![[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}
Expression
,
Irene Kyrmizi, and
Iannis Talianidis*
Institute of Molecular Biology and Biotechnology, Foundation for Research and Technology Hellas, Herakleion, Crete, Greece
Received 17 February 2006/ Returned for modification 3 April 2006/ Accepted 17 July 2006
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
expression. During MAP kinase signaling, the recruitment of HNF-3ß and HNF-1 to the HNF-4 enhancer and RNA polymerase II to the proximal HNF-4 promoter was compromised. CBP, Brg1, and TFIIB were also dissociated from the HNF-4 regulatory regions, and the enhancer-promoter complex was disrupted. Interestingly, the extent of nucleosome acetylation did not decrease at either regulatory region, and HNF-6 and HNF-1, as well as components of the TFIID, remained associated with the proximal promoter during the repressed state. The results point to an absolute requirement of enhancer-promoter communication for maintaining the active state of the HNF-4 gene and provide evidence for a molecular bookmarking mechanism, which may contribute to the prevention of permanent silencing of the locus during the repressed state. |
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
HNF-4 is a principal member of the hepatic transcription factor network. Mouse embryos lacking HNF-4 die before completing gastrulation due to its crucial role in extraembryonic, visceral endoderm function (3). Studies in mice where the early lethal phenotype is circumvented, either by complementation with tetraploid embryo-derived visceral endoderm or by inactivating HNF-4 specifically at the hepatoblast stage, have revealed that HNF-4 is dispensable for hepatocyte specification but is essential for subsequent steps of differentiation and the development of normal liver architecture during morphogenesis (8, 22, 25). The pivotal role of HNF-4 in the maintenance of the differentiated hepatic phenotype is highlighted by the severe metabolic defects in mice where HNF-4 was inactivated in the adult liver (12) and by the exceptionally high number of potential direct target genes revealed by genome-scale target search studies (24). In the adult human liver, HNF-4 was found to occupy ca. 12% of the genes represented in a 13K DNA microarray and ca. 42% of those bound by RNA polymerase II (pol-II) (24). These studies established HNF-4 as a regulator of several biological pathways, which raises the importance of understanding how its activity and expression are regulated. Previous analyses have revealed that HNF-4 activity is subject to regulation by phosphorylation (13, 19, 39), acetylation (33), and protein-protein interactions with other factors or by coregulators (6, 31, 33). Further complexity in the control of HNF-4-dependent genes arises from the existence of several HNF-4 isoforms generated by alternative splicing (7, 18). The mechanism involved in the transcriptional regulation of the HNF-4 gene has also been studied in great detail. Two main regulatory regions have been identified: the proximal promoter and a distant enhancer located around 6.5 kb upstream of the transcription start site (1, 10, 11). It has been shown that activation of the HNF-4 gene requires the synergistic action of HNF-1 and HNF-6 on the proximal promoter, which communicates via a looping mechanism with a distant enhancer bound by HNF-1, HNF-3ß (FoxA2), and C/EBP (10, 11). Although the main steps of the dynamic mechanism involved in the initial activation of the HNF-4 gene have been comprehended in great detail, it is not clear whether the enhancer-promoter complex is necessary for the maintenance of transcription after the gene has been activated. Furthermore, given the high number of HNF-4 targets playing roles in diverse biological pathways, the question of whether HNF-4 expression can be modulated in response to different environmental signals is of considerable interest.
Here we have investigated the potential role of mitogen-activated protein (MAP) kinase regulation on HNF-4 expression, a signaling pathway involved in the control of fundamental cellular processes, including cell growth and survival, differentiation, and metabolism (15, 26). We demonstrate that Erk2 activation downregulates HNF-4 expression in human hepatoma cells at the transcriptional level via a mechanism that involves loss of C/EBP expression and concomitant disruption of the HNF-4 enhancer-promoter complex.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
gene and its 5' deletion derivatives have been described (10). The plasmid Enhancer-Luc was constructed by inserting a PCR-amplified DNA fragment containing the –6.1- to –6.7-kb region of the human HNF-4 upstream region into the BglII site of pGl-luc plasmid containing the region of the human HNF-4 gene from nucleotides (nt) –40 to +67. pCMV-C/EBP, pCMV-HNF-1, and pCMV-HNF-3ß have been described elsewhere (10, 35). pCMV-MEK1 was from Stratagene, and pCMV-Raf-BXB was obtained from G. Mavrothalassitis. Site-directed mutagenesis of the footprinted regions of the human HNF-4 enhancer was performed by the GeneEditor kit (Promega) according to the manufacturer's instructions. The following oligonucleotides were used for mutagenesis: HNF-3 site mutant, 5'-CTC TCT TTG GTA AGG ATC CGG ATT TGC TCA GGA CCC AGC; DR-1 site mutant, 5'-GGG GGA ACA AGC AGA CTA TGT CGA CTT GAG CAA AGC CTC; C/EBP site mutant, 5'-GGC CAG CGG CCT GGA TCC TAA CCC TGG AGG CC; and HNF-1 site mutant: 5'-CTC ATG CCC AGT CTA GAT TGG AAG GCA AAA TCA ACA GGC.
Cell culture RT-PCR, transfections, and Western blot assays. HepG2 and HuH7 cells were grown in Dulbecco modified Eagle medium supplemented with 10% fetal calf serum. The cells were seeded in 60-mm culture plates and treated with 1 µg of phorbol myristate acetate (PMA) (Calbiochem)/ml or 20 µM PD98059 (Biomol) for different time intervals. Total RNAs were prepared by extraction with TRIzol (Invitrogen) reagent and treated with 10 U of DNase I (Invitrogen) for 30 min at 37°C before cDNA synthesis, which was performed by SuperScript reverse transcriptase (RT; Invitrogen) using an oligo(dT) primer. The cDNAs were used directly for real-time PCR amplification as described in reference (16, 17). Primer sets used for RT-PCR have been described (10, 11, 17).
For Western blot analysis, cells were lysed in a buffer containing 50 mM Tris-HCl (pH 7.5), 1% NP-40, 0.25% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM NaF, and 1 mM phenylmethylsulfonyl fluoride, supplemented with protease inhibitor cocktail (Roche). The extracts were then separated by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis and analyzed in Western blots as described previously (16, 35). Quantitative assessment of Western blot signals was performed with a Fujifilm LAS-1000 luminescent image analyzer.
For luciferase reporter assays, cells in six-well plates were transfected with 0.5 µg of promoter-reporter DNA, 0.25 µg of cytomegalovirus (CMV) expression vectors, and 0.25 µg of pCMV-ß-Gal plasmid by the calcium phosphate coprecipitation method. At 12 h after transfection, the cells were harvested and lysed as described previously (10). Luciferase activities were measured by using the luciferase assay kit from Promega, and the values were normalized to ß-galactosidase values.
siRNA-mediated knockdown of C/EBP.
To interfere with endogenous C/EBP expression in HepG2 and HuH7 cells, we used a double-stranded RNA with the following sequences: sense strand, 5'-GAA GUC GGU GGA CAA GAA CUU; and antisense strand, 5'-GUU CUU GUC CAC CGA CUU CUU. Control small interfering RNA (siRNA) containing scrambled sequence was purchased from Santa Cruz Biotechnology. The cells were transfected with the double-stranded RNA at 100 nM by using the jetSI-ENDO kit (Polyplus) and analyzed 72 h after transfection.
ChIP and chromosome conformation capture (3C) assays. Formaldehyde cross-linking of HepG2 cells and chromatin chromatin immunoprecipitations (ChIPs) were performed as described previously (10, 11, 34). For real-time PCR analysis of the HNF-4 enhancer and promoter regions, the following primer sets were used: enhancer, 5' (–6756 nt)-GGC TCT GAC ACT GC AGA GTT CTA GAA C and 5' (–6393 nt)-CCA AAC TTA CCC AGC TGC TAA TCA TTG C; promoter, 5' (–444 nt)-TCG AGG CAG CCT TAT CTC TGC AAA AGC and 5' (–97 nt)-TCG AGG GGT GGG GGT AAT GGT TAA TCG G; and control downstream region, 5' (+3731 nt)-AAT GCG GGA GGG CCC GGA CAT CTC CAG C and 5' (+3963 nt)-CCC ACC ATC CAC GCC CAT CCT CAC CTG G.
3C experiments were performed as described earlier but with minor modifications (38). Cells were treated with 2% of formaldehyde for 10 min to cross-link chromatin. After quenching with 0.125 M glycine, nuclei were prepared by incubating the cells in a buffer containing 10 mM Tris-HCl (pH 8.0), 10 mM NaCl, 0.2% NP-40, and protease inhibitor cocktail for 1 h at 4°C. The nuclei were harvested by centrifugation and resuspended in restriction enzyme buffer containing 0.3% SDS. After incubation at 37°C for 1 h, Triton X-100 was added to 1.8% final concentration, followed by further incubation for 1 h to sequester the SDS. Aliquots of 106 nuclei were digested by 500 U of BglII and 600 U of BclI at 37°C for 18 h. The digestion was continued for an additional 3 h after the addition of fresh enzyme. Under these conditions, the efficiencies of BglII and BclI cleavage at the studied locations were ca. 80 and 50%, respectively. The enzymes were inactivated by the addition of SDS to a final concentration of 1.6% and incubation at 65°C for 15 min. An aliquot of chromatin (containing 2 µg of DNA) was then diluted in a buffer containing 50 mM Tris-HCl (pH 8.0), 10 mM MgCl2, 10 mM dithiothreitol, and 1 mM ATP to a final DNA concentration of 2.5 µg/ml and after, the addition of Triton X-100 to 1%, was incubated at 37°C for 1 h. The DNAs were ligated by using 40 U of T4 DNA ligase at 16°C overnight. Cross-links were reversed by incubating the samples at 65°C for 16 h, and the DNA was purified by sequential RNase A and proteinase K treatments, followed by phenol-chloroform extractions and ethanol precipitations. To detect the intramolecular ligation between the cross-linked HNF-4 enhancer and promoter fragments, PCR amplifications (30 cycles) were performed in the presence of 1 µCi of [32P]dATP and [32P]dCTP using the following primer sets: primer A, 5' (–7521 nt)-TTA ACT TCC AGG GTT GTC ATG; and primer B, 5' (–251 nt)-CCA GCA GTT GTA ATT AGC ACC.
Antibodies.
Anti-pErk2 and anti-p42 were obtained from Cell Signaling. The antibodies against RNA pol-II, Brg-1, CBP, HNF-1, HNF-3ß, HNF-6, C/EBP, TBP, TFIIB, and TAF-1 were from Santa Cruz Biotechnology. Acetyl-H3 and acetyl-H4 antibodies were from Upstate Biotechnology. The HNF-4 antibody has been described previously (10). The specificity of each antibody was verified by immunoprecipitation-Western blot assays using cross-linked extracts.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
mRNA levels after 6 h of PMA treatment (Fig. 1B). At later time points of PMA treatment, however, the amount of HNF-4 mRNA increased to somewhat higher than the levels of untreated controls. Specific inhibition of the MEK-ERK pathway by PD98059, resulted in a twofold increase in HNF-4 mRNA levels, reaching a maximum at 6 h posttreatment, whereas cotreatment of the cells with both PMA and PD98059 abolished changes in either direction (Fig. 1B). The steady-state levels of HNF-4 protein changed in a similar fashion in PMA- and PD98059-treated cells, except for a shift in the time course of PMA-mediated repression, which was maximal at 12 h posttreatment (Fig. 1C). The time difference between maximum repression of RNA and protein levels is expected if protein stability is taken into account.
MAP kinase signaling inhibits the activity of the HNF-4 enhancer.
In order to investigate the mechanism of the MAP kinase-mediated repressive effect on HNF-4 expression, we first performed transient-transfection assays in HepG2 cells with a construct containing the 12-kb upstream sequence of the HNF-4 gene. PMA treatment reduced reporter activity to 24% of the control, whereas PD98059 stimulated it by a factor of 3.4 (Fig. 2). Overexpression of constitutively active Raf kinase (Raf-BXB) or MEK1, two kinases acting downstream of PKC and upstream of Erk2, decreased the activity of the reporter to 28 and 10% of the control, respectively. These results further demonstrate that the downregulation of HNF-4 mRNA upon PMA treatment is mediated by the activated PKC-Raf-MEK1-Erk2 pathway, which represses HNF-4 expression at the transcriptional level.
|
Downregulation of C/EBP upon MAP kinase signaling is involved in the molecular mechanism of HNF-4 repression.
Previous studies on the mouse and human HNF-4 enhancer have identified four cis-acting elements as binding sites of HNF-1, C/EBP, nuclear hormone receptors (DR-1), and the HNF-3 family of transcription factors (1, 11). In order to identify the specific factor(s) involved in the MAP kinase-mediated effect, we tested the activity of individual binding site mutants (Fig. 3A) in cells treated with PMA or PD98059 and in cells overexpressing Raf-BXB or MEK1. None of the introduced mutations alone eliminated completely the activity of the HNF-4 enhancer-promoter construct, which suggests that at least in transient-transfection assays there is some functional redundancy between the factors binding to the enhancer region. Mutations in the HNF-1, C/EBP, DR-1, and HNF-3 binding sites reduced promoter activity to 50, 32, 88, and 70% of the control, respectively (Fig. 3B). We note, however, that mutation of the C/EBP binding site resulted in the highest degree of repression, which was not significantly different from the repression level observed with the wild-type construct after PMA treatment. More importantly, we found that the activity of the mutant C/EBP site-containing construct was not affected by PMA and PD98059 treatment or overexpressions of Raf-BXB and MEK1, whereas the activities of the other mutants were affected in a way similar to the wild-type construct (Fig. 3B). These results suggest that the effect of MAP kinase signaling on HNF-4 expression is mediated through the modulation of the action of C/EBP factors on the HNF-4 enhancer.
|
expression upon the induction of MAP kinases or a modulation of its activity via phosphorylation. The results of our subsequent experiments corroborate the former scenario. First, overexpression of C/EBP transactivated the 6.7 kb-luc reporter five- to sevenfold in either untreated, PMA-treated, or Raf-BXB- or MEK1-transfected cells (Fig. 3C). This shows that exogenously added C/EBP can reverse the repression caused by PMA treatment or by overexpression of the upstream kinases (Fig. 3C). As expected, the DNA binding-deficient R289A mutant form of C/EBP (23) could not transactivate the 6.7 kb-luc reporter or reverse MAPK-mediated repression (Fig. 3C). Second, the mRNA and protein levels of endogenous C/EBP transiently decreased in HepG2 cells treated with PMA (Fig. 4A and B, panels on the right). The kinetics of the changes observed on C/EBP mRNA and protein levels were in good correlation with those of the phosphorylated Erk2 protein and HNF-4 mRNA shown in Fig. 1. A substantial decrease (to ca. 20% of the control) in CEBP expression was already evident by 3 h posttreatment with PMA, whereas maximal reduction (to ca. 3% of the control) was observed at the 6-h time point. At later time periods (12 to 48 h), when the phosphorylated Erk2 signal disappeared, C/EBP expression gradually returned to the original pretreatment levels (Fig. 4). When the cells were treated with PD98059, C/EBP mRNA and protein levels increased about 2- and 2.5-fold, respectively (Fig. 4). In addition, PD98059 could reverse PMA-mediated repression of C/EBP (Fig. 4).
|
and HNF-4 in PD98059-treated cells also suggests that their expression in control HepG2 cells is not maximal, probably due to the presence of basal mitogenic signals in the serum-containing growth media. Relevant to this, one would expect that in cells treated simultaneously with both PMA (an inducer) and PD98059 (an inhibitor of Erk2 phosphorylation) the effect of the inhibitor would predominate and, as a result, increased expression of C/EBP and HNF-4 would be expected. Instead, we observed expression levels similar to those of untreated cells (Fig. 1 and 4). One possible explanation for the observed difference between the effects of PMA+PD98059 and of PD98059 alone could be that, besides the activation of the MEK1-Erk2 pathway, in PMA+PD98059-treated cells the PMA may also induce parallel kinase pathways affecting the expression of C/EBP and HNF-4. Although this possibility cannot be excluded entirely, we note that a faint, but visible band corresponding to phosphorylated Erk2 was observable in cells exposed simultaneously to the two drugs (Fig. 1A). This fits with the other results, which showed a positive correlation between the amounts of active MEK1-Erk2 proteins in the cells and the expression of C/EBP and HNF-4. We therefore conclude that MEK1-Erk2 activation is the main PMA-induced pathway responsible for the downregulation of C/EBP and HNF-4.
We also tested the expression of the other two factors (HNF-1 and HNF-3ß), which bind to the HNF-4 enhancer. HNF-3ß mRNA and protein levels remained constant after treatment with either PMA or PD98059. On the other hand, the mRNA and protein levels of HNF-1 were somewhat decreased (to ca. 60 to 70% of the control) in PMA-treated cells and increased (
1.4-fold above the control) in PD98059-treated cells (Fig. 4A and B). We note, however, that at 3 or 6 h after PMA treatment when phosphorylated Erk2 could be detected and HNF-4 mRNA levels decreased to 51 and 18%, respectively (Fig. 1A and B), no significant changes could be observed in HNF-1 expression (Fig. 4A and B). Maximal HNF-1 repression was seen at the 12-h time point, when HNF-4 protein had already dropped to its lowest levels (Fig. 1C and 4). These kinetics suggest that PMA-mediated repression of HNF-1 is a consequence, rather than the cause, of changes in HNF-4 expression, which is consistent with the fact that HNF-1 is a bona fide target gene of HNF-4 (20, 21).
Taken together, these results show that the mechanism of repression of HNF-4 involves compromised activity of the HNF-4 enhancer, which can be reversed by ectopic expression of C/EBP, the levels of which decrease during MAP kinase signaling. Furthermore, the effects of the MAPK modulators (PMA and PD98059) on C/EBP and HNF-4 expression could be reproduced in another human hepatoma cell line (HuH7), suggesting that the observations are not due to potential peculiarities specifically related to HepG2 cells. In HuH7 cells both drugs induced changes in C/EBP and HNF-4 mRNA levels in a similar direction and with similar kinetics as were observed in HepG2 cells (Fig. 5A).
|
in HepG2 and HuH7 cells results in decreased expression of HNF-4.
To further assess the direct role of C/EBP on HNF-4 expression, we used RNAi-mediated approach. Transfection of HepG2 and HuH7 cells by a C/EBP-specific dsRNA resulted in a decrease of C/EBP mRNA levels corresponding to 16 and 23% of the endogenous levels, respectively (Fig. 5B), and to barely detectable endogenous C/EBP protein levels (Fig. 5C). In both cell lines a parallel decrease of HNF-4 mRNA and protein levels was observed (Fig. 5B and C). These results confirm the essential role of C/EBP on HNF-4 expression.
MAP kinase signaling disrupts the HNF-4 enhancer-promoter complex.
In order to investigate the potential MAP kinase-induced changes in factor occupancy and in the configuration of the HNF-4 regulatory region in vivo, we performed ChIP assays. Untreated and PMA-treated HepG2 cells were cross-linked with formaldehyde and, after sonication, the soluble chromatin was immunoprecipitated with various antibodies against factors binding to the enhancer or the promoter. The purified DNA from the immunoprecipitates was used for PCR analysis with primer sets amplifying the proximal promoter and the distal enhancer region of the HNF-4 gene. In control, untreated HepG2 cells, the DNA immunoprecipitated by antibodies against factors which specifically bind to the enhancer (C/EBP and HNF-3ß) or the proximal promoter (HNF-6, RNA pol-II, TFIIB, TBP, and TAF-1) contained sequences encompassing both the enhancer and the promoter region (Fig. 6 and 7). The simultaneous presence of the two DNA fragments in the immunoprecipitates of this variety of factors demonstrates that, in HepG2 cells, the enhancer and promoter regions are in close proximity and form a higher-order complex by looping out the intervening DNA. In cells treated with PMA for 6 h, when HNF-4 expression was repressed, we did not observe association of HNF-3ß, C/EBP, RNA pol-II, CBP, Brg-1, and TFIIB with either regulatory region. However, ChIP signals significantly above background levels were detected in HNF-1, HNF-6, TBP, and TAF-1 immunoprecipitates on the proximal promoter but not on the distant enhancer (Fig. 6 and 7). Since HNF-1 has binding sites at both regions, whereas HNF-6, TBP, and TAF-1 are recruited to the proximal promoter, these results are indicative of a linear conformation of the HNF-4 regulatory region during MAP kinase-induced repression conditions. The unaffected associations of HNF-1, HNF-6, and the components of TFIID with the promoter are of special interest, since they may be part of a mechanism that keeps the locus in an open state during repression. In line with this scenario is the finding that histone 3 and 4 acetylation levels were not significantly changed in PMA-treated cells (Fig. 6). In experiments where HepG2 cells were treated with PMA for 24 h, at which time the HNF-4 expression is restored, the ChIP signals obtained with all of the antibodies were similar to those detected in untreated control cells (Fig. 6 and 7).
|
|
is expressed, but not in the nonexpressing HeLa cell line (Fig. 8B). This signal was strictly dependent on cross-linking and DNA ligation, confirming the specificity of the approach. Importantly, the PCR product was also detectable in cells treated with PMA for 24 h, when HNF-4 is reexpressed, but not in cells treated with PMA for 6 h or in cells transfected with C/EBP siRNA, where HNF-4 expression is downregulated (Fig. 8B). These observations are in good correlation with the conclusions drawn from the data of ChIP assays regarding the role of C/EBP in MAP kinase-induced configuration changes of the HNF-4 regulatory regions.
|
is reexpressed, and the communication between the enhancer and the promoter is restored with parallel activation of the transcription of the HNF-4 gene. |
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
is a member of the nuclear receptor superfamily and regulates the transcription of a wide range of genes involved in different metabolic pathways, hepatocyte differentiation, and liver organogenesis. Previous studies employing genome-wide location analysis have identified a surprisingly large number of gene regulatory regions occupied by HNF-4 in human livers. Recruitment of HNF-4 was observed in ca. 12% of the genes represented in a human 13K array, which is about 5 to 10 times higher than those identified for any other transcription factor thus far (24). Although the role of HNF-4 in the regulation of the identified genes still needs to be validated by expression and genetic analysis, the huge number of its potential targets represents convincing evidence for the role of HNF-4 in the regulation of diverse biological pathways. Combined with the severe phenotypes observed in liver-specific HNF-4 KO models (12, 25), these findings indicate that any changes in the activity and/or the expression of HNF-4 will have important consequences in the pattern of expression of hepatic genes.
HNF-4 expression is modulated by MAP kinase signaling.
Although HNF-4 is a highly expressed, constitutively active gene in human hepatocytes, previous studies have provided some indications that its expression could be modulated by extracellular signals. For example, we have shown that in HepG2 cells, retinoic acid signaling can potentiate HNF-4 transcription, which is mediated by the action of liganded RXR/RAR on a bona fide hormone response element located at the proximal promoter of the HNF-4 gene (10). More recently, it was found that bile acids could suppress HNF-4 expression via an unknown molecular mechanism (27). Because bile acids can induce various signaling pathways, including PKC activation (5), we sought to determine whether the PKC-MAP kinase cascade can directly modulate HNF-4 expression. We found that treatment of HepG2 cells with PMA, a potent inducer of PKC, dramatically decreased endogenous HNF-4 mRNA levels. Several lines of evidence presented here suggest that the observed effect is mediated by the activation of the downstream Raf-MEK1-Erk2 signal transduction pathway. First, the levels of HNF-4 mRNA decreased only at the time periods when activated Erk2 was observed. Second, treatment of the cells with PD98059, a specific inhibitor of MEK1, increased the amounts of HNF-4 mRNA and counteracted PMA-dependent repression. Third, overexpression of a constitutively active Raf kinase mutant, or MEK1, inhibited the activity of a luciferase reporter driven by the upstream HNF-4 regulatory region in transient-transfection assays. These latter experiments also suggested that MAP kinase signaling downregulates HNF-4 expression at the transcriptional level. In vivo support for this notion was provided by ChIP experiments, which showed the dissociation of RNA pol-II from the HNF-4 promoter and the disassembly of the enhancer-promoter complex at the HNF-4 regulatory region in PMA-treated cells.
Our results also show that impaired HNF-4 enhancer activity emanating from MAP kinase-induced downregulation of C/EBP constitutes the molecular basis of the observed effects. In support of this we found that C/EBP mRNA and protein levels were dramatically reduced at time periods when phosphorylated Erk2 was detected. In addition, PD98059 treatment increased C/EBP expression and counteracted PMA-dependent downregulation. Importantly, these changes followed a kinetic pattern similar to that observed for HNF-4 mRNA. Finally, overexpression of C/EBP could restore the transcriptional activity of the HNF-4 upstream regulatory region in PMA-treated cells, and RNAi-mediated knockdown of C/EBP resulted in a decrease in endogenous HNF-4 expression.
Signaling through the Raf-MEK1-Erk2 pathway has been implicated in various cellular processes, including proliferation, differentiation, senescence, and apoptosis (15, 26). C/EBP is a strong inhibitor of cell proliferation, which uses multiple pathways to cause growth arrest (14, 32, 36, 37, 42). Although the molecular mechanism by which PMA treatment inhibited C/EBP was not a subject of the present study, a potential link between MAP kinase-induced cell proliferation and inhibition of C/EBP expression could be considered. Logarithmically growing HepG2 and HuH7 cells, however, express high levels of C/EBP, suggesting that in this cell line its antiproliferative activity is suppressed. In agreement with this are the recent findings showing that phosphorylation of C/EBP at serine 193 is required for its growth-inhibitory activity and that this modification is erased by protein phosphatase 2A in proliferating liver cells and a variety of hepatoma cell lines (40, 41). The dephosphorylated form of C/EBP was actually found to accelerate proliferation (41). Furthermore, we have not seen differences in PMA-mediated inhibition of either C/EBP or HNF-4 expression between experiments in which the cells were treated with PMA at the stage of logarithmic growth or confluence (data not shown). Therefore, MAP kinase induced proliferation as a mechanism of C/EBP repression in PMA-treated HepG2 cells is unlikely.
C/EBP has been shown to be a direct substrate of Erk2, which can phosphorylate it at serine 21 in myeloid cells (29). Phosphorylation at Ser21 blocks the ability of C/EBP to induce granulopoiesis of bipotential myeloid progenitor cells but does not affect monocyte differentiation or adipogenesis (29). Ser21 phosphorylation was observed at very early (15 to 30 min) times after PMA induction (29). In HepG2 cells, however, we could not detect this modification by Western blotting experiments with a phospho-Ser21-specific C/EBP antibody, even after very short time treatments with PMA (data not shown). Although we cannot entirely exclude the possibility that Ser-21 phosphorylation of C/EBP may occur in hepatic cells soon after stimulation, our results showing C/EBP mRNA and protein levels dropping dramatically 3 and 6 h after PMA treatment provide a clear demonstration that the mechanism of HNF-4 repression involves the loss of C/EBP rather than modulation of its activity via posttranslational modification. Nevertheless, the findings raise the intriguing possibility that C/EBP expression and activity could be modulated by cell-specific regulatory mechanisms, which have different sensitivities to MAP kinase signaling. Consistent with this notion are the studies in other systems demonstrating that activation of the MEK1-Erk2 pathway during the initial phase of adipocyte differentiation enhances C/EBP expression (28).
Disruption of the HNF-4 enhancer-promoter complex during MAP kinase signaling. We have shown previously that, during the initial activation of the HNF-4 gene in differentiating enterocytes, orderly assembled regulatory complexes at the proximal promoter and at the distant enhancer region associate with each other to form a higher-order complex by looping out the intervening DNA (11). The formation of the enhancer-promoter complex triggered a critical nucleosome remodeling event at the transcription start site that allowed the escape of RNA pol-II from the promoter (11).
Here, we used the same ChIP-based approach to study the HNF-4 enhancer-promoter complex in HepG2 cells, which constitutively express the gene. The immunoprecipitated chromatin obtained by antibodies recognizing several factors specifically associated with the enhancer (C/EBP and HNF-3ß), or the proximal promoter (HNF-6, RNA pol-II, TFIIB, TBP, and TAF-1), contained both enhancer and promoter sequences. The efficient cross-linking of the HNF-4 enhancer and promoter suggests that in HepG2 cells the two regions are in close proximity, bridged by the factors associated with them. MAP kinase activation disrupted this interaction, and the HNF-4 regulatory region attained a linear conformation. The enhancer-promoter complex was reformed at time points when phosphorylated Erk2 disappeared and transcription of the HNF-4 gene resumed. This change in the three-dimensional configuration of the HNF-4 regulatory region was further confirmed by 3C assays. Association between the HNF-4 enhancer and proximal promoter could be readily detected in untreated HepG2 cells but not in cells where C/EBP was downregulated either via RNAi knockdown or MAP kinase activation. A schematic overview of the results is presented in Fig. 9.
|
, one of the factors recruited to the HNF-4 enhancer, is depleted in MAP kinase-activated cells, together with those showing that ectopically expressed C/EBP can reverse HNF-4 repression, points to the pivotal role of C/EBP in the stability and formation of the higher-order enhancer-promoter complex. Although the exact interaction partner(s) of the enhancer-bound C/EBP at the HNF-4 proximal promoter is not known, we speculate that HNF-6 could be a good candidate. C/EBP has been shown to physically interact with HNF-6 and that this association is necessary for CBP recruitment to the HNF-3ß promoter (44). Alternatively, C/EBP may be necessary to hold together the components constituting the HNF-4 enhancer complex (HNF-1, HNF-3ß, CBP, and Brg1), which is indicated by the dissociation of all of these factors from the enhancer region during MAP kinase activation.
"Molecular bookmarks" decorate the HNF-4 regulatory region during the repressed state.
MAP kinase activation resulted in the dissociation of RNA pol-II, TFIIB, CBP, Brg-1, and HNF-3ß from the HNF-4 regulatory regions. Interestingly, however, HNF-6, HNF-1, and components of the TFIID complex remained associated with the proximal promoter during the repressed state. As a result of this, the HNF-4 regulatory region adopts a linear configuration in which the enhancer is cleared but the proximal promoter remains occupied by transcription factors (Fig. 9). Furthermore, the extent of acetylation of the surrounding nucleosomes did not decrease upon repression. We have shown previously that histone acetylation marks at the promoters and the coding regions of several hepatic genes, including the HNF-4 gene, remain stable during mitotic and -amanitin-mediated transcriptional inactivation (17). It was proposed that, together with other modifications, acetylation marks may provide means for the cells to "remember" the locations of the genome where transcription should resume after the cells exit mitosis, thus contributing to the faithful maintenance of cell-specific gene expression patterns (17). The stably associated factors and histone acetylation marks at the HNF-4 regulatory region during MAP kinase repression may have a similar "bookmarking" function. They could prevent the permanent silencing of the locus and provide an open state of the local chromatin, competent for rapid reassembly of the RNA pol-II machinery, once the repressive signal declines.
|
ACKNOWLEDGMENTS |
|---|
This study was supported by a fellowship from the Onassis Foundation to I.K. and grants from GSRT (PENED-01ED509 and PENED-03ED542) and EU (LSHG-CT-2004-502950 and MTKD-CT2005 029610).
|
FOOTNOTES |
|---|
P.H. and I.K. contributed equally to this study.
Present address: Hubrecht Laboratory, Netherlands Institute for Developmental Biology, Uppsalalaan 8, 3584 CT Utrecht, The Netherlands.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 promoter is activated by glucocorticoids and liver-enriched transcription factors. Nucleic Acids Res. 29:3495-3505.2. Cereghini, S. 1996. Liver-enriched transcription factors and hepatocyte differentiation. FASEB J. 10:267-282.[Abstract]
3. Chen, W. S., K. Manova, D. C. Weinstein, S. A. Duncan, A. S. Plump, V. R. Prezioso, R. F. Bachvarova, and J. E. Darnell, Jr. 1994. Disruption of the HNF-4 gene, expressed in visceral endoderm, leads to cell death in embryonic ectoderm and impaired gastrulation of mouse embryos. Genes Dev. 8:2466-2477.
4. Costa, R. H., V. V. Kalinichenko, A. X. Holterman, and X. Wang. 2003. Transcription factors in liver development, differentiation, and regeneration. Hepatology 38:1331-1347.[Medline]
5. De Fabiani, E., N. Mitro, A. C. Anzulovich, A. Pinelli, G. Galli, and M. Crestani. 2001. The negative effects of bile acids and tumor necrosis factor-alpha on the transcription of cholesterol 7-hydroxylase gene (CYP7A1) converge to hepatic nuclear factor-4: a novel mechanism of feedback regulation of bile acid synthesis mediated by nuclear receptors. J. Biol. Chem. 276:30708-30716.
6. Dell, H., and M. Hadzopoulou-Cladaras. 1999. CREB-binding protein is a transcriptional coactivator for hepatocyte nuclear factor-4 and enhances apolipoprotein gene expression. J. Biol. Chem. 274:9013-9021.
7. Drewes, T., S. Senkel, B. Holewa, and G. U. Ryffel. 1996. Human hepatocyte nuclear factor 4 isoforms are encoded by distinct and differentially expressed genes. Mol. Cell. Biol. 16:925-931.[Abstract]
8. Duncan, S. A., A. Nagy, and W. Chan. 1997. Murine gastrulation requires HNF-4 regulated gene expression in the visceral endoderm: tetraploid rescue of Hnf-4–/– embryos. Development 124:279-287.[Abstract]
9. Duncan, S. A., M. A. Navas, D. Dufort, J. Rossant, and M. Stoffel. 1998. Regulation of a transcription factor network required for differentiation and metabolism. Science 281:692-695.
10. Hatzis, P., and I. Talianidis. 2001. Regulatory mechanisms controlling human hepatocyte nuclear factor 4 gene expression. Mol. Cell. Biol. 21:7320-7330.
11. Hatzis, P., and I. Talianidis. 2002. Dynamics of enhancer-promoter communication during differentiation-induced gene activation. Mol. Cell 10:1467-1477.[CrossRef][Medline]
12. Hayhurst, G. P., Y. H. Lee, G. Lambert, J. M. Ward, and F. J. Gonzalez. 2001. Hepatocyte nuclear factor 4 (nuclear receptor 2A1) is essential for maintenance of hepatic gene expression and lipid homeostasis. Mol. Cell. Biol. 21:1393-1403.
13. Jiang, G., L. Nepomuceno, Q. Yang, and F. M. Sladek. 1997. Serine/threonine phosphorylation of orphan receptor hepatocyte nuclear factor 4. Arch. Biochem. Biophys. 340:1-9.[CrossRef][Medline]
14. Johnson, P. F. 2005. Molecular stop signs: regulation of cell-cycle arrest by C/EBP transcription factors. J. Cell Sci. 118:2545-2555.
15. Kolch, W. 2005. Coordinating ERK/MAPK signalling through scaffolds and inhibitors. Nat. Rev. Mol. Cell. Biol. 6:827-837.[CrossRef][Medline]
16. Kouskouti, A., E. Scheer, A. Staub, L. Tora, and I. Talianidis. 2004. Gene-specific modulation of TAF10 function by SET9-mediated methylation. Mol. Cell 14:175-182.[CrossRef][Medline]
17. Kouskouti, A., and I. Talianidis. 2005. Histone modifications defining active genes persist after transcriptional and mitotic inactivation. EMBO J. 24:347-357.[CrossRef][Medline]
18. Kritis, A. A., A. Argyrokastritis, N. K. Moschonas, S. Power, N. Katrakili, V. I. Zannis, S. Cereghini, and I. Talianidis. 1996. Isolation and characterization of a third isoform of human hepatocyte nuclear factor 4. Gene 173:275-280.[CrossRef][Medline]
19. Ktistaki, E., N. T. Ktistakis, E. Papadogeorgaki, and I. Talianidis. 1995. Recruitment of hepatocyte nuclear factor 4 into specific intranuclear compartments depends on tyrosine phosphorylation that affects its DNA-binding and transactivation potential. Proc. Natl. Acad. Sci. USA 92:9876-9880.
20. Ktistaki, E., and I. Talianidis. 1997. Modulation of hepatic gene expression by hepatocyte nuclear factor 1. Science 277:109-112.
21. Kuo, C. J., P. B. Conley, L. Chen, F. M. Sladek, J. E. Darnell, Jr., and G. R. Crabtree. 1992. A transcriptional hierarchy involved in mammalian cell-type specification. Nature 355:457-461.[CrossRef][Medline]
22. Li, J., G. Ning, and S. A. Duncan. 2000. Mammalian hepatocyte differentiation requires the transcription factor HNF-4. Genes Dev. 14:464-474.
23. Miller, M., J. D. Shuman, T. Sebastian, Z. Dauter, and P. F. Johnson. 2003. Structural basis for DNA recognition by the basic region leucine zipper transcription factor CCAAT/enhancer-binding protein alpha. J. Biol. Chem. 278:15178-15184.
24. Odom, D. T., N. Zizlsperger, D. B. Gordon, G. W. Bell, N. J. Rinaldi, H. L. Murray, T. L. Volkert, J. Schreiber, P. A. Rolfe, D. K. Gifford, E. Fraenkel, G. I. Bell, and R. A. Young. 2004. Control of pancreas and liver gene expression by HNF transcription factors. Science 303:1378-1381.
25. Parviz, F., C. Matullo, W. D. Garrison, L. Savatski, J. W. Adamson, G. Ning, K. H. Kaestner, J. M. Rossi, K. S. Zaret, and S. A. Duncan. 2003. Hepatocyte nuclear factor 4 controls the development of a hepatic epithelium and liver morphogenesis. Nat. Genet. 34:292-296.[CrossRef][Medline]
26. Pearson, G., F. Robinson, T. Beers Gibson, B. E. Xu, M. Karandikar, K. Berman, and M. H. Cobb. 2001. Mitogen-activated protein (MAP) kinase pathways: regulation and physiological functions. Endocrinol. Rev. 22:153-183.
27. Popowski, K., J. J. Eloranta, M. Saborowski, M. Fried, P. J. Meier, and G. A. Kullak-Ublick. 2005. The human organic anion transporter 2 gene is transactivated by hepatocyte nuclear factor-4 alpha and suppressed by bile acids. Mol. Pharmacol. 67:1629-1638.
28. Prusty, D., B. H. Park, K. E. Davis, and S. R. Farmer. 2002. Activation of MEK/ERK signaling promotes adipogenesis by enhancing peroxisome proliferator-activated receptor gamma (PPAR
) and C/EBP gene expression during the differentiation of 3T3-L1 preadipocytes. J. Biol. Chem. 277:46226-46232.
29. Ross, S. E., H. S. Radomska, B. Wu, P. Zhang, J. N. Winnay, L. Bajnok, W. S. Wright, F. Schaufele, D. G. Tenen, and O. A. MacDougald. 2004. Phosphorylation of C/EBP inhibits granulopoiesis. Mol. Cell. Biol. 24:675-686.
30. Rubins, N. E., J. R. Friedman, P. P. Le, L. Zhang, J. Brestelli, and K. H. Kaestner. 2005. Transcriptional networks in the liver: hepatocyte nuclear factor 6 function is largely independent of Foxa2. Mol. Cell. Biol. 25:7069-7077.
31. Sladek, F. M., M. D. Ruse, Jr., L. Nepomuceno, S. M. Huang, and M. R. Stallcup. 1999. Modulation of transcriptional activation and coactivator interaction by a splicing variation in the F domain of nuclear receptor hepatocyte nuclear factor 4alpha1. Mol. Cell. Biol. 19:6509-6522.
32. Slomiany, B. A., K. L. D'Arigo, M. M. Kelly, and D. T. Kurtz. 2000. C/EBP inhibits cell growth via direct repression of E2F-DP-mediated transcription. Mol. Cell. Biol. 20:5986-5997.
33. Soutoglou, E., N. Katrakili, and I. Talianidis. 2000. Acetylation regulates transcription factor activity at multiple levels. Mol. Cell 5:745-751.[CrossRef][Medline]
34. Soutoglou, E., and I. Talianidis. 2002. Coordination of PIC assembly and chromatin remodeling during differentiation-induced gene activation. Science 295:1901-1904.
35. Soutoglou, E., B. Viollet, M. Vaxillaire, M. Yaniv, M. Pontoglio, and I. Talianidis. 2001. Transcription factor-dependent regulation of CBP and P/CAF histone acetyltransferase activity. EMBO J. 20:1984-1992.[CrossRef][Medline]
36. Timchenko, N. A., T. E. Harris, M. Wilde, T. A. Bilyeu, B. L. Burgess-Beusse, M. J. Finegold, and G. J. Darlington. 1997. CCAAT/enhancer binding protein alpha regulates p21 protein and hepatocyte proliferation in newborn mice. Mol. Cell. Biol. 17:7353-7361.[Abstract]
37. Timchenko, N. A., M. Wilde, M. Nakanishi, J. R. Smith, and G. J. Darlington. 1996. CCAAT/enhancer-binding protein alpha (C/EBP) inhibits cell proliferation through the p21 (WAF-1/CIP-1/SDI-1) protein. Genes Dev. 10:804-815.
38. Tolhuis, B., R. J. Palstra, E. Splinter, F. Grosveld, and W. de Laat. 2002. Looping and interaction between hypersensitive sites in the active beta-globin locus. Mol. Cell 10:1453-1465.[CrossRef][Medline]
39. Viollet, B., A. Kahn, and M. Raymondjean. 1997. Protein kinase A-dependent phosphorylation modulates DNA-binding activity of hepatocyte nuclear factor 4. Mol. Cell. Biol. 17:4208-4219.[Abstract]
40. Wang, G. L., P. Iakova, M. Wilde, S. Awad, and N. A. Timchenko. 2004. Liver tumors escape negative control of proliferation via PI3K/Akt-mediated block of C/EBP growth inhibitory activity. Genes Dev. 18:912-925.
41. Wang, G. L., and N. A. Timchenko. 2005. Dephosphorylated C/EBP accelerates cell proliferation through sequestering retinoblastoma protein. Mol. Cell. Biol. 25:1325-1338.
42. Wang, H., P. Iakova, M. Wilde, A. Welm, T. Goode, W. J. Roesler, and N. A. Timchenko. 2001. C/EBP arrests cell proliferation through direct inhibition of Cdk2 and Cdk4. Mol. Cell 8:817-828.[CrossRef][Medline]
43. Watt, A. J., W. D. Garrison, and S. A. Duncan. 2003. HNF4: a central regulator of hepatocyte differentiation and function. Hepatology 37:1249-1253.[CrossRef][Medline]
44. Yoshida, Y., D. E. Hughes, F. M. Rausa III, I. M. Kim, Y. Tan, G. J. Darlington, and R. H. Costa. 2006. C/EBP and HNF6 protein complex formation stimulates HNF6-dependent transcription by CBP coactivator recruitment in HepG2 cells. Hepatology 43:276-286.[CrossRef][Medline]
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
