| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
LIVER AND BILIARY TRACT
-hydroxylase gene transcription
Department of Biochemistry and Molecular Pathology, Northeastern Ohio University College of Medicine, Rootstown, Ohio
Submitted 14 June 2004 ; accepted in final form 25 August 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-hydroxylase (CYP7A1) in bile acid synthesis in the liver. Rifampicin, a human PXR agonist, inhibits bile acid synthesis and has been used to treat cholestatic diseases. The objective of this study is to elucidate the mechanism by which PXR inhibits CYP7A1 gene transcription. The mRNA expression levels of CYP7A1 and several nuclear receptors known to regulate the CYP7A1 gene were assayed in human primary hepatocytes by quantitative real-time PCR (Q-PCR). Rifampicin reduced CYP7A1 and small heterodimer partner (SHP; NR02B) mRNA expression suggesting that SHP was not involved in PXR inhibition of CYP7A1. Rifampicin inhibited CYP7A1 reporter activity and a PXR binding site was localized to the bile acid response element-I. Mammalian two-hybrid assays revealed that PXR interacted with hepatic nuclear factor 4 (HNF4, NR2A1) and rifampicin was required. Coimmunoprecipitation assay confirmed PXR interaction with HNF4. PXR also interacted with peroxisome proliferator-activated receptor 
coactivator (PGC-1), which interacted with HNF4 and induced CYP7A1 gene transcription. Rifampicin enhanced PXR interaction with HNF4 and reduced PGC-1 interaction with HNF4. Chromatin immunoprecipitation assay showed that PXR, HNF4, and PGC-1 bound to CYP7A1 chromatin, and rifampicin dissociated PGC-1 from chromatin. These results suggest that activation of PXR by rifampicin promotes PXR interaction with HNF4 and blocks PGC-1 activation with HNF4 and results in inhibition of CYP7A1 gene transcription. Rifampicin inhibition of bile acid synthesis may be a protective mechanism against drug and bile acid-induced cholestasis.
bile acid synthesis; gene regulation; nuclear receptors; peroxisome proliferator-activated receptor coactivator
-hydroxylase (CYP3A4), the rate-limiting enzyme in the bile acid biosynthetic pathway. This tightly regulated feedback mechanism is necessary because bile acids are highly toxic and cholesterol is needed for maintaining cellular structure and function. Recent studies (4) have revealed that bile acid feedback has far-reaching impacts on liver metabolism; it not only regulates bile acid synthesis in the digestive system but also regulates cholesterol, lipoprotein, triglyceride, and glucose metabolisms. The discovery that lithocholic acid (LCA) is an endogenous ligand for pregnane X receptor (PXR) suggests that bile acids may also regulate drug metabolism in the liver and intestine by induction of CYP450 enzymes (27, 34). PXR and its human ortholog steroid and xenobiotic receptor (SXR) induce the CYP3A, CYP2B, and CYP2C families of steroid- and drug-metabolizing enzymes in the liver and intestine (17). Despite the high-sequence identity in the ligand-binding domain between human and mouse PXR, they are very different in ligand selectivity. Pregnenolone-16-carbonitrile (PCN) and dexamethasone are strong agonists for mouse PXR but weaker agonists for human PXR. In contrast, rifampicin, an antibiotic that has been used to treat intrahepatic cholestasis, activates human PXR, but not mouse PXR (18). Previous studies (6, 20) from this laboratory show that PCN and dexamethasone strongly reduced CYP7A1 activity, mRNA, and protein expression levels in rat livers. This finding was recently confirmed by reports that PCN inhibited Cyp7a1 mRNA levels in mouse liver but had no effect on Cyp7a1 mRNA expression in Pxr knockout mice (27). These results suggest that PXR may mediate LCA and PCN inhibition of mouse Cyp7a1 gene transcription. It was proposed that activation of PXR by drugs and bile acids might protect the liver against drug and bile acid-induced cholestasis (27, 34).
Our previous studies identified two bile acid response elements (BAREs): BARE-I and BARE-II. These two BAREs are essential for basal transcriptional activity and also for conferring bile acid feedback inhibition (reviewed in Ref. 3). They contain several AGGTCA-like repeating sequences, which are potential binding sites for nuclear receptors. A direct repeat spaced by four bases (DR4) in the rat BARE-I is a binding site for an orphan receptor, chicken ovalbumin upstream transcription factor-II (COUP-TFII; NR2F2) and also for liver orphan receptor (LXR; NR1H3). However, COUP-TFII and LXR do not bind to the human BARE-I due to the lack of a DR4 motif (5). The factor that binds to the human BARE-I has not been identified. The BARE-II contains an 18-bp sequence that is completely conserved in rats and humans. Hepatic nuclear factor (HNF)4 binds to a DR1, and human -fetoprotein transcription factor (FTF; NR5A2) or mouse liver-related homolog (LRH) binds to an overlapping half-site sequence in the BARE-II. HNF4 is the only nuclear receptor that is able to stimulate the human CYP7A1 gene; all other factors tested (FTF, COUP-TFII, LXR) inhibited the human CYP7A1 gene (2, 5, 29).
Bile acids are known to activate nuclear receptor farnesoid X receptor (FXR; NR1H4), which binds to the inverted repeat (IR1) sequences and induces several genes involved in lipid metabolism (reviewed in Ref. 4). However, FXR inhibits the CYP7A1 gene by inducing a negative receptor, small heterodimer partnet (SHP), which interacts with FTF and inhibits its transactivation of the CYP7A1 gene (3). The FXR/SHP/FTF mechanism is consistent with lacking inhibition of Cyp7a1 and induction of Shp mRNA expression in Fxr null mice fed with bile acids (26). However, bile acid feeding still inhibits Cyp7a1 expression in Shp null mice (15, 33), suggesting that mechanisms alternative to the FXR/SHP/FTF pathway must exist for bile acids to inhibit Cyp7a1 mRNA expression in Shp null mice (33). The alternative mechanisms include the cytokine/MAPK/JNK pathway, protein kinase C/cJun pathway, FXR/FGF19/FGFR4 pathway, and the PXR-mediated pathway (4).
Rifampicin has been used to treat pruritus of intrahepatic cholestasis and primary biliary cirrhosis (1, 12). Rifampicin induces 6-hydroxylation of bile acids by CYP3A4. The objective of this research is to study how PXR inhibit CYP7A1 gene transcription. First, we applied quantitative real-time PCR (Q-PCR) to assay the mRNA expression of CYP7A1 and several nuclear receptors that are known to regulate CYP7A1 expression in human primary hepatocytes and established a model for studying the human CYP7A1 gene regulation. We used reporter assays, mutagenesis, Q-PCR, and EMSAs to identify a PXR response element in the human CYP7A1 gene. Mammalian two-hybrid assay and coimmunoprecipitation (co-IP) assay were used to study PXR interaction with HNF4 or peroxisome proliferator-activated receptor coactivator (PGC-1). Chromatin immunoprecipitation assay (ChIP) was used to verify HNF4 and PXR binding to the native CYP7A1 chromatin. Results revealed a novel mechanism for PXR, HNF4, and PGC-1 regulation of CYP7A1 gene transcription in bile acid- and drug-induced cholestasis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Reporter and receptor expression plasmids.
Human CYP7A1/luciferase (Luc) reporter constructs ph-1887/Luc, ph-298/Luc, and ph-150/Luc were constructed as described previously (32). Expression plasmid for steroid receptor coactivator (SRC)-1 (PCR3.1-hSRC-1A) was obtained from M.-J. Tsai (Baylor College of Medicine, Houston, TX). Expression plasmid for human PGC-1 (pcDNA3/HA-PGC-1) was obtained from A. Kralli (The Scripp Research Institute, La Jolla, CA). Expression plasmid for human FTF (pCDM8-hFTF) was obtained from D. Moore (Baylor College), expression plasmids for retinoid X receptor (RXR; pcDNA3-hRXR), HNF4 (pCMX-rHNF4), and human PXR (pSG5-hPXR) (19) were provided by R. Evans (Scripps Research Institute), W. Chin (Eli Lily Research Laboratories, Indianapolis, ID), and S. Kliewer (University of Texas Southwestern Medical Center, Dallas, TX), respectively.
For mammalian two-hybrid assays, the reporter used was 5xUAS-TK-Luc, which contains five copies of the upstream activating sequence (UAS) fused upstream of a thymidine kinase minimum promoter (TK) and the luciferase reporter gene (30). Two-hybrid constructs contained ligand-binding domain (LBD) of nuclear receptors or nuclear receptor-interacting domain (RID) of coactivators fused to Gal4-DNA binding domain (DBD) or VP16-activation domain (AD) vector. Mammalian two-hybrid fusion plasmids used were as follows: pCMX-VP16-rHNF4 [D + E regions, amino acids (AA) 125–364] from D. Moore (Baylor College of Medicine); pCMX-VP16-hCPF (D + E regions) from B. Shen (Tularik, South San Francisco, CA); Gal4-PXR-LBD (AA 107–434), Gal4-SRC-1-RID (AA 595–780), and VP16-PXR-LBD (AA 107–437) from A. Takeshita (Toranomon Hospital, Tokyo, Japan) (30), and Gal4-PGC-1 [full length (FL)] from A. Kralli (The Scripp Research Institute). Gal4-HNF4 fusion plasmids pBx-HNF4-FL (AA 1–455) and pBx-HNF4-130–455 (D + E + F regions) were obtained from I. Talianidis (Institute of Molecular Biology and Biotechnology Foundation for Resarch and Technology, Hellas, Herakleion Crete, Greece), and pcDNA3X-Gal4-HNF4-1–129 (A/B and DBD-C) was obtained from M. Crestani (University of Milan, Italy).
RNA isolation and Q-PCR.
Primary hepatocytes were plated in six-well plates. Cells were treated with rifampicin (10 µM; Sigma, St. Louis, MO) and chenodeoxycholic acid (CDCA) (30 µM) (Sigma) in serum-free media and grown for 24 h. Total RNA was isolated from the cells using Tri-Reagent (Sigma) according to the manufacturer's protocol. Reverse-transcription reactions were performed using RETROscript kit according to manufacturer's instructions (Ambion, Austin, TX). For Q-PCR, samples were prepared according to the PCR SYBRgreen Master Mix 2X protocol (Applied Biosystems, Foster City, CA). Amplification of human highly basic protein (21) or ubiquitin C (16) was used in the same reactions of as an internal reference gene. Primers were designed using Primer Express 1.5 (ABI). Quantitative PCR analysis was conducted on the ABI Prism 7700 sequence-detection system. Relative mRNA expression was quantified using the comparative threshold cycle (Ct; 
Ct) method according to the ABI manual.
Transient transfection assay.
HepG2 cells were grown to 
80% confluence in 24-well tissue culture plates. Unless otherwise indicated, in transient transfection assay, human CYP7A1/Luciferase reporter plasmid (1 µg) was transfected with expression plasmid (0.5 µg) using the calcium phosphate-DNA coprecipitation method. The pCMV
-galactosidase plasmid (0.1 µg) was transfected as an internal standard for normalizing the transfection efficiency in each assay. In some samples, empty expression vectors were added to equalize the total amounts of plasmid DNA transfected in each assay. Four hours after transfection, cells were incubated in serum-free media and treated with LCA (Steraloids, Newport, RI) or rifampicin in the concentrations indicated in each experiment for 40 h. Cells were harvested for assay of luciferase reporter activity using Luciferase Assay System (Promega). Luciferase activity was determined using a Lumat LB 9501 luminometer (Berthold Systems, Pittsburgh, PA) and normalized by dividing the relative light units by -galactosidase activity. Each assay was done in triplicate, and individual experiments were repeated at least three times. Data are plotted as means ± SD. Statistical analyses of treated vs. untreated controls (vehicle or empty plasmids) were performed using Student's t-test. A P < 0.05 is considered statistically significant.
Cell viability assay. Cell viability assays were performed using TOX1 kit (Sigma). HepG2 cells or human primary hepatocytes (HH1165) were cultured in 48-well plates and treated with increasing amounts of LCA (1 to 15 µM), CDCA (10–50 µM) or rifampicin (5–20 µM) for 48 h. 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was added to the media and incubated for an additional 2 h. The resulting purple crystals from MTT were dissolved in MTT dissolving solution, and absorbance was measured at 570 nm using a spectrophotometer.
Site-directed mutagenesis. Mutations were introduced into ph-298/Luc plasmid using PCR-based QuikChange Site-directed Mutagenesis Kit (Stratagene, La Jolla, CA). Two complementary oligonucleotides containing the mutations (see Fig. 4B) were used as PCR primers. PCR reactions were set up according to the manufacturer's instruction using 50 ng of template DNA and 125 ng of primers. PCR cycling parameters were set as follows: denaturing at 95°C for 2 min, followed by 18 cycles at 95°C for 30 s, 55°C for 1 min, and 68°C for 18 min. The reaction mixture was digested by Dpn I for 2 h to remove the template DNA and transformed into XL1-Blue super competent cells (Stratagene) for selection of mutant clones. Mutations in each clone were confirmed by DNA sequencing.
|
-gal plasmid (0.25 µg) into HepG2 cells. Cells were treated with rifampicin (10 µM), LCA (10 µM), or vehicle (DMSO). Luciferase activity was assayed as described in Transient transfection assay.
EMSA.
Nuclear receptors were synthesized in vitro using Quick-coupled Transcription/Translation Systems (Promega) programmed with receptor expression plasmids according to the manufacturer's instruction. Double-stranded synthetic probes for the PXR binding site of CYP3A4, BARE-I and BARE-II of the human CYP7A1 gene were prepared by heating equal molar amounts of complementary oligonucleotides to 95°C in 2x SSC (0.5 M NaCl, 15 mM sodium citrate, pH 7.0) and then slowly cooled to room temperature. The resulting double-stranded fragments were labeled with [32P]dCTP by filling in reaction using the Klenow fragment of DNA polymerase I. Labeled double-stranded probes were purified through G-50 spin columns. For each binding reaction, 50,000 counts/min (cpm) of labeled probe and 5 l of in vitro translated protein were incubated at room temperature for 20 min in 20 µl of binding buffer containing 12 mM HEPES, pH 7.9, 50 mM KCl, 1 mM EDTA, 1 mM dithiothreitol, 15% glycerol, and 2 µg of poly (dI-dC). Samples were loaded on a 5% polyacrylamine gel, and electrophoresis was performed at room temperature at constant 200 V for 1.5 h. The gel was dried and autoradiographed using Phosphor Imager 445Si (Molecular Dynamics, Sunnyvale, CA).
ChIP assay.
Primary human hepatocytes were obtained in T75 culture flasks, and ChIP assays were performed on the same day the cells were received. HepG2 cells were grown in 100-mm culture dishes to 80% confluence. Ten micrograms of PXR, RXR, FTF, HNF4, and HA tagged-PGC-1 expression plasmids were transfected using the calcium phosphate-DNA coprecipitation method as described in Transient transfection assay. Four hours after transfection, cells were incubated in serum-free media containing vehicle (DMSO) or 10 µM rifampicin for 40 h. ChIP assays were performed using ChIP assay kit (Upstate Biotec, NY) according to manufacturer's protocol. Briefly, cells were cross-linked in 1% formaldehyde for 10 min and washed with ice-cold PBS containing protease inhibitors (1 mM PMSF, 1 µg/ml aprotinin, and 1 µg/ml pepstanin A, Sigma) twice. Cells were scraped and incubated in 1% SDS lysis buffer (1% SDS, 10 mM EDTA, 50 mM Tris·HCl, pH 8.1) for 30 min on ice and sonicated using a Branson sonifier 250 with a microtip setting 6 for 15-s pulses at 40% output for a total of 1.5 min to break the DNA into 0.2- to 2-kb fragments. Cell lysates were collected by centrifugation and diluted 10-fold in ChIP dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris·HCl, pH 8.1, 167 mM NaCl). Ten percent of the cell lysates were saved and used as the "input" for immunoprecipitation. After the diluted cell lysate was precleared with protein A-agarose, DNA-protein complexes were precipitated by incubating the cell lysates with 10 µg of antibodies (goat HNF4, #6556; goat PXR, #7737; goat PGC-1, #5815; goat CPF, #5995, from Santa Cruz Biotechnology, Santa Cruz, CA) against target proteins overnight followed by a 3-h incubation with protein A or G-agarose beads for each treatment. Protein A or G was added as a nonimmunoprecipitation control. The beads were washed with low-salt wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris·HCl, pH 8.1, 150 mM NaCl) once, high-salt wash buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris·HCl, pH 8.1, 500 mM NaCl) once, LiCl wash buffer [0.25 M LiCl, 1% Nonidet P-40 (NP-40), 1% deoxycholate, 1 mM EDTA, and 10 mM Tris·HCl, pH 8.1] once, and buffer with 10 mM Tris·HCl and 1 mM EDTA, pH 8.0, twice. Samples were eluted twice with 250 µl of ChIP elution buffer (1% SDS, 0.1 M NaHCO3), and the eluates were combined. The cross-links were reversed by adjusting NaCl concentration to 200 mM and incubating the eluates at 65°C for 4 h followed by a 1-h incubation in 0.04 µg/µl proteinase K. DNA was extracted using phenol/chloroform and precipitated using isopropanol. A 391-bp DNA fragment containing BARE-I and BARE-II was PCR amplified for 45 cycles and analyzed on a 2% agarose gel. The PCR primers used were as follows: forward primer: 5'- ATCACCGTCTCTCTGGCAAAGCAC; reverse primer: 5'-CCATTAACTTGAGCTTGGTTGACAAAG. The amplified fragment of 391 bp is from –432 to –41 containing the BARE-I and BARE-II.
Co-IP assay.
HepG2 cells were cultured in T150 flasks and transfected with 10 g of PXR expression plasmid using Ca+2-phosphate coimmunoprecipitation method. Human primary hepatocytes (#HH1148) in T75 flasks were treated with either vehicle (DMSO) or rifampicin (10 µM) for 24 h, collected, and incubated in modified RIPA buffer (50 mM Tris-HCl, 1% NP-40, 0.25%-deoxycholate, 150 mM NaCl, 1 mM EDTA) containing protease inhibitors (1 mM PMSF, 1 µg/ml aprotinin, 1 µg/ml pepstanin, Sigma) for 30 min. Clear cell lyates were collected by centrifugation at 10,000 g at 4°C for 15 min. Clear lysates were precleared with protein G beads and incubated with 15 µg of rabbit anti-HNF4 antibody (Santa Cruz Biotechnology) at 4°C with rotation overnight, followed by an additional incubation for 2 h with protein G beads. The beads were then washed three times with RIPA buffer and were boiled in 2x protein loading buffer for 5 min. Samples were divided into equal amounts and_loaded on SDS PAGE gels for Western immunoblot analysis using goat antibodies against HNF4 (#6556, Santa Cruz Biotechnology), PGC-1 (#5815), and PXR (#7737).
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
mRNAs are expressed in human primary hepatocytes.
We first used Q-PCR assays to detect mRNA expression levels of CYP7A1 and several transcription factors known to regulate CYP7A1 transcription in human primary hepatocytes. The averaging Ct (±SD) determined from seven hepatocyte preparations (HH1088, HH1089, HH1115, HH1117, HH1118, HH1119, and HH1122) wereas follows: CYP7A1 (16.38 ± 2.17), SHP (14.37 ± 1.47), PXR (9.60 ± 0.99), FTF (8.71 ± 1.16), COUP-TFII (7.68 ± 0.75), FXR (6.42 ± 1.13), PGC-1 (6.30 ± 0.99), HNF4 (5.92 ± 0.97), and Prox1 (5.33 ± 0.97). The Ct is the fractional cycle number at which the amount of amplified target reaches a fixed threshold. The Ct value is exponentially related to copy number. High Ct value indicates low copy number. The Ct value is the Ct of the target gene minus the Ct of the reference gene (ubiquitin C). These data indicated that CYP7A1 mRNA levels were very low but detectable by this assay. SHP mRNA levels were 100-fold higher than CYP7A1. PXR mRNA levels were lower than FXR, HNF4, and PGC-1, which were abundant in human primary hepatocytes. These experiments revealed that human primary hepatocytes expressed very low levels of CYP7A1 but abundant nuclear receptors and a coactivator known to regulate CYP7A1 gene expression. Therefore, the human primary hepatocyte is suitable for studying CYP7A1 gene regulation by nuclear receptors.
Rifampicin and CDCA reduce CYP7A1 mRNA expression in human primary hepatocytes.
Q-PCR was then used to assay the effect of CDCA, an FXR ligand, and rifampicin, a human PXR ligand, on CYP7A1 mRNA expression levels in human primary hepatocytes (HH1106). Comparative Ct (–Ct) method was used to quantify the change of mRNA expression levels by CDCA and rifampicin (Table 1). CDCA (30 µM) markedly reduced CYP7A1 mRNA expression by 95%, and rifampicin (10 µM) reduced CYP7A1 mRNA expression by 80%. As a positive control for CDCA induction, SHP mRNA levels were determined. As expected, CDCA induced SHP mRNA by sixfold. The observed inverse relationship of CYP7A1 and SHP mRNA expression levels by bile acid is consistent with the hypothesis that SHP mechanism mediates bile acid inhibition of CYP7A1 gene transcription. Interestingly, rifampicin completely abolished SHP mRNA expression. As a positive control of PXR induction, CYP3A4 mRNA expression levels were measured in human primary hepatocytes. The CYP3A4 mRNA levels were too low to be detected in vehicle and CDCA-treated human hepatocytes (Ct = 40). Rifampicin remarkably induced CYP3A4 mRNA levels by more than 50-fold, consistent with a previous report (9). The strong induction of SHP by CDCA and CYP3A4 by rifampicin indicates that these compounds at the concentrations tested did not cause toxic effect on human primary hepatocytes as demonstrated by MTT assays. The lack of an inverse relationship between CYP7A1 and SHP mRNA expression by rifampicin provides strong evidence that SHP is not involved in PXR inhibition of CYP7A1. These results may suggest that ligand-activated PXR reduced SHP gene transcription in human livers.
|
in HepG2 cells for reporter assays. As shown in Fig. 1B, rifampicin strongly reduced the reporter activity by 80% in PXR-expressing HepG2 cells. These results suggest that PXR is involved in rifampicin inhibition of the CYP7A1 gene.
|
. Reporter activities of all three constructs were inhibited 80% by rifampicin (data not shown). It is concluded that the PXR response element is located in the region downstream of –150, which contains both BARE-I and BARE-II. We thus performed EMSA using oligonucleotide probes designed based on the nucleotide sequences in the BARE-I and BARE-II. When the human BARE-I probe was used for EMSA (Fig. 2A), a mixture of in vitro synthesized PXR and RXR shifted a single band (Fig. 2B). Neither PXR nor RXR alone bound the BARE-I probe (Fig. 2B). A probe that contains a strong PXR binding site in the human CYP3A4 gene (ER6; Fig. 2A) was used as a positive control for PXR binding (Fig. 2B). Excess unlabeled BARE-I or unlabeled CYP3A4 oligonucleotide completely abolished the shifted band. A mutant BARE-I probe, which has mutations introduced to alter the AGCTCA half-site sequence, did not bind PXR/RXR. Figure 2C shows that unlabeled CYP3A4 oligonucleotide could completely compete out PXR binding to the labeled BARE-I probe at 100-fold excess. On the other hand, 200-fold excess of unlabeled BARE-I was required to completely complete out PXR binding to the labeled 3A4 probe. PXR is known to bind various motifs including DR3, DR4, and ER6 (11). Analysis of the nucleotide sequences of human BARE-I revealed a putative DR3 motif (Fig. 2A) that is known to bind human PXR (17).
|
, rifampicin did not significantly inhibit the mutant reporter activity (Fig. 3B). These results confirmed that the BARE-I was a functional negative PXR response element. It was also concluded that the BARE-II was not a PXR response element, because PXR did not bind to the BARE-II (data not shown) and rifampicin did not inhibit the mutant reporter that had an intact BARE-II (Fig. 3B).
|
and PGC-1 stimulation of human CYP7A1 gene.
We then studied the effect of rifampicin on CYP7A1 reporter activity affected by HNF4, PXR, and PGC-1. As shown in Fig. 4, HNF4 stimulated human CYP7A1 reporter activity by fivefold (lane 2). PGC-1 alone stimulated human CYP7A1 reporter activity by threefold (lane 3). Combination of HNF4 and PGC-1 stimulated activity by eightfold (lane 5). It is remarkable that addition of rifampicin almost completely abolished reporter activity in PXR expression HepG2 cells (lanes 4 and 6–8). These data revealed that PXR was a strong repressor of the CYP7A1 gene when activated by a ligand, whereas HNF4 and PGC-1 are positive regulators of the human CYP7A1 gene.
PXR interacts with HNF4 and PGC-1.
We reported previously that HNF4 bound to the BARE-II and COUP-TFII bound to the BARE-I synergistically stimulated rat CYP7A1 gene transcription (29). We proposed a model that factors bound to the BARE-I and BARE-II interact and regulate CYP7A1 gene transcription. We therefore wanted to test the hypothesis that PXR, which binds to the BARE-I, might also interact with HNF4 to regulate the human CYP7A1 gene. We used mammalian two-hybrid assay to study the interaction between Gal4-PXR-LBD fusion and VP16-HNF4 fusion in HepG2 cells. In this assay, Gal4-PXR-LBD fusion protein binds to the Gal4-DBD (DNA binding site or UAS) in a luciferase reporter (Gal4–5XUAS-TK-Luc), which has a TK minimal promoter, and expresses low basal activity in mammalian cells. If Gal4-PXR fusion protein interacts with VP16-HNF4 fusion protein, then the Gal4 reporter activity should be stimulated. The effect of a PXR ligand on the interaction also can be tested in this assay. The two-hybrid assays in Fig. 5A show that Gal4-PXR-LBD (AA 107–434) interacts strongly with VP16-HNF4 (D + BD, AA 125–364) only in the presence of rifampicin. As a control for specificity of interaction, Gal4-PXR interacts very weakly with VP16-FTF. It should be noted that the FTF binding site overlaps with the HNF4 binding site in the BARE-II. We then used Gal4-HNF4 and VP16-PXR hybrids in two-hybrid assays. Figure 5B shows that Gal4-HNF4 full-length and VP16-PXR-LBD interact without addition of a HNF4 ligand (4-fold of the background). It should be noted that HNF4 has intrinsic activity, and the endogenous HNF4 ligand is not known for certain. Addition of rifampicin somewhat enhanced the interaction. The interactions between Gal4-HNF4 and VP16-PXR fusion proteins were much weaker than between Gal4-PXR/VP16-HNF4 hybrids (Fig. 5A), as indicated by much lower reporter activities. The NH2-terminal region (A/B + DBD; AA 1–129) of HNF4 did not_interact with PXR, whereas the COOH-terminal region (LBD; AA 130–455) of HNF4 did. These data revealed that the LBD of HNF4 interacts with LBD of PXR.
|
and rifampicin attenuates the PGC-1 interaction with HNF4 and PXR.
PGC-1 was originally identified as a coactivator of PPAR. Since then, several other nuclear receptors including HNF4, LXR, RXR, and Estrogen receptor (ER), and ER-related receptor were found to interact with PGC-1 and regulate gluconeogenesis and energy metabolism (23, 25). It is not known whether PGC-1 interacts with PXR. We therefore performed mammalian two-hybrid assays to study PXR interaction with PGC-1. Figure 6A shows that Gal4-PGC-1 strongly interacts with VP16-PXR as indicated by a 400-fold stimulation of Gal4 reporter activity over Gal4 basal activity. A ubiquitous steroid receptor coactivator, SRC-1, interacts much weaker (20-fold) with VP16-PXR. Because coactivator is known to interact with nuclear receptor by binding to the LBD and ligand enhances the interaction, we then studied the effect of rifampicin on PXR interaction with HNF4 and PGC-1. Figure 6B shows that rifampicin induced a fourfold stimulation of PXR and PGC-1 interaction. Interestingly, overexpression of HNF4 strongly inhibited PXR and PGC-1 interaction only when a PXR ligand was added. This indicates that rifampicin promotes PXR and HNF4 interaction and excludes PGC-1 from interaction with PXR. Figure 6C shows that Gal4-PGC-1 interacts strongly with VP-16-HNF4 (500-fold), and rifampicin did not have any effect on the interaction. This is expected because rifampicin is not a ligand of HNF4. When PXR expression plasmid was cotransfected, addition of rifampicin strongly inhibited HNF4 and PGC-1 interaction. These results also suggest that rifampicin promotes PXR interaction with HNF4 and prevents PGC-1 interaction with HNF4. These results reveal for the first time that PGC-1 is a strong coactivator of PXR and rifampicin promotes PXR interaction with HNF4 and blocks PGC-1 from interaction with HNF4 or PXR.
|
and PXR bind to the CYP7A1 gene and interact with PGC-1 in the native CYP7A1 chromatin.
To further test the binding and interaction of HNF4, PXR, and PGC-1 to the CYP7A1 gene in vivo, we performed ChIP assays using human primary hepatocytes and HepG2 cells. PCR primers were designed to amplify a 391-bp fragment that contains both BARE-I and BARE-II sequences. Figure 7A shows that an antibody against PXR or HNF4 precipitated the CYP7A1 chromatins in primary human hepatocytes. For ChIP assay using HepG2 cells, PGC-1, HNF4, and PXR/RXR were overexpressed to increase receptor expression levels. Figure 7B shows that anti-PXR or HNF4 antibody precipitates CYP7A1 chromatin. These results confirmed that PXR and HNF4 bound to the native CYP7A1 chromatin. PGC-1 antibody also precipitated the CYP7A1 chromatin (Fig. 7B). Because PGC-1 is a coactivator that lacks a DBD, PGC-1 must interact with either HNF4 and/or PXR in the CYP7A1 chromatin. We also did ChIP assay using an antibody against FTF and confirmed that FTF bound to the BARE-II. We then studied the effect of rifampicin on the binding of these transcription factors to the CYP7A1 chromatin. Rifampicin treatments did not affect the binding of HNF4 and PXR to the CYP7A1 chromatin, but PGC-1 was absent from the chromatin (Fig. 7B). As a control, ChIP assays showed that rifampicin did not alter FTF binding to the chromatin. These data are consistent with our mammalian two-hybrid assay data, which suggest that rifampicin disrupts PGC-1 interaction with HNF4 or PXR.
|
interaction with PXR and PGC-12.
To further confirm the physical interaction between HNF4 and PXR in cells, anti-HNF4 antibody was used to immunoprecipitate HNF4 from cell extracts isolated from HepG2 and human primary hepatocytes. If PXR interacts with HNF4 in these cells, PXR should be co-IP by anti-HNF4 antibody and detected by Western immunoblot analysis using antibody against PXR. Because PXR protein levels are very low in HepG2 cells, HepG2 cells were transfected with PXR expression plasmid. Fig. 8A shows that antibody against PXR detected small amounts of PXR in HepG2 cell extracts precipitated by anti-HNF4 antibody. Rifampicin treatment markedly increased PXR protein in the IP. Anti-HNF4 antibody detected similar amounts of HNF4 in vehicle and rifampicin-treated cell extracts. We also did co-IP assay using human primary hepatocytes (no overexpression of PXR). Figure 8B shows co-IP assay using anti-HNF4 antibody to precipitate cell extracts from human primary hepatocytes. Anti-PXR antibody detected a small amount of PXR protein in the IP. Rifampicin pretreatment of hepatocytes markedly increased the amounts of PXR co-IP with the HNF4. Anti-PGC-1 antibody detected PGC-1 protein in the IP, and rifampicin pretreatment markedly reduced the amounts of PGC-1 protein in the IP. These data confirmed that rifampicin enhanced PXR interaction with HNF4 as revealed by mammalian two-hybrid assays shown in Fig. 5A. Rifampicin reduced PGC-1 co-IP with HNF4. These results are consistent with the finding from two-hybrid assays (Fig. 6, B and C) and ChIP assays (Fig. 7B) that rifampicin promoted PXR interaction with HNF4 and excluded PGC-1 from interaction with HNF4.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
. Q-PCR analysis reveals that CYP7A1 mRNA expression levels are repressed in human primary hepatocytes, likely because the negative effects of several negative regulators of the human CYP7A1 gene, FTF, COUP-TFII, PXR, Prox1 (see below), and SHP, may dominant over the positive effect of activators HNF4 and PGC-1. Q-PCR analysis showed that CDCA strongly reduced CYP7A1 but induced SHP mRNA expression in human primary hepatocytes. This inverse relationship of CDCA effect on CYP7A1 and SHP expression is consistent with the FXR/SHP mechanism of bile acid inhibition of CYP7A1 gene transcription. In contrast, rifampicin strongly reduced both CYP7A1 and SHP mRNA expression in human primary hepatocytes. This result provides strong evidence that SHP is not involved in PXR inhibition of CYP7A1 because reduction of a negative factor should result in induction of CYP7A1 expression. Our preliminary results show that rifampicin-activated PXR inhibits human SHP reporter activity (data not shown). In bile acid or drug-induced cholestasis, PXR may be activated to induce CYP3A4, which catalyzes 6-hydroxylation of bile acids for renal excretion. PXR also induces dehydroepiandrosterone sulfotransferase that excretes sulfate-conjugated bile acids. Our findings suggest that PXR inhibition of CYP7A1 gene transcription may be an additional mechanism for reducing intrahepatic bile acid concentrations and further protect the liver from toxic effects of bile acids and xenobiotics. PXR inhibition of SHP may also be a part of the PXR protection mechanism, because SHP could interact with PXR to inhibit CYP3A4 induction (22).
Our cell-based mammalian two-hybrid assay, co-IP, and ChIP assays all confirm the interactions of PXR with HNF4 and PGC-1. The interaction between PXR and HNF4 is novel and is consistent with our model that factors bound to the BARE-I and BARE-II interact and coordinately regulate basal transcription of the CYP7A1 gene (8). HNF4 is known to play a central role in regulation of bile acid synthesis (3). There are reports that HNF4 is required for PXR-mediated xenobiotic induction of CYP3A4 in mice (31) and that HNF4 regulates PXR expression and xenobiotic response during fetal liver development (14). Therefore, the PXR and HNF4 interaction may be physiologically important in the coordinated regulation of xenobiotic and bile acid metabolisms. Both two-hybrid assay and co-IP assay using HepG2 and primary hepatocyte extracts show that rifampicin markedly enhances PXR interaction with HNF4. The interaction between PXR and PGC-1 is as strong as between HNF4 and PGC-1. This is the first report that PXR is a target of PGC-1. Rifampicin is not required for PXR to interact with PGC-1 and does not affect HNF4 and PXR binding to CYP7A1 chromatin. On the other hand, rifampicin dissociates PGC-1 from CYP7A1 chromatin likely due to enhancing PXR interaction with HNF4.
Our results suggest a model for PXR, HNF4, and PGC-1 regulation of the CYP7A1 gene (Fig. 9). HNF4 and PGC-1 are the most important transactivators of the human CYP7A1 gene. PGC-1 interacts with HNF4, which binds to the BARE-II and activates human CYP7A1 gene. According to this study, PXR binds to the BARE-I and interacts with PGC-1 but may not be important in basal transcription of the CYP7A1 gene. It should be noted that our model does not suggest a ternary complex formation among PXR, HNF4, and PGC-1. PGC-1 is a coactivator that binds to the LBD of both PXR and HNF4. Therefore, it is not likely that PGC-1 interacts simultaneously with both nuclear receptors. Ubiquitous coactivators, SRC-1 and CBP, are known to interact with PXR and HNF4; however, our data show a much stronger interaction of PXR with PGC-1 than SRC-1. Nuclear receptor/coactivator complexes may facilitate the recruitment of histone acetytransferase to remodel the chromatin to allow recruitment of other coactivator/mediator complexes and assemble the general transcriptional machinery that regulates RNA polymerase II activity (10). On the basis of our results, we propose a model that activation of PXR by rifampicin would allow PXR to interact directly with HNF4 and exclude PGC-1 from interacting with HNF4, resulting in suppression of the CYP7A1 gene transcription. Corepressors, such as SMRT and NCoR1, have been shown to inhibit nuclear receptor activity (7). Our preliminary results show that the interaction of SMRT and NCoR1 with HNF4 and PXR is very weak. It is also possible that rifampicin-activated PXR may compete with HNF4 for PGC-1, thus inhibiting CYP7A1 gene transcription. This mechanism of PXR inhibition of gene transcription is not likely because PGC-1 interaction with PXR or HNF4 does not require a ligand and rifampicin-activated PXR and HNF4 is known to strongly induce, not inhibit, CYP3A4 gene transcription.
|
, or PXR. Another recent report (28) shows that Prox1 interacts with FTF and inhibits SHP gene expression and that Prox1 specifically interacts with histone deacetylase (HDAC)3. HDACs are known to form complexes with nuclear receptors and corepressors and inhibit gene transcription (13). The roles of these transcriptional repressors in mediating bile acid inhibition of CYP7A1 gene transcription remain to be elucidated.
In summary, this study provides the first evidence that PXR is able to interact with HNF4 and PGC-1 and plays a role in mediating rifampicin and bile acid inhibition of CYP7A1 gene transcription. On ligand binding, PXR interacts with HNF4, blocks PGC-1 interaction with HNF4, and results in inhibition of the CYP7A1 gene. The PXR pathway may be important for preventing bile acid and drug-induced cholestasis by inhibiting bile acid synthesis as well as stimulating bile acid transport, excretion, and detoxification. Thus PXR plays a central role in coordinated regulation of bile acid homeostasis and drug metabolism. This may be the mechanism for using rifampicin and steroids to treat pruritus of intrahepatic cholestasis and primary biliary cirrhosis (1, 12). Efficacious PXR agonists may be developed for treating these chronic liver diseases.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-hydroxylase gene transcription by bile acids. J Lipid Res 42: 1402–1412, 2001.-hydroxylase gene (CYP7A1) transcription by the liver orphan receptor (LXR). Gene 262: 257–265, 2001.[CrossRef][Web of Science][Medline]
-hydroxylase in the liver:Purification of cholesterol 7-hydroxylase and the immunochemical evidence for the induction of cholesterol 7-hydroxylase by cholestyramine and circadian rhythm. J Biol Chem 265: 3889–3897, 1990.-hydroxylase gene (CYP7A) by nuclear hormone receptors. J Lipid Res 39: 2192–2200, 1998. in control of the pregnane X receptor during fetal liver development. Hepatology 37: 1375–1384, 2003.[CrossRef][Web of Science][Medline]
-hydroxylase in the liver: cDNA cloning, sequencing and regulation of cholesterol 7-hydroxylase mRNA. J Biol Chem 265: 12012–12019, 1990.-hydroxylase (CYP7a). Endocrinology 143: 2548–2558., 2002. (PGC-1): transcriptional coactivator and metabolic regulator. Endocr Rev 24: 78–90, 2003.-hydroxylase gene. Mol Endocrinol 18: 2424–2439, 2004. (ERR). J Biol Chem 278: 9013–9018, 2003.-hydroxylase gene (CYP7A). J Lipid Res 41: 1–11, 2000. determines PXR- and CAR-mediated xenobiotic induction of CYP3A4. Nat Med 9: 220–224, 2003.[CrossRef][Web of Science][Medline]
-hydroxylase gene (CYP7A) in HepG2 cells. J Lipid Res 37: 1831–1841, 1996.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  K. T. Burke, P. S. Horn, P. Tso, J. E. Heubi, and L. A. Woollett Hepatic bile acid metabolism in the neonatal hamster: expansion of the bile acid pool parallels increased Cyp7a1 expression levels Am J Physiol Gastrointest Liver Physiol, July 1, 2009; 297(1): G144 - G151. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Han and J. Y. L. Chiang Mechanism of Vitamin D Receptor Inhibition of Cholesterol 7{alpha}-Hydroxylase Gene Transcription in Human Hepatocytes Drug Metab. Dispos., March 1, 2009; 37(3): 469 - 478. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Hariparsad, X. Chu, J. Yabut, P. Labhart, D. P. Hartley, X. Dai, and R. Evers Identification of pregnane-X receptor target genes and coactivator and corepressor binding to promoter elements in human hepatocytes Nucleic Acids Res., March 1, 2009; 37(4): 1160 - 1173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Li, H. Ma, and J. Y. L. Chiang TGF{beta}1, TNF{alpha}, and insulin signaling crosstalk in regulation of the rat cholesterol 7{alpha}-hydroxylase gene expression J. Lipid Res., September 1, 2008; 49(9): 1981 - 1989. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Ning, S. Dial, Y. Sun, J. Wang, J. Yang, and L. Guo Systematic and Simultaneous Gene Profiling of 84 Drug-Metabolizing Genes in Primary Human Hepatocytes J Biomol Screen, March 1, 2008; 13(3): 194 - 201. [Abstract] [PDF]
|
|
|
|
 |
|
 H.-L. Fang, S. C. Strom, E. Ellis, Z. Duanmu, J. Fu, Z. Duniec-Dmuchowski, C. N. Falany, J. L. Falany, T. A. Kocarek, and M. Runge-Morris Positive and Negative Regulation of Human Hepatic Hydroxysteroid Sulfotransferase (SULT2A1) Gene Transcription by Rifampicin: Roles of Hepatocyte Nuclear Factor 4{alpha} and Pregnane X Receptor J. Pharmacol. Exp. Ther., November 1, 2007; 323(2): 586 - 598. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Li, W. Chen, and J. Y. L. Chiang PXR induces CYP27A1 and regulates cholesterol metabolism in the intestine J. Lipid Res., February 1, 2007; 48(2): 373 - 384. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Zollner, M. Wagner, T. Moustafa, P. Fickert, D. Silbert, J. Gumhold, A. Fuchsbichler, E. Halilbasic, H. Denk, H.-U. Marschall, et al. Coordinated induction of bile acid detoxification and alternative elimination in mice: role of FXR-regulated organic solute transporter-{alpha}/beta in the adaptive response to bile acids Am J Physiol Gastrointest Liver Physiol, May 1, 2006; 290(5): G923 - G932. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
