Introduction

Top Abstract Introduction Results Discussion References

Multidrug resistance-associated protein 3 (Mrp3/MRP3¹; Abcc3/ABCC3) mediates the ATP-dependent transport of anionic compounds including glucuronide conjugates, glutathione conjugates, and sulfate conjugates of some bile salts, and thus exhibits overlapping substrate specificity with Mrp2. Mrp3 also transports monovalent bile salts (Hirohashi et al., 2000). In rats, Mrp3 is expressed predominantly in intestine, lung, and kidney; hepatic Mrp3 expression is very low (Hirohashi et al., 1998, 2000). However, phenobarbital (PB) can markedly induce Mrp3 expression in liver. Induction of Mrp3 may significantly alter the hepatobiliary disposition of a variety of anionic compounds (Takenaka et al., 1995; Xiong et al., 2000).

PB is known as a prototype of a large group of structurally unrelated chemicals that induce many cytochrome P450 (P450) genes including CYP2A, CYP2B, CYP2C, CYP2H, CYP3A (Sueyoshi and Negishi, 2001). Among those genes, CYP2B genes have been studied most intensively because they are induced most effectively by PB. PB response elements have been identified in the promoter regions of CYP2B genes (Trottier et al., 1995; Honkakoski and Negishi, 1997); the core enhancer sequence is a 51-base pairs DNA fragment called the phenobarbital-responsive enhancer module (Honkakoski et al., 1998). Subsequently, it was established that the nuclear receptor CAR binds to phenobarbital-responsive enhancer module and regulates the expression of CYP2B genes (Honkakoski et al., 1998; Kawamoto et al., 1999; Wei et al., 2000). The mechanism of P450 induction by PB has not been fully elucidated. However, PB is known to stimulate both the nuclear translocation and nuclear activation of CAR (Sueyoshi and Negishi, 2001). PB also activates human pregnane X receptor (PXR). However, PB has little or no activity in the activation of rat PXR (Jones et al., 2000; Moore et al., 2000). CAR and PXR belong to a nuclear receptor family called nuclear hormone receptors or nuclear orphan receptors (NORs). Increasing evidence suggests that NORs may be involved in the regulation of some hepatobiliary transporters in addition to P450 enzymes (Muller, 2000; Kast et al., 2002). Interestingly, recent studies indicated that induction of both CYP3A and Mdr1 P-glycoprotein by rifampicin was mediated by PXR (Kliewer et al., 1998; Geick et al., 2001). The coregulation of a P450 enzyme and a hepatic transporter led to the hypothesis that the PB induction of CYP2B genes and Mrp3 in rats may be mediated by the same nuclear receptor, CAR.

Strain- and gender-selectivity in the induction of CYP2B genes have been documented in rodents (Blouin et al., 1993; Larsen et al., 1994). The obese (fa/fa) Zucker rat is a genetically obese rat strain due to an autosomal recessive mutation in the leptin receptor gene, which results in decreased leptin responsiveness (Phillips et al., 1996). Blouin et al. (1993) reported that PB induction of hepatic CYP2B1 and CYP2B2 was ~3-fold lower in obese compared with lean Zucker rats. Because CYP2B induction is mediated by CAR, it was hypothesized that CAR expression in obese Zucker rat livers may be impaired, and thus obese Zucker rats may be a useful model to investigate the role of CAR in the induction of Mrp3. Wistar Kyoto (WKY) rats exhibit sexually dimorphic induction of CYP2B1 and CYP2B2 (Larsen et al., 1994). The impaired induction of CYP2B genes in female WKY rat livers may be attributed primarily to decreased levels of CAR protein in these livers (Yoshinari et al., 2001). Female WKY rats were used as a supplementary animal model to test the hypothesis that CAR mediates Mrp3 induction. In addition, an in vitro model, mouse CAR-expressing HepG2 (g2car-3) cells, was used to investigate the role of CAR in the induction of Mrp3.

Experimental Procedures

Materials. PB was purchased from Sigma-Aldrich (St. Louis, MO). TRIzol reagent and SuperScript preamplification system were purchased from Invitrogen (Carlsbad, CA). Advantage 2 PCR kit was purchased from BD Biosciences Clontech (Palo Alto, CA). Anti-RXR antibody was obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). Anti-CAR antibodies were prepared in our laboratory (Yoshinari et al., 2001). Anti-Mrp3 antiserum was a gift from Dr. Yuichi Sugiyama (Tokyo, Japan). Oligonucleotides were synthesized with ABI 392 DNA/RNA synthesizer (Applied Biosystems, Foster City, CA). All other chemicals were of analytical reagent grade.

Animals. Zucker rats (6-8 weeks) and WKY rats (5-6 weeks) were obtained from Charles River Laboratories, Inc. (Raleigh, NC). Rats were maintained on a 12-h light/dark cycle. Access to rat chow and water was allowed ad libitum. Rats were allowed to acclimate for 5 to 7 days prior to experimentation. The Institutional Animal Care and Use Committee of the University of North Carolina at Chapel Hill approved all procedures. For CYP2B induction studies, lean and obese Zucker rats were injected with PB (100 mg/kg body weight) intraperitoneally. Control rats received vehicle (saline) only. Rats were sacrificed 3 h after dosing. For Mrp3 induction studies, lean Zucker rats, obese Zucker rats, and WKY rats received PB (75, 45, and 75 mg/kg/d, respectively) orally for 4 days (Brouwer et al., 1984). Control rats received saline. Rats were sacrificed 24 h after the last dose.

Cell Culture. HepG2 and g2car-3 (Sueyoshi et al., 1999) cells were cultured in minimal essential medium supplemented with 10% fetal bovine serum. The cells were grown at more than 90% confluence and treated with 1 to 5 mM PB or 250 nM 1,4-bis[2-(3,5-dichloropyridyloxy)]benzene (TCPOBOP) for 4 days.

Western Blotting. Preparation of nuclear extracts from rat livers was carried out as described previously (Sueyoshi et al., 1995). Nuclear extracts (20 µg of protein) were separated on 10% SDS-polyacrylamide gel and transferred to Immobilon-P (Millipore, Bedford, MA). The membrane was immunostained with anti-CAR or anti- RXR antibodies. To examine the expression of Mrp3, liver homogenate and liver crude membrane fractions were prepared as described by Bergwerk et al. (1996). Aliquots of liver homogenate (80 µg of protein) or crude membrane fractions (60 µg of protein) were separated on 4-10% SDS-polyacrylamide gel and transferred to polyvinylidene difluoride membrane. The membrane was probed with anti-Mrp3 antiserum.

RT-PCR. Preparation of total RNA, synthesis of cDNA, and PCR amplification were performed as described previously (Sueyoshi et al., 1999; Yoshinari et al., 2001). Primer sequences for CYP2B1 and CYP2B2 mRNA amplification were described previously (Li and Kupfer, 1998). A pair of oligonucleotides, 5'-TCTCACTCAACACTACGTTC-3'/5'-CTGGGAAAGGATCCAAGCCTGGG-3' and 5'-AGGACCCCATCCTGTTCT-3'/5'-CTGGAGAATCAAATTCAG-3' were used for CAR and MRP3 mRNA amplification, respectively. Numbers of amplification cycles were 20 to 23 for CYP2B1/2, 32 for CAR, 24 for MRP3, 32 for CYP2B6, and 18 for -actin. All amplified PCR products were subcloned into pCR2.1-TOPO (Invitrogen) and sequences were verified.

High-Performance Liquid Chromatography Assays. Plasma samples were collected from each rat when rats were sacrificed. Samples were stored at 20°C until assay. PB concentrations in plasma samples were determined by a high-performance liquid chromatography method described by Studenberg and Brouwer (1992).

	Results

Top Abstract Introduction Results Discussion References

CAR Expression in Zucker Rat Livers. CAR mRNA levels were significantly lower in obese compared with lean Zucker rat livers (Fig. 1A). No CAR protein was detectable in the nuclear extracts from saline-treated obese rat livers. In contrast, the constitutive levels of nuclear CAR in the lean rat livers were easily detected. In response to PB treatment (3 h), nuclear CAR levels were significantly increased in both lean and obese rat livers. However, significantly higher CAR levels were observed in PB-treated lean Zucker rat livers. Nuclear levels of RXR, a NOR that forms a heterodimer with CAR, were not different between lean and obese Zucker rat livers and were not affected by PB treatment. In addition to CAR, expression of other NORs including liver X receptor, farnesoid X receptor (FXR), and peroxisome proliferator-activated receptor  in obese Zucker rat livers also was examined in the present study. No significant differences in mRNA levels were observed between lean and obese Zucker rats (data not shown).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Differences in CAR protein and mRNA levels and CYP2B1/2B2 induction by PB between lean and obese Zucker rat livers. Nuclear extracts (20 µg of protein) prepared from saline- or PB-treated (3 h) Zucker rat livers were separated by 10% SDS-PAGE and probed with anti-CAR and anti-RXR antibodies. Total RNA was prepared from saline- or PB-treated (3 h) Zucker rat livers and CAR, CYP2B1, CYP2B2, and -actin mRNA were amplified by RT-PCR and separated by agarose gel electrophoresis; panels show the reverse images of ethidium bromide-stained gels. The arrow indicates the full-length CAR mRNA. The other bands are splicing variants of CAR mRNA. C, saline-treated; PB, PB-treated; PAGE, polyacrylamide gel electrophoresis.

Induction of CYP2B1 and CYP2B2 in Zucker Rats. CYP2B1 and CYP2B2 mRNA levels also were examined in the livers of lean and obese Zucker rats treated with PB (3 h). The constitutive levels of CYP2B1 mRNA were very low in both obese and lean rat livers (Fig. 1B). The constitutive levels of CYP2B2 mRNA in obese rat livers were about half of that in the lean rat livers. After PB treatment (3 h), CYP2B1 and CYP2B2 mRNA were increased significantly in both lean and obese rat livers. However, induction was much greater in lean compared with obese rat livers. The pattern of CYP2B1 and CYP2B2 induction by PB correlates with the different nuclear CAR levels in the lean and obese rat livers.

Induction of Mrp3 in Zucker Rats. Lean and obese Zucker rats were treated with PB for 4 days to examine the induction of Mrp3. Lean and obese rats received different doses of PB because of the difference in the disposition of PB between these two groups of rats (Brouwer et al., 1984). As expected, plasma PB concentrations (mean ± S.D.) measured at the time when rats were sacrificed, were not statistically different between lean and obese rats (52 ± 11 µg/ml and 48 ± 8 µg/ml, respectively), indicating that these rats were exposed to similar concentrations of PB during the 4-day treatment. The constitutive levels of Mrp3 in lean Zucker rat livers were barely detectable at either the mRNA or protein level (Figs. 2 and 3). The Mrp3 levels in obese Zucker rat livers also were very low but were notably higher than Mrp3 levels in lean rat livers. PB treatment (4 days) markedly induced the expression of Mrp3 in lean rat livers. In contrast, PB treatment only slightly induced Mrp3 in obese rat livers. As a control, PB induced CYP2B1 to a greater extent in lean compared with obese Zucker rat livers (Fig. 3). Prolonged PB treatment did not seem to alter the expression of CAR. CAR mRNA levels were still much lower in obese compared with lean rat livers (Fig. 3).

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Western blot analysis of Mrp3 expression in lean and obese Zucker rat livers. A, liver crude membrane fractions (60 µg of protein) prepared from saline- and PB-treated (×4 days) lean and obese Zucker rats were separated by SDS-PAGE and probed with anti-Mrp3 antiserum and anti-actin antibody. B, densitometric analysis of the blot in Fig. 2A. Levels of Mrp3 were normalized by the levels of -actin. The Mrp3 level in saline-treated lean Zucker rat livers was set equal to 1. PAGE, polyacrylamide gel electrophoresis.

View larger version (43K): [in this window] [in a new window]   Fig. 3.   RT-PCR analysis of Mrp3, CYP2B1, and CAR expression in lean and obese Zucker rat livers. Zucker lean and obese rats were treated with saline or PB for 4 days. Total RNA was prepared from the livers and Mrp3, CAR, CYP2B1, and -actin mRNA were amplified by RT-PCR and separated by agarose gel electrophoresis. The arrow indicates the full-length CAR mRNA. The other bands are splicing variants of CAR mRNA.

Induction of Mrp3 in WKY Rats. Induction of hepatic Mrp3 also was examined in male and female WKY rats following 4 days of PB treatment. At both the mRNA and protein levels, the constitutive expression of Mrp3 was not detectable in male WKY rats. In contrast, low levels of Mrp3 were detected in saline-treated female rat livers. PB treatment significantly induced Mrp3 in both male and female WKY rats to similar levels (Fig. 4). As previously reported, CYP2B1 induction by PB was significantly lower in female relative to male WKY rat livers (Fig. 4B). Nuclear levels of CAR after 4 days of PB treatment also were examined. Low basal levels of CAR were detected in the nuclear extracts from male WKY rat livers whereas no measurable levels were detected in female WKY rat livers (Fig. 5). PB treatment increased nuclear levels of CAR in both male and female WKY rats. However, the increase was much higher in male compared with female WKY rats (Fig. 5).

View larger version (57K): [in this window] [in a new window]   Fig. 4.   Protein and mRNA levels of Mrp3 in male and female WKY rat livers. WKY rats were treated with saline or PB for 4 days. A, liver homogenates (80 µg protein) were separated by 4-12% SDS-PAGE and probed with anti-Mrp3 antiserum and anti-actin antibody. B, Mrp3 and -actin mRNA were amplified by RT-PCR from total RNA prepared from these livers. PAGE, polyacrylamide gel electrophoresis.

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Levels of CAR in total nuclear extracts from WKY rat livers. Total nuclear extracts (20 µg protein) prepared from saline- or PB-treated (×4 days) WKY rats were separated by 10% SDS-PAGE and probed with anti-CAR antibody and anti-RXR antibody. PAGE, polyacrylamide gel electrophoresis.

Induction of CYP2B and MRP3 in HepG2 and g2car-3 Cells. The role of CAR in the induction of MRP3 also was studied in HepG2 cells and g2car-3 cells. HepG2 and g2car-3 cells were treated with 1 to 5 mM PB or 250 nM TCPOBOP for 4 days, and the mRNA levels of human MRP3 and CYP2B6 were examined by RT-PCR. PB induced MRP3 in both HepG2 and g2car-3 cells in a concentration-dependent manner (Fig. 6). However, no significant differences in MRP3 levels were observed between HepG2 and g2car-3 cells in the presence or absence of inducers (PB or TCPOBOP). CYP2B6 mRNA levels were significantly higher in g2car-3 cells than in HepG2 cells (Fig. 6A). CYP2B6 expression was not influenced by PB treatment in either cell line. TCPOBOP significantly induced CYP2B6 but not MRP3 expression in g2car-3 cells.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   RT-PCR analysis of MRP3 and CYP2B6 expression in HepG2 and g2car-3 cells. A, HepG2 and g2car-3 cells were treated with blank medium (lane 0), or medium with PB (1 mM, lane 1; 2 mM, lane 2; or 5 mM, lane 5), or 250 nM TCPOBOP (lane T) for 4 days. Total RNA was prepared from the cells, and MRP3, CYP2B6, and -actin mRNA were amplified by RT-PCR. Data are representative of three independent experiments. B, densitometric analysis of MRP3 bands in Fig. 6A. Levels of MRP3 were normalized by the levels of -actin. The MRP3 level in untreated HepG2 cells was set equal to 1.

	Discussion

Top Abstract Introduction Results Discussion References

In the present study, we have examined the potential role of CAR in the induction of Mrp3 by PB in vivo and in vitro. The obese (fa/fa) Zucker rat has been used as a model to study obesity. The fatty locus (fa/fa) has a mis-sense mutation in the coding region of the leptin receptor gene that diminishes but does not completely eliminate responsiveness to leptin (Phillips et al., 1996). In addition to obesity, obese Zucker rats also develop hypercholesterolemia, hyperlipidemia, hyperglycemia, and changes in endocrine homeostasis (Bray, 1977). Blouin and coworkers (1993) previously reported that induction of hepatic CYP2B1 and CYP2B2 by PB was impaired in these rats. Recognition that CYP2B genes are regulated by CAR led to the hypothesis that the function of CAR is impaired in obese Zucker rat livers.

The present study revealed that the expression of CAR is impaired in obese Zucker rat livers. As a consequence, the nuclear levels of CAR were lower both under basal conditions and after PB treatment, which correlated with the impaired induction of CYP2B1 and CYP2B2 by PB (3 h) in these livers. This finding was consistent with previous reports that the induction of these genes is mediated by CAR (Honkakoski et al., 1998; Kawamoto et al., 1999). PB treatment for 4 days significantly increased Mrp3 in lean Zucker rat livers. In contrast, the induction was marginal in obese rat livers. The same induction pattern was observed for CYP2B1. The same pattern in the induction of Mrp3 and CYP2B1 by PB in Zucker rat livers suggested that CAR may play a role in the induction of Mrp3 by PB. However, since CAR deficiency is neither the primary nor the only defect in obese Zucker rats, the observations may be confounded by other deficiencies. Additional evidence was needed to support or refute this hypothesis.

Recently, gender-dependent expression of CAR was revealed in WKY rat livers. Although the levels of CAR mRNA are similar between male and female WKY rat livers, the levels of CAR protein in the cytosol and nucleus were extremely low in female rat livers. Gender-dependent expression of CAR has been suggested as the primarily reason for the sexually dimorphic induction of CYP2B genes by PB in WKY rats (Yoshinari et al., 2001). In the present study, significant impairment in the PB induction of CYP2B1 but not Mrp3 was observed in female compared with male WKY rat livers. The same extent of induction of Mrp3 by PB in male and female WKY rat livers, regardless of the marked gender difference in CAR expression in WKY rats, suggested that CAR does not play a key role in the induction of Mrp3.

In addition to the animal models, Mrp3 induction also was examined in g2car-3 cells, because PB induces not only rat Mrp3 but also human MRP3 (Kiuchi et al., 1998). Although significant species differences in the ligand specificity for CAR have been observed (Moore et al., 2000), PB activates human, rat, and mouse CAR. In addition, the DNA-binding domain of mouse and rat CAR is identical and shares 95.5% homology with the DNA-binding domain of human CAR (Jones et al., 2000). Both human and mouse CAR-expressing HepG2 cells have proved to be useful models in elucidating the role of CAR in the induction of human CY2B6 and human bilirubin UDP-glucuronosyltransferase 1A1 (UGT1A1) by PB (Kawamoto et al., 1999; Sueyoshi et al., 1999; Sugatani et al., 2001). Consistent with the previous report, significant differences in CYP2B6 expression were observed between HepG2 cells and g2car-3 cells (Sueyoshi et al., 1999). Higher basal levels of CYP2B6 expression in g2car-3 cells were due to the constitutive activation of CAR in these cells. PB did not further increase CYP2B6 expression in g2car-3 cells, suggesting that PB could not further activate CAR in these cells. Sueyoshi et al. (1999) reported that the induction of CYP2B6 by PB in g2car-3 cells was concentration-dependent in the presence of 3-androstanol, an endogenous CAR antagonist, which suggested that PB induced CAR-mediated transactivation by displacing 3-androstanol. In contrast to CYP2B6 expression, no significant difference was observed in the expression of MRP3 between HepG2 and g2car-3 cells. In both cell lines, dose-dependent induction of MRP3 by PB was evident, consistent with a previous report in HepG2 cells (Kiuchi et al., 1998). The similar pattern and extent of MRP3 induction by PB in both cell lines suggested that PB induction of MRP3 is mediated by factors other than CAR. The CAR-mediated induction of CYP2B by TCPOBOP has been reported previously (Tzameli et al., 2000). The observation that TCPOBOP significantly induced CYP2B6 but not MRP3 in g2car-3 cells further supports the conclusion that CAR does not play a key role in the induction of human MRP3 by PB. Although we cannot exclude the possibility that human MRP3 and rat Mrp3 are regulated differently by CAR, the in vitro study strongly suggested that CAR is not a key mediator in PB induction of Mrp3/MRP3.

Taken together, data from WKY rats and HepG2 cells suggested that CAR does not play a key role in the induction of Mrp3/MRP3 by PB. The impaired induction of Mrp3 in obese Zucker rats likely is due to another deficiency associated with the mutation in the leptin receptor gene. It is unlikely that PB induction of rat Mrp3 is mediated by PXR or other NORs, because PB has little or no activity in the activation of rat PXR and other NORs. The 5'-flanking regions of human MRP3 have been cloned and characterized (Fromm et al., 1999; Takada et al., 2000). Sequence analysis revealed the presence of consensus binding sites for a number of transcriptional factors including Sp1, AP1, AP2, AP3, N-myc, CCAAT/enhancer-binding protein, hepatic nuclear factor-5, and nuclear factor-B. Takada et al. (2000) determined that the region between 127 and 22 nucleotides, which includes the TATA-less box and Sp1 binding sites, is important for the expression of human MRP3. Whether this region also is important for the induction of MRP3 is not known. Alterations in transcriptional factors in obese Zucker rat livers have been reported. Higher constitutive AP1 activity was observed in obese Zucker rat livers, which was not activated further by PB treatment. In contrast, signal transducer and activator of transcription activity was lower in obese rat livers (Roe et al., 1998). Investigation of the potential roles of AP1 and signal transducer and activator of transcription in Mrp3 induction, along with investigation of other transcription factors that exhibit altered expression in obese Zucker rat livers, may elucidate the mechanisms of Mrp3 induction.

Increasing evidence suggests that hepatobiliary transporters, like P450 enzymes, are subject to regulation by NORs. For example, expression of the bile salt export pump is regulated by FXR (Sinal et al., 2000), and expression of MDR1 and Oatp2 is regulated by PXR (Geick et al., 2001; Staudinger et al., 2001). Kast et al. (2002) recently reported that Mrp2/MRP2 is regulated by PXR, FXR, and CAR. However, many of the experiments were conducted in vitro, and the results may not be consistent with in vivo observations. For example, Mrp2 was induced by pregnenolone 16-carbonitrile (PCN; PXR ligand) and PB (CAR activator) in primary cultured rat hepatocytes (Kast et al., 2002). However, neither PCN nor PB induces Mrp2 mRNA levels in rat in vivo (Ogawa et al., 2000; Vore M, personal communication). Likewise, induction of Mdr1 by PCN was observed in vitro but not in vivo (Salphati and Benet, 1998; Jones et al., 2000). The discrepancies between the results from in vivo and in vitro studies suggest that some regulatory mechanisms demonstrated in vitro may not play a dominant role in the physiologically intact system, or are relevant only under extreme conditions.

In addition to the difference in Mrp3 induction, the constitutive levels of Mrp3 were higher at both the mRNA and protein levels in obese relative to lean Zucker rat livers. This higher basal Mrp3 expression may be associated with the obese state. The metabolism and excretion of bile salts and bile lipids in obese Zucker rats are significantly different compared with their lean counterparts (Bray, 1977; VanPatten et al., 2001). This may result in the accumulation of endogenous Mrp3 inducer(s), similar to cholestatic conditions.

Gender differences in the constitutive levels of Mrp3 also were observed in WKY rat livers. This is the first report of gender differences in Mrp3 expression. These gender differences in Mrp3 expression suggest that the disposition of Mrp3 substrates (e.g., glucuronide conjugates, glutathione conjugates, bile salts) may differ between male and female WKY rats. Gender differences in the expression and induction of other hepatobiliary transport proteins, including Mrp2, Mdr2, and Ntcp, have been documented in monkeys or rodents (Kauffmann et al., 1998; Salphati and Benet, 1998; Simon et al., 1999).

In conclusion, results from the present study demonstrate that the expression of CAR is deficient in obese Zucker rat livers, which is consistent with the impaired CYP2B induction by PB in this obese rat strain. Mrp3 induction by PB also was impaired in obese Zucker rat livers. However, data from WKY rats and mouse CAR-expressing HepG2 cells indicate that CAR does not play a key role in the induction of Mrp3. Other alteration(s) in obese Zucker rat livers are responsible for the impaired induction of Mrp3 by PB.