Transcriptional Regulation by Triiodothyronine of the UDP-glucuronosyltransferase Family 1 Gene Complex in Rat Liver

Transcription

1 THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 272, No. 27, Issue of July 4, pp , by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. Transcriptional Regulation by Triiodothyronine of the UDP-glucuronosyltransferase Family 1 Gene Complex in Rat Liver COMPARISON WITH INDUCTION BY 3-METHYLCHOLANTHRENE* (Received for publication, January 31, 1997, and in revised form, April 30, 1997) Taoufik Masmoudi, Abdelmadjid K. Hihi, Manuel Vázquez, Yves Artur, Béatrice Desvergne, Walter Wahli, and Hervé Goudonnet From the Laboratoire de Biochimie Pharmacologique, Faculté de Pharmacie, Université de Bourgogne, 7. Bv. Jeanne d Arc, Dijon 21033, France and the Institut de Biologie Animale, Bâtiment de Biologie, Université de Lausanne,CH-1015 Lausanne, Switzerland This study demonstrates that the expression of the phenol UDP-glucuronosyltransferase 1 gene (UGT1A1) is regulated at the transcriptional level by thyroid hormone in rat liver. Following 3,5,3 -triiodo-l-thyronine (T 3 ) stimulation in vivo, there is a gradual increase in the amount of UGT1A1 mrna with maximum levels reached 24 h after treatment. In comparison, induction with the specific inducer, 3-methylcholanthrene (3-MC), results in maximal levels of UGT1A1 mrna after 8hof treatment. In primary hepatocyte cultures, the stimulatory effect of both T 3 and 3-MC is also observed. This induction is suppressed by the RNA synthesis inhibitor actinomycin D, indicating that neither inducer acts at the level of mrna stabilization. Indeed, nuclear run-on assays show a 3-fold increase in UGT1A1 transcription after T 3 treatment and a 6-fold increase after 3-MC stimulation. This transcriptional induction by T 3 is prevented by cycloheximide in primary hepatocyte cultures, while 3-MC stimulation is only partially affected after prolonged treatment with the protein synthesis inhibitor. Together, these data provide evidence for a transcriptional control of UGT1A1 synthesis and indicate that T 3 and 3-MC use different activation mechanisms. Stimulation of the UGT1A1 gene by T 3 requires de novo protein synthesis, while 3-MC-dependent activation is the result of a direct action of the compound, most likely via the aromatic hydrocarbon receptor complex. UDP-glucuronosyltransferases (UGTs) 1 form a family of microsomal membrane-bound enzymes that catalyze the binding of glucuronic acid on molecules with aromatic groups or aliphatic hydroxy, amino, carboxy, and mercapto groups. Multiple forms of UGTs have been observed in several species, including * This work was supported in part by grants from the Conseil Régional de Bourgogne (Groupement d Intérêt Scientifique, Toxicologie Cellulaire-Dijon, France), from the Swiss National Science Foundation, and from the Etat de Vaud. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Supported by a fellowship from the European Science Foundation. To whom correspondence should be addressed. Tel.: ; Fax: The abbreviations used are: UGT, UDP-glucuronosyltransferase; T 3, 3,5,3 -triiodo-l-thyronine; 3-MC, 3-methylcholanthrene; RT-PCR, reverse transcription-polymerase chain reaction; Ah, aromatic hydrocarbon; TR, thyroid hormone receptor; TRE, thyroid hormone response element; XRE, xenobiotic response element; bp, base pair(s). This paper is available on line at man (1). The regulation of their synthesis appears to be an important determinant of drug and xenobiotic detoxification and elimination (2). Endogenous compounds, such as retinoic acid, the steroid hormones estrone, and 4-hydroxyestrone, and the thyroid hormones can also be inactivated by glucuronidation (3). Based on similarities of their predicted amino acid sequences, two families of UGT have been characterized. Family 2, which will not be analyzed in the present work, is composed of multiple steroid UGTs with broad substrate specificity. Family 1 consists of several forms of UGT1 that catalyze glucuronidation of phenol and bilirubin and that are synthesized from the phenol-bilirubin gene complex. The members of the UGT1 gene complex differ in their first exon but share four common exons, i.e. from the second to the fifth. The variable first exon codes for the N terminus of the proteins, which determines substrate specificity. The different UGT1 isozymes result from differential promoter usage and thus from splicing of different first exons to the common exons 2 5. At least 13 and 9 different exons 1 have been identified in the human and rat gene complex, respectively (4 6). The differential regulation of gene expression by xenobiotics and endobiotics resulting in the synthesis of different UGT isozymes is not well characterized. However, one of the UGT1 phenol-glucuronidating isoenzymes is specifically induced by 3-methylcholanthrene (3-MC) in rat liver (6). It contains an amino-terminal portion encoded by a first exon called A1, has been identified as a 4-nitrophenol UGT and is usually named UGT1A1 (7, 8). Recently, we have reported that 3,5,3 triiodo- L-thyronine (T 3 ) also modulates the expression of the UGT1A1 isozyme (9, 10), whereby 4-nitrophenol glucuronidation is selectively increased. In agreement with this observation, this glucuronidation is decreased in thyroidectomized rats. Using quantitative RT-PCR and specific probes of the exon A1 of UGT1A1, we have found that these variations correlate with concomitant changes in the corresponding mrna levels (11). However, the molecular basis and the physiological significance of this phenomenon remain to be elucidated. Interestingly, previous studies have implicated the UGT1A1 isoenzyme in the glucuronidation of thyroid hormones (12 14). Therefore, further investigation of the effect of thyroid hormone on UGT1A1 expression is required to determine the mechanism by which thyroid hormones regulate their own metabolism and stimulate drug glucuronidation and elimination from the liver. In the present study, we examined whether regulation of UGT1A1 synthesis by thyroid hormone is due to increased rates of gene transcription or to increased mrna stability.

2 17172 Hormonal Regulation of the UGT1A1 Gene in Rat Liver Nuclear run-on assays with isolated hepatic nuclei from whole animals and studies using inhibitors of RNA and protein synthesis in primary hepatocyte cultures revealed that thyroid hormones regulate transcription of the UGT1A1 gene without affecting the half-life of the mrna. Furthermore, the hormonedependent increase in the transcription rate is dependent upon de novo protein synthesis. We also observed that 3-MC-stimulated UGT1A1 gene transcription was affected by cycloheximide only after prolonged treatment (16 h). MATERIALS AND METHODS Preparation of the DNA Probes UGT1A1 DNA and -actin DNA probes were synthesized by PCR as described previously (11). Amplified cdnas corresponding to the expected 507-bp fragment (nucleotides 6 513) of UGT1A1 exon 1 and 222-bp fragment (nucleotide ) of -actin were inserted into the SmaI site of the pbluescript vector (Stratagene), creating pbs-ugt1a1 and pbs- act, respectively. Northern Blot Analysis Total RNA was extracted from Wistar rat liver as described by the supplier of the extraction kit (Eurobio, Les Ulis, France), and RNA integrity was assessed by agarose gel electrophoresis. Total RNA (50 g) was denatured, electrophoresed on a formaldehyde, 1% agarose gel and transferred onto a Gene Screen membrane (NEN Life Science Products) using 3 M sodium chloride and 0.15 M sodium citrate (20 SSC). Membranes were baked at 80 C for 2 h and prehybridized overnight at 65 C in a rotary hybridization incubator with the following solution: 0.2% polyvinylpyrrolidone (M r 40,000), 0.2% bovine serum albumin, 0.2% Ficoll (M r 400,000), 0.05 M Tris-HCl (ph 7.5), 1 M NaCl, 0.1% sodium pyrophosphate, 1% SDS, 10% dextran sulfate (M r 500,000) and denatured salmon sperm DNA (100 g/ml). The membranes were then hybridized for 48 h at 65 C with the same buffer plus the appropriate cdna probe labeled with [ - 32 P]dCTP by random priming as described by the supplier of the labeling kit (Appligene). After the hybridization reaction, membranes were washed twice in 2 SSC at room temperature for 5 min with constant agitation, twice in 2 SSC with 1% SDS at 65 C for 30 min, and twice in 0.1 SSC at room temperature for 30 min and subjected to autoradiography at 70 C for 3 days. The relative amounts of mrna were estimated by densitometric scanning of the Northern blot autoradiograms. The blots were rehybridized with a -actin probe, and the amount of actin mrna in each lane was used as a standard to quantify UGT1A1 mrna. Nuclear Run-on Transcription Assay All steps were performed at 4 C. Nuclei from Wistar rat liver were freshly prepared by sucrose gradient centrifugation essentially as described by Corthésy and Wahli (15) with some modifications as described below. The pelleted nuclei were resuspended at a concentration of in 200 l of transcription mix (92 mm Tris-HCl (ph 8.0), 20 mm NaCl, 4 mm MnCl 2,10mM phosphocreatine, 0.35 M (NH 4 ) 2 SO 4, 1.44 units of human placental ribonuclease inhibitor (Promega, Zurich, Switzerland), 5 g of creatine phosphokinase, plus, for each reaction, 1 mm each of ATP, GTP, and CTP, and 100 Ci [ - 32 P]UTP (400 Ci/mM, 20 Ci/ l)). In vitro elongation of nascent mrna was carried out at 26 C for 30 min. The reaction was stopped by addition of 4 l of MgCl 2 at 0.5 M and 3 l of DNase I (2 mg/ml, Promega) and followed by an additional incubation (for 5 min at 26 C) to remove template DNA. The mixture was deproteinized by treatment with 20 g/ml proteinase K in the presence of 100 g/ml Escherichia coli trna (Boehringer, Rotkreuz, Switzerland) for 30 min at 37 C. Nuclear RNA, including the RNA labeled during the elongation reaction, was purified as described by Bender et al. (16). 5 g of pbs-ugt1a1 and pbs- act were linearized with EcoRI and XbaI, respectively, and denatured before immobilization on a gene screen membrane using a dot-blot apparatus. Then, the filters were baked at 80 C for 2 h. Prior to hybridization with radiolabeled nuclear transcripts, the filters were treated for 6hat65 Cin50mMTris-HCl (ph 8.0), 0.3 M NaCl, 10 mm EDTA, 0.2% SDS, 0.2% Ficoll, 0.2% polyvinylpyrrolidone, 1% sodium pyrophosphate, and 1% E. coli trna. Hybridization reactions were performed in 5-ml propylene tubes (Falcon) in a final volume of 1 ml containing 50 mm Tris-HCl (ph 8.0), 0.3 M NaCl, 10 mm EDTA, 0.2% SDS, 0.02% Ficoll, 0.02% polyvinylpyrrolidone, 0.1% sodium pyrophosphate, and 0.1% E. coli trna in the presence of cpm of labeled mrna, at 65 C for 3 days using a rotary hybridization incubator. After hybridization, the filters were first washed twice 30 min at 65 C with 2 SSC buffer. The filters were then washed in the same buffer containing 0.4 g/ml of RNase A and 10 units/ml of RNase T1 at 37 C for 15 min to hydrolyze the nonhybridized mrna. Finally, the filters were washed at 37 C for 1hin1mM Tris-HCl, ph 7.5, 150 mm NaCl, 1 mm EDTA, 1% SDS, and 50 g/ml of FIG. 1. A, Northern blot analysis of UGT1A1 mrna levels using a radiolabeled UGT1A1 cdna probe. B, hybridization of the same membrane with a -actin probe for the purpose of normalization. For experimental conditions see Materials and Methods. Abbreviations for lane identification are: C, control; Co, corn oil; 3MC, 3-methylcholanthrene, and T 3, 3,5,3 -triiodo-l-thyronine. The positions of 18 and 28 S ribosomal RNAs used as molecular weight standards are indicated. proteinase K and subjected to autoradiography. The relative amounts of 32 P-labeled mrnas bound to the filters were estimated by laser densitometric scanning of the autoradiograms. Hepatocyte Primary Cultures Rat hepatocytes were isolated by collagenase perfusion of livers from g Wistar rats (17), and cell viability was higher than 85% by trypan blue exclusion test. The hepatocytes were cultured in monolayer ( cells/cm 2 ) in Williams E medium (Life Technologies, Inc.) supplemented with 5% fetal calf serum (10%, v/v), fatty acid-free bovine serum albumin (0.2%, w/v), NaHCO 3 (26 mm), glutamine (2 mm), glucose (3 g/liter), and antibiotics, at 37 C in a humidified atmosphere of 5% C0 2, 95% air. After a 4-h incubation period, the medium was replaced by serum-free medium, and treatments with T 3 (0.1 M in 0.9% NaCl), 3-MC (10 M in corn oil), and vehicles alone as control were started. RT-PCR Analysis Complementary DNA was synthesized from RNA samples by mixing 1 g of total RNA, 100 pmol of random hexamers as primers in the presence of 50 mm Tris-HCl buffer (ph 8.3), 75 mm KCl, 3mMMgCl 2,10mMdithiothreitol, 200 units of Moloney murine leukemia virus reverse transcriptase, 40 units of RNase inhibitor, and 1 mm of each dntp in a total volume of 20 l. Samples were incubated at 37 C for 60 min and then diluted to 100 l with sterile diethylpyrocarbonate-treated H 2 O. The reverse transcriptase was inactivated by heating at 95 C for 5 min. A 10- l aliquot of the reverse transcription reaction was then used for subsequent PCR co-amplification and for quantitative analysis with specific primers for UGT1A1 isoform and for the -actin internal standard according to the sequences determined by Iyanagi et al. (7) and Nudel et al. (19), as described previously (11). Absence of genomic DNA contamination was verified by using a control tube supplemented with 10 g of RNase. RESULTS In Vivo Induction of Liver UGT1A1 mrna by Thyroid Hormone Rats were treated intraperitoneally with T 3 dissolved in 0.9% NaCl at a dose of 50 g/kg/day for 8 days and the relative amounts of liver UGT1A1 mrna were assayed by Northern blot analysis. In parallel, we monitored UGT1A1 mrna levels in rats treated with 3-MC, a strong UGT1A1 inducer, and with vehicles alone as positive and negative controls, respectively. Hybridization of mrna to the UGT1A1 cdna probe reveals a prominent transcript of about 2.5 kilobase pairs (Fig. 1A). Following treatment with T 3 and 3-MC, the levels of the UGT1A1 transcript increase two and four times, respectively, when normalized to the actin signal (Fig. 1B). These results are in agreement with the quantitative RT-PCR analysis reported previously (11) and correlate well with the increase in both the UGT activity toward 4-nitrophenol and the UGT1A1 protein levels as measured by immunoblot analysis (11).

3 Hormonal Regulation of the UGT1A1 Gene in Rat Liver FIG. 2.Time-dependent accumulation of UGT1A1 mrna in rat liver after a single injection of L-T 3 (500 g/kg of body mass), 3-MC (80 mg/kg of body mass), or vehicles (NaCl, 0.9% and corn oil, respectively). RT-PCR analysis was done with 1 g total RNA. A, pictures of acrylamide gels loaded with 10- l aliquots of PCR reactions containing co-amplified products of the rat UGT1A1 (507 bp) and -actin (220 bp) transcripts, which are revealed by silver staining. B, graphic representation of the results (means S.E.) of three independent experiments. The amounts of amplified -actin mrna products were used as internal standard to determine UGT1A1 mrna levels. Std lanes refer to lanes containing DNA molecular weight markers (from 24 to 726 bp; phi X 174 HinfI DNA marker, Promega). FIG. 3. Induction of the transcriptional rate of the UGT1A1 gene by T 3 and 3-MC in rat liver. Nuclei were prepared from rat liver 16 h after injection of 500 g/kg of body mass T 3 or 80 mg/kg 3-MC, and the relative amounts of in vitro labeled nascent UGT1A1 and -actin transcripts were determined by dot hybridization. The autoradiograms were scanned, and the levels of UGT1A1 transcripts determined by using -actin transcript levels as reference. The indicated -fold induction (F.I.) in the relative transcriptional rate of the UGT1A1 gene in T 3 - or 3-MC-treated rats corresponds to the means S.E. of three independent experiments. pbluescript DNA was introduced as a negative control of the hybridization experiments. Ctrl., control animals (treated with vehicle alone). *, significantly different (p 0.05) from control animals. To study further T 3 induction of UGT1A1 expression, the time course of UGT1A1 mrna accumulation in response to T 3 (500 g/kg in one injection) and to 3-MC (80 mg/kg in one injection) as a positive control, was followed in rat liver by semiquantitative RT-PCR analysis. Injection of the vehicles alone served as negative controls. Fig. 2 shows the levels of UGT1A1 mrna at different time points after treatment using as reference the -actin mrna levels, which remain unchanged in the rat liver after T 3 and 3-MC treatments. There is a gradual T 3 -induced increase in UGT1A1 mrna, with a maximum level found at 24 h (about 3-fold), which then declines to reach nearly control levels at 48 h after treatment (Fig. 2B). In contrast, 3-MC leads to a quicker and stronger increase of the UGT1A1 mrna level that is already close to maximum 8 h after treatment (5-fold increase). The induced level remains high up to 36 h and decreases thereafter (Fig. 2B). NaCl or corn oil alone has no effect. These results provide evidence that both T 3 and 3-MC are able to increase UGT1A1 mrna levels in an inducer-specific and time-dependent manner and that thyroid hormone is not as strong an inducer as 3-MC. Transcriptional Induction of UGT1A1 Expression by T 3 To determine the molecular mechanism by which accumulation of UGT1A1 mrna is regulated in rat liver, nuclear run-on assays were performed first using hepatocyte nuclei isolated 16 h after a single injection of T 3 (500 g/kg of body mass) or 3-MC (80 mg/kg of body mass). In this assay, only transcripts that are already initiated in vivo are faithfully elongated in vitro, giving an accurate estimation of the relative transcriptional activity at the time point of nuclei isolation. The radiolabeled nascent RNA transcripts were isolated and hybridized to immobilized UGT1A1 and -actin cdna. Densitometric analysis of the resulting hybridization signals reveal that both T 3 and 3-MC treatments augment the levels of nascent UGT1A1 transcripts relative to actin transcripts, reflecting increased transcription rates by a factor of 3 and 6 for T 3 and 3-MC, respectively (Fig. 3). These results indicate that T 3 regulates UGT1A1 expression at the transcriptional level and confirm results obtained previously with 3-MC (8). The difference between the transcription induction factors of T 3 and 3-MC correlate well with the difference in induced levels of UGT1A1 mrna described above (compare results of Fig. 2 with those of Fig. 3). Effect of T 3 on UGT1A1 mrna Stability Next, we tested whether T 3 stimulation is confined to induction of transcription only or also involves an increase in mrna stability, a mechanism used in the control of some thyroid hormone-responsive genes. To this end, hepatocytes in primary culture were treated for 6 and 16 h with T 3 in the presence or absence of the RNA synthesis inhibitor actinomycin D (Fig. 4). In parallel, the same experiment was performed in cultures treated with 3-MC (Fig. 5), as well as with the vehicles alone as controls. The decrease in UGT1A1 mrna levels after inhibition of RNA synthesis was measured by quantitative RT-PCR as a function of time. The results obtained with T 3 and 3-MC are very similar (Figs. 4A and 5A): UGT1A1 mrna decays slowly and independently of the presence or absence of the inducers. In Figs. 4B and 5B, the amounts of UGT1A1 mrna (log values) are plotted as function of time, allowing the determination of the RNA half-life (t) assuming first-order kinetics. The values obtained with the T 3 experiments are h(r 0.994) in T 3 -treated hepatocytes and 7 1h(r 0.993) in control hepatocytes and, thus, are not significantly different from each other (Fig. 4B). This result indicates that T 3 does not exert its control on UGT1A1 mrna levels by RNA stabilization. Similarly, the 3-MC treatment has no effect on mrna stability, since we determined half-lives of 7 1 h and 8 1 h in treated and nontreated cells, respectively (Fig. 5B). T 3 -induced UGT1A1 Transcription Is Dependent on de Novo Protein Synthesis The slow increase in UGT1A1 mrna levels after T 3 injection suggested that the transcriptional activation of the UGT1A1 gene might require de novo protein synthesis. This possibility was tested using hepatocytes in primary culture incubated for 6 and 16 h with 0.1 mm T 3 alone or in the presence of 10 mg/ml cycloheximide, an inhibitor of protein synthesis. In the presence of cycloheximide, the induction of the UGT1A1 gene by T 3 is blocked (Fig. 4, A and C), providing evidence that de novo protein synthesis is required for UGT1A1 mrna induction. A different result is observed with 3-MC, since cycloheximide treatment does not suppress UGT1A1 mrna stimulation after 6 h of treatment and has only a weak effect after a 16-h treatment (Fig. 5, A and C). These results suggest that de novo synthesis is not required for stimulation by 3-MC but that after prolonged treatment one factor(s) becomes rate-limiting and has to be produced de novo to maintain

4 17174 Hormonal Regulation of the UGT1A1 Gene in Rat Liver FIG. 4. Actinomycin D and cycloheximide effects on UGT1A1 inducibility by T 3. Hepatocytes in primary cultures were maintained in the absence ( ) or presence ( ) oft 3 and actinomycin D at 5 g/ml (Act.D) or cycloheximide at 10 g/ml (CHX) as indicated. None of these inhibitors was added in control cultures (Ctrl). RNA (1 g) isolated from the cultures harvested after 6 or 16 h of treatment was subjected to RT-PCR analysis using two sets of primers specific for UGT1A1 and -actin. A, the products of amplification were separated by polyacrylamide gel electrophoresis and visualized by silver staining as in Fig. 2. B, the intensity of the silver-stained bands, shown in A, was quantified by densitometry. The percentages of UGT1A1 mrna remaining after an actinomycin D treatment of 6 or 16 h were calculated and plotted on a semilogarithmic scale versus the duration of the treatment. The UGT1A1 mrna levels were calculated at various times by using the formula described by Chelly et al. (20): A UGT /A act Y UGT (1 R act ) n / Y act (1 R UGT ) n, where A is the initial amount of material, R the efficiency, n the number of cycles, and Y the area intensity. The values given represent the mean S.E. from three independent experiments. A linear regression was used to calculate the UGT1A1 mrna half-life by extrapolation and the values given represent the mean S.E. of three independent experiments. E, control hepatocytes;, T 3 -treated hepatocytes. C, effect of T 3 and/or cycloheximide on UGT1A1 mrna levels in hepatocyte cultures. The UGTA1 mrna levels were calculated as in B. The -fold induction (F.I.) of UGT1A1 mrna was calculated as the ratio of the values corresponding to hormone-treated hepatocytes over those of untreated cells. Values are the mean S.E. obtained from three independent experiments. (*, p 0.05; **, p 0.01). full induction. Importantly, we observe that cycloheximide treatment, even for 16 h, does not affect -actin gene expression. FIG. 5. Actinomycin D and cycloheximide effects on UGT1A1 inductibility by 3-MC. Hepatocytes were cultured in the absence ( ) or presence ( ) of 3-MC for 6 or 16 h, and actinomycin D at 5 g/ml (Act.D) or cycloheximide at 10 g/ml (CHX) was added when indicated. The methods are as described in the legend of Fig. 4. DISCUSSION In this work we provide evidence that T 3 regulates UGT1A1 gene expression at the transcriptional level and that de novo protein synthesis is required for this stimulation. These results offer a molecular explanation for our recent observation that administration of T 3 to adult rats enhances the activity of phenol UGT in the liver (11). First, these conclusions are based on a nuclear run on assay using isolated liver nuclei which showed that T 3 appears to act on UGT1A1 gene expression mainly at the transcriptional level by increasing the rate of synthesis of UGT1A1 mrna. Furthermore, the slow progressive induction of UGT1A1 by thyroid hormone is consistent with de novo protein synthesis, which is required for this transcriptional regulation. Second, we observed no change in the stability of UGT1A1 mrna after T 3 treatment, which suggest that post-transcriptional regulation does not contribute to the T 3 -dependent increase in UGT1A1 activity. T 3 regulates the expression of several other target genes at the transcriptional level. These include the genes encoding rat growth hormone (21), rat malic enzyme (22), phosphoenolpyruvate carboxykinase (23), chicken fatty acid synthase (24), and rat apolipoprotein AI (25, 26). T 3 mediates its effects through nuclear T 3 receptors (TRs) that are ligand-dependent transcription factors and that bind to thyroid hormone response elements (TREs) in the promoter of these genes (27, 28). In addition, the TRs bind to TREs as homodimers (29, 30) or heterodimers with accessory proteins such as the retinoid X receptor (31, 32). Interestingly, our results suggest that the mode of action of T 3 on UGT1A1 gene expression is different, since de novo protein synthesis is required for stimulation. Although our observations do not formally exclude a direct action of T 3 on UGT1A1 expression via the thyroid hormone receptor, they clearly show that if it were to occur it would not be sufficient, since the hormone is controlling the de novo production of a factor that is required for the stimulation of UGT1A1 gene transcription. The biological significance of the T 3 enhancement of the transcriptional rate of the UGT1A1 gene, and the subsequent increase in the expression of the corresponding isozyme, remain speculative. Indeed, little is known about the effects of UGT1A1 activity on thyroid hormone homeostasis in rats. Nevertheless, glucuronidation has been found to be a major path-

5 Hormonal Regulation of the UGT1A1 Gene in Rat Liver way for biotransformation of the iodothyronines (33, 34), conjugates being subsequently excreted in the bile. Recent studies reported that multiple UGT isoenzymes are involved in this pathway. Thyroxine (T 4 ), the main secretory product of the thyroid gland is glucuronidated by 4-nitrophenol (UGT1A1) and bilirubin UGTs, while T 3, which is produced by outer ring deiodination of T 4 in peripheral tissues, is glucuronidated by androsterone UGT (14, 15). In addition, Visser et al. (13) reported that the MC-type inducers also increase T 3 glucuronidation. These findings suggest that the T 3 -stimulated increase in UGT1A1 activity following T 3 treatment could lead to a lowering of the T 4 and T 3 levels in blood and, thus, participate in the regulation of thyroid hormone action. The physiological or pathological relevance of thyroid stimulation of UGT1A1 gene expression could now also be explored further in Gunn rats in which this gene is non-functional. Comparison of the response to T 3 and 3-MC stimulation is interesting, since it has been reported that 3-MC is an effective inducer of UGT1A1 gene expression also acting at the transcriptional level (8). Herein, nuclear run on assays on isolated liver nuclei confirm this finding and show that 3-MC enhances the transcriptional rate of the UGT1A1 gene. In addition, we present original data providing important information on the mechanism of 3-MC action: 3-MC has no effect at the posttranscriptional level, and although de novo protein synthesis is not required for transcriptional induction, a factor(s) that is most likely unstable, becomes rate-limiting, and its continuous synthesis is necessary for full induction. The nature of this factor(s) remains to be elucidated. 3-MC is known to act via a cytosolic protein, called aromatic hydrocarbon (Ah) receptor. 3-MC binds to the Ah receptor, and the Ah receptor-ligand complex translocates into the nucleus after interaction with the Ah receptor nuclear translocator. Once in the nucleus, the activated complex binds to the AhRE (also termed xenobiotic or dioxin response elements (XREs or DREs, respectively) and turns on transcription of the Ah-responsive genes (reviewed in Refs ). Interestingly, with respect to our results, it has been demonstrated that induction of mrna encoding mouse Ugt1.6, which corresponds to the rat UGT1A1, was dependent upon the Ah receptor complex (18), which recognizes one or more XREs in the UGT1A1 promoter (8). In conclusion, the data presented herein further our understanding of the regulation of UGT1A1 gene expression and demonstrate a major difference in the mode of action of T 3 and 3-MC. The effect of 3-MC appears to be direct, in the sense that it is independent of de novo protein synthesis during the initial phase of the stimulatory process. However, a rate-limiting factor is required to obtain maximal induction, and it needs to be produced by de novo synthesis. In comparison, the regulating action of T 3 depends on a factor being synthesized after hormonal treatment. The identification of this factor opens a new research avenue on UGT1A1 gene expression control. Acknowledgments We are grateful to Dr. R. Planells (INSERM U 38, Marseille, France) for helpful technical advice and to Dr. E. Beale (Institut de Biologie Animale, Lausanne, Switzerland) for reading the manuscript. REFERENCES 1. Burchell, B., Nebert, D. W., Nelson, D. R., Iyanagi, T., Jansen, P. L. M., Lancet, D., Mulder, G. J., Chowdhury, J. R., Siest, G., Tephly, T. R., and Mackenzie, P. I. (1991) DNA Cell Biol. 10, Dutton, G. J. (1980) Glucuronidation of Drugs and Other Compounds, pp , CRC press, Boca Raton, FL 3. Burchell, B., and Coughtrie, M. W. H. (1989) Pharmacol. Ther. 43, Ritter, J. K., Chen, F., Sheen, Y. Y., Tran, H. M., Kimura, S., Yeatman, M. T., and Owens, I. S. (1992) J. Biol. Chem. 267, Ciotti, M., Yeatman, M. T., Sokol, R. J., and Owens, I. S. (1995) J. Biol. Chem. 270, Emi, Y., Ikushiro, S., and Iyanagi, T. (1995) J. Biochem. (Tokyo) 117, Iyanagi, T., Haniu, M., Sogawa, K., Fujii-Kuriyama, Y., Watanabe, S., Shively, J. E., and Anan, K. F. (1986) J. Biol. Chem. 261, Emi, Y., Ikushiro, S., and Iyanagi, T. (1996) J. Biol. Chem. 271, Goudonnet, H., Magdalou, J., Mounié, J., Naoumi, A., Viriot, M. L., Escousse, A., Siest, G., and Truchot, R. (1990) Biochim. Biophys. Acta 1035, Pacot, C., Charmoillaux, M., Goudonnet, H., Truchot, R. C., and Latruffe, N. (1993) Biochem. Pharmacol. 45, Masmoudi, T., Planells, R., Mounié, J., Artur, Y., Magdalou, J., and Goudonnet, H. (1996) FEBS Lett. 379, Visser, T. J., Kaptein, E., Van Toor, H., Van RAAIJ, J. A. G. M., Van Den Berg, K. J., Tjong Tjin Joe, C., Van Engelen, J. G. M., and Brouwer, A. (1993) Endocrinology 133, Beetstra, J. B., Van Engelen, J. G. M., Karels, P., Van Der Hoek, H. J., De Jong, M., Docter, R., Krenning, E. P., Hennemann, G., Brouwer, A., and Visser, T. J. (1991) Endocrinology 128, Visser, T. J., Kaptein, E., and Harpur, E. S. (1991) Biochem. Pharmacol. 42, Corthésy, B., and Wahli, W. (1990) Transcription Factors DNA Binding Proteins (Antonucci, T., Chang, S., El-Gewely, M. R., Elices, M. J., Engelke, D. R., Gorman, C., Graham, D., Landik, R. G., Mavromara-Nazos, D., Merlino, G. T., Murray, J. A. H., Vlahos, C. J., and Willis, E., eds) Vol. 2, pp , Wiley-Liss, Inc., New York 16. Bender, T. P., and Greenberg, M. E. (1995) in Current Protocols in Molecular Biology (Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A and Struhl, K., eds) Vol. 26, pp. 1 11, John Wiley & Sons, Inc., New York 17. Berry, M. N., and Friend, D. S. (1969) J. Cell Biol. 43, Lamb, J. G., Straub, P., and Tukey, R. H. (1994) Biochemistry 33, Nudel, U., Zakut, R., Shani, M., Neuman, S., Levy, Z., and Yaffe, D. (1983) Nucleic Acid Res. 11, Chelly, J., Kaplan, J. C., Maire, P., Gautron, S., and Kahn, A. (1988) Nature 333, Glass, C. K., Franco, R., Weinberger, C., Albert, V. R., Evans, R. M., and Rosenfeld, R. M. (1987) Nature 329, Desvergne, B., Petty, K. J., and Nikodem, V. M. (1991) J. Biol. Chem. 266, Giralt, M., Park, E. A., Gurney, A. L., Liu, J., Hakimi, P., and Hanson, R. W. (1991) J. Biol. Chem. 266, Stapleton, S. R., Mitchell, D. A., Salati, L. M., and Goodridge, A. G. (1990) J. Biol. Chem. 265, Strobl, W., Chan, L., and Patsch, W. (1992) Atherosclerosis 97, Romney, J. S., Chan, J., Carr, F. E., Mooradian, A. D., and Wong, N. C. (1992) Mol. Endocrinol. 6, Evans, R. M. (1988) Science 240, Brent, G. A., Moore, D. D., and Larsen, P. R. (1991) Annu. Rev. Physiol. 53, Lazar, M., Berrodin, T. J., and Harding, H. P. (1991) Mol. Cell. Biol. 11, Brent, G. A., Williams, G. R., Harney, J. W., Forman, B. M., Samuels, H. H., Moore, D. D., and Larsen, P. R. (1992) Mol. Endocrinol. 6, Kliewer, S. A., Umesono, K., Mangelsdorf, D. J., and Evans, R. M. (1992) Nature 355, Zhang, X. K., Hoffman, B., Tran, P. B., Graupner, G., and Pfahl, M. (1992) Nature 355, Visser, T. J. (1990) in The Thyroid Gland (Greer, M. A., ed) pp , Raven Press, New York 34. Curran, P. J., and DeGroot, L. J. (1991) Endocrinol. Rev. 12, Vasiliou, V., Puga, A., and Nebert, D. W. (1993) Adv. Exp. Med. Biol. 328, Nebert, D. W., Puga, A., and Vasiliou, V. (1993) Ann. N. Y. Acad. Sci. 685, Swanson, H. I., and Bradfield, C. A. (1993) Pharmacogenetics 3, Whitlock, J. P., Jr. (1994) Chem. Res. Toxicol. 6,