Cholestasis leads to hepatic and systemic accumulation

Transcription

1 CAR and PXR Agonists Stimulate Hepatic Bile Acid and Bilirubin Detoxification and Elimination Pathways in Mice Martin Wagner, 1 Emina Halilbasic, 1 Hanns-Ulrich Marschall, 2 Gernot Zollner, 1 Peter Fickert, 1 Cord Langner, 3 Kurt Zatloukal, 3 Helmut Denk, 3 and Michael Trauner 1 Induction of hepatic phase I/II detoxification enzymes and alternative excretory pumps may limit hepatocellular accumulation of toxic biliary compounds in cholestasis. Because the nuclear xenobiotic receptors constitutive androstane receptor (CAR) and pregnane X receptor (PXR) regulate involved enzymes and transporters, we aimed to induce adaptive alternative pathways with different CAR and PXR agonists in vivo. Mice were treated with the CAR agonists phenobarbital and 1,4-bis-[2-(3,5-dichlorpyridyloxy)]benzene, as well as the PXR agonists atorvastatin and pregnenolone-16 -carbonitrile. Hepatic bile acid and bilirubin-metabolizing/detoxifying enzymes (Cyp2b10, Cyp3a11, Ugt1a1, Sult2a1), their regulatory nuclear receptors (CAR, PXR, farnesoid X receptor), and bile acid/organic anion and lipid transporters (Ntcp, Oatp1,2,4, Bsep, Mrp2-4, Mdr2, Abcg5/8, Asbt) in the liver and kidney were analyzed via reverse-transcriptase polymerase chain reaction and Western blotting. Potential functional relevance was tested in common bile duct ligation (CBDL). CAR agonists induced Mrp2-4 and Oatp2; PXR agonists induced only Mrp3 and Oatp2. Both PXR and CAR agonists profoundly stimulated bile acid hydroxylating/detoxifying enzymes Cyp3a11 and Cyp2b10. In addition, CAR agonists upregulated bile acid sulfating Sult2a1 and bilirubin-glucuronidating Ugt1a1. These changes were accompanied by reduced serum levels of bilirubin and bile acids in healthy and CBDL mice and by increased levels of polyhydroxylated bile acids in serum and urine of cholestatic mice. Atorvastatin significantly increased Oatp2, Mdr2, and Asbt, while other transporters and enzymes were moderately affected. In conclusion, administration of specific CAR or PXR ligands results in coordinated stimulation of major hepatic bile acid/bilirubin metabolizing and detoxifying enzymes and hepatic key alternative efflux systems, effects that are predicted to counteract cholestasis. (HEPATOLOGY 2005;42: ) See Editorial on Page 266 Abbreviations: FXR, farnesoid X receptor; CAR, constitutive androstane receptor; PXR, pregnane X receptor; CBDL, common bile duct ligation; PCN, pregnenolone- 16 -carbonitrile; TCPOBOP, 1,4-bis-[2-(3,5-dichlorpyridyloxy)]benzene; DMSO, dimethyl sulfoxide; PB, phenobarbital; mrna, messenger RNA; PPAR-, peroxisome proliferator-activated receptor. From the 1 Laboratory of Experimental and Molecular Hepatology, Division of Gastroenterology and Hepatology, Department of Medicine, Medical University, Graz, Austria; 2 Karolinska Institutet, Department of Medicine, Karolinska University Hospital Huddinge, Stockholm, Sweden; and 3 Department of Pathology, Medical University, Graz, Austria. Received November 25, 2004; accepted May 12, Supported by grant P (M.T.) from the Austrian Science Foundation and by grants from Karolinska Institutet, Ruth and Richard Julins Fund, and the Swedish Medical Association (H.-U.M.). Address reprint requests to: Michael Trauner, M.D., Laboratory of Experimental and Molecular Hepatology, Division of Gastroenterology and Hepatology, Department of Medicine, Medical University Graz, Auenbruggerplatz 15, A-8036 Graz, Austria. michael.trauner@meduni-graz.at; fax: (43) Copyright 2005 by the American Association for the Study of Liver Diseases. Published online in Wiley InterScience ( DOI /hep Potential conflict of interest: Nothing to report. Cholestasis leads to hepatic and systemic accumulation of potentially toxic biliary compounds such as bile acids and bilirubin, resulting in liver damage and jaundice. 1 Spontaneous anticholestatic defense mechanisms of hepatocytes comprise downregulation of import systems (e.g., Ntcp, Oatp1, Oatp4) 2 and adaptive induction of basolateral alternative export pumps (e.g., Mrp3, Mrp4). 3,4 This adaptive response may reduce hepatocellular accumulation of bile acids and bilirubin and thus may limit cholestatic liver injury. In addition, detoxification of accumulating biliary compounds via phase I hydroxylation and phase II conjugation may counteract cholestatic liver damage by rendering hydrophobic substrates less toxic and better soluble for biliary and urinary excretion. 5,6 However, intrinsic hepatocellular adaptive induction of transporters and enzymes in cholestasis is too weak to prevent ongoing liver injury. 4,7 420

2 HEPATOLOGY, Vol. 42, No. 2, 2005 WAGNER ET AL. 421 In addition to the classical bile acid receptor, farnesoid X receptor (FXR), also the xenobiotic receptors, constitutive androstane receptor (CAR) and pregnane X receptor (PXR), critically participate in the regulation of genes involved in the detoxification and transport of bile acids and bilirubin. 8 Manipulation of PXR and CAR may therefore represent an attractive pharmacological option for further stimulation of otherwise weak, intrinsically induced detoxification and export pathways in cholestasis. 4,7 The effects of CAR and PXR ligands on the orchestra of bile acid/bilirubin transport systems and detoxifying enzymes in concert with their regulating nuclear receptors have so far not been studied in vivo. We therefore aimed to coordinately stimulate adaptive detoxification (phase I and II) and hepatic (phase III) as well as renal elimination pathways for biliary compounds with different CAR and PXR agonists in vivo and to test the potential functional relevance in a mouse model of cholestasis. This study provides evidence that (1) CAR and PXR ligands have coordinated effects on phase I, phase II, and phase III adaptive reactions at both RNA and protein levels; (2) these effects can also be induced by agonists relevant for human therapy; and (3) these CAR and PXR effects reduce serum bilirubin and bile acid levels in cholestasis. Materials and Methods Animals. Six- to eight-week-old Swiss albino mice (Institute of Laboratory Animal Research, Medical University Vienna, Himberg, Austria) were housed with a 12:12-hour light/dark cycle and permitted ad libitum consumption of water and a standard mouse diet. The experimental protocols were approved by the local Animal Care and Use Committee, according to criteria outlined in the Guide for the Care and Use of Laboratory Animals prepared by the National Academy of Sciences, as published by the National Institutes of Health (NIH publication 86-23, revised 1985). Materials. Pregnenolone-16 -carbonitrile (PCN), 1,4- bis-[2-(3,5-dichlorpyridyloxy)]benzene (TCPOBOP), and dimethyl sulfoxide (DMSO) were purchased from Sigma (Steinheim, Germany); phenobarbital (PB)-free acid was purchased from the institutional pharmacy; atorvastatin was purchased from Goedecke GmbH (Freiburg, Germany). Experiments and Protocols in Healthy (Noncholestatic) Animals. Mice were treated at the same time points in the morning with intraperitoneal injections of the clinically used CAR ligand PB (100 mg/kg body weight, 9 dissolved in DMSO); the specific, more potent rodent CAR ligand TCPOBOP (3 mg/kg body weight, 9 dissolved in corn oil); and the specific, potent, rodent PXR ligand PCN (75 mg/kg body weight, 10 dissolved in DMSO) over a period of 3 days. As a vehicle, 100 L of DMSO (for comparison in the PB and PCN group) or corn oil (for comparison in the TCPOBOP group) were applied in the same manner. Atorvastatin as a less potent but for the treatment of human cholestatic liver diseases, clinically more relevant PXR agonist was fed as a 0.1% enriched standard diet (wt/wt) for 7 days 14 and compared with control mice fed a standard diet. Twenty-four hours after the last dose, the livers and kidneys were excised under general anesthesia (400 mg avertin/kg body weight intraperitoneally). Aliquots were frozen in liquid nitrogen for further total RNA and protein preparation or were fixed in 4% neutral buffered formaldehyde solution and paraffin-embedded for light microscopy. Serum samples from each mouse were stored at 70 C. Body weight and liver weight were obtained at the time of death. Experiments and Protocols for Common Bile Duct Ligated Animals. To further study the functional consequences of potential PXR and CAR effects on transporter and enzyme expression, mice were pretreated with specific and potent CAR (TCPOBOP) and PXR (PCN) agonists and their respective vehicles for 3 days as described above. Twenty-four hours after the last dose, common bile duct ligation (CBDL) was performed for another 24 hours as previously described, 15 and animals were placed into metabolic cages to collect urine for this period. In addition, the effects of atorvastatin were studied after pretreatment for 7 days, followed by 3-day CBDL with continuation of atorvastatin treatment. The livers were excised under general anesthesia with avertin (400 mg/kg body weight intraperitoneally) 24 hours after surgery (5 to 6 animals were studied in each group). Aliquots of liver tissue were fixed in 4% neutral buffered formaldehyde solution and paraffin-embedded for light microscopy. Serum and urine samples from each mouse were stored at 70 C. Serum Alanine Aminotransferase and Total Bilirubin Measurements. Alanine aminotransferase and bilirubin were analyzed with a Hitachi 917 analyzer (Boehringer Mannheim, Mannheim, Germany). Bile Acid Analysis. Total serum bile acids were estimated using a commercially 3 -hydroxysteroid dehydrogenase assay (Ecoline S ; DiaSys, Holzheim, Germany). For a detailed analysis, bile acids were extracted from serum ( 0.1 ml/animal) and also of urine (2-5 ml/animal/24 hours) using solid-phase extraction as previously described, 16 in the case of serum after disrupting protein adsorption by incubating with 1 ml 0.5-mol/L triethylammoniumsulfate at 64 C

3 422 WAGNER ET AL. HEPATOLOGY, August 2005 Table 1. Mouse Primer and Probe Sets Gene Forward Primer Reverse Primer Probe Accession No. 28sRNA cggctcttcctatcattgtg cctgtctcacgacggtctaa caagcgttggattgttcaccca X00525 Abcg5 caactccttcaggaccccaag aggctggtggatggtgacaat aaccacaggactggactgcatgactgc NM_ Abcg8 gacagcttcacagcccacaa gcctgaagatgtcagagcga caggctggtgctcatctccctcc NM_ Aox gggagtgctacgggttacatg ccgatatccccaacagtgatg * AF Asbt ggaactggctccaatatcctg gttcccgagtcaacccacat ctgaggtccatgtcgccatctatccaat NM_ Bsep gagtggtggacagaagcaaa tgaggtagccatgtccagaa cgcgccctcatacggaaacc AF CAR ggaggaccagatctcccttc atttcattgccactcccaag NM_ Cyp2b10 caatggggaacgttggaaga tgatgcactggaagaggaac ttcgtagattctctctggccaccatgaga NM_ Cyp3a11 ccaccagtagcacactttcc ttccatctccatcacagtatca ctctgcccaacaaggcacctcc NM_ FXR ggcctctgggtaccactaca aagaaacatggcctccactg NM_ Mdr2 atcctatgcactggccttctggt gaaagcatcaatacagggggcag cttcttctcaatcctcatcggggctttcagtgt NM_ Mrp2 gcttcccatggtgatctctt atcatcgcttcccaggtact cagtcatccaggccagcgtctc AF Mrp4 ttagatgggcctctggttct gcccacaattccaaccttt actgcgctcatcaagtccaggg W54702 Ntcp caccatggagttcagcaaga agcactgaggggcatgatac aggctcacttctggaagcccaaa U95131 Oatp1 gtcttacgagtgtgctccagat ggaatactgcctctgaagtggatt aatggatttgccagtacatttaccttcttgccc NM_ Oatp2 gacggctcagtgttcattc cttctagctggtccctctt aacgagccatagagcacatgactattaaca AB Oatp4 gatccttcacttacctgttcaa cctaaaaacattccacttgccata agacagcatcgcaggccaactttctg NM_ PXR cccatcaacgtagaggagga gggggttggtagttccagat AF Sult2a1 ggaaggaccacgactcataac gattcttcacaaggtttgtgttacc cccatccatctcttctccaagtctttcttcag L02335 Ugt1a1 tctgagccctgcatctatctg ccccagaggcgttgacata tggtataaattgccttcagaaaaagcccctatc L02333 NOTE. Fluorogenic probes for TaqMan RT-PCR are 5 FAM- and 3 TAMRA-labeled. *Real-time polymerase chain reaction using the Sybr Green assay was performed. Competitive reverse-transcriptase polymerase chain reaction was performed as previously described. 17 for 30 minutes. Equipment and conditions used for electrospray mass spectrometry and sample purification via anion-exchange chromatography, hydrolysis via cholylglycine hydrolase, and conversion to methyl ester trimethylsilyl ether derivatives for gas chromatography-mass spectrometry were the same as previously described in detail for the quantification of bile acids in individual human serum and urine samples. 16 Liver Histology. For conventional light microscopy, formalin-fixed livers were embedded in paraffin, and 4- m thick sections were stained with hematoxylin-eosin. The sections were coded and examined by a pathologist (C.L.) who was unaware of the animals genotype and treatment. RNA Isolation, Reverse-Transcription, and Taq- Man Real-Time Polymerase Chain Reaction. Total hepatic and renal RNA was isolated and reverse-transcribed into complementary DNA as previously described. 17 Real-time PCR was performed as previously described 4 with the exception of acyl-coenzyme A oxidase-1, which was determined under the same conditions but using the Sybr Green PCR Master Mix (Applied Biosystems, Warrington, UK) assay with a subsequent melting curve analysis. Messenger RNA (mrna) levels of FXR, PXR, and CAR were determined using a protocol previously described. 7 Primers and probes are given in Table 1. Mrp3 and Cyp7a1 primers and probes were published elsewhere. 18,19 Preparation of Liver and Kidney and Analysis of Transporter Protein Levels via Western Blotting. Crude liver and kidney membranes were prepared as previously described. 4,7 Transporter protein levels were determined using polyclonal antibodies against Ntcp (dilution, 1:2,500; kindly provided by Dr. Bruno Stieger, Zurich, Switzerland), Bsep (dilution, 1:7,500; kindly provided by Dr. Renxue Wang, Vancouver, Canada), 20 Mrp2 (dilution, 1:1,000; kindly provided by Dr. Bruno Stieger), Mrp3 (dilution, 1:1,000; kindly provided by Dr. Dietrich Keppler, Heidelberg, Germany), and Mrp4 (dilution, 1:1,000; kindly provided by Dr. Dietrich Keppler) as previously described. 4,7 Blots were reprobed with an anti -actin antibody (dilution, 1:5,000; Sigma) to confirm the specificity of the observed changes in transporter protein levels. Statistical Analysis. Four to six animals of each group were studied in parallel. Data are reported as the mean SEM. For statistical analysis, the treatment group was compared with the respective vehicle group using the Student t test (Sigmastat statistic program; Jandel Scientific, San Rafael, CA). A P value of less than.05 was considered significant. Results To study the in vivo effects of CAR and PXR activation on a functionally linked set of genes involved in adaptive bile acid/bilirubin transport and metabolism, mice were treated with different potent receptor agonists and com-

4 HEPATOLOGY, Vol. 42, No. 2, 2005 WAGNER ET AL. 423 Table 2. Serum Biochemistry in Healthy (Noncholestatic) Animals Intervention Total Bilirubin (mg/dl) Serum Bile Acids ( mol/l) Alanine Aminotransferase (U/L) LW/100 g Body Weight Control DMSO * * 65 7* ND Corn oil * ND PB TCPOBOP PCN Atorvastatin Fig. 1. Hepatic phase I and II detoxification enzymes, hepatobiliary transport systems (phase III), and nuclear receptors investigated in this study. Bile acid uptake is mediated via the basolateral Na /taurochalate cotransporter Ntcp and the organic anion transporters Oatp1, Oatp2, and Oatp4. Canalicular bile acid export is mediated by Bsep and to a lesser extent by Mrp2. Mdr2 acts as a canalicular phospholipid flippase, while the two-half transporters Abcg5/8 are responsible for canalicular cholesterol export. Alternative bile acid export is mediated via basolateral Mrp3 and Mrp4. Hydroxylation of bile acids is mediated by the phase I detoxifying cytochrome p450 enzymes Cyp3a11 and potentially Cyp2b10; the phase II conjugation is mediated by Sult2a1 and Ugt1a1. Cyp7a1 represents the key enzyme for the metabolism of cholesterol to bile acids. Although FXR and PXR are normally located within the nucleus, CAR is translocated into the nucleus only upon ligand activation. BA, bile acid; GSH, glutathione; Chol, cholesterol; FXR, farnesoid X receptor; CAR, constitutive androstane receptor; PXR, pregnane X receptor; Bili, bilirubin; OA, organic anions; BA-OH, hydroxylated BA; Bili-Glc, glucuronidated bilirubin; BA-SO 3, sulfated BA; PL, phospholipids. NOTE. Mice were treated with CAR and PXR agonists and their respective vehicles as described in Materials and Methods. Values are expressed as the mean SEM. Control: standard chow-fed mice. Abbreviations: ND, not determined; LW, liver weight. *P.05, vehicle (DMSO, corn oil) versus control (Student t test). P.05, agonist (PB, TCPOBOP, atorvastatin, PCN) versus respective vehicle (Student t test). pared with vehicles as control groups. Phase I and II detoxification enzymes, transport (phase III) systems, and nuclear receptors investigated in this study are summarized in Fig. 1. No significant differences in gene expression levels between the different vehicle-treated groups (DMSO or corn oil) and naive controls were observed, except for an induction of Sult2a1 by DMSO compared with naive controls (8.9-fold; P.032) (not shown). Alanine aminotransferase levels in the DMSO vehicle group were slightly higher compared with untreated controls (65 7 U/L vs U/L, respectively; P.037) and also slightly elevated in the TCPOBOP group compared with its vehicle (corn oil) group (70 8 U/L vs U/L, respectively; P.021) (Table 2). No significant histological changes in the vehicle groups compared with controls were observed. However, PB-, PCN-, and atorvastatin-treated animals showed slightly enlarged hepatocytes with a fine granular eosinophilic plasma within the central zone of hepatic lobules, consistent with adaptive changes due to enzymatic induction. In TCPOBOP-treated animals, these morphological changes were more pronounced. In line with the livermitogenic activity of CAR and PXR ligands, 10,21 the liver weight versus body weight ratio increased in the rank order: control PCN atorvastatin TCPOBOP (Table 2). Effects of CAR and PXR Ligands on Hepatobiliary Transport Systems for Bile Acids, Bilirubin, Phospholipids, and Cholesterol. mrna levels of the main hepatic bile acid uptake (Ntcp) and export (Bsep) systems remained unchanged after treatment with any of the four agonists (Fig. 2A). However, Bsep protein levels were significantly (2.8-fold) induced by PB, effects that were not observed after administration of the more specific CAR ligand TCPOBOP (Fig. 2B). Mrp3 and Mrp4 mrna levels were upregulated 4.1- and 3.0-fold, respectively, by PB, while the more potent CAR ligand TCPOBOP significantly induced mrna levels of Mrp2, Mrp3, and Mrp4 3.4-, 5.0-, and 8.6-fold, respectively (Fig. 3A). Both CAR ligands also significantly induced Mrp2, Mrp3, and Mrp4 protein expression levels (Fig. 3B). In contrast to the broad effects of CAR agonists on Mrp transporter expression, the selective PXR ligand PCN induced only Mrp3 mrna (4.2-fold) and protein levels (1.7-fold) and atorvastatin significantly induced only Mrp3 mrna 2.6-fold (Fig. 3). Oatp2 mrna was significantly induced by PXR agonists PCN (2.2-fold) and atorvastatin (2.4-fold) and CAR agonists PB (2.4-fold) and TCPOBOP (2.0-fold) (Fig. 4A). However, other Oatp family members particularly Oatp1 and Oatp4 remained unaffected (Fig. 4A). The canalicular cholesterol half-transporters Abcg5/8 showed a trend toward higher levels after atorvastatin treatment but did not change significantly with any of the four agonists (Fig. 4B). Atorvastatin significantly induced mrna levels of the cholangiocellular bile acid reuptake system

5 424 WAGNER ET AL. HEPATOLOGY, August 2005 (Mrp2, Mrp4) induction has been observed in cholestatic rats and mice, 23,24 we next addressed the question of whether Mrp2-4 (highly induced by TCPOBOP in the liver) are also induced in the kidney by this compound. Although Mrp2 mrna and Mrp4 mrna were both induced 2.0- and 2.1-fold, respectively (Fig. 5A), no differences in protein levels were observed (Fig. 5B). In addition, Mrp3, which is not constitutively expressed in male kidneys, remained undetectable after TCPOBOP stimulation (data not shown). Effects of CAR and PXR Ligands on Phase I and Phase II Detoxification Enzymes. The classical phase I target gene of CAR activation, Cyp2b10, showed a robust induction by both CAR agonists in the rank order of potency (Fig. 6A). In addition, the potent PXR agonist PCN induced Cyp2b10. In line with this, the phase I target gene of PXR, Cyp3a11, was induced by PCN (3.7- Fig. 2. Effects of CAR (PB, TCPOBOP) and PXR (atorvastatin, PCN) ligands on Ntcp and Bsep expression. Total RNA and liver membranes from vehicles, PB-treated (black bars), TCPOBOP-treated (dark gray bars), atorvastatin-treated (white bars), and PCN-treated (light gray bars) animals were isolated as described in Materials and Methods and analyzed via (A) reverse-transcriptase polymerase chain reaction and (B) Western blotting. (A) Ntcp and Bsep steady-state mrna levels are unchanged to corresponding vehicle-treated controls (dashed line) following treatment with all four agonists. Values are expressed as the percentage of respective vehicle controls (100%, represented as dashed line) SEM (n 4 to 6 in each group). (B) Ntcp protein amounts are unchanged after treatment with all four agonists. Bsep protein is significantly increased after PB treatment but is unchanged after TCPOBOP administration. -Actin for confirmation of the specificity of observed changes in transporter protein levels is unchanged. Data are expressed as the fold change compared with the respective vehicle (n 4to6in each group). *Respective vehicle versus treatment (P.05). mrna, messenger RNA; PB, phenobarbital; TCPOBOP, 1,4-bis-[2-(3,5-dichlorpyridyloxy)]benzene; PCN, pregnenolone-16 -carbonitrile. Asbt (3.4-fold) as well as Mdr2 (2.3-fold) (Fig. 4B). Because statins were reported to activate peroxisome proliferator-activated receptor (PPAR- ), 22 we also determined mrna levels of the classical PPAR- target gene acyl-coenzyme A oxidase-1. Only atorvastatin was able to significantly induce acyl-coenzyme A oxidase-1 (1.9-fold), while treatment with PB, TCPOBOP, and PCN did not show any induction (data not shown). Taken together, these findings indicate that CAR and to a lesser extent PXR agonists stimulate primarily the expression of hepatic export systems for bile acids and bilirubin (i.e., Mrp2-4 and Oatp2, respectively) at the mrna and protein levels and that some transporter effects of atorvastatin (e.g., Asbt, Mdr2) might be derived from the capacity to activate PPAR-. Because active transport might also play a role in renal bile acid elimination, and some adaptive renal transporter Fig. 3. Effects of CAR (PB, TCPOBOP) and PXR (atorvastatin, PCN) ligands on Mrp expression. Total RNA and liver membranes from vehicles, PB-treated (black bars), TCPOBOP-treated (dark gray bars), atorvastatintreated (white bars), and PCN-treated (light gray bars) animals were isolated as described in Materials and Methods and analyzed via (A) reverse-transcriptase polymerase chain reaction and (B) Western blotting. (A) Mrp2 mrna levels are significantly induced by TCPOBOP, Mrp3 mrna levels are significantly induced by all four agonists, and Mrp4 mrna levels are significantly induced by both CAR agonists (PB and TCPOBOP). Values are expressed as the percentage of respective vehicle controls (100%, represented as dashed line) SEM (n 4 to 6 in each group). *Respective vehicle versus treatment (P.05). (B) Mrp2-4 protein is significantly induced by both CAR agonists (PB and TCPOBOP). In addition, Mrp3 protein is also significantly increased by the PXR ligand PCN. Data are expressed as the fold change compared with the respective vehicle (n 4 to 6 in each group). mrna, messenger RNA; PB, phenobarbital; TCPOBOP, 1,4-bis-[2-(3,5-dichlorpyridyloxy)]benzene; PCN, pregnenolone-16 -carbonitrile.

6 HEPATOLOGY, Vol. 42, No. 2, 2005 WAGNER ET AL. 425 Fig. 4. Effects of CAR (PB, TCPOBOP) and PXR (atorvastatin, PCN) ligands on Oatp, Asbt, and lipid transporter expression. Total RNA from vehicles, PB-treated (black bars), TCPOBOP-treated (dark gray bars), atorvastatin-treated (white bars) and PCN-treated (light gray bars) animals were isolated as described in Materials and Methods and analyzed via reverse-transcriptase polymerase chain reaction for (A) mrna levels of Oatps and (B) lipid transporters, including bile acid transporting Asbt. (A) Oatp2 mrna levels are significantly induced by both CAR (PB, TCPOBOP) and both PXR (atorvastatin, PCN) agonists, while Oatp1 and Oatp4 mrna is unchanged. (B) Only atorvastatin significantly induces Asbt and Mdr2 mrna levels, while the other agonists have no impact. Abcg5/8 are not significantly induced following treatment with any of the four agonists. Values are expressed as the percentage of respective vehicle controls (100%, represented as dashed line) SEM (n 4to 6 in each group). *Respective vehicle versus treatment (P.05). mrna, messenger RNA; PB, phenobarbital; TCPOBOP, 1,4-bis-[2-(3,5- dichlorpyridyloxy)]benzene; PCN, pregnenolone-16 -carbonitrile. fold) and to a lesser extent by the CAR agonists PB (2.8- fold) and TCPOBOP (3.1-fold). Atorvastatin did not significantly induce Cyp3a11 (Fig. 6A). The phase II target gene, bile acid sulfotransferase Sult2a1, was induced by all four agonists but was statistically significant only after TCPOBOP administration (28-fold) (Fig. 6A). The bilirubin glucuronosyltransferase Ugt1a1 was significantly upregulated by both CAR agonists (2.7- and 3.1- fold by PB and TCPOBOP, respectively) and PCN (1.5- fold) (Fig. 6A). The key enzyme of bile acid synthesis, Cyp7a1, was reduced by PCN to 74% (without reaching statistical significance) and TCPOBOP (38%; P.05) (not shown). Taken together, these findings indicate that phase I hydroxylation is induced by both CAR and PXR agonists, while phase II conjugation mainly represents a target for CAR agonists. Effects of CAR and PXR Ligands on Expression of Nuclear Receptors. To further test whether CAR and PXR agonists may also regulate their own expression or that of the classical nuclear bile acid receptor FXR, mrna levels of FXR, PXR, and CAR were also determined. We found a significant induction of PXR mrna levels (1.7-fold) after administration of TCPOBOP, while mrna levels of CAR and FXR remained unchanged after treatment with any of the four agonists (Fig. 6B). Effects of CAR and PXR Ligands on Serum Bile Acid and Bilirubin Levels and Bile Acid Composition. To determine whether the observed mrna and protein changes were accompanied by relevant functional changes, we measured serum bilirubin as well as serum and urine bile acids in healthy and cholestatic mice pretreated with the agonists prior to CBDL as described in Materials and Methods. Both CAR ligands (PB and TCPOBOP) significantly decreased serum bilirubin levels already in healthy, noncholestatic animals. Moreover, PB significantly decreased serum bile acid levels in these mice, while TCPOBOP-treated animals showed a trend toward lower bile acid levels. PCN and atorvastatin treatment did not affect these parameters in healthy mice with normal bilirubin and bile acid levels (Table 2). In CBDL mice, TCPOBOP significantly reduced serum bilirubin and bile acid levels and increased the excretion of bile acids in urine (Table 3). Simultaneously, there was a shift Fig. 5. Effects of the CAR ligand TCPOBOP on renal Mrp2 and Mrp4 expression. Total RNA and kidney membranes from vehicles (black bars) and TCPOBOP-treated (gray bars) animals were isolated as described in Materials and Methods and analyzed via (A) reverse-transcriptase polymerase chain reaction and (B) Western blotting. (A) Renal Mrp2 (without becoming statistically significant) and Mrp4 mrna levels (statistically significant) are induced by TCPOBOP. Values are expressed as the percentage of respective vehicle controls (100%) SEM (n 5 in each group). *Respective vehicle versus treatment (P.05). (B) Renal Mrp2, Mrp4, and -actin protein is unchanged by TCPOBOP. Data are expressed as the fold change compared with the respective vehicle (n 4 to 5 in each group). mrna, messenger RNA; TCPOBOP, 1,4-bis-[2- (3,5-dichlorpyridyloxy)]benzene.

7 426 WAGNER ET AL. HEPATOLOGY, August 2005 Fig. 6. Effects of CAR (PB, TCPOBOP) and PXR (atorvastatin, PCN) ligands on phase I and phase II and on nuclear receptor expression. Total RNA from vehicles, PB-treated (black bars), TCPOBOP-treated (dark gray bars), atorvastatin-treated (white bars), and PCN-treated (light gray bars) animals were isolated as described in Materials and Methods and analyzed via reverse-transcriptase polymerase chain reaction for mrna levels of (A) phase I and phase II and (B) nuclear receptors. (A) Phase I detoxifying enzymes Cyp2b10 and Cyp3a11 are significantly induced by both CAR agonists, PB and TCPOBOP, and the potent PXR agonist, PCN, while the effects of atorvastatin on their expression levels are not statistically significant. Among the phase II detoxifying enzymes, Sult2a1 is significantly stimulated by the potent CAR agonist TCPOBOP, and Ugt1a1 is significantly stimulated by both CAR agonists (PB, TCPOBOP) and PCN. Note the logarithmic scale. (B) The CAR agonist TCPOBOP significantly stimulates expression of the nuclear receptor PXR. FXR and CAR remain unchanged compared with vehicle. Values are expressed as the percentage of respective vehicle controls (100%, represented as dashed line) SEM (n 4 to 6 in each group). *Respective vehicle versus treatment (P.05). mrna, messenger RNA; PB, phenobarbital; TCPOBOP, 1,4-bis-[2-(3,5-dichlorpyridyloxy)]benzene; PCN, pregnenolone-16 -carbonitrile; FXR, farnesoid X receptor; PXR, pregnane X receptor; CAR, constitutive androstane receptor. from trihydroxylated primary bile acids (triols) to less toxic pentahydroxylated bile acids (pentols) in serum and urine (Table 4). PCN significantly reduced serum bile acids, showed a trend toward higher urinary bile acid excretion, and again showed a shift from tetrols to pentols in serum and urine (Tables 3 and 4). Atorvastatin also significantly reduced serum bile acids from 1, mol/l in CBDL controls to mol/l (Table 3). In contrast to noncholestatic animals, alanine aminotransferase values significantly increased, despite improvement of cholestasis, which most likely reflects hepatotoxicity of the CAR and PXR ligands tested under cholestatic conditions (Table 3). Taken together, these results clearly demonstrate the functional impact of CAR and PXR stimulation on the metabolism and urinary elimination of bile acids and bilirubin in health and disease as predicted by the expression changes. Discussion Recent reports have revealed that the xenobiotic receptors CAR and PXR also critically regulate metabolic pathways for endobiotics, including bile acids and bilirubin, which in turn can bind as activating ligands to these nuclear receptors. 25 Thus, activation of nuclear receptors by accumulating cholephiles could explain adaptive changes in hepatobiliary transporter and metabolizing enzyme expression in cholestasis. 4,7,9,26 This intrinsic induction, however, is ineffective in fully preventing cholestatic liver injury. 4,7 In contrast to FXR, expression of CAR and PXR is preserved during noninflammatory cholestasis, 9,27 thus providing a good rationale for pharmacological interference with bile acid/bilirubin transport and metabolism via these nuclear receptors. We herein tested the effects of less potent CAR/PXR receptor ligands (phenobarbital/ atorvastatin), which are also agonists for the corresponding human receptors, as well as of more potent receptor ligands specifically activating murine/rodent nuclear receptors (TCPOBOP/PCN) on expression levels of bile acid/bilirubin-metabolizing enzymes and transport systems. Furthermore, we investigated their functional impact on bile acid and bilirubin metabolism in healthy (noncholestatic) and cholestatic mice. The results indicate that CAR and to a lesser extent PXR are involved in the coordinated regulation of several steps of bile acid/ bilirubin detoxification and make both receptors potential targets for future therapeutic strategies. Coordinated Induction of Adaptive Export Systems and Alternative Detoxification Enzymes by CAR and PXR Ligands. The alternative hepatocellular efflux pumps Mrp3 and Mrp4 (phase III) were induced by both CAR (Mrp3 and Mrp4) and PXR agonists (Mrp3). Both transporters constitutively expressed at very low levels on the basolateral membrane of hepatocytes are upregulated under various cholestatic conditions, 4,26,28-30 implying a role as intrinsic escape routes for biliary constituents. Phase I bile acid hydroxylating enzymes Cyp2b10 and Cyp3a11 were also strongly induced by both CAR and PXR agonists. In cholestatic liver disease in humans, hydroxylation of bile acids is already spontaneously observed, which renders bile acids more hydrophilic and thus less toxic. 6,31 Functionally, CAR and PXR stimulation was followed by a shift from primary cholic acid and

8 HEPATOLOGY, Vol. 42, No. 2, 2005 WAGNER ET AL. 427 Table 3. Serum Biochemistry in CBDL Mice Pretreated With CAR and PXR Agonists Total Bilirubin (mg/dl) Serum Bile Acids ( mol/l) Urinary Bile Acids (mg/24 hr) Alanine Aminotransferase (U/L) Mortality 24-hr CBDL Control , ND 1, /6 DMSO , (n 3) 1, /6 Corn oil , (n 3) 1, /5 TCPOBOP * * * (n 3) 3, * 0/6 PCN * (n 3) 3,239 1,306* 2/7 3-d CBDL Control , ND /5 Atorvastatin * ND 1, * 0/5 NOTE. Mice were pretreated for 3 days with CAR (TCPOBOP) and PXR (PCN) agonists and their respective vehicles prior to 24-hour CBDL as described in Materials and Methods. Atorvastatin was prefed for 7 days followed by 3-day CBDL with continuation of atorvastatin feeding. Values are expressed as the mean SEM. Control: standard chow-fed mice. Abbreviation: ND, not determined. *P.05, agonist (TCPOBOP, PCN, atorvastatin) versus respective vehicle (Student t test). muricholic acids (triols) to polyhydroxylated bile acids (pentols) in serum in CBDL mice. In urine, the less toxic and better water-soluble tetrols dominated in all studied groups. Again in urine, a shift toward even more hydroxylated pentols was observed after TCPOBOP and PCN treatment. Polyhydroxylated bile acids are obviously more rapidly cleared from the circulation as reflected by reductions in serum bile acid levels. Phase II bile acid sulfating enzyme Sult2a1 showed a robust induction by the potent CAR ligand TCPOBOP. Of interest, sulfated bile acids are thought to be excellent substrates for CARinducible Mrp4. 32 However, sulfated bile acids were not found to increase under the studied conditions due to the fact that sulfation is not a common metabolic pathway to detoxify intrinsic bile acids in mice. Neither we nor others Table 4. Amount of Polyhydroxylated Bile Acids in Serum and Urine After CBDL Pretreated With CAR and PXR Agonists 24-hr CBDL Triols (%) Tetrols (%) Pentols (%) Serum Control Not detectable DMSO Not detectable Corn oil Not detectable TCPOBOP 53 2* 33 1* * PCN 74 1* * Urine Control ND ND ND DMSO Corn oil TCPOBOP 20 3* * PCN 22 1* * NOTE. Sulfated and glucuronidated bile acids were not found to increase under these conditions. Mice were pretreated for 3 days with CAR (TCPOBOP) and PXR (PCN) agonists and their respective vehicles prior to 24-hour CBDL as described in Materials and Methods. Values are expressed as the mean SEM. Control: standard chow-fed mice. Abbreviation: ND, not determined. *P.05, agonist (TCPOBOP, PCN) versus respective vehicle (Student t test). have found such conjugates in naive or cholestatic conditions. In mice, sulfation was described only for lithocholic acid (LCA) during LCA feeding. 33 Normally, LCA is virtually absent in mice. 4,23,34 However, in contrast to mice, sulfation of bile acids is a relevant metabolic pathway in humans reflected by increased amounts in serum and urine of patients with cholestasis. 5,6 Thus, a picture is emerging in which coordinated induction of hepatic bile acid hydroxylation (i.e., Cyp2b10 and Cyp3a11; phase I) and potentially sulfation (i.e., Sult2a1; phase II) together with overexpression of adaptive bile acid export systems (i.e., Mrp3 and Mrp4; phase III) may represent therapeutically inducible, alternative pathways. Moreover, additional stimulatory effects of CAR agonists (TCPOBOP) on the expression levels of the bile acid receptor PXR might amplify such alternative pathways. In line with the concept of a coordinated regulation of transport and metabolism, Ugt1a1 (selectively conjugating bilirubin) was also induced by both CAR agonists and to a lesser extent by the PXR ligand PCN, in concert with the induction of bilirubin conjugate transport systems Mrp2 and Mrp3. 35,36 This was accompanied by a reduction in serum bilirubin levels in cholestatic (CBDL) mice and even in healthy (noncholestatic) mice with normal bilirubin levels. However, induction of renal bile acid transporters Mrp2 and Mrp4, which is observed under cholestatic conditions, 23,24 was only moderate for mrna and absent for protein expression after TCPOBOP treatment. This may be explained by lower renal expression of CAR and PXR. The high amount of polyhydroxylated bile acids in urine during treatment with PXR and CAR agonists may be due to enhanced passive glomerular filtration of these compounds, because protein binding decreases with degree of hydroxylation. 37

9 428 WAGNER ET AL. HEPATOLOGY, August 2005 Modulation of Other Hepatobiliary Transport Systems by CAR and PXR Agonists. The organic anion transporter Oatp2 was induced by any CAR and PXR agonist tested. Although Oatp2 acts mainly as a bidirectional exchange system for various organic compounds, including bile acids, 38 it was also demonstrated that Oatp2 may be involved in the export of organic solutes, 39 thus potentially assisting alternative, basolateral bile acid efflux via Mrp3 and Mrp4. However, the more relevant bile acid transporters Oatp1 and Oatp4 were not affected by any of the administered agonists. In line with these findings, the major bile acid uptake system Ntcp, which is responsible for more than 80% of bile acid uptake, remained unchanged. Bsep was significantly induced only by PB on posttranscriptional levels. This difference in Bsep expression between both CAR agonists may be due to additional CARindependent effects of PB or PB metabolites. 40 The effects of PB on bile acid dependent bile flow (mediated via Bsep) are controversial In summary, these data suggest that orthograde transport systems (i.e., Oatp1, Oatp4, Ntcp, Bsep) are less affected, while alternative/ adaptive transporters (i.e., Oatp2, Mrp3, Mrp4) are selectively induced by pharmacological CAR and, to a lesser extent, PXR agonists. Of interest, phospholipid flippase Mdr2 was induced by atorvastatin, an effect not seen by the specific CAR and PXR agonists. PPAR- agonists were recently shown to stimulate Mdr2 overexpression, 44 suggesting that this finding could rather be related to the PPAR- activating capacity of atorvastatin. 22 Because an imbalance in the biliary phospholipid and bile acid ratio in Mdr2 knockout mice leads to a cholestatic phenotype resembling primary sclerosing cholangitis, 15,17 it is attractive to speculate that induction of Mdr2 and biliary phospholipid secretion by statins might protect bile ducts against toxic bile acids In addition, Asbt was only induced by atorvastatin, which analogous to human ASBT could be related to PPAR- stimulation. 45 Induced cholangiocellular Asbt could favor reabsorption of toxic bile acids from stagnant bile in (obstructed) bile ducts. Some of the effects of atorvastatin may also involve the SREBP-2 (sterol regulatory element-binding protein-2) system, as hypothesized for Mdr2. 14 These findings, together with the marginal Cyp3a11 induction and the significant induction of the prototypic PPAR- target gene acyl-coenzyme A oxidase-1, in fact indicate a more relevant role for atorvastatin as a PPAR- agonist in rodents. However, because human PXR is activated by statins to a higher degree than murine PXR, 46 the PXR effects of statins may be more pronounced in humans. Could Combination Therapy With CAR and PXR Agonists Represent a Future Option for the Treatment of Cholestasis? PXR and CAR agonists have already been used in the pre ursodeoxycholic acid area for the treatment of cholestatic disorders and its symptoms (e.g., pruritus) before knowing their exact mode of action. As such, PB has been used in several cholestatic disorders, predominantly for the treatment of pruritus. 8 In primary biliary cirrhosis, PB reduces pruritus without major effects on biochemical markers of cholestasis. 8 Other studies using PB showed partial improvement of cholestasis in only some patients. 8 Rifampicin (which now is known to be a human PXR agonist) was also mainly used for the treatment of pruritus, but in most cases effects on cholestatic parameters in primary biliary cirrhosis again were only minor. 8 However, data obtained with the more specific receptor agonists in rodents should not be easily extrapolated to the human situation, because there may be considerable differences in the ligand specificity among different species. 46,47 Therefore, nuclear receptor based chemical libraries are used to discover novel selective nuclear receptor activating ligands. 48 Fig. 7. Schematic summary of the effects of CAR and PXR agonists. CAR and PXR agonists do not affect major basolateral uptake (i.e., Ntcp, Oatp1, Oatp4) and canalicular export (i.e., Bsep) systems for bile acid. CAR agonists selectively stimulate bile acid/bilirubin phase I hydroxylation (i.e., Cyp3a11, Cyp2b10), phase II conjugation (i.e., Sult2a1, Ugt1a1), and coordinated alternative phase III export for bile acids (i.e., Mrp3, Mrp4, potentially Oatp2) and bilirubin (i.e., Mrp2). In addition, CAR agonists induce expression of PXR and reduce Cyp7a1 expression. PXR agonists stimulate a smaller set of phase I/II enzymes (i.e., Cyp3a11, Cyp2b10, Ugt1a1) and phase III transport systems (i.e., Mrp3, Oatp2). BA, bile acid; GSH, glutathione; Chol, cholesterol; FXR, farnesoid X receptor; CAR, constitutive androstane receptor; PXR, pregnane X receptor; Bili, bilirubin; OA, organic anions; BA-OH, hydroxylated BA; Bili-Glc, glucuronidated bilirubin; BA-SO 3, sulfated BA; PL, phospholipids.

10 HEPATOLOGY, Vol. 42, No. 2, 2005 WAGNER ET AL. 429 However, nuclear receptor agonists can also have hepatotoxic side effects in cholestasis. In line with clinical reports on rifampicin-induced hepatotoxicity in patients with primary biliary cirrhosis, 49 CAR and PXR agonists in cholestatic (CBDL) mice resulted in increased serum aminotransferase levels despite an improvement of cholestasis in the present study. Although allowing a proof of principle in the present study, CBDL may not represent the ideal model to test pharmacological strategies for the treatment of cholestasis, and further long-term studies in more suitable mouse models of chronic cholestasis (e.g., an Mdr2 knockout mouse model of sclerosing cholangitis) are needed. 15,17 In conclusion, additional stimulation of PXR and CAR in cholestasis with potent and specific ligands should counteract accumulating bile acids/bilirubin via amplification of partly existing adaptive pathways and recruitment of new adaptive pathways of bile acid hydroxylation (via Cyp3a11 and possibly Cyp2b10); sulfation (via Sult2a1); bilirubin conjugation (via Ugt1a1); and facilitation of rapid alternative excretion of bile acid and bilirubin conjugates (via Mrp2-4) (Fig. 7). Acknowledgment: We gratefully acknowledge Dr. W. Erwa (Graz, Austria) and colleagues for performing liver function tests. The excellent technical assistance of Dagmar Silbert, Judith Gumhold, and Andrea Fuchsbichler (Graz, Austria) is also gratefully acknowledged. References 1. Trauner M, Meier PJ, Boyer JL. Molecular pathogenesis of cholestasis. N Engl J Med 1998;339: Lee J, Boyer JL. Molecular alterations in hepatocyte transport mechanisms in acquired cholestatic liver disorders. Semin Liver Dis 2000;20: Zollner G, Fickert P, Silbert D, Fuchsbichler A, Marschall HU, Zatloukal K, et al. Adaptive changes in hepatobiliary transporter expression in primary biliary cirrhosis. J Hepatol 2003;38: Wagner M, Fickert P, Zollner G, Fuchsbichler A, Silbert D, Tsybrovskyy O, et al. Role of farnesoid X receptor in determining hepatic ABC transporter expression and liver injury in bile duct-ligated mice. Gastroenterology 2003;125: van Berge Henegouwen GP, Brandt KH, Eyssen H, Parmentier G. Sulphated and unsulphated bile acids in serum, bile, and urine of patients with cholestasis. Gut 1976;17: Bremmelgaard A, Sjovall J. Bile acid profiles in urine of patients with liver diseases. Eur J Clin Invest 1979;9: Fickert P, Zollner G, Fuchsbichler A, Stumptner C, Pojer C, Zenz R, et al. Effects of ursodeoxycholic and cholic acid feeding on hepatocellular transporter expression in mouse liver. Gastroenterology 2001;121: Trauner M, Wagner M, Fickert P, Zollner G. Molecular regulation of hepatobiliary transport systems: clinical implications for understanding and treating cholestasis. J Clin Gastroenterol 2005;39:S111 S Guo GL, Lambert G, Negishi M, Ward JM, Brewer HB Jr, Kliewer SA, et al. Complementary roles of farnesoid X receptor, pregnane X receptor, and constitutive androstane receptor in protection against bile acid toxicity. J Biol Chem 2003;278: Guo GL, Johnson DR, Klaassen CD. Postnatal expression and induction by pregnenolone-16alpha-carbonitrile of the organic anion-transporting polypeptide 2 in rat liver. Drug Metab Dispos 2002;30: Ritzel U, Leonhardt U, Nather M, Schafer G, Armstrong VW, Ramadori G. Simvastatin in primary biliary cirrhosis: effects on serum lipids and distinct disease markers. J Hepatol 2002;36: Kamisako T, Adachi Y. Marked improvement in cholestasis and hypercholesterolemia with simvastatin in a patient with primary biliary cirrhosis. Am J Gastroenterol 1995;90: Kurihara T, Akimoto M, Abe K, Ishiguro H, Niimi A, Maeda A, et al. Experimental use of pravastatin in patients with primary biliary cirrhosis associated with hypercholesterolemia. Clin Ther 1993;15: Hooiveld GJ, Vos TA, Scheffer GL, Van GH, Koning H, Bloks V, et al. 3-Hydroxy-3-methylglutaryl-coenzyme A reductase inhibitors (statins) induce hepatic expression of the phospholipid translocase mdr2 in rats. Gastroenterology 1999;117: Fickert P, Zollner G, Fuchsbichler A, Stumptner C, Weiglein AH, Lammert F, et al. Ursodeoxycholic acid aggravates bile infarcts in bile ductligated and Mdr2 knockout mice via disruption of cholangioles. Gastroenterology 2002;123: Marschall HU, Broome U, Einarsson C, Alvelius G, Thomas HG, Matern S. Isoursodeoxycholic acid: metabolism and therapeutic effects in primary biliary cirrhosis. J Lipid Res 2001;42: Fickert P, Fuchsbichler A, Wagner M, Zollner G, Kaser A, Tilg H, et al. Regurgitation of bile acids from leaky bile ducts causes sclerosing cholangitis in Mdr2 (Abcb4) knockout mice. Gastroenterology 2004;127: Bohan A, Chen WS, Denson LA, Held MA, Boyer JL. Tumor necrosis factor alpha-dependent up-regulation of Lrh-1 and Mrp3(Abcc3) reduces liver injury in obstructive cholestasis. J Biol Chem 2003;278: Plosch T, Kok T, Bloks VW, Smit MJ, Havinga R, Chimini G, et al. Increased hepatobiliary and fecal cholesterol excretion upon activation of the liver X receptor is independent of ABCA1. J Biol Chem 2002;277: Wang R, Salem M, Yousef IM, Tuchweber B, Lam P, Childs SJ, et al. Targeted inactivation of sister of P-glycoprotein gene (spgp) in mice results in nonprogressive but persistent intrahepatic cholestasis. Proc Natl Acad SciUSA2001;98: Columbano A, Ledda-Columbano GM, Pibiri M, Concas D, Reddy JK, Rao MS. Peroxisome proliferator-activated receptor-alpha mice show enhanced hepatocyte proliferation in response to the hepatomitogen 1,4-bis [2-(3,5-dichloropyridyloxy)] benzene, a ligand of constitutive androstane receptor. HEPATOLOGY 2001;34: Landrier JF, Thomas C, Grober J, Duez H, Percevault F, Souidi M, et al. Statin induction of liver fatty acid-binding protein (L-FABP) gene expression is peroxisome proliferator-activated receptor-alpha-dependent. J Biol Chem 2004;279: Lee J, Azzaroli F, Wang L, Soroka CJ, Gigliozzi A, Setchell KD, et al. Adaptive regulation of bile salt transporters in kidney and liver in obstructive cholestasis in the rat. Gastroenterology 2001;121: Wagner M, Fickert P, Marschall HU, Zollner G, Silbert D, Fuchsbichler A, et al. Alternative exretory routes for accumulating bile acids via adaptive renal ABC transporter expression in common bile duct-ligated (CBDL) mice [Abstract]. Gastroenterology 2004;126(Suppl 2):A Trauner M, Boyer JL. Bile salt transporters: molecular characterization, function, and regulation. Physiol Rev 2003;83: Zollner G, Fickert P, Fuchsbichler A, Silbert D, Wagner M, Arbeiter S, et al. Role of nuclear bile acid receptor, FXR, in adaptive ABC transporter regulation by cholic and ursodeoxycholic acid in mouse liver, kidney and intestine. J Hepatol 2003;39: Denson LA, Bohan A, Bajwa HJ, Held MA, Boyer JL. Alterations in nuclear hormone receptors (NHR) are associated with changes in the expression of multidrug resistance proteins Mrp2 and Mrp3 in liver and kidney in cholestasis [Abstract]. HEPATOLOGY 2001;34(Suppl 2): Soroka CJ, Lee JM, Azzaroli F, Boyer JL. Cellular localization and upregulation of multidrug resistance-associated protein 3 in hepatocytes and cholangiocytes during obstructive cholestasis in rat liver. HEPATOLOGY 2001;33: