionally regulated by pregnane X receptor (PXR), which is a ligand-dependent transcription factor. In this study, we report that NF-κB activation by lipopolysaccharide and tumor necrosis factor-α plays a pivotal role in the suppression of cyp3a4 through interactions of NF-κB with the PXR·retinoid X receptor (RXR) complex. Inhibition of NF-κB by NF-κB-specific suppressor SRIκBα reversed the suppressive effects of lipopolysaccharide and tumor necrosis factor-α. Furthermore, we showed that NF-κB p65 disrupted the association of the PXR·RXRα complex with DNA sequences as determined by electrophoretic mobility shift assay and chromatin immunoprecipitation assays. NF-κB p65 directly interacted with the DNA-binding domain of RXRα and may prevent its binding to the consensus DNA sequences, thus inhibiting the transactivation by the PXR·RXRα complex. This mechanism of suppression by NF-κB activation may be extended to other nuclear receptor-regulated systems where RXRα is a dimerization partner.

Inflammatory responses and infections suppress the biotransformation of drugs and decrease the hepatointestinal capacity of drug clearance. This results in alterations of therapeutic indices and increases the toxicity of certain administered drugs. Inflammatory responses also play important roles in liver pathological conditions such as drug-induced hepatitis and cholestatic diseases (1,2). The mechanisms of these clinically important effects have not been well understood.
In human liver, the first pass of biotransformation is mainly carried out by cytochrome P-450 (CYP) 2 3A4, which is the predominant isoform of monooxygenases that are expressed in the adult hepatointestinal system. It is estimated that CYP3A4 is responsible for the metabolism of over 50% of drugs in use today, many of which are either metabolically activated and/or metabolically broken down (detoxified) through this enzyme. Therefore, transcriptional and post-transcriptional alterations of CYP3A4 activity have direct effects on the efficacy of drugs and detoxification of xenobiotics (reviewed in Refs. 3 and 4).
Recent molecular and pharmacological studies have demonstrated that transcriptional activation of cyp3a4 is mediated by the nuclear receptor PXR (pregnane X receptor). The rodent PXR (5) and its human homolog hPXR (6), also known as steroid and xenobiotic receptor (7) or hPAR (8), were identified as xenobiotic receptors that can be activated by certain xenobiotics and endobiotics. PXR regulates the expression of cyp3a4 by associating with its obligate partner RXR, and the heterodimer binds to the nuclear receptor response elements found in the regulatory regions of these genes. Genes that are regulated by PXR include multiple drug-resistant genes such as MDR1 (9) and MRP2 (10) as well as genes involved in metabolism and transport of endogenous molecules, including bilirubin, bile acids, thyroid hormone, fatty acids, and steroids (11)(12)(13). PXR⅐RXR can also interact with pathways regulated by other nuclear receptors such as the constitutive androstane receptor⅐RXR by mutual binding to the consensus regulatory DNA sequences, thus forming a redundant, compensatory network for the metabolism and disposition of xenobiotic and endobiotics (14).
The mechanisms of cyp3a4 suppression caused by inflammatory responses and infections have been investigated (15,16). Several aspects of the transcriptional regulation may be involved including decreases of PXR and RXR mRNA levels or induction of the liver inhibitory protein, which suppresses cyp3a4 through a distal flanking region (17). It is likely that modulation of transcriptional activation by several pathways leads to down-regulation of PXR-regulated gene expression.
It has been shown that most inflammatory cytokines induced during sepsis and aseptic responses lead to suppression of CYP3A4 gene expression. We hypothesize that there may be immediate, early events at transcriptional level where the effects of the proinflammatory responses converge. One of the critical responses to acute infections and inflammations is the activation of NF-B (18 -20), which has pleiotropic functions and has been shown to down-regulate the transcriptional activity of multiple steroid/nuclear receptors (21). NF-B regulates innate as well as adaptive immune systems. One of the pivotal functions of NF-B is its swift activation in response to LPS or proinflammatory cytokines, which is an evolutionally conserved defensive mechanism against infections. The classic NF-B consists of p65 (RelA) * This work was supported in part by NIEHS, National Institutes of Health Grants ES 09859 and ES 09106 and American Heart Association Grant 0355131Y. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1  and p50 heterodimer, and it is activated in response to various stimuli including LPS, TNF-␣, double-stranded RNA, and UV radiation. In this study, we investigate the role of NF-B in regulation of the transcriptional activity of the PXR⅐RXR␣ complex in an attempt to address the mechanism of suppression of cyp3a4 by LPS and proinflammatory cytokine TNF-␣. The results reveal that NF-B plays an important role in suppression of PXR⅐RXR␣-regulated gene expression by interfering with the binding of PXR⅐RXR␣ to the regulatory DNA sequences. The mechanism may have general implications in gene expressions regulated by nuclear receptors where RXR␣ is a common dimerization partner.

MATERIALS AND METHODS
Oligonucleotides as the PCR primers and ER6 EMSA probe, the DNA modifying enzymes, and Lipofectamine were from Invitrogen. Dulbecco's modified Eagle's medium was from Invitrogen or HyClone (Logan, UT), fetal bovine serum was from Atlanta Biologicals (Lawrenceville, GA). Plasmid DNA purification kits, rifampicin, lipopolysachride, and monoclonal antibody against the FLAG tag were from Sigma. Recombinant human TNF-␣ was purchased from Roche Applied Science (Indianapolis, IN). The polyclonal antibodies against RXR␣ and p65 were from Santa Cruz (Santa Cruz, CA). The human HepG2 cell line was purchased from the American Type Culture Collection (Manassas, VA).
Cell Culture and Transient Transfection-For primary human hepatocyte culture, cell suspension was purchased from Cambrex Bio-Science (Walkersville, MD). The donor of the human hepatocytes was a 26-year-old male without heart disease or hypertension. Serological tests showed negative for human immunodeficiency virus, types 1 and 2, hepatitis B surface antigen (HBsAg), hepatitis C virus, human T-cell lymphotropic I/II virus, and syphilis. Upon arrival the cells were resuspended in Dulbecco's modified Eagle's medium containing 5% fetal bovine serum, antibiotics, 4 g/ml insulin, and 1 mol/liter dexamethasone, plated in collagen-coated plate for attachment, and then maintained in Williams' E medium containing ITSϩ (insulin, transferrin, selenium, borine serum albumin, and linoleic acid), 0.1 mol/liter dexamethasone and antibiotics overnight for recovery, and then the cells were treated with Me 2 SO, RIF, RIFϩLPS, and RIFϩTNF-␣. 24 h after the treatment, the cells were harvested for isolation of total RNA for real time reverse transcription-PCR analysis. HepG2 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% antibiotic-antimycotic (100 units/ml penicillin G sodium, 100 g/ml streptomycin sulfate, and 0.25 g/ml amphotericin B) in 5% CO 2 at 37°C.
For transient transfection, HepG2 cells were seeded in 12-well plates at 30% confluence. On the next day transfection was performed using Lipofectamine (Invitrogen). 6 h after transfection, cells were treated with RIF and other reagents. 48 h later, cells were harvested to determine the luciferase activity using the luciferase assay system (Promega). Conclusions were made based on three or more independent transfection experiments.
In Vitro Transcription Coupled to Translation and EMSA-Human PXR and human RXR␣ polypeptides were generated by in vitro transcription coupled to translation using TNT-coupled reticulocyte lysate system (Promega, Madison, WI). Oligonucleotides used for EMSA were the ER6 consensus sequences in cyp3a4 promoter region as described (14). The double-stranded oligonucleotide was labeled with [␣-32 P]dCTP using Klenow enzyme (USB Corp., Cleveland, OH). For EMSA assay, PXR and RXR␣ and recombinant p50 (Promega) and p65 (produced by baculoviral expressions) in various combinations were incubated for 30 min in a reaction mixture containing 40 mM KCl, 1 mM MgCl 2 , 0.1 mM EGTA, 0.5 mM dithiothreitol, 20 mM Hepes, pH 7.9, 4% Ficoll (400 K), and ϳ30000 cpm of radiolabeled double-stranded oligonucleotide probe. After incubation for 30 min at room temperature, the reaction mixtures were separated by electrophoresis in 4.5% nondenaturing polyacrylamide gel. The results were recorded by autoradiography.
Immunocytochemistry-Primary human hepatocytes growing in 24-well plates were treated with LPS or TNF-␣ for 1 h. The cells were washed three times with cold PBS and then fixed with fresh 4% formaldehyde in PBS for 10 min at room temperature. After washing three times with PBS, the cells were permeablized with 0.2% Triton X-100 for 10 min at room temperature. After washing with PBS (three times for 5 min each), the cells were blocked with 5% bovine serum albumin in PBS/Tween 20 for 1 h at room temperature. Then primary antibody against p65 (Santa Cruz, sc-109X) diluted (1:500) in PBS/Tween 20 was added, and the reaction was incubated at room temperature for 1 h. After three washes with PBS/Tween 20, 10 min each, secondary antibody conjugated with Alexa Fluo-568 (Molecular Probe, A11011) diluted in PBS/Tween 20 (1:1000) was added and incubated for 1 h at room temperature. The cells were washed with PBS/Tween 20 three times for 10 min each. 4Ј,6Ј-Diamino-2-phenylindole was added to stain the cells. The images were visualized, and representative views of the cells were recorded by fluorescence microscopy with an Olympus IX71 microscope.
GST Pull-down Analysis-The GST pull-down assay was essentially as described (22). [ 35 S]Methionine-labeled full-length p65 protein was generated with a TNT-coupled reticulocyte lysate system (Promega) using the T7 promoter-driven cDNA plasmid as the template. PCRgenerated cDNA fragments of RXR␣ corresponding to the domains of RXR␣ (see Fig. 5A for details and the sequences of PCR primer used are available upon request) were inserted into pGEX-5X-3 (Amersham Biosciences), yielding the expression plasmids for GST-RXR␣ fusion peptides. The plasmids were expressed in Escherichia coli (BL21), and fusion polypeptides were purified with the glutathione-Sepharose 4B affinity matrix (Amersham Biosciences) according to the manufacturer's instructions. Ten micrograms of each fusion polypeptides (estimated by comparison with bovine serum albumin in an SDS-PAGE gel with Coomassie staining) was incubated with 20 l of radiolabeled p65 in a total of 250 l of binding reaction buffer (20 mM Hepes, pH 7.9, 1% Triton X-100, 20 mM dithiothreitol, 0.5% bovine serum albumin, and 100 mM KCl) for 2 h at 4°C. After incubation, the beads were washed with the same buffer without bovine serum albumin five times. The bound proteins were eluted by boiling in the SDS-PAGE sample buffer and resolved by 8% SDS-PAGE gel electrophoresis. The signals were detected by autoradiography.
Chromatin Immunoprecipitation (ChIP) Assay-The ChIP assay was based on published procedure with modification (11). HepG2 cells were transfected with FLAG-tagged PXR and pGL3-3A4-Luc and were maintained in 10-cm plates under standard cell culture conditions. At 95% confluence formaldehyde was added directly to tissue culture medium to a final concentration of 1% for cross-linking, and the plates were incubated for 15 min at room temperature on a rocker. The crosslinking reaction was stopped by adding glycine to a final concentration of 0.125 M. The plates were incubated at room temperature for 5 min. The plates were then rinsed twice with ice-cold phosphate-buffered saline. The cells were scraped off the plates and collected into 50-ml conical tubes by centrifugation (600 ϫ g for 5 min at 4°C), and the pellet was washed once with phosphate-buffered saline containing 1 mM phenylmethylsulfonyl fluoride and resuspended in 2 ml of cell lysis buffer (5 mM PIPES, pH 8, 1 mM EDTA, 0.5 mM EGTA, 85 mM KCl, 0.5% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, and 5 g/ml each of leupeptin and aprotinin) and incubated for on ice for 10 min. The cells were homogenized on ice using an B type pestle by processing in a Dounce homogenizer 200 times to aid the release of nuclei. The nuclei were collected by centrifugation (5000 ϫ g for 10 min at 4°C) and then resuspended them in nuclei lysis buffer (50 mM Tris-HCl, pH 8.1, 10 mM EDTA, 0.5 mM EGTA, 1% SDS, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, 5 g/ml each of leupeptin and aprotinin) and incubated again on ice for 10 min. The samples were sonicated into DNA fragments of 0.5-1.5 kilobase pairs (checked by agarose gel electrophoresis/ethidium bromide staining) and microcentrifuged at 14,000 rpm for 10 min at 4°C. The supernatant was cleared by incubation with Staph A cells (2.5 g/sample; Roche Applied Science) for 15 min and AG beads for 30 -60 min sequentially at 4°C on a rotating platform. The supernatant was aliquoted after centrifugation at 12,000 ϫ g for 5 min to the clean tubes. Appropriate antibodies (1 g each) were added to the aliquots and then 25 l of precleared 50% protein A/G beads (Amersham Biosciences) was added. The final volume of each sample was adjusted to no more than 500 l with the same amount of immunoprecipitation dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-Cl, pH 8.1, 167 mM NaCl, 100 g/ml sonicated salmon sperm DNA) as the nuclei lysis buffer. The mixtures were incubated on the rotating platform at 4°C, overnight. After incubation, the beads were collected by centrifugation at 5000 rpm for 1 min in a microcentrifuge, and pellets were washed once with 1 ml of 1ϫ dialysis buffer (2 mM EDTA; 50 mM Tris-Cl, pH 8.0) with 100 g/ml sonicated salmon sperm DNA, twice with 1ϫ dialysis buffer and three times with 1 ml of immunoprecipitation wash buffer (100 mM Tris-Cl, pH 9.0, 500 mM LiCl, 1% Nonidet P-40, 1% deoxycholic acid) for 10 min with rotation. After the wash, 200 l of protein kinase digestion buffer (50 mM Tris, pH 8.0, 1 mM EDTA, 100 mM NaCl, 0.5% SDS, 100 mg/ml proteinase K) was added to each sample, and the reaction was incubated at 55°C for 3 h and then at 65°C for 6 h to reverse the cross-linking. The sample was extracted once with phenol-chloroformisoamyl alcohol and precipitated with ethanol in the presence of 20 g of glycogen overnight. The precipitated pellets were collected by centrifugation at 14,000 ϫ g in microcentrifuge, and the pellets were resuspended in 20 l of TE buffer. Aliquots from each tube were amplified by PCR, and PCR products were separated by 1.2% agarose gel electrophoresis and visualized by ethidium bromide staining. The PCR primer pairs were 5Ј-TTGGACTCCCCAGTAACATTG-3Ј and 5Ј-TGCATGGAGCTTT-CCTGC-3Ј for amplifying the cyp3a4 promoter region and 5Ј-ACTCAT-GTCCCAATTAAAGGTC-3Ј and 5Ј-TGTTCTTGTCAGAAGTTCAG-C-3Ј for amplifying the enhancer module.

Suppression of PXR-mediated Gene Activation by LPS and TNF-␣ in
Human Liver Cells-The effects of LPS and TNF-␣ on the expression of PXR, RXR␣, and CYP3A4 were investigated in a primary human hepatocyte cell culture model by quantitative real time PCR. Treatment of the hepatocytes with the prototypical human PXR agonist RIF induced a 34-fold increase in CYP3A4 mRNA; the RIF-induced CYP3A4 mRNA levels were suppressed by more than 50 and 90% after cotreatment with either TNF-␣ (2 ng/ml, 24 h) and LPS (5 g/ml, 24 h), respectively (Fig.  1A). In contrast, PXR mRNA levels were unchanged by TNF-␣ treatment, and there was an approximately 30% decrease in hPXR mRNA in LPS-treated samples (Fig. 1B). RXR␣ mRNA levels were not significantly changed after treatments with either LPS or TNF-␣ (Fig. 1C). Activation of NF-B by LPS or TNF-␣ was confirmed by immunocytochemistry for p65 nuclear translocation (Fig. 1D). The RNA samples were also analyzed by microarray profiling, and the results were consistent with those obtained by the quantitative PCR, with respect to the changes of PXR, RXR␣ levels, and the suppression of cyp3a4 by LPS or TNF-␣ (data not shown).
To further investigate the effects of proinflammatory agents on the transcriptional activity of PXR and to avoid donor variability in PXRregulated genes in the human primary hepatocytes, we constructed the luciferase reporter gene driven by PXR-responsive enhancer modules for analysis of the PXR-regulated gene expression in a human hepatoma cell line (HepG2) based on the published information ( Fig. 2A) (23). HepG2 cells were transiently cotransfected with pGL3-3A4-Luc and hPXR expression plasmids pCI-hPXR. The transfected cells were then treated with RIF alone or cotreated with RIF and LPS or RIF and TNF-␣. TNF-␣ and LPS caused significant suppressions of the luciferase gene expression (Fig. 2B) that are consistent with the results from the primary human hepatocyte culture model (Fig. 1). These results using the HepG2 cell line also confirmed the utility of the HepG2 cell culture model in analysis of PXR-regulated transcription.

NF-B Plays a Critical Role in Down-regulation of cyp3a4 Expression by Inflammatory Mediators-NF-B
is an immediate early gene, which is activated in response to various stress stimuli including infections and inflammatory responses. NF-B plays a pivotal role in mediating the pathological effects of TNF-␣ and LPS. It has been demonstrated that NF-B regulates several nuclear/steroid receptors through physical and function interactions, resulting in transrepression of the gene expressions regulated by these receptors (24) (reviewed in Ref. 21). To test the role of NF-B in mediating the suppression of PXR transcriptional activity, we first transiently cotransfected NF-B p65 with PXR-driven luciferase reporter gene in HepG2 cells. Coexpression of NF-B p65 potently suppressed PXR-driven luciferase reporter gene activity, suggesting a role for NF-B in mediating suppression (Fig. 3). To further demonstrate that NF-B is specifically involved in the suppression of cyp3a4 expression, we coexpressed the NF-B super repressor, SRIB␣, in transient transfection assays and analyzed the effects of NF-B inhibition on TNF-␣-and LPS-treated cells. SRIB␣ is a mutant of IB␣ with a serine to alanine mutation at residues 32 and 36. These mutations render the IB␣ unable to be phosphorylated at serines 32 and 36 and therefore resistant to degradation by the proteosome pathway, thus causing constitutive inhibition of NF-B. In transient transfection assays, HepG2 cells were cotransfected with plasmids pCI-PXR, pGL3-3A4-Luc reporter gene and increasing amounts of SRIB␣ expression plasmid. As expected, activation of NF-B by either TNF-␣ or LPS caused suppression of the reporter gene activity. However, the LPS or TNF␣-induced suppression of reporter gene was reversed by coexpression of SRIB␣ (Fig. 3), indicating that NF-B activation was directly responsible for the suppression of the PXR-regulated gene expression.

NF-B Regulates PXR Transcriptional Activity by Disrupting the Association between PXR⅐RXR␣ Complex and DNA Sequences-It has
been shown that NF-B regulates the transcriptional activity of steroid/ nuclear receptors through direct protein-protein interaction. Na et al. (25) reported that NF-B directly interacts with RXR. The association of NF-B with nuclear receptors may potentially have a functional impact on the transcriptional activity of the PXR⅐RXR complex. One possible effect is that the binding of p65 with RXR␣ may interfere with the formation of the enhancersome consisting of the PXR⅐RXR complex and consensus DNA sequences. To test this hypothesis, we performed EMSA. PXR and RXR␣ proteins were generated through in vitro transcription coupled to translation. PXR and RXR␣ bound to the ER6 probe as dimer (Fig. 4, lanes 5 and 6). Addition of the recombinant p65 protein disrupted the binding of PXR⅐RXR␣ to the consensus ER6 FIGURE 1. The effects of LPS and TNF-␣ on RIF-induced CYP3A4, PXR, and RXR␣ mRNA levels in primary human hepatocytes. Primary human hepatocytes were treated with either RIF or cotreated with LPS (5 g/ml) or TNF-␣ (2 ng/ml) for 24 h. The total RNA was isolated and relative mRNA levels of CYP3A4, PXR, and RXR␣ were quantified by real time reverse transcription-PCR. The levels of changes of CYP3A4, PXR, and RXR␣ mRNA were normalized with ␤-actin (housekeeping gene) and presented in A, B, and C, respectively. # and *, statistically significant difference (p Ͻ 0.01 and p Ͻ 0.05, respectively) compared with RIF treatment. The data are the means Ϯ S.D. of three independent real time PCR results. D, nuclear translocation of p65 indicated activation of NF-B as determined by immunocytochemistry staining with antibody against p65. sequence (Fig. 4, compare lane 5 with lanes 10 and 11). Interestingly, disruption by p65 could be reversed upon the addition of p50 protein, which is a cognate p65 partner known to negatively regulate p65 activity (Fig. 4, compare lane 10 with lane 12). As expected, the addition of bovine serum albumin had no effects (compare lane 5 with 7) on retarded band formation, suggesting that the p65 disrupted the binding of the PXR⅐RXR␣ complex to DNA in this assay.
The results of EMSA are consistent with the hypothesis that association between RXR␣ and p65 prevents RXR␣ binding to the DNA sequences. To further analyze the interaction between RXR␣ and p65, we mapped the domains of RXR␣ responsible for association with p65 using GST pull-down assay. The known functional modular domains were fused with GST in various combinations and expressed as fusion peptides in E. coli (Fig. 5, A and C). The p65 was radiolabeled by in vitro transcription-coupled translation in the presence of [ 35 S]methionine. Interestingly, the DNA-binding domain (domain C, amino acid 135-200) is critical in mediating the RXR␣-p65 interaction, because domain C and fusion peptides that contain domain C all interacted with the radiolabeled p65, whereas the peptides that do not contain domain C did not associate with p65. The association of p65 with RXR␣ DNAbinding domain may generate steric hindrance, blocking the binding of RXR␣ to DNA, thus causing inhibition of the gene expression.
To further test this hypothesis in vivo, we performed a ChIP assay on HepG2 cells with transfected FLAG-tagged PXR. LPS and TNF-␣ treatments as well as transient coexpression of p65 decreased the association of RXR␣ with the regulatory regions of the cyp3a4 (Fig. 6), supporting the hypothesis that NF-B interferes with the binding of the PXR⅐RXR␣ complex to regulatory sequences of cyp3a4, thereby inhibiting the PXRregulated gene expression.

DISCUSSION
CYP3A4 is a predominant human liver monooxygenase metabolizing more than half of the drugs in use today. Transcriptional and posttranscriptional regulations of the expression of this enzyme are of great importance in therapeutic application as well as the development of therapeutics. Recent studies have demonstrated that the liganddependent transcription factor hPXR plays a pivotal role in coordinated regulation of cyp3a4, conjugation enzymes, and transporters at the transcriptional level (reviewed in Refs. 26 and 27); therefore, it is impor-  (23). B, suppression of PXR-mediated pGL3-3A4-Luc luciferase reporter gene activity by TNF-␣ and LPS. HepG2 cells were transiently cotransfected with pCI-hPXR and pGL3-3A4-Luc reporter plasmids, and after 6 h, the transfected cells were cotreated with either RIF ϩ TNF-␣ or RIF ϩ LPS for 24 h, respectively. The luciferase activity was assayed 48 h after the treatments. # and *, statistically significant difference ( p Ͻ 0.01 and p Ͻ 0.05, respectively) compared with RIF treatment. All of the data are the means Ϯ S.D. of triplicate transfections representative of three independent experiments.

FIGURE 3. NF-B activation is responsible for the suppression of PXR-regulated gene expression by inflammatory agents.
HepG2 cells were seeded in 12-well plates, and the cells were transiently cotransfected with plasmids of pCI-hPXR (0.2 g), pGL3-3A4-Luc (0.5 g). The transcriptional activity of PXR was either suppressed by coexpression with p65 or treatment with LPS or TNF-␣. HepG2 cells were cotransfected with SRIB␣ as indicated. The cells were harvested 48 h after the TNF or LPS treatment for luciferase activity determination. # and *, statistically significant difference (p Ͻ 0.01 and p Ͻ 0.05, respectively) compared with RIF treatment; **, statistically significant difference (p Ͻ 0.05) compared with corresponding treatment without transfection of SRIB␣. All of the data are the means Ϯ S.D. of triplicate transfections representative of three independent experiments. tant to analyze physiological and pathological conditions that may impact the PXR activity.
Infections and inflammatory responses have long been observed to suppress hepato-intestinal cytochromes P-450 as well as phase II enzymes, resulting in reduced capacity of drug clearance in both humans and experimental animals (reviewed in Refs. 16 and 28). These clinically important phenomena have been investigated extensively. Several mechanisms have been proposed to explain the infection-and inflammation-induced suppression of cyp3a4 expression. For example, it has been observed that LPS treatment down-regulates the PXR mRNA levels in cells and animals (29), and this may potentially result in suppression of cyp3a4 expression. However, the levels of the nuclear receptors may not be an accurate gauge in evaluating their transcriptional activity. Using real time quantitative PCR and microarray profiling with LPS-and TNF-␣-treated primary human hepatocytes, we found that a slight decrease of PXR mRNA and RXR␣ mRNA level was essentially unchanged (Fig. 1). The marginal decrease in PXR mRNA may not account for the dramatic suppression of the CYP3A4 mRNA by LPS and TNF-␣ (Fig. 1). Using PXR-and PPAR␣-deficient mice, Richardson and Morgan (30) have shown that endotoxin caused approximately the same levels of suppression of P-450 in KO mice as in the wild type, suggesting that nuclear receptors PXR and PPAR␣ are not required for regulating the LPS-imposed suppression of the cytochromes P-450 including CYP3As, at least in the animals whose P-450s have not been induced by exogenous agents. However, because there have been extensive cross-talks between nuclear receptors, the compensatory roles of other nuclear receptors in mediating the LPS-induced suppression remains to be investigated. This is especially true in view of the current finding that the NF-B-mediated suppression of the nuclear receptors may be through a general mechanism where the functions of a common partner (RXR) for nuclear receptors is being compromised upon NF-B activation.
Recent studies have shown that the DNA sequences at approximately Ϫ5.95 kilobase of cyp3a4 regulatory region contains CCAAT/enhancer sequences, which can be regulated by liver inhibitory protein (17), thereby causing suppression. Liver-enriched transcription factor has also been shown to mediate the LPS suppressive effects of the organic anion transporting peptide 4 (31). In our current studies, we found that PXR-directed luciferase reporter gene without the CCAAT/enhancer sequences was also suppressed by NF-B activation (Fig. 2), and inhibition of NF-B alleviated the suppression (Fig. 3), suggesting that disruption of the binding of PXR⅐RXR␣ complex to the consensus sequences (Figs. 4 -6) is an important mechanism in addition to the regulation by liver inhibitory protein. It is highly likely that more than one mechanism may be responsible for the suppression of cyp3a4 gene expression.
A common transcriptional response to the challenges of infection and inflammation is the induction of immediate early genes. One of these genes is the pleiotropic transcription factor NF-B, which is activated in response to various proinflammatory stimuli. NF-B has been shown to interact with the nuclear/steroid receptor, Ah receptor (21,24,32) and modulates the transcriptional activity of these receptors (reviewed in Refs. 21 and 33). In the mouse LPS-induced CNS inflammation model, it has been shown that Toll-like receptor regulates the suppression of the hepatic cytochromes P-450 by LPS (34,35). These studies suggested that NF-B is involved in regulation of the hepatic P-450. Although it is unknown whether NF-B activation plays a role in  7 and 8). The protein-DNA complexes were separated by nondenatured 4% polyacrylamide gel electrophoresis, and the results were recorded by autoradiography. The experiments were repeated with consistent results.  (39) and deletion mutants used for GST pull-down assays. B, GST pull-down analysis of the domains of RXR␣ that interact with p65. In lanes 1-6 the radiolabeled p65 was incubated with various recombinant GST-RXR␣ fusion peptides as indicated in A. Lanes 9 and 10 are in the input controls with 1/10 of the radiolabeled p65 and luciferase, respectively. Lanes 7 is the full-length GST-RXR␣ pulldowned luciferase (negative control). After the washes, the p65 associated with the GST-RXR␣ fusion peptides was separated by SDS-PAGE, and radioactivity was detected by autoradiography. C, Coomassie Blue staining of the GST fusion peptides used in GST pull-down assay. MW, molecular standard. the transcriptional activity of PXR, it was found that the common dimerization partner RXR for the nuclear receptors interacted with NF-B (25). We hypothesized that NF-B may play a direct role in suppression of cyp3a4 expression and developed a PXR-driven luciferase assay using HepG2 cell culture model for the analysis of transcriptional regulation of PXR by proinflammatory agents. In comparison with human primary hepatocyte culture, the magnitudes of PXR activation by rifampicin or transrepression of PXR by NF-B activation were lower, which may be due to the clonal nature of the immortalized cell line. It is well known that hepatocytes lose certain aspects of xenobiotic responses in ex vivo culture conditions. These quantitative differences notwithstanding, the HepG2-based culture model has allowed us to analyze the transcriptional regulation by PXR and overcome certain drawbacks associated with using human primary hepatocyte culture, such as the donor variability and cost.
In this study, the important role of NF-B in the suppression of cyp3a4 is demonstrated based on the following results: (i) TNF-␣ and LPS treatments of human primary hepatocytes resulted in activation of the NF-B and coincided with the down-regulation of cyp3a4, and in luciferase reporter gene assay, activation of NF-B suppressed the PXRdriven luciferase reporter gene activity; (ii) TNF-␣-and LPS-imposed repression of cyp3a4 promoter activity was reversed by the NF-B super repressor (SRIB␣), thus demonstrating the specific involvement of NF-B.
To further elucidate the mechanism underlying the suppression of cyp3a4 by NF-B, we performed EMSA, GST pull-down, and ChIP assays to test the interaction between NF-B and PXR⅐RXR␣ complexes. Using EMSA assay, we found that binding of PXR⅐RXR␣ heterodimer to the ER6 consensus sequences was inhibited by p65. The inhibitory effects of p65 on the binding of PXR⅐RXR␣ to ER6 were alleviated by p50, which is the cognate partner for p65, suggesting that the inhibitory effect of p65 could be competitively decreased by p50, which is consistent with the hypothesis that p65 interferes with the association of PXR⅐RXR␣ with DNA sequences (Fig. 4). This notion was further strengthened by the observation that the association between RXR and NF-B p65 was mediated through the RXR DNA-binding domain as determined by GST pull-down assay (Fig. 5).
Furthermore, using the ChIP assay, we found that the association of RXR␣ with the regulatory regions of cyp3a4 was disrupted upon activation of NF-B by either by LPS treatment or transient expression of p65, suggesting that the association between PXR⅐RXR␣ complex with DNA sequences was disrupted by NF-B in vivo (Fig. 6). FIGURE 6. Effects of NF-B activation on the associations of PXR and RXR␣ with the regulatory regions of cyp3a4 determined by the ChIP assay. The HepG2 cells were cotransfected with 3ϫFLAG-tagged hPXR( Fig. 2A), pGL3-3A4-Luc. The NF-B was activated by either cotransfection with p65 or treatment of the cells with LPS or TNF-␣. The cells were formaldehyde cross-linked, and associations of FLAG-tagged PXR, RXR␣ with DNA sequences were determined by ChIP assay. The regions of PCR amplification are indicated in the lower panel. Three independent ChIP assays were performed, and typical results are shown. I.P., immunoprecipitation. Transcriptional activation of gene expression consists of multiple interconnected yet distinct steps involving a constellation of transcriptional factors at different steps. For example, in regulation of cyp1a1 gene expression, the regulatory steps that have been investigated include histone remodeling and modifications (32,36), recruitments of coactivator (37) and mediator complexes (38), and recruitment of the positive transcriptional elongation factor, which leads to phosphorylation of the C-terminal domain of the large subunit of RNA polymerase II (22). It is highly likely that transcriptional regulation of cyp3a4 is also subjected to regulation at these critical steps by various signaling mechanisms including NF-B activation.
Taken together, these in vitro and in vivo results suggest that activation of NF-B results in disruption of the interaction of the PXR⅐RXR␣ complex with the consensus DNA sequences in the regulatory regions of cyp3a4, thus providing a mechanistic explanation for the observed suppression of cyp3a4 by LPS, proinflammatory cytokines, and other stress signals that are known to induce NF-B. The mechanism is depicted showing that NF-B activation by physiological and pathological stimuli leads to its translocation into the nucleus where it interrupts the binding of PXR⅐RXR␣ complex to the cognate consensus DNA sequences, thereby causing transcriptional suppression (Fig. 7). Because RXR␣ binding is interfered with by NF-B, this mechanism of suppression by NF-B may be extended to other nuclear receptor-regulated systems where RXR␣ is a dimerization partner.