Organ-specific alterations in RARα:RXRα abundance regulate rat Mrp2 (Abcc2) expression in obstructive cholestasis

GASTROENTEROLOGY 2002;123:599 – 607

BASIC – LIVER, PANCREAS, AND BILIARY TRACT Organ-Specific Alterations in RAR␣:RXR␣ Abundance Regulate Rat Mrp2 (Abcc2) Expression in Obstructive Cholestasis LEE A. DENSON,*,‡ ALAN BOHAN,‡,§ MATTHEW A. HELD,* and JAMES L. BOYER‡,§ *Department of Pediatrics and Yale Child Health Research Center, and the §Department of Internal Medicine and ‡Yale Liver Center, Yale University School of Medicine, New Haven, Connecticut

Background & Aims: Obstructive cholestasis is associated with adaptive changes in expression of hepatocyte transport proteins. These include a significant reduction in hepatic expression of Mrp2 (Abcc2), the principal canalicular multispecific organic anion transporter. Renal Mrp2 expression is preserved under these conditions. We have recently reported that the rat Mrp2 promoter is activated by RAR␣:RXR␣, and that interleukin 1␤ (IL-1␤) repressed promoter activity via this element. We hypothesized that cytokines, which are up-regulated in obstructive cholestasis, would reduce nuclear RAR␣: RXR␣ levels, and that this would be associated with suppression of hepatic Mrp2 expression. Methods: Male Sprague-Dawley rats were subjected to bile duct ligation (BDL) or sham surgery, and liver and kidney RNA and protein were isolated. Primary rat hepatocytes were treated with bile acids, retinoids, or cytokines, and RNA and protein were isolated. Mrp2 and RAR␣:RXR␣ protein abundance and activity were assessed by using electrophoretic mobility shift assays (EMSA) and immunoblots. IL-1␤ abundance was determined by enzymelinked immunosorbent assay. RAR␣, RXR␣, and Mrp2 RNA levels were determined by using ribonuclease protection assays (RPA). Results: Mrp2 down-regulation and IL-1␤ up-regulation were observed in liver after BDL. This was temporally associated with down-regulation of liver RAR␣:RXR␣ nuclear protein levels and binding to the Mrp2 promoter cis element. Renal RAR␣:RXR␣ and Mrp2 expression were preserved under these conditions. IL-1␤ treatment of primary hepatocytes reduced Mrp2 and RXR␣ expression. Conclusions: Organ-specific regulation of Mrp2 expression in obstructive cholestasis is associated with cytokine-dependent alterations in RAR␣:RXR␣ nuclear receptors. Preservation of renal Mrp2 expression may permit urinary excretion of toxic organic anions and xenobiotics under conditions in which biliary excretion is impaired.

iliary excretion of amphiphilic anionic conjugates is a critical physiologic function of the liver.1 The multidrug resistance protein 2 (Mrp2 or Abcc2) is the

B

principal, hepatic, canalicular, multispecific, organic, anion transporter, and mediates excretion of bilirubin and conjugates of lipophilic substances via bile.2 Although Mrp2 is also expressed on the apical surface of renal proximal tubules, its role in the kidney is less well understood.3 However, in obstructive cholestasis, in which biliary transport of Mrp2 substrates is impaired, adaptive mechanisms involving renal Mrp2 expression may permit urinary excretion of these potentially toxic compounds.4,5 The pronounced down-regulation of hepatic, but not renal, Mrp2 expression in obstructive cholestasis has suggested that substances that accumulate within the hepatocyte under these conditions may specifically regulate this gene.6,7 These could include bile acids, cytokines, bilirubin, or other Mrp2 substrates.8 –10 We have previously characterized a RAR␣:RXR␣ cis element in the rat Mrp2 promoter that regulates promoter induction by rexinoids and suppression by cytokines including interleukin 1␤ (IL-1␤).11 Other nuclear receptor ligands, including phenobarbital, rifampicin, and bile acids, also recently have been shown to activate the Mrp2 promoter.10 These compounds, which are ligands for the CAR (phenobarbital), PXR (rifampicin), or FXR (bile acids) nuclear receptors, induce Mrp2 expression via a common cis element that shares one half-site with the previously reported RAR␣:RXR␣ element. Conversely, in models of acute inflammation, lipopolysaccharide administration and hepatic Kupffer-cell activation suppress Mrp2 expression and function, and in vitro studies have shown that TNF␣ and IL-1␤ mediate these effects.7,12,13 Overall, Mrp2 expression would therefore be predicted to be Abbreviations used in this paper: BDL, bile duct ligation; CDCA, chenodeoxycholic acid; EMSA, electrophoretic mobility shift assays; IL-1␤, interleukin 1␤; RARE, RAR␣:RXR␣ cis element; RPA, ribonuclease protection assays; TNF, tumor necrosis factor. © 2002 by the American Gastroenterological Association 0016-5085/02/$35.00 doi:10.1053/gast.2002.34758

600

DENSON ET AL.

dependent on relative cellular levels of inflammatory cytokines, nuclear hormones, and their respective receptors. RXR␣ is a central heterodimeric regulator of hepatic bile acid synthesis and transport, and fatty acid, lipid, and xenobiotic metabolism.14 Although RXR␣ is also expressed in the renal proximal and distal tubules, its function in the kidney is less well understood.15 Recent studies have begun to elucidate the manner in which nuclear receptors including RXR␣ are regulated in cholestatic liver diseases. In cholestatic liver fibrosis induced by bile duct ligation (BDL), retinoid signaling is significantly down-regulated in stellate cells, the activated cells that produce collagen in chronic liver diseases. This is caused by down-regulation of RXR␣ and RAR␣ RNA expression and a reduction in retinoid ligands, which then leads to decreased expression of RXR␣:RAR␣ target genes.16 We have reported that inflammatory cholestasis induced by LPS administration has also been associated with down-regulation of hepatic RXR␣: RAR␣ proteins and target genes, including Mrp2 and the basolateral bile acid transporter, Ntcp.7,17 Moreover, TNF␣ has been shown to suppress activity of the RXR␤ promoter in vitro.18 Inflammatory cytokines including TNF␣, IL-1␤, and IL-6 are induced in liver after BDL, and for IL-1␤ this induction persists at significant levels for time points up to 2 weeks later.19 However, whether bile acids, cytokines, or retinoids directly regulated hepatocyte nuclear RXR␣:RAR␣ levels and Mrp2 expression in obstructive cholestasis was not known. We hypothesized that BDL would suppress hepatic Mrp2 expression, and that this would be temporally associated with down-regulation of RXR␣:RAR␣. In this study, we have shown a cytokine-dependent reduction in hepatic nuclear RXR␣:RAR␣ levels, which is temporally associated with down-regulation of Mrp2 expression.

Materials and Methods Materials Cell culture medium and reagents were obtained from Gibco Life Technologies, Inc. (Gaithersburg, MD). Mouse recombinant tumor necrosis factor ␣ (TNF␣) and IL-1␤ were obtained from R&D Systems (Minneapolis, MN). Chenodeoxycholic acid (CDCA) was obtained from Sigma (St. Louis, MO). The synthetic RXR␣ ligand, LG364, was provided by Dr. P. Heyman (Ligand Pharmaceuticals, San Diego, CA). RXR␣ (catalog no. sc-553), RAR␣ (catalog no. sc-551), and SH-PTP1 (catalog no. sc-287) polyclonal antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). The Mrp2 antibody has been described previously.7

GASTROENTEROLOGY Vol. 123, No. 2

Bile Duct Ligation Male Sprague-Dawley rats (body weight 175–250 g) were obtained from Charles River (Wilmington, MA) and maintained in a Yale Animal Care Facility. All rats were housed in a temperature- and humidity-controlled environment under a constant light cycle where they had free access to water. The protocol was approved by the Yale Animal Care and Use Committee and all animals received humane care as outlined in the Guide for Care and Use of Laboratory Animals (National Institutes of Health publication 86-23, revised 1985). Common BDL was performed under sterile conditions as previously described.7 The common bile duct was exposed, ligated twice close to the hilum of the liver immediately below the bifurcation, and then cut between the ligatures. Control animals underwent sham surgery in which the bile duct was exposed but not ligated. The animals were exsanguinated and the livers and kidneys were harvested at 1, 3, and 14 days after sham or BDL surgery.

Cytokine Enzyme-Linked Immunosorbent Assay Liver and kidney tissue homogenates were prepared and enzyme-linked immunosorbent assay was performed for rat IL-1␤ as per the manufacturer’s recommendations (R&D Systems).

Rat Hepatocyte Isolation and Cell Culture Hepatocytes were isolated from livers after collagenase digestion. The cells were suspended in Williams E Medium containing 10% fetal bovine serum, 1 ␮mol/L dexamethasone (Sigma), 250 nU/mL insulin (Sigma), and 1 ⫻ penicillinstreptomycin-glutamine (Gibco). The cells were plated (4 – 6 million cells/plate) on Matrigel (BD Biosciences, Bedford, MA). After adhesion the cells were treated with 100 ng/mL IL-1␤, 100 ng/mL TNF␣, 50 ␮mol/L CDCA, or 100 nmol/L LG364 alone. Control cells were treated with serum-free medium and all cells were recovered after 24 hours.

RNAse Protection Assays Total RNA was extracted by using TRIzol (Gibco Life Technologies). Concentration and purity were confirmed by spectrophotometry (Beckmann DU-640B; Beckmann Instruments, Palo Alto, CA). RNA was stored at ⫺80°C. Plasmids containing nuclear receptor complementary DNA (cDNA) (RXR␣ and RAR␣) or rat liver RNA were used to generate cDNA templates for subsequent probe synthesis and RNAse protection assay (RPA) analysis by using the following primers (Gibco-BRL, Rockville, MD): RXR␣:RAR␣ electrophoretic mobility shift assay (EMSA) probe (mrp2 RARE): rat mrp2 nt ⫺422 GGGTATTTAACATCTCTGTGAACTC ⫺398. Primers for generation of cDNA templates for RPA probes: hRXR␣s: CTCCCCGGGACAGCTGCATTCTCC; hT7RXR␣r: TAATACGACTCACTATAGGCGGCAGGTGTAGGTCAGGTCCTTG; hRAR␣s: GGCTGCAAGGGCTTCTTCCGCCGC; hT7RAR␣r: TAATACGACTCACTATAGGGCAGATCCG-

August 2002

CAGGATCAGGATGTC; rmrp2s: CTCCTACGGTTTCCAGATTGTCCT; rT7mrp2r: TAATACGACTCACTATAGGGACTTGATCAACCACGATTTGGGA; m28Ss: AGAAGGGCAAAAGCTCGCTT; mT728Sr: TAATACGACTCACTATAGGAGCAGGATTACCATGGCAAC. Where necessary, for EMSA, double-stranded oligonucleotides were generated by annealing of synthetic oligonucleotides with the respective complementary sequences. The RPA probes were synthesized and assays were performed as previously described. Briefly, antisense RNA probes were transcribed from respective cDNA templates by using T7 RNA polymerase and the protocol supplied with the Maxiscript kit (Ambion, Austin, TX). Biotinylated deoxycytidine triphosphate was incorporated into the transcription reaction and the full-length transcript for each cDNA template was purified and excised from a denaturing 5% polyacrylamide gel. The probe was eluted at 37°C into 0.5 mol/L ammonium acetate, 1 mmol/L ethylenediaminetetraacetic acid, and 0.1% sodium dodecyl sulfate solution. RPAs were performed by using the RPA III kit (Ambion) and between 5 and 20 mcg of RNA was isolated from rat liver or kidney. Each probe was hybridized with the RNA for 12 hours at 42°C. Unhybridized RNA was digested by RNAse T1 and diluted 1:100 for 30 minutes at 37°C. The protected hybrids were then resolved on a denaturing 5% polyacrylamide gel and subsequently transferred to a Brightstar Plus nylon membrane (Ambion). Detection was performed by using the Brightstar Biodetect kit (Ambion) and exposure to Biomax ML film (Kodak, Rochester, NY). A 28S RPA probe was used to control for loading and band signal intensity was quantified by densitometry.

Electrophoretic Mobility Shift Assays and Western Blot Analyses Nuclear proteins were prepared from rat liver and kidney by using the NE-PER kit (Pierce, Rockford, IL) as per the manufacturer’s recommendations. Membrane proteins were prepared as per previously published methods.7 Nuclear proteins were prepared from primary hepatocytes as previously described.20 The EMSA probe for the Mrp2 RARE was endlabeled with DIG-11– deoxyuridine triphosphate and EMSA was performed by using the DIG Gel Shift Kit, according to the manufacturer’s protocol (Boehringer Mannheim, Indianapolis, IN). Briefly, 10 ␮g of nuclear extract per sample was incubated in binding buffer with labeled oligonucleotide for 30 minutes, on ice. For competition assays, a 100-fold excess of unlabeled oligonucleotide was added to the binding reaction coincident with addition of the labeled oligonucleotide. For supershift assays, 2 ␮g of polyclonal antibody was added to the binding reaction for 2 hours on ice before addition of the labeled oligonucleotide. The samples were then resolved through a nondenaturing 6% polyacrylamide gel and subsequently transferred to a nylon membrane by using an electroblotter (Bio-Rad, Richmond, CA). After chemiluminescent detection, the membrane was exposed to Biomax ML Photographic Film (Kodak) and quantitation was assessed by densitometry (Molecular Dynamics, Sunnyvale, CA). For Western

MRP2 REGULATION IN OBSTRUCTIVE CHOLESTASIS

601

immunoblot analysis, 20 – 40 ␮g of nuclear protein was loaded on 7.5% or 12% sodium dodecyl sulfate–polyacrylamide gels. After electrophoresis, gels were subjected to electroblot onto nitrocellulose membranes and uniformity of loading and transfer was confirmed by Ponceau staining. The membranes were blocked overnight in TBS-Tween complete (20 mmol/L TrisHCL, 150 mmol/L NaCl, 5% nonfat dry milk, and 0.1% Tween-20) at 4°C. Blots were incubated for 1 hour at room temperature with polyclonal immunoglobulin G antibodies (RAR␣ 1:1000, RXR␣ 1:2000, Mrp2 1:10,000, SH-PTP1 1:1000). Blots were then washed and incubated with horseradish peroxidase conjugate antibody (Santa Cruz Biotechnologies). The immune complexes were detected with the Western Blot Chemiluminescence Reagent Plus Kit (NEN Life Science Products, Wilmington, DE) and quantified by using densitometry.

Statistical Analysis Data were expressed as the mean ⫾ standard deviation of experiments with at least 3 independent treatments per group. Differences among experimental groups were analyzed by analysis of variance (ANOVA). Where there were differences among the groups (P ⬍ 0.05), these were subjected to multiple comparisons against all other groups by using unpaired Student t test (InStat 2.03 software package; GraphPad Software, San Diego, CA). P ⬍ 0.05 was considered significant.

Results Obstructive Cholestasis Reduces Liver, but not Kidney, Expression of Mrp2 Prior studies from our laboratory and others have shown suppression of Mrp2 RNA and protein expression in the liver in obstructive cholestasis.7,21–23 We performed RPA by using total RNA from liver and kidney of rats subjected to BDL for 1, 3, and 14 days to determine whether changes in Mrp2 expression in this model were organ specific. As shown in Figure 1, we confirmed that Mrp2 RNA expression was reduced in liver after BDL, to 66% ⫾ 22% of sham levels by day 3, and to 44% ⫾ 8% of sham levels by day 14. In contrast to these changes in hepatic expression, Mrp2 expression was preserved in the kidney throughout the time course. Immunoblotting showed that Mrp2 protein abundance was reduced by 1 day after BDL in liver to 4% ⫾ 0.6% of sham levels (Figures 1C and D). These results confirmed organ-specific regulation of Mrp2 RNA and protein expression in obstructive cholestasis, which could be owing to differences in expression and/or function of nuclear hormone receptors. Because we had previously reported that inflammatory cytokines, including IL-1␤, which is induced by BDL, suppress the rat Mrp2 promoter via down-regulation of an RAR␣:RXR␣ cis ele-

602

DENSON ET AL.

GASTROENTEROLOGY Vol. 123, No. 2

Figure 1. Mrp2 expression is suppressed in liver and preserved in kidney in obstructive cholestasis. (A) RPA was performed by using total RNA from 1-, 3-, and 14-day sham and BDL liver and kidney and probes for rat Mrp2 and 28S. (C) Immunoblot was performed by using membrane proteins from 1-day sham and BDL liver and a polyclonal Mrp2 antibody. Representative blots at each time point are shown. (B, D) Mrp2 signal intensity relative to 28S for the RPA, or Mrp2 alone for the immunoblot, was determined by densitometry and is shown. N ⫽ 3 at each time point for sham and BDL, *P ⬍ 0.01 by ANOVA, for BDL value vs. sham control. (B) . . .■. . ., BDL kidney; —■—, BDL liver. (D) OD, optical density.

ment (RARE), we then examined RAR␣:RXR␣ expression after BDL.11,19 Nuclear Protein Binding to the Rat Mrp2 Promoter RAR␣:RXR␣ cis Element Is Reduced After BDL We have previously shown that cytokine treatment of HepG2 cells reduced binding of nuclear proteins to rat Ntcp and Mrp2 RARE, and that this suppressed promoter activity.11 We therefore examined protein: DNA binding with the rat Mrp2 RARE after BDL. We found that protein binding on EMSA was significantly reduced by 3 days after BDL to 52% ⫾ 15% of sham levels; this correlated temporally with the observed reduction in hepatic Mrp2 RNA expression (see Figures 1B and 2B). By comparison, protein binding was preserved in kidney under these conditions (Figure 2A). We had previously shown that RAR␣ and RXR␣ bind the Mrp2 promoter RARE cis element in liver.11 To confirm the specificity of binding in the current study, we performed competition with an excess of unlabeled RARE or unrelated Sp1/Sp3 oligonucleotides, and supershift with RXR and RAR isoform–specific antibodies. As shown (Figure 2C), the observed complex was specific, and contained RXR␣, RXR␤, and RAR␣. The reduction in RARE binding could account for the observed downregulation of Mrp2 RNA expression. Reduced binding to the Mrp2 RARE could be caused by reduced expres-

sion of RXR␣ or RAR␣ at the RNA or protein level, or to posttranslational changes affecting RAR␣:RXR␣ DNA binding or nuclear import. We therefore next examined expression of RAR␣ and RXR␣ after BDL in liver and kidney. RAR␣ and RXR␣ Nuclear Protein Abundance Is Reduced in Liver After Bile Duct Ligation As shown in Figure 3A, hepatic nuclear RXR␣ and RAR␣ protein abundance was significantly reduced beginning 1 day after BDL to 42% ⫾ 22% and 52% ⫾ 30% of sham levels, respectively. This persisted throughout the time course, with RXR␣ and RAR␣ protein abundance reduced to 43% ⫾ 18% and undetectable levels, respectively, by day 14. These changes preceded the observed reduction in RXR␣ and RAR␣ RNA expression, which was not significant until 14 days after BDL (see Figure 4), showing a significant posttranscriptional suppression of retinoid receptor expression after BDL. The observed reduction in RAR␣:RXR␣ nuclear levels would then be expected to lead to down-regulation of target genes, including Mrp2, which are regulated by retinoid receptors. In contrast to the significant reduction in RAR␣:RXR␣ nuclear levels in the liver after BDL, we found that RAR␣:RXR␣ nuclear levels were relatively preserved in kidney (see Figure 3C). This could account for the relative preservation of renal RAR␣:

August 2002

MRP2 REGULATION IN OBSTRUCTIVE CHOLESTASIS

603

Figure 2. RAR␣:RXR␣ binding to the Mrp2 promoter is reduced after BDL. (A) EMSA was performed by using nuclear protein from 1-, 3-, and 14-day sham and BDL liver and 14-day sham and BDL kidney and a probe for the rat Mrp2 promoter RAR␣:RXR␣ cis element. Representative blots at each time point are shown. (B) Signal intensity was determined by densitometry and normalized to sham values at each time point. N ⫽ 3 at each time point for sham and BDL, *P ⬍ 0.01 by ANOVA for BDL value vs. sham control. (C) EMSA competition and supershift assays were also performed as shown. COMP, oligonucleotide competitor; SP, specific 100-fold excess of RARE unlabeled oligonucleotide; NSP, nonspecific 100-fold excess of unrelated Sp1/Sp3 consensus unlabeled oligonucleotide, L2A; AB, antibody used for supershift assay.20

RXR␣ DNA binding and Mrp2 expression under these conditions. To confirm overall protein loading and stability, the level of SH-PTP1 (a nuclear tyrosine phosphatase) expression was also determined by using these liver nuclear extracts. As shown in Figure 4A, SH-PTP1 nuclear protein levels were the same in sham and BDL samples, confirming overall integrity of the protein samples. RPA was then used to determine whether these alterations in retinoid-receptor protein levels were caused by changes in RXR␣ and RAR␣ RNA expression. RAR␣ and RXR␣ RNA Expression Is Not Reduced in Liver Until 14 Days After Bile Duct Ligation RPA was performed by using total RNA from liver and kidney harvested 1, 3, or 14 days after sham surgery or BDL. As shown in Figure 4, RNA expression of RAR␣ and RXR␣ was reduced to 28% ⫾ 7% and 59% ⫾ 4%, respectively, relative to sham levels, by 14 days after BDL. In contrast to this, RNA expression of RXR␣ and RAR␣ were preserved in the kidney at all time points after BDL (data not shown). Taken together, these results indicated that primarily posttranscriptional

Figure 3. RAR␣ and RXR␣ nuclear protein abundance is reduced after BDL in liver and preserved in kidney. Immunoblots were performed by using nuclear protein from (A) 1-, 3-, and 14-day sham and BDL liver and (C) kidney and polyclonal antibodies for RXR␣, RAR␣, and SHPTP1. Representative blots at each time point are shown. (B) Signal intensity was determined by densitometry and normalized to sham values at each time point. N ⫽ 3 at each time point for sham and BDL, *P ⬍ 0.01 by ANOVA for BDL value vs. sham control. OD, optical density normalized to sham value. (B) Solid line, RXR␣; dotted line, RAR␣.

604

DENSON ET AL.

Figure 4. RAR␣ and RXR␣ RNA expression is reduced after BDL in liver. RPA was performed by using total RNA from 1-, 3-, and 14-day sham and BDL liver and probes for RXR␣, RAR␣, and 28S. (A) Representative blots at each time point are shown. (B) Signal intensity was determined by densitometry and normalized to sham values at each time point. N ⫽ 3 at each time point for sham and BDL, *P ⬍ 0.01 by ANOVA for BDL values vs. sham control. (B) . . .■. . ., RAR␣; —■—, RXR␣.

GASTROENTEROLOGY Vol. 123, No. 2

pression. As shown in Figure 6, treatment with IL-1␤ significantly reduced nuclear protein abundance of RXR␣ in primary rat hepatocytes, to 33% ⫾ 28% of control values. Treatment with CDCA did not yield a significant reduction. Treatment with TNF␣ or LG364 led to a modest reduction or induction of RXR␣ protein abundance, respectively (data not shown). Treatment with these agents did not affect RAR␣ protein abundance (Figure 6A). However, IL-1␤ treatment did reduce both RAR␣ (67% ⫾ 10% of control values) and RXR␣ (48% ⫾ 4% of control values) RNA expression, as shown in Figure 7. These data confirmed that an inflammatory cytokine, IL-1␤, which is induced in the liver in obstructive cholestasis, can significantly suppress Mrp2 RNA and RXR␣ protein expression. This mechanism could contribute to the down-regulation of Mrp2 expression, which was observed in the liver after BDL.

suppression of nuclear RAR␣:RXR␣ by substances that accumulate within the hepatocyte after BDL, including cytokines, could account for the observed reduction in Mrp2 expression. We therefore examined IL-1␤ abundance in liver and kidney after BDL and the effect of cytokine treatment on Mrp2 and RXR␣:RAR␣ protein expression in primary rat hepatocytes to test more directly this potential mechanism. IL-1␤ Suppresses Mrp2 and RXR␣ Expression in Primary Rat Hepatocytes Previous studies have shown hepatic induction of inflammatory cytokines including TNF␣, IL-6, and IL-1␤ after BDL.19,24,25 Moreover, TNF␣ and IL-1␤ have been shown to directly suppress Mrp2 RNA expression in primary rat hepatocytes.12 We also found that IL-1␤ was up-regulated in 1-day BDL liver, relative to sham liver and sham or BDL kidney (Figure 5A). We have observed previously that the concentration of bile acids including CDCA increased in liver after BDL, and that CDCA cotreatment prevented retinoid-receptor up-regulation of Ntcp RNA expression in primary rat hepatocytes.28 However, a synthetic Fxr agonist has recently been shown to up-regulate Mrp2 RNA expression in primary rat hepatocytes.10 Therefore, primary rat hepatocytes were treated with either IL-1␤ or CDCA for 24 hours, and RNA and nuclear proteins were isolated. As shown in Figure 5B, treatment with IL-1␤ significantly reduced Mrp2 RNA expression, to 26% ⫾ 5% of control values. Treatment with CDCA did not alter Mrp2 ex-

Figure 5. Mrp2 RNA expression is reduced by IL-1␤ in primary rat hepatocytes. (A) Enzyme-linked immunosorbent assay for rat IL-1␤ was performed by using tissue homogenates from 1-day sham and BDL liver and kidney. (B) RPA for Mrp2 was performed by using total RNA isolated from primary rat hepatocytes 24 hours after treatment with IL-1␤ (100 ng/mL) or CDCA (50 ␮mol/L). Representative blots are shown. (C) Signal intensity was determined by densitometry and is shown. N ⫽ 4 –5 for each treatment condition, *P ⬍ 0.01 by ANOVA for IL-1␤ abundance in BDL liver vs. sham liver and for Mrp2 RNA level in IL-1␤–treated hepatocytes vs. untreated hepatocytes. PG/MG, picogram IL-1␤/mg tissue; OD, optical density.

August 2002

Figure 6. RXR␣ nuclear protein abundance is reduced by IL-1␤ in primary rat hepatocytes. (A) Immunoblots were performed by using nuclear protein isolated from primary rat hepatocytes 24 hours after treatment with IL-1␤ (100 ng/mL) or CDCA (50 ␮mol/L) and polyclonal antibodies for RXR␣, RAR␣, and SH-PTP1. Representative blots are shown. (B) Signal intensity was determined by densitometry and is shown. N ⫽ 4 –5 for each treatment condition, *P ⬍ 0.01 by ANOVA for RXR␣ abundance in IL-1␤–treated hepatocytes vs. untreated hepatocytes. OD, optical density; C, control.

Discussion The multidrug resistance protein 2 (Mrp2 or Abcc2) is the principal hepatic canalicular multispecific organic anion transporter, and it mediates excretion of bilirubin-diglucuronide and a variety of conjugates of lipophilic substances via bile under normal conditions. Our group and others have previously characterized down-regulation of hepatic Mrp2 RNA and protein expression and function in experimental models of cholestatic liver disease.7,21–23 More recently, we and other investigators have observed that, in obstructive cholestasis, renal Mrp2 expression is preserved, providing a means for urinary excretion of potentially toxic hepatic Mrp2 substrates.5,23 We have found that Mrp2 protein abundance is increased by 2-fold in kidney in 14-day BDL rats; this occurs in the absence of a change in Mrp2 RNA expression.5 In agreement with this, Tanaka et al.23 recently reported that Mrp2 protein abundance was increased by 4- to 5-fold in kidney between 1 and 5 days after BDL. This was associated with a modest initial increase in Mrp2 RNA expression that returned to baseline by 5 days after BDL.23 Mrp2 expression and promoter activity is reduced by inflammatory cytokines, indicating a potential mechanism for acute-phase downregulation of the gene.11,12 However, molecular mechanisms that could account for the observed organ-specific regulation of Mrp2 in obstructive cholestasis were not known. In this study, we have identified a mechanism to account for organ-specific regulation of Mrp2 expression in obstructive cholestasis, via IL-1␤– dependent alterations in regulatory RAR␣:RXR␣ nuclear receptors.

MRP2 REGULATION IN OBSTRUCTIVE CHOLESTASIS

605

The characterization of the class II nuclear receptors (for which RXR␣ is the obligate heterodimeric partner) in normal hepatic physiology over the past few years has contributed greatly to our understanding of regulation of hepatobiliary transport systems.26 Importantly, nuclear hormones that are substrates for these transporters, such as bile acids, have also been shown to be ligands for regulatory nuclear-hormone receptors that control their gene expression.10,14 In this manner, intrahepatic accumulation of a given transporter substrate can induce expression of the associated transporter in a feed-forward manner, avoiding potentially toxic intracellular accumulation of these compounds. We had previously described up-regulation of the rat Mrp2 promoter by the RAR␣: RXR␣ nuclear receptor; recently, this observation has been extended to include up-regulation of Mrp2 by nuclear receptors for bile acids (FXR:RXR␣), phenobarbital (CAR:RXR␣), and rifampicin (PXR:RXR␣).10 Interestingly, this appears to involve a shared cis element that extends over a 31-bp region of the proximal promoter, and includes overlapping DR5 (RAR␣:RXR␣) and ER8 (FXR:RXR␣, PXR:RXR␣, and CAR:RXR␣) elements. These studies have shown the manner in which Mrp2 expression and function could be induced in response to intracellular accumulation of bile acids or xenobiotics. However, the manner in which Mrp2 expression was specifically reduced in the liver in disease

Figure 7. RXR␣ and RAR␣ RNA expression is reduced by IL-1␤ in primary rat hepatocytes. RPA was performed by using total RNA isolated from primary rat hepatocytes 24 hours after treatment with IL-1␤ (100 ng/mL) or CDCA (50 ␮mol/L) and probes for RXR␣, RAR␣, and 28S. (A) Representative blots are shown. (B) Signal intensity was determined by densitometry. N ⫽ 4 –5 for each treatment condition, *P ⬍ 0.01 by ANOVA for RAR␣ or RXR␣ RNA level in IL-1␤–treated hepatocytes vs. untreated hepatocytes, and for RXR␣ RNA level in CDCA-treated hepatocytes vs. untreated hepatocytes. OD, optical density.

606

DENSON ET AL.

states, including obstructive cholestasis, had not been described. We had previously determined that the rat Mrp2 promoter was suppressed by IL-1␤ via down-regulation of binding of RAR␣:RXR␣ to the positive-acting cis element.11 Because inflammatory cytokines are known to be induced in the liver in obstructive cholestasis, we hypothesized that cytokine-dependent reduction in RAR␣:RXR␣ binding to the Mrp2 promoter cis element would be associated with a reduction in gene expression.19,24,25 This hypothesis would be consistent with an increasing number of studies that have characterized class II nuclear receptors as negative acute-phase proteins.27 Under these conditions, reduced nuclear levels of class II nuclear receptors may become limiting, leading to down-regulation of nuclear hormone– dependent target genes, despite intracellular accumulation of nuclear hormones including bile acids and xenobiotics.10,28 Lipopolysaccharide administration has also been associated with significant inhibition of RXR␣ and RAR␣ expression, at both the protein and RNA levels, and leads to down-regulation of Mrp2 expression.7,27 In this study, we found that these nuclear receptors were initially suppressed at the protein level after BDL, with relative preservation of RNA expression. The in vivo reduction in Mrp2 and RXR␣ expression was reproduced by IL-1␤ treatment of primary rat hepatocytes, indicating that inflammatory cytokines may mediate this change. In contrast, RAR␣ abundance was not reduced by IL-1␤ treatment in vitro, despite a profound reduction in the BDL model. This finding may indicate that alternate cytokines or cytokine signaling pathways mediate this effect in vivo. Alternately, we did observe that basal RAR␣ protein expression was significantly lower in the cell culture than the animal model; this may reflect reduction in RAR␣ abundance to a constitutive low level of expression in cell culture, which would no longer be regulated by cytokines. RXR␣ is a phosphoprotein, and so could be the target of multiple cellular signaling pathways that are known to be activated by cytokines.29 In some studies, inducible phosphorylation of RXR␣ has been shown to affect protein stability and/or function, although these results have not been uniform.30,31 Whether inflammatory cytokines are regulating RXR␣ stability or nuclear localization in obstructive cholestasis is currently under study. The reduction in liver RXR␣ abundance in obstructive cholestasis has important implications for metabolic functions that are now known to be regulated by class II nuclear receptors for which RXR␣ is the obligate heterodimeric partner.14 These include bile acid synthesis

GASTROENTEROLOGY Vol. 123, No. 2

and transport, cytochrome P450 drug metabolism, and cholesterol and fatty acid metabolism.26 It would be predicted that reductions in RXR␣ would lead to reduced expression of target genes that control rate-limiting steps in these pathways. For example, induction of Mrp2 by compounds such as phenobarbital and rifampicin may be impaired in disease states, including obstructive cholestasis, in which the regulatory nuclear receptors are reduced. This will have important implications for the design of drug therapy based on activating nuclear receptors. In terms of the specific role of Mrp2 in biliary excretion, adaptive up-regulation of the basolateral transport protein Mrp3, which shares similar substrate specificity, and relative preservation of renal Mrp2 expression, may provide an alternate pathway for excretion of these compounds from the body in obstructive cholestasis.5,23,32 In this respect, down-regulation of nuclear receptors by cytokines may adaptively alter these synthetic and transport systems in disease states, in much the same way that the specific nuclear hormones control the expression of rate-limiting genes under more physiologic conditions.

References 1. Nathanson MH, Boyer JL. Mechanisms and regulation of bile secretion. Hepatology 1991;14:551–566. 2. Meier PJ. Molecular mechanisms of hepatic bile salt transport from sinusoidal blood into bile. Am J Physiol 1995;269:G801– G812. 3. Van Aubel RA, Masereeuw R, Russel FG. Molecular pharmacology of renal organic anion transporters. Am J Physiol 2000;279: F216 –F232. 4. Laouari D, Yang R, Veau C, Blanke I, Friedlander G. Two apical multidrug transporters, p-gp and mrp2, are differently altered in chronic renal failure. Am J Physiol 2001;280:F636 –F645. 5. Lee J, Azzaroli F, Wang L, Soroka CJ, Gigliozzi A, Setchell KDR, Kramer W, Boyer JL. Adaptive regulation of bile salt transporters in kidney and liver in obstructive cholestasis in the rat. Gastroenterology 2001;121:1473–1484. 6. Kubitz R, Wettstein M, Warskulat U, Haussinger D. Regulation of the multidrug resistance protein 2 in the rat liver by lipopolysaccharide and dexamethasone. Gastroenterology 1999;116:401– 410. 7. Trauner M, Arrese M, Soroka CJ, Ananthanarayanan M, Koeppel TA, Schlosser SF, Suchy FJ, Keppler D, Boyer JL. The rat canalicular conjugate export pump (mrp2) is down-regulated in intrahepatic and obstructive cholestasis. Gastroenterology 1997;113: 255–264. 8. Schrenk D, Baus PR, Ermel N, Klein C, Vorderstemann B, Kauffmann HM. Up-regulation of transporters of the mrp family by drugs and toxins. Toxicol Lett 2001;120:51–57. 9. Fickert P, Zollner G, Fuchsbichler A, Stumptner C, Pojer C, Zenz R, Lammert F, Stieger B, Meier PJ, Zatloukal K, Denk H, Trauner M. Effects of ursodeoxycholic and cholic acid feeding on hepatocellular transporter expression in mouse liver. Gastroenterology 2001;121:170 –183. 10. Kast HR, Goodwin B, Tarr PT, Jones SA, Anisfeld AM, Stoltz CM, Tontonoz P, Kliewer S, Willson TM, Edwards PA. Regulation of multidrug resistance-associated protein 2 (abcc2) by the nuclear receptors pregnane x receptor, farnesoid x-activated receptor,

August 2002

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

and constitutive androstane receptor. J Biol Chem 2002;277: 2908 –2915. Denson LA, Auld KL, Schiek DS, Mcclure MH, Mangelsdorf DJ, Karpen SJ. Interleukin-1beta suppresses retinoid transactivation of two hepatic transporter genes involved in bile formation. J Biol Chem 2000;275:8835– 8843. Nakamura J, Nishida T, Hayashi K, Kawada N, Ueshima S, Sugiyama Y, Ito T, Sobue K, Matsuda H. Kupffer cell-mediated down regulation of rat hepatic cmoat/mrp2 gene expression. Biochem Biophys Res Commun 1999;255:143–149. Vos TA, Hooiveld GJ, Koning H, Childs S, Meijer DK, Moshage H, Jansen PL, Muller M. Up-regulation of the multidrug resistance genes, mrp1 and mdr1b, and down-regulation of the organic anion transporter, mrp2, and the bile salt transporter, spgp, in endotoxemic rat liver. Hepatology 1998;28:1637–1644. Wan YJ, An D, Cai Y, Repa JJ, Hung-Po Chen T, Flores M, Postic C, Magnuson MA, Chen J, Chien KR, French S, Mangelsdorf DJ, Sucov HM. Hepatocyte-specific mutation establishes retinoid x receptor alpha as a heterodimeric integrator of multiple physiological processes in the liver. Mol Cell Biol 2000;20:4436 – 4444. Sugawara A, Sanno N, Takahashi N, Osamura RY, Abe K. Retinoid x receptors in the kidney: their protein expression and functional significance. Endocrinology 1997;138:3175–3180. Ohata M, Lin M, Satre M, Tsukamoto H. Diminished retinoic acid signaling in hepatic stellate cells in cholestatic liver fibrosis. Am J Physiol 1997;272:G589 –G596. Trauner M, Arrese M, Lee H, Boyer JL, Karpen SJ. Endotoxin downregulates rat hepatic ntcp gene expression via decreased activity of critical transcription factors. J Clin Invest 1998;101: 2092–2100. Sugawara A, Uruno A, Nagata T, Taketo MM, Takeuchi K, Ito S. Characterization of mouse retinoid x receptor (rxr)-beta gene promoter: negative regulation by tumor necrosis factor (tnf)-alpha. Endocrinology 1998;139:3030 –3033. Sewnath ME, Van Der Poll T, Ten Kate FJ, Van Noorden CJ, Gouma DJ. Interleukin-1 receptor type i gene-deficient bile ductligated mice are partially protected against endotoxin. Hepatology 2002;35:149 –158. Denson LA, Menon RK, Shaufl A, Bajwa HS, Williams CR, Karpen SJ. Tnf-alpha downregulates murine hepatic growth hormone receptor expression by inhibiting sp1 and sp3 binding. J Clin Invest 2001;107:1451–1458. Lee J, Boyer JL. Molecular alterations in hepatocyte transport mechanisms in acquired cholestatic liver disorders [in process citation]. Semin Liver Dis 2000;20:373–384. Paulusma CC, Kothe MJ, Bakker CT, Bosma PJ, Van Bokhoven I, Van Marle J, Bolder U, Tytgat GN, Oude Elferink RP. Zonal downregulation and redistribution of the multidrug resistance protein 2 during bile duct ligation in rat liver. Hepatology 2000;31:684 – 693. Tanaka Y, Kobayashi Y, Gabazza EC, Higuchi K, Kamisako T, Kuroda M, Takeuchi K, Iwasa M, Kaito M, Adachi Y. Increased renal expression of bilirubin glucuronide transporters in a rat

MRP2 REGULATION IN OBSTRUCTIVE CHOLESTASIS

24.

25.

26.

27.

28.

29.

30.

31.

32.

607

model of obstructive jaundice. Am J Physiol 2002;282:G656 – G662. Liu TZ, Lee KT, Chern CL, Cheng JT, Stern A, Tsai LY. Free radical-triggered hepatic injury of experimental obstructive jaundice of rats involves overproduction of proinflammatory cytokines and enhanced activation of nuclear factor kappaB. Ann Clin Lab Sci 2001;31:383–390. Plebani M, Panozzo MP, Basso D, De Paoli M, Biasin R, Infantolino D. Cytokines and the progression of liver damage in experimental bile duct ligation. Clin Exp Pharmacol Physiol 1999;26: 358 –363. Lu TT, Repa JJ, Mangelsdorf DJ. Orphan nuclear receptors as elixirs and fixers of sterol metabolism. J Biol Chem 2001;276: 37735–37738. Beigneux AP, Moser AH, Shigenaga JK, Grunfeld C, Feingold KR. The acute phase response is associated with retinoid x receptor repression in rodent liver. J Biol Chem 2000;275:16390 – 16399. Setchell KD, Rodrigues CM, Clerici C, Solinas A, Morelli A, Gartung C, Boyer J. Bile acid concentrations in human and rat liver tissue and in hepatocyte nuclei. Gastroenterology 1997;112: 226 –235. Lee HY, Suh YA, Robinson MJ, Clifford JL, Hong WK, Woodgett JR, Cobb MH, Mangelsdorf DJ, Kurie JM. Stress pathway activation induces phosphorylation of retinoid x receptor. J Biol Chem 2000;275:32193–32199. Kopf E, Plassat JL, Vivat V, de The H, Chambon P, Rochette-Egly C. Dimerization with retinoid x receptors and phosphorylation modulate the retinoic acid-induced degradation of retinoic acid receptors alpha and gamma through the ubiquitin-proteasome pathway. J Biol Chem 2000;275:33280 –33288. Adam-Stitah S, Penna L, Chambon P, Rochette-Egly C. Hyperphosphorylation of the retinoid x receptor alpha by activated c-jun nh2-terminal kinases. J Biol Chem 1999;274:18932–18941. Soroka CJ, Lee JM, Azzaroli F, Boyer JL. Cellular localization and up-regulation of multidrug resistance-associated protein 3 in hepatocytes and cholangiocytes during obstructive cholestasis in rat liver. Hepatology 2001;33:783–791.

Received December 12, 2001. Accepted May 9, 2002. Address requests for reprints to: Lee A. Denson, M.D., Yale Child Health Research Center, Department of Pediatrics, 464 Congress Avenue, P.O. Box 208081, New Haven, Connecticut 06520-8081. e-mail: [email protected]; fax: (203) 737-5972. Supported by National Institutes of Health grant DK02700 (to L.A.D.), DK25636 (to J.L.B.), Charles H. Hood Foundation (to L.A.D.), and Yale Liver Center (DK P30-34989). Previously published in abstract form (Hepatology 2001;34:367A) and presented at the annual meeting of the American Association for the Study of Liver Diseases. The authors acknowledge the excellent technical support of Himmat Bajwa.