Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Glucuronidation is a major pathway of drug metabolism in mammals. Many xenobiotics, including drugs and environmental pollutants, as well as endogenous substances such as bilirubin and steroid hormones, are glucuronidated and then excreted via either the urine or the bile (Mackenzie, 1995). In general, the resulting glucuronides are end-products of metabolism that have lost the pharmacological activities of the original compound. However, some glucuronides such as the potent analgesic, morphine-6-glucuronide (Shimomura et al., 1971), are more active than their parent aglycones. The UDP glucuronosyltransferases (UGTs) that catalyze this reaction have been classified into two families, designated UGT1 and UGT2 (Mackenzie, 1995; Mackenzie et al., 1997). The UGT2 family has been subdivided into UGT2A forms, which are olfactory specific, UGT2B forms, and a rabbit UGT2C form. To date, the human UGT2B subfamily consists of six forms (reviewed in Mackenzie et al., 1997). Human UGT2B7 is a major form in liver, where it catalyzes the glucuronidation of bile acids, steroids, and many foreign compounds possessing a hydroxyl or carboxylic acid moiety such as nonsteroidal anti-inflammatory drugs (Ritter et al., 1990; Jin et al., 1993, 1997). UGT2B7 contributes to the glucuronidation of 3-hydroxysteroids (Jin et al., 1997) and appears to be the main enzyme responsible for morphine-6-glucuronide formation in humans (Coffman et al., 1997). Interindividual differences in UGT2B7 mRNA levels have been observed (Chen et al., 1993). Because these differences may be a determinant of the pharmacological significance of morphine-6-glucuronide as a potent metabolite of morphine, it is important that they are identified. An understanding of the transcriptional regulation of the UGT2B7 gene should help to identify these factors and should provide a framework to search for regulatory polymorphisms that affect UGT2B7 expression.

As a first step toward understanding the mechanisms that regulate UGT2B7 expression, we have isolated and characterized cosmid clones encoding the UGT2B7 gene. The transcription start site was mapped and binding sites for nuclear proteins in the proximal promoter of the UGT2B7 gene were identified by DNase I footprint assay. The importance of one of these binding sites, termed region A (2 to 42) in the regulation of the UGT2B7 gene was assessed by functional and DNA binding assays.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials

Restriction enzymes and calf intestinal phosphatase were obtained from New England Biolabs (Beverly, MA). Poly(dI-dC) was purchased from Boehringer Mannheim (Indianapolis, IN), dNTPs and DNase I from Pharmacia (Piscataway, NJ), [³⁵S]dATP from Amersham, [-³²P]ATP from Bresatec (Adelaide, Australia), Taq polymerase from Perkin-Elmer (Norwalk, CT), and ELONGASE from Life Technologies (Rockville, MD). pGL3-basic, pGL3-control, pRL-TK, and the Dual-Luciferase detection kit were purchased from Promega (Madison, WI). The hepatic nuclear factor (HNF)-1 and Oct-1 expression plasmids were kind gifts from Dr. Gerald Crabtree (Stanford University, Stanford, CA) and Dr. Rick Sturm (University of Queensland, Brisbane, Australia), respectively. Antibodies specific for HNF1, HNF1, and Oct-1, and specific blocking peptide to each antibody were obtained from Santa Cruz Biotechnology (Santa Cruz, CA).

Methods

Isolation of the UGT2B7 Gene. Two genomic clones, COS-16 and COS-17, containing the human UGT2B7 gene were isolated from a lymphocyte genomic DNA library constructed in the Spcos2 cosmid vector (Kimura et al., 1989) with ³²P-labeled UGT2B7 variant cDNA (Jin et al., 1993) as a probe. The cosmid clones were analyzed by restriction mapping. All UGT2B genes characterized to date have the same exon/intron structure. On the basis of this knowledge, oligonucleotides to UGT2B7 cDNA sequence were designed to amplify across putative UGT2B7 introns. The sizes of the polymerase chain reaction (PCR) products were determined on agarose gels by comparison with appropriate DNA size markers, and exon/intron boundaries were identified by sequencing the PCR products. A 1.7-kb HindIII fragment of COS-17 that contained the proximal promoter was subcloned into pBluescriptII-SK(+) (Stratagene, La Jolla, CA) and sequenced.

Determination of the UGT2B7 Transcription Start Site. The transcription start site was determined by RNase protection and primer extension analyses. RNase protection analysis was performed according to the RPA II RNase Protection Assay Kit protocol (Ambion AS, Austin, TX). Human liver polyA(+) RNA was used with and without the addition of RNase A. After denaturing at 80°C, reaction products were separated on a 6% denaturing polyacrylamide gel, together with products of known UGT2B7 DNA sequence, which were used as marker. After electrophoresis for 3 h at 60 W, gels were processed for autoradiography.

Footprint Analysis. Footprint analysis was performed according to the method of Kroeger and Abraham (1997), by using Dynabeads M-280 Streptavidin-coated beads (Dynal Inc., Oslo, Norway) and biotinylated primers. A 229-base pair (bp) fragment of the UGT2B7 promoter (171 to +58 bp) was generated by PCR with primer sets for both the sense (5'-GCACTCATAAAGATAAAAGG-3', biotinylated-5'-CTTGGTGCAATGCAATGCTT-3') and anti-sense (biotinylated-5'-GCACTCATAAAGATAAAAGG-3', 5'-CTTGGTGCAATGCAA- TGCTT-3') orientations. The transcription factor binding regions were analyzed by DNase I footprint assay with human hepatoma (HepG2) cell nuclear extracts, by using 50,000 cpm of ³²P-labeled sense or anti-sense probe. Nuclear extracts were prepared as previously described (Hansen et al., 1997) and aliquots were stored at 80°C before use. Samples were treated with 1 U of DNase I for 5 min before isolation of the ³²P-labeled DNA/magnetic bead conjugate. After heating in formamide loading buffer, samples were separated on 6% acrylamide gels. Sequencing ladders were generated with the same sense and anti-sense primers used in the synthesis of the footprinting probes and run in the same gel as footprint reactions. This enabled direct determination of the regions of sequence involved in DNA-protein interactions.

Construction and Expression of UGT2B7 Promoter Constructs. Constructs containing 5' deletions of the 800/+58 UGT2B7 promoter fragment were generated by PCR. The following oligonucleotides with KpnI sites (underlined) were used to define the 5' ends of the deletion constructs: 275 (5'- CGGGGTACCAGATCTGTCACTGCTA- CTG-3'), 44 (5'-CGGGGTACCAAGGGTTACATTTTAACTTCTTG-3') and 27 (5'-CGGGGTACCTTCTTGGCTAATTTATCTTTGG-3'). An oligonucleotide spanning +39 to +57 of UGT2B7 gene (5'-CGGGGTACCTGGTGCAATGCAATGCTTG-3') was used in PCR to define the 3' end of each of these UGT2B7 promoter deletion fragments. The region A mutant constructs were synthesized by PCR. In the case of the 44 m construct, the forward primer contained a mutated HNF1 site (5'-CGGGGTACCAAGGGTTACATTTGCCCTTCTTG-3'). For the 275 m construct, the HNF1 site was similarly altered by using complementary oligonucleotides (5'-AAGGGTTACATTTGCCCTTCTTG-3' and 5'-CA- AGAAGGGCAAATGTAACCCTT-3'), and the mutations were introduced by sequential PCR steps (Cormack, 1991). The PCR was performed with ELONGASE (Life Technologies, Rockville, MD). The oligonucleotide primers were synthesized by Bresatec (Adelaide, Australia). All fragments were digested with KpnI and subcloned into pGL3-basic vector at compatible sites (Promega, Madison, WC). DNA sequencing was carried out on all constructs to ensure that no undesired mutations had been introduced during DNA amplification by ELONGASE. The pRL-TK vector containing the Herpes simplex virus thymidine kinase gene promoter was used as an internal control in all transfection experiments. Transfections were performed as previously described (Hansen et al., 1997). Cells were harvested at 68 h post-transfection, and promoter activities in cell lysates were determined by the Dual-Luciferase Reporter Assay System (Promega) in 96-well plates with a Packard TopCount luminescense and scintillation counter.

Gel Shift Assay. Complementary oligonucleotides corresponding to the HNF1/AT-rich element in region A of the UGT2B7 gene (Fig. 2, 48 to 23, 26 bp) were synthesized by Life Technologies. These were as follows: UGT2B7 HNF1/AT (5'-TTATAAGGGTTACATTTTAACTTCTT-3', 5'-AAGAAGTTAAAATGTAACCC-TTATAA-3'). The remaining oligonucleotides containing the consensus binding sites for HNF1, Oct-1 and cAMP-response element-binding protein (CREB), were synthesized by Bresatec (Adelaide, Australia) as follows: HNF1con (5'-TCAGGTTAATCATTAACGATCT-3', 5'-AGATCGTTAATGATTAACCTGA-3'), Oct-1con (5'-TGTCGAATGCAAATCACTAGAA-3', 5'-TTCTAGTGATTTGCATTCGACA-3'), Oct-1 m with a mutated Oct-1 binding site (5'-TGTCGAATGCAAGCCACTAGAA-3', 5'-TTCTAGTGGCTTGCATTCGACA-3'), and CREB (5'-AGAGATTGCCTGACGTCAGAGAGCTAG-3', 5'-CTAGCTCTCTGACG- TCAGGCAATCTCT-3'). Double-stranded oligonucleotides were end-labeled with T4 polynucleotide kinase and [-³²P]ATP after annealing of complementary oligonucleotides. Conditions for gel shift assay were as described previously by Hansen et al. (1998) with the use of 6-µg cell nuclear extracts. For supershifts, either 1 or 2 µl of antibody was added after addition of ³²P-labeled oligonucleotide, and samples were incubated for 20 min at room temperature. Each blocking peptide to the antibody was added before addition of ³²P-labeled oligonucleotide and was preincubated with nuclear extracts for 5 min at room temperature.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Characterization of the UGT2B7 Gene And Its Proximal Promoter. Cosmid clones that hybridized to ³²P-labeled UGT2B7 cDNA were isolated from a human genomic DNA library. Restriction mapping and sequencing was performed on these clones. One genomic clone, COS-17, which contained the UGT2B7 coding region, was selected for additional analysis. Exon/intron boundaries and intron sizes were determined by PCR and sequencing to show that the UGT2B7 gene consists of 6 exons and extends over 16 kb as depicted in Fig. 1.

View larger version (7K): [in this window] [in a new window]   Fig. 1.   Organization of the human UGT2B7 gene. The position and sizes of the 6 exons (black boxes) and introns (connecting lines) are depicted. The PCR products that were generated using COS 17 as template and oligonucleotides to the ends of exons (black diamonds) are indicated. The length and partial sequences of these PCR products were used to determine intron lengths and intron/exon boundaries. HindIII restriction sites are denoted.

View larger version (70K): [in this window] [in a new window]   Fig. 2.   Nucleotide sequence of the UGT2B7 gene proximal promoter. The three footprints (regions A, B, and C) are underlined. The translation start codon is boxed. The designated transcription start site is shown as the enlarged base. The HNF1-binding site in region A is shown in italics.

View larger version (48K): [in this window] [in a new window]   Fig. 3.   RNase protection analysis of UGT2B7 RNA. Human liver poly(A+) RNA (5 µg) was incubated with 32P-labeled UGT2B7 RNA transcript generated from a recombinant plasmid containing the region 45 to +166 of the UGT2B7 gene. The sample was untreated () or treated with RNase A (+) and the products were separated by electrophoresis. A sequencing ladder generated from the corresponding region of the UGT2B7 gene was used as a size marker. The protected fragment corresponding to the position of the UGT2B7 transcription start site is indicated by the arrow.

View larger version (58K): [in this window] [in a new window]   Fig. 4.   DNase I footprint analysis of the UGT2B7 gene promoter. Incubation of 32P-labeled sense or anti-sense probe was performed in the presence of nuclear extracts from HepG2 cells (50 µg protein) with (+) and without () added DNase I. Regions resistant to DNase I digestion (regions A, B, and C) are denoted. A sequencing ladder generated with the sense primer employed in PCR of the footprinting probes is also shown. Sequence of the footprinted regions A, B, and C is denoted in Fig. 2.

Functional Analysis of UGT2B7 Gene Promoter Activity In HepG2 Cells. To determine whether the HNF1/AT-rich element in region A contributed toward UGT2B7 gene promoter activity, various lengths of the UGT2B7 gene promoter were prepared by PCR and subcloned into the promoterless pGL3-basic vector. The promoter constructs were designated 2B7 275/+57 (HNF1/AT-rich element in region A, regions B and C present), 2B7 275 m/+57 (point mutations introduced into the HNF1-like element in region A), 2B7 44/+57 (HNF1/AT-rich element in region A present), 2B7 44 m/+57 (point mutations introduced into the HNF1-like element in region A) and 2B7 27/+57 (HNF1/AT-rich element absent; Fig. 5A). The ability of these constructs to drive the firefly luciferase reporter gene was tested by transfection into human liver hepatoma HepG2 cells, which contain HNF1 and Oct-1 transcription factors. The UGT2B7 constructs containing region A were active in the HepG2 cell line (Fig. 5B). The longer UGT2B7 gene promoter (275/+57) was about 12-fold more active than the shorter promoter (44/+57). When region A was absent, as in the 2B7 27/+57 construct, or mutated, as in the 2B7 44 m/+57 and 275 m/+constructs, UGT2B7 gene promoter activity was comparable to or less than that of the promoterless control, pGL3-basic.

View larger version (20K): [in this window] [in a new window]   Fig. 5.   Relative luciferase activity of transfected UGT2B7 promoter deletion constructs. In A, UGT2B7 deletion and mutation constructs used in transfection analysis are shown diagrammatically. The location of the HNF1/AT-rich site where transcription factors bind is indicated. The point mutations introduced are shown (underlined). In B, the relative strength of the UGT2B7 promoter with and without the HNF1/AT-rich region is demonstrated. The promoterless pGL3-basic is included as a reference. In C, 2 µg of the UGT2B7-firefly luciferase constructs were cotransfected with 40 ng of Renilla luciferase reporter plasmid (internal control) in the presence of 0.2 µg of pCMV5 (control) or 0.2 µg of HNF1 and/or Oct-1 expression vectors. Relative firefly luciferase activities were calculated as the mean of triplicate assays normalized to Renilla luciferase activities with S.E. Comparisons were made with the UGT2B7 44/+57 construct. The promoter activity of the UGT2B7 44/+57 construct cotransfected with HNF1 and Oct-1 expression vectors was 18% of a firefly luciferase reporter plasmid containing the SV40 promoter and enhancer (positive control). The results from a typical experiment are shown.

HNF1 Binds to Region A of the UGT2B7 Gene Proximal Promoter. As shown above, the HNF1/AT-rich element in region A of the human UGT2B7 gene promoter binds nuclear proteins and is necessary for HNF1/Oct-1-mediated enhancement of UGT2B7 promoter activity. To determine whether this region binds HNF1 specifically, gel shift analysis was carried out using ³²P-labeled double-stranded oligonucleotide containing the HNF1/AT-rich sequence (UGT2B7 HNF1/AT) and HepG2 nuclear extracts. Nuclear proteins that bound specifically to this oligonucleotide were detected as a doublet on autoradiographs (Fig. 6, arrow). Binding of the labeled oligonucleotide to the complex was substantially reduced, with excess unlabeled UGT2B7 HNF1/AT or HNF1-consensus oligonucleotides. In contrast, the binding was not significantly reduced in the presence of Oct-1-consensus oligonucleotide (up to 50-fold molar excess) or oligonucleotide containing point mutations (Hansen et al. 1997) that interrupted the HNF1-binding site (data not shown).

View larger version (99K): [in this window] [in a new window]   Fig. 6.   Gel electrophoresis mobility shift assay of the UGT2B7-HNF1/AT element with various competitors. 32P-labeled oligonucleotide (35,000 cpm per lane) was incubated with 6 µg of HepG2 nuclear extracts followed by electrophoretic resolution on a 4% nondenaturing polyacrylamide gel. The competition studies were performed using 25- and 50-fold molar excess of unlabeled double-stranded oligonucleotides for either HNF1 consensus (HNF1con), UGT2B7-HNF1/AT, or the Oct-1 consensus sequence. The sequences of the oligonucleotides are described in Methods.

View larger version (35K): [in this window] [in a new window]   Fig. 7.   Gel electrophoresis mobility and supershift assay of the UGT2B7 HNF1/AT element with HepG2 nuclear extracts. 32P-labeled UGT2B7-HNF1/AT oligonucleotide (50,000 cpm per lane) was incubated with 6 µg of nuclear extract followed by electrophoretic resolution on a 3.5% nondenaturing polyacrylamide gel. Supershifts were performed in the presence of anti-HNF1 (2 µg) or anti-HNF1 (1 µg) anti-peptide antibody. The competition studies were performed with specific blocking peptide (2 µg) for each antibody.

Oct-1 Is Part of the Protein Complex That Binds to Region A in the UGT2B7 Gene Proximal Promoter. As demonstrated by the functional assays described previously, Oct-1 only activates the UGT2B7 promoter in the presence of HNF1. To determine whether Oct-1 is part of the protein complex that binds to region A, supershift analysis with specific antibody to Oct-1 was performed. Specificity of the Oct-1 antibody was first demonstrated by its capacity to recognize the protein binding to an oligonucleotide containing a consensus Oct-1 binding site and its inability to recognize protein that binds to an unrelated site, a consensus cAMP response element binding site (Fig. 8). This antibody caused a supershift in part of the protein complex that binds to radiolabeled UGT2B7 HNF1/AT oligonucleotide (Fig. 9, A and B). The supershift was abolished in the presence of peptide that was used to generate the Oct-1-specific antibody but was not affected by peptide to the related factor, Oct-2 (Fig. 9A) or by HNF1 peptide (Fig. 9B). As shown in Fig. 6, the oligonucleotide containing the Oct-1 consensus binding site does not compete with binding of nuclear proteins, including HNF1 and Oct-1, to the UGT2B7 HNF1/AT rich site. This suggests that Oct-1 does not bind directly to this site. This was further confirmed by gel shift analyses with LNCaP cells, which do not contain HNF1 (Fig. 10). Nuclear extracts from LNCaP cells did not bind to the UGT2B7 HNF1/Oct-1 site (Fig. 10, lanes 4-6), even though Oct-1 was present in these cells, as shown by the capacity of the Oct-1 consensus oligonucleotide to bind a protein recognized by the Oct-1 specific antibody (Fig. 10, lanes 9 and 10). However, as described above, nuclear proteins containing HNF1 and Oct-1 in HepG2 cells bound to the UGT2B7 HNF1/AT-rich site (Fig. 10, lanes 1-3). These data indicate that Oct-1 stimulates the activity of the UGT2B7 proximal promoter via interactions with HNF1 or other proteins in the complex rather than by direct binding to DNA.

View larger version (27K): [in this window] [in a new window]   Fig. 8.   Specificity of Oct-1 antibody in supershift assays. 32P-labeled Oct-1 or CREB consensus oligonucleotide (35,000 cpm per lane) was incubated with 6 µg of HepG2 nuclear extract followed by electrophoretic resolution on a 3.5% nondenaturing polyacrylamide gel. Cold competitor oligonucleotide was included at 50-fold molar excess. Anti-Oct-1 (1 µg) antibody was added where indicated.

View larger version (43K): [in this window] [in a new window]   Fig. 9.   Specificity of the interaction of Oct-1 with the UGT2B7 HNF1/AT element. 32P-labeled UGT2B7 HNF1/AT oligonucleotide (35,000 cpm per lane) was incubated with 6 µg of HepG2 nuclear extracts followed by electrophoretic resolution on a 4% nondenaturing polyacrylamide gel. In A, the competitor oligonucleotide, UGT2B7-HNF/AT, was added as shown. The remaining reactions contained 1 µg of anti-Oct-1 or peptide competitor as indicated. In B, anti-Oct-1, alone or in combination with Oct-1 or HNF1 competitor peptides, was included in gel shift reactions.

View larger version (30K): [in this window] [in a new window]   Fig. 10.   Binding of HNF1 and Oct-1 to the UGT2B7 HNF1/AT element in liver versus prostate cell extracts. Either HepG2 (6 µg) or LNCaP (4 µg) nuclear extracts were incubated with 100,000 cpm of the indicated 32P-labeled oligonucleotides containing the UGT2B7 HNF1/AT-rich region or the Oct-1 consensus binding site. Antibodies to either HNF1 (2 µg) or Oct-1 (1 µg) were included, as shown, for both extracts.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

In this paper we have characterized the UGT2B7 gene and shown that its proximal promoter is activated by the liver-enriched factor HNF1 and the ubiquitous factor, Oct-1. The UGT2B7 gene consists of 6 exons and extends over 16 kb. It is thus similar in arrangement to other UGT2B genes including UGT2B1, UGT2B2, UGT2B4, and UGT2B17 (Mackenzie and Rodbourn, 1990; Haque et al., 1991; Beaulieu et al., 1997; Monaghan et al., 1997). However, in contrast to these genes, which have a TATA box approximately 30 bp upstream from the transcription start site, UGT2B7 does not have a canonical TATA box or other sequences that are known to specify the site of transcription initiation, such as the initiator element, a 17-bp element encompassing the transcription initiation site. We have, however, identified a region (region A) adjacent to the transcription start site that binds nuclear proteins and is most likely responsible for establishing the preinitiation complex. The 5'-end of region A contains an HNF1/AT-rich site. The position of this site just 27 bp upstream from the transcription start site is unusual, as genes containing a functional HNF1 site have the site positioned at least 40 bp or more from the transcription start site. These genes include the UGT2B1 gene (Mackenzie and Rodbourn, 1990; Hansen et al., 1997); the bilirubin UDP glucuronosyltransferase gene, UGT1A1 (Ueyama et al., 1997); the albumin, -fetoprotein, and transthyretin genes (Courtois et al., 1988); the high affinity Na⁺/glucose cotransporter (Rhoads et al., 1998); CYP2C1, CYP2C2 (Kim and Kemper, 1991), CYP2E1 (Liu and Gonzalez, 1995), and CYP3A2 genes (Legraverand et al., 1994); and the mouse guanylin precursor gene (Sciaki et al., 1994). The unusual position of the HNF/AT-rich site so close to the transcription start site suggests that it may be important in the recruitment of the general transcription machinery to the promoter. Indeed, if this site is removed or mutated, UGT2B7 promoter activity is lowered significantly. This is particularly evident when the longer promoter construct (275/+57) is used in transfection experiments. The longer promoter is about 12-fold more active than the shorter promoter (44/+57) and contains two other regions that bind transcription factors (regions B and C). These factors most probably contribute to the increased constitutive activity of the UGT2B1 promoter. Nevertheless, the activity of this longer promoter is all but abolished when the HNF/AT-rich site (region A) is mutated.

We have shown that the liver-enriched transcription factor, HNF1, binds to the HNF1/AT-rich site and activates UGT2B7 promoter activity. Although this HNF1-mediated activation is enhanced by Oct-1, Oct-1 alone does not activate the promoter or bind directly to the HNF1/AT-rich site. As supershift analysis demonstrates that Oct-1 is part of the complex that binds to the HNF1/AT-rich site, it is logical to assume that Oct-1 enhances promoter activity by directly interacting with HNF1 rather than binding to the DNA. Oct-1 is known to bind to the basal transcription factor TFIIB (Nakshatri et al., 1995) and hence may act as the conduit for recruiting the preinitiation complex to the transcription start site. Although many genes are regulated by HNF1 and Oct-1, there is only one report demonstrating the interaction of these two factors in gene regulation. The interaction between HNF1 and Oct-1 occurred when their respective binding sites were adjacent to each other and was required for activation of the hepatitis B virus preS1 promoter (Zhou and Yen, 1991). The binding of HNF1 and Oct-1 in a complex to a single DNA binding site, as observed for the UGT2B7 proximal promoter, has not been described before.

HNF1 and Oct-1 are both homeodomain proteins of the Helix-Turn-Helix superclass of transcription factors. HNF1 binds as a homodimer or as a heterodimer with HNF1 to regulate gene transcription in the liver, kidney, and intestine (Frain et al., 1989; Rey-Campos et al., 1991; Cereghini et al., 1998). This binding may be stabilized via the direct interaction of a dimer of DCoH (dimerization cofactor of HNF1) with the HNF dimer to form a tetramer. The dimer of DCoH does not contact the DNA directly, but promotes the interaction of HNF1 with suboptimal DNA target sequences such as the ₁-antitrypsin gene TATA box region (Rhee et al., 1997). Oct-1 may have a role similar to that of DCoH as a coactivator by stabilizing the interaction of HNF1 with the UGT2B7 promoter and enhancing HNF1-mediated transcription activation. Oct-1 has a bipartite DNA binding domain called the POU domain, which can bind as a monomer to DNA (Herr and Cleary, 1995). In contrast to HNF1, Oct-1 can interact with several other factors including Oct-2 and Pit-1 (Herr and Cleary, 1995), the glucocorticoid receptor (Chandran et al., 1999), OBF-1 (Sauter and Matthias, 1998), MAT1 (Inamoto et al., 1997), VP16 (Babb et al., 1997), nuclear matrix proteins (Kim et al., 1996), Sp1 (Strom et al., 1996), NF1 (O'Connor and Bernard, 1995), the Vitamin D-3 Receptor (Liu and Freedman, 1994), and AP-1 (Pfeuffer et al., 1994) to regulate gene transcription. All these interactions involve the binding of Oct-1 to its cognate octamer binding site on DNA and do not depend on the recruitment of Oct-1 to other sites via an interaction with the factor bound at those sites, as is proposed in the regulation of the UGT2B7 gene promoter.

In summary, we have shown that a region adjacent to the UGT2B7 transcription start site is necessary for promoter activity and that HNF1 binds to this region to activate the UGT2B7 proximal promoter. Furthermore, we show that Oct-1 acts as a coactivator to enhance the transcriptional activity of the UGT2B7 gene proximal promoter by HNF1. This interaction between HNF1 and Oct-1 may fine-tune the expression of UGT2B7. Subsequent experiments will be directed toward identifying factors that bind to regions upstream of region A, (e.g., regions B and C) in the UGT2B7 promoter to obtain a more complete model of how this gene is regulated. However, it is apparent that these upstream activating factors that bind to regions B and C still require a functional interaction between HNF1 and region A for full activity. An understanding of the transcriptional regulation of the UGT2B7 gene should provide new information and strategies to elucidate the basis of potential polymorphic variations in the expression of UGT2B7.