VPAC1 expression is regulated by FXR agonists in the human gallbladder epithelium

Authors


  • Potential conflict of interest: Nothing to report.

Abstract

Vasoactive intestinal peptide receptor-1 (VPAC1) is the high-affinity receptor of vasoactive intestinal peptide (VIP), a major regulator of bile secretion. To better define the level at which VPAC1 stimulates bile secretion, we examined its expression in the different cell types participating in bile formation (i.e., hepatocytes, bile duct, and gallbladder epithelial cells). Because VPAC1 expression was previously shown to be regulated by nuclear receptors, we tested the hypothesis that it may be regulated by the farnesoid X receptor (FXR). Quantitative RT-PCR and immunoblot analyses of cell isolates indicated that VPAC1 is expressed in all three cell types lining the human biliary tree, with predominant expression in the gallbladder. In primary cultures of human gallbladder epithelial cells, VIP induced cAMP production and chloride secretion. Analysis of the VPAC1 gene revealed the presence of potential FXR response element sequences, and both FXR and RXRα expressions were detected in gallbladder epithelial cells. In these cells, the FXR pharmacological agonist GW4064 upregulated VPAC1 expression in a dose-dependent manner, and this effect was antagonized by the RXRα ligand, 9-cis retinoic acid. Chenodeoxycholate activated endogenous FXR in gallbladder epithelial cells, as ascertained by electromobility shift assay and upregulation of the FXR target gene, small heterodimer partner. Chenodeoxycholate also provoked an increase in VPAC1 mRNA and protein content in these cells. In conclusion, FXR agonists may increase gallbladder fluid secretion through transcriptional activation of VPAC1, which may contribute to the regulation of bile secretion by bile salts and to a protective effect of FXR pharmacological agonists in gallstone disease. (HEPATOLOGY 2005.)

The regulation of bile secretion occurs at different levels of the biliary tree. Bile is formed primarily in hepatocyte canaculi by an osmotic process resulting from active bile salt secretion.1 In bile ducts and in the gallbladder, bile is then modified by the absorption or secretion of water and ions.2, 3 Thus, bile delivered to the intestine results from vectorial transport occurring in the different epithelial cell types lining the biliary tree, i.e., hepatocytes, intrahepatic bile duct, and gallbladder epithelial cells. Among regulatory peptides, the vasoactive intestinal peptide (VIP) induces bile secretion by stimulating transport activities both in hepatocytes and in biliary epithelial cells. In hepatocytes, VIP induces an increase in bile salt–dependent bile secretion.4 In bile ducts, VIP stimulates a bicarbonate-rich secretion with higher potency than all other secretagogues, including secretin.4 In the gallbladder, VIP is also a potent inducer of fluid and anion secretion.5, 6 In most systems, VIP effects are mediated by cyclic adenosine monophosphate (cAMP) production after the activation of high- or low-affinity receptors, namely, the vasoactive intestinal peptide receptor-1 (VPAC1) and the vasoactive intestinal peptide receptor-2 (VPAC2), respectively.7, 8 Among these two types of VIP receptors, only VPAC1 is expressed in the liver.7

In cancer cell lines, VPAC1 gene transcription is regulated by members of the nuclear receptor superfamily.9–11 The nuclear receptor superfamily comprises the farnesoid X receptor (FXR) for which bile salts serve as natural ligands.12 Ligand-bound FXR regulates gene transcription on heterodimerization with the retinoid X receptor alpha (RXRα), a permissive nuclear receptor activated by 9-cis retinoic acid (9-cis Ra).13 The FXR/RXRα heterodimer controls different aspects of bile salt synthesis and transport in hepatocytes and in enterocytes,14–16 in part through transcriptional activation of the short heterodimer partner (SHP).15, 17, 18 In animal models, the synthetic FXR agonist, GW4064,19 causes significant protection against cholestatic liver injury through the regulation of genes ensuring bile formation in hepatocytes.20 Although no FXR expression has been reported in bile ducts, bile salts are able to induce ductal secretion in rats through increased expression of the secretin receptor.21 We have previously shown that bile salts are able to regulate secretory activities in human gallbladder epithelial cells.22, 23 However, whether bile salts regulate transcriptional activities in human biliary epithelial cells remains to be determined.

We first examined VPAC1 expression in the different cell types lining the human biliary tree, i.e., hepatocytes, intrahepatic bile duct, and gallbladder epithelial cells. We hypothesized that FXR-mediated regulation could occur in biliary epithelial cells and that VPAC1 gene could be a target of this regulation. Because VPAC1 highest expression was detected in the gallbladder, we examined the effect of FXR agonists on VPAC1 expression in primary cultures of human gallbladder epithelial cells. The results suggest that bile salts may bind to nuclear receptors in the human gallbladder epithelium and modulate the expression of VPAC1, a major regulator of secretory functions in this epithelium.

Abbreviations:

VIP, vasoactive intestinal peptide; cAMP, cyclic adenosine monophosphate; VPAC1, vasoactive intestinal peptide receptor-1; VPAC2, vasoactive intestinal peptide receptor-2; FXR, farnesoid X receptor; RXRα, retinoid X receptor; 9-cis Ra, 9-cis retinoic acid; SHP, small heterodimer partner; DMEM, Dulbecco's modified Eagle medium; CDC, chenodeoxycholic acid; DTT, dithiothreitol; NTCP, Na+-taurocholate cotransporting polypeptide; CFTR, cystic fibrosis transmembrane conductance regulator; HSR, human secretin receptor; RT-PCR, reverse transcription polymerase chain reaction; EMSA, electrophoretic mobility shift assay; IR-1, bile salt response element sequences.

Materials and Methods

Reagents.

Dulbecco's modified Eagle medium (DMEM)/Ham's F12 (1:1) mixture was purchased from Life Technologies (Cergy Pontoise, France), Ultroser G, from Biosepra (Villeneuve-la-Garenne, France), and human type IV collagen, from Tebu (Le Perray-en-Yvelines, France). Protease type XIV from Streptomyces griseus, 9-cis retinoic acid, and VIP were provided by Sigma (Saint-Quentin Fallavier, France). Collagenase type D was purchased from Boehringer Mannheim (Mannheim, Germany). Chenodeoxycholate (CDC) (99 % pure) was obtained from Calbiochem (Meudon, France). Ribonuclease inhibitor RNazine was purchased from Promega (Charbonnières, France); Moloney murine leukemia virus reverse transcriptase, from Life Technologies; and Taq DNA polymerase, from Perkin-Elmer (Les Ulis, France).

Cell Isolation and Culture.

The procedure to obtain human samples was in accordance with current French legislation.

Hepatocytes and intrahepatic bile duct preparations were isolated from human liver samples with minor modifications to previously described protocols.24, 25 Briefly, human liver samples (15-60 g) were perfused with 0.5 mmol/L EDTA, and with collagenase type D (0.8 UI/mL) at 37°C. After hepatocyte isolation, the biliary tree was incubated thrice in collagenase type D (3 UI/mL) supplemented with 0.1 % (wt/vol) protease type XIV at 37°C for 30 minutes. We have previously shown that more than 90% of cells forming bile duct segments isolated by this technique are biliary epithelial cells as ascertained by cytokeratin 19 immunolabeling and by gamma glutamyltransferase (γ-GT) staining.25 Isolated hepatocytes and intrahepatic bile ducts were stored at −80°C until use.

Gallbladder samples displayed no significant histological abnormality. Epithelial cells were isolated by incubation in 0.075% (wt/vol) protease type XIV for 12 hours at 4°C, as described.26 Cells were either fresh frozen at −80°C or suspended in DMEM/Ham's F12 containing 1 mmol/L Ca2+, supplemented with 2% (wt/vol) Ultroser G, 7 g/L D-glucose, 14 mmol/L NaHCO3, and 200,000 IU 200 mg/L penicillin G-streptomycin, pH 7.4, and plated in 12-well culture dishes coated with human type IV collagen. The cells were incubated under air/CO2 95/5, at 37°C. The culture medium was renewed every 48 hours.

Reverse Transcription Polymerase Chain Reaction.

Total RNA was extracted from freshly isolated and cultured cells treated or not with GW4064 (10 μmol/L) (a pharmacological FXR agonist provided by Stacey Jones, GlaxoSmithKline, Research Triangle Park, NC), CDC (50 μmol/L), 9-cis Ra (1 μmol/L), GW4064 with 9-cis Ra, or CDC with 9-cis Ra for 21 hours, using RNA plus lysis solution (Quantum, Montreuil-sous-Bois, France). One microgram total RNA was denatured by heating at 72°C for 10 minutes, and then incubated in 25 μL reaction buffer containing 10 mmol/L dithiothreitol (DTT), 0.5 mmol/L deoxyribonucleotide triphosphate (dNTP), 20 U RNazine, 5 μmol/L random hexamers, and 200 U Moloney murine leukemia virus reverse transcriptase. Reverse transcription was allowed to proceed for 1 hour at 37°C. Amplifications were achieved using 1.25 U Taq DNA polymerase and 0.5 μmol/L selected primers designed to amplify Na+-taurocholate cotransport protein (NTCP), cystic fibrosis transmembrane conductance regulator (CFTR), anion exchanger 2 (AE2), human secretin receptor (HSR), VPAC1, FXR, RXRα, SHP, and β-actin27 (Table 1). Polymerase chain reaction (PCR) products after completion of 38 cycles were separated by electrophoresis through a 2% agarose gel stained with ethidium bromide. The authenticity of PCR products was verified by enzymatic restriction. VPAC1 PCR products were purified and calibrated as a number of copies to be used as a scale for quantification in real-time PCR experiments. Quantitative real-time PCR was performed using the Sybr Green PCR Core Reagents Kit (PE Applied Biosystems, Foster City, CA) on a PE Applied Biosystems 7700 Sequence Detector. VPAC1 primers were designed according to published human cDNA sequences 5′ GAT GAC AAG GCA GCG AGT TTG 3′ (VPAC1-ts) and 5′ AGA AGG GTG GCG AGG GAC 3′ (VPAC1-tas) that amplify a 101-bp cDNA fragment (nucleotides 296-396) of VPAC1 (Genbank database, accession no. NM_004624). The 18S primers were the following: 5′ GAG CGA AAG CAT TTG CCA AG 3′ (sense) and 5′ GGC ATC GTT TAT GGT CGG AA 3′ (antisense) (Genbank database, accession no. NG_002801). Quantitative real-time PCR reactions were run with 200 nmol/L VPAC1-ts and VPAC1-tas primers and 50 nmol/L of each 18S sense and antisense primers. A scale of calibrated total RNA (PE Applied Biosystems) was reverse transcribed and used as a reference for 18S amplification. The amplification conditions on the PE Applied Biosystems 7700 sequence detector were the following: 2 minutes at 50°C, 10 minutes at 95°C, 40 cycles of 95°C for 15 seconds, and 60°C for 1 minute. Data were collected and analyzed with Sequence Detector v1.7 software (PE Biosystems). Data were expressed as a ratio of the number of copy of VPAC1 mRNA by μg RNA.

Table 1. Sequences of PCR Primers
GeneForward Primer (5′–3′)Reverse Primer (5′–3′)Genebank Database Accession No.
NTCPTCC TGG TTC TCA TTC CTT GCTAT GGC AAT GAG GAG AAG CCL21893
CFTRACT GGA GCA GGC AAG ACT TCACAG TGT GAT TCC ACC TTC TCNM_000492
AE2AGA TGC TCA AGA AGC GAG AGGGTG GTT GCT GCT ACA GAA CGAU62290
HSRACA TCC ACA TGC ACC TGT TCAG AAA GTG TCT GGC AAT AGNM_002980
VPAC1AGT GTG ACT ATG TGC AGA TGAGG TGC ATG TGG ATG TAG TTNM_004624
FXRAAC AGA ACA AGT GGC AGG ATCGCA TTC AGC CAA CAT TCC CANM_005123
RXRαATC CCA CAC TTC TCA GAG CTCAG CAT CTC CAT AAG GAA GGX52773
SHPCTC AAG AAG ATT CTG CTG GAAAA GAA GAG GTC CCC CAA GCAAB058644
β-ActinATC ATG TTT GAG ACC TTC AATTG CGC TCA GGA GGA GCA ATE00829

cAMP Assay.

Gallbladder epithelial cells at days 5 to 6 of primary culture were incubated for 10 minutes at 37°C, with increasing concentrations of VIP (10-13 mol/L to 10-7 mol/L) in DMEM/Ham's F12. At the end of the experiment, ice-cold DMEM/Ham's F12 containing digitonin (40 mmol/L) and 3-Isobutyl-1-methylxanthine (IBMX) (1 mmol/L) was added. Cyclic AMP was assayed in the supernatant by a commercial RIA (NEN Life Science Products, Paris, France). Protein content in cell samples was determined by BCA-protein assay (Pierce).

Chloride Efflux Assay.

Chloride efflux was measured as described.28 Gallbladder epithelial cells at days 5 to 6 of primary culture were loaded with 36Cl (5 μCi/mL) in efflux buffer containing 140 mmol/L NaCl, 4 mmol/L KCl, 1 mmol/L KH2PO4, 2 mmol/L MgCl2, 1 mmol/L CaCl2, 10 mmol/L glucose, and 10 mmol/L HEPES, pH 7.4, for 1 hour at 37°C. The cells were washed rapidly thrice with 1 mL isotope-free buffer, which was then replaced at 2-minute intervals, before VIP (10-7 mol/L) was added in 1 mL efflux buffer. At the end of the experiment, cells were solubilized in 1 mol/L NaOH, and samples were counted for radioactivity. The efflux was calculated as the ratio of radioactivity in the efflux sample at a given time to the total radioactivity present in the cells during the previous 2-minute interval.

Electrophoretic Mobility Shift Assay.

Nuclear extracts were prepared as described by Bertrand et al.29 Briefly, gallbladder epithelial cells at days 5 to 6 of primary culture incubated for 21 hours at 37°C with CDC (50 μmol/L), 9-cis Ra (1 μmol/L), or a combination of CDC with 9-cis Ra. Cells were lysed in buffer A (Hepes 10 mmol/L, MgCl2 1.5 mmol/L, KCl 10 mmol/L, DTT 0.5 mmol/L, and protease inhibitor cocktail, pH 7.9) supplemented with Triton X-100 0.1 % (v/v). After homogenization in a tight-fitting Dounce homogenizer, cell lysates were maintained 10 minutes on ice and centrifuged at 2,000g for 5 minutes. The nuclear pellet was resuspended in buffer B (Hepes 20 mmol/L, MgCl2 1.5 mmol/L, NaCl 420 mmol/L, EDTA 0.2 mmol/L, DTT 0.5 mmol/L, glycerol 25% (p/v), protease inhibitor cocktail, pH 7,9). After 2 hours at 4°C under constant agitation, nuclear debris were centrifuged at 13,000g for 5 minutes, and the supernatant (nuclear extract) was stored at −80°C until analysis by electrophoretic mobility shift assay (EMSA). The double-stranded FXR probe: 5′ AAG GTC AAT GAC CTT 3′; 3′ AAG GTC ATT GAC CTT 5′30 was annealed and end-labeled by using the DNA polymerase 1 Klenow fragment in the presence of [α-32P]adenosine triphosphate in a reaction mixture containing 10 mmol/L Tris-HCl, pH 7.9, 10 mmol/L MgCl, 1 mmol/L DTT, and 1 mmol/L EDTA. Unincorporated nucleotides were removed by filtration through a column. Binding reactions were carried out in binding reaction mixture (Hepes 20 mmol/L [pH 7.5], KCl 50 mmol/L, MgCl2 1.5 mmol/L, EDTA 2 mmol/L, DTT 1 mmol/L, 5% glycerol, 2 μg poly(dI-dC) and 1% nonidet NP-40) containing 5 μg nuclear proteins and the FXR probe (80,000 counts/min). Samples were fractionated by electrophoresis on a 5% non-denaturing polyacrylamide gel in TBE buffer. After electrophoresis, gels were dried and exposed to radiosensitive films.

Immunoblotting.

Gallbladder epithelial cells at days 5 to 6 of primary culture were incubated with either CDC (50 μmol/L), 9-cis Ra (1 μmol/L) or CDC with 9-cis Ra for 21 hours. Cells were lysed in a buffer composed of 50 mmol/L Tris (pH 7.4), 150 mmol/L NaCl, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, and 1% NP-40. Protein extracts (25 μg) were subjected to electrophoresis through a 10% sodium dodecyl sulfate polyacrylamide gel, and then transferred to nitrocellulose membranes. Immunoblotting was performed using an affinity purified goat polyclonal antibody raised against a peptide mapping near the amino terminus of SHP (Santa Cruz Biotechnology, Le Perray en Yvelines, France), a mouse monoclonal antibody raised against a recombinant VPAC1 protein (Exalpha Biologicals Inc., Boston, MA) and a mouse monoclonal antibody raised against the amino terminus of β-actin (Sigma). Immunoreactivity was revealed by enhanced chemiluminescence using an ECL kit (Amersham, Les Ulis, France).

Statistical Analysis.

Comparisons were made using the Student t test and the paired Wilcoxon signed rank test. A P value of less than .05 was considered significant.

Results

Expression of VPAC1 Gene Along the Human Biliary Tree.

To better define the contribution of VPAC1 to the regulation of bile formation, we examined its expression along with that of other genes involved in bile secretion (hepatobiliary transporters and HSR) at different levels of the human biliary tree. Gene expressions were analyzed by reverse transcription (RT)-PCR in freshly isolated hepatocytes, intrahepatic bile duct, and gallbladder epithelial cells issued from the same donor. As expected, Na+-taurocholate cotransporting polypeptide (NTCP) expression was detected only in hepatocytes. CFTR expression was detected in both bile duct and gallbladder epithelial cells and was absent from hepatocytes. AE2 transcripts were detected in all three cell types (Fig. 1A). HSR expression was restricted to bile duct epithelial cells, whereas VPAC1 expression was detected in all three cell types (Fig. 1A). As shown in Fig. 1B, VPAC1 transcripts, as determined by quantitative real-time RT-PCR, were more abundant in bile duct epithelial cells (570 ± 341 copies/μg RNA) than in hepatocytes (112 ± 75 copies/μg RNA). They were yet more abundant in gallbladder epithelial cells (3,337 ± 731 copies/μg RNA) than in bile duct epithelial cells. The differential expression of VPAC1 in the three cell types was confirmed at the protein level by immunoblot analyses (Fig. 1C). Together with previous physiological studies,31, 32 these results indicate that VIP-induced regulations of bile secretion mainly take place in bile duct and gallbladder epithelial cells.

Figure 1.

VPAC1 expression in cell types of the human biliary tree. (A) Total RNA from human hepatocytes (lane 1), intrahepatic bile duct epithelial cells (lane 2), and gallbladder epithelial cells (lane 3) issued from the same subject was submitted to RT-PCR with primers designated to amplify NTCP, CFTR, AE2, HSR, VPAC1, or β-actin. Representative gels of cell preparations obtained from 3 different subjects are shown. (B) Quantification of VPAC1 mRNA by real-time RT-PCR in human hepatocytes (column 1), intrahepatic bile duct epithelial cells (column 2), and gallbladder epithelial cells (column 3). Data are expressed in copy number per μg RNA and represent means ± SEM of 3 experiments performed in duplicate. (C) Proteins from human hepatocytes (lane 1), intrahepatic bile duct epithelial cells (lane 2), and gallbladder epithelial cells (lane 3) issued from the same subject were submitted to VPAC1 immunoblot–ECL analysis. Representative gels of cell preparations obtained from three different subjects are shown. VPAC1, vasoactive intestinal peptide receptor-1; RT-PCR, reverse transcription polymerase chain reaction; NTCP, Na+-taurocholate cotransporting polypeptide; CFTR, cystic fibrosis transmembrane conductance regulator; AE2, anion exchanger 2; HSR, human secretin receptor.

Signaling of Ligand-Activated VPAC1 in Gallbladder Epithelial Cells.

Whether VIP regulation is mediated by a cAMP pathway in biliary epithelial cell is an issue of conflicting results.4, 31 To address this question, we measured cAMP production and chloride secretion in primary cultures of human gallbladder epithelial cells exposed to VIP. In gallbladder epithelial cells, VIP elicited a dose-dependent increase in cAMP content (Fig. 2A) and caused a significant rise in chloride secretion (Fig. 2B). No expression of the low-affinity VIP receptor, VPAC2, could be detected by RT-PCR in gallbladder epithelial cells (data not shown), providing evidence that VPAC1, coupled to a cAMP-dependent pathway, mediates VIP-induced secretion in gallbladder epithelial cells.

Figure 2.

Effect of VIP on human gallbladder epithelial cells. (A) Gallbladder epithelial cells were incubated with increasing VIP concentrations (10-13 mol/L to 10-7 mol/L). Cyclic AMP content was measured by RIA and was reported to cell protein content. Data represent means ± SEM of at least three experiments performed in duplicate. *P < .05 versus condition without VIP. (B) Gallbladder epithelial cells were loaded with 36Cl. After washing, 1 mL efflux buffer was added and removed at 2-minute intervals for counting. Cells were exposed at 6 minutes (arrow) to VIP (0.1 μmol/L). Chloride secretion was significantly induced by VIP (P < .05, n = 5). The figure shows the results of a typical experiment. VIP, vasoactive intestinal peptide; cAMP, cyclic adenosine monophosphate.

Regulation of VPAC1 Gene Expression by GW4064 in Gallbladder Epithelial Cells.

Subsequent investigations were aimed to test the potential of the bile salt nuclear receptor, FXR, to control VPAC1 gene expression in gallbladder epithelial cells. Analysis of the human VPAC1 receptor gene33 showed that this gene contains potential bile salt response element sequences (IR-1) within the flanking region located upstream of the coding sequence (Fig. 3A). As shown in Fig. 3B, these IR-1–like sequences of the VPAC1 gene share homology with previously described IR-1 consensus sequence (IR-1 con) and IR-1 sequences of various FXR target genes.16–18, 34–36 These findings indicate that VPAC1 gene is a potential FXR target gene. Furthermore, as observed in hepatocytes, expression of FXR, of RXRα and of SHP, a classical FXR target gene,17, 18 were detected in intrahepatic bile duct and gallbladder epithelial cells (Fig. 4). These observations suggest that FXR activation can occur and may target VPAC1 gene in biliary epithelial cells.

Figure 3.

IR-1–like sequence in the VPAC1 gene. (A) The nucleotide sequences corresponding to putative FXR/RXRα binding site at −919, +33, and +156 in the human VPAC1 gene33 are shown. Numbering starts from transcription start site. (B) Alignement of IR-1 consensus sequences from known FXR target genes16–18, 34–36 with putative VPAC1 IR-1–like sequence (−919 to −906). Nucleotides shared between IR-1 sequences are shown in upper case boldface type. IR-1, bile salt response element sequences; VPAC1, vasoactive intestinal peptide receptor-1; FXR, farnesoid X receptor; RXRα, retinoid X receptor alpha.

Figure 4.

Expression of FXR, RXRα, and SHP in cell types of the human biliary tree. Total RNA from human hepatocytes (lane 1), bile duct epithelial cells (lane 2), and gallbladder epithelial cells (lane 3) issued from the same subject was submitted to RT-PCR with primers designated to amplify FXR, RXRα, and SHP. Representative gels of cell preparations obtained from 3 different subjects are shown. FXR, farnesoid X receptor; RXRα, retinoid X receptor alpha; SHP, small heterodimer partner; RT-PCR, reverse transcription polymerase chain reaction.

To further examine this possibility, gallbladder epithelial cells in primary culture were incubated with increasing concentrations of the synthetic FXR ligand, GW 4064. As shown in Fig. 5A, VPAC1 gene expression assessed by real-time RT-PCR increased in a dose-dependent manner in response to GW4064. Previous studies have shown that RXRα activation by 9-cis Ra was able to modulate FXR-induced gene regulation.34, 37–39 To determine the effect of RXRα activation on FXR-mediated VPAC1 gene regulation, we compared VPAC1 expression levels in cells exposed to GW4064 with or without 9-cis Ra. As reported for the regulation of the bile salt export pump gene in hepatocytes,39 the addition of 9-cis Ra prevented the induction of VPAC1 gene expression by GW4064 (103 ± 6 vs. 179 ± 29% of control; P < .05) (Fig. 5B). These findings indicate that in gallbladder epithelial cells, FXR regulates VPAC1 gene expression and that this transcriptional effect of FXR is antagonized by RXR agonists.

Figure 5.

Effect of FXR agonists on VPAC1 mRNA expression. Gallbladder epithelial cells were treated with (A) increasing concentrations of GW4064 (1-20 μmol/L) and (B) GW 4064 (10 μmol/L) or GW 4064 in combination with 9-cis Ra (1 μmol/L). Total RNA was subjected to real-time RT-PCR with primers designated to amplify VPAC1. Data, expressed in percentage of control, represent means ± SEM of 3 experiments performed in duplicate. *P < .05 versus control. **P < .05 versus condition without 9-cis Ra. FXR, farnesoid X receptor; VPAC1, vasoactive intestinal peptide receptor-1; 9-cis Ra, 9-cis retinoic acid; RT-PCR, reverse transcription polymerase chain reaction.

Effects of CDC on FXR Activation and VPAC1 Gene Expression in Gallbladder Epithelial Cells.

CDC, which was previously identified as the principal FXR activator in bile,40 was selected as a natural test ligand to ascertain the effect of FXR activation by bile salts on VPAC1 expression.

The ability of CDC to activate endogenous FXR in gallbladder epithelial cells was assessed by EMSA experiments using a consensus FXR binding sequence. Nuclear extracts from untreated gallbladder epithelial cells induced a significant band shift (Fig. 6A). As previously observed in FXR-expressing tissue,41 this result indicates a constitutive binding activity of FXR in gallbladder epithelial cells. Furthermore, in the presence of CDC, the formation of this DNA–protein complex was increased. A similar effect was observed when 9-cis Ra was added together with CDC. Protein expression levels of SHP were used to ascertain that FXR activation was functional in gallbladder epithelial cells. Consistent with EMSA experiments, SHP expression was increased in cells exposed to CDC alone or in combination with 9-cis Ra (Fig. 6B).

Figure 6.

Activation of FXR by CDC. (A) Gallbladder epithelial cells were treated with CDC (50 μmol/L) alone or in combination with 9-cis Ra (1 μmol/L). Nuclear extracts were incubated with an FXR-specific probe and subjected to electrophoretic mobility shift assay. Representative gel obtained from 3 different experiments. (B) Gallbladder epithelial cells were treated with CDC (50 μmol/L) alone or in combination with 9-cis Ra (1 μmol/L). Proteins were subjected to SHP immunoblot-ECL analysis. Representative gels obtained from 3 different experiments. FXR, farnesoid X receptor; CDC, chenodeoxycholic acid; 9-cis Ra, 9-cis retinoic acid; SHP, small heterodimer partner.

We next addressed the effect of CDC on VPAC1 expression. As shown in Fig. 7A, CDC induced a significant increase in VPAC1 transcript levels (P < .05). Other major bile salts of human bile, such as cholate, deoxycholate, as well as the glyco and tauroconjugates of CDC, also showed ability to increase VPAC1 expression (152 ± 26; 113 ± 4; 111 ± 5; 112 ± 6 in percentage of control, n = 3, respectively). As observed with GW4064, the addition of 9-cis Ra prevented the induction of CDC-elicited VPAC1 expression (Fig. 7A). Moreover, CDC-induced expression of the VPAC1 gene resulted in increased protein levels, and this effect was reduced by the addition of 9-cis Ra (Fig. 7B). We conclude from these results that in gallbladder epithelial cells, bile salts are able to control VPAC1 expression through an FXR-mediated regulation that can be antagonized by RXRα agonists.

Figure 7.

Effect of CDC on VPAC1 mRNA expression. (A) Gallbladder epithelial cells were treated with CDC (50 μmol/L) alone or in combination with 9-cis Ra (1 μmol/L). Total RNA was subjected to real-time RT-PCR with primers designated to amplify VPAC1. Data, expressed in percentage of control, represent means ± SEM of 3 experiments performed in duplicate. *P < .05 versus control. **P < .05 versus condition without 9-cis Ra. (B) Gallbladder epithelial cells were treated with CDC (50 μmol/L) alone or in combination with 9-cis Ra (1 μmol/L). Proteins were subjected to VPAC1 immunoblot-ECL analysis. Representative gels obtained from 3 different experiments. CDC, chenodeoxyxholic acid; VPAC1, vasoactive intestinal peptide receptor-1; 9-cis Ra, 9-cis retinoic acid; RT-PCR, reverse transcription polymerase chain reaction.

Discussion

This study shows that VPAC1 is expressed in all major cell types participating in bile formation. VPAC1 displays a gradient of expression along the human biliary tree, the gallbladder showing the highest level of expression. Whereas bile duct epithelial cells express both VPAC1 and HSR, we found that gallbladder epithelial cells express only VPAC1. In primary cultures of gallbladder epithelial cells, VIP elicited both cAMP production and chloride secretion. In line with previous studies,4–6, 31, 32 these observations suggest that VIP through VPAC1 activation is a major regulator of ductular secretion and of cAMP-dependent hydroelectrolytic secretion in the gallbladder.

We have previously shown that, in human gallbladder epithelial cells, bile salts stimulate chloride secretion through the activation of calcium-regulated channels and through the potentiation of receptor-induced adenylyl cyclase activity, which leads to enhanced activation of cAMP-regulated channels.22, 23 Bile salts have also been shown to increase ductal secretion in rats through the transcriptional control of cAMP-coupled receptor.21 Thus, we hypothesized that bile salts may have regulatory effects on VPAC1 expression in gallbladder epithelial cells. Analysis of the human VPAC1 gene showed the presence of potential FXR response elements in the 5′ end and promoter region of the gene. Furthermore, we found that gallbladder epithelial cells, as well as bile duct epithelial cells, express both FXR and RXRα. In primary cultures of gallbladder epithelial cells, pharmacological activation of FXR by GW4064 induced a significant increase in VPAC1 transcripts. In physiological settings, FXR activation mainly arises from bile salts present in bile. Bile salt concentrations in human bile ducts42 and in the gallbladder43 are higher than the half-maximal effective concentration required for bile salt activation of FXR, 50 μmol/L.14 Thus, bile salts have the potential to induce VPAC1 transcriptional regulations through FXR both in bile duct and gallbladder epithelial cells. Human bile is mostly composed of chenodeoxycholate, cholate, and deoxycholate (38%, 36%, and 16% of the total bile salt pool, respectively), whereas other bile salts, such as ursodeoxycholate, lithocholate, and sulfolithocholate, are present in trace amounts.44 Even though all major bile salts have the ability to activate FXR in vitro, as shown here and in previous studies,12, 14 CDC has been identified by sequential purification as the principal FXR activator present in bile.40 In gallbladder epithelial cells, the ability of CDC to activate endogenous FXR was demonstrated by gel-shift experiments and by an increase in SHP protein expression. We also show that in these cells CDC regulates VPAC1 expression both at the transcript and protein level. As reported for the bile salt export pump gene,39 the induction of VPAC1 expression by FXR was antagonized by concomitant activation of RXRα by 9-cis Ra. Consistent with previous results showing that GW4064 was more potent than CDC in inducing gene expression in hepatocytes,45 we show that CDC induces VPAC1 gene expression to a lesser extent than GW4064.

In animal models of cholestasis, GW4064 provides protection against liver injury through FXR activation.20 Moreover, administration of GW4064 to mice fed a lithogenic diet prevents the development of cholesterol gallstone disease.46 These protective effects have been attributed to the transcriptional regulation of genes involved in the metabolic and transport functions of hepatocytes. Our findings suggest that therapeutic effects of FXR activation may also result from regulations occurring in bile duct and gallbladder epithelial cells. We postulate on the basis of the current data that in animal models of cholestasis, GW4064 treatment may modulate ductal secretion through the control of gene expression. As demonstrated in gallbladder epithelial cells, GW4064 may increase VPAC1 expression through FXR activation in bile duct epithelial cells and thus favor choleresis by enhancing ductal secretion. Furthermore, we anticipate that in cholesterol gallstone disease FXR-induced VPAC1 expression would enhance the stimulation of fluid secretion in gallbladder. The resulting increase in bile dilution is expected to reduce cholesterol supersaturation and crystallization.

In conclusion, we have shown that VPAC1 is a major receptor regulating secretory functions in the gallbladder and that VPAC1 expression in the gallbladder is upregulated by FXR activation. These results suggest that bile salts through FXR-regulated VPAC1 expression could favor VIP-induced bile delivery to the intestine after feeding. Moreover, they indicate that in pathological settings, such as in cholesterol gallstone disease, pharmacological agonists of FXR may exert preventive effects through increased bile dilution in the gallbladder.

Acknowledgements

The authors thank Stacey A Jones, Nuclear Receptor Functional Analysis, GlaxoSmithKline, Research Triangle Park, NC, for kindly providing the GW4064.

Ancillary