Peroxisome Proliferator-activated Receptor α Induces Hepatic Expression of the Human Bile Acid Glucuronidating UDP-glucuronosyltransferase 2B4 Enzyme*

Glucuronidation, a major metabolic pathway for a large variety of endobiotics and xenobiotics, is catalyzed by enzymes belonging to the UDP-glucuronosyltransferase (UGT) family. Among UGT enzymes, UGT2B4 conjugates a large variety of endogenous and exogenous molecules and is considered to be the major bile acid conjugating UGT enzyme in human liver. In the present study, we identify UGT2B4 as a novel target gene of the nuclear receptor peroxisome proliferator-activated receptor α (PPARα), which mediates the hypolipidemic action of fibrates. Incubation of human hepatocytes or hepatoblastoma HepG2 and Huh7 cells with synthetic PPARα agonists, fenofibric acid, or Wy 14643 resulted in an increase of UGT2B4 mRNA levels. Furthermore, treatment of HepG2 cells with Wy 14643 induced the glucuronidation of hyodeoxycholic acid, a specific bile acid UGT2B4 substrate. Analysis of UGT2B mRNA and protein levels in PPARα wild type and null mice revealed that PPARα regulates both basal and fibrate-induced expression of these enzymes in rodents also. Finally, a PPAR response element was identified in the UGT2B4 promoter by site-directed mutagenesis and electromobility shift assays. These results demonstrate that PPARα agonists may control the catabolism of cytotoxic bile acids and reinforce recent data indicating that PPARα, which has been largely implicated in the control of lipid and cholesterol metabolism, is also an important modulator of the metabolism of endobiotics and xenobiotics in human hepatocytes.

Glucuronide conjugation is a major metabolic pathway for numerous endogenous and exogenous compounds, including bile acids (BA), 1 bilirubin, steroids, drugs, and environmental pollutants (1). This reaction consists in the transfer of the glucuronosyl group from UDP-glucuronic acid to the acceptor molecule (1). The addition of the glucuronosyl group on a compound results in a more water-soluble molecule, which can be excreted into bile or urine. Glucuronidation is catalyzed by enzymes belonging to the UDP-glucuronosyltransferase (UGT) family, and based on primary structure homology, UGT proteins have been divided into two major subfamilies, UGT1A and UGT2B (2). In humans, seven members of the UGT2B subfamily have been characterized: UGT2B4, UGT2B7, UGT2B10, UGT2B11, UGT2B15, UGT2B17, and UGT2B28 (3,4).
Among the UGT2B enzymes, UGT2B4 catalyzes the glucuronide conjugation of various molecules, including BAs, 5␣reduced androgens, catecholesterogens, and phenolic and monoterpenoid compounds (4 -7). A certain degree of overlapping substrate specificity exists among the UGT2Bs, and these compounds are also conjugated by other UGT2B isoforms. However, various studies established the crucial role that UGT2B4 plays in hepatic BA glucuronide conjugation. Pillot et al. (7) carried out immunoprecipitation studies to demonstrate the strict substrate specificity of UGT2B4 for the 6␣-hydroxylated BA hyodeoxycholic acid (HDCA) in human liver. Furthermore, no or low glucuronidation activity of HDCA was observed in colon where UGT2B4 is not expressed (8,9). Finally, a recent study revealed that UGT2B4 expression is positively regulated by the BA sensor farnesoid X-receptor (FXR) and suggested that UGT2B4 induction by BAs may be part of a negative feedback mechanism by which BAs limit their biological activity and control their intracellular levels to avoid a pathophysiological accumulation (10).
An important consequence of BA glucuronidation is the introduction of an additional negative charge in the molecule that allows their transport by conjugate transporters such as the multidrug-resistance related proteins, MRP2 (ABCC2) and MRP3 (ABCC3), which are present in liver (11,12), and favors their excretion in urine. Whereas BAs are biological detergents with numerous important functions, these compounds are inherently cytotoxic and perturbations in their normal synthesis, transport, or secretion can result in a variety of pathophysiological conditions including intrahepatic cholestasis (13). During their enterohepatic circulation, BAs undergo several metabolic alterations, including glucuronide conjugation at ring hydroxyl groups (7,14). The most abundant glucuronide conjugate reported in human plasma is the primary BA chenodeoxycholic acid (CDCA) glucuronide followed by the secondary lithocholic acid (LCA) glucuronide (7,15). In the urine of cholestatic patients, the proportion of BA glucuronide metabolites increases to up to 35% of total BAs (16,17) and HDCA is exclusively found as a glucuronide derivative (18).
Peroxisome proliferator-activated receptors (PPARs) belong to the family of nuclear receptors that are ligand-activated transcription factors. Three distinct types of PPARs have been identified as PPAR␣, PPAR␦ (or PPAR␤), and PPAR␥. Each isotype is encoded by a distinct gene and shows different distribution patterns (19,20). Upon ligand activation, PPARs regulate gene transcription by dimerizing with the retinoid X-receptor (RXR) and binding to PPAR response elements (PPREs) within the regulatory regions of target genes (19). These PPREs usually consist of a direct repeat of the hexanucleotide AGGTCA sequence separated by one or two nucleotides (DR1 or DR2) (19). Furthermore, PPARs can also negatively interfere with pro-inflammatory transcription factor pathways by a mechanism termed transrepression (21). PPAR␣ is highly expressed in various tissues such as liver, muscle, kidney, and heart where it stimulates the ␤-oxidative degradation of fatty acids (22). Natural eicosanoids derived from arachidonic acid via the lipoxygenase pathway, such as 8-hydroxytetraenoic acid, 15-hydroxytetraenoic acid, and leukotriene B4 as well as oxidized phospholipids, activate PPAR␣ (23)(24)(25). The hypolipidemic fibrates (gemfibrozil, bezafibrate, ciprofibrate, and fenofibrate) are synthetic PPAR␣ ligands used in the treatment of dyslipidemia (23).
Recent findings indicate that PPAR␣ also regulates BA synthesis and transport. In cultured rat hepatocytes, PPAR␣ agonists decrease bile acid synthesis and suppress the expression of two key BA-synthesizing enzymes, the cytochrome P450 cholesterol 7␣-hydroxylase (CYP)7A1 and the sterol 27-hydroxylase (CYP27), which is paralleled by a similar reduction of their respective activities (26). By contrast, ligand-activated PPAR␣ stimulates the expression and activity of the murine sterol 12␣-hydroxylase enzyme (CYP8B1), a hepatic microsomal enzyme that acts as a branch point in the bile acid synthetic pathway, determining the ratio of cholic acid/CDCA (27). In human hepatoma HepG2 cells, the PPAR␣ ligand Wy 14643 suppresses CYP7A1 gene promoter activity (28). Furthermore, in mouse liver, treatment with the PPAR␣ agonist, ciprofibrate, results in a decreased expression of the bile salt transporters, such as Na ϩ -taurocholate co-transporting polypeptide 1, Na ϩ -independent organic anion-transporting polypeptide (Oatp1), and the bile salt export pump (29). By contrast, ciprofibrate activation of PPAR␣ induces the promoter activity of human apical sodium-dependent bile salt transporter (ASBT) gene in human colon carcinoma Caco2 cells (30). Overall, these data suggest that PPAR␣ activation may result in a decreased BAs synthesis and secretion into bile. Furthermore, several studies in both humans and animals reported that treatment with PPAR␣ activators results in enhanced glucuronidation activity and UGT expression (31,32). Indeed, clofibrate induces the bilirubin-conjugating UGT1A1 protein in microsomes from rat liver (32).
Since PPAR␣ is an important regulator of BAs synthesis and transport and considering the major role that UGT2B4 plays in hepatic glucuronidation of BAs, we investigated in the present study whether hepatic UGT2B4 expression and activity are regulated by PPAR␣. Our results demonstrate that PPAR␣ activation results in the induction of UGT2B4 gene expression in human primary hepatocytes and human hepatoblastoma HepG2 and Huh7 cells. The induction of UGT2B4 gene expression is accompanied by an increased glucuronidation activity of HDCA. This positive regulation occurs at the transcriptional level via binding of PPAR␣ to a DR1 response element located at Ϫ1193 bp in the promoter region of the UGT2B4 gene.

EXPERIMENTAL PROCEDURES
Animal Studies-Animal studies were performed in compliance with European Union specifications regarding the use of laboratory animals. Details of experimental conditions have been described previously (33). Male Sv/129 homozygous wild type (ϩ/ϩ) (n ϭ 6) and PPAR␣ null (Ϫ/Ϫ) (n ϭ 6) mice (34) (a kind gift of Dr. F. Gonzalez, National Cancer Institute, National Institutes of Health, Bethesda, MD) were fed for 17 days with a standard mouse chow diet containing 0.2% (wt/wt) fenofibrate or not. At the end of the treatment period, the animals were fasted for 4 h and sacrificed and livers were removed immediately, weighed, rinsed in 0.9% (w/v) NaCl, frozen in liquid nitrogen, and stored at Ϫ80°C until total RNA or microsome preparation.
Materials-UDP-glucuronic acid (UDPGA), leupeptin, pepstatin, phosphatidylcholine, and BAs were obtained from Sigma. Human hepatoblastoma HepG2 cells were from the American Type Culture Collection (Manassas, VA). Restriction enzymes and other molecular biology reagents were from New England Biolabs (distributed by Ozyme, Saint-Quentin, France), Stratagene (La Jolla, CA), Promega, and Roche Applied Science. Protein assay reagents were obtained from Bio-Rad.
Cell Culture-Human primary hepatocytes were isolated as described previously (35) and incubated for the indicated times in William's E medium containing fenofibric acid (250 M). Human hepatoma HepG2 and Huh7 cells were grown as described previously (36,37). For RNA analyses, 10 6 HepG2 or Huh7 cells were treated with Wy 14643 at the indicated concentrations in the presence or absence of 75 M CDCA for 24 h. In all of the experiments, controls were incubated with an identical volume of Me 2 SO (vehicle).
RNA Analysis-Total RNA was isolated from mice liver, primary human hepatocytes, HepG2, and Huh7 cells using TRIzol (Invitrogen). Northern blot analyses were performed as described previously (38) using human UGT2B4 and 36B4 cDNAs as probes. For quantitative RT-PCR analyses of UGT2B4 gene expression, RNA was reverse-transcribed using random hexamer primers and 200 units of Moloney murine leukemia virus reverse transcriptase (Invitrogen). Reverse-transcribed UGT2B4 and 28 S cDNAs were quantified by real-time PCR on a MX4000 apparatus (Stratagene) using specific primers for UGT2B4 and 28 S as described previously (36,39). PCR amplifications were performed in a volume of 25 l containing 100 nM of each primer, 4 mM MgCl 2 , the Brilliant Quantitative PCR Core reagent kit mixture (Stratagene), and SYBR Green 0.33X (Sigma). The conditions were 95°C for 10 min followed by 40 cycles of 30 s at 95°C, 30 s at 60°C, and 30 s at 72°C. UGT2B4 mRNA levels were normalized to 28 S mRNA (36).
Transient Transfection Assays-60 ϫ 10 3 HepG2 or Huh7 cells were transfected with 100 ng of the indicated luciferase reporter plasmids, 50 ng of the pCMV-␤-galactosidase expression vector, and with or without 30 ng of the pSG5-PPAR␣ plasmid. All of the samples were comple-mented with pBS-SKϩ plasmid (Stratagene) to an identical amount of 500 ng/well. Cells were transfected with ExGen reagent (Euromedex) for 6 h at 37°C and subsequently incubated overnight with Dulbecco's modified Eagle's medium, 0.2% fetal bovine serum and then treated for 24 h with either Me 2 SO (vehicle) or Wy 14643 (50 M) as indicated.
Electrophoretic Mobility Shift Assays (EMSA)-EMSA using in vitro produced PPAR␣ and RXR were performed as described previously (40) using the radiolabeled probes B4-PPREwt, 5Ј-TAAGATGAACTTTAA-TCTTGTAAC-3Ј; B4-PPREmt5Ј, 5Ј-TAAGATAAAATTTAATCTTGTA-AC-3Ј; and B4-PPREmt3Ј, 5Ј-TAAGATGAACTTTAAAATTGTAAC-3Ј (where underlined nucleotides represent response element half-sites and bases in boldface are mutated). For supershift experiments, 0.2 g of the anti-PPAR␣ antibody (Santa-Cruz Biotechnology) was preincubated for 20 min in the binding buffer before the addition of PPAR␣ and RXR proteins. For competition experiments, the unlabeled oligonucleotides were included in the binding reaction at the indicated excess concentrations over the probe just before adding the labeled oligonucleotide.
Microsome Purification and Western Blot Analysis-Microsomal proteins were purified from wild type or PPAR␣-null mouse livers as previously described (41). Microsome pellets were resuspended in 300 l of homogenization buffer, and the protein content was determined using Bradford reagent (Bio-Rad) and bovine serum albumin for standard curves. Samples were aliquoted and kept at Ϫ 80°C until Western blot analysis or glucuronidation assays. For Western blot, 25 g of microsomal proteins were separated on a 10% SDS-polyacrylamide gel. The gel was transferred onto a nitrocellulose membrane, which was then hybridized with the anti-UGT2B antibody (dilution, 1/2000). An anti-rabbit IgG antibody conjugated with peroxidase was used as secondary antibody (dilution, 1/10000), and the resulting immunocom-plexes were visualized using the Western blot Chemiluminescence Reagent Plus as specified by the manufacturer (PerkinElmer Life Sciences).
Glucuronidation Assay-HepG2 cells were resuspended in Tris-buffered saline containing 0.5 mM dithiothreitol and homogenized using a Brinkman Polytron. Enzyme assays were performed as described previously (5). 100 g of cell homogenate were incubated with 25 M [ 14 C]UDP-glucuronic acid, 2 mM unlabeled UDPGA, and 200 M HDCA in a final volume of 100 l of glucuronidation assay buffer for 8 h (5). Assays were terminated by adding 100 l of methanol, and the samples were centrifuged at 14,000 rpm for 2 min to remove the precipitated proteins. 100 l of glucuronidation assays were applied onto a thin layer chromatography (TLC) plate (Merck) and migrated using a toluene: methanol:acetic acid (7:3:1) mixture. The extent of HDCA glucuronidation was analyzed and quantified by PhosphorImager analysis.
Statistical Analyses-A nonparametric Mann-Whitney test was used to analyze for significant difference between the experimental groups. Analyses of variance (ANOVA) and Tukey post-hoc tests were used for analysis of the effects of simultaneous treatment with FXR and PPAR␣ agonists.

PPAR␣ Activators Induce UGT2B4 Expression in Human Hepatocytes
Primary human hepatocytes were treated with fenofibric acid (250 M) for 24 h, and UGT2B mRNA levels were determined by Northern blot analysis. A significant increase in UGT2B mRNA was observed in fenofibric acid-treated cells compared with vehicle-treated cells (Fig. 1a). With the cDNA sequences of the different human UGT2B isoforms being Ͼ85% homologous, a UGT2B4-specific real time RT-PCR method was used to specifically quantify UGT2B4 mRNA levels in hepatocytes treated for 6, 12, 24, or 48 h with fenofibric acid (250 M) (Fig. 1b). Fenofibric acid rapidly induced UGT2B4 expression, because a maximal 10-fold increase in the concentration of UGT2B4 transcripts was observed within 12 h (Fig. 1b).
To further characterize the PPAR␣-dependent regulation of UGT2B4 expression, human hepatoma HepG2 and Huh7 cells were incubated in the presence of increasing concentrations of the PPAR␣ ligand Wy 14643 (Fig. 2, a and b). UGT2B4 mRNA levels were induced in a dose-dependent manner to a maximum of 2.7-and 2.3-fold activation in HepG2 and Huh7 cells, respectively (Fig. 2, a and b).

PPAR␣ Activators Induce UGT2B4 Activity in HepG2 Cells
To determine whether PPAR␣ activation of UGT2B4 expression modifies its activity, HepG2 cells were treated with Wy 14643 (75 M) for 36 h and their glucuronidation activity was analyzed using the UGT2B4-specific substrate HDCA (Fig. 3a). Treatment with Wy 14643 provoked a 3-fold increase in HDCA glucuronidation (Fig. 3b), thus demonstrating that PPAR␣ agonists induce UGT2B4 activity.

PPAR␣ Gene Disruption Abolishes Fibrate Induction of UGT2B mRNA and Protein Levels in Mouse Liver
The PPAR␣-dependent induction of UGT2B expression was measured in male Sv/129 homozygous wild type (ϩ/ϩ) and PPAR␣-null (Ϫ/Ϫ) mice by Northern blotting (Fig. 4a). In wild type mice, fenofibrate treatment resulted in an ϳ5-fold increase of UGT2B mRNA levels compared with vehicle-treated animals. Interestingly, UGT2B transcripts were undetectable by this method in both fenofibrate-and vehicle-treated PPAR␣null mice. As control, 36B4 mRNA levels were similar in all of the groups. These data indicate that PPAR␣ is a crucial regulator of basal and fibrate-activated murine UGT2B gene expression.
To determine whether PPAR␣ also regulates murine UGT2B protein levels, liver microsomes from wild type and PPAR␣null mice were subjected to Western blot analysis using an anti-UGT2B antibody. In wild type mice, a pronounced increase in UGT2B protein levels was observed in fenofibratetreated compared with vehicle-treated animals (Fig. 4b). As for their mRNAs, UGT2B protein levels were almost undetectable in liver microsomes from PPAR␣-null mice and treatment with fenofibrate failed to increase UGT2B protein concentration (Fig. 4b). In fact, longer exposure of the Western blot revealed the presence of low amounts of UGT2Bs in PPAR␣-null mice, which were not affected by fenofibrate treatment (data not shown). These results clearly demonstrate that, similar to human UGT2B4, murine UGT2B enzymes are positively regulated PPAR␣ target genes.

PPAR␣ Activates the UGT2B4 Gene Promoter
To decipher the molecular mechanisms of human UGT2B4 induction by PPAR␣ activators, a 2.4-kb fragment of the UGT2B4 gene promoter cloned in front of the pGL3-luciferase reporter gene was transfected into HepG2 cells in the presence or absence of a PPAR␣ expression plasmid. Transfected cells were subsequently treated with the PPAR␣ ligand Wy 14643 (Fig. 5). Wy 14643 alone slightly induced UGT2B4 promoter activity, whereas co-transfection of PPAR␣ significantly enhanced Wy 14643-induced promoter activity to ϳ3-fold in HepG2 cells (Fig. 5). To localize the region within the UGT2B4 promoter that confers transcriptional responsiveness to PPAR␣ ligands, serial deletions from Ϫ2084 to Ϫ524 bp of the UGT2B4 promoter were also co-transfected with or without the expression vector for PPAR␣ (Fig. 5). A marked increase in reporter activities of the two larger fragments (Ϫ2084 and Ϫ1214 bp) was observed in HepG2 cells treated with Wy 14643, and co- transfection of PPAR␣ further increased the activities of the two constructs (Fig. 5). Further 5Ј deletion (Ϫ1149 and Ϫ524 bp) constructs were no longer induced by Wy 14643-activated PPAR␣, indicating that the region between Ϫ1214 and Ϫ1149 bp mediates the effect of PPAR␣ ligands on the UGT2B4 promoter. Identical results were obtained when these reported constructs were co-transfected with or without PPAR␣ in Huh7 cells (data not shown).

Identification of a PPRE within the UGT2B4 Gene Promoter
Consensus DR1 sites have been previously reported to bind the PPAR␣/RXR heterodimer (19). A computer-assisted analysis (42) of the Ϫ1214/Ϫ1149 region of the UGT2B4 promoter revealed the presence of a degenerated DR1 sequence, TGAACTTTAATCT, at positions from Ϫ1193 to Ϫ1180. To test whether this site mediates the induction by PPAR␣, mutations were introduced in this site in the context of the Ϫ2400 bp UGT2B4 promoter constructs (Fig. 6a). Mutation of this site abolished the induction of UGT2B4 promoter activity by Wy 14643-activated PPAR␣. Furthermore, the UGT2B4 DR1 site was cloned in multiple copies upstream of the luciferase reporter gene driven by the heterologous thymidine kinase promoter TKpGL3 and subsequently transfected in the presence or absence of the pSG5-PPAR␣ in HepG2 cells, which were treated or not with Wy 14643 (Fig. 6b). Reporter activity was increased upon co-transfection of constructs containing six copies of the wild type DR1 with the PPAR␣ plasmid. This activity was further enhanced by the addition of the PPAR␣ ligand (Fig.  6b). By contrast, no change in activity was observed when either the empty TKpGL3 vector (Fig. 6b) or the TKpGL3 vector containing three copies of the mutated DR1 (data not shown) was transfected. These results indicate that the Ϫ1193 to Ϫ1180 site in the UGT2B4 promoter is a positive PPRE.
To demonstrate that PPAR␣ binds to the PPRE identified in the UGT2B4 gene promoter, EMSAs were performed using a probe spanning nucleotides from Ϫ1199 to Ϫ1175 (B4-PPREwt) in the presence of in vitro translated PPAR␣ and RXR proteins (Fig. 7). As expected, neither RXR nor PPAR␣ alone bound the probe (Fig. 7a, lanes 2 and 3). By contrast, a clear shift was observed when this oligonucleotide was incubated in the presence of both RXR and PPAR␣ (Fig. 7a, lane 4). Furthermore, this complex was supershifted by the anti-PPAR␣ antibody (lane 5), thus demonstrating that the PPAR␣/RXR heterodimer specifically binds the Ϫ1193 DR1 site. By contrast, no protein-DNA complex was observed when mutated probes in the 5Ј and 3Ј half-sites (B4-PPREmt5Ј and B4-PPREmt3Ј, respectively) were tested (Fig. 7b, lanes 5-12). For competition experiments, increasing amounts (1-, 10-, and 50-fold excess) of unlabeled oligonucleotides encompassing either a consensus DR1 site (DR1cons), the B4-PPREwt, or the B4-PPREmt5Ј were added to binding reactions containing PPAR␣ in the presence of RXR (Fig. 7c). PPAR␣ binding to the B4-PPREwt was competed by the DR1 consensus site and by the B4-PPREwt (Fig. 7c). By contrast, the mutated B4-PPREmt3Ј did not compete for PPAR␣ binding to the DR1. Taken together, these data demonstrate that PPAR␣ binds to the PPRE site at nucleotides from Ϫ1193 to Ϫ1180 in the UGT2B4 gene promoter.

PPAR␣ and FXR Activators Additively Induce UGT2B4 Expression
We previously reported that CDCA-activated FXR positively regulates the expression of UGT2B4 in human hepatocytes and HepG2 cells (10). To test whether ligand-activated FXR and PPAR␣ can cooperate to regulate UGT2B4 expression, HepG2 cells were treated for 24 h with Wy 14643, CDCA, or both Wy 14643 and CDCA together. As expected, UGT2B4 mRNA levels were induced 2.6-fold by Wy 14643, whereas CDCA-induced UGT2B4 gene expression was ϳ10-fold (Fig. 8). Interestingly, cells treated with both PPAR␣ and FXR activators contained 14-fold higher concentrations of UGT2B4 transcripts, indicating that the two receptors coordinately regulate UGT2B4 gene expression.

DISCUSSION
In this study, we identify the human UGT2B4 enzyme as a positively regulated PPAR␣ target gene. UGT2B4 induction by fibrates occurs via PPAR␣ binding to a PPRE in the UGT2B4 promoter. Furthermore, we show that fenofibrate induces hepatic UGT2B mRNA and protein levels only in Sv/129 wild type FIG. 7. PPAR␣ binds to the PPRE in the UGT2B4 promoter. a and b, EMSAs were performed with end-labeled wild type or mutated B4-PPRE probes as indicated in the presence of unprogrammed reticulocyte lysate, RXR, PPAR␣, or both RXR and PPAR␣ as indicated. Supershift experiments were carried out using the anti-PPAR␣ antibody (0.2 g). c, competition EMSAs on radiolabeled B4-PPRE probe were performed by adding 1-, 10-, or 50-fold molar excess of the indicated cold consensus DR1 (DR1cons), B4-PPREwt, or B4-PPREmt5Ј oligonucleotides in EMSA with unprogrammed reticulocyte lysate, RXR, and/or PPAR␣. mice, whereas a drastically lowered expression of UGT2Bs is observed in livers from PPAR␣-null mice treated or not with fenofibrate. This observation demonstrates that PPAR␣ is a crucial regulator of both human UGT2B4 and murine UGT2B enzyme expression. Interestingly, PPAR␣ gene disruption also critically reduced the basal expression of mitochondrial fatty acid-metabolizing enzymes such as very long chain acyl-CoA dehydrogenase, long chain acyl-CoA dehydrogenase, and long chain acyl-CoA synthetase enzymes (43). Thus, the present findings demonstrate that in mice, PPAR␣ plays a crucial role in the constitutive expression of not only mitochondrial fatty acid-metabolizing enzymes but also microsomal UGT2B enzymes.
Considering the major role that UGT2B4 plays in BA glucuronidation, we hypothesized that UGT2B4 induction following PPAR␣ activation may affect BA glucuronidation in HepG2 cells. Indeed, we observed that Wy 14643-dependent PPAR␣ activation provoked a 2-fold increase of HDCA-glucuronidation activity. HDCA is a 6␣-hydroxylated metabolite of LCA, which is primarily excreted as a glucuronide derivative in urine (18,44). Because of its high degree of lipophilicity, LCA is a potent cholestatic agent and possesses an elevated cytotoxicity (45,46). However, conjugation of LCA with sulfate, a conjugation reaction catalyzed by the dehydroepiandrosterone sulfotransferase (SULT2A1) enzyme, allows an increased hydrosolubility of LCA and facilitates its biliary excretion (47)(48)(49). In addition to sulfation, LCA is efficiently 6␣-hydroxylated into HDCA by the hepatic CYP3A4 enzyme, and this modification facilitates its glucuronidation by UGT2B4 at the 6␣-hydroxy position prior to renal excretion (50). Thus, glucuronidation of HDCA has been proposed as an alternative mechanism for reducing the hepatic toxicity of monohydroxylated LCA (44,50). Recent studies indicate that the BA sensors pregnane X-receptor (PXR) and FXR play important roles in LCA detoxification. As such, activation of PXR induces both SULT2A1 and CYP3A4 expression, whereas BA-activated FXR stimulates SULT2A1 and UGT2B4 expression (10,44,49,51). Results from the present study prove that PPAR␣ also participates in the control of LCA detoxification in addition to PXR and FXR (Fig. 9). Furthermore, PPAR␣, FXR, and PXR inhibit CYP7A1 expression (26,52,53), thus suggesting that the three receptors may cooperate to control BA homeostasis and detoxification by both reducing BA synthesis and inducing their metabolism (Fig. 9). Recently, PPAR␣ was identified as a FXR target gene, thus providing molecular evidence for a cross-talk between the FXR and PPAR␣ transcriptional pathways in humans (54). Considering that UGT2B4 expression is also up-regulated upon CDCA activation of FXR (10), we investigated whether this cross-talk between FXR and PPAR␣ can affect UGT2B4 expression in HepG2 cells. We observed that upon ligand activation, PPAR␣ and FXR act in concert to stimulate BA glucuronidation. Overall, these results demonstrate that FXR and PPAR␣ control not only the same BA-metabolizing enzyme but also share cooperative activity to induce BA glucuronidation catalyzed by UGT2B4. It would be interesting to determine whether a sim- ilar cross-talk also exists between PXR and PPAR␣ and/or FXR.
UGT2B4 is considered to be the specific BA-conjugating UGT enzyme in human liver, although it also participates to the glucuronide conjugation of a wide variety of endogenous or exogenous compounds. As such, various C19-steroids such as androstane-3␣,17␤-diol are substrates for UGT2B4 (5,6). In the Helsinki Heart Study population, gemfibrozil treatment resulted in a 3-fold elevation of plasma 3␣,17␤-diol glucuronide levels (31), which may reflect an increased expression and activity of UGT2B4 in these patients. A recent study in nonhuman primates revealed that UGT enzymes expressed in androgen target tissues glucuronidate, preferentially C19-steroids (55), suggesting that UGTs participate in the control of intracellular levels of the active androgen. It would be of interest to determine whether PPAR␣ activation also affects androgen glucuronidation in a tissue such as the prostate where both PPAR␣ and UGT2B4 are expressed (5,56). Based on the present study, it is tempting to speculate that fibrate treatment may induce androgen glucuronidation and that PPAR␣ can be a potential regulator of androgen levels in such a tissue.
UGT2B4 is also involved in the inactivation of various xenobiotics, such as phenolic and monoterpenoid compounds (4,5,7). Interestingly, Kok et al. (29) reported that ciprofibrate induces the hepatic expression of the multidrug resistance (Mdr2) gene in a PPAR␣-dependent manner in mice. P-glycoprotein, the Mdr2 gene product, is a hepatocyte transporter located on the canalicular membrane (57), which has a broad substrate specificity that encompasses glucuronide conjugates of a variety of endobiotics and xenobiotics (29,(57)(58)(59). Thus, by stimulating both glucuronidation and transport, PPAR␣ appears to be a key factor for the elimination of many endogenous and exogenous glucuronide derivatives from the liver, at least, in rodents. The role of different nuclear receptors, such as PXR and constitutive androstane receptor, in the control of xenobiotic metabolizing enzyme expression has been fully characterized, whereas PPAR␣ received less attention regarding xenobiotic metabolizing enzyme regulation. Nevertheless, the present findings added to previous reports indicate that PPAR␣ is also an important xenobiotic sensor that regulates both phase I (CYP1A2, 2A1, 2B1, and 2B2) and phase II (glutathione S-transferase A1, glutathione S-transferase M2, UGT1A9, and UGT2B4) enzymes (60 -63).
In conclusion, the present study illustrates for the first time the implication of PPAR␣ in the control of BA glucuronidation and more generally reinforces the role of this nuclear receptor as a regulator of endobiotic and xenobiotic metabolism.