α1-Fetoprotein Transcription Factor Is Required for the Expression of Sterol 12α-Hydroxylase, the Specific Enzyme for Cholic Acid Synthesis

Cholesterol conversion to bile acids occurs via the “classic” (neutral) or the “alternative” (acidic) bile acid biosynthesis pathways. Sterol 12α-hydroxylase/CYP8b1 is the specific enzyme required for cholic acid synthesis. The levels of this enzyme determine the ratio of cholic acid to chenodeoxycholic acid and thus the hydrophobicity of the circulating bile acid pool. Expression of the 12α-hydroxylase gene is tightly down-regulated by hydrophobic bile acids. In this study, we report the characterization of two DNA elements that are required for both the 12α-hydroxylase promoter activity and bile acid-mediated regulation. Mutation of these elements suppresses 12α-hydroxylase promoter activity. Mutations of any other part of the promoter do not alter substantially the promoter activity or alter regulation by bile acids relative to the wild type promoter. These two DNA elements bind α1-fetoprotein transcription factor (FTF), a member of the nuclear receptor family. We also show that overexpression of FTF in a non-liver cell line activates the sterol 12α-hydroxylase promoter. These studies demonstrate the crucial role of FTF for the expression and regulation of a critical gene in the bile acid biosynthetic pathways.

Proper control of intracellular and circulating cholesterol levels is essential for maintaining cholesterol homeostasis, since metabolic disarrangement can lead to many degenerative conditions with a probable genetic component, such as atherosclerosis, cholestasis, and cholesterol gallstone disease. Because nearly 50% of the body cholesterol is catabolized to bile acids, this pathway plays an important role in the cholesterol homeostasis of mammals. Current evidence suggests a decreased bile acid output as the major contributing factor in the production of lithogenic bile. This leads to an abnormal ratio of cholesterol to bile acids and lecithin, which is a major risk factor for cholesterol gallstone formation (1).
Cholesterol conversion to bile acids occurs via the "classic" (neutral) or the "alternative" (acidic) bile acid biosynthesis pathways (2). Cholic acid and chenodeoxycholic acid (CDCA) 1 are the end products of these pathways (Fig. 1) and the major primary bile acids found in most vertebrates. Cholic acid is hydroxylated at position 12␣, whereas CDCA is not. There are three enzymes that play major regulatory roles in these two pathways. Cholesterol 7␣-hydroxylase/CYP7a1 (7␣-hydroxylase) is the rate-limiting enzyme in the classic pathway. Sterol 27-hydroxylase/CYP27 is the first enzyme in the alternative pathway. Sterol 12␣-hydroxylase/CYP8b1 (12␣-hydroxylase) is the specific enzyme for cholic acid synthesis and determines the ratio of cholic acid to chenodeoxycholic acid and thus the hydrophobicity of the circulating pool.
The altered ratio of cholic to CDCA has been postulated to play a role in cholesterol gallstone formation (3). Suppression of 12␣-hydroxylase by specific inhibitors has been suggested as a possible therapeutic strategy for dissolution of cholesterol gallstone (3). Because chenodeoxycholic acid is a more potent suppressor of HMG-CoA reductase and 7␣-hydroxylase than cholic acid (3,4), the relative activity of 12␣-hydroxylase may play an important role in the regulation of hepatic cholesterol homeostasis. The alteration of cholic/CDCA ratio affects biliary cholesterol and phospholipid secretion, thus altering intestinal cholesterol absorption and receptor-mediated lipoprotein uptake by the hepatocyte (4).
It is well documented that bile acids exert negative feedback regulation on their own synthesis (5). Interruption of the enterohepatic circulation, by biliary diversion or by feeding bile acid-binding resins (cholestyramine), enhances cholesterol and bile acid synthesis (6). Conversely, feeding bile acids suppresses bile acid and cholesterol synthesis (7). Bile acids negatively regulate the transcription of the 7␣-hydroxylase gene, which controls output from the classic pathway (8). Two bile acid response elements have been localized within the 5Ј-flanking region of the rat gene (9, 10), but the factors that mediate regulation have not been characterized. It remains to be demonstrated whether these elements are indeed involved in this regulation. Similarly, bile acids also down-regulate transcription of the sterol 27-hydroxylase gene (11), and it has been reported that hepatocyte nuclear factor 1 (HNF-1) is involved in this regulation (12). However, the molecular mechanisms involved in that regulation are not well defined.
More recently, it has been shown that a transcriptional fac-tor, named the CYP7A promoter-binding factor (CPF), is required for the expression of the 7␣-hydroxylase gene (13). This factor is a member of the Ftz-F1 family of the class IV orphan nuclear receptor superfamily (14). CPF binds to the region of the 7␣-hydroxylase promoter previously characterized as a bile acid response element (10), which suggests that CPF might play a role in the bile acid-mediated down-regulation of 7␣hydroxylase transcription. The cDNAs and genes encoding the rabbit, rat, and human 12␣-hydroxylase enzyme have been cloned (15)(16)(17), and studies of the molecular basis of its regulation are now feasible. It is expressed exclusively in the liver, since it corresponds to a liver-specific process. Recently, Vlahcevic et al. (18) showed that expression of the 12␣-hydroxylase gene is tightly regulated in a similar fashion to the 7␣-hydroxylase gene. Cholestyramine-fed rats contained approximately 2-fold more 12␣hydroxylase mRNA than control rats, whereas deoxycholate-or cholate-fed rats had undetectable levels of 12␣-hydroxylase mRNA (at least 10-fold less than controls). Interestingly, cholesterol feeding decreased the amount of 12␣-hydroxylase mRNA by about 2-fold. Similar regulation was observed for both enzymatic and transcriptional activity, which indicates that both bile acids and cholesterol regulate the expression of the 12␣-hydroxylase gene mainly at the transcriptional level.
In this study, we have characterized the 12␣-hydroxylase promoter. We have localized two DNA elements that are required for both expression of the 12␣-hydroxylase promoter and bile acid-mediated regulation. These elements bind ␣ 1 -fetoprotein transcription factor (FTF) (19), a member of the Ftz-F1 family of receptors. Rat FTF is the ortholog of mouse liver receptor homolog 1 2 and human PHR-1 (21). We have also shown that when FTF is expressed in a non-liver cell line, the 12␣-hydroxylase promoter becomes active, demonstrating the critical role of FTF for 12␣-hydroxylase promoter activity. These studies demonstrate the crucial role of FTF in the bile acid biosynthetic pathway.

EXPERIMENTAL PROCEDURES
Materials-Reagents used in DNA cloning and sequencing were from New England Biolabs, Roche Molecular Biochemicals, U.S. Biochemical Corp., or Life Technologies, Inc. Common laboratory chemicals were from Fisher, Sigma, or Bio-Rad. The luciferase promoterless vector, pGL3-Basic, was purchased from Promega. Oligonucleotides were prepared in the Medical College of Virginia DNA Synthesis Facility by the phosphoramidite method on an automated DNA synthesizer. pCI-FTF, an expression plasmid that contains the human FTF cDNA in the expression vector pCI (Promega), was a generous gift from Dr. Bélanger (19). pCMX, a plasmid for expression in mammalian cells and in vitro, contains the cytomegalovirus and the T7 promoters and was a gift from Dr. Ronald M. Evans. Anti-FTF antibodies were a gift from Dr. David W. Russell and were raised against a peptide corresponding to amino acids 180 -1197 of the DNA binding domain.
General Methods-Standard recombinant DNA procedures were carried out essentially as described (8). DNA sequencing was done by the dideoxy chain termination method using DNA fragments subcloned into M13 vectors or with double-stranded clones and the universal primer or sequence-specific primers with reagents from U.S. Biochemical Corp.
Genomic Cloning and Characterization-A rat genomic library was obtained from CLONTECH. The library was plated and screened with a doubled-stranded probe made by reverse transcriptase-polymerase chain reaction and contained a 200-nt fragment from the 5Ј-end of the rabbit 12␣-hydroxylase cDNA. The polymerase chain reaction primers were synthesized based on published DNA sequence (22). Southern blot analysis was done to characterize the clone. About 3 kilobases was sequenced by the dideoxy method.

FIG. 1. Bile acid biosynthetic pathways.
A highly schematic depiction of the bile acid metabolic pathways is shown, and the key enzymes for both pathways are noted. Conversion of cholesterol to 7␣-hydroxycholesterol by 7␣-hydroxylase is the initial and rate-limiting step in the classic pathway. Hydroxylation at position 27 is the initial step of the acidic pathway. Cholic acid and chenodeoxycholic acid are the major products of these pathways and are commonly referred to as the primary bile acids. constructs were confirmed by DNA sequencing.
Transient Transfection and Luciferase Assays-HepG2 and CV-1 cells were obtained from American Type Culture Collection. Both cell types were transfected with Lipofectin (Life Technologies, Inc.), using 1.5 g of total DNA in 35-mm plates. HepG2 were transfected with 25 ng of test plasmid and 5 ng of pCMV-Gal (a plasmid containing the human cytomegalovirus promoter in from of the bacteria galactosidase gene) to normalize for transfection efficiencies. CV-1 cells were transfected with 200 ng of test plasmid, 50 ng of pCMV-Gal, and the indicated amounts of pCI-FTF, an expression plasmid that contains the FTF cDNA driven by the cytomegalovirus promoter. After 16 h, the DNA was removed, and where indicated, 100 M CDCA was added. Cells were harvested 48 h later, and luciferase and ␤-galactosidase assays were performed with a kit from Tropix (Bedford, MA), according to the manufacturer's protocol. Average values are for the number of experiments indicated.
In Vitro Transcription/Translation and HepG2 Nuclear Extracts-Transcription/translation of cDNAs encoding FTF or of the growth hormone receptor (GHR) as a control was performed using the TNT T7-coupled rabbit reticulocyte lysate system according to the manufacturer's protocol (Promega). HepG2 nuclear extracts were prepared as indicated (24).
Electrophoretic Mobility Shift Analysis-DNA binding reactions were set up in 50 mM KCl, 20 mM Tris-HCl (pH 8.0), 0.2 mM EDTA, 4% Ficoll, 1.0 g of poly (dI-dC), 4 l of the translated protein, and a 1500-fold molar excess of an irrelevant single-stranded DNA, in a final volume of 20 l on ice. After a 15-min incubation, 320 fmol of the indicated 32 P-labeled DNA probes (ϳ2 ϫ 10 5 cpm) were added. All probes were adjusted to the same specific radioactivity. After incubation for 20 min on ice, samples were loaded onto a 4.5% polyacrylamide gel and subjected to electrophoresis at 4°C. Gels were dried and exposed to XAR-5 (Eastman Kodak Co.). For supershift experiments, 1 l of the crude serum was used.

RESULTS
The DNA sequence of the 5Ј-flanking region of the 12␣hydroxylase gene is shown in Fig. 2. The transcriptional initiation site was located by primer extension techniques (data not shown) and is numbered ϩ1. This sequence contains a TATA box-like element, two consensus binding sites for HNF-3, a liver-specific DNA-binding protein (16), and two sterol regula-tory elements (SREs) that are implicated in cholesterol-mediated regulation of the transcription of several genes (17). One stretch of DNA located between nucleotides Ϫ63 and Ϫ48 contains two potential sites for the liver-specific nuclear receptor FTF (19).
To functionally characterize the 12␣-hydroxylase promoter, we prepared a chimeric gene (pGL3-R12␣-865, as shown in Fig.  3) by fusing the SacI-SacI fragment (Fig. 2), which contains 865 base pairs of the 5Ј-flanking region and 33 base pairs of 5Ј-untranslated region, to the coding region of the luciferase gene. We then used HepG2 cells as recipient cells to transfect this chimeric gene. After transfection, cells were treated with or without 100 M CDCA, which was shown to suppress expression of the endogenous gene in primary hepatocytes (data not shown), mimicking the bile acid-mediated regulation that has been described in vivo (18). Cells were harvested for luciferase and galactosidase activities as explained under "Experimental Procedures." As shown in Fig. 3, pGL3-R12␣-865 had promoter activity (100-fold above background levels) that was regulated approximately 5-fold upon the addition of CDCA.
To narrow down the promoter region involved in both transcriptional activity and bile acid-mediated regulation, we made the three 5Ј-deletion constructs shown in Fig. 3. All three constructs showed regulated activity, indicating that all of the DNA elements required for both promoter activity and bile acid-mediated regulation are located in the first 106 nucleotides of the 12␣-hydroxylase promoter.
To further characterize the 12␣-hydroxylase promoter, we mutated the first 106 nucleotides in blocks of approximately 20 nucleotides each as shown in Fig. 2 (brackets A-G). The results of these experiments are shown in Fig. 4A, and the actual mutations introduced in mutants D, E, and F are shown in Fig.  4B. All mutants exhibited promoter activity and bile acidmediated regulation, except when nucleotides Ϫ62 to Ϫ49 were mutated. This region contains two putative elements with homology to the rat FTF site, a member of the Drosophila orphan receptor fushi tarazu F1 (Ftz-F1) (25), and we named them FTF sites. These elements are also homologous to the recently described CPF site for the 7␣-hydroxylase gene (26), another nuclear receptor of the same family. These homologies are shown in Fig. 5.
Based on these homologies, we hypothesized that there are two FTF elements within the 12␣-hydroxylase promoter located at positions Ϫ63 to Ϫ55 and Ϫ56 to Ϫ48. The data in Fig.  4A show that both mutants D and F, which have the two potential FTF sites independently mutated, and mutant E, which has both sites mutated, had very low, if detectable, activities. This suggests that both elements are required for activity. Fig. 6 shows that in vitro synthesized FTF binds to the 12␣-hydroxylase promoter (lanes 2 and 3). When in vitro made FTF was preincubated with antibodies raised against a peptide corresponding to the DNA binding domain of FTF, binding was mostly abolished (lane 5), demonstrating that the protein binding to the 12␣-hydroxylase probe is in fact FTF. Preimmune serum did not produce any effect (lane 4). Mutation of 5 nucleotides (Ϫ62 to Ϫ58) within the first site diminished binding about 5-fold (lanes 7 and 8). Mutation of 4 nucleotides (Ϫ52 to Ϫ49) within the second putative site also diminished binding by about 5-fold (Fig. 6A, lanes 13 and 14). Mutation of 8 nucleotides (Ϫ56 to Ϫ49) that modifies both sites abolished binding completely (lanes 10 and 11). As a negative control, we used in vitro synthesized GHR (lanes 1, 6, 9, and 12).
To determine the specificity of the binding and if the 7␣hydroxylase CPF site binds to the same protein, we performed the competition experiment shown in Fig. 6B. Wild type probe competed as expected (lanes 3-5). A probe made from the rat 7␣-hydroxylase promoter sequence (Ϫ150 to Ϫ131) containing the described CPF site (13) competed even better than the 12␣-hydroxylase probe itself (lanes 6 -8), suggesting that the 7␣-hydroxylase CPF site has higher affinity for FTF than the 12␣-hydroxylase sites. A mutated 12␣-hydroxylase fragment (mutant E), with nucleotides mutated in both FTF sites, did not compete for binding (lanes 9 -11), confirming the specificity of the binding.
To demonstrate that the protein binding to the 12␣-hydroxylase FTF sites exists in liver cells, HepG2 nuclear extracts were used for binding experiments (Fig. 6C, lanes 7-9). The binding activity found in HepG2 cells was also inhibited by incubation with a specific antibody against FTF (lane 9), and the DNA-protein complexes formed exhibited mobility identical to that of the complex formed by in vitro made FTF (lanes 4 -6). As a control, we used a probe from the rat 7␣-hydroxylase promoter known to bind CPF, a highly homologous protein to FTF (13). Both in vitro made FTF and HepG2 nuclear extracts bind the 7␣-hydroxylase probe, and that binding was specific as indicated by the fact that anti-FTF antibodies also prevented binding (Fig. 6C, lanes 10 -18). The 7␣-hydroxylase probe binds FTF at higher affinity that the 12␣-hydroxylase probe, in agreement with the competition experiments shown in Fig. 6B.
To further demonstrate the key role of FTF for 12␣-hydroxylase promoter activity, we overexpressed FTF in the kidney cell line CV-1 (Fig. 7). Since 12␣-hydroxylase is expressed only in the liver, pGL3-R12␣-865 was inactive in CV-1 cells as expected. However, when we cotransfected pCI-FTF, pGL3-R12␣-865 became active. The level of activation was nearly 13-fold when 200 ng of pCI-FTF was used. As a control, we used pGL3-Basic and pGL3-R12␣-865 mutant E, which had no promoter activity in HepG2 cells (Fig. 4). The activities from these two plasmids were the same as pGL3-R12␣-865 when no pCI-FTF was included in the transfection. Overexpression of FTF produced only a slight activation (2-3-fold) of both the promoterless plasmid pGL3-Basic and the pGL3-R12␣-865 mutant E. For comparison, a construct containing the rat 7␣-hydroxylase promoter was activated only 8-fold when 2-fold more pCI-FTF (400 ng) was cotransfected (data not shown), suggesting a greater activity of FTF for the 12␣-hydroxylase promoter than the 7␣-hydroxylase promoter. DISCUSSION The data presented in this study demonstrate that FTF is a factor required for 12␣-hydroxylase expression. Several lines of evidence support this conclusion. First, mutagenesis of either of its two binding sites abolished promoter activity completely in HepG2 cells (Fig. 4); therefore, both sites are required for promoter activity. Second, in vitro-made FTF binds specifically to these sites (Fig. 6), and the same binding activity was observed in nuclear extracts prepared from HepG2 cells. Third, expression of FTF in a non-liver cell line activates the 12␣hydroxylase promoter, which is otherwise inactive (Fig. 7). Additionally, this study also demonstrates that all of the DNA elements required for the bile acid-mediated regulation of 12␣hydroxylase promoter activity are located in the first 106 nucleotides of the 5Ј-flanking region (Fig. 3).
The 12␣-hydroxylase promoter sequence also has two other potential regulatory elements. Two putative HNF-3 sites are located at approximately nucleotides Ϫ455 and Ϫ385 (Fig. 2). HNF-3 sites are found in the promoter of some liver-specific genes and are required for the expression of those genes (27). Additionally, two SRE sites are located at approximately nucleotides Ϫ315 and Ϫ328. SREs are regulatory elements found in cholesterol and fatty acid-regulated genes (28). However, The DNA fragment from the 5Ј-flanking region of the rat CYPb1 gene is indicated by a thin line with the XhoI site as a reference point. The luciferase coding sequence is denoted by the hatched bar, and the TATA-like box is indicated. A scheme of the different 5Ј deletion constructs used is shown. HepG2 cells were transfected with the indicated construct and treated with chenodeoxycholic acid as described under "Experimental Procedures." Relative transcription was determined by normalizing luciferase activity to ␤-galactosidase activity. The data were normalized to the activity produced by the construct containing the longest promoter fragment, pGL3-R12␣-865, in cells grown in the absence of CDCA and represent the averages of n experiments Ϯ S.D. bp, base pairs. neither the HNF-3 sites nor the SRE sites seem to be required for either promoter activity or regulation by bile acids, since deletion of these sites has very little effect, if any (Fig. 3). It is conceivable that the SRE sites are involved in the cholesterolmediated regulation of 12␣-hydroxylase transcription that has been observed in rats (18), although this was not the objective of the present studies.
Sterol 12␣-hydroxylase is the second gene involved in bile acid biosynthesis that has been shown to require a member of the Ftz-F1 family of the class IV orphan nuclear receptor superfamily for its expression. Recently, it has been shown that another member of the same family, CPF, is required for 7␣-hydroxylase expression (13). CPF also activates the 12␣-hydroxylase promoter. In transactivation experiments similar to these shown in Fig. 6, CPF also activated 12␣-hydroxylase promoter activity, although to a lesser extent (data not shown). It is possible that several members of the Ftz-F1 family have similar activity on genes involved in bile acid biosynthesis. In fact, it has been shown that at least both CPF and CPF variant 1 (another member of the same family) were active on the 7␣-hydroxylase promoter (13). This suggests a crucial role of this family of factors in the control of bile acid biosynthesis.
An interesting issue is the apparent existence of two FTF binding sites within the 12␣-hydroxylase promoter. These two FTF sites were located based on homology with other FTF sites, and they overlap by the last 2 nucleotides of the first site based on published consensus sequences (13,19,25). Although DNA probes containing an individual mutation of either site still bind FTF weakly (Fig. 6A), both sites are required for an active promoter (Fig. 4A). Nuclear receptors of the Ftz-F1 family are known to bind as monomers (19), and given the migration pattern of the protein-probe complex, only one protein molecule seemed to bind per probe molecule. Thus, mutant D, which has only the first site mutated (Fig. 4B), has some binding capability ( Fig. 6A) but has no promoter activity (Fig. 4A). Mutant F, which has 4 nucleotides mutated within the second site, has a binding capability similar to that of mutant D (Fig. 6A) and has some promoter activity (Fig. 4A), but much lower than wild type. Mutant E, which has 8 nucleotides mutated across both binding sites, lost all binding (Fig. 6A) and promoter activity. The most likely explanation to this issue is that binding of FTF to the 12␣-hydroxylase promoter site requires more nucleotides than the binding of FTF or homologous factors to other promoters, and in fact, there may be only one extended binding site.
Whether the activity of FTF is regulated by small molecules in a way similar to most nuclear receptors is still unknown. Attempts in our laboratory to identify a bile acid molecule or derivative that could act as ligand and/or regulator of FTF have not been successful. It is possible that there is another unidentified factor of the same family that is specific for genes of the bile acid biosynthetic pathways and that this factor has a ligand binding activity for bile acids or a derivative.
Besides the similarity between the 12␣-hydroxylase and 7␣hydroxylase promoters in the requirement for FTF or a homologous factor such as CPF for its expression, these two genes FIG. 4. Expression and bile acid-mediated regulation of 12␣hydroxylase promoter/luciferase block mutants in HepG2 cells. A, HepG2 cells were transfected with the indicated constructs and treated with CDCA as described under "Experimental Procedures." Capital letters refer to the promoter sections mutated as shown in Fig.  1. Negative numbers in the mutation column refer to the first and last nucleotides mutated in each construct. The data were normalized to the activity produced by the wild type construct, pGL3-R12␣-865, in cells grown in the absence of CDCA and represent the averages of n experiments Ϯ S.D. ND, not detectable. The sequence containing the two FTF sites (indicated by dashed boxes) is shown at the bottom, with the nucleotides mutated in mutants D, E, and F noted by brackets. B shows the mutations (lowercase) introduced in mutants D, E, and F (also used in the experiments shown in Fig. 6).
FIG. 5. Homology between the 12␣-hydroxylase promoter FTF sites and other homologous binding sites. Nucleotides Ϫ63 to Ϫ48 from the rat 12␣-hydroxylase promoter are shown at the top, with the two FTF sites boxed. These two sites are aligned with the rat FTF site (19); the Drosophila Ftz-F1 consensus binding site (25); and the human, rat, and hamster 7␣-hydroxylase CPF sites (13). seem to differ in the factors required for bile acid regulation of their transcription. In experiments using HepG2 cells and with CDCA, the 7␣-hydroxylase promoter required exogenous far-nesol X receptor in order to show bile acid-mediated regulation of its activity (20), whereas the 12␣-hydroxylase promoter did not show this requirement (Figs. 3 and 4). FIG. 6. FTF binds to the 12␣-hydroxylase promoter. Gel shift experiments were performed as described under "Experimental Procedures," using the indicated amounts of in vitro synthesized FTF protein or in vitro made GHR as a control. In A, five different probes were used, the wild type 12␣-hydroxylase sequence from nucleotides Ϫ62 to Ϫ58 (lanes 1-5) and the three indicated mutants (lanes 6 -14). Preimmune or immune serum against FTF was obtained and used as indicated under "Experimental Procedures" (lanes 4, 5, 18, and 19). An arrow points to the retarded band. The sequence containing the two FTF sites (indicated by dashed boxes) is shown at the bottom, with the nucleotides mutated in mutants D, E, and F noted by brackets. In B, increasing amounts of the indicated competitor DNA was used. In C, both in vitro FTF (ivFTF) and in vitro GHR (ivGHR) as a control and HepG2 nuclear extracts were used for binding.
Although this study was not specifically directed to study the factors involved in the bile acid-mediated regulation of 12␣hydroxylase expression, our data strongly suggest that FTF is also implicated in that regulation. This is based on the fact that only mutations within the FTF binding sites abolished regulation. All deletion promoter constructs (Fig. 3) and all block mutant constructs (Fig. 4) are regulated by bile acids except for the two constructs that abolished FTF binding. It could be argued that elimination of FTF binding renders the promoter inactive, and no conclusion could be drawn about its regulation. However, mutation of the rest of the promoter did not alter bile acid-mediated regulation, and therefore the FTF sites should be involved in the regulation either directly or indirectly. Whether the FTF binding sites or FTF itself is capable of interacting with a bile acid-regulated factor is still unknown. In conclusion, this study demonstrates the key role of FTF or its homologues in the regulation of bile acid biosynthesis and should help to elucidate the molecular mechanisms involved in the bile acid-mediated down-regulation of gene transcription, a process poorly understood to date. FIG. 7. Overexpression of FTF provides 12␣-hydroxylase promoter activity in a non-liver cell line. pCI-FTF was cotransfected in CV-1 cells together with pGL3-Basic, pGL3-R12␣-865, or pGL3-R12␣-865 mutant E (Fig. 3) with nucleotides Ϫ56 to Ϫ49 mutated. Cells were harvested and analyzed for luciferase and ␤-galactosidase activities as described under "Experimental Procedures." The data were normalized to the activity of each construct in the absence of pCI-FTF and represents the averages of three experiments. Error bars denote the S.D. value.