JBC Origene Your Gene Company

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M411520200 on April 13, 2005 Originally published In Press as doi:10.1074/jbc.M411520200 on April 11, 2005

J. Biol. Chem., Vol. 280, Issue 24, 23232-23242, June 17, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
280/24/23232    most recent
M411520200v2
M411520200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McCarthy, T. C.
Right arrow Articles by Sinal, C. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McCarthy, T. C.
Right arrow Articles by Sinal, C. J.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Vitamin D Receptor-dependent Regulation of Colon Multidrug Resistance-associated Protein 3 Gene Expression by Bile Acids*

Tanya C. McCarthy, Xiufeng Li, and Christopher J. Sinal{ddagger}

From the Department of Pharmacology, Dalhousie University, Halifax, Nova Scotia B3H 1X5, Canada

Received for publication, October 8, 2004 , and in revised form, March 10, 2005.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The multidrug resistance-associated protein 3 (MRP3) is a multispecific anion transporter that is capable of transporting a number of conjugated and unconjugated bile acids. Expression of the MRP3 gene is increased during pathological states associated with elevated bile acid concentrations indicating a role for this transporter in adaptive and homeostatic bile acid metabolism. Analysis of Mrp3 mRNA levels in various mouse tissues with known relevance and/or exposure to bile acids revealed the highest levels of basal expression in the colon followed in order by the liver, duodenum, jejunum, ileum, and kidney. Functional analysis of a murine Mrp3 promoter reporter construct revealed vitamin D receptor (VDR)-dependent activation by 1,25-dihydroxyvitamin D3 (VD3), 9-cis-retinoic acid (RA), and the cholestatic secondary bile acid, lithocholic acid (LCA). Using a series of deletion constructs combined with sequence analysis, a candidate VDR response element (VDRE) was identified between -1028 and -1014 bp of the Mrp3 promoter. Activation of the Mrp3 promoter in response to VD3, RA, or LCA, as well as binding of VDR/RXR heterodimers, was attenuated substantially by mutation of this VDRE. Treatment of mice with VD3 or LCA demonstrated in vivo modulation of the Mrp3 gene in colon but not in the liver. Reduction of endogenous VDR expression in colon adenocarcinoma MCA-38 cells by siRNA transfection was associated with reduced constitutive and inducible expression of the Mrp3 gene. These data support a regulatory role for the VDR in the protection of colon cells from bile acid toxicity through regulation of the Mrp3 expression.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The ATP-binding cassette (ABC)1 transporter superfamily is comprised of a number of active transport proteins, which function in the translocation of a variety of molecules across biological membranes (1–4). Variations in the structure of certain ABC transporter genes are the underlying cause, or contribute to a number of human diseases including cystic fibrosis, Dubin-Johnson syndrome, Tangier disease, and progressive familial intrahepatic cholestasis (2, 5–7). Whereas some ABC transporters exhibit a high degree of substrate selectivity and have clear physiological roles, many others transport a diverse range of exogenous and endogenous substrates and have less well defined physiological functions. The overexpression of ABC transporters with wide substrate selectivity can result in an increased efflux of chemotherapeutic agents from within cells and is one of the primary causes of multidrug resistance in cancer cell lines and tumors (1, 8). The multidrug resistance-associated protein 3 (Mrp3), also known as Abcc3, is capable of transporting a wide range of substrates including anti-cancer drugs, organic anions, as well as glucuronide and sulfate conjugates of a number of endogenous and exogenous compounds (9–11). The physiological function of this transporter remains largely uncharacterized, however, the ability of Mrp3 to transport bile acids and bile acid conjugates suggests a role in normal or adaptive bile acid homeostasis (10, 11).

The hepatic biosynthesis of bile acids from cholesterol involves a cascade of sequential enzymatic reactions that occur in multiple subcellular compartments. The primary bile acids cholic acid (CA) and chenodeoxycholic acid (CDCA) are the major bile acids synthesized in human liver (12). Bile acids are conjugated, primarily with taurine and glycine, and are subsequently transported across the hepatocyte canalicular membrane by the bile salt export pump, a member of the ABC family (Abcb11) (13). After secretion into the intestine, bile acids function as biological detergents and facilitate the emulsification and absorption of dietary lipids and lipid-soluble vitamins. Approximately 95% of secreted bile acids are reabsorbed from the intestine and returned to the liver via the enterohepatic circulation. Bile acids can also be biotransformed by intestinal bacteria by the processes of deconjugation and reduction to form free bile acids and secondary bile acids, respectively. Deconjugated bile acids can be reabsorbed, returned to the liver, and reconjugated for utilization. Secondary bile acids are also reabsorbed and reconjugated to some extent, however, the concentrations of these hepatotoxic bile acids, particularly lithocholic acid (LCA), must be limited because of the highly cholestatic nature of these compounds.

Expression of MRP3/Mrp3 is induced under pathological conditions associated with elevated bile acid concentrations including intrahepatic cholestasis, Dubin-Johnson syndrome as well as a number of animal models of cholestasis (14–17). The ability of Mrp3 to transport free and sulfate conjugates of LCA is consistent with a protective role for this protein in the detoxification of cholestatic bile acids (10). Expression of human MRP3 was reported to be up-regulated by a number of bile acids in the human enterocyte Caco-2 cells (18). In this model, bile acids were found to increase the expression of the liver receptor homologue-1 transcription factor, which in turn, activated transcription of the MRP3 promoter (18). More recently, rat hepatic Mrp3 induction in response to bile duct ligation was both found to be dependent upon the activation of tumor necrosis factor-{alpha} signaling pathways (19). Together, these data provide an indirect mechanism by which inflammatory responses provoked by bile acid accumulation lead to increased MRP3/Mrp3 gene expression. Whereas this model provides a plausible mechanism by which bile acids can activate MRP3 expression, a complementary direct mechanism for MRP3 modulation by bile acids may exist. Bile acids are ligand activators of a number of nuclear hormone receptors (NHRs), including the farnesoid X receptor (FXR), pregnane X receptor (PXR), and the VDR (20–25). Through the direct regulation of genes encoding proteins involved in bile acid metabolism and transport, these NHRs have complementary and overlapping roles in the regulation of bile acid homeostasis and protection against bile acid toxicity (26, 27). Thus, we hypothesized that bile acid-activated NHRs could directly regulate expression of the Mrp3 gene. To test this hypothesis, the murine Mrp3 promoter was cloned, characterized, and tested in a series of transfection assays for regulation by NHRs known to be activated by bile acids. The results of this study identify VDR as a regulator of basal and inducible Mrp3 expression in the colon.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Animals and Treatments—Male C57BL/6J mice were obtained from our breeding colony and were housed under a 12-h light/dark cycle. Mice were allowed water and standard rodent chow ad libitum prior to, and during treatment. All experiments were performed with male mice weighing 25–30 g at 9–11 weeks of age. Treatment consisted of daily gavage of 2 µg of 1{alpha},25-(OH)2VD3 (Sigma) or 10 mg of LCA (Sigma) suspended in corn oil (0.1 ml) for a total of 4 days. Mice were anesthetized with an intraperitoneal injection of 80 mg/kg sodium pentobarbital (CMTC Pharmaceuticals, Cambridge, Ontario, Canada) 24 h after the last treatment. Tissues were removed rapidly and snap frozen in liquid nitrogen and stored at -80 °C prior to RNA extraction. All protocols and procedures were approved by the Dalhousie University Committee on Laboratory Animals and are in accordance with the Canadian Council on Animal Care guidelines.

Rapid Amplification of cDNA Ends (RACE) and Cloning of Full-length Murine Mrp3 cDNA—Mouse liver and colon total RNA was prepared using TRIzol reagent as per the supplier's (Invitrogen, Burlington, ON, Canada) instructions. 5'-RACE was performed using the SMART RACE cDNA Amplification Kit as per the supplier's (Clontech, Palo Alto, CA) instructions. Total RNA (1 µg) from liver or colon was used as the starting template and Race-R3, an oligonucleotide (all oligonucleotides purchased from Sigma) of the sequence 5'-GTGGAGCTCAGGAACATCAGTCCGGC-3' was used as the gene-specific primer. Analysis of the PCR on a 1% TAE-agarose gel revealed a single band of ~1150 bp. The PCR product was cloned into the pCRII vector using a TOPO-TA Cloning Kit as per the supplier's (Invitrogen) instructions. Plasmid DNA was prepared from 16 positive colonies (8 liver, 8 colon) using a Qiaprep Spin Miniprep kit (Qiagen, Mississauga, ON, Canada) and sequenced by the Center for Functional Microbial Genomics and Host Defense DNA Sequencing Facility (Halifax, NS, Canada). The longest clone was used to design an upstream primer (MRP3-F1) corresponding to the extreme 5' end of the Mrp3 cDNA (5'-AGATCTGCTGGGGCTGAGCTGAACTGA-3'). This primer was then used in a 3'-RACE reaction as per the supplier's (Clontech) instructions. The PCR product was cloned into the pCRII vector and positive clones were selected and sequenced. The clones exhibited >99% sequence homology and a consensus full-length Mrp3 mRNA sequence was derived from eight individual clones.

Mrp3 Promoter and Nuclear Receptor Expression Constructs—To identify the promoter region of the murine Mrp3 gene, the 5'-end of the full-length Mrp3 mRNA sequence was aligned with the mouse genome using the Ensembl Genome Browser (www.ensembl.org). Based upon this sequence information, the following primers were designed (restriction sites are underlined): MRP3PROM5.0-F1, 5'-AAAAGCTAGCTTTAGGCTGTCTTGGAACCCAT-3'; MRP3PROM2.5-F1, 5'-AAAAGCTAGCTTCTTTCTTCCTCCGCCATGGTCCC-3'; MRP3PROM1.1-F1, 5'-AAAAGCTAGCAGAGAGAGAAGGAGGGGAACCCAGC-3'; MRP3PROM1.0-F1, 5'-AAAAGCTAGCGTTTACAGAGTGAGTTCAAGGGTAG-3'; MRP3PROM0.5-F1, 5'-AAAAGCTAGCAAGTGGGGAAGAGAAGTGCCCTGGG-3'; and MRP3PROM-R1, 5-AAAAAAGCTTTTCAGCTCAGCCCCAGCAGATCT-3'. Genomic DNA was prepared from male C57BL/6J mouse liver using a DNeasy genomic DNA preparation kit (Qiagen) and used as a template for Elongase (Invitrogen) PCR using the following primer pairs: MRP3PROM5.0-F1/MRP3PROM-R1 (-4953/+23), MRP3PROM2.5-F1/MRP3PROM-R1 (-2500/+23), MRP3PROM1.1-F1/MRP3PROM-R1 (-1158/+23) MRP3PROM1.0-F1/MRP3PROM-R1 (-1002/+23), and MRP3PROM0.5-F1/MRP3PROM-R1 (-541/+23). The PCR products were digested with NheI/HindIII and ligated into the pGL3-Basic vector (Promega, Madison, WI) that had previously been digested with the same enzyme pair. For construction of the murine nuclear receptor expression plasmids, the following primer pairs were used: mCAR-F1, 5'-AAAAGAATTCCAGGAGACCATGACAGCTATGCTA-3'; mCAR-R1, 5'-AAAAGGATCCCAAGCCTGGGCCTCAACTGCAAAT-3'; mFXR-F1, 5'-AAAAGGATCCGCTAAGGATGGTGATGCAGTTTCAG-3'; mFXR-R1, 5'-AAAAGATGTGCAGTGATGGACACCAGTGGGATCCGGATCC-3'; mPXR-F1, 5'-AAAAGAATTCAACCTAGAGATGAGACCTGAGGAG; mPXR-R1, 5'-AAAAGAATTCCCACTCAGCCATCTGTGCTGCTAA-3'; mVDR-F1, 5'-AAAAGAATTCCTTCAGGGATGGAGGCA; and mVDR-R1, 5'-AAAAGGATCCTGGTCAGGAGATCTCATTGCCGAA-3'. Mouse liver total RNA (5 µg) was reverse transcribed using Superscript II Reverse Transcriptase and oligo(dT)25 as a primer according to the supplier's instructions. The appropriate nuclear receptor primer pairs were used in an Elongase (Invitrogen) PCR with 1 µlof the reverse transcription reaction as an amplification template. The PCR products were purified using a Qiaquick PCR purification kit (Qiagen), digested with appropriate restriction enzymes, purified again, and ligated into the pSG5 vector (Stratagene, La Jolla, CA) that had previously been digested with the same restriction enzymes as the insert. All promoter and nuclear receptor expression constructs were verified by restriction mapping and sequencing. The plasmids pSG5-RXR{alpha} and pSG5-mPPAR{alpha} were a generous gift of Frank Gonzalez (NCI, National Institute of Health, Bethesda, MD).

Site-directed Mutagenesis and Mrp3 Promoter Construct—Site-directed mutagenesis of the Mrp3 promoter was performed using a QuikChange II Site-directed Mutagenesis Kit according to the supplier's (Stratagene) instructions. Complimentary primers of the sequence (mutated residues underlined), MRP3MUT(+), 5'-GGCAAGCAGGTCTTTGATTGCGAGGCCAGCTTGG-3' and MRP3MUT(-), 5'-CCAAGCTGGCCTCGCAATCAAAGACCTGCTTGCC-3', corresponding to bases -1035/-1002 of the Mrp3 promoter were synthesized. These primers, along with 25 µg of pGL3-MRP3(1.1) as a template, were used for site-directed mutagenesis to generate pGL3-MRP3(1.1)mut. To construct pGL3-Promoter-VDRE, oligonucleotides MRP3PROM1.1-F1 and MRP3PROM1.0-R1 (5'-AAAAAGATCTCTTGAACTCACTCTGTAAACCAAGC-3') were used to amplify the region of the Mrp3 promoter corresponding to bases -1158/-983. The PCR products were digested with NheI/BglII and ligated into the pGL3-Promoter vector (Promega) that had previously been digested with the same enzymes. All constructs were verified by restriction mapping and sequencing.

Electromobility Shift Analysis (EMSA)—Murine VDR and human RXR{alpha} were synthesized in vitro by programming the TNT-coupled transcription/translation system (Promega) with 2 µg of pSG5-mVDR or pSG5-RXR{alpha}. Oligonucleotides of the following sequences (DR3 elements underlined): OCVDRE(+), 5'-GCACTGGGTGAATGAGGACATTACT-3'; MRP3VDRE(+), 5'-CAAGCAGGTCTTTGAGGTCGAGGCC-3'; MUTMRP3VDRE(+), 5'-CAAGCAGGTCTTTGATTGCGAGGCC-3'; along with oligonucleotides of complimentary sequences were synthesized (Integrated DNA Technologies Inc., Coralville, IA). The oligonucleotides were mixed (50 ng/µl final concentration) and denatured by heating to 95 °C for 10 min in 0.01 M Tris-Cl, 5 mM MgCl2 (pH 7.9) and allowed to anneal by slowly cooling to room temperature. The annealed oligonucleotides were end labeled with [{gamma}-32P]ATP using T4 polynucleotide kinase according to the supplier's (Fermentas) instructions. In a total volume of 20 µl of binding buffer (25 mM Tris-Cl (pH 7.5), 40 mM KCl, 0.5 mM MgCl2, 0.1 mM EDTA, 5 mM dithiothreitol, 10% glycerol), the following components were combined: 2 µg of poly(dI-dC) and 1.5 µl of each in vitro translation reaction. Where appropriate, cold competitor oligonucleotide (10-, 100-, or 500-fold excess) or 1 µg of VDR antibody (catalogue number sc-1008X; Santa Cruz Biotechnology, Santa Cruz, CA) were also included. After a 20-min incubation at room temperature, 50,000 cpm of the labeled oligonucleotide was added and incubation was continued for a further 30 min. Samples were analyzed on 5% non-denaturing polyacrylamide gel, containing 2.5% glycerol, in 0.4x TBE (1x = 89 mM Tris-Cl, 89 mM boric acid, 2 mM EDTA). After drying, the gels were exposed to phosphorimager screen cassettes and visualized using a Storm 840 Phosphorimager system (Amersham Biosciences).

Cell Culture and Transfection—Human hepatocellular carcinoma (HepG2), human colon adenocarcinoma (HT-29), and murine adenocarcinoma MCA-38 cells were maintained in a complete medium consisting of phenol red-free Dulbecco's modified Eagle's medium (Hyclone, Logan, UT) supplemented with 4 mM L-glutamine (Hyclone), 10% heat inactivated charcoal dextran-stripped fetal bovine serum (Gemini Bioproducts, Woodland, CA), 1 mM sodium pyruvate (Sigma), and 100 units/ml penicillin, 100 µg/ml streptomycin (Sigma) at 37 °C in 95% air, 5% CO2. For transfections, HepG2 or MCA-38 cells were plated at a density of 130,000 cells/ml (0.5 ml/well) in 24-well plates and transfected 18–24 h later with (per well) 150 ng of luciferase reporter construct, 50 ng of nuclear receptor expression construct, 125 ng of pCMV-{beta}gal and pBSK to a total of 500 ng and using TransIT-LT1 (Mirus, Madison, WI; HepG2) or Lipofectamine 2000 (Invitrogen; MCA-38) as per the supplier's instructions. Cells were treated 18–24 h after transfection with Me2SO (control) or the test compounds (all except GW4064 from Sigma): androstenol, pregnenolone 16{alpha}-carbonitrile, 9-cis-retinoic acid (RA), 1,4-bis-[2-(3,5-dichloropyridyloxy)]benzene, GW4064 (a generous gift from Timothy Willson, GlaxoSmithKline), clofibrate, dexamethasone, VD3, CA, CDCA, ursodeoxycholic acid, deoxycholic acid (DCA), LCA, ketolithocholic acid (KLCA), or diketolithocholic acid (dKLCA) at the indicated concentrations. All test compounds were solubilized in Me2SO and diluted in media to a maximum final concentration of 0.5% (v/v) Me2SO. 18–24 h post-treatment cells were harvested using Reporter Lysis Buffer (Promega) as per supplier's instructions. Cell lysates were assayed for luciferase activity using the Luciferase Assay System (Promega) and measured using a Luminoskan Ascent (Thermo Labsystems, Franklin, MA). All luciferase values were corrected for lysate {beta}-galactosidase activity measured colorimetrically on a Power-WaveX microplate spectrophotometer (Bio-Tek Instruments, Winooski, VT).

Gene Expression Analysis—Total RNA was isolated from murine tissues and cultured cells using TRIzol reagent as per the supplier's (Invitrogen) instructions. Total RNA (5 µg) was reverse transcribed using Superscript II Reverse Transcriptase (Invitrogen) with random hexamers pd(N)6 (Amersham Biosciences) according to the supplier's instructions. The synthesized cDNA was then amplified by quantitative PCR using a Stratagene MX3000p thermocycler in a total volume of 25 µl with Brilliant SYBR Green QPCR Master Mix (Stratagene, Cedar Creek, TX). The following primer pairs were used for quantitative PCR analysis: mMRP3PCR-F1, 5'-AGAGCTGGGCTCCAAGTTCT-3'; mMRP3PCR-R1, 5'-TGGTGTCTCAGGTAAAACAGGTAGCA-3'; mVDR-F1, 5'-ACCCTGGTGACTTTGACCG-3'; mVDR-R1, 5'-GGCAATCTCCATTGAAGGGG-3'; mGAPDH-F1, 5'-GAAGGTCGGTGTGAACGGATTTGGC-3'; mGAPDH-F2, 5'-TTGATGTTAGTGGGGTCTCGCTCCTG-3'; mCalbindin3-F, 5'-ATGTGTGCTGAGAAGTCTCCT-3'; mCalbindin3-R 5'-CGCCATTCTTATCCAGCTCCTT-3'; mCyp3a11-F1, 5'-TGGTCAAACGCCTCTCCTTG-3'; mCyp3a11 5'-TGAATGTGGGGGACAGCAAAG-3'; mSt2a2-F1, 5'-TAACTTACCCCAAGTCAGGAACG-3'; mSt2a2-R1, 5'-ATGGGAAGATGGGAGGTTATGA-3'; hMRP3-F1, 5'-CCTGCCCCTGTTTTCTTTGTC-3'; hMRP3-R1, 5'-TTGTGTCGTGCCGTCTGCTTTTCC-3'; hGAPDH-F1, 5'-TGGAAATCCCATCACCATCT-3'; hGAPDH-R1, 5'-GTCTTCTGGGTGGCAGTGAT-3'. Thermal cycling conditions were identical for each primer pair and were as follows: a single cycle of 94 °C for 10 min, followed by 40 cycles of denaturation at 94 °C for 30 s, annealing at 60 °C for 18 s, and elongation at 72 °C for 30 s. Melting curves were generated from 60 to 94 °C at the end of the PCR protocol to ensure the amplification of a single product. The PCR products were then separated on a 2.5% agarose gel and visualized by ethidium bromide staining to ensure the formation of a single product at the appropriate size was generated. Relative CT values were obtained by the {Delta}{Delta}CT method (28) using a threshold of 10 S.D. above background for CT.

RNA Interference Analysis—PCR amplification was used to generate a 570-bp murine VDR cDNA and a 549-bp Aequorea victoria green fluorescent protein (GFP) cDNA from the appropriate plasmid vectors (pSG5-mVDR and pEGFP-C1). The primers incorporated a 3-bp stabilizer and T7 promoter sequence (lowercase) at the 5'-ends. The primer sequences were as follows: mVDR siRNA-F1, 5'-gcgtaatacgactcactataggTCTGAGGAGCAACAGCACATTATC-3'; mVDR siRNA-R1, 5'-gcgtaatacgactcactataggGCCCCACCTGGAACTTTATGAG-3'; eGFP siRNA-F1, 5'-gcgtaatacgactcactataggGCAAGCTGACCCTGAAGTTCATC-3'; eGFP siRNA-R1, 5'-gcgtaatacgactcactataggAACTCCAGCAGGACCATGTGATCG-3'. After purification (GeneElute PCR purification kit, Sigma), 1 µg of the PCR products were used to generate double-stranded RNA using the RNAMaxx high yield transcription kit as per the supplier's instructions (Stratagene). The double-stranded RNA was cleaved in vitro to small (21–26 bp) siRNA molecules using recombinant Dicer as per the supplier's instructions (Stratagene), and subsequently purified using a MicroSpin G-25 column (Amersham Biosciences) followed by a Microcon YM-100 column (Millipore, Billerica, MA). Murine colon adenocarcinoma cells (MCA-38) were maintained in a complete medium identical to that described for HepG2 cell culture. The cells were plated at a density of 130,000 cells/ml (1.0 ml/well) in 12-well plates and transfected 18–24 h later with 0, 10, or 25 nM siRNA using 2.5 µl/well GeneEraser siRNA transfection reagent as per the supplier's instructions (Stratagene). After 24 h, cells were treated with Me2SO or VD3 at the indicated concentrations and after a further 24 h, total RNA was prepared using TRIzol reagent and gene expression analysis was performed as described above.

Statistics—Values are reported as mean ± S.D. Comparisons between experimental groups were made using the Student's t test for independent samples (two-tailed).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Cloning and Expression of Murine Mrp3—To determine empirically the transcription start site of the murine Mrp3 gene, 5'-RACE was performed using mouse liver and colon RNA as templates. Analysis of clones derived from the 5'-RACE reactions for each tissue (8 clones per tissue) revealed a very homogenous group exhibiting >99% sequence identity and a maximum of 7 bp difference in the length of sequence obtained at the 5'-end for the shortest versus longest clones. To generate a full-length Mrp3 cDNA, a gene-specific primer was designed based upon the extreme 5'-sequence of the largest 5'-RACE clone and was used in a 3'-RACE reaction. Sequencing of the resulting Mrp3 cDNA indicated a clone with an additional 66 bp at the 5'-end compared with the longest murine Mrp3 sequence (BC048825 [GenBank] ) present with GenBankTM at the time that this study was conducted. Alignment of the full-length Mrp3 cDNA (4998 bp) with the mouse genome using the Ensembl Genome Browser (www.ensembl.org) allowed the assignment of exons and the production of a map of the murine Mrp3 gene (Fig. 1A). Similar to the human gene (29), the murine Mrp3 gene consist of 31 exons, all of which contain coding sequence. The murine Mrp3 gene is located on chromosome 11 and spans 48,814 bp. A single open reading frame of 4,569 bp encoding a protein of 1522 amino acids was identified. The predicted protein exhibits 92 and 80% overall sequence identity to the rat and human isoforms, respectively (Fig. 1B). Real-time PCR analysis of the expression level of Mrp3 mRNA in various tissues with known relevance and/or exposure to bile acids revealed the highest levels of basal expression in the colon followed in order by liver, duodenum, jejunum, ileum, and kidney (Fig. 1C).

Cloning and Regulation of the Mrp3 Promoter—After establishing the transcription start site by 5'-RACE, oligonucleotide primers were designed to amplify ~5 kb of the regulatory region (-4953/+23) of the murine Mrp3 gene. This sequence was placed upstream of the luciferase structural gene of pGL3-basic to create the pGL3-MRP3(5.0) promoter reporter construct. The function of nuclear receptors CAR, FXR, PPAR{alpha}, PXR, and VDR have been reported to be directly and/or indirectly affected by bile acids (23–27, 30). The ability of these nuclear receptors to modulate the expression of this promoter construct was tested by transient transfection of HepG2 cells with pGL3-MRP3(5.0) and expression constructs for various nuclear receptors. As all of these receptors function as heterodimers with RXR, an expression construct for this receptor was also included in all co-transfections. As shown in Fig. 2A, co-transfection with an RXR expression construct alone or co-transfection with expression constructs for VDR and RXR were found to activate expression of pGL3-MRP3(5.0) greater than 2-fold upon treatment with the corresponding ligands (RA or VD3). Co-transfection with expression constructs for CAR and RXR and treatment with androstenol reduced expression of pGL3-MRP3(5.0) by ~40% versus Me2SO control. Cotransfection of the remaining nuclear receptor expression constructs was without effect in either the presence or absence of ligand. To further explore the activation of the Mrp3 promoter by VDR and RXR ligands, HepG2 cells were co-transfected with the reporter construct and expression constructs for RXR and VDR followed by treatment with the corresponding ligands. Treatment with RA activated the expression of pGL3-MRP3(5.0) ~2.5-fold when the cells were co-transfected with expression constructs for RXR or VDR and RXR (Fig. 2B). Activation, to a lesser extent, was seen for RA-treated cells transfected with an expression construct for VDR alone, however, no activation was observed for co-transfection with empty expression vector (pSG5). Treatment with VD3 activated pGL3-MRP3(5.0) ~8-fold when co-transfected with expression constructs for VDR and RXR. A small amount (~2-fold) of activation was observed with co-transfection of the VDR expression construct alone. Treatment of cells co-transfected with expression constructs for VDR and RXR with RA and VD3 resulted in no further activation beyond that seen for VD3 treatment alone. None of the transfections or the ligand treatments had a significant effect on the expression of the promoterless reporter construct pGL3-Basic.



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 1.
Characterization and expression of the murine Mrp3 gene. A, exon (boxes) assignment and genomic structure of the murine Mrp3 gene based upon alignment of the full-length Mrp3 cDNA sequence with the murine genome. B, amino acid length (aa) and identity (%) of the predicted protein sequences for mouse, rat, and human Mrp3/MRP3. C, real-time quantitative reverse transcriptase PCR analysis of Mrp3 mRNA levels in tissues relevant to bile acid homeostasis. Mrp3 mRNA levels were corrected for Gapdh mRNA levels and normalized to the value for liver. All values are mean ± S.D., n = 4.

 
To investigate the ability of bile acids to activate the Mrp3 promoter in a NHR-dependent manner, HepG2 cells were co-transfected with the pGL3-MRP3(5.0) and expression constructs for VDR and RXR. The effect of co-transfection with an expression construct for the Na+-taurocholate co-transporting polypeptide was also determined. Na+-taurocholate co-transporting polypeptide is found on the basolateral membrane of hepatocytes, and is the major protein responsible for transporting bile acids from the blood into the liver, and exhibits affinity for both conjugated and unconjugated bile acids (31, 32). As shown in Fig. 3A, co-transfection with VDR and RXR expression constructs activated the expression of pGL3-MRP3(5.0) upon treatment with VD3, LCA, KLCA, or dKLCA. However, the primary bile acids CA, CDCA, DCA, and ursodeoxycholic acid were unable to activate pGL3-MRP3(5.0). Co-transfection of the Na+-taurocholate co-transporting polypeptide expression construct was without effect in the presence or absence of ligand. To determine a dose-response relationship between bile acid concentration and activation of the Mrp3 promoter construct, the secondary bile acids, LCA, KLCA, and dKLCA, were tested for activation of the pGL3-MRP3(5.0) construct in the presence of VDR and RXR over a range of concentrations. As shown in Fig. 3B, LCA was the most potent bile acid producing a significant 2.5-fold activation of pGL3-MRP3(5.0) at a concentration of 10 µM. Significant activation of the reporter construct by KLCA or dKLCA required higher concentrations, 30 or 50 µM, respectively. Treatment with LCA also produced the greatest magnitude of promoter activation (~4.5-fold versus Me2SO) over the range of concentrations studied. A small degree of cell toxicity was observed at the highest (100 µM) concentration of bile acids tested, however, at lower concentrations toxicity was not evident. As LCA was the most potent and efficacious of the bile acids tested in these assays, this compound was used in all subsequent experiments as a model for secondary bile acid regulation of the Mrp3 promoter. Overall, VD3 was the most potent and efficacious of the ligands studied with significant activation of the pGL3-MRP3(5.0) construct achieved at 1 nM and a maximum activation of 6–7-fold versus Me2SO at concentrations of 30 nM and greater (Fig. 3C). These data demonstrate that the Mrp3 promoter can be activated by VD3 and cholestatic secondary bile acids in a VDR-dependent manner.

Identification and Analysis of cis-Acting Regulatory DNA Elements—To identify the region of the Mrp3 promoter and cis-acting elements that conferred NHR-dependent activation by VD3, RA, and LCA, a series of reporter deletion constructs were generated. As shown in Fig. 4A, co-transfection with expression constructs for VDR and RXR activated expression of pGL3-MRP3(5.0), pGL3-MRP3(2.5), and pGL3-MRP3(1.1) by ~8-, 4-, or 2.5-fold versus Me2SO upon treatment with VD3, LCA, or RA, respectively. In contrast, co-transfection of VDR and RXR expression constructs failed to activate the expression of pGL3-MRP3(1.0) and pGL3-MRP3(0.5) upon treatment with VD3, LCA, or RA. Whereas the inducibility of the Mrp3 promoter was clearly affected by the length of promoter sequence, the basal activity of all reporter constructs was similar at ~7-fold greater than the promoterless pGL3-Basic plasmid. Therefore, the response element conferring VDR-dependent activation of the Mrp3 promoter was hypothesized to lie between -1158/-1052 bp upstream of the Mrp3 transcription start site.



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 2.
Regulation of the Mrp3 promoter by NHR ligands. A, HepG2 cells were co-transfected with a luciferase reporter construct containing the murine Mrp3 promoter (-4953/+23) and expression constructs for the indicated NHRs. 24 h after transfection, the cells were treated with the following ligands: RXR, ligand A, 1 µM RA; CAR/RXR, ligand A, 10 µM 1,4-bis-[2-(3,5-dichloropyridyloxy)]benzene, ligand B, 50 µM androstenol; FXR/RXR, ligand A, 1 µM GW4064, ligand B, 50 µM CDCA; PPAR{alpha}/RXR, ligand A, 50 µM clofibrate, ligand B, 50 µM Wy-14,643; PXR/RXR, ligand A, 50 µM dexamethasone, ligand B, 50 µM pregnenolone 16{alpha}-carbonitrile; VDR/RXR, ligand A, 250 nM VD3. All values were corrected for {beta}-galactosidase activity and normalized to the values obtained in the absence of co-transfection with an NHR expression construct (none) and treatment with Me2SO. All values are mean ± S.D., n = 3. *, p < 0.05 versus Me2SO for each NHR. B, HepG2 cells were co-transfected with a promoterless luciferase reporter construct (pGL3-Basic) or a construct containing the murine Mrp3 promoter (pGL3-MRP3(5.0)) and expression constructs for the indicated NHRs. 24 h after transfection, the cells were treated with the following ligands: 1 µM RA and/or 250 nM VD3. All values were corrected for {beta}-galactosidase activity and expressed as arbitrary relative luciferase units (RLU). All values are mean ± S.D., n = 4. *, p < 0.05 versus Me2SO for each group.

 
Sequencing and analysis of the pGL3-MRP3(1.1) construct indicated the presence of a candidate direct repeat-3 (DR3) VDRE located between bases -1028 and -1014 of the Mrp3 promoter (Fig. 5A). This candidate Mrp3 VDRE differs from the consensus DR3 sequence by single base changes in each of the half-sites. To further investigate the functionality of the candidate VDRE, a 175-bp region (-1158/-983 bp) of the Mrp3 promoter that contained this element was tested for the ability to enhance the expression of a heterologous promoter. This region was placed upstream of the SV40 promoter of the pGL3-Promoter construct to generate the pGL3-Promoter-VDRE reporter construct. As shown in Fig. 5B, co-transfection of VDR and RXR expression constructs activated the pGL3-Promoter-VDRE reporter construct ~4-, 2.5-, or 1.5-fold upon treatment with VD3, LCA, or RA, respectively. In contrast, co-transfection of VDR and RXR expression constructs had no impact upon the expression of the parent pGL3-Promoter construct regardless of treatment. In a complementary set of experiments, a 3-base change in one of the half-sites of the candidate DR3 sequence (Fig. 5A) was achieved through site-directed mutagenesis of the pGL3-MRP3(1.1) construct. Mutation of the candidate VDRE resulted in a substantial reduction in the activation of the mutant (pGL3-MRP3(1.1)mut) versus parent (pGL3-MRP3(1.1)) construct upon cotransfection with VDR and RXR expression constructs and treatment with VD3, LCA, or RA (Fig. 5C).



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 3.
VDR-dependent regulation of the Mrp3 promoter by secondary bile acids. A, HepG2 cells were co-transfected with a murine Mrp3 promoter luciferase reporter construct (pGL3-MRP3(5.0)), expression constructs for VDR and RXR, and where indicated, an empty expression construct or an expression construct for Na+-taurocholate co-transporting polypeptide. 24 h after transfection, the cells were treated with 50 µM of indicated bile acids. VD3 was used at 250 nM. All values were corrected for {beta}-galactosidase activity and expressed as arbitrary relative luciferase units (RLU). All values are mean ± S.D., n = 4. *, p < 0.05 versus Me2SO for each group. B, HepG2 cells were co-transfected with the murine Mrp3 promoter luciferase reporter construct (pGL3-MRP3(5.0)), and expression constructs for VDR and RXR. 24 h after transfection the cells were treated with bile acids at the indicated concentrations. All values were corrected for {beta}-galactosidase activity and normalized to that obtained for Me2SO. All values are mean ± S.D., n = 4. *, p < 0.05 versus Me2SO. C, HepG2 cells were co-transfected with the murine Mrp3 promoter luciferase reporter construct (pGL3-MRP3(5.0)), and expression constructs for VDR and RXR. 24 h after transfection, the cells were treated with VD3 at the indicated concentrations. All values were corrected for {beta}-galactosidase activity and normalized to that obtained for Me2SO. All values are mean ± S.D., n = 3. *, p < 0.05 versus Me2SO.

 



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 4.
Deletion analysis of the Mrp3 promoter. HepG2 cells were co-transfected with a promoterless luciferase reporter construct (pGL3-Basic) or with the indicated Mrp3 promoter luciferase reporter constructs and expression constructs for VDR and RXR. 24 h after transfection, the cells were treated with 250 nM VD3, 50 µM LCA, or 1 µM RA. All values were corrected for {beta}-galactosidase activity and expressed as arbitrary relative luciferase units (RLU). All values are mean ± S.D., n = 4. *, p < 0.05 versus Me2SO for each group.

 
After establishing that the putative VDRE element could confer VDR-dependent activation of the Mrp3 promoter by VD3, RA, and LCA, experiments were designed to test the binding of VDR/RXR heterodimers to this sequence. Double-stranded oligonucleotides (Fig. 6A) corresponding to the candidate Mrp3 DR3 and 5 bp of distal and proximal flanking sequence from the promoter (MRP3 VDRE); a mutant MRP3 DR3 with a 3-bp change in one of the NHR half-sites (mutMRP3 VDRE); and, a control oligonucleotide corresponding to the VDRE of the rat osteocalcin gene (OC VDRE) (33) were used for these studies. These oligonucleotides were 32P-end-labeled and used as probes in a series of EMSA in combination with in vitro translated VDR and RXR proteins. Incubation of the VDR and RXR proteins with the OC VDRE oligonucleotide resulted in the formation of a protein-DNA complex as evidenced by the presence of a labeled band delayed in electrophoretic migration compared with the free probe (Fig. 6B). Formation of this complex was dependent upon the presence of both proteins and was reduced substantially by the inclusion of a 100-fold excess of unlabeled OC VDRE oligonucleotide in the binding reaction. A similar protein-DNA complex, differing from the OC VDRE complex in intensity only, was formed upon incubation of VDR and RXR proteins with the labeled Mrp3 VDRE oligonucleotide. Effective competition of this complex was achieved by a 100-fold excess of unlabeled OC VDRE oligonucleotide. Similarly, inclusion of an excess of the unlabeled MRP3 VDRE, but notmutMRP3VDREoligonucleotide, resulted in a concentration-dependent reduction in intensity of the protein-DNA complex formed with VDR and RXR using labeled OC VDRE as the probe (Fig. 7A). As the VDR and RXR proteins used for the EMSA were generated by in vitro translation and thus, were contaminated with a number of reticulocyte proteins, supershift analysis was performed to verify the identity of the protein-DNA complex formed with the labeled MRP3 VDRE oligonucleotide. Inclusion of a VDR antibody in the binding reaction resulted in the formation of an additional labeled complex with substantially retarded electrophoretic mobility compared with that formed in the absence of antibody (Fig. 7B). Furthermore, the formation of this complex was dependent upon the presence of VDR and RXR proteins in the binding reaction as evidenced by the absence of a supershift band with the VDR antibody without the VDR or RXR proteins. These data demonstrate that activation of the Mrp3 gene in response to treatment with VD3 and bile acids occurs through a direct mechanism mediated by binding of the VDR to a DR3 response element present within the promoter region of Mrp3 gene.

In Vivo Modulation of Murine Mrp3—To determine whether the VDR-dependent regulation of the Mrp3 gene occurs in vivo, mice were administered vehicle (corn oil), VD3 or LCA (2 µg or 10 mg, respectively) by oral gavage, once per day, for 4 consecutive days. Modulation of Mrp3 gene expression in response to these treatments was measured by quantitation of Mrp3 mRNA levels by real-time quantitative PCR of total RNA prepared from liver and colon. These tissues were chosen for study for three reasons: the highest constitutive levels of Mrp3 mRNA are contained within these tissues (Fig. 1C); the levels of VDR mRNA expression are over 3 orders of magnitude greater in colon versus liver (Fig. 8A); and, these tissues are routinely exposed to high levels of bile acids and are physiologically relevant to bile acid homeostasis. Treatment of mice with LCA or VD3 resulted in significant increases of Mrp3 mRNA levels in colon, but not liver (Fig. 8B). As a positive control for VDR-dependent regulation, measurement of calbindin-D9K (Cabd9k) mRNA levels was also performed. Similar to the effect on the Mrp3 gene, Cabd9k mRNA levels were increased significantly in colon, but not liver, of LCA- and VD3-treated mice (Fig. 8C). The genes encoding cytochrome P450 3a11 (Cyp3a11) and dehydroepiandrosterone-preferring sulfotransferase (Sult2a2 in mouse) are known to be regulated in a VDR-dependent manner and to be involved in the protection of cells against LCA toxicity (23–25, 33–35). Increased levels of the mRNAs for these genes were also observed in the colon of VD3 versus vehicle-treated mice. Taken together, these data confirm that colon Mrp3 expression, as well as that of other genes with a protective role in bile acid toxicity (Cyp3a11 and Sult2a2), can be activated in vivo by VD3 or cholestatic bile acids such as LCA.

Mrp3 Regulation in Murine and Human Colon Cell Lines— To further investigate the role of the VDR in regulating colon Mrp3 gene expression, transfection of siRNA was used to reduce VDR mRNA in murine colon adenocarcinoma MCA-38 cells. By targeting the primers used for real-time quantitative PCR analysis to a region 3' of the region targeted by the VDR siRNA, it was possible to measure the effects of siRNA treatment at the mRNA level. Transfection of VDR siRNA resulted in a dose-dependent reduction of VDR mRNA levels in MCA-38 cells 48 h post-transfection (Fig. 9A). A maximum reduction to ~25% of the level of VDR mRNA in mock transfected cells was observed for transfection with 25 nM VDR siRNA. In contrast, no significant reduction of VDR mRNA levels was observed after transfection with up to 25 nM siRNA targeted to the jellyfish (A. victoria) GFP mRNA. Interestingly, VDR, but not GFP siRNA significantly reduced the apparent autoregulation of VDR gene expression that occurred upon treatment with VD3. As a positive control for alterations in VDR function as a consequence of the reduced VDR mRNA levels, induction of Cabd9k expression was measured by real-time quantitative PCR analysis (Fig. 9B). Transfection with 25 nM VDR siRNA reduced the increase of Cabd9k mRNA levels in response to VD3 treatment by ~50% compared with mock transfected cells. In contrast, constitutive levels of Cabd9k expression were not affected by VDR siRNA transfection. Transfection of GFP siRNA was without effect on the constitutive or inducible expression of Cabd9k. Similar to the effect on Cabd9k mRNA levels, transfection of 25 nM VDR siRNA reduced VD3-induced levels of Mrp3 mRNA to ~50% of that seen for mock transfected cells (Fig. 9C). In contrast to the effects on Cabd9k expression, VDR siRNA transfection also reduced the constitutive levels of Mrp3 mRNA by ~50% compared with mock transfected cells. Transfection of GFP siRNA did not affect the constitutive or inducible expression of Mrp3. As a control for nonspecific effects of siRNA transfection, Gapdh mRNA levels were monitored and were found to be unaffected by transfection of GFP or VDR siRNA (Fig. 9D).



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 5.
A DR3 in the Mrp3 promoter confers VDR-dependent activation by VD3 and bile acids. A, nucleotide sequence of a consensus DR3, the candidate VDRE located in the murine Mrp3 promoter and a mutant VDRE obtained by site-directed mutagenesis. B, HepG2 cells were transfected with an enhancerless SV40 luciferase reporter construct or a reporter construct containing a SV40 promoter and a single copy of the region corresponding to -1158/-983 bp of the murine Mrp3 promoter. Cells were co-transfected with an empty expression construct (none) or expression constructs for VDR and RXR (VDR/RXR). 24 h after transfection, the cells were treated with 50 nM VD3, 50 µM LCA, or 1 µM RA. All values were corrected for {beta}-galactosidase activity normalized to the Me2SO values for each group. All values are mean ± S.D., n = 4. *, p < 0.05 versus Me2SO for each group. C, HepG2 cells were co-transfected with the murine Mrp3 promoter luciferase reporter construct (pGL3-MRP3(1.1)) or a promoter construct in which the VDRE was mutated to the sequence indicated in A. Cells were co-transfected with expression constructs for VDR and RXR and 24 h after transfection, were treated with 50 nM VD3, 250 µM LCA, or 1 µM RA. All values were corrected for {beta}-galactosidase activity and normalized to that obtained for Me2SO. All values are mean ± S.D., n = 4. *, p < 0.05 versus Me2SO for each group. #, p < 0.05 versus same treatment for pGL3-MRP3(1.1).

 



View larger version (33K):
[in this window]
[in a new window]
  		FIG. 6.
VDR/RXR heterodimers bind to the Mrp3 VDRE. EMSAs were performed using in vitro translated VDR and RXR as described under "Materials and Methods." A, nucleotide sequence of the oligonucleotides used in these assays. For each of the oligonucleotides, only the plus strand is shown. VDRE half-sites are underlined. B, the indicated 32P-labeled oligonucleotides (probe) were incubated in the presence or absence of in vitro translated VDR and/or RXR and protein-DNA complexes were resolved on a 5% non-denaturing polyacrylamide gel. Where indicated, an unlabeled oligonucleotide competitor (COMP) was also added to the binding reactions at a concentration equal to 100-fold that of the probe.

 
To determine whether the regulatory properties of the Mrp3 promoter constructs in HepG2 cells were conserved in colon cancer cells, the pGL3-MRP3(1.1) promoter reporter construct was transiently transfected into MCA-38 cells. Without cotransfection of VDR and RXR expression constructs, expression of the pGL3-MRP3(1.1) construct was increased significantly in response to treatment with 1 µM RA or 1 µM RA + 200 nM VD3, but not 200 nM VD3 alone (Fig. 9E). Upon co-transfection with VDR and RXR expression constructs, significant increases in reporter gene expression were observed for all treatments. Endogenous Mrp3 mRNA levels were also measured in human colon adenocarcinoma HT-29 cells (Fig. 9F). Similar to the results for MCA-38 cells (Fig. 9C), VD3 treatment increased Mrp3 mRNA levels by ~60% in HT-29 cells.



View larger version (32K):
[in this window]
[in a new window]
  		FIG. 7.
Specificity of VDR/RXR binding to the Mrp3 VDRE. EMSAs were performed using in vitro translated VDR and RXR as described under "Materials and Methods." A, competition assays in which the OC VDRE was 32P-labeled and used as a probe. Where indicated, unlabeled MRP3 VDRE or mutMRP3 VDRE oligonucleotides were also included in the binding reactions at concentrations in 10-, 100-, or 500-fold excess relative to the OC VDRE. Where indicated, in vitro translated VDR and/or RXR were also included. Protein-DNA complexes were resolved on a 5% non-denaturing polyacrylamide gel. B, supershift assays were performed using 32P-labeled MRP3 VDRE as a probe and where indicated, in vitro translated VDR and/or RXR and a monoclonal antibody against murine VDR. Protein-DNA complexes were resolved on a 5% non-denaturing polyacrylamide gel. The supershifted complex is indicated by the arrow.

 



View larger version (35K):
[in this window]
[in a new window]
  		FIG. 8.
VDR-dependent regulation of the murine Mrp3 gene in vivo. Mice were treated daily with 2 µg of VD3 or 10 mg of LCA by oral gavage, for a total of 4 days. 24 h after the final dose, liver and colon were harvested from each animal and used for preparation of total RNA. After reverse transcription, liver and colon VDR (A), Mrp3 (B), and Cabd9k (C) mRNA levels were analyzed by real-time quantitative reverse transcriptase-PCR. D, Cyp3a11 and Sult2a2 mRNA levels were also analyzed in mouse colon. All values were corrected for Gapdh mRNA levels and normalized to that for liver (A) or the appropriate vehicle-treated control (B–D). All values are mean ± S.D., n = 5. *, p < 0.05 versus vehicle for each group.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
The results of the present study provide compelling evidence for a novel VDR-dependent mechanism by which bile acid inducible expression of the murine Mrp3 gene is achieved. Mrp3 mRNA levels were found to be highest in mouse colon, a tissue commonly exposed to the highly cholestatic secondary bile acid, LCA. Functional analysis of the murine Mrp3 promoter revealed activation by secondary bile acids and VD3, a process dependent upon the presence of a functional VDR. Deletion mapping of the Mrp3 promoter identified the region of the Mrp3 promoter that conferred VDR-dependent activation by VD3 or secondary bile acids. Within this region, an imperfect DR3 element was found that could confer VDR responsiveness upon a heterologous promoter and be bound with high specificity by VDR/RXR heterodimers. Mutation of this DR3 element markedly reduced VDR-dependent activation of the Mrp3 promoter by VD3 and secondary bile acids. In vivo evidence for the relevance of this mechanism was derived from the up-regulation of Mrp3 by VD3 and LCA in colon, a tissue that expresses high levels of the VDR, but not liver, a tissue with extremely low expression of the VDR. This activation of colon Mrp3 expression occurred concomitantly with increased expression of Cyp3a11 and Sult2a2, genes also believed to confer protection against LCA toxicity. Interestingly, siRNA-mediated knock-down of VDR mRNA levels in murine colon cancer cells suggested a role for this receptor in both the constitutive and inducible expression of the Mrp3 gene.



View larger version (42K):
[in this window]
[in a new window]
  		FIG. 9.
VDR-dependent regulation of Mrp3/MRP3 gene in murine and human colon cell lines. Murine adenocarcinoma MCA-38 cells were mock transfected (none) or transfected with the indicated concentrations of GFP- or VDR-specific siRNA. After 24 h, the cells were treated with Me2SO or 250 nM VD3. The cells were harvested 24 h later, total RNA was prepared and analyzed for VDR (A), Cabd9k (B), Mrp3 (C), and Gapdh (D) mRNA levels by real-time quantitative reverse transcriptase-PCR. All values were corrected for GAPDH mRNA levels and normalized to that for mock transfected cells treated with Me2SO. All values are mean ± S.D., n = 3. *, p < 0.05 versus mock transfected for each treatment group. E, MCA-38 cells were co-transfected with a murine Mrp3 promoter luciferase reporter construct (pGL3-MRP3(1.1)), an empty expression vector (none), or expression constructs for VDR and RXR. 24 h after transfection, the cells were treated with Me2SO, 250 nM VD3, 1 µM RA, or 250 nM VD3 plus 1 µM RA. All values were corrected for {beta}-galactosidase activity and expressed relative to the corresponding Me2SO treated control. All values are mean ± S.D., n = 3. *, p < 0.05 versus Me2SO for each group. F, human colon adenocarcinoma HT-29 cells were treated with Me2SO or 250 nM VD3. 24 h later total RNA was prepared and analyzed for MRP3 mRNA levels by real-time quantitative reverse transcriptase-PCR. All values were corrected for GAPDH mRNA levels and normalized to cells treated with Me2SO. All values are mean ± S.D., n = 3. *, p < 0.05 versus Me2SO (vehicle) control.

 
A broad range of physiological roles has been described for the VDR and the target genes regulated by this NHR. The most well established functions are the control of calcium homeostasis and regulation of bone resorption (36). A large body of evidence derived from molecular, cellular, animal, and epidemiological studies has indicated physiological roles for the VDR beyond these primary functions. For example, VD3 inhibits cell proliferation in various types of cancer cells and may provide protection against prostate and colon cancers (37–39). Accumulating evidence also indicates that the VDR regulates the differentiation of cells of the immune system and thereby, has an important systemic immunomodulatory function (40, 41). Among the newest recognized functions for the VDR is as a regulator of bile acid homeostasis. This aspect of VDR biology differs from other functions of this NHR in that the relevant biological responses can be elicited in response to VD3 or bile acid ligands. Consistent with the data presented in this study, the VDR is preferentially activated by cholestatic secondary versus primary bile acids (25). Bile acid activation of the VDR induces the expression of CYP3A genes, which encode cytochrome P450 enzymes capable of the hydroxylation and thereby, detoxification of LCA (25, 42, 43). Similar to the findings regarding the VDR-dependent regulation of Mrp3 expression in the present study, VDR-dependent activation of Cyp3A expression by LCA is conferred by a DR3 motif located in the regulatory region of the CYP3A gene (25). Recently, dehydroepiandrosterone preferring sulfotransferase (Sult2a2), the exclusive sulfating enzyme for LCA (44), was described as a target gene for VDR-dependent regulation (33). The sulfate conjugate of LCA is much less cytotoxic than LCA and is more readily eliminated in the feces (12, 45–47). Moreover, a role for Sult2a2 in protection against LCA-induced hepatotoxicity has been demonstrated in mice (34, 35). MRP3 is a low affinity transporter capable of transporting a wide range of organic anion substrates including bile acid sulfates (9, 10). The expression of MRP3/Mrp3 is induced under pathological conditions associated with elevated bile acid concentrations including intrahepatic cholestasis, Dubin-Johnson syndrome, as well as a number of animal models of cholestasis (14–17). This induction of MRP3 may confer cytoprotection and/or be part of a homeostatic mechanism to maintain normal bile acid metabolism in cells exposed to elevated levels of bile acids. The VDR-dependent constitutive and inducible expression of the Mrp3 gene by VD3 and LCA reported in this study are consistent with this proposition.



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 10.
A model for the regulation of LCA detoxification in colon cells. LCA, derived from the deconjugation and/or dehydroxylation of the primary bile acid CDCA passes readily into the colon. Passive diffusion of LCA into colon cells results in activation of VDR/RXR heterodimers and subsequently induction of Cyp3a11, Sult2a2, and Mrp3 expression. LCA is then hydroxylated by Cyp3a11 and/or sulfated by Sult2a2 leading to a reduction in the inherent toxicity of this bile acid. Mrp3 and other multispecific organic anion transporters such as Mrp2 may efflux LCA conjugates into the blood and colon lumen, respectively. The primary function of Mrp3-mediated efflux of LCA sulfate conjugates into the blood is to protect colon cells from LCA toxicity. However, a secondary benefit of this activity may be to promote the systemic elimination of LCA sulfate conjugates via the liver and kidney.

 
In addition to the VDR, bile acids are ligand activators of other NHRs involved in normal or adaptive bile acid metabolism. FXR serves as an intracellular bile acid sensor in the liver and intestine and is critical for the feedback regulation of hepatic bile acid biosynthesis and feedforward regulation of hepatic bile acid efflux and intestinal bile acid uptake (48). PXR has been identified as a receptor for secondary bile acids and to mediate a hepatoprotective induction of CYP3A and organic anion transporting polypeptide-2 gene expression in response to exposure to LCA (23, 24). Pharmacologic activation of CAR protects against LCA-induced hepatotoxicity primarily by activating the expression of CYP3A and MRP3 genes (26, 27, 34), however, it is not yet clear if bile acids directly regulate the transcriptional activity of this receptor. Thus, at least four distinct NHRs contribute to the maintenance of bile acid homeostasis through the regulation of overlapping subsets of genes that confer protection against the accumulation of toxic intracellular concentrations of bile acids. Induction of MRP3 gene expression during intrahepatic cholestasis most likely contributes to the efflux of toxic bile acids from hepatocytes under these conditions. The lack of hepatic Mrp3 induction after VD3 or LCA treatment in the present study argues against a role for the VDR in this response. Instead, the much higher levels of VDR expression and the increase of colon Mrp3 mRNA levels in response to VD3 or LCA indicates that the physiologically relevant site for VDR-dependent regulation of Mrp3 is the gastrointestinal tract. In addition to the inducibility of Mrp3 in the colon, it is likely that VDR-dependent activation of this gene by bile acids also contributes to the higher constitutive expression of this transporter in colon versus liver, in vivo. This is supported by the reduction of constitutive and inducible Mrp3 expression in response to VDR mRNA knock-down in MCA-38 colon cancer cells.

Fecal bile acid concentrations are often elevated in populations at high risk for colon cancer (49, 50) and numerous studies have implicated bile acids as tumor promoters (51–54). LCA is the most toxic bile acid toward colon cells and has been shown to increase the number of tumors formed in chemically induced animal models of colon carcinogenesis (51, 52). Multiple complementary mechanisms may contribute to the protection of colon cells from the toxic effects of LCA. For example, CYP3A-dependent hydroxylation and Sult2a2-dependent sulfation increases the water solubility, reduces the inherent toxicity, and facilitates the elimination of LCA (23–25, 33–35). Similar to Mrp3, the genes encoding Cyp3a and Sult2a2 enzymes are regulated by bile acids in a VDR-dependent manner (25, 33, 34) and have been shown to be up-regulated in the colon upon treatment with VD3 in this study. LCA is derived from the deconjugation and/or dehydroxylation of the primary bile acid CDCA at the C-7 position by intestinal bacteria (5, 12). Unlike most primary bile acids that are efficiently reabsorbed in the ileum, LCA passes readily into the colon and is generally present at a level of 2–3.5 µg/mg of fecal dry weight in healthy individuals (55). The data from the present study provide important new information regarding the mechanisms by which colon cells are protected against the high concentrations of this toxic bile acid. Under this scenario (Fig. 10), passive diffusion of LCA into colon cells results in activation of VDR/RXR heterodimers and subsequently induction of Cyp3a11, Sult2a2, and Mrp3 expression. LCA is then hydroxylated by Cyp3a11 and/or sulfated by Sult2a2 leading to a reduction in the inherent toxicity of this bile acid. MRP3/Mrp3 expression has been localized to the basolateral face of enterocytes (56) and thus, may function subsequently to transport LCA conjugates from within the cell into the blood. Other multispecific organic anion transporters such as Mrp2 may efflux LCA conjugates into the colon lumen and thus, act in concert with Mrp3 to lower the intracellular concentration of LCA conjugates (1). Whereas the primary function of Mrp3-mediated efflux of LCA sulfate conjugates into the blood is to protect colon cells from LCA toxicity, a secondary systemic benefit may also be possible. Once returned to the liver, LCA sulfates may be further conjugated with glycine or taurine and subsequently secreted into the bile (12). As conjugates of LCA are poorly reabsorbed, this enterohepatic path of LCA reabsorption, metabolism, and transport may facilitate the clearance of this toxic secondary bile acid through the feces and/or urine. The ability of the VDR to bind LCA and activate the transcription of genes critical for this process places this NHR as a key regulator of bile acid homeostasis within colon cells. Animal studies have identified a chemopreventive effect for VD3 in colon cancer and polymorphisms of the human VDR gene have been associated with a modified risk for the development of colon cancer (37–39). Further study will help to determine whether altered regulation of bile acid metabolism in colon cells contributes to these effects.

In conclusion, we have identified a novel VDR-dependent pathway by which cholestatic secondary bile acids activate the expression of the murine Mrp3 gene. This mechanism most likely contributes to the protection of colon cells from the toxic effects of LCA, but given the extremely low levels of VDR expression in the liver, is not likely to account for the adaptive increase in hepatic Mrp3 expression during cholestasis. These data indicate that therapeutic activation of the VDR may protect against the toxicity and tumor promoting effects of bile acids within the colon through the coordinate activation of the Mrp3, Cyp3a, and Sult2a genes.


   FOOTNOTES
 
* This work was supported by a grant from the Canadian Institutes for Health Research. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} To whom correspondence should be addressed: Dept. of Pharmacology, Dalhousie University, 5850 College St., Halifax, Nova Scotia B3H 1X5, Canada. Tel.: 902-494-2347; Fax: 902-494-1388; E-mail: csinal{at}dal.ca.

1 The abbreviations used are: ABC, ATP-binding cassette; CA, cholic acid; CAR, constitutive androstane receptor; CDCA, chenodeoxycholic acid; CYP3A, cytochrome P450 3A; DCA, deoxycholic acid; dKLCA, diketolithocholic acid; DR3, direct repeat-3; EMSA, electromobility shift assay; FXR, farnesoid X receptor; KLCA, ketolithocholic acid; LCA, lithocholic acid; MRP3, multidrug resistance-associated protein 3; NHR, nuclear hormone receptor; OC, rat osteocalcin gene; PPAR{alpha}, peroxisome proliferator-activated receptor-{alpha}; PXR, pregnane X receptor; RA, 9-cis-retinoic acid; RACE, rapid amplification of cDNA ends; RXR, retinoid X receptor; Sult2a2, dehydroepiandrosterone-preferring sulfotransferase; VD3, 1{alpha},25-dihydroxyvitamin D3; VDR, vitamin D receptor; VDRE, vitamin D receptor response element; GFP, green fluorescent protein; siRNA, small interfering RNA; hGAPDH, human glyceraldehyde-3-phosphate dehydrogenase; CABD9K, calbindin-D9K. Back



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Chang, G. (2003) FEBS Lett. 555, 102-105[CrossRef][Medline] [Order article via Infotrieve]
  2. Schmitt, L., and Tampe, R. (2002) Curr. Opin. Struct. Biol. 12, 754-760[CrossRef][Medline] [Order article via Infotrieve]
  3. Ambudkar, S. V., Kimchi-Sarfaty, C., Sauna, Z. E., and Gottesman, M. M. (2003) Oncogene 22, 7468-7485[CrossRef][Medline] [Order article via Infotrieve]
  4. Brousseau, M. E. (2003) Curr. Opin. Lipidol. 14, 35-40[CrossRef][Medline] [Order article via Infotrieve]
  5. Tomer, G., and Shneider, B. L. (2003) Gastroenterol. Clin. North Am. 32, 839-855, vi[CrossRef][Medline] [Order article via Infotrieve]
  6. Oram, J. F. (2002) Trends Mol. Med. 8, 168-173[CrossRef][Medline] [Order article via Infotrieve]
  7. Lewis, M. J., Lewis, E. H., 3rd, Amos, J. A., and Tsongalis, G. J. (2003) Am. J. Clin. Pathol. 120, (suppl.) S3-S13[Medline] [Order article via Infotrieve]
  8. Kruh, G. D., and Belinsky, M. G. (2003) Oncogene 22, 7537-7552[CrossRef][Medline] [Order article via Infotrieve]
  9. Hirohashi, T., Suzuki, H., and Sugiyama, Y. (1999) J. Biol. Chem. 274, 15181-15185[Abstract/Free Full Text]
  10. Hirohashi, T., Suzuki, H., Takikawa, H., and Sugiyama, Y. (2000) J. Biol. Chem. 275, 2905-2910[Abstract/Free Full Text]
  11. Kool, M., van der Linden, M., de Haas, M., Scheffer, G. L., de Vree, J. M., Smith, A. J., Jansen, G., Peters, G. J., Ponne, N., Scheper, R. J., Elferink, R. P., Baas, F., and Borst, P. (1999) Proc. Natl. Acad. Sci. U. S. A. 96, 6914-6919[Abstract/Free Full Text]
  12. Hofmann, A. F. (1999) Arch. Intern. Med. 159, 2647-2658[Abstract/Free Full Text]
  13. Wang, R., Salem, M., Yousef, I. M., Tuchweber, B., Lam, P., Childs, S. J., Helgason, C. D., Ackerley, C., Phillips, M. J., and Ling, V. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 2011-2016[Abstract/Free Full Text]
  14. Konig, J., Rost, D., Cui, Y., and Keppler, D. (1999) Hepatology 29, 1156-1163[CrossRef][Medline] [Order article via Infotrieve]
  15. Soroka, C. J., Lee, J. M., Azzaroli, F., and Boyer, J. L. (2001) Hepatology 33, 783-791[CrossRef][Medline] [Order article via Infotrieve]
  16. Scheffer, G. L., Kool, M., de Haas, M., de Vree, J. M., Pijnenborg, A. C., Bosman, D. K., Elferink, R. P., van der Valk, P., Borst, P., and Scheper, R. J. (2002) Lab. Investig. 82, 193-201[Medline] [Order article via Infotrieve]
  17. Ros, J. E., Libbrecht, L., Geuken, M., Jansen, P. L., and Roskams, T. A. (2003) J. Pathol. 200, 553-560[CrossRef][Medline] [Order article via Infotrieve]
  18. Inokuchi, A., Hinoshita, E., Iwamoto, Y., Kohno, K., Kuwano, M., and Uchiumi, T. (2001) J. Biol. Chem. 276, 46822-46829[Abstract/Free Full Text]
  19. Bohan, A., Chen, W. S., Denson, L. A., Held, M. A., and Boyer, J. L. (2003) J. Biol. Chem. 278, 36688-366