Abstract
The aim of this study was to determine the role of pregnane X receptor (PXR) in the induction of UDP-glucuronosyltransferases (UGTs) by pregnenolone-16α-carbonitrile (PCN). Four- to six-month-old male wild-type and PXR-null mice received control or PCN-treated (1500 ppm) diet for 21 days. On day 22, livers were taken to prepare microsomes and total RNA to determine UGT activity and mRNA levels, respectively. In wild-type mice, PCN treatment significantly increased UGT activities toward bilirubin, 1-naphthol, chloramphenicol, thyroxine, and triiodothyronine. On control diet, the UGT activities toward the above substrates (except for 1-naphthol) in the PXR-null mice were significantly higher than those of wild-type mice. However, UGT activities in PXR-null mice were not increased by PCN. In agreement with the above findings, mRNA levels of mouse Ugt1a1 and Ugt1a9, which are involved in the glucuronidation of bilirubin and phenolic compounds, were increased about 100% in wild-type mice following PCN treatment, whereas the expression of Ugt1a2, 1a6, and 2b5 was not affected. In contrast, PCN treatment had no effect on the mRNA levels of these UGTs in PXR-null mice. Taken together, these results indicate that PCN treatment induces glucuronidation in mouse liver, and that PXR regulates constitutive and PCN-inducible expression of some UGTs.
Glucuronidation is a phase II biotransformation pathway that plays a major role in the metabolism and elimination of hydrophobic compounds, including environmental pollutants, drugs, steroid hormones, bilirubin, and thyroid hormones [thyroxine (T42) and triiodothyronine (T3)] (Burchell et al., 1998). This reaction is catalyzed by UDP-glucuronosyltransferases (UGTs), which are located in endoplasmic reticulum of liver and other tissues. Molecular cloning identified two families of UGTs, UGT1 and UGT2 (Mackenzie et al., 1997). UGT1 family members are encoded by UGT1 gene locus, which can potentially generate up to 12 isozymes with unique amino-terminal sequences but an identical carboxyl-terminal sequence by a mechanism of exon sharing (Ritter et al., 1992; Emi et al., 1995). In contrast, UGT2 family members appear to be encoded by independent genes.
Treatment of rats with prototypical microsomal enzyme inducers (MEIs), such as 3-methylcholanthrene (3-MC), polychlorinated biphenyls, phenobarbital (PB), pregnenolone-16α-carbonitrile (PCN), and clofibrate results in differential induction of UGT activities in liver. 3-MC and other cytochrome P450 1A enzyme inducers increase the glucuronidation of planar chemicals, such as 1-naphthol and 4-nitrophenol. PB and other CYP2B enzyme inducers tend to induce the glucuronidation of bulky chemicals, such as chloramphenicol and morphine (Bock et al., 1973). PCN and other CYP3A enzyme inducers induce UGT activities toward digitoxigenin-monodigitoxoside and bilirubin, whereas clofibrate and other CYP4A enzyme inducers increase the glucuronidation of bilirubin but not digitoxigeninmonodigitoxoside (Watkins et al., 1982; Watkins and Klaassen, 1982). The differential induction of UGTs in rats was the subject of intense research for many years, contributing greatly to the characterization of UGT isozymes.
The recent discovery of several receptors within the cell that function as ligand-activated transcription factors has shed significant light on the molecular mechanisms for the up-regulation of many phase I and phase II biotransformation enzymes following treatment with MEIs. For instance, the cytosolic aryl hydrocarbon receptor is involved in the induction of CYP1A1 by polycyclic aromatic hydrocarbons (Whitlock et al., 1996), whereas three members of the nuclear receptor family, the constitutive androstane receptor (CAR), the pregnane X receptor (PXR), and the peroxisome proliferator-activated receptor α mediate the induction of CYP2B (Honkakoski et al., 1998), CYP3A (Kliewer et al., 1998), and CYP4A (Muerhoff et al., 1992) by xenobiotics, respectively. Generally, the transcriptional regulation via CAR, PXR, and peroxisome proliferator-activated receptor α involves the formation of heterodimers between these receptors and 9-cis retinoic acid receptor, binding of the heterodimer to response elements in the regulatory region of target genes, and ligand-dependent trans-activation of gene expression.
The involvement of the aforementioned receptors in the induction of the UGTs by MEIs is not well characterized in comparison to cytochromes P450, in spite of some recent progress, such as the recent characterization of a PB response enhancer module in human UGT1A1 gene that can be activated by PB in the presence of CAR (Sugatani et al., 2001). Although the induction of hepatic UGT activities by the potent PXR ligand PCN in rats is well documented in the literature (Watkins et al., 1982; Watkins and Klaassen, 1982), few studies have been conducted to determine the effects of PCN on mRNA levels of UGTs in rodents and other species, or the role of PXR in the regulation of UGTs, until recently. In a study that was aimed at exploring the regulatory roles of PXR and CAR in different phases of xenobiotic metabolism, Maglich et al. (2002) showed that PCN increases mRNA levels of mouse Ugt1a1 in a PXR-dependent manner. However, it remains unclear whether other mouse UGTs are similarly up-regulated by PCN and whether increases in levels of gene transcripts are coupled with corresponding changes in levels of protein and enzyme activity.
The present study was conducted to address these important issues by determining the effects of PCN on microsomal glucuronidation of five representative aglycones (i.e., bilirubin, 1-naphthol, chloramphenicol, T4, and T3) and mRNA levels of UGT genes in liver of wild-type and PXR-null mice. According to the nomenclature of mouse UGTs (Mackenzie et al., 1997), cDNA sequence information for five mouse UGT genes is available. Four of them belong to the UGT1 family, namely mouse Ugt1a1, 1a2, 1a6, and 1a9, and the other one is Ugt2b5. Consequently, gene-specific probe sets for these UGTs were generated to measure mRNA levels using QuantiGene signal amplification assay. Up-regulation of UGT gene transcripts by PCN was related to inducibility of hepatic glucuronidation activities. To our knowledge, this is the first comprehensive study on the inducibility of the entire collection of cloned mouse UGTs by PCN and possible involvement of PXR in UGT induction.
Materials and Methods
Materials. UDP-glucuronic acid (UDP-GA), PCN, T4, T3, 1-naphthol, chloramphenicol, 1-[1-14C]-naphthol, and 3-[(3-chlolamidopropyl) dimethylammonio]-1-propanesulfonic acid (CHAPS) were purchased from Sigma-Aldrich (St. Louis, MO). Bilirubin and ethyl anthranilate were purchased from Aldrich Chemical (Milwaukee, WI). [125 I]-T4, [125I]-T3, and d-threo-[dichloroacetyl-1, 2-14C]-chloramphenicol were obtained from PerkinElmer Life Sciences (Boston, MA). All other reagents used were acquired from the above suppliers and were of reagent grade or better.
Animals and Treatment. Wild-type and PXR-null mice as described by Staudinger et al. (2001a) were bred in the laboratory animal facilities of the University of Kansas Medical Center. Animals were housed in polypropylene cages (less than four animals per cage) on corn-cob bedding. Animal room temperature was maintained at approximately 21°C with a 12-h light cycle. Four- to six-month-old male mice were fed either control or PCN-treated (1,500 ppm) diet for 21 days. Animals had free access to feed and water. Feed consumption and body weights were recorded every 2 days. On day 22, mice were decapitated. Livers were immediately removed and snap-frozen in liquid nitrogen. Tissue samples were stored at –80°C until assayed.
Preparation of Microsomes. Microsomes were prepared by ultracentrifugation (100,000g for 60 min) of the postmitochondrial supernatant (10,000g for 20 min) from a 10% liver homogenate (w/v), prepared in 50 mM Tris-HCl (pH 7.4) containing 150 mM KCl as described by Lu and Levin (1972). The microsomal fraction was washed by homogenization in 10 mM EDTA (pH 7.4) containing 150 mM KCl. After ultracentrifugation, the resulting pellets were covered with a small volume of 250 mM sucrose and stored at –80°C.
UGT Activity toward Bilirubin. The method of Heirwegh et al. (1972) was followed to determine UGT activity toward bilirubin. The microsomal pellets were suspended in 0.25 M sucrose (0.4 g eq. wet weight of liver/ml). The resulting preparation was further diluted 1:1 with digitonin solution (5.4 mg/ml) and agitated for at least 20 min at 4°C before use. Reaction mixtures contained 200 μl of 0.5 M triethanolamine-HCl buffer (pH 7.7), 40 μl of 125 mM MgCl2, 200 μl of bilirubin-albumin mixture (0.25 mg/ml), 200 μl of digitonin-activated microsomal preparation, and 20 μl of UDP-GA (77 mM). After incubation at 37°C for 15 min, reactions were stopped by adding 2 ml of glycine-HCl buffer (pH 2.7). The tubes were placed in a water bath at 25°C. Ethyl anthranilate diazo reagent (1 ml) was added, and diazo-coupling was allowed to proceed at 25°C for 30 min. The reaction was terminated by an 8-min incubation with 0.5 ml of freshly prepared ascorbic acid (100 mg/ml). Tubes were placed on ice to cool. The contents of each tube were then shaken vigorously (30 times) with 2 ml of 2-pentanone/n-butyl acetate (17:3, v/v). The tubes were placed in –20°C freezer for a few min. The contents of each tube were then vortexed thoroughly. Following centrifugation (1,000g for 5 min), absorbance of pigment in the resulting organic phases was measured at 546 nm using the extraction solvent as a blank reference.
UGT Activities toward 1-Naphthol and Chloramphenicol. UGT activities toward 1-naphthol and chloramphenicol were assayed in both native and activated microsomal preparation according to the method of Hazelton et al. (1985) with minor modification. Briefly, the microsomal pellets were suspended in 0.25 M sucrose (1 g eq. wet weight of liver/ml). A portion of this suspension was diluted 1:1 with 0.25 M sucrose, and the resulting preparation was termed “native” microsomes. A second portion of the suspension was diluted with an equal volume of 0.25 M sucrose containing 16 mM CHAPS. This preparation was termed “activated” microsomes. Both of the diluted microsomal preparations were agitated for at least 20 min at 4°C. The reaction mixture for both assays was 0.2 M Tris-HCl (pH 7.5), 10 mM MgCl2, 2.2 mM saccharic acid-1,4-lactone, 1-naphthol (0.5 mM, 0.04 μCi), or chloramphenicol (2.0 mM, 0.4 μCi), and microsomal preparation (0.1 mg and 0.25 mg of protein for 1-naphthol and chloramphenicol assay, respectively) in a final volume of 0.5 ml. Reactions were initiated by the addition of 4 mM UDP-GA. Blank controls received water and 0.25 M sucrose with or without CHAPS instead of UDP-GA and microsomal preparation. After incubation at 37°C for 15 min (1-naphthol) or 20 min (chloramphenicol), reactions were stopped by the addition of ice-cold ethanol (1-naphthol) or water (chloramphenicol). The latter was then heated in boiling water for 45 s. The parent compounds were separated from the glucuronidated products by extraction with chloroform (1-naphthol) and isoamyl acetate (chloramphenicol). After extraction, radioactivity in an aliquot of the aqueous phase was determined with a liquid scintillation counter.
UGT Activities toward T4 and T3. UGT activities toward T4 and T3 were determined as described by Hood and Klaassen (2000). Reaction mixture (final volume, 150 μl) was made up of 75 mM Tris-HCl, 7.5 mM MgCl2, 30 mM UDP-GA, 1 μM T4 or T3 (approximately 100,000 cpm), and 0.1 mM propylthiouracil (to inhibit outer-ring deiodinase activity). Reactions were started by adding 50 μl of protein (final concentration of 250 μg/ml for T4 and 81 μg/ml for T3) and incubated at 37°C for 60 min. The reactions were stopped by adding 200 μl of ice-cold methanol. 125I-T4-glucuronide or 125I-T3-glucuronide was separated from unconjugated T4 or T3 with a Sephadex LH-20 column (1-ml bed volume) (Amersham Biosciences Inc., Piscataway, NJ). The amount of 125I-T4 or 125I-T3 in the eluates was quantified using a Packard gamma counting system (PerkinElmer Life Sciences, Boston, MA).
Isolation of Total RNA. Total RNA was isolated using RNAzol Bee reagent (Tel-Test Inc., Friendswood, TX) according to the manufacturer's instruction and resuspended in diethyl pyrocarbonate-treated water. The concentration and purity of total RNA in each sample were assessed by measuring absorbance at 260 and 280 nm. The integrity of RNA samples was verified by checking the integrity of 18S and 28S rRNA separated by formaldehydeagarose gel electrophoresis.
Design of Specific Oligonucleotide Probe Sets for the Analysis of Mouse UGT mRNA Levels Using QuantiGene Signal Amplification Assay. The nomenclature for mouse UGTs as described by Mackenzie et al. (1997) was followed in this study. The GenBank accession numbers for the mouse UGTs examined, namely Ugt1a1, 1a2, 1a6, 1a9, and 2b5, are listed in Table 1, despite the existence of multiple GenBank accession numbers for the same or portions of the same coding sequence for most of these genes. For UGT1A family members, cDNA sequences were subjected to multiple alignment analysis (CLUSTAL W; available at http://www.ddbj.nig.ac.jp) and only the 5′-variable regions were used as the target sequences for probe design. For mouse Ugt2b5, the entire cDNA sequence was used for probe design. These target sequences were analyzed by ProbeDesigner software version 2.0 (Bayer Corp.-Diagnostics Div., Tarrytown, NY) to generate three groups of oligonucleotide probes (i.e., capture extender, label extender, and blocker probe). Their functions in the assay were described previously (Hartley and Klaassen, 2000). All probes were designed with a Tm of approximately 63°C to allow hybridization under constant conditions (i.e., 53°C). To ensure minimal cross-reactivity with other mouse sequences, each candidate probe was submitted to the National Center for Biotechnological Information for nucleotide comparison by the basic local alignment search tool (BLASTn). Oligonucleotides with a high degree of similarity (>80%) to other mouse gene transcripts were eliminated from the design. Detailed information about the final probe sets is listed in Table 1. Each probe were synthesized on a 50-nmol synthesis scale by Operon Technologies (Alameda, CA) and obtained desalted and lyophilized. Probes were diluted in 0.5 ml of 1× Tris EDTA buffer (pH 8.0) and stored at –20°C.
List of oligonucleotide probes generated for the analysis of mRNA levels of mouse UGTs using QuantiGene signal amplification assay.
QuantiGene Signal Amplification Assay. The application of QuantiGene signal amplification assay to the measurement of mRNA levels of drug-metabolizing enzymes was extensively described and validated by Hartley and Klaassen (2000). Briefly, capture extenders, label extenders, and blocker probes for each specific UGT were combined and diluted to 50, 200, and 100 fmol/μl, respectively, in the lysis buffer. All reagents for analysis (i.e., lysis buffer, capture hybridization buffer, amplifier/label probe buffer, and substrate solution) were supplied in the assay kit (Bayer Diagnostics). Total RNA (10 μg) was added to each well of a 96-well plate along with 50 μl of capture hybridization buffer and 50 μl of each diluted probe set. RNA was allowed to hybridize for at least 16 h at 53°C. Plates were then rinsed twice with wash buffer (400 μl). Amplifier molecules (100 μl) diluted in amplifier/label probe buffer (1:1,000) were added to each well, and plates were incubated at 46°C for 60 min. Plates were rinsed again with wash buffer followed by the addition of 100 μl of label probe (1:1,000 in amplifier/label buffer) to each well. After incubation at 46°C for 60 min, plates were rinsed twice with wash buffer. Alkaline phosphatase-mediated luminescence was triggered by the addition of a dioxetane substrate solution (100 μl/well). The enzymatic reaction was allowed to proceed for 30 min at 37°C, and luminescence was measured with the Quantiplex 320 bDNA Luminometer (Bayer Diagnostics) interfaced with Quantiplex Data Management Software version 5.02 (Bayer Diagnostics) for analysis of luminescence from 96-well plates.
Statistical Analysis. Results were expressed as means ± standard error of means (S.E.). Treatment groups consisted of 6 to 8 animals. Difference between groups was analyzed using analysis of variance followed by Newman-Keuls test. p values less than 0.05 were considered statistically significant.
Results
PCN treatment in rodents results in an increase in liver mass (Japundzic et al., 1974). This effect is absent in PXR-null mice, suggesting that the nuclear receptor is required for PCN-induced hepatomegaly (Staudinger et al., 2001b). In the present study, liver-to-body-weight ratio in wild-type mice was significantly increased following 21-day PCN treatment. PCN treatment did not increase liver mass in the null mice (Fig. 1). These data indicate the absence of functional PXR in the null mice. However, a significant increase in liver-to-body-weight ratio was observed for PXR-null mice in comparison to control wild-type mice but was not reported in an earlier study (Staudinger et al., 2001b). The basis for this difference is unclear.
Effect of PCN treatment on liver-to-body weight ratio in wild-type and PXR-null mice.
Values are expressed as the mean ± S.E. (n = 6–8). Bars not sharing at least one letter are statistically different from each other (p < 0.05).
Effect of PCN Treatment on Hepatic UGT Activities. Hepatic microsomal glucuronidation of five different aglycones (i.e., bilirubin, 1-naphthol, chloramphenicol, T4, and T3) was measured to determine the effects of PCN on UGT activities. As shown in Fig. 2, a 2.1-fold increase in hepatic UGT activity toward bilirubin was seen in wild-type mice following PCN treatment. PCN treatment, however, had no effect on bilirubin UGT activity in PXR-null mice, although the basal activity in the null mice was about 125% higher than that of wild-type mice.
Effect of PCN treatment on hepatic UDP-glucuronosyltransferase activity toward bilirubin in wild-type and PXR-null mice.
Values are expressed as the mean ± S.E. (n = 6–8). Bars not sharing at least one letter are statistically different from each other (p < 0.05).
The glucuronidation of 1-naphthol and chloramphenicol was measured using native and detergent-treated microsomal preparations as described previously (Hazelton et al., 1985). UGT activity to 1-naphthol was increased by 360 to 500% by the detergent treatment, whereas that to chloramphenicol was increased by only 45% (Fig. 3). Nevertheless, PCN increased the UGT activities to 1-naphthol and chloramphenicol by about 40 and 60%, respectively, in both native and activated microsomal preparations. In contrast, 1-naphthol and chloramphenicol UGT activities in PXR-null mice were not increased by PCN. Furthermore, basal levels of UGT activity to chloramphenicol, but not 1-naphthol, in the null mice were significantly higher than those of wild-type mice.
Effect of PCN treatment on hepatic UDP-glucuronosyltransferase activities toward 1-naphthol and chloramphenicol in wild-type and PXR-null mice.
Enzyme activities were measured in both native and detergent-treated (activated) microsomes as described under Material and Methods. Values are expressed as the mean ± S.E. (n = 6–8). Bars not sharing at least one letter are statistically different from each other (p < 0.05).
Glucuronidation of T4 and T3 results in inactivation of their biological activities and facilitates their elimination from the body. In wild-type mice, PCN increased T4 and T3 glucuronidation by 80 and 40%, respectively (Fig. 4). However, no difference in glucuronidation of thyroid hormones was observed between groups of control and PCN-treated PXR-null mice. UGT activity toward T4 and T3 of PXR-null mice on control diet was about 35 and 30% higher than that of wild-type controls, respectively.
Effect of PCN treatment hepatic UDP-glucuronosyltransferase activities toward thyroxine (T4) and triiodothyronine (T3) in wild-type and PXR-null mice.
Values are expressed as the mean ± S.E. (n = 6–8). Bars not sharing at least one letter are statistically different from each other (p < 0.05).
Effect of PCN Treatment on Hepatic mRNA Levels of UGTs. Despite the across-the-board increase in hepatic glucuronidation of a range of aglycones shown above, little information on the role of PXR in the regulation of a particular mouse UGT isozyme can be deduced from these results. Therefore, oligonucleotide probe sets specific to the known mouse UGT genes were designed and used to determine the effects of PCN on mRNA levels of these genes using the QuantiGene signal amplification assay.
In wild-type mice, PCN increased the hepatic mRNA levels of Ugt1a1 and 1a9 by 190 and 80%, respectively, but had no effect on those of Ugt1a2 and 1a6 (Fig. 5). In PXR-null mice, the mRNA levels of these four UGT1 family members were not significantly increased by PCN treatment. Furthermore, Ugt1a1 mRNA levels of the null mice on control diet were 100% higher than those of wild-type controls. Therefore, the PXR-dependent increase in Ugt1a1 and 1a9 gene transcripts and the higher basal levels of Ugt1a1 message seen in the null mice as compared with the wild-type mice are in accordance with the effects of PCN on hepatic UGT activities as well as the higher constitutive glucuronidation activities seen in the null mice. The value of relative light unit detected using the Ugt1a2 probeset was very low (less than 1); thus, suggesting that this gene is not as highly expressed in liver as the other members of UGT1 family examined.
Effect of PCN treatment on mRNA levels of mouse UGT1 family members, namely, Ugt1a1, 1a2, 1a6, and 1a9 in liver of wild-type and PXR-null mice.
Values are expressed as the mean ± S.E. (n = 6–8). Bars with different letters are statistically different from each other (p < 0.05).
As shown in Fig. 6, Ugt2b5 mRNA levels in liver of both wild-type and PXR-null mice were not affected by PCN treatment. Also, no difference in the basal levels of Ugt2b5 gene transcript was observed between wild-type and null mice.
Effect of PCN treatment on mRNA levels of mouse Ugt2b5 in liver of wild-type and PXR-null mice.
Values are expressed as the mean ± S.E. (n = 6–8). Bars with different letters are statistically different from each other (p < 0.05).
Discussion
PCN belongs to a group of steroids known as “catatoxic steroids”, which are named after their ability to afford protection against various types of intoxication by accelerating the elimination of harmful chemicals from the body (Kourounakis et al., 1977). Treatment of rodents with PCN results in the induction of drug metabolism enzymes and transport systems including CYP3A (Kliewer et al., 1998), UGTs (Watkins and Klaassen, 1982; Hazelton and Klaassen, 1988), sulfotransferases (Liu and Klaassen, 1996), organic anion transporting polypeptide 2 (Guo et al., 2002), and multidrug resistance protein 2 (Johnson and Klaassen, 2002). The involvement of the nuclear receptor PXR in the up-regulation of CYP3A, Oatp2, and sulfotransferase by PCN treatment was recently established (Staudinger et al., 2001a; Sonoda et al., 2002). Along the same line, the data presented herein showed that hepatic UGT activities to the five representative aglycones as well as the steady-state mRNA levels of Ugt1a1 and 1a9 were increased in a PXR-dependent manner, suggesting that induction of the major phase II biotransformation family of enzymes, namely the UGTs, is also under control by PXR. Thus, these findings, along with reports by others (Kliewer et al., 1998; Staudinger et al., 2001a; Sonoda et al., 2002) support the notion that PXR is an important component of the body's adaptive defense mechanism against toxic substances.
In the present study, basal levels of hepatic UGT activities toward bilirubin, chloramphenicol, T4, and T3 were significantly higher in PXR-null mice than wild-type mice. This is accompanied by about a doubling in basal levels of Ugt1a1 gene transcripts in null mice as compared with wild types, but the difference is marginally insignificant (p = 0.072). Similarly, Staudinger et al. (2001a,b) reported a slight but statistically significant increase in Cyp3a11 mRNA levels and activity as indicated by testosterone 6β-hydroxylation activity and zoxazolamine resistance in the same line of PXR-null mice. In contrast, reports by Xie et al. (2000) showed that ablation of PXR function has no effect on the basal levels of Cyp3a11 gene transcription and CYP3A activity in another independently developed PXR-null mouse model. Despite the lack of insight into the basis for the discrepancy between the two lines of PXR-null mice, the findings of the present study, together with those of Staudinger et al. (2001a,b), support possible involvement of PXR in the control of constitutive expression of biotransformation enzymes such as UGTs and CYP3A.
The induction of hepatic UGT activities to bilirubin, 1-naphthol, chloramphenicol, T4, and T3 by PCN treatment, seen in the present study does not agree with the findings of a previous study (Viollon-Abadie et al., 1999), which showed no induction of UGT activities toward bilirubin, p-nitrophenol, androsterone, T4, and T3, in OF-1 mice treated with PCN p.o. for 14 days. The results of the present study, however, agree with the dose-dependent increase in hepatic glucuronidation of chloramphenicol, 1-naphthol, and T4 in B6C3F1 mice given PCN-treated diets for 21 days (Hood et al., 2003). Furthermore, Maglich et al. (2002) observed a PXR-dependent induction of UGT1A1 in the same line of wild-type and PXR-null mice following PCN treatment (two i.p. injections within 24 h). In the present study, the estimated daily dose of PCN is 125–175 mg/kg (average body weight of the animals, 30 g; daily food intake, 2.5–3.5 g per animal). This dose range of PCN is close to that used in the other three previous studies (about 100 mg of PCN/kg). Although the dosage and duration of PCN treatment in the current study and the study by Allen et al. (submitted) are similar to those in the study of Viollon-Abadie et al. (1999), the results of the two studies that used either B6C3F1 mice or PXR-null and wild-type mice differ drastically from those of the study that used OF-1 mice. In contrast, a PXR-dependent induction of UGT(s) was observed in the same line of PXR-null and wild-type mice treated either acutely (Maglich et al., 2002) or sub-chronically with PCN (the present study). Therefore, it seems that a strain difference in the control of UGT induction following PCN treatment may exist in mice. This conclusion is analogous to the difference in induction of phenol UGT activities observed in C57BL/6 and DBA/2 mice, which are responsive and nonresponsive to 3-MC-type inducers, respectively (Bock et al., 1982). Further studies are necessary to elucidate possible mechanisms underlying this strain difference in UGT induction by PCN.
Generally, UGT1 isozymes are categorized into the following two groups based on substrate preferences: bilirubin UGT and 3-MC inducible phenol UGT. UGT1A1 is the most important glucuronosyltransferase for bilirubin glucuronidation in human and rats, whereas human UGT1A6 and 1A9 as well as rat UGT1A6 and 1A7 are the major transferases for phenol glucuronidation (Burchell et al., 1998). Human UGT1A1 is inducible by PB. In contrast, bilirubin glucuronidation in rats is not readily inducible by PB but can be greatly induced by the hypolipidemic agent clofibrate as well as PCN (Bock et al., 1973; Watkins et al., 1982; Watkins and Klaassen, 1982). In the present study, PCN treatment increased hepatic UGT activities to bilirubin, and the phenolic compound 1-naphthol in mice. The increases in UGT activities were accompanied by parallel changes in mRNA levels of Ugt1a1 and Ugt1a9 but not Ugt1a2, 1a6, and 2b5. These findings agree with the predicted function of mouse UGT1A1 as bilirubin UGT and mouse UGT1A9 as phenol UGT, based on their high sequence similarity to rat and human bilirubin and phenol UGT isoform(s) (Kong et al., 1993). However, an in vitro enzyme expression study, so far, has not been conducted to confirm the substrate specificity of these two isozymes. Interestingly, mouse UGT1A6, a known phenol UGT (Lamb et al., 1994), was not induced by PCN in the present study. Therefore, it seems that the two phenol UGTs in mice, namely UGT1A6 and UGT1A9, are differentially regulated by PCN. The existence of distinct promoters for the multiple first exons encoding the variable regions of UGT1 family members is the underlying mechanism for the differential induction of these isozymes.
Hepatic microsomal glucuronidation of chloramphenicol is not impaired in Gunn rats, which lack UGT1 enzyme activities due to a genetic mutation in the common region of UGT1 (Watkins and Klaassen, 1982). These findings suggest that the glucuronidation of chloramphenicol is catalyzed by UGTs other than those belonging to the UGT1 family. In the present study, hepatic UGT activity to chloramphenicol was induced by PCN in wild-type mice but not in the null mice. However, PCN treatment had no effect on mRNA levels of Ugt2b5, a mouse UGT outside the UGT1 family (Kimura and Owens, 1987). Thus, the identity of the UGT isozyme(s) involved in chloramphenicol glucuronidation remains unclear. The most likely explanation for this observation is that at least one other UGT2 family member is involved in the glucuronidation of chloramphenicol, and is inducible by PCN via the PXR regulatory pathway.
Glucuronidation of thyroid hormones and subsequent excretion of the conjugates into bile serve as the major route for the elimination of T4 and T3. In rats, UGT activity to T4 is increased following treatment with PCN, PB, 3-MC, and polychlorinated biphenyl, among which only PCN and PB produced an increase in T3-UGT activity (Barter and Klaassen, 1992; Liu et al., 1995; Hood and Klaassen, 2000). Although all of these compounds decrease serum T4 concentrations, only PCN and PB are associated with increase in circulating concentrations of thyroid stimulating hormone (TSH) (Barter and Klaassen, 1994; Liu et al., 1995), which regulates thyroid hormone synthesis. Increased serum TSH concentrations are of concern because sustained increases in TSH have been associated with chemicals such as propylthiouracil and potassium percholorate that induce thyroid cancer in lab animals (Kanno et al., 1990; Hood et al., 1999). Thus, it would be of interest to determine which UGT isozyme(s) catalyze(s) the glucuronidation of thyroid hormones, and how xenobiotics induce UGT activities toward thyroid hormones. With such information, a screening assay for potential thyroid tumor promoters could be developed. More than one UGT isozyme is thought to be responsible for glucuronidation of thyroid hormones in liver. In rats, T4 glucuronidation is catalyzed by both bilirubin UGT and 3-MC-inducible phenol UGT, whereas androsterone-metabolizing UGT2B2 may be the isozyme for T3 glucuronidation (Visser et al., 1993). In humans, convincing results showed that UGT1A1 and 1A9 catalyze T4 glucuronidation, whereas studies on the role of UGTs in T3 metabolism remain inconclusive (Findlay et al., 2000). In contrast, little is known about the identity of UGTs responsible for the glucuronidation of thyroid hormones in mice. The present study shows for the first time that treatment of mice with PCN increased hepatic UGT activities toward T4 and T3 in a PXR-dependent manner, which is correlated with changes in mRNA levels of mouse Ugt1a1 and Ugt1a9. These results suggest that thyroid hormones may be glucuronidated in mice by bilirubin and phenol UGTs, which are regulated by the ligand-activated transcription factor PXR. It should be noted that our results cannot exclude the possibility that thyroid hormones are glucuronidated by some unknown PCN-inducible UGTs. Future studies are needed to determine whether induction of thyroid hormone glucuronidation by PCN in mice would be associated with disturbed thyroid homeostasis, as seen in rats.
In conclusion, the present study provides evidence supporting the involvement of PXR in the constitutive expression and induction of UGTs by PCN. In light of the major role of hepatic glucuronidation played in the metabolism of a wide range of endogenous and xenobiotic compounds, these findings further emphasize the pivotal role of PXR in the control of biotransformation processes.
Footnotes
-
↵1 Current address: Department of Pharmacology and Toxicology, University of Kansas, Lawrence, KS 66045.
-
↵2 Abbreviations used are: T4, thyroxine; T3, triiodothyronine; UGTs, UDP-glucuronosyltransferases; MEIs, microsomal enzyme inducers; 3-MC, 3-methylcholanthrene; PB, phenobarbital; PCN, pregnenolone-16α-carbonitrile; CAR, constitutive androstane receptor; PXR, pregnane X receptor; RXR, 9-cis retinoic acid receptor; GA, glucuronic acid; CHAPS, 3-[(3-chlolamidopropyl) dimethylammonio]-1-propanesulfonic acid; TSH, thyroid stimulating hormone; UGTs, UDP-glucuronosyltransferases.
-
This work was supported by National Institutes of Health Grant ES-08156 to Curtis D. Klaassen.
- Received January 7, 2003.
- Accepted March 25, 2003.
- The American Society for Pharmacology and Experimental Therapeutics
