Organic solute transporter (OSTα-OSTβ) is a novel heteromeric bile acid and sterol transporter expressed at the basolateral membranes of epithelium in the ileum, kidney, and liver. To determine whether OSTα-OSTβ undergoes farnesoid X receptor (FXR)-dependent adaptive regulation following cholestatic liver injury, mRNA and protein expression levels were analyzed in patients with primary biliary cirrhosis (PBC) and following common bile duct ligation (CBDL) in rats and Fxr null and wild-type mice. Hepatic OSTα and OSTβ mRNA increased 3- and 32-fold, respectively, in patients with PBC compared with controls, whereas expression of Ostα and Ostβ also increased in the liver of rats and mice following CBDL. In contrast, expression of Ostα and Ostβ mRNA was generally lower in Fxr null mice, and CBDL failed to enhance expression of Ostα and Ostβ compared with wild-type mice. HepG2 cells treated for 24 h with chenodeoxycholic acid, a selective FXR ligand, had higher levels of OSTα and OSTβ mRNA and protein. Increases in OST protein were visualized by confocal microscopy at the plasma membrane. These results indicate that expression of Ostα and Ostβ are highly regulated in response to cholestasis and that this response is dependent on the FXR bile acid receptor.
- bile acid and steroid transporter
- primary biliary cirrhosis
- bile acid reabsorption
organic solute transporter (OSTα-OSTβ) is a heteromeric transporter expressed on the basolateral membrane of epithelial cells of human ileum, liver, and kidney that functions to extrude bile acids and other organic solutes and steroids from these and other tissues (2, 3, 7, 22, 29). OSTα-OSTβ is known to transport estrone 3-sulfate, dehydroepiandrosterone 3-sulfate, taurocholate, digoxin, and prostaglandin E2 (22, 29), as well as a variety of other bile acids (3). This transporter was originally cloned from the liver of an evolutionarily ancient vertebrate, the small skate Leucoraja erinacea (29), and was subsequently discovered to have orthologs in mammals, including humans (22). Recent studies (7) indicate that murine Ostα-Ostβ is a basolateral transporter that is highly expressed in the epithelial cells of the terminal ileum of the mouse and is responsible for the export of bile acids to mesenteric blood following their uptake by the apical sodium-dependent bile acid transporter Asbt (Slc10A2). Subsequent studies have demonstrated that both OSTα and OSTβ are also expressed in the kidney and in cholangiocytes in rodents and human, where bile acids are also taken up by ASBT (3). Although expression in the kidney and liver varies between species, localization of Ostα-Ostβ at the basolateral membrane of the mouse and rat proximal renal tubule and cholangiocyte also supports its role as a major bile acid efflux transporter in these cells (3).
It is now well established, based on studies in rodents as well as human cholestatic liver disease, that other bile acid transporters, including NTCP (SLC10A1), bile salt export pump (BSEP; ABCB11), and ASBT (SLC10A2), all undergo adaptive responses both to cholestatic liver injury and to bile acid feeding and that these responses can be interpreted as protecting the tissue from the toxic effects of bile acid accumulation (5, 26, 27). Furthermore, studies in common bile duct-ligated (CBDL) mice indicate that these changes in bile acid transporter expression are regulated by the bile acid-activated nuclear receptor farnesoid X receptor (FXR, NR1H4) (28, 32). Therefore, we hypothesized that the subunits of OST might also undergo adaptive regulation in response to cholestatic liver injury and be regulated by FXR. To examine this question, the expression of OSTα-OSTβ was examined in patients with cholestatic liver injury and after CBDL in mice and rats. In addition, to determine whether these adaptive responses depended on Fxr, CBDL was also carried out in Fxr−/− mice, and human hepatoma cells were treated with chenodeoxycholic acid (CDCA), a potent FXR ligand. Our findings indicate that cholestasis results in a significant adaptive response in OSTα-OSTβ in the liver of humans and rodents, although the levels of expression vary between subunits and species. Furthermore, the results indicate that these responses are critically dependent on the nuclear receptor FXR. These findings are consistent with the hypothesis that OSTα-OSTβ is a basolateral bile salt transporter that is likely to have a significant role in protecting the liver from the accumulation of bile salts during cholestatic liver injury.
C57/BL6 mice with targeted disruption of Fxr and their wild-type littermates were kindly provided by Dr. Frank Gonzales [National Institutes of Health (NIH)] (23). Male Sprague-Dawley rats (200–275 g body wt) were obtained from Charles River (Wilmington, MA), and C57/BL6 mice were purchased from Jackson Laboratories (Bar Harbor, ME). All animals were housed in a temperature- and humidity-controlled environment under a constant light cycle where they had free access to water and food. CBDL was performed under sterile conditions as previously described (11, 25). Briefly, the common bile duct was exposed, ligated twice close to the liver hilus immediately below the bifurcation, and then cut between the ligatures. Control animals underwent sham surgery in which the bile duct was exposed but not ligated. Tissue was harvested 3, 7, and/or 14 days after surgery. In some cases, livers were collagenase digested and hepatocytes were separated from nonparenchymal, bile duct-enriched tissue as previously described (21).
All experimental protocols were approved by the local Animal Care and Use Committee, according to criteria outlined in the “Guidelines for the Care and Use of Laboratory Animals” prepared by the National Academy of Sciences, as published by the NIH (publication 86–23, revised 1985).
Tissue specimens and patients characteristics.
Fifteen liver specimens comprising samples from patients with primary biliary cirrhosis (PBC; n = 11), and controls without liver disease (n = 4) were analyzed. PBC specimens were obtained during liver transplantation. The diagnosis of PBC was based on standard criteria (13, 14). Six patients had PBC stage III (PBC III), and five had stage IV (PBC IV) according to Ludwig (19). Most PBC patients (4 of 6 PBC III; 3 of 5 PBC IV) received standard ursodeoxycholic acid (UDCA) treatment. Normal liver tissue was obtained from patients undergoing resection of liver metastases. All of the specimens were analyzed previously (31). Liver tissue was immediately snap frozen and stored in liquid nitrogen until analysis. All patients had given their informed consent for the study, and the experimental protocol was approved by the local ethics committees in accordance with the ethical guidelines of the 1975 Declaration of Helsinki.
HepG2 cell experiments.
The human hepatocellular carcinoma cell line HepG2 was acquired from ATCC (Manassas, VA) and used without testing for mycoplasmal contamination. Cells were cultured in DMEM containing 10% FBS and 1% penicillin-streptomycin at 37°C with 5% CO2. When cells reached ∼70% confluence, they were washed and cultured in fresh, serum-depleted medium in the presence or absence of 50 μM CDCA (Sigma, St. Louis, MO). Twenty-four hours after CDCA addition, cells were collected by directly applying TRIzol (Invitrogen, Carlsbad, CA) to the culture dish for total RNA isolation and quantitative PCR was preformed as described below.
For immunoblot analyses, HepG2 cells were washed with PBS before being scraped in 1 ml of ice-cold homogenizing buffer containing 10 mM Tris·HCl (pH 7.4), 10 mM KCl, 1.5 mM MgCl2, 10 μl/ml protease inhibitor cocktail (Sigma), 2 mM PMSF, and 0.68 mM EDTA. Cells were homogenized with 35 strokes using type A pestle of a Dounce homogenizer while in ice. Sucrose was added to a final concentration of 250 mM, and the homogenate was centrifuged at 800 g for 20 min at 4°C. The supernatant was collected and stored at −80°C, and protein concentration was determined by the Lowry assay (18). Cell homogenates were dissolved in Pierce lane marker sample buffer (Pierce, Rockford, IL) supplemented with β-mercaptoethanol, incubated at 37°C for 30 min, brought to room temperature, and subjected to SDS-PAGE.
For immunofluorescence analysis, HepG2 cells were fixed with cold methanol 24 h after addition of 50 μM CDCA or 100 μM UDCA. Cells were incubated 20 min in blocking buffer (PBS, 1% BSA, 0.05% Triton X-100) followed by 2 h in antibody to OSTα and OSTβ diluted 1:150 in blocking buffer. After being washed 30–45 min in PBS-0.05% Triton X-100, Alexa 594 anti-rabbit IgG (Invitrogen) was incubated for 1 h at room temperature. Fluorescence was visualized with a Zeiss LSM510 (Carl Zeiss, Thornwood, NY) confocal microscope, and images were processed with Photoshop (Adobe, Mountainview, CA).
Total RNA was isolated from liver using TRIzol reagent according to manufacturer's protocol (Invitrogen). Total RNA (5 μg) was reverse transcribed using Pro-Star first-strand kit (Stratagene, La Jolla, CA) with oligo(dT) according to the supplier's protocol. Quantitative PCR was performed using Applied Biosystems 7500 DNA sequence detector system in a total volume of 20 μl with either Platinum SYBR green master mix, Platinum quantitative PCR supermix-UDG (Invitrogen), or TaqMan universal master mix (Applied Biosystems, Foster City, CA). Specific primer pairs and probes were designed (Table 1) or purchased (TaqMan gene expression assays, Applied Biosytems) and used for quantitative PCR analysis. The detection procedure was as follows: 50°C for 2 min, 95°C 10 min, then 40 cycles at 95°C for 15 s, followed by 60°C for 1 min. At the end of the PCR, a dissociation curve was set from 60 to 95°C to ensure the quantification/amplification of single PCR products if SYBR green was the detector. Relative target quantity was determined from the standard curve and normalized to either 18S ribosomal RNA or GAPDH expression.
Preparation of liver membranes.
Liver tissue was homogenized in ice-cold homogenizing buffer containing 10 mM Tris·HCl (pH 7.4), 10 mM KCl, 1.5 mM MgCl2, complete protease inhibitor cocktail (Roche, Indianapolis, IN), 2 mM PMSF, and 0.68 mM EDTA. Sucrose was added to a final concentration of 250 mM, and the homogenate was centrifuged at 800 g for 20 min at 4°C. The supernatant was recovered and further centrifuged at 100,000 g for 1 h to obtain a total membrane fraction.
Western blot analysis.
Similar amounts of protein (20–100 μg) were subjected to SDS-PAGE. Equal protein loading was confirmed by Coomassie blue staining of gels and Ponceau S staining of membranes. After electrotransfer to nitrocellulose membranes (Bio-Rad, Hercules, CA), blots were blocked with 5% nonfat milk in Tris-buffered saline containing 0.1% Tween 20 for 1 h at room temperature and incubated overnight at 4°C with primary antibodies. OSTα and OSTβ polyclonal antibodies were generated as previously described (3). Monoclonal anti-β-actin antibodies and horseradish peroxidase-conjugated secondary antibodies were from Sigma, and enhanced chemiluminescence reagents were from Amersham Pharmacia Biotech (Piscataway, NJ). Densitometry was performed with Multi-Analyst software (Bio-Rad).
Data are expressed as the means ± SD. Differences between specific groups were determined using an unpaired Student's t-test and analyzed by the Instat software package (GraphPad Software, San Diego, CA). Data without error bars represent pooled samples.
OST/Ostα and OST/Ostβ mRNA and protein are induced in human and rodent cholestatic liver
mRNA and protein samples from human cholestatic livers were obtained from a series of patients with primary biliary cirrhosis (31). Both OSTα and OSTβ are more highly expressed in normal human liver than rodent livers (3), and the hepatic expression of these subunits in humans is more comparable with the marine skate from which the original genes were cloned (29). Figure 1, A and B, demonstrates that there are significant increases in the mRNA and protein of both these subunits in patients with primary biliary cirrhosis with a threefold increase in OSTα mRNA and a marked 32-fold increase in OSTβ mRNA compared with control human liver. OST protein expression also increased approximately two- to threefold for the α-subunit and four- to fivefold for the β-subunit (Fig. 1B).
After bile duct ligation in the rat, Ostα mRNA in liver was reduced at days 3 and 7 but unchanged compared with sham controls at day 14 (Fig. 2A). In contrast to Ostα mRNA, Ostα protein was significantly upregulated fourfold at day 14 (Fig. 2B). Remarkably, Ostβ mRNA was upregulated 100-fold at day 7 and ∼70-fold at day 14 (Fig. 2A). Although Ostβ protein was not detectable by Western blotting of a whole rat liver total membrane fraction, it could be weakly detected by immunofluorescence in cholangiocytes and hepatocytes (data not shown). Quantitative PCR analysis of mRNA obtained from isolated rat hepatocytes and from rat bile duct tissue demonstrated that the upregulation of Ostβ occurs predominantly in hepatocytes rather than in cholangiocytes (Fig. 2C).
Figure 3A illustrates the Ostα and Ostβ mRNA response to day 3, 7, and 14 bile duct ligation in the mouse. There were no significant changes in Ostα mRNA, whereas Ostβ mRNA increased by ∼20-fold at day 3, 21-fold at day 7, and 2.5-fold after day 14 following common bile duct obstruction. However, a progressive increase in Ostα protein was seen over the 14-day time course. As was the case in rat liver, Ostβ protein was undetectable in total membrane fractions of whole mouse liver, although it was weakly visible by immunofluorescence in the basolateral membrane in cholangiocytes (data not shown).
The absence of Fxr prevents the upregulation of Ostβ in CBDL mice.
To determine whether the changes in expression of these two subunits are dependent on the action of the nuclear receptor Fxr, studies were carried out before and after bile duct ligation in control and Fxr−/− mice. Bile duct ligation in Fxr−/− mice failed to enhance expression of Ostα and Ostβ compared with wild-type mice (Fig. 4). As seen in Fig. 4A, there were no significant changes in hepatic Ostα mRNA levels after 3 days of CBDL in either the Fxr+/+ or Fxr−/− mouse. However, there was a small but significant twofold increase at day 7 in wild-type CBDL mice but not Fxr−/− day-7 CBDL mice. Furthermore, the highly significant changes in Ostβ mRNA expression in wild-type mice were completely abrogated 3 and 7 days following CBDL in the Fxr−/− mouse, indicating complete dependence of the induced expression of the β-subunit on the presence of Fxr (Fig. 4B).
Bile acid regulation of OSTα and OSTβ expression in HepG2 cells.
To further test the hypothesis that FXR is involved in the cholestasis-induced changes in OSTα and OSTβ expression, transcript abundance of OSTα and OSTβ was measured in HepG2 cells treated with the FXR ligand CDCA. OSTα and OSTβ mRNA were markedly increased in cells treated for 24 h with CDCA (Fig. 5A). The effect on OSTβ mRNA (∼70-fold increase) was larger than on OSTα (∼18-fold increase), consistent with the in vivo studies in human and rodent liver described previously. BSEP and SHP mRNA levels were also found to be increased after CDCA treatment (Fig. 5A), consistent with previous reports that they are FXR regulated (12, 23). These changes in mRNA were accompanied by large increases in OSTβ protein and smaller changes in OSTα protein (Fig. 5B). Note that the OSTα protein appears on Western blots as two bands of ∼36 and 40 kDa. These probably represent the unmodified and glycosylated forms of the polypeptide, respectively (7). Immunofluorescent studies of HepG2 cells treated with 50 μM CDCA revealed a large increase in staining of the plasma membrane for OSTα and OSTβ (Fig. 6, B and E) that was not evident in untreated cells (Fig. 6, A and D) or cells treated with 100 μM UDCA (Fig. 6, C and F), suggesting both an upregulation and a possible redistribution of the protein to the plasma membrane.
The present studies provide significant and novel insights into the expression of OSTα and OSTβ in cholestasis in humans and rodents and provide evidence that these adaptive responses are regulated by the nuclear receptor FXR. OSTα and OSTβ are two recently described proteins whose simultaneous expression at the basolateral membrane of cells results in a novel organic solute carrier (2, 3, 7, 22, 29). In the ileum, where these solute carriers are most highly expressed, OSTα and OSTβ presumably function physiologically as critical components of the enterohepatic circulation of bile acids by transporting them from the enterocytes to the splanchnic circulation for subsequent first pass extraction by bile salt transporters in the liver. This conclusion is based on the ability of the bile salt taurocholate to move from luminal to the basolateral compartments of MDCK cells in transwells only when both subunits are expressed (7) and by efflux of taurocholate in Xenopus oocytes expressing OSTα and OSTβ (3). Bile acids can also be transported into cholangiocytes and the renal proximal tubule by ASBT (6, 17). Cholangiocyte ASBT presumably contributes to the cholehepatic shunting of bile salts, which is more prominent in cholestatic liver injury where bile ducts are proliferated, whereas the renal ASBT normally recycles bile acids from the lumen of the proximal tubule after they are filtered at the glomerulus (17). Localization of Ostα and Ostβ to the basolateral membranes of cholangiocytes and proximal tubule of the kidney (3) strongly suggests that these proteins form the functional transporter that completes the cholehepatic and renal-systemic shunting of bile acids and possibly other organic solutes and steroids.
The present study extends these prior observations by examining the effects of cholestasis on the expression of these two subunits in cholestatic liver disease in human and in rodent models of obstructive cholestasis. Although the findings reveal species differences in the magnitude of this response, they support the above-predicted physiological role. The expression of OSTα-OSTβ in the human liver is higher than in rodent liver, as previously reported (7, 22). Thus, in the present study, the most significant results come from the changes of expression of this transporter in human livers from patients with primary biliary cirrhosis. Both α- and β-subunits are upregulated in these patients both at the mRNA level and in Western blot analysis of tissue protein. In the rat and mouse, Ostα protein is also significantly increased in the liver after CBDL. Competition transport studies indicate that sulfated and/or amidated bile acids are more powerful inhibitors of transport than the corresponding parent compounds (22, 29), suggesting that these are better substrates for Ostα-Ostβ. Thus the present findings in cholestatic liver are consistent with the known substrate specificity of Ost. Furthermore, in HepG2 cells, exposure to CDCA induced a dramatic increase in the amount of the OSTs located at the plasma membrane, suggesting that increases in tissue levels of bile acids are providing the stimulus for a possible translocation to the plasma membrane. Thus bile acids may have both transcriptional as well as posttranscriptional effects on OST expression.
Differences were observed after cholestasis between the level of expression of OST/Ostα and OST/Ostβ both within and between species, which are not currently explained. However, it should be noted that neither the mechanism by which OST/Ostα and OST/Ostβ interact, their individual roles in generating a functional complex at the plasma membrane, nor their individual roles in solute transport have yet been identified. Although both proteins are required to elicit transport activity, there is as yet no evidence that they combine in any particular stoichiometry. Indeed, studies of Dawson et al. (7) indicate that Ostβ may function as a chaperone or catalyst for delivery of mouse Ostα to the plasma membrane because coexpression of Ostα and Ostβ was required to convert the Ostα subunit to a mature N-glycosylated endo H-resistant form. This suggests that coexpression facilitates the movement of Ostα through the Golgi apparatus.
Previous studies (26, 28) clearly demonstrate that all other major transporters for bile acids in the enterohepatic circulation undergo adaptive regulation in the cholestatic liver in a manner that would help protect these tissues from bile acid accumulation and tissue injury. Furthermore, several of these changes in transporter expression are either directly or indirectly regulated by Fxr, the bile acid nuclear receptor (5, 28). Thus NTCP and Asbt are indirectly downregulated by the action of the short heterodimeric protein SHP-1 on their promoters (8), and expression of SHP-1 is positively upregulated by bile acid activation of FXR (12). In contrast, BSEP and the ileal bile salt binding protein (I-BABP) contain FXR response elements in their promoters, so their expression is directly and positively upregulated by bile acid activation of FXR (23). Thus it is logical to expect that if OSTα-OSTβ is an important component of the enterohepatic, cholehepatic, and renal-systemic circulations of bile acids, its expression should also be regulated by FXR. To assess this directly, we used the Fxr−/− mouse model in which it has been demonstrated that adaptive responses of Bsep, Shp-1, Ntcp, and the I-babp after bile duct ligation or cholic acid feeding depend on the presence of Fxr (28). The results of the present study clearly confirm this expectation by demonstrating that the induction of mouse liver Ostβ mRNA by CBDL is completely abolished in the Fxr−/− mouse. Regulation of both mouse and human OST/Ostα and OST/Ostβ by FXR has also been recently confirmed (10, 15, 16, 33). Thus it is likely that OSTα-OSTβ is the key transporter whose induction at the basolateral membrane of hepatocytes, cholangiocytes, and proximal tubule epithelia contributes to the adaptive response seen in cholestasis.
Several MRP/Mrps are also located at the basolateral membrane of hepatocytes and cholangiocytes, and their expression is upregulated in human cholestatic liver disease and obstructive cholestasis in rats and mice (9, 24, 31). Furthermore, MRP3/Mrp3 and MRP4/Mrp4 both are capable of transporting divalent bile acid conjugates in in vitro expression systems (1, 26). Although the functional significance of changing levels of expression of different transporters is difficult to compare, the quantitative increases seen in the present studies for the Osts after bile duct ligation in the mouse are much higher than previously observed for mouse Mrp3 and Mrp4 (28). Previous studies have shown that the induction of OSTα-OSTβ is also more profound than the induction of MRP3 and MRP4 in human cholestatic liver. Notably, the OSTs undergo extensive regulation at the mRNA level, which was not seen for human MRP3 and MRP4 (31). Although studies of the effects of bile duct ligation on Mrp3 and Mrp4 expression in their respective knockout mouse models suggest that both contribute to liver tissue protection [with Mrp4 having the greater role as a bile acid conjugate efflux pump (4, 20, 30)], Ostα-Ostβ may well provide the predominant response. Future studies in Ostα-Ostβ knockout mice may address these issues. The present studies provide the first evidence for a basolateral transporter to be directly regulated by FXR in cholestasis. Mrp3 and Mrp4 are Fxr independent and are even induced to a higher extent in cholestatic and bile acid-challenged Fxr knockout mice (28). Thus the OSTs may be the long-sought missing link in the adaptive response to toxic bile acids.
It remains unclear why there are such profound differences in expression of OST/Ostα and OST/Ostβ between species. Although the original transcripts were cloned from a marine skate liver library as part of an effort to identify novel organic solute transporters, message levels were found for these transcripts in a wide variety of skate tissues (22, 29). In humans, the highest expression of OSTα is liver > testes > intestinal tissue > kidney, whereas the β-subunit is expressed most highly in testes > intestine > kidney > adrenal gland > liver, although it should be noted that ileum was not specifically selected in those studies (22). Nevertheless, the widespread expression of these proteins in human tissues and their ability to transport a wide range of organic solutes, including the steroids taurocholate, estrone 3-sulfate, dehydroepiandrosterone sulfate, and digoxin as well as the eicosanoid prostaglandin E2, suggests that bile acid transport is not their only function (3, 22, 29). Although future studies should examine these questions, the present findings strongly support a major role for OST/Ostα and OST/Ostβ as part of the adaptive response of bile acid transport proteins to limit the recycling and tissue accumulation of bile acids during cholestasis, particularly in human liver where these subunits are more highly expressed.
This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grants DK-25636 (to J. L. Boyer) and DK-P30–34989 (the Yale Liver Center), the NIEHS MFBS Center Grant ES-03828, Grant P18613-B05 from the Austrian Science Fund (to M. Trauner), and Grants DK-067214 and ES-01247 (to N. Ballatori).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society