HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 275, Issue 23, 17814-17820, June 9, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/23/17814    most recent M909842199v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Yeagley, D. Articles by Quinn, P. G. Search for Related Content PubMed PubMed Citation Articles by Yeagley, D. Articles by Quinn, P. G. Social Bookmarking                      What's this?

Characterization of Elements Mediating Regulation of Phosphoenolpyruvate Carboxykinase Gene Transcription by Protein Kinase A and Insulin

IDENTIFICATION OF A DISTINCT COMPLEX FORMED IN CELLS THAT MEDIATE INSULIN INHIBITION^*

David Yeagley, Jonathan Moll^§, Charles A. Vinson^§, and Patrick G. Quinn^¶

From the  Department of Cellular and Molecular Physiology, The Pennsylvania State University College of Medicine, Hershey, Pennsylvania 17033 and the ^§ Laboratory of Metabolism, NCI, National Institutes of Health, Bethesda, Maryland 20892

Received for publication, December 13, 1999, and in revised form, March 13, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The in vivo pattern of induction of phosphoenolpyruvate carboxykinase (PEPCK) gene transcription by cAMP and its inhibition by insulin is reproduced in H4IIe cells and is mediated by a bipartite cAMP/insulin response unit (C/IRU) consisting of a cAMP response element (95/87) and an upstream enhancer, AC (271/225). Studies in HepG2 cells showed that binding of AP-1 and CAAT/enhancer-binding protein (C/EBP) to AC is required for induction by cAMP, but insulin did not inhibit cAMP-induced PEPCK expression in HepG2 cells. Binding of H4IIe nuclear proteins to an AC element probe was inhibited by antibodies or a consensus site for C/EBP, but not AP-1. Transfection with dominant negative bZIP factors, which prevent endogenous factors from binding to DNA, showed that elimination of cAMP regulatory element-binding protein CREB or C/EBP activity blocked induction by protein kinase A (PKA), whereas elimination of AP-1 activity had no effect. In addition, promoters with multiple CREB sites, or a single CREB site and multiple C/EBP sites, mediated PKA induction, but this was inhibited to no greater extent than basal activity was by insulin. These results indicate that an AC factor other than C/EBP must mediate insulin inhibition. An A-site probe (265/247) or a probe across the middle of the AC element (256/237) competed for complexes formed by factors other than AP-1 or C/EBP. However, analysis of competitor oligonucleotides and antibodies for candidate factors failed to identify other factors. Scanning mutations throughout the AC element interfered with induction but allowed us to define five overlapping sites for regulatory factors in AC and to design probes binding just one or two factors. Comparison of the protein-DNA complexes formed on these smaller probes revealed that a specific complex present in rat liver and H4IIe cell nuclear extracts differed from those formed by HepG2 cell nuclear extracts. Our results suggest that multiple factors binding the AC element of the C/IRU interact with each other and CREB to regulate PEPCK induction by cAMP and inhibition by insulin and that the unique factor expressed in H4IIe cells is a candidate for involvement in insulin regulation of PKA-induced PEPCK gene transcription.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Phosphoenolpyruvate carboxykinase (PEPCK)¹ catalyzes a committed and rate-limiting step in hepatic gluconeogenesis, so its activity must be carefully controlled to maintain blood glucose levels within normal limits. The primary means of modulating PEPCK activity, which is proportional to the rate at which its gene is transcribed, is through hormonal control of PEPCK gene transcription (1). In particular, the pancreatic hormones, glucagon and insulin, induce and inhibit PEPCK gene transcription, respectively. Glucagon stimulates the synthesis of cAMP, cAMP activates protein kinase A (PKA), and PKA phosphorylates the PEPCK promoter-associated cAMP regulatory element-binding protein (CREB) on Ser¹³³, which is necessary but not sufficient for induction of PEPCK transcription (2). Insulin activates phosphatidylinositol 3-kinase, which activates downstream kinases that have been implicated in inhibition of PEPCK transcription, protein kinase B (3) and others (1, 4-6). However, the transcription factor(s) whose binding or activity is modified by insulin to inhibit glucagon/cAMP-induced PEPCK gene transcription has not been identified.

We recently showed that induction of PEPCK transcription in response to cAMP and inhibition in response to insulin can be reconstituted in H4IIe cells by two promoter elements that form a cAMP/insulin response unit (C/IRU) (2). Regulation of induction requires a CREB-binding site and an upstream enhancer, the AC element. The AC element has been characterized in HepG2 cells, where AP-1 and C/EBP bind to a 5' A-site and 3' C-site, respectively (7-10). We show here that insulin cannot inhibit induction by cAMP in HepG2 cells. In contrast, transcription from endogenous and exogenous PEPCK promoters in response to both cAMP and insulin is regulated in H4IIe cells, as in vivo (2, 5, 11). Therefore, we examined the potential roles of the A-site and C-site and of AP-1 and C/EBP factors in regulating PEPCK induction by cAMP and inhibition by insulin in H4IIe cells. The binding of AP-1 and C/EBP factors, which are abundantly expressed in H4IIe cells, to the AC element was evaluated with competitor oligonucleotides and factor-specific antibodies. The effects of expression of dominant negative mutants of CREB, C/EBP, and AP-1 factors on regulation of PEPCK gene transcription were determined to assess the physiological contributions of these factors.

Several factors are predicted to bind to AC, based on the similarity of their recognition sites to the enhancer sequence (13). Of particular interest among these were factors that are regulated by phosphorylation or are involved in liver-specific gene expression. We examined the role of several AP-1/CREB family members that are regulated by phosphorylation, including Jun, Fos, JunD, ATF2, and CREB. Several of the HNFs, which contribute to liver-specific expression of PEPCK and other genes (14-16), have potential binding sites within AC. A potential binding site for NF-Y was also identified, as were sites for HMG/SRY factors. Of particular interest was a potential site for FKHR, whose phosphorylation by protein kinase B has been demonstrated to be responsible for insulin inhibition of basal IGFBP-1 expression (17, 18). Involvement of these factors was tested by competition with consensus binding sites and/or antibodies against specific factors.

A series of scanning mutations throughout the AC enhancer were analyzed for their effects on function of a minimal PEPCK promoter (AC-G4T) that mediates regulation. All scanning mutations disrupted regulation, indicating that factors associated with several overlapping sites within AC are required for induction by PKA and inhibition of this induction by insulin in H4IIe cells, which recapitulate the pattern of hormonal regulation seen in vivo (1). Analysis of the competition by oligonucleotides containing scanning mutations allowed us to map the protein recognition sites within AC that are required for regulation and to design smaller probes, binding only one or two factors. These smaller probes were used to compare the binding of nuclear proteins derived from rat liver, H4IIe or HepG2 cells. We found that rat liver and H4IIe nuclear proteins formed a unique complex, not found with HepG2 nuclear proteins, on two related sites within AC. Because this complex forms only in cells in which insulin inhibits cAMP-induced PEPCK expression, the factor forming it is a good candidate for involvement in regulation by insulin.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

H4IIe Cell Culture and Transfection Analyses-- H4IIe cells were grown and transfected as described previously (19, 20). In brief, the cells were transfected in solution with 10 µg of luciferase (firefly) reporter plus 1 µg of each expression plasmid plus 1 µg of pRL-SV (Renilla luciferase; Promega Corp.) reporter to correct for differences in transfection efficiency. Half of the cells were seeded into each of two 60-mm dishes, one of which served as a control while the other was treated with 10 nM insulin. Where indicated, cells were cotransfected with an expression vector for the catalytic subunit of PKA, obtained from R. Maurer (Oregon Health Sciences University) (21). After 4 h, the cells were treated with 20% Me₂SO for 3 min and washed in phosphate-buffered saline, and then medium with or without 10 nM insulin was added for the remaining 20 h. Cells were harvested with trypsin/EDTA and lysed, and luciferase activities for the firefly and Renilla luciferase reporters were measured with the dual luciferase kit (Promega), using an ALL Monolight 3010 dual injector luminometer. Variations in PEPCK promoter-firefly luciferase activity caused by variations in transfection efficiency were corrected for by measuring Renilla luciferase activity in the same sample. Values were normalized for transfection efficiency, and the mean was computed for several experiments. Data shown in figures were obtained from independent transfection experiments performed with different preparations of the various plasmids. All figures represent several transfection experiments, each of which was normalized to the untreated control and the data combined for analysis. The number of experiments for each figure is indicated in the figure legends.

Expression Vectors-- The CRG (CREB-GAL4 fusion protein) expression vector contains the activation domain of CREB (amino acids 1-277) fused to amino acids 4-147 of the GAL4 DNA-binding domain and has been described previously (22). The PKA expression vector, RSV-C, contains the cDNA for the catalytic subunit of protein kinase A under control of the Rous sarcoma virus promoter (23). The A-ZIP series of expression vectors has been described before (24-26). The cDNAs encoding the A-ZIP factors were subcloned into an Rous sarcoma virus expression vector for these studies because the cytomegalovirus promoter of the original expression vector contains several CREs that would have confounded interpretation of these experiments.

Reporter Vectors-- The PQ-Luc luciferase reporter plasmid is based on the promoterless PGL3-basic vector encoding firefly luciferase, obtained from Promega Corp. and modified to accept PEPCK promoter fragments as described previously (8). The PEPCK and G4-PEPCK promoters were described previously (20), as were the 5XGT and AC-G4T promoters (2). Oligonucleotides for the A element or C element (see Fig. 2) containing HindIII ends were obtained from Life Technologies, Inc. and subcloned into AC-G4T-Luc. Oligonucleotides of AC element truncations and clustered point mutations containing HindIII ends were obtained from Life Technologies, Inc. and subcloned into AC-G4T-Luc and G4-PEPCK-Luc. Promoter fragments containing the desired mutation were sequenced in their entirety prior to transfection and mobility shift analysis.

Mobility Shift Experiments-- Nuclear extracts were prepared from H4IIe cells by a modification of the method of Hurst et al. (27), as described previously (2). The double-stranded AC element DNA probe contained HindIII ends (agcttCGGTCAAAGTTTAGTCAATCAAACGTTGTGTAAGGACTCt). The AC element probe was labeled by filling in the overhang with Klenow in the presence of [-³²P]dATP. 10 fmol of radiolabeled DNA probe was incubated with 5 µg of nuclear protein for 15 min at room temperature, as described previously (28). The reactions were electrophoresed in a 6% polyacrylamide gel (25 mM Tris, 190 mM glycine, 1 mM EDTA; 150V; 3.5 h), which was dried and analyzed by autoradiography. For competition studies, a 50-fold molar excess of unlabeled, duplex oligonucleotide was mixed with the radiolabeled probe prior to addition of nuclear protein. The competitors used were: AC, same as above; A-site, agcttCGGTCAAAGTTTAGTCAAt; C-site, agcttCAAACGTTGTGTAAGGACTCt; mAC, aggcttTTTAGTCAATCAAACGTTGTt; and the SM series, presented in Fig. 4A. The competitors for wild type FKHR (consensus sites are underlined and bold), CACTAGCAAAACAAACTTATTTTGAACAC, and its mutant, AmBm, CACTAGCCCCGGGAACTTAGGGGTAACAC, contained the duplicated FKHR sites of the IGFBP-1 promoter (17, 29). The sequences of other competitors were obtained by comparing factor-binding sites for various promoters with consensus sites in TF Search (13) to obtain the following binding sites: AP-1, AGCTTGCATGAGTCAGACTAGT (30, 31); C/EBP, AGCTTGCTTGCGCAACTAGT (30, 32, 33); CREB, TCGACCCCTGACGTCAGAGGCG (28, 32); HNF-1, GGTTAATGATTAACA (32, 34); HNF-3, AGCTTAAAGCAAACAT (30, 35); NF-Y, ACTGATTGGTTAGT, (36); SRY, GTTAACAATTGA, (37, 38). Several antibodies for supershift experiments were obtained in high concentration from Santa Cruz Biotechnology (c-Fos, sc-52x; c-Jun, sc-169x; C/EBP, sc-61x; C/EBP, sc-150x; C/EBP, sc151x; ATF2, sc-187x; SRY, sc-8235X; and JunD, sc-74x). Anti-CREB was provided by D. Ginty (Baltimore, MD), anti-HNF-3 was from R. Costa (Chicago, IL), anti-NF-Y was from R. Mantovani (Milan, Italy), and anti-HNF-6 was provided by F. LeMaigre (Brussels, Belgium).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Characterization of Regulation by the C/IRU or Its Component Parts-- As illustrated in Fig. 1, opposing regulation of PEPCK transcription by cAMP and insulin can be reconstituted with a minimal promoter (40/+1) fused to a binding site for CREB (supplied here as G4, which binds CREB-GAL4, CRG) and an enhancer located between nucleotides 271 and 230, the AC element. Expression of a phosphorylation defective CREB activation domain (CRG-S133A) failed to mediate induction, indicating that CREB phosphorylation is required for regulation. Transcription from a PEPCK promoter under the control of multiple CREB-binding sites (5XGT) provided modest induction, but insulin inhibited this induction to no greater extent than it inhibits basal expression, indicating that, although necessary for induction, CREB does not mediate insulin inhibition. In a similar manner, a promoter containing four copies of the C/EBP site in place of AC mediated modest induction but not insulin inhibition of this induction. Because the C/IRU mediates both induction and inhibition, a factor other than C/EBP, bound at the AC element, must mediate insulin inhibition of PKA-activated transcription.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Characterization of regulation by the intact C/IRU and its components. A, maps of the PEPCK promoters indicating the elements that are present. B, H4IIe cells were cotransfected with the indicated CRG expression vector and luciferase reporters, or with PEPCK-Luc, in the absence and presence of PKAc expression vector as described under "Experimental Procedures." Each precipitate was split into two dishes, and half of them were treated with 10 nM insulin for the final 20 h of the experiment. The results shown represent the mean ± S.E. of at least four independent experiments. C, calcium phosphate precipitates of PEPCK-Luc in the absence and presence of a PKAc expression vector were prepared and split between H4IIe and HepG2 cell suspensions. Each suspension was split into two dishes, one of which was treated with 10 nM insulin for the final 20 h of the experiment. The results shown represent the means ± S.E. of four independent experiments. Con, control; Ins, insulin.

Insulin Inhibits PKA-induced PEPCK Gene Transcription in H4IIe Cells but Not in HepG2 Cells-- We transfected both H4IIe and HepG2 cells with the same calcium phosphate precipitates, containing PEPCK-Luc /+, a PKA expression vector, and treated the cells with or without 10 nM insulin to assess hormonal regulation of the PEPCK promoter (Fig. 1C). In H4IIe cells, transcription of PEPCK-Luc was induced by cotransfection with PKA, and induction was inhibited by concomitant treatment of the cells with insulin. This result is consistent with previous studies of PEPCK gene regulation, using run-on transcription and transfection assays in the H4IIe cell line or determination of mRNA amount in livers from hormone-treated animals (39). In HepG2 cells, PEPCK-Luc was strongly induced by PKA, as previously reported (8). However, insulin failed to inhibit induction of PEPCK-Luc by PKA in these cells. Because insulin activates apparently identical signaling pathways in these cell lines (17), this result suggested that different factors may mediate regulation in the two cell lines.

C/EBP Binds the C-site, but Factors Distinct from AP-1 or C/EBP Bind the A-site and a Central Region of AC-- We examined the binding of H4IIe nuclear proteins to an AC element probe in electrophoretic mobility shift assays (EMSA) to determine the contribution of AP-1 and C/EBP factors to the pattern of protein-DNA interactions observed with H4IIe cell nuclear extracts. H4IIe nuclear proteins formed 11 discreet complexes with the labeled AC element in EMSA (Fig. 2). To address the nature of the protein-DNA complexes present in the EMSA, we evaluated the ability of oligonucleotides containing various consensus sequences or motifs within the AC element to compete for protein binding (Fig. 2). The AP-1 consensus sequence was ineffective as a competitor for specific complexes formed on the AC element. In contrast, either a C/EBP consensus site or the C-site competed for the same more rapidly migrating protein-DNA complexes (bands 4, 6, and 8-11). In agreement with these results, antibodies directed against Fos and Jun proteins were without effect upon protein-DNA complex formation (Table I). On the other hand, antibody specific for C/EBP supershifted protein-DNA complexes corresponding to bands 4 and 6, whereas antibody to C/EBP supershifted protein-DNA complexes corresponding to bands 4 and 8-11 (Table I). An A-site competitor abrogated formation of the most slowly migrating protein-DNA complexes (bands 1-3). Binding of proteins in complexes represented by bands 5 and 7 was not diminished by either the A-site or C-site competitor. In addition, band 7 was reduced by a competitor derived from the middle of the AC element (mAC), comprising the 3' half of the A-site and the 5' half of the C-site, indicating that this central site in AC may be necessary for PEPCK gene regulation.

View larger version (96K): [in this window] [in a new window]   Fig. 2.   Effects of binding site competitors on protein-DNA complex formation on an AC element probe. A, the sequence of the AC element is shown, together with the boundaries of the A, mAC, and C probes used in the competition studies below. Also shown are putative AP-1- and C/EBP-binding sites within the AC element. B, nuclear extracts were prepared from H4IIe cells and incubated with 10 fmol of radiolabeled AC element probe, in the absence or presence of the indicated competitor oligonucleotides. The binding reactions were electrophoresed on a 6% native polyacrylamide gel to produce this autoradiograph. Results of a typical experiment are shown. Similar results were obtained with independent preparations of nuclear extracts.

CREB and C/EBP, but Not AP-1, Are Required for Induction by PKA-- To investigate the functional contribution of AP-1 and C/EBP factors to the regulation of PEPCK transcription by cAMP and insulin, dominant negative variants of these factors were transfected into H4IIe cells. Isoforms of transcription factors that contain only the DNA-binding domain compete for factor binding, displacing the endogenous, full-length factor and abrogating regulation by it, as has been demonstrated for both CREB and C/EBP (40, 41). However, these experiments are confounded by the possibility that the overexpressed DNA-binding domain will displace other factors whose binding sites overlap with the targeted site. In response to this problem, we recently developed a series of dominant negative factors targeting the basic region of the leucine zipper (bZIP) family proteins, of which AP-1, C/EBP, and CREB are members (42). Proteins of the bZIP family dimerize through their leucine zipper domains and bind to the major groove of DNA through contacts within the adjacent basic regions, as described in the "scissors grip" model (43). The dominant negative A-ZIP factors contain an acidic domain complementary in charge distribution to the basic region of the factor targeted (24-26). As a result, when the A-ZIP factor dimerizes with a wild type factor to form a coiled coil through the leucine zipper region, the respective acidic and basic regions continue the formation of a very stable helical structure that engages the basic region of the wild type factor and prevents it from binding to DNA (24-26). In this way, the contribution of the endogenous factor to transcription regulation is removed without affecting other factors that may bind overlapping or adjacent sites in the regulatory region. As seen in Fig. 3, overexpression of the empty vector in which the A-ZIP factors are expressed had no effect upon the extent or pattern of regulation. However, expression of A-CREB or A-C/EBP essentially abolished induction by PKA. Expression of A-Fos had no effect upon hormonal regulation of PEPCK-luciferase. Thus, CREB and C/EBP, but not AP-1, are essential for induction by PKA.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Effects of cotransfection of dominant negative CREB, C/EBP, and AP-1 factors upon hormonal regulation mediated by the PEPCK promoter. H4IIe cells were cotransfected with the indicated A-ZIP expression vector(s) and a luciferase reporter gene under control of the complete PEPCK promoter in the absence and presence of a PKAc expression vector as described under "Experimental Procedures." Each precipitate was split into two dishes, and half of them were treated with 10 nM insulin for the final 20 h of the experiment. The results shown represent the means ± S.E. of six independent experiments. Con, control; Ins, insulin.

Analysis of Predicted Regulatory Sites in AC-- The AC element sequence was analyzed with the TFSEARCH program (13) to guide analysis of competition with consensus sites and antibodies against putative factors. We concentrated on examining the binding of factors that are involved in liver-enriched gene expression or are expressed in liver and known to be modified by phosphorylation. No further consideration was given to factors whose expression is restricted from liver, such as the caudal related factors (44). The potential involvement of AP-1, CREB, HNF, HMG/SRY, NF-Y, and FKHR factors was assessed by examining competition for complexes formed by H4IIe nuclear extracts on the AC element probe. Oligonucleotides encoding consensus binding sites for the factors listed in Table I were tested as competitors for binding to AC. Antibodies for factors recognizing AP-1 or cAMP response element sites (Fos, c-Jun, JunD, ATF2, and CREB), the CCAAT box (C/EBP and NF-Y), AT-rich sequences (the architectural factor SRY), or HNF sites (HNF-1, HNF-3, and HNF-6) were ineffective as competitors in displacing specific complexes formed on the AC element. All of the antibodies used here recognized their antigens on Western blots. Some had been demonstrated to supershift their antigens by other investigators (HNF-3, NF-Y and HNF-6; Refs. 45-47), and others were demonstrated to recognize native factors by immunoprecipitation (data not shown). This analysis eliminated the obvious choices for regulatory factors binding the AC element.

Definition of Overlapping Factor-binding Sites by Analysis of Scanning Mutations-- To analyze the binding sites within AC that are actually utilized by H4IIe nuclear extracts, an overlapping set of scanning mutations (4 nucleotides each) that covers the unmapped portion of the AC element up to the C/EBP-binding site was created and tested (Fig. 4A). Transition mutations (A  G, C  T) were made to affect protein-DNA interactions within the major groove while minimizing changes in DNA structure. When analyzed for function by transfection analysis, each of these mutated sequences abrogated PKA induction (Fig. 4B), indicating that the entire sequence is necessary for regulation.

View larger version (40K): [in this window] [in a new window]   Fig. 4.   Effect of AC scanning mutations on hormonal regulation. A, the sequence of the AC element is shown with the location of the scanning mutations (SM1-SM9) indicated in bold type. B, AC-G4T plasmids containing wild type AC or SM mutations in AC were cotransfected with PKA and treated with 10 nM insulin, as indicated. The means ± S.E. of six independent experiments are shown. Con, control; Ins, insulin.

Therefore, we turned our attention to analysis of competition of these mutated sequences for the formation of specific complexes on the AC probe (Fig. 5A). EMSA analysis using these mutated sequences as competitors for binding to an AC probe, together with analysis of binding to the A and mAC probes (Fig. 5B), allowed for the definition of four overlapping binding sites, in addition to the C/EBP site (Fig. 5C). In the SM competition assay (Fig. 5A), ineffective competition results in appearance of the complexes that would ordinarily form on the site. This is most easily seen for SM9, which disrupts the C/EBP site, resulting in the appearance of complexes 4, 6, and 8-11, all of which have been shown to bind C/EBP isoforms (Fig. 2). Analysis of competition for a labeled AC probe showed that SM 2-4 and 6-7 failed to compete for bands 1-3. In addition, complex 7 was seen weakly in the presence of SM1-4 and more strongly in SM 5-8. SM 5-8 also failed to compete for complex 5. 

View larger version (54K): [in this window] [in a new window]   Fig. 5.   Effect of AC scanning mutation competitors on binding to an AC probe. A, nuclear extracts were prepared from H4IIe cells and incubated with 10 fmol of radiolabeled AC element probe, in the absence or presence of a 50-fold excess of the indicated competitor oligonucleotides. The binding reactions were electrophoresed on a 6% native polyacrylamide gel to produce this autoradiograph. Results of a typical experiment are shown. Similar results were obtained with independent preparations of nuclear extracts. B, 10 fmol of oligonucleotide probe, either the A oligo (Fig. 2A, nucleotides 1-19) or the mAC oligo (Fig. 2A, nucleotides 10-29), were incubated with H4IIe nuclear extract in the absence or presence of a 50-fold excess of the indicated unlabeled oligonucleotide. The binding reactions were electrophoresed on a 6% native polyacrylamide gel to produce this autoradiograph. Results of a typical experiment are shown. Similar results were obtained with independent preparations of nuclear extracts. C, a composite representation of the data in A and B illustrates the limits of binding sites defined by competition with SM oligonucleotides. nt, nucleotides.

To refine this analysis, we labeled the A and mAC probes (nucleotides 1-19 and 10-29, respectively, as shown in Fig. 2) to determine which complexes could form independently on these probes and to determine whether they would exhibit cross-competition. Complexes 1-3 and 7 formed on the A probe, whereas complexes 5 and 7 formed on the mAC probe (Fig. 5B). Either probe could compete for complex 7 on the alternate probe, but A did not compete for complex 5 and mAC did not compete for complexes 1-3, demonstrating the specificity of competition. The results of this analysis, which were confirmed by additional competition experiments, are summarized in Fig. 5C. Complexes 1-3 form on a site that can be interrupted by SM5, rather than forming on related sites, because they form on A but not on mAC. Complex 7 forms on related sites within A and mAC, as shown by its ability to form on either probe and its sensitivity to cross-competition with these probes. Both competition and direct binding experiments indicated that the site in mAC binds the factor forming complex 7 more avidly than the site in A. In contrast, complex 5 forms on an mAC site that cannot be distinguished with these mutations from the complex 7 site, but no formation of complex 5 was detected on the A-site.

A Factor Unique to H4IIe Cells and Rat Liver Forms Complex 7-- To determine which of the AC-binding factors is most likely to be involved in insulin regulation, we compared the binding of nuclear extracts derived from rat liver, H4IIe or HepG2 cells (Fig. 6). PEPCK transcription is induced by cAMP in all of these cells, but PKA-induced PEPCK transcription is inhibited by insulin only in H4IIe cells and rat liver. The formation of complexes 1-3 and 5 and all C/EBP-containing complexes (bands 4, 6, and 8-11) was indistinguishable between H4IIe and HepG2 nuclear extracts (data not shown). However, the formation of complex 7 appeared to be different and was examined in more detail with smaller probes identified by analysis of SM competition: nucleotides 1-13 (disrupted in SM 1-4), which bind complex 7, and nucleotides 13-25 (disrupted in SM 5-8), which bind complexes 5 and 7. Rat liver nuclear extracts were also included in these assays for comparison. EMSA analysis of factors binding a probe encompassing nucleotides 1-13 showed that complex 7 was formed in H4IIe and rat liver nuclear extracts, whereas more quickly and more slowly migrating complexes were formed by HepG2 nuclear extracts (Fig. 6A). When binding to a probe encompassing nucleotides 13-25 was analyzed, an apparently identical complex 5 was formed by H4IIe, HepG2, and rat liver nuclear extracts. However, the factor forming complex 7 was unique to H4IIe cells and rat liver, which mediate insulin inhibition of cAMP-induced PEPCK expression. Again, more quickly and more slowly migrating complexes were formed by HepG2 nuclear extracts. The complex 7 bands that formed on the two probes by nuclear extracts from H4IIe cells or rat liver were apparently identical, but the complex formed on nucleotides 13-25 was stronger than that formed on the nucleotides 1-13 probe, which is consistent with nucleotides 13-25 being a stronger competitor for complex 7. The complexes that formed on these related sites by factors in nuclear extracts from HepG2 cells were also apparently identical, but they differed from the complex 7 formed by H4IIe cells and rat liver. Because insulin inhibits PKA-induced PEPCK transcription in H4IIe cells and in vivo, but not in HepG2 cells, this unique factor forming complex 7 is a good candidate to be involved in the inhibitory effect of insulin.

View larger version (27K): [in this window] [in a new window]   Fig. 6.   Differences in factor binding in H4IIe, HepG2, and rat liver nuclear extracts. A, nuclear extracts were prepared from H4IIe cells, HepG2 cells, or rat liver and incubated with 10 fmol of either radiolabeled 1-13 or 13-25 probe, as indicated. The binding reactions were electrophoresed on a 6% native polyacrylamide gel to produce this autoradiograph. Results of a typical experiment are shown. B, schematic representation of factors binding to nucleotides 1-13 and 13-25 in H4IIe cells and rat liver, as presented in A. Related sequence motifs in the two probes are underlined. nt, nucleotides; NE, nuclear extract.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The results presented here indicate that the AC element of the PEPCK gene binds C/EBP and other factors that interact with CREB to mediate induction of transcription by PKA and inhibition by insulin. Hormonal regulation of this minimal PEPCK promoter containing the C/IRU reproduces regulation of the endogenous PEPCK gene in vivo (1, 2, 5, 11). HepG2 cells utilize AP-1 and C/EBP as accessory factors with CREB to mediate induction by PKA (7-10). Thus, the demonstration that insulin inhibits PKA induction of PEPCK-Luc in H4IIe cells, but not in HepG2 cells, suggested that factors other than AP-1 or C/EBP are required to mediate insulin inhibition. Both functional and DNA binding studies indicate that AP-1 factors are not involved in regulation in H4IIe cells and that C/EBP factors play a role in induction of PEPCK by PKA but not in insulin inhibition. Our competition studies provide evidence that factors other than AP-1 or C/EBP bind to 5' and internal sites in AC. Given that the C/IRU mediates insulin inhibition of induction and that neither CREB nor C/EBP can account for this, these other factors must play a role in insulin inhibition through the C/IRU. A likely candidate for an insulin regulatory factor is the unique factor (complex 7) present in H4IIe and rat liver cells, in which insulin inhibits PKA-activated PEPCK transcription, but not in HepG2 cells, where insulin has no such effect.

We previously demonstrated that the AC element and a CREB-binding site comprise a C/IRU for the PEPCK gene that mediates both induction by cAMP and inhibition by insulin in H4IIe cells (2). Because the AC element had been reported to mediate induction in HepG2 cells through AP-1 and C/EBP sites and because these factors are expressed in liver and regulated by protein kinases, we investigated their role in mediating opposing regulation of the PEPCK gene by cAMP and insulin in H4IIe cells. We found no evidence for the involvement of AP-1 factors in regulation by PKA or insulin through the AC element in H4IIe cells. Binding of H4IIe nuclear factors was unaffected by competition with a consensus AP-1 site or antibodies recognizing various AP-1 factors (Fig. 2 and Table I). In addition, expression of wild type or phosphorylation-deficient forms of Jun, either alone or with c-Fos, had no effect upon regulation (data not shown). Most importantly, expression of dominant negative AP-1 (A-Fos) had no effect upon regulation of the PEPCK promoter by cAMP or insulin. This is in contrast to the result obtained in HepG2 cells, where A-Fos abrogated induction by PKA as effectively as A-CREB or A-C/EBP (10), which again indicates that different factors are involved in regulation in HepG2 and H4IIe cells. On the other hand, our data on the role of C/EBP in regulation of induction by cAMP in H4IIe cells are in agreement with those from the HepG2 studies. Either C/EBP or C/EBP can interact with CREB to mediate regulation of the PEPCK gene by cAMP in HepG2 cells (8, 9, 48). Both C/EBP and C/EBP isoforms are present in H4IIe nuclear extracts and bind to the 3' site in the AC element, as judged by competition and antibody supershifting EMSA experiments (Fig. 2 and Table I). Most significantly, interference with C/EBP binding by expression of A-C/EBP drastically reduced induction by PKA. Although multiple C-sites augmented induction by PKA, this was inhibited to no greater extent than basal expression by insulin, as with the CREB-dependent 5XGT promoter (Fig. 1) (2). Insulin can promote dephosphorylation of C/EBP factors in adipocytes (12, 49). However, neither insulin nor cAMP had any effect on the phosphorylation of either C/EBP or C/EBP in H4IIe cells (data not shown). Thus, phosphorylation and modification of C/EBP activity is not required for it to contribute to induction, and it has no role in inhibition. Overall, the functional and binding data suggest that C/EBP and/or C/EBP transcription factors bind the AC element and interact with CREB to confer induction of the PEPCK gene by cAMP, but not inhibition by insulin. Although disruption of 5' or central sequence elements by scanning mutagenesis abolished induction by PKA, the entire AC element can be inverted without loss of regulation by PKA and insulin (data not shown). Thus, the A- and C-sites do not operate independently, but the AC element acts as a classical enhancer, in concert with a CREB-binding site, to provide regulation by both PKA and insulin when linked to a minimal PEPCK promoter.

Given that CREB and C/EBP cannot mediate insulin inhibition, the formation of other complexes on the AC element by H4IIe nuclear proteins is highly significant. The results presented here demonstrate that the AC element of the PEPCK C/IRU contains five overlapping sites for transcription factors. All of these sites are required for function of the enhancer element, indicating that the interaction of regulatory factors bound throughout the AC element is required for regulation. Complexes 1-3 form on the A-site but do not involve AP-1 factors. Complexes 4 and 6 contained C/EBP, whereas complexes 4 and 8-11 contained C/EBP. Clearly, only one of these C/EBP complexes will form on the single site in AC in vivo. In addition, factors bound within AC will undoubtedly influence each other's binding and stabilize specific complexes. It is highly significant that the complexes represented by bands 5 and 7 form on the center of the AC element, which cannot be interrupted, and do not bind AP-1 or C/EBP factors. In addition, one or more factors forming complexes 1-3 on the A-site is also required for regulation. Given that neither CREB nor C/EBP mediates insulin inhibition, it is likely that one or more of these other factors play an essential role in insulin regulation through the C/IRU.

All but one of the complexes formed on the AC enhancer were apparently identical in nuclear extracts derived from H4IIe or HepG2 cells, indicating that many of the same factors are utilized to regulate induction by cAMP/PKA in these cells. The critical difference in transcription factor utilization is that AP-1 was reported to be essential for regulation in HepG2 cells, but AP-1 is not required in H4IIe cells (10). In contrast, the factor forming complex 7 is uniquely expressed in H4IIe cells, which mediate insulin inhibition of PKA-induced PEPCK expression. We identified two related sites for complex 7 within the AC enhancer (Fig. 6B). By inference from the binding and competition studies, i.e. complex 7 binds nucleotides 1-13, and this is disrupted most strongly by SM 2 or 3, the complex 7-binding site within nucleotides 1-13 must contain at least TCAAAGT. Likewise, the site in nucleotides 13-25 must contain at least CAATCAA. When flanking sequences are considered, a common site of C(G/A)(G/A)TCAAA(G/C)(T/G)T emerges that is most closely related to a homeodomain recognition site. TFSEARCH (13) predicts that cdx factors should recognize this site, but to date the expression of cdx has been shown to be restricted to nonhepatic tissues (44). Alternatively, the cut/homeodomain factor HNF-6 was reported recently to bind the AC element (47), but antibody against HNF-6 did not supershift any of the complexes formed by H4IIe nuclear extracts. Thus, a unique homeodomain factor or a related factor may be predicted to form complex 7.

Based on the studies presented here and in our earlier work (2), we postulate a model involving separate mechanisms for insulin inhibition of basal and PKA-induced PEPCK gene transcription. Insulin efficiently inhibits cAMP-induced expression of AC-G4T but cannot effectively inhibit induction mediated by multiple copies of C/EBP plus CREB or by multiple copies of CREB alone. Thus, our data indicate that insulin inhibits cAMP/PKA-induction of PEPCK gene transcription mediated by the C/IRU by a mechanism distinct from that employed to inhibit basal gene expression and promiscuous induction mediated by fragments of the C/IRU (5XGT, 4C-GT; Fig. 1 and Ref. 2). The fact that insulin inhibits basal activity and that stimulated by CREB/CEBP equally suggests that this effect is exerted on the general factors of the RNA polymerase II complex, because this inhibition is independent of the nature of the activator. In contrast, insulin inhibition of cAMP-induced, C/IRU-mediated induction is robust, suggesting that insulin specifically neutralizes or reverses the activity of one or more C/IRU factors, independently of any effects on the polymerase complex. Our results suggest that this factor may very well be the factor forming complex 7, which is unique to cells in which insulin inhibits induction of PEPCK expression by cAMP, H4IIe, and rat liver. The nature of this factor and its role in regulation by insulin will have to await further biochemical and genetic characterization.

	ACKNOWLEDGEMENTS

We thank Jennifer Smith for technical assistance; M. Fried for advice on EMSA; R. Maurer (Oregon Health Sciences University), J. Holt (Vanderbilt University), S. McKnight (University of Texas Health Science Center), and J. Woodgett (University of Montreal) for plasmids; and R. Costa (University of Illinois), R. Mantovani (University of Milan), D. Ginty (Johns Hopkins University), and F. LeMaigre (Université Catholique de Louvain) for antibodies.

	FOOTNOTES

^* This work was supported by NIDDK, National Institutes of Health Grant R01 43871.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ To whom correspondence should be addressed: Dept. of Cellular and Molecular Physiology, H166, Pennsylvania State University, College of Medicine, 500 University Dr., Hershey, PA 17033. Tel.: 717-531-6182; Fax: 717-531-7667; E-mail: pquinn@psu.edu.

Published, JBC Papers in Press, March 28, 2000, DOI 10.1074/jbc.M909842199

	ABBREVIATIONS

The abbreviations used are: PEPCK, phosphoenolpyruvate carboxykinase; PKA, protein kinase A; CREB, cAMP regulatory element-binding protein; C/EBP, CAAT enhancer-binding protein; C/IRU, cAMP/insulin response unit; EMSA, electrophoretic mobility shift assay(s).

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				W.-S. Au, L. Lu, C.-M. Yeung, C.-C. Liu, O. G Wong, L. Lai, H.-f. Kung, and M. C Lin Hepatocyte nuclear factor 1 binding element within the promoter of microsomal triglyceride transfer protein (MTTP) gene is crucial for MTTP basal expression and insulin responsiveness J. Mol. Endocrinol., October 1, 2008; 41(4): 229 - 238. [Abstract] [Full Text] [PDF]

				B Elo, C M Villano, D Govorko, and L A White Larval zebrafish as a model for glucose metabolism: expression of phosphoenolpyruvate carboxykinase as a marker for exposure to anti-diabetic compounds J. Mol. Endocrinol., April 1, 2007; 38(4): 433 - 440. [Abstract] [Full Text] [PDF]

				A. V. Cybulsky, T. Takano, J. Papillon, W. Hao, A. Mancini, J. A. Di Battista, and M. I. Cybulsky The 3'-untranslated region of the Ste20-like kinase SLK regulates SLK expression Am J Physiol Renal Physiol, February 1, 2007; 292(2): F845 - F852. [Abstract] [Full Text] [PDF]

				D. Yeagley and P. G. Quinn 3',5'-Cyclic Adenosine Monophosphate Response Element-Binding Protein and CCAAT Enhancer-Binding Protein Are Dispensable for Insulin Inhibition of Phosphoenolpyruvate Carboxykinase Transcription and for Its Synergistic Induction by Protein Kinase A and Glucocorticoids Mol. Endocrinol., April 1, 2005; 19(4): 913 - 924. [Abstract] [Full Text] [PDF]

				M. E. Scassa, A. S. Guberman, J. M. Ceruti, and E. T. Canepa Hepatic Nuclear Factor 3 and Nuclear Factor 1 Regulate 5-Aminolevulinate Synthase Gene Expression and Are Involved in Insulin Repression J. Biol. Chem., July 2, 2004; 279(27): 28082 - 28092. [Abstract] [Full Text] [PDF]

				C. Anderwald, G. Muller, G. Koca, C. Furnsinn, W. Waldhausl, and M. Roden Short-Term Leptin-Dependent Inhibition of Hepatic Gluconeogenesis Is Mediated by Insulin Receptor Substrate-2 Mol. Endocrinol., July 1, 2002; 16(7): 1612 - 1628. [Abstract] [Full Text] [PDF]

				L. A. Hornbuckle, D. S. Edgerton, J. E. Ayala, C. A. Svitek, J. K. Oeser, D. W. Neal, S. Cardin, A. D. Cherrington, and R. M. O'Brien Selective tonic inhibition of G-6-Pase catalytic subunit, but not G-6-P transporter, gene expression by insulin in vivo Am J Physiol Endocrinol Metab, October 1, 2001; 281(4): E713 - E725. [Abstract] [Full Text] [PDF]

				D. Schmoll, K. S. Walker, D. R. Alessi, R. Grempler, A. Burchell, S. Guo, R. Walther, and T. G. Unterman Regulation of Glucose-6-phosphatase Gene Expression by Protein Kinase Balpha and the Forkhead Transcription Factor FKHR. EVIDENCE FOR INSULIN RESPONSE UNIT-DEPENDENT AND -INDEPENDENT EFFECTS OF INSULIN ON PROMOTER ACTIVITY J. Biol. Chem., November 10, 2000; 275(46): 36324 - 36333. [Abstract] [Full Text] [PDF]

A. K. Ghosh, R. Lacson, P. Liu, S. B. Cichy, A. Danilkovich, S. Guo, and T. G. Unterman A Nucleoprotein Complex containing CCAAT/Enhancer-binding Protein beta Interacts with an Insulin Response Sequence in the Insulin-like Growth Factor-binding Protein-1 Gene and Contributes to Insulin-regulated Gene Expression J. Biol. Chem., March 9, 2001; 276(11): 8507 - 8515. [Abstract] [Full Text] [PDF]