Two CCAAT/enhancer binding protein sites in the cytochrome P4503A1 locus Potencial role in the glucocorticoid response Elsa Rodrigues 1 , Marie-Jose ´ Vilarem 2 , Vera Ribeiro 1, *, Patrick Maurel 2 and Maria C. Lechner 1 1 Molecular Biology Unit, Faculty of Pharmacy, University of Lisbon, Portugal; 2 Unite ´ 128 INSERM, Montpellier, France Induction of CYP3A genes by the ligand-activated preg- nane-X-receptor (PXR) involves the interaction of other as yet unidentified liver transcription factors. Here we show that the CYP3A1 promoter contains two active sites con- trolled by the CCAAT/enhancer-binding protein a (C/ EBPa), previously shown to regulate a number of liver stress response genes. We have identified two functional C/EBP binding sites at the CYP3A1 promoter that confer luciferase activity to C/EBPa cotransfected CHO cells. When inserted upstream of a thymidine kinase promoter, oligonucleotides corresponding to these elements ()350/)311 and )628/ )608), increase reporter gene expression when cotransfected with a C/EBPa expression vector. Point mutations in the most conserved nucleotides in either element prevent binding of C/EBPa and abolish transactivation of the CYP3A1 promoter. Moreover, we demonstrate that C/EBPa accu- mulates in the rat liver nuclei in response to dexamethasone, and that under these conditions C/EBPa binds to the CYP3A1 promoter elements. Our results suggest a correla- tion between transcription of C/EBPa, nuclear protein function and induction of CYP3A1 by dexamethasone in the liver. They also support the notion that C/EBPa participates in the up-regulation of the CYP3A1 gene in response to synthetic glucocorticoids. Keywords: cytochrome P450; CYP3A1 locus; C/EBP; regu- latory elements; glucocorticoid induction. Mammalian hepatic phenotypes are controlled through the concerted action of a number of liver-enriched transcription factors that act in cooperation with a number of ubiquitous and ligand-activated factors to direct the selective expression of liver-specific genes. The activity of each target gene is modulated by dynamic arrays of counterpart transcription factors that act upon liver differentiation and that are stimulated by endogenous as well as by exogenous ligands [1]. Cytochrome P450 genes, members of the CYP1–4 fami- lies, are abundantly expressed in the liver and are subjected to complex regulatory networks that depend markedly on the ontological development, and on the hormonal and nutritional status of the animals [2]. Moreover, P450 mono- oxygenases, as major body interfaces, are adaptive enzymes highly responsive to induction and repression by environ- mental and xenobiotic agents in general. Recently, important advances have been made in under- standing the mechanism of action of prototype inducers that control the expression of hepatic P450 enzymes, namely those mediated by the nuclear receptors AhR, CAR, PPAR and PXR (reviewed, [3]). However, the role of these ligand- activated receptors can be enhanced, reduced or inhibited by the availability of other transcription factors that participate in the organization of the transcription initiation complexes in the native DNA context of each CYP locus. Such func- tional interactions have not been fully defined, but their elu- cidation is essential to understand the molecular basis of the marked physiological variations often observed in the in vivo response of each CYP gene to common inducing agents. CYP3A1 has been widely investigated as a model of a xenobiotic up-regulated liver expressed gene. Transcription of CYP3A1 is virtually undetected in the adult rat liver, but markedly induced by structurally diverse agents, including the synthetic glucocorticoid agonist dexamethasone [4] and the antagonist pregnenolone 16a-carbonitrile [5]. Such paradoxical behaviour of the rat CYP3A1, and of other orthologous genes, was shown to involve the pregnane X receptor (PXR) [6–8], a member of the steroid hormone receptor family. The mechanism of the glucocorticoid- mediated activation of CYP3A1 relies on a dexamethasone responsive unit in the promoter region of this gene [9,10]. This unit comprises elements that are targeted by the activated PXR, as well as by COUP-TFs and HNF4 [11,12]. However, in vitro experiments have shown that mouse [6], Correspondence to M. C. Lechner, Molecular Biology Unit, Faculty of Pharmacy, University of Lisbon, Av. Prof Gama Pinto, 1649–003 Lisbon, Portugal. Fax: + 351 21 7946491, Tel.: + 351 21 7946490, E-mail: clechner@ff.ul.pt Abbreviations: C/EBP, CCAAT/enhancer binding protein; C/EBP cons, C/EBP consensus oligonucleotide; CYP, cytochrome P450; PXR, pregnane-X-receptor; AhR, aryl hydrocarbon receptor; CAR, constitutive androstane receptor; PPAR, peroxisome proliferator- activated receptor; RXR, retinoid-X-receptor; COUP-TF, chicken ovalbumin upstream promoter transcription factor; HNF4, hepato- cyte nuclear factor 4; TK, thymidine kinase; EMSA, electrophoretic mobility-shift assay. *Present address: Biochemistry Laboratory, Chemistry Departmental Area, Faculty of Sciences and Technology, University of Algarve, Campus de Gambelas 8000-117 Faro, Portugal. (Received 30 September 2002, revised 14 November 2002, accepted 5 December 2002) Eur. J. Biochem. 270, 556–564 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03413.x rat [13] and human [8,14] PXR receptors are moderately activated by dexamethasone. This result is in contrast with the strong in vivo transcriptional induction by this gluco- corticoid [5,15–17], and suggests that PXR does not fully explain the pattern of CYP3A1 induction. The importance of other transcription factors for the positive regulation by dexamethasone is also suggested by the observation that full CYP3A1 induction can not be achieved in CV1 cells, that lack liver-enriched transcription factors presumably needed to elicit the response [18]. Other factors structurally unrelated to PXR have been demonstrated to activate transcription of liver-expressed genes, namely as potentiators of glucocorticoid responses. CCAAT/enhancer-binding proteins (C/EBP) were shown to play a role in adipocyte differentiation, as well as in the liver acute-phase response (review, [19]). These factors create multiple possibilities of combinatorial gene regulation in the liver, by strategies that involve heterodimerization with other b-zip proteins, and cross talk interactions with other factors, such as the glucocorticoid receptor [20]. P450 genes, namely Cyp2D5 [21,22], CYP2B [23–25], CYP2C12 [26] and human CYP3A4 [27], have been shown to be the targets of C/EBP factors in liver cells. In the present work we investigated the presence of cis- acting elements in the CYP3A1 promoter. Dynamic trans- fection–transactivation assays were performed to evaluate the potential activity of recombinant C/EBPa on the previ- ously cloned CYP3A1 promoter region [28]. Two C/EBP binding sites were identified by proteinÆDNA gel mobility- shift assays. To assess the biological significance of these regulatory elements in the response to glucocorticoids, the concurrence of the CYP3A1 and the C/EBP responsiveness was investigated in the rat liver upon synthetic glucocorticoid administration. The hypothesis of a cause–effect relationship is suggested by the time-course analysis of C/EBPa and CYP3A1 mRNA and protein accumulation in the liver. Our in vitro and the in vivo results indicate that C/EBPa is an activator of the CYP3A1 gene promoter and suggest a role for this liver enriched transcription factor in the transcrip- tional activation of CYP3A1 by synthetic glucocorticoids. Materials and methods Animals and treatments Male adult Wistar rats, bred at the Gulbenkian Institute animal house, Ociras, Portugal, were used in this investiga- tion. Rats were maintained with standard chow and water ad libitum, until 24 h before hormone treatment. Dexameth- asone 21-phosphate (Sigma) was given intragastrically, in aqueous solution, in a dose of 40 mg per kg body mass, and the animals were sacrificed at different time points after drug administration. Groups of three rats were used for each sample. The livers were excised immediately after decapi- tation and pooled for cell fractionation or RNA extrac- tion. All procedures were carried out in accordance with European regulations concerning animal experimentation. Plasmids Several different fragments derived from the 5¢ flanking region of the CYP3A1 gene (GenBank accession no. X62086) [28], were previously subcloned upstream of the thymidine kinase gene [29] in the luciferase expression vector pT81Luc [30], and analysed in transient transfection experiments. Constructs bearing multiple element copies were made using oligonucleotides encompassing the two distinct CYP3A1 5¢ sites 3A1-300 ()350/)331, 5¢- GTCCTTCTGTAATGGTGTG-3¢), or 3A1-600 ()629/ )608, 5¢-TGCAGGATTGCAGAAGTCTATT-3¢).These were ligated with the SmaI-digested pT81Luc vector and analysed in transient transfection experiments. All con- structs were verified by DNA sequencing. The pMSV/EBPa and pMSV/EBPb expression vectors were kindly provided by S. L. McKnight (University of Texas South-Western Medical School, TX, USA). Cell culture CHO cell line (hamster ovary epithelial) was maintained in Ham’s F12 medium and the COS-7 cell cultures were maintained in high glucose Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% (w/v) heat inactivated fetal bovine serum and maintained at 37 °Cin humidified 5% (v/v) CO 2 . Transactivation assays To minimize variations in transfection efficiency, replicates were transfected in single batch suspension with FuGENE (Roche Molecular Biochemicals), according to the manu- facturer’s instructions. Plates containing 150 000 cells were cotransfected with 0.5 lg of CYP3A1T81Luc plasmid and different amounts of each expression vector. Cells were inoculated in 24-well plates and maintained for 48 h. These cells were harvested and lysed in reporter lysis buffer (Promega, Madison, WI, USA). Cell extracts were assayed for luciferase activity and protein content (BCA reagent, Pierce, Rockfort, IL, USA). The cell extracts were normal- ized for the total amount of protein prior to performing the luciferase assay, as C/EBP expression vector was shown to severely repress the expression of b-galactosidase in the transfected CHO cell line. Site-directed mutagenesis The plasmid 0.8T81Luc served as the template for site- directed mutagenesis using the QuickChange Site Directed Mutagenesis Kit (Stratagene, CA, USA). All reactions were performed according to the protocol provided by the manufacturer. The oligonucleotides used for site-directed mutagenesis of the C/EBP binding sites were as follows (mutated bases underlined): site 3A1-300: 5¢-GGAGA AAGTCC GTCTATGGTGGTGTGCAGATGACACAG TTTTGGC-3¢; site 3A1-600: 5¢-GCCTCTGCTCTGTA AGTGCAGGA CCGTAGAGGTCTATTACTTATG-3¢. mRNA analysis Total liver RNA was extracted by a modification of the LiCl/ urea method [31], and samples (20 lg) eletrophoresed on formaldehyde-agarose gels for Northern blot analysis. The subsequently generated Northern blot nylon membranes were incubated with the specific oligonucleotide probes Ó FEBS 2003 Two C/EBP sites in the CYP3A1 locus (Eur. J. Biochem. 270) 557 CYP3A1 (5¢-TGTGCGGGTCCCAAATCCGT-3¢) and C/EBPa (5¢-GCACGAGACGTCTATAGACA-3¢) end- labelled with [c- 32 P] dATP using T4 polynucleotide kinase. Nuclear extracts Isolation of liver nuclei both from control and dexameth- asone-treated rats was carried out [32] and nuclear extracts were prepared as described previously [33]. Recombinant C/EBPa used in the gel mobility-shift assays were obtained by transfecting 3.0 · 10 6 COS-7 cells with 30 lgofpMSV/EBPa. Cells were seeded in 100-mm plates and maintained for 48 h for nuclear extract preparation [33]. Western blot analysis Five micrograms of liver nuclear protein were electrophore- sed on 10% SDS/polyacrylamide gels and electroblotted onto Imobilon P (Millipore, Bedford, MA, USA). After visua- lization of the transferred proteins by amido black staining, the membranes were incubated with an anti-C/EBPa Ig (14AA, Santa Cruz Biotechnology). Results were quantified using the EAGLE - EYE software package (Stratagene). Electrophoretic mobility-shift assay (EMSA) Double-stranded DNA probes and competitors were gen- erated by annealing from the following complementary single-stranded oligonucleotides: C/EBP consensus oligo- nucleotide (C/EBPcons), 5¢-TGCAGATTCCGCAATCTG CA-3¢ [34]; 3A1–300, 5¢-AGTCCTTCTGTAATGGTG TG-3¢; 3A1–600, 5¢-TGCAGGATTGCAGAAGTCTA TT-3¢; 3A1-700, 5¢-AATTTTGGTGGATAGATAT AG-3¢; m3A1-300, 5¢-AGTCC GTCTATGGTGGTGTG-3¢; m3 A1-600, 5¢-TGCAGGA CCGTAGAGGTCTATT-3¢ (mutated bases are underlined). The oligonucleotides 3A1- 300 (position )350 to )331), 3A1-600 (position )629 to )608) and 3A1-700 (position )766 to )746) encompass distinct CYP3A1 promoter regions. DNA (6 pmol) was end- labelled with [c- 32 P] dATP using T4 polynucleotide kinase, and unincorporated nucleotides were removed by Sephadex G50 filtration. The binding reactions were performed in a total volume of 20 lL and contained 2–5 lg of nuclear extract, 10 m M Hepes buffer, pH 8.0, 0.1 m M EDTA, 2 m M dithiothreitol, 17.5% (v/v) glycerol, 40 m M spermidine, 40 m M MgCl 2 ,0.5lgofdIdC,1lg salmon sperm DNA and 0.5–2 ng of oligonucleotide probe. In competition assays excess unlabeled oligonucleotide was preincubated (30 min) at 4 °C, prior to incubation with each probe for additional 20 min. Supershift reactions were performed with 1 lLof anti-C/EBP (14AA, Santa Cruz Biotechnology) which was added to the reaction media, that were kept on ice for 30 min before addition of the probe. ProteinÆDNA complexes were resolved on 5% (w/v) nondenaturing polyacrylamide gels (acrylamide/bisacrylamide 29 : 1, v/v) in 0.5· Tris/borate/ ECTA buffer (45 m M Tris/borate, 1 m M EDTA). The gels were eletrophoresed for 2.5 h, at 30 mA. Statistical analysis Statistical analyses were performed using the Student’s t-test and the ANOVA one-way test with the Tukey HSD posthoc test for unequal N (Spjotvoll/Stoline test). All analysis were performed using the StatSoft Inc. (1995) STATISTICA FOR WINDOWS software. Results Identification of C/EBP-responsive sequences in the CYP3A1 gene promoter CYP3A1T81Luc recombinants containing the 5¢ upstream region of the CYP3A1 gene were constructed and cotrans- fected into CHO cells together with the C/EBP expression vectors pMSV/EBPa or pMSV/EBPb.TheCHOcellline, which is devoid of hepato-specific transcription factors, has been used previously to characterize C/EBP-dependent gene expression [26,35]. We found that C/EBPa significantly Fig. 1. Transactivation of CYP3AT81Luc constructs with vectors pMSV/EBPa or pMSV/EBPb in the CHO cell line. Cotransfections were carried out using 0.5 lg of CYP3AT81Luc and increasing con- centrations (0.025–0.2 lg) of C/EBP expression vectors, or empty vector. (A) Diagramatic representation of the luciferase constructs containing CYP3A1 5¢-flanking DNA sequences, upstream of luci- ferase cDNA. Tk, thymidine kinase promoter; lucif, luciferase cDNA. (B) Cotransfection of CYP3AT81Luc with 0.15 lgofpMSV/EBPa. (C) Cotransfection of p0.8T81Luc with increasing concentrations of pMSV/EBPa or pMSV/EBPb. The normalized luciferase activities are expressed as mean values ± SD of duplicates for a minimum of three experiments. *P < 0.05, **P <0.01,***P < 0.001 significant dif- ferences between cells cotransfected with pMSV and pMSV/EBPa (Student’s t-test). 558 E. Rodrigues et al.(Eur. J. Biochem. 270) Ó FEBS 2003 increases the luciferase reporter gene activity of the CYP3A1T81Luc (Fig. 1B). The strongest transactivation level (approximately sevenfold) was observed in the pre- sence of the p0.8T81-Luc construct (Student’s t-test, P < 0.001), which contains approximately 800 bp of the 5¢ upstream region of the CYP3A1 locus (Fig. 1B). Deletion of the DNA segment )166 to )811 severely reduced transactivation by C/EBPa. Moreover, transactivation of the p0.8T81Luc is shown to be dose-dependent with the C/EBPa expression vector (Fig. 1C). C/EBP-binding activity in rat liver has been described as involving both C/EBPa and C/EBPb. However, when pMSV/EBPb was assayed in cotransfection experiments with the CYP3A1 recombinant p0.8T81Luc no transacti- vation was observed (Fig. 1C). Moreover, C/EBPb inhibits the C/EBPa -stimulated expression of the p0.8T81Luc (data not shown). Characterization of the C/EBP binding sequences in the CYP3A1 promoter region We used gel mobility-shift assays, with the goal of identi- fying active C/EBPa binding sites within the CYP3A1 locus. First, oligonucleotides encompassing the putative C/EBP responsive elements in p0.8T81Luc were synthesized. These were named 3A1-300, 3A1-600 and 3A1-700 (Fig. 2A). C/EBP proteins were then over-produced in COS-7 cells, and the nuclear extracts used to characterize their binding activities to 3A1-300, 3A1-600 or 3A1-700 by EMSA. The specificity of the complex(es) formed when a C/EBP consensus oligonucleotide (C/EBPcons) was used, was verified by means of a supershift assay using an anti-C/ EBPa Ig (Fig. 2B; arrowhead). Both 3A1-300 and 3A1-600 oligonucleotides competed efficiently for C/EBPa binding to the C/EBPcons preventing the formation of the complex at a 50-fold molar excess (Fig. 2B). The C/EBPconsÆ C/EBPa complex was not competed by 3A1-700 or by an unrelated DNA sequence (Fig. 2B). Supershift assays, with an anti-C/EBPa Ig and radio- labelled oligonucleotides 3A1-600 and 3A1-300 (data not shown) confirmed the specificity of C/EBPa protein binding (Fig. 2C; arrowhead). We note that the 3A1-600ÆC/EBPa complex was competed by C/EBPcons and by cold self, while a mutant form of this oligonucleotide, differing in Fig. 2. Characterization of the binding activities to sites 3A1-300, 3A1- 600 and 3A1-700 in C/EBPa over-expressed COS-7 cell nuclear extracts. (A) DNA sequence of the CYP3A1 5¢-flanking region cor- responding to construct p0.8T81Luc. The oligonucleotides containing the putative C/EBP responsive elements used in the gel mobility-shift assays are underlined. The oligonucleotide 3A1-300 corresponds to the )350/)331 sequence of the CYP3A1 cDNA, the 3A1-600 to the )629/ )608 sequence and the 3A1-700 to nucleotides )767/)747. (B) EMSA was performed using a radiolabeled double-stranded oligonucleotide corresponding to the C/EBP consensus (C/EBPcons) as a probe. Competition experiments were performed by adding a five- or 50-fold excess of unlabelled double-stranded oligonucleotides corresponding to cold-self, a nonspecific tubulin sequence, and 3A1-300, 3A1-600 or 3A1-700. (C) EMSA was performed using a radiolabeled double- stranded oligonucleotide corresponding to site 3A1-600 as a probe. Competition experiments were performed by adding a five- or 50-fold excess of unlabelled double-stranded oligonucleotides corresponding to C/EBPcons, cold self, mutant 3A1-600 (m3A1-600), and a non- specific tubulin sequence. Supershift experiments were performed using an anti-C/EBPa Ig. Symbol ÔsÕ in panel B and C denotes the position of the C/EBP containing complex and the arrowhead ÔssÕ the position of the DNAÆprotein complex shifted by the specific antibody. Ó FEBS 2003 Two C/EBP sites in the CYP3A1 locus (Eur. J. Biochem. 270) 559 3 bp, did not. The results suggest the presence of two C/EBPa binding sites at positions )300 and )600 within the CYP3A1 locus. Activation of multimerized CYP3A1-C/EBP responsive elements by C/EBPa in transfection assays To determine whether C/EBPa binding to either site 3A1- 300 or 3A1-600 resulted in transcriptional activation, we analysed the ability of each specific site to confer C/EBP responsiveness to the thymidine kinase promoter of the pT81Luc recombinant. We performed the cotransfection of CHO cells with pMSV/EBPa and single recombinants containing two or three synthetic copies of 3A1-300 or 3A1- 600 oligonucleotides. As shown in Fig. 3, the C/EBP elements found in the CYP3A promoter are functional and autonomous units that confer C/EBP responsiveness to a heterologous promoter. Functional analysis of C/EBP binding sites in CYP3A1 promoter by site directed mutagenesis The functional relevance of sites 3A1-300 and 3A1-600 for CYP3A1 induction by C/EBPa was investigated by individual or combined mutagenesis of the two sites in p0.8T81Luc, followed by cotransfection of C/EBPa expression vector into CHO cells. As shown in Fig. 4, mutation of site 3A1-300 (m300) and/or 3A1-600 (m600) significantly reduced activation of the p0.8T81Luc by C/EBPa to approximately 50% of the wild-type level (wt) ( ANOVA one-way test: F ¼ 12.29, d.f. ¼ 9, P < 0001). The posthoc comparisons revealed that significant differ- ences were found between the wt and each mutant (Tukey HSD for unequal N: m300, P < 0.05; m600, P <0.01; dm300/600, P < 0.01). Western blot analysis of the 42 kDa C/EBPa protein in the cotransfection experiments proved that the decrease in CYP3A1 activation is not due to a reduction in C/EBPa expression (Fig. 4B). Taken together, these results strongly indicate that both sites 3A1- 300 and 3A1-600 are functional elements of the CYP3A1 promoter. Time-course analysis of the in vivo expression of C/EBPa upon glucocorticoid administration The time-course variation of the relative concentrations of C/EBPa mRNAandproteinwasmonitoredintheratliver. In parallel we also followed the accumulation of the CYP3A1 mRNA, upon addition of dexamethasone (Fig. 5). C/EBPa mRNA concentration increases approximately fivefold over the control value between 0.5 and 4 h after treatment, and decreases to about fourfold of the control value by 21 h after dexamethasone administration. Such an increase clearly precedes the marked induction of hepatic CYP3A1 mRNA found to occur between 4 and 21 h after treatment (Fig. 5A). We found the accumulation of both isoforms of C/EBPa (p42, of 42 kDa and p30, of 30 kDa) to occur concomit- antly with the increase in the relative abundance of the corresponding mRNA. However, the two isoforms accu- mulate with different kinetic profiles (Fig. 5B). Maximal induction (an increase of about 2.5-fold over the basal level) is observed for p30 between 2 and 4 h upon dexamethasone administration, with a gradual decrease to the control value 21 h after treatment. In contrast the levels of the p42 isoform increases from hour 2 onwards, reaching constant Fig. 3. Ability of multimerized CYP3A1-C/EBP responsive elements to confer activation by C/EBPa in the CHO cell line. Cotransfections were carried out with 0.5 lgof3A1-C/EBP-RE-T81Luc and 0.15 lgof pMSV/EBPa expression vector, or empty vector. The normalized luciferase activities are expressed as mean values ± SD of duplicates for a minimum of three experiments. Fig. 4. Functional analysis of the C/EBP binding sites in the CYP3A1 promoter by cotransfection studies using the p0.8T81Luc reporter plas- mid altered by site-directed mutagenesis. (A) Cotransfections were carried out using 0.5 lg of wild-type (wt) or the different mutated promoter reporter plasmids (mut300, mut600 and dm300/600) and 0.15 lg of pMSV/EBPa expression vector, or empty vector. As no significant differences in the basal activity of the different mutated reporter plasmids were found, normalize luciferase activities are expressed as mean values ± S.D. of fold induction of duplicates for a minimum of three experiments. *P < 0.05, **P < 0.01 significant differences in transactivation due to mutation of 3A1-300 or/and 3A1- 600 sites ( ANOVA one-way test: F ¼ 12.29, d.f. ¼ 9, p < 0001; posthoc Tukey HSD test for unequal N: m300, P < 0.05; m600, P <0.01; dm300/600, P < 0.01). (B) Result of a representative Western blot analysis of the 42 kDa C/EBPa protein in the cotransfection experi- ments: lanes (–) with empty vector and (+) with 0.15 lgofpMSV/ EBPa. Fifty lL of total protein extract were analysed by SDS/PAGE (10%, w/v), the resolved proteins transferred to poly(vinylidene difluoride) membranes, and the membranes incubated with a specific anti-C/EBPa Ig. 560 E. Rodrigues et al.(Eur. J. Biochem. 270) Ó FEBS 2003 levels of about twice the control value, which persisted 21 h after treatment. The relative differences observed for the increase in C/EBPa mRNA (approximately fivefold) or protein levels (approximately twofold) is consistent with the previous demonstration of a post-transcriptional regulation of C/EBP protein expression. Actually, a previous report has shown that the stimulatory effect of cAMP on C/EBPb mRNA levels does not result in the alteration of protein levels [36], while others have reported a discrepancy between C/EBPb protein and mRNA levels in various tissues [37]. Our results demonstrate that dexamethasone is an inducer of C/EBPa, exerting a positive effect on the steady-state levels of both the mRNA and protein levels in the liver. The increase in the levels of both hepatic C/ EBPa isoforms clearly precedes induction of the CYP3A1 mRNA, suggesting that these factors may play an active role in the transcriptional activation of the CYP3A1 gene by dexamethasone. In vivo CYP3A1-C/EBPa binding activity We investigated the presence and relative concentration of active C/EBP proteins in the hepatic cell nuclei as a function of the animal treatment. The results of the binding activities of nuclear proteins extracted from control and from dexamethasone treated rats to site 3A1-300 are shown in Fig. 6. A significant increase in the cellular C/EBP protein capable of binding to site 3A1-300 was systematically observed in the living hepatic cell nuclei 2 h after treatment with dexamethasone. The anti-C/EBPa Ig displaces the majority of the complex (Fig. 6; arrowhead) confirming the ability of this regulatory factor to participate in the transcriptional activation of CYP3A1 in vivo. Similar results were obtained for the 3A1-600 site (data not shown). These results strongly suggest the requirement of C/EBPa for the dexamethasone induced transcriptional activation of the CYP3A1 gene. Discussion It has long been recognized that induction of hepatic CYP3A genes by synthetic glucocorticoids differs from the physiological responses that are mediated by the glucocor- ticoid receptor [38]. More recently, the identification of PXR as the transducer of the CYP3A1 adaptive responses to xenobiotics, ligands of this novel receptor, has defined an alternative steroid regulatory pathway [39]. However, other critical transcription factors are likely to participate in the organization of the transcription initiation complex in the de novo activation of the CYP3A1 locus. For instance, the observation that the full induction of CYP3A1 by glucocorticoid cannot be achieved in CV1 cells, which lack a number of hepatic transcription factors [18], illustrates the need for other partner proteins in the synthetic glucocorticoid-induced CYP3A activation in the liver. We have identified two C/EBP sites in the CYP3A1 promoter that are specifically recognized by the C/EBPa transcription factor. These two cis-elements, 3A1-300 and 3A1-600, share 50% and 60% identity with the C/EBP consensus sequence. Moreover, point mutations of the most conserved nucleotides in the cis-element completely preven- ted binding of C/EBPa, thereby abolishing transactivation of the CYP3A1 promoter. Accordingly, the dynamic study of the C/EBP elements performed by transfection-transac- tivation assays, clearly demonstrated the presence of active C/EBP regulatory elements in the 5¢ upstream region of the CYP3A1 gene. Fig. 5. Time-course variation of rat liver CYP3A1 and C/EBPa levels in response to dexamethasone. Total RNA and nuclear extracts were isolated from pools of control or dexamethasone-treated adult rat liver (40 mg per kg body mass) killed at various times after dexamethasone administration. The blots shown are representative of results obtained in three independent experiments. (A) Total RNA (20 lg) was ana- lysed by Northern blotting for CYP3A1 and C/EBPa mRNA. Pho- tographs of the ethidium bromide-stained total RNA, including the 18S and 28S ribosomal RNA bands, used to generate the Northern blots are shown. (B) Western blot analysis of C/EBPa. Liver nuclear proteins (5 lg) were analysed by SDS/PAGE (10%, w/v), transferred to poly(vinylidene difluoride) membranes, and the membranes incu- bated with a specific anti-C/EBPa Ig. Protein loading control using Amido Black staining of the membrane is shown in the bottom panel. Results were quantified using the EAGLE - EYE software package (Stratagene). Ó FEBS 2003 Two C/EBP sites in the CYP3A1 locus (Eur. J. Biochem. 270) 561 While all C/EBP proteins can interact with the same DNA target sequences, C/EBPa but not C/EBPb specifi- cally transactivates the CYP3A1 promoter by binding to the 3A1-300 and 3A1-600 promoter elements. Replacement of one C/EBP factor by another has been proposed as a modulation mechanism of particular genes at different stages of proliferation and differentiation [40,41]. Multiple homodimeric or heterodimeric combinations can occur in vivo to control the biogenesis of a variety of proteins displaying distinct physiological properties. The differential occupancy of the C/EBP promoter element in a particular gene may be dictated by the relative abundance and affinity of the different homo- or heterodimers, as well as by interactions of the different C/EBP proteins with other factors binding to neighbouring regulatory sites. The in vivo observations of the induction of C/EBPa mRNA and protein preceding the accumulation of CYP3A1 mRNA in rat liver suggest the importance of the characterized C/EBPa promoter elements in the up-regulation of this gene. This is consistent with the previously described threefold increase of the C/EBPa mRNA concentration in rat hepatoma cells H4IIE within 4 h of dexamethasone treatment [36]. Transcriptional induction of C/EBP genes by dexamethasone has also been shown in the rat intestinal epithelial crypt cell line IEC-6, in which a rapid (30 min) increase of C/EBPa and C/EBPb mRNA was observed [42]. Others have shown that induction of C/EBPb mRNA by dexamethasone begins 30 min after treatment [43,44]. Dexamethasone transiently reduces the levels of C/EBPa in 3T3L1 adipocytes, as well as in the white adipose tissue [45], while it does not affect the C/EBPa levels in the rat intestinal epithelium crypt IEC-6 cell line [42], indicating that the effect of glucocorticoids on C/EBPa expression may be cell type-specific. Liver differentiation and development is under the physiological control of glucocorticoid hormones that induce the expression of several transcription factors in the hepatocyte. C/EBP concentration was recently shown to vary as a function of the differentiation of long-term human primary hepatocytes in culture [46]. The requirement of C/EBP isoforms for the mediation of hormone-induced expression as been reported for hepatic genes [25], namely P450 genes [47,48]. Synergy in glucocorticoid responsiveness has been well characterized in the case of the phosphoenol- pyruvate carboxykinase gene promoter [49]. Several other hormone responses have also been demonstrated to depend on C/EBP proteins, particularly the a-isoform [50]. The presence of glucocorticoid responsive elements that require C/EBPa for maximal hormone activation has been des- cribed for the gene encoding carbamoylphosphate synthe- tase I as well as other liver expressed genes [35]. The role of C/EBPa in regulating CYP3A gene expres- sion has been previously investigated. C/EBPa, along with D-element binding protein, was shown to increase expres- sion of a construct containing the )169 bp to +11 bp fragment of the CYP3A4 promoter [27]. The present results first point to the importance of C/EBPa for the CYP3A1 induction by dexamethasone and suggest a cause–effect relationship between the variation of C/EBPa and CYP3A1 mRNA in rat liver. This casual link is further supported by the net increase of the functional C/EBPa proteins found in the liver nuclei of the dexamethasone pretreated animals that display the characteristic induction of CYP3A1. Actually, the C/EBPa protein that accumulates in the liver nuclei is capable of recognizing the two cis-acting elements 3A1-300 and 3A1-600 of the CYP3A1 gene promoter as shown here using DNAÆprotein binding assays. The link between dexamethasone treatment, C/EBP induction and CYP3A1 transcriptional activation should nevertheless be unequivocally demonstrated not to be merely a phenomena association. The in vivo and in vitro evidences here described are consistent with two alternative hypotheses for an active role of C/EBPa in the CYP3A1 dexamethasone-responsiveness. Induction of the CYP3A1 could be driven by functional cross-talk interactions of the Fig. 6. In vivo CYP3A1-C/EBP binding activity. EMSA of 5 lg liver nuclear extract proteins from dexamethasone-treated rats using as a probe a radiolabeled double-stranded oligonucleotide corresponding to site 3A1-300. Supershift experiments were performed with a specific anti-C/EBPa Ig. The (s) symbol denotes the position of the C/EBP containing complex and the arrowhead (ss) the position of the DNA- protein shifted by the specific antibody. 562 E. Rodrigues et al.(Eur. J. Biochem. 270) Ó FEBS 2003 activated PXR with C/EBPa, necessary for the CYP3A1 locus full response. Similarly, the glucocorticoid-induced activation of the CYP3A1 promoter has already been shown to primarily reflect the increase in the levels of other transcription factors, namely RXRa [18,51]. An alternative pathway operating for the positive modulation of CYP3A1 expression by glucocorticoids can also be considered based on our results. Such a pathway has been recently proposed by other authors based on the effect of a mutation within the promoter region of the Human CYP3A4, which interrupts a putative complex binding site for both C/EBPa and HNF-3 [48]. The described mutation interferes with the ability of the CYP3A4 promoter to respond to glucocorticoids but does not affect the response to the potent CYP3A4 inducer and PXR-ligand, rifampicin. The nature of a C/EBP mediated response or the putative interactions between PXR and C/EBPa should be further investigated at a molecular level, namely through dynamic transfection–transactivation assay as well as by in vivo footprinting analysis. These studies are necessary for a more complete definition of the transduction pathway triggered by the synthethic glucocorticoids in the hepatic cell that directly modulate the expression of the CYP3A genes. Acknowledgements Supported by Fundac¸ a ˜ oparaaCieˆ ncia e Tecnologia – grant PRAXIS 4/4.1/BD/2999 (E.R); project PRAXIS XXI 2/2.1/BIA/234/94 – and by the Franco-Portuguese Scientific Cooperation Program INSERM/ ICCTI. References 1. Zaret, K.S. (1996) Molecular genetics of early liver development. Annu. Rev. Physiol. 58, 231–251. 2. Gonzalez, F.J. & Lee, Y.H. (1996) Constitutive expression of hepatic cytochrome P450 genes. Faseb J. 10, 1112–1117. 3. Waxman, D.J. (1999) P450 gene induction by structurally diverse xenochemicals: central role of nuclear receptors CAR, PXR, PPAR. Arch. Biochem. Biophys. 369, 11–23. 4. Pereira, T.M. & Lechner, M.C. (1995) Differential regulation of the cytochrome P450 3A1 gene transcription by dexa- methasone in immature and adult rat liver. Eur J. Biochem. 229, 171–177. 5. Gonzalez,F.J.,Song,B.J.&Hardwick,J.P.(1986)Pregnenolone 16a-carbonitrile-inducible P-450 gene family: gene conversion and differential regulation. Mol. Cell Biol. 6, 2969–2976. 6. Kliewer, S.A., Moore, J.T., Wade, L., Staudinger, J.L., Watson, M.A., Jones, S.A., McKee, D.D., Oliver, B.B., Willson, T.M., Zetterstrom, R.H., Perlmann, T. & Lehmann, J.M. (1998) An orphan nuclear receptor activated by pregnanes defines a novel steroid signaling pathway. Cell 92, 73–82. 7. Lehmann, J.M., McKee, D.D., Watson, M.A., Willson, T.M., Moore, J.T. & Kliewer, S.A. (1998) The human orphan nuclear receptor PXR is activated by compounds that regulate CYP3A4 gene expression and cause drug interactions. J. Clin. Invest 102, 1016–1023. 8. Blumberg, B., Sabbagh, W. Jr., Juguilon, H., Bolado, J. Jr., van Meter,C.M.,Ong,E.S.&Evans,R.M.(1998)SXR,anovel steroid and xenobiotic-sensing nuclear receptor. Genes Dev 12, 3195–3205. 9. Huss, J.M., Wang, S.I., Astrom, A., McQuiddy, P. & Kasper, C.B. (1996) Dexamethasone responsiveness of a major gluco- corticoid-inducible CYP3A gene is mediated by elements unrelated to a glucocorticoid receptor binding motif. Proc. Natl Acad. Sci. U S A 93, 4666–4670. 10. Quattrochi, L.C., Mills, A.S., Barwick, J.L., Yockey, C.B. & Guzelian, P.S. (1995) A novel cis-acting element in a liver cytochrome P450, 3A gene confers synergistic induction by glucocorticoids plus antiglucocorticoids. J. Biol. Chem 270, 28917–28923. 11. Huss, J.M. & Kasper, C.B. (1998) Nuclear receptor involvement in the regulation of rat cytochrome P450 3A23 expression. J. Biol. Chem 273, 16155–16162. 12. Ogino, M., Nagata, K., Miyata, M. & Yamazoe, Y. (1999) Hepatocyte nuclear factor 4-mediated activation of rat CYP3A1 gene and its modes of modulation by apolipoprotein AI regulatory protein I and v-ErbA-related protein 3. Arch. Biochem. Biophys. 362, 32–37. 13. Jones, S.A., Moore, L.B., Shenk, J.L., Wisely, G.B., Hamilton, G.A., McKee, D.D., Tomkinson, N.C., LeCluyse, E.L., Lambert, M.H., Willson, T.M., Kliewer, S.A. & Moore, J.T. (2000) The pregnane X receptor: a promiscuous xenobiotic receptor that has diverged during evolution. Mol Endocrinol. 14, 27–39. 14. Bertilsson, G., Heidrich, J., Svensson, K., Asman, M., Jendeberg, L., Sydow-Backman, M., Ohlsson, R., Postlind, H., Blomquist, P. & Berkenstam, A. (1998) Identification of a human nuclear receptor defines a new signaling pathway for CYP3A induction. Proc.NatlAcad.Sci.USA95, 12208–12213. 15. Yanagimoto,T.,Itoh,S.,Sawada,M.&Kamataki,T.(1997) Mouse cytochrome P450 (Cyp3a11): predominant expression in liver and capacity to activate aflatoxin. B1, Arch. Biochem. Biophys. 340, 215–218. 16. Wrighton, S.A., Schuetz, E.G., Watkins, P.B., Maurel, P., Barwick,J.,Bailey,B.S.,Hartle,H.T.,Young,B.&Guzelian,P. (1985) Demonstration in multiple species of inducible hepatic cytochromes P-450 and their mRNAs related to the glucocorti- coid-inducible cytochrome P-450 of the rat. Mol Pharmacol. 28, 312–321. 17. Watkins, P.B., Murray, S.A., Winkelman, L.G., Heuman, D.M., Wrighton, S.A. & Guzelian, P.S. (1989) Erythromycin breath test as an assay of glucocorticoid-inducible liver cytochromes P-450. Studies in rats and patients. J. Clin. Invest 83, 688–697. 18. Huss, J.M. & Kasper, C.B. (2000) Two-stage glucocorticoid induction of CYP3A23 through both the glucocorticoid and pregnane X receptors. Mol Pharmacol. 58, 48–57. 19. Lekstrom-Himes, J. & Xanthopoulos, K.G. (1998) Biological role of the CCAAT/enhancer-binding protein family of transcription factors. J. Biol. Chem 273, 28545–28548. 20. Nishio, Y., Isshiki, H., Kishimoto, T. & Akira, S. (1993) A nuclear factor for interleukin-6 expression (NF-IL6) and the glucocorti- coid receptor synergistically activate transcription of the rat a1-acid glycoprotein gene via direct protein–protein interaction. MolCellBiol.13, 1854–1862. 21. Lee,Y.H.,Yano,M.,Liu,S.Y.,Matsunaga,E.,Johnson,P.F.& Gonzalez, F.J. (1994) A novel cis-acting element controlling the rat CYP2D5 gene and requiring cooperativity between C/EBPb andanSp1factor.MolCellBiol.14, 1383–1394. 22. Lee, Y.H., Williams, S.C., Baer, M., Sterneck, E., Gonzalez, F.J. & Johnson, P.F. (1997) The ability of C/EBP beta but not C/EBP alpha to synergize with an Sp1 protein is specified by the leucine zipper and activation domain. MolCellBiol.17, 2038–2047. 23. Luc, P.V., Adesnik, M., Ganguly, S. & Shaw, P.M. (1996) Tran- scriptional regulation of the CYP2B1 and CYP2B2 genes by C/EBP-related proteins. Biochem. Pharmacol. 51, 345–356. 24. Park, Y. & Kemper, B. (1996) The CYP2B1 proximal promoter contains a functional C/EBP regulatory element. DNA Cell Biol. 15, 693–701. Ó FEBS 2003 Two C/EBP sites in the CYP3A1 locus (Eur. J. Biochem. 270) 563 25. Cassel, T.N., Gustafsson, J.A. & Nord, M. (2000) CYP2B1 is regulated by C/EBPa and C/EBPd inlung epithelial cells. Mol Cell Biol.Res.Commun3, 42–47. 26. Tollet, P., Lahuna, O., Ahlgren, R., Mode, A. & Gustafsson, J.A. (1995) CCAAT/enhancer-binding protein-alpha-dependent trans- activation of CYP2C12 in rat hepatocytes. Mol Endocrinol. 9, 1771–1781. 27. Ourlin, J.C., Jounaidi, Y., Maurel, P. & Vilarem, M.J. (1997) Role of the liver-enriched transcription factors C/EBPa and DBP in the expression of human CYP3A4 and CYP3A7. J. Hepatol. 26, 54–62. 28. Telhada, M.B., Pereira, T.M. & Lechner, M.C. (1992) Effect of dexamethasone and phenobarbital on run-on transcription rate and CYP3A mRNA concentration in rat liver: changes during development. Arch. Biochem. Biophys. 298, 715–725. 29. Pereira, T.M., Carlstedt-Duke, J., Lechner, M.C. & Gustafsson, J.A. (1998) Identification of a functional glucocorticoid response element in the CYP3A1/IGC2 gene. DNA Cell Biol. 17, 39–49. 30. Nordeen, S.K. (1988) Luciferase reporter gene vectors for analysis of promoters and enhancers. Biotechniques 6, 454–458. 31. Auffray, C. & Rougeon, F. (1980) Purification of mouse immunoglobulin heavy-chain messenger RNAs from total myeloma tumor RNA. Eur J. Biochem. 107, 303–314. 32. Hattori, M., Tugores, A., Veloz, L., Karin, M. & Brenner, D.A. (1990) A simplified method for the preparation of transcriptionally active liver nuclear extracts. DNA Cell Biol. 9, 777–781. 33. Schreiber, E., Matthias, P., Muller, M.M. & Schaffner, W. (1989) Rapid detection of octamer binding proteins with Ômini-extractsÕ, prepared from a small number of cells. Nucleic Acids Res. 17, 6419. 34. Johnson, P.F. (1993) Identification of C/EBP basic region residues involved in DNA sequence recognition and half-site spacing preference. MolCellBiol.13, 6919–6930. 35. Christoffels, V.M., Grange, T., Kaestner, K.H., Cole, T.J., Darlington, G.J., Croniger, C.M. & Lamers, W.H. (1998) Gluco- corticoid receptor, C/EBP, HNF3, and protein kinase A coordi- nately activate the glucocorticoid response unit of the carbamoylphosphate synthetase. I Gene, Mol Cell Biol. 18, 6305– 6315. 36. Crosson, S.M., Davies, G.F. & Roesler, W.J. (1997) Hepatic expression of CCAAT/enhancer binding protein a: hormonal and metabolic regulation in rats. Diabetologia 40, 1117–1124. 37. Descombes, P., Chojkier, M., Lichtsteiner, S., Falvey, E. & Schi- bler, U. (1990) LAP, a novel member of the C/EBP gene family, encodes a liver-enriched transcriptional activator protein, Genes Dev 4, 1541–1551. 38. Schuetz, E.G., Wrighton, S.A., Barwick, J.L. & Guzelian, P.S. (1984) Induction of cytochrome P-450 by glucocorticoids in rat liver. I. Evidence that glucocorticoids and pregnenolone 16a-car- bonitrile regulate de novo synthesis of a common form of cyto- chrome P-450 in cultures of adult rat hepatocytes and in the liver in vivo. J. Biol. Chem 259, 1999–2006. 39. Kliewer, S.A., Lehmann, J.M. & Willson, T.M. (1999) Orphan nuclear receptors: shifting endocrinology into reverse. Science 284, 757–760. 40. Cao, Z., Umek, R.M. & McKnight, S.L. (1991) Regulated expression of three C/EBP isoforms during adipose conversion of 3T3-L1 cells. Genes Dev. 5, 1538–1552. 41. Poli, V., Mancini, F.P. & Cortese, R. (1990) IL-6DBP, a nuclear protein involved in interleukin-6 signal transduction, defines a new family of leucine zipper proteins related to C/EBP. Cell 63, 643– 653. 42. Boudreau, F., Blais, S. & Asselin, C. (1996) Regulation of CCAAT/enhancer binding protein isoforms by serum and gluco- corticoids in the rat intestinal epithelial crypt cell line IEC-6. Exp Cell Res. 222, 1–9. 43. Matsuno, F., Chowdhury, S., Gotoh, T., Iwase, K., Matsuzaki, H., Takatsuki, K., Mori, M. & Takiguchi, M. (1996) Induction of the C/EBPb gene by dexamethasone and glucagon in primary- cultured rat hepatocytes. J. Biochem. (Tokyo) 119, 524–532. 44. Gotoh, T., Chowdhury, S., Takiguchi, M. & Mori, M. (1997) The glucocorticoid-responsive gene cascade. Activation of the rat arginase gene through induction of C/EBPb. J. Biol. Chem 272, 3694–3698. 45. MacDougald,O.A.,Cornelius,P.,Lin,F.T.,Chen,S.S.&Lane, M.D. (1994) Glucocorticoids reciprocally regulate expression of the CCAAT/enhancer- binding protein a and d genes in 3T3-L1 adipocytes and white adipose tissue. J. Biol. Chem 269, 19041– 19047. 46. Ferrini, J.B., Rodrigues, E., Dulic, V., Pichard-Garcia, L., Fabr, J.M.,Blanc,P.&Maurel,P.(2001)ExpressionandDNA-binding activity of C/EBPa and C/EBPb in human liver and differentiated primary hepatocytes. J. Hepatol. 35, 170–177. 47. Yamada, K., Duong, D.T., Scott, D.K., Wang, J.C. & Granner, D.K. (1999) CCAAT/enhancer-binding protein beta is an acces- sory factor for the glucocorticoid response from the cAMP response element in the rat phosphoenolpyruvate carboxykinase gene promoter. J. Biol. Chem 274, 5880–5887. 48. El-Sankary, W., Bombail, V., Gibson, G.G. & Plant, N. (2002) Glucocorticoid-mediated induction of CYP3A4 is decreased by disruption of a protein:DNA interaction distinct from the preg- nane X receptor response element. Drug Metab Dispos 30, 1029– 1034. 49. Croniger, C., Leahy, P., Reshef, L. & Hanson, R.W. (1998) C/EBP and the control of phosphoenolpyruvate carboxykinase gene transcription in the liver. J. Biol. Chem 273, 31629–31632. 50. Crosson, S.M. & Roesler, W.J. (2000) Hormonal regulation of the phosphoenolpyruvate carboxykinase gene. Role specific CCAAT/enhancer-binding protein isoforms. J. Biol. Chem 275, 5804–5809. 51. Pascussi, J.M., Drocourt, L., Fabre, J.M., Maurel, P. & Vilarem, M.J. (2000) Dexamethasone induces pregnane X receptor and retinoid X receptor-a expression in human hepatocytes: synergistic increase of CYP3A4 induction by pregnane X receptor activators. Mol. Pharmacol. 58, 361–372. 564 E. Rodrigues et al.(Eur. J. Biochem. 270) Ó FEBS 2003 . Two CCAAT/enhancer binding protein sites in the cytochrome P4503A1 locus Potencial role in the glucocorticoid response Elsa Rodrigues 1 ,. need for other partner proteins in the synthetic glucocorticoid- induced CYP3A activation in the liver. We have identified two C/EBP sites in the CYP3A1 promoter