![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Molecular and Cellular Biology, January 2003, p. 437-449, Vol. 23, No. 2
0270-7306/03/$08.00+0 DOI: 10.1128/MCB.23.2.437-449.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Department of Molecular Genetics, College of Medicine, University of Illinois at Chicago, Chicago, Illinois 60607
Received 30 May 2002/ Returned for modification 6 August 2002/ Accepted 18 October 2002
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) and FoxA2 (HNF-3ß) proteins share 93% homology in the winged-helix motif, bind to similar target DNA sequences (23, 34, 44), and potentially regulate transcription of numerous genes critical for liver function (8). The FoxA1 and FoxA2 proteins also regulate transcription of epithelial cell genes critical for function of the lung and pancreas (9, 19, 38). The FoxA proteins possess homology in the N-terminal and C-terminal transcriptional activation domains (35, 39), the latter of which contains the functionally important region II and III sequences that are conserved in the Fox protein family (9). Interestingly, the FoxA2 region II sequences have been shown to associate with the human homolog of the Drosophila Groucho transcriptional repressors (transducin-like Enhancer of split), which suggests the possibility that FoxA proteins may also function to inhibit transcription in cell types expressing Groucho proteins (51).
Hepatocyte nuclear factor 6 (HNF-6), belonging to the family of One Cut transcription factors (also known as OC-1), possesses a C-terminal DNA binding domain composed of a single Cut motif and a divergent Homeodomain (24, 25, 28, 40). The HNF-6 protein possesses an N-terminal transcriptional activation domain, and the Cut-Homeodomain sequences were shown to interact with the CREB-binding protein (CBP) in vitro through the LXXLL motif within the Cut domain and the M and F amino acid residues within the Homeodomain (26). In cotransfection assays, transcriptional activation of a 6X HNF6-TATA luciferase reporter gene was stimulated by the intact HNF-6 protein and a four- to eightfold excess of CBP expression vector (26). These studies suggested that full-length HNF-6 protein may potentially recruit the p300/CBP coactivator proteins to stimulate transcription through an HNF-6 binding site. Interestingly, FoxA2 and HNF-6 proteins potentially regulate expression of similar target genes critical for hepatocyte function (8, 44).
In previous studies we used transgenic mice or recombinant adenovirus infection to increase hepatic expression of FoxA2, which caused diminished hepatocyte glycogen levels and reduced expression of glucose homeostasis genes and was associated with decreased HNF-6 levels (42, 50). Furthermore, we demonstrated that mouse tail vein injections of AdFoxA2 and AdHNF-6 together served to increase hepatic glycogen levels and expression of glucose transporter Glut-2 (48).
We therefore conducted the present study to determine whether the HNF-6 protein would potentiate FoxA2 transcriptional activation in cotransfection assays. We show that through association between the Cut-Homeodomain and winged-helix DNA binding domains, the HNF-6 Cut-Homeodomain functioned as a coactivator protein for FoxA2 transcriptional activation by recruiting the p300/CBP coactivator proteins. This was supported by the fact that FoxA2 transcriptional synergy with HNF-6 was inhibited by mutation of the Cut domain LXXLL motif (HNF-6 L350A), which was unable to recruit the p300/CBP proteins (26). Interestingly, HNF-6 was unable to stimulate either FoxA1 or FoxA3 transcriptional activity in cotransfection assays or efficiently bind to the FoxA1 winged-helix motif, indicating that HNF-6 specifically interacted with the FoxA2 protein. Moreover, with the HNF-6-regulated Glut-2 promoter, we observed that FoxA2 interaction prevented HNF-6 from binding to its DNA recognition sequence, thereby significantly diminishing HNF-6 transcriptional activation. These data suggest that at a FoxA-specific site, HNF-6 serves as a coactivator protein to enhance FoxA2 transcription, whereas at an HNF-6-specific site, FoxA2 represses HNF-6 transcription by inhibiting its DNA binding activity.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The 6X HNF-6 TATA-luciferase reporter construct consisted of six copies of the HNF-6 binding site derived from the FoxA2 promoter (-141 to -127 bp) driving expression of the TATA box luciferase gene and was described previously (40). The 6X FoxA TATA-luciferase utilized six copies of the FoxA2 binding site (TTTGTTTGTTTG) from the cdx-2 promoter region, and this sequence was selective for FoxA2 binding (44, 53).
The glutathione S-transferase (GST) fusion protein with the FoxA2 (amino acids 144 to 279) and FoxA1 (amino acids 134 to 290) winged-helix DNA binding domains were described previously (34). GST fusion proteins with the FoxA1 and FoxA2 N-terminal regions (amino acids 7 to 103 and 7 to 94) and FoxA2 C-terminal region (amino acid 361 to 458) were described previously (17, 35).
Adenovirus vectors that used the CMV promoter to drive expression of either the mouse HNF-6 (AdHNF6) or rat FoxA2 (AdFoxA2; AdHNF3ß) cDNA were described previously (48-50). For all the green fluorescent protein (GFP)-HNF-6 fusion proteins, the HNF-6 fragments were PCR generated, digested with EcoRI and BamHI, and then ligated into the appropriate GFP expression vector, except for the HNF-6 N-terminal fragment which was digested with EcoRI and XbaI. The PCR-amplified HNF-6 fragments were verified by DNA sequencing.
The following sense and antisense primers were used for PCR amplification of the designated HNF-6 coding regions: HNF-6 FL (1 to 465), 5'-CGCGAATTCATGAACGCACAGCTGACC-3' and 5'-CGCGGATCCTGCTTTGGTACAAGTGCT-3'; HNF-6 Cut-Homeo (289 to 465), 5'-CGCGAATTCATGGAAGAGATCAATACC-3' and 5'-CGCGGATCCTGCTTTGGTACAAGTGCT-3'; HNF-6 N-Term (1 to 288), 5'-CGCGAATTCATGAACGCACAGCTGACC-3' and 5'-CGCTCTAGACTGCCCTGAATTACTTCC-3'; GFP HNF-6 
Homeo (1 to 384), 5'-CGCGAATTCATGAACGCACAGCTGACC-3' and 5'-CGCGGATCCGGTGTTGCCTCTGTCCTT-3'; HNF-6 Cut (289 to 384), 5'-CGCGAATTCATGGAAGAGATCAATACC-3' and 5'-CGCGGATCCGGTGTTGCCTCTGTCCTT-3'; and HNF-6 Homeo (385 to 465), 5'-CGCGAATTCCCCAAAAAGCCCAGGCTG-3' and 5'-CGCGGATCCTGCTTTGGTACAAGTGCT-3'.
Cell culture and transient transfection. HepG2 cells were maintained in F12 medium supplemented with 10% fetal bovine serum, essential amino acids, penicillin-streptomycin, and insulin as previously described (44). For transient transfection, HepG2 cells were plated in six-well plates and transfected with Fugene 6 reagent (Roche) according to the manufacturer's protocol. Cells were transfected with 250 ng of CMV-HNF6 or deletion mutants, CMV-FoxA1, or -FoxA2, or FoxA3 with 1.6 µg of a 6X FoxA TATA-luciferase reporter with 30 ng of CMV-Renilla internal control.
We also performed these cotransfection experiments with either the -134 FoxA2 promoter luciferase construct (36) or the -188 Glut2 promoter luciferase construct (48). Total transfected DNA was kept constant with empty CMV vector. Twenty-four hours posttransfection, cells were prepared for dual luciferase assays (Promega). Luciferase activity was determined as fold induction over cells transfected with empty CMV vector, normalized to Renilla activity. Experiments were performed at least two to six times in triplicate unless otherwise stated. Statistical analysis was performed with Microsoft Excel tools.
Immunofluorescence. HepG2 cells were plated on untreated glass coverslips and transfected with 500 ng of full-length GFP HNF-6 or mutant constructs with the Fugene 6 (Roche) reagent according to the manufacturer's protocol. Twenty-four hours posttransfection, the coverslips were washed three times in phosphate-buffered saline (PBS), and the cells were fixed in 3% paraformaldehyde for 15 min at room temperature. After several rinses in PBS, the coverslips were mounted on glass slides with mounting medium (15% [wt/vol] Vinol 205 polyvinyl alcohol, 33% [vol/vol] glycerol, 0.1% azide in PBS, pH 8.5). Cellular fluorescence was visualized with a Zeiss Axioplan microscope.
In vitro pulldown assay. Production of GST fusion proteins was described previously (34). For the binding reaction, 15 µl of 35S-labeled in vitro transcribed and translated protein was incubated with 20 µl of glutathione-Sepharose beads (preadsorbed with 2.5 µg of various GST-FoxA fusion proteins) in 500 µl of binding buffer for 2 h at 4°C with constant rotation. After 2 h of binding, the beads were washed three times with 500 µl of wash buffer. The pelleted beads were then resuspended with sodium dodecyl sulfate (SDS) sample buffer plus 1% ß-mercaptoethanol and boiled for 5 min, after which the supernatant was loaded on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and processed for autoradiography. The binding buffer consisted of 85 mM KCl, 20 mM HEPES, pH 7.9, 1.5 mM MgCl2, 0.2 mM EDTA, 0.5 mM dithiothreitol, 20% glycerol, 0.1% NP-40 detergent, 0.2 mM phenylmethylsulfonyl fluoride, and protease inhibitors (Roche). The wash buffer was identical to the binding buffer except that it contained 150 mM KCl.
Western and coimmunoprecipitation with transfected HepG2 nuclear extracts.
HepG2 cells were cotransfected with 5 µg of full-length GFP-HNF6 (1 to 465) or GFP-Homeo HNF-6 (1 to 384) with pcDNA3.1 FoxA1 or FoxA2 tagged with a C-terminal V5 epitope or CMV-CBP. Twenty-four hours posttransfection, cells were washed twice with cold PBS, collected, and prepared for nuclear extracts. Nuclear extracts were prepared as described with slight modifications (36, 45). Briefly, cells were resuspended with hypotonic buffer (10 mM HEPES, pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM dithiothreitol, and 0.2 mM phenylmethylsulfonyl fluoride supplemented with protease complete inhibitor tablets), incubated on ice for 20 min, and then ruptured by 30 passes in a Dounce homogenizer.
Nuclear pellets were washed with hypotonic buffer, and nuclear proteins were extracted by addition of high-salt buffer (hypotonic buffer containing 20% glycerol, 0.5 M KCl, 0.2 mM EDTA, and 0.1% NP-40) and incubated for 40 min at 4°C with occasional agitation. After centrifugation, the Bradford reagent (Bio-Rad) was used to determine protein concentrations of the nuclear extracts. For immunoprecipitation, 850 µg of nuclear extracts was diluted to a final salt concentration of 85 mM KCl and allowed to equilibrate for 45 min. The extracts were first cleared with protein A-Sepharose and then incubated with 1 µl of either V5 (Invitrogen) or CBP (C-1; Santa Cruz Biotech) monoclonal antibodies for 16 h at 4°C. The extracts were then immunoprecipitated by the addition of protein A-Sepharose. Beads were washed extensively with wash buffer (hypotonic buffer containing 150 mM KCl). The immunoprecipitates were eluted by the addition of SDS sample buffer and boiling for 5 min at 100°C. Eluted proteins were resolved by SDS-PAGE and transferred to nitrocellulose membranes for Western blot analysis. For Western blot analysis, a monoclonal GFP antibody (Clontech) was incubated at a dilution of 1:1,000 overnight at 4°C.
Adenovirus infection of HepG2 cells and electrophoretic mobility shift assays. HepG2 cells were infected with a constant amount of adenovirus expressing the mouse HNF-6 cDNA (AdHNF-6; 10 PFU per cell) with increasing amounts of AdFoxA2 (AdHNF-3ß; 2, 10, or 50 PFU per cell) as described previously (48-50). HepG2 cells were also infected with a constant amount of AdFoxA2 (10 PFU per cell) with increasing amounts of AdHNF-6 (0.2-, 1-, or 5-fold excess). The AdLacZ adenovirus (CMV-LacZ) was used to keep the total amount of infecting adenovirus at 60 PFU per cell and used to eliminate secondary competition between CMV promoters. Nuclear extracts were prepared 24 h following infection and used for electrophoretic mobility shift assays with the HNF-6 DNA binding site from the Glut-2 promoter (48) or the FoxA DNA binding site from the cdx2 promoter as described previously (37, 53). We also included control lanes containing a 100-fold excess of unlabeled DNA competitor and HepG2 nuclear extracts infected with either AdHNF-6 or AdFoxA2. Western blot analysis of AdFoxA2 and AdHNF-6 HepG2 nuclear extracts with HNF-6 antibody was performed as described previously (42, 52).
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Consistent with this finding, in cotransfection assays with the 6X FoxA-TATA luciferase plasmid, the CMV-FoxA2 cDNA expression vector elicited transcriptional activation of the reporter gene, whereas cotransfection of the CMV-HNF-6 plasmid did not activate this promoter (Fig. 1B). Interestingly, combining the FoxA2 and HNF-6 expression vectors in HepG2 transient transcription assays caused a statistically significant increase in expression of the 6X FoxA-TATA luciferase gene (Fig. 1B, P = 0.0003). In contrast, this HNF-6 transcriptional synergy was not found with either the related FoxA1 (HNF-3) or FoxA3 (HNF-3
) expression vector (Fig. 1A). Interestingly, the HNF-6 Cut-Homeodomain (Fig. 1A, 289 to 465) sequences alone were sufficient to stimulate FoxA2 transcriptional activity to levels similar to full-length HNF-6 protein (Fig. 1B).
|
Homeo 1 to 384) abrogated stimulation of FoxA2 transcription, and the HNF-6 Homeo protein functioned as a dominant negative inhibitor of the HNF-6 transcriptional synergy with FoxA2 in cotransfection assays (Fig. 1B). Moreover, when an HNF-6-dependent reporter (6X HNF-6 TATA luciferase) was used, the HNF-6 Homeo mutant exhibited a 75% reduction in transcriptional activation of the reporter gene and the Cut-Homeodomain sequences were transcriptionally inactive compared to full-length HNF-6 (Fig. 1C). These results suggested that the HNF-6 Homeodomain sequences were essential for potentiating transcriptional activity of both a FoxA-dependent and an HNF-6-dependent reporter gene. Furthermore, although the HNF-6 Cut-Homeodomain sequences mediated recruitment of the p300/CBP coactivator proteins (26, 40), our data suggested that they were unable to do so when bound to DNA. These data suggested that HNF-6 Cut-Homeodomain mediated stimulation of FoxA2-dependent transcription through protein-protein association. To further assess the effects of FoxA2 and HNF-6 in the context of an intact promoter, we performed transcription assays with the -134 bp FoxA2 promoter luciferase reporter gene, which possesses a FoxA binding site but lacks a high-affinity HNF-6 binding site (36, 40, 44). Transfection with CMV HNF-6 mildly stimulated expression of the FoxA2 promoter (Fig. 1D), presumably through a potential weak-affinity HNF-6 DNA binding site found at -64 to -52 bp (ACctTgGATTTAA). In spite of the fact that the CMV-FoxA2 expression vector alone was unable to elicit strong transcriptional activation of the FoxA2 promoter, coexpression of both FoxA2 and HNF-6 proteins provided synergistic transcriptional activation of the -134 bp FoxA2 promoter region (Fig. 1D). Consistent with HNF-6 mediating synergistic transcriptional activity with FoxA2, the HNF-6 Cut-Homeodomain protein alone was unable to stimulate expression of the -134 bp FoxA2 promoter, but when the HNF-6 Cut-Homeodomain and FoxA2 expression vectors were combined, they provided transcriptional synergy of the FoxA2 promoter (Fig. 1D). These data suggested that HNF-6 mediated stimulation of FoxA2 transcriptional activation through FoxA specific DNA sites within an intact promoter region.
To determine whether the nuclear localization sequences were contained in the Cut-Homeodomain, we used mammalian expression vectors in which the GFP was fused to the N terminus of various HNF-6 protein regions (Fig. 2A). Transfection of HepG2 cells with either GFP HNF-6 full-length (HNF6 FL) or GFP-HNF-6 Cut-Homeodomain (HNF6 Cut-Homeo) fusion protein showed equivalent nuclear fluorescence (Fig. 2B to C). The GFP HNF-6 N-terminal fusion protein (HNF6 N-Term) displayed cytoplasmic fluorescence in HepG2 cell transfections (Fig. 2D), suggesting that the nuclear localization sequences were contained within the HNF-6 Cut-Homeodomain motif. Furthermore, predominant nuclear fluorescence was found with GFP fusion proteins that contained either the HNF-6 Cut-domain (Fig. 2E and F; HNF6 Homeo and HNF6 Cut) or the HNF-6 Homeodomain (Fig. 2G, HNF6 Homeo) in HepG2 cell transfections. These data suggested that the inability of the HNF6 Homeo expression construct to synergistically potentiate FoxA2 transcriptional activity in cotransfection assays was not due to a problem with nuclear localization.
|
|
We next prepared nuclear extracts from HepG2 cells cotransfected with expression vectors containing the GFP protein fused to either HNF-6 FL (1 to 465) or HNF-6 Homeo (1 to 384) (see Fig. 1A) and a V5 epitope-tagged FoxA2 or FoxA1 cDNA. These nuclear extracts were subjected to immunoprecipitation with a V5 monoclonal antibody followed by Western blot analysis with a monoclonal GFP antibody (Fig. 4A). All of the transfected GFP-HNF-6 fusion proteins were expressed as determined by Western blot analysis of the transfected HepG2 nuclear extracts with the GFP monoclonal antibody (Fig. 4A, left panel). Both GFP-HNF-6 full-length and GFP HNF-6 Homeo proteins were able to coimmunoprecipate with the FoxA2 protein (Fig. 4A, right panel), suggesting that the HNF-6 Homeodomain sequence was dispensable for FoxA2 protein association. Interestingly, we were unable to coimmunoprecipitate FoxA1 and GFP full-length HNF-6 from transfected HepG2 nuclear extracts (Fig. 4A, right panel), a finding consistent with the inability of HNF-6 to enhance FoxA1 transcriptional activity in cotransfection assays (Fig. 2A). These data demonstrate that the HNF-6 protein associates with FoxA2 but not FoxA1 in nuclear extracts prepared from transfected HepG2 cells.
|
Homeo (1 to 384) and the CMV-CBP expression vector. These nuclear extracts were subjected to immunoprecipitation with CBP antibody followed by Western blot analysis with a monoclonal GFP antibody (Fig. 4B). Only the GFP-HNF-6 full-length protein was able to coimmunoprecipate with the CBP protein (Fig. 4B), suggesting that the HNF-6 Homeodomain was required for interaction with the CBP protein, a finding consistent with the inability of HNF-6 Homeo protein to synergistically stimulate FoxA2 transcription. Recruitment of the p300/CBP proteins is required for HNF-6 transcriptional synergy with FoxA2. Because we found that the HNF-6 protein could immunoprecipitate with the CBP protein from HepG2 nuclear extracts, we next examined whether HNF-6 mediates FoxA2 transcriptional activation by recruiting the p300/CBP coactivator proteins. To test this hypothesis, we cotransfected the 6X FoxA TATA luciferase reporter plasmid with the FoxA2 and mutant HNF-6 L350A expression constructs, the latter of which can interact with FoxA2 (Fig. 3C) but can no longer recruit the p300/CEBP protein (26). These experiments demonstrated that the mutant HNF-6 L350A was unable to stimulate FoxA2 transcription and functioned as a dominant negative protein to inhibit FoxA2 and HNF-6 transcriptional synergy (Fig. 5A).
|
These data confirm that the HNF-6 protein functions to recruit the p300/CBP coactivators, which mediate synergistic activation of a FoxA-dependent promoter expression. Interestingly, neither E1A nor CBP HAT (-) expression vectors inhibited HNF-6 mediated activation of the 6X FoxA-TATA luciferase reporter gene (Fig. 5B). These results suggest that when HNF-6 protein binds to its target DNA sequence, the HNF-6 Cut-Homeodomain sequences are unable to recruit the p300/CBP protein and HNF-6 therefore mediates transcriptional activation through its N-terminal domain (26).
FoxA2 inhibits HNF-6 stimulation of the Glut-2 promoter by inhibiting HNF-6 DNA binding. Previous cotransfection and DNA binding studies demonstrated that the Glut-2 promoter is transcriptionally activated by HNF-6 but not by the FoxA (HNF-3) proteins (48). Consistent with these studies, HNF-6 cotransfection stimulated expression of the -188 Glut-2 promoter, whereas FoxA2 did not (Fig. 6). Interestingly, cotransfection of equal amounts of FoxA2 and HNF-6 expression vectors caused a significant reduction in HNF-6 activation of the Glut-2 promoter (Fig. 6).
|
|
Taken together, our studies suggested that association between FoxA2 and HNF-6 DNA binding domains disrupted HNF-6 DNA binding, thereby inhibiting transcriptional activation of its target genes. Conversely, FoxA2 protein association with the HNF-6 protein synergistically activated FoxA2 transcription through HNF-6 mediated recruitment of the p300/CBP coactivator proteins.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We propose a model in which, depending on the target sequence, FoxA2 and HNF-6 protein interaction either synergistically stimulates or represses transcription (Fig. 8). On a FoxA-dependent promoter, HNF-6 functions as a coactivator protein to recruit the p300/CBP proteins to the promoter, thereby synergistically activating FoxA2-dependent transcription. An alternative possibility that cannot be ruled out is that both the FoxA2 and HNF-6 proteins are required for recruitment of the p300/CBP coactivator proteins. The p300/CBP proteins possess histone acetyltransferase activity that enhances transcription by acetylating the lysine residues of histone proteins, thereby diminishing their association with DNA and through p300/CBP protein interaction with the general transcriptional machinery (2, 3, 32, 33, 43). Conversely, on the HNF-6-specific glut-2 promoter, FoxA2 interaction repressed HNF-6 transcriptional activation by diminishing the ability of HNF-6 to bind its DNA target sequence (Fig. 8).
|
In the context of the intact HNF-6 protein, transcriptional activation involves the N-terminal STP box domain and the C-terminal Cut-Homeodomain, the latter of which has been shown to interact with the p300/CBP histone acetyltransferase proteins in vitro with GST-CBP pulldown assays (26). The p300/CBP interaction depends on two motifs, the LXXLL motif found in the Cut domain and two amino acid residues in the Homeodomain (26). In our current study, we found that in HepG2 cotransfection assays, the HNF-6 Cut-Homeodomain sequences were sufficient to synergistically stimulate FoxA2 transcriptional activity and that recruitment of the p300/CBP was essential for this transcriptional synergy. This was supported by the fact that transcriptional synergy between FoxA2 and HNF-6 was inhibited by mutation of the Cut domain LXXLL motif (HNF-6 L350A), which is essential for recruiting the p300/CBP proteins (26). Furthermore, FoxA2 and HNF-6 transcriptional synergy was abrogated by inhibition of p300/CBP activity by either coexpression of either the E1A protein or a CBP HAT (-) dominant negative protein.
These results suggest that FoxA2 and HNF-6 protein association was conducive for the recruitment of the p300/CBP proteins. However, we showed that the HNF-6 Cut-Homeodomain domain alone was unable to activate the 6X HNF-6 TATA luciferase reporter gene, suggesting that the HNF-6 Cut-Homeodomain sequences alone were unable to recruit p300/CBP proteins when they were binding to DNA. Furthermore, inhibition of p300/CBP activity by E1A protein or CBP HAT (-) did not influence the ability of full-length HNF-6 to stimulate transcription of the 6X HNF-6 TATA luciferase reporter gene. This result suggested that HNF-6 protein binding to DNA precludes the Cut-Homeodomain ability to recruit the p300/CBP coactivator proteins, suggesting that HNF-6 mediates transcriptional activation through its N-terminal domain when bound to its target sequence (26).
Our current studies suggest that this stimulation of HNF-6 transcriptional activation may involve a mechanism other than recruitment of CBP by the Cut-Homeodomain sequences when HNF-6 protein is bound to its DNA binding site. Moreover, we found that the highest HNF-6 transcriptional synergy with FoxA2 occurred in low-passage HepG2 cells and that the fold induction diminished with dedifferentiation as a result of successive passage number (data not shown), suggesting reduced levels of an additional factor required for this transcriptional synergy.
Our coimmunoprecipitation experiments indicated that the HNF-6 Cut domain was sufficient for interaction with the FoxA2 protein, but that retention of both the HNF-6 Cut and Homeodomain was essential for efficient interaction with the CBP protein in vivo. Furthermore, the HNF-6 protein lacking the Homeodomain motif functioned as a dominant negative inhibitor of FoxA2 transcriptional synergy with HNF-6. These findings are consistent with cotransfection assays demonstrating that the HNF-6 Cut and Homeodomain sequences were indispensable for synergistically stimulating FoxA2 transcriptional activation. It is interesting that many of the strong-affinity FoxA binding sites also bound HNF-6 with weak affinity, suggesting the hypothesis that these dual binding sequences may enhance recruitment of HNF-6 protein and its association with FoxA2 (8, 44).
Consistent with this hypothesis, mutation of the transthyretin FoxA/HNF-6 binding site to a sequence that only binds FoxA (HNF-3) protein diminished expression of the transthyretin promoter in HepG2 cell transfection assays (44). Furthermore, we showed that HNF-6 protein was required for FoxA2 to efficiently stimulate transcription of the FoxA2 promoter through its autoregulatory binding site (36). This suggests that expression of HNF-6 during early liver development (24, 31, 40) plays an important role in maintaining hepatocyte transcription of the FoxA2 promoter. Numerous promoter studies have indicated that collaboration between multiple hepatic transcription factors maintains hepatocyte-specific transcription (1, 6, 7, 10, 11, 13, 16, 27 to 30, 36, 41, 44, 45), but the mechanisms involved in this transcriptional synergy remain unknown. Our study suggests that one mechanism for transcriptional synergy involves FoxA2 association with HNF-6 protein, with HNF-6 functioning as a coactivator protein to recruit the p300/CBP histone acetyltransferase proteins to promoters containing a FoxA DNA binding site.
We demonstrated that FoxA2 interaction with HNF-6 inhibits binding activity and transcriptional activation of HNF-6 target genes. This implies that the ratio of active HNF-6 and FoxA2 proteins alters expression levels of HNF-6 target genes. For example, increasing FoxA2 protein in vivo diminished expression of the HNF-6 target genes glut-2 and HNF-4 (15, 42, 50). This appears likely to occur through diminished HNF-6 protein levels as well as through inhibition of HNF-6 DNA binding activity. Consistent with this mechanism, the adenovirus-mediated increase of both FoxA2 and HNF-6 caused elevated hepatic levels of Glut-2 and HNF-4 and was associated with increase in hepatocyte glycogen storage (48). Increased hepatic levels of FoxA2 also diminished expression of known FoxA2 target genes, which coincided with diminished HNF-6 protein levels (42, 50).
One potential mechanism for diminished expression of FoxA2 target genes involves diminished HNF-6 coactivator levels, which may facilitate FoxA2 C-terminal interaction with Groucho repressor proteins (51), thereby causing repression of FoxA2 target genes. This is supported by the fact that mouse liver cDNA array studies have demonstrated that expression of the mouse Drosophila Groucho homolog Enhancer of Split Protein 1 (ESP1) is detectable in postnatal mouse liver (50). Furthermore, protein kinase A has been shown to phosphorylate the HNF-6 carboxyl terminus and mediate transcriptional induction of the glucose 6-phospatase promoter through an HNF-6 DNA binding site (47). It is tempting to speculate that phosphorylation of the HNF-6 Cut-Homeodomain sequences may alter the interaction with the FoxA2 winged-helix domain and thus regulate HNF-6 coactivator protein function.
Although FoxA2, FoxA3, and FoxA1 share strong homology in the winged-helix DNA binding domain, FoxA1 interacted only weakly with HNF-6 homeodomain in GST pulldown assays and was unable to interact at all with HNF-6 in coimmunoprecipitation assays with transfected nuclear extracts. This result is consistent with the fact that the HNF-6 protein failed to exhibit transcriptional synergy with either FoxA1 or FoxA3 in HepG2 cell cotransfection assays. Although these FoxA proteins share related DNA target genes, they differ in ability to activate transcription because only the transcriptional activity of FoxA2 is stimulated by HNF-6 protein interactions. In the mouse embryo, FoxA2 and HNF-6 are expressed at the onset of liver, gall bladder, and pancreatic development and their embryonic liver expression is transiently diminished between 13 and 15 days postcoitum, suggesting that their hepatic expression may be coordinately regulated (24, 31, 40). Recent analysis of hnf6-/- mice demonstrates that HNF-6 expression is required for gall bladder development and that the mice exhibit severe defects in formation of intrahepatic bile ducts, causing biliary cholestasis (5, 18). It is interesting to speculate that HNF-6 deficiency may limit transcriptional activity of the FoxA2 protein and contribute to defective formation of the gall bladder and the hepatic biliary tree.
Recent studies demonstrated that interactions between the Pou-Homeodomain Oct4 and FoxD3 proteins converted FoxD3 into a transcriptional repressor of the FoxA1 and FoxA2 promoters, thus providing a mechanism for inhibition of FoxA expression in embryonic stem cells and prevention of cellular differentiation (14). The Oct4 Homeodomain mediated these protein interactions through the FoxD3 winged-helix motif, but Oct4 was unable to interact with the FoxA1 and FoxA2 proteins and influence their transcriptional activity (14). In another study, interaction between growth hormone-inducible Stat5b and FoxA2 (HNF-3ß) inhibited FoxA2 DNA binding and transcriptional activity and also prevented activation of Stat5B by tyrosine phosphorylation (37). Taken together, these studies indicate that the transcriptional activity of Fox proteins can be influenced through winged-helix interactions with distinct transcription factors that function to either stimulate or repress their transcriptional activities. Our studies demonstrate that the winged-helix domain not only mediates DNA binding and nuclear localization (22, 39) but also functions as a protein-protein association motif to recruit transcriptional coactivator proteins.
In summary, our transfection and protein binding studies suggest a model in which, depending on the target sequence, FoxA2 and HNF-6 protein interaction can either synergistically stimulate or repress transcription (Fig. 8). On a FoxA-dependent promoter, HNF-6 and FoxA2 protein interaction synergistically stimulates transcription by HNF-6-mediated recruitment of the p300/CBP histone acetyltransferase proteins, which also interact with the basal RNA polymerase II transcriptional machinery. On an HNF-6-specific site, the association between the HNF-6 and FoxA2 proteins interferes with the binding of HNF-6 to its target DNA sequence, thereby causing reduced transcription of HNF-6 dependent target genes.
|
ACKNOWLEDGMENTS |
|---|
We thank K. Zaret, R. Storti, S. Ackerman, K. Colley, P. Raychaudhuri, and H. Kiyokawa for helpful suggestions and V. Kalinichenko, Y. Zhou, and P. Raychaudhuri for critically reviewing the manuscript. We thank P. Raychaudhuri (University of Illinois at Chicago) for the E1A (S12) expression vector, I. Talianidis (Institute of Molecular Biology and Biotechnology, Crete, Greece) for the CBP HAT (-) expression vector and F. Lemaigre (Université Catholique de Louvain, Brussels, Belgium) for providing us with the mutant HNF-6 L350A expression vectors.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2. Chan, H. M., and N. B. La Thangue. 2001. p300/CBP proteins: HATs for transcriptional bridges and scaffolds. J. Cell Sci. 114:2363-2373.
3. Chen, H., M. Tini, and R. M. Evans. 2001. HATs on and beyond chromatin. Curr. Opin. Cell. Biol. 13:218-224.[CrossRef][Medline]
4. Clark, K. L., E. D. Halay, E. Lai, and S. K. Burley. 1993. Co-crystal structure of the HNF-3/fork head DNA-recognition motif resembles histone H5. Nature 364:412-420.[CrossRef][Medline]
5. Clotman, F., V. J. Lannoy, M. Reber, S. Cereghini, D. Cassiman, P. Jacquemin, T. Roskams, G. G. Rousseau, and F. P. Lemaigre. 2002. The onecut transcription factor HNF6 is required for normal development of the biliary tract. Development 129:1819-1828.
6. Costa, R. H., and D. R. Grayson. 1991. Site-directed mutagenesis of hepatocyte nuclear factor (HNF) binding sites in the mouse transthyretin (transthyretin) promoter reveal synergistic interactions with its enhancer region. Nucleic Acids Res. 19:4139-4145.
7. Costa, R. H., D. R. Grayson, and J. E. Darnell, Jr. 1989. Multiple hepatocyte-enriched nuclear factors function in the regulation of transthyretin and 1-antitrypsin genes. Mol. Cell. Biol. 9:1415-1425.
8. Costa, R. H., A.-X. Holterman, F. M. Rausa, and G. Adami. 2001. Gene regulation and in vivo function of liver transcription factors, p. 59-75. In I. M. Arias et al. (ed.), The liver: biology and pathobiology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
9. Costa, R. H., V. V. Kalinichenko, and L. Lim. 2001. Transcription factors in mouse lung development and function. Am. J. Physiol. Lung Cell Mol. Physiol. 280:L823-L838.
10. Courtois, G., S. Baumhueter, and G. R. Crabtree. 1988. Purified hepatocyte nuclear factor 1 interacts with a family of hepatocyte-specific promoters. Proc. Natl. Acad. Sci. USA 85:7937-7941.
11. DiPersio, C. M., D. A. Jackson, and K. S. Zaret. 1991. The extracellular matrix coordinately modulates liver transcription factors and hepatocyte morphology. Mol. Cell. Biol. 11:4405-4414.
12. Duncan, S. A. 2000. Transcriptional regulation of liver development. Dev. Dyn. 219:131-142.[CrossRef][Medline]
13. Grayson, D. R., R. H. Costa, K. G. Xanthopoulos, and J. E. Darnell. 1988. One factor recognizes the liver-specific enhancers in 1-antitrypsin and transthyretin genes. Science 239:786-788.
14. Guo, Y., R. Costa, H. Ramsey, T. Starnes, G. Vance, K. Robertson, M. Kelley, R. Reinbold, H. Scholer, and R. Hromas. 2002. The embryonic stem cell transcription factors Oct-4 and FoxD3 interact to regulate endodermal-specific promoter expression. Proc. Natl. Acad. Sci. USA 99:3663-3667.
15. Hatzis, P., and I. Talianidis. 2001. Regulatory mechanisms controlling human hepatocyte nuclear factor 4 gene expression. Mol. Cell. Biol. 21:7320-7330.
16. Herbst, R. S., U. Nielsch, F. Sladek, E. Lai, L. E. Babiss, and J. E. Darnell, Jr. 1991. Differential regulation of hepatocyte-enriched transcription factors explains changes in albumin and transthyretin gene expression among hepatoma cells. New Biol. 3:289-296.[Medline]
17. Jacob, A., S. Budhiraja, X. Qian, D. Clevidence, R. H. Costa, and R. R. Reichel. 1994. Retinoic acid-mediated activation of HNF-3 during EC stem cell differentiation. Nucleic Acids Res. 22:2126-2133.
18. Jacquemin, P., S. M. Durviaux, J. Jensen, C. Godfraind, G. Gradwohl, F. Guillemot, O. D. Madsen, P. Carmeliet, M. Dewerchin, D. Collen, G. G. Rousseau, and F. P. Lemaigre. 2000. Transcription factor hepatocyte nuclear factor 6 regulates pancreatic endocrine cell differentiation and controls expression of the proendocrine gene ngn3. Mol. Cell. Biol. 20:4445-4454.
19. Kaestner, K. H. 2000. The hepatocyte nuclear factor 3 (HNF3 or FOXA) family in metabolism. Trends Endocrinol. Metab. 11:281-285.[CrossRef][Medline]
20. Kaestner, K. H., W. Knochel, and D. E. Martinez. 2000. Unified nomenclature for the winged-helix/forkhead transcription factors. Genes Dev. 14:142-146.
21. Kaufmann, E., and W. Knochel. 1996. Five years on the wings of fork head. Mech. Dev. 57:3-20.[CrossRef][Medline]
22. Lai, E., V. R. Prezioso, E. Smith, O. Litvin, R. H. Costa, and J. E. Darnell, Jr. 1990. HNF-3A, a hepatocyte-enriched transcription factor of novel structure is regulated transcriptionally. Genes Dev. 4:1427-1436.
23. Lai, E., V. R. Prezioso, W. F. Tao, W. S. Chen, and J. E. Darnell, Jr. 1991. Hepatocyte nuclear factor 3 belongs to a gene family in mammals that is homologous to the Drosophila homeotic gene fork head. Genes Dev. 5:416-427.
24. Landry, C., F. Clotman, T. Hioki, H. Oda, J. J. Picard, F. P. Lemaigre, and G. G. Rousseau. 1997. HNF-6 is expressed in endoderm derivatives and nervous system of the mouse embryo and participates to the cross-regulatory network of liver-enriched transcription factors. Dev. Biol. 192:247-257.[CrossRef][Medline]
25. Lannoy, V. J., T. R. Burglin, G. G. Rousseau, and F. P. Lemaigre. 1998. Isoforms of hepatocyte nuclear factor-6 differ in DNA-binding properties, contain a bifunctional homeodomain, and define the new ONECUT class of homeodomain proteins. J. Biol. Chem. 273:13552-13562.
26. Lannoy, V. J., A. Rodolosse, C. E. Pierreux, G. G. Rousseau, and F. P. Lemaigre. 2000. Transcriptional stimulation by hepatocyte nuclear factor-6. Target-specific recruitment of either CREB-binding protein (CBP) or p300/CBP-associated factor (p/CAF). J. Biol. Chem. 275:22098-22103.
27. Lemaigre, F. P., S. M. Durviaux, and G. G. Rousseau. 1993. Liver-specific factor binding to the liver promoter of a 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase gene. J. Biol. Chem. 268:19896-19905.
28. Lemaigre, F. P., S. M. Durviaux, O. Truong, V. J. Lannoy, J. J. Hsuan, and G. G. Rousseau. 1996. Hepatocyte nuclear factor 6, a transcription factor that contains a novel type of homeodomain and a single cut domain. Proc. Natl. Acad. Sci. USA 93:9460-9464.
29. Liu, J. K., C. M. DiPersio, and K. S. Zaret. 1991. Extracellular signals that regulate liver transcription factors during hepatic differentiation in vitro. Mol. Cell. Biol. 11:773-784.
30. Monaci, P., A. Nicosia, and R. Cortese. 1988. Two different liver-specific factors stimulate in vitro transcription from the human 1-antitrypsin promoter. EMBO J. 7:2075-2087.[Medline]
31. Monaghan, A. P., K. H. Kaestner, E. Grau, and G. Schutz. 1993. Postimplantation expression patterns indicate a role for the mouse forkhead/HNF-3 , ß and genes in determination of the definitive endoderm, chordamesoderm and neuroectoderm. Development 119:567-578.[Abstract]
32. Naar, A. M., B. D. Lemon, and R. Tjian. 2001. Transcriptional coactivator complexes. Annu. Rev. Biochem. 70:475-501.[CrossRef][Medline]
33. Narlikar, G. J., H. Y. Fan, and R. E. Kingston. 2002. Cooperation between complexes that regulate chromatin structure and transcription. Cell 108:475-487.[CrossRef][Medline]
34. Overdier, D. G., A. Porcella, and R. H. Costa. 1994. The DNA-binding specificity of the hepatocyte nuclear factor 3/forkhead domain is influenced by amino acid residues adjacent to the recognition helix. Mol. Cell. Biol. 14:2755-2766.
35. Pani, L., D. G. Overdier, A. Porcella, X. Qian, E. Lai, and R. H. Costa. 1992. Hepatocyte nuclear factor 3ß contains two transcriptional activation domains, one of which is novel and conserved with the Drosophila forkhead protein. Mol. Cell. Biol. 12:3723-3732.
36. Pani, L., X. B. Qian, D. Clevidence, and R. H. Costa. 1992. The restricted promoter activity of the liver transcription factor hepatocyte nuclear factor 3ß involves a cell-specific factor and positive autoactivation. Mol. Cell. Biol. 12:552-562.
37. Park, S. H., and D. J. Waxman. 2001. Inhibitory cross-talk between STAT5b and liver nuclear factor HNF3ß. Impact on the regulation of growth hormone pulse-stimulated, male-specific liver cytochrome P-450 gene expression. J. Biol. Chem. 276:43031-43039.
38. Perl, A. K., and J. A. Whitsett. 1999. Molecular mechanisms controlling lung morphogenesis. Clin. Genet. 56:14-27.[CrossRef][Medline]
39. Qian, X., and R. H. Costa. 1995. Analysis of HNF-3ß protein domains required for transcriptional activation and nuclear targeting. Nucleic Acids Res. 23:1184-1191.
40. Rausa, F., U. Samadani, H. Ye, L. Lim, C. F. Fletcher, N. A. Jenkins, N. G. Copeland, and R. H. Costa. 1997. The cut-homeodomain transcriptional activator HNF-6 is coexpressed with its target gene HNF-3ß in the developing murine liver and pancreas. Dev. Biol. 192:228-246.[CrossRef][Medline]
41. Rausa, F. M., L. Galarneau, B. l. L, and R. H. Costa. 1999. The nuclear receptor fetoprotein transcription factor is coexpressed with its target gene HNF-3ß in the developing murine liver intestine and pancreas. Mech. Dev. 89:185-188.[CrossRef][Medline]
42. Rausa, F. M., Y. Tan, H. Zhou, K. Yoo, D. B. Stolz, S. Watkins, R. R. Franks, T. G. Unterman, and R. H. Costa. 2000. Elevated levels of HNF-3ß in mouse hepatocytes influence expression of genes involved in bile acid and glucose homeostasis. Mol. Cell. Biol. 20:8264-8282.
43. Roth, S. Y., J. M. Denu, and C. D. Allis. 2001. Histone acetyltransferases. Annu. Rev. Biochem. 70:81-120.[CrossRef][Medline]
44. Samadani, U., and R. H. Costa. 1996. The transcriptional activator hepatocyte nuclear factor six regulates liver gene expression. Mol. Cell. Biol. 16:6273-6284.[Abstract]
45. Samadani, U., A. Porcella, L. Pani, P. F. Johnson, J. Burch, R. Pine, and R. H. Costa. 1995. Cytokine regulation of the liver transcription factor HNF-3ß is mediated by the C/EBP family and interferon regulatory factor 1. Cell Growth Differ. 6:879-890.[Abstract]
46. Soutoglou, E., N. Katrakili, and I. Talianidis. 2000. Acetylation regulates transcription factor activity at multiple levels. Mol. Cell 5:745-751.[CrossRef][Medline]
47. Streeper, R. S., L. A. Hornbuckle, C. A. Svitek, J. K. Goldman, J. K. Oeser, and R. M. O'Brien. 2001. Protein kinase A phosphorylates hepatocyte nuclear factor-6 and stimulates glucose-6-phosphatase catalytic subunit gene transcription. J. Biol. Chem. 276:19111-19118.
48. Tan, Y., G. Adami, and R. H. Costa. 2002. Maintaining HNF-6 expression prevents AdHNF3ß Mediated decrease in hepatic levels of Glut2 and glycogen. Hepatology 35:790-798.[CrossRef][Medline]
49. Tan, Y., R. H. Costa, I. Kovesdi, and R. R. Reichel. 2001. Adenovirus-mediated increase of HNF-3 levels stimulates transthyretin and sonic hedgehog expression which is associated with F9 cell differentiation toward the visceral endoderm lineage. Gene Expr. 9:237-248.[Medline]
50. Tan, Y., D. E. Hughes, X. Wang, and R. H. Costa. 2002. Adenovirus-mediated hepatic increase in HNF-3ß or HNF-3 shows differences in levels of liver glycogen and gene expression. Hepatology 35:30-39.[CrossRef][Medline]
51. Wang, J. C., M. Waltner-Law, K. Yamada, H. Osawa, S. Stifani, and D. K. Granner. 2000. Transducin-like enhancer of split proteins, the human homologs of Drosophila groucho, interact with hepatic nuclear factor 3ß. J. Biol. Chem. 275:18418-18423.
52. Wang, X., E. Quail, N.-J. Hung, Y. Tan, H. Ye, and R. H. Costa. 2001. Increased levels of forkhead box M1B transcription factor in transgenic mouse hepatocytes prevents age-related proliferation defects in regenerating liver. Proc. Natl. Acad. Sci. USA 98:11468-11473.
53. Ye, H., T. F. Kelly, U. Samadani, L. Lim, S. Rubio, D. G. Overdier, K. A. Roebuck, and R. H. Costa. 1997. Hepatocyte nuclear factor 3/fork head homolog 11 is expressed in proliferating epithelial and mesenchymal cells of embryonic and adult tissues. Mol. Cell. Biol. 17:1626-1641.[Abstract]
54. Zaret, K. 1999. Developmental competence of the gut endoderm: genetic potentiation by GATA and HNF3/fork head proteins. Dev. Biol. 209:1-10.[CrossRef][Medline]
This article has been cited by other articles:
