Journal of Nutrition OpenSOurce Diets- www.ResearchDiets.com

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Savon, S. P.
Right arrow Articles by Hanson, R. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Savon, S. P.
Right arrow Articles by Hanson, R. W.

The Journal of Nutrition Vol. 127 No. 2 February 1997, pp. 276-285
Copyright ©1997 by the American Society for Nutritional Sciences

The Promoter Regulatory Regions of the Genes for the Cytosolic Form of Phosphoenolpyruvate Carboxykinase (GTP) from the Chicken and the Rat Have Different Species-Specific Roles in Gluconeogenesis1,2

Summer P. Savon, Parvin Hakimi3, Deborah R. Crawford3, Dwight J. Klemm, Austin L. Gurney, and Richard W. Hanson4

Department of Biochemistry, Case Western Reserve University School of Medicine, Cleveland, OH 44106-4935

ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
FOOTNOTES
LITERATURE CITED

ABSTRACT

Hepatic expression of the gene for phosphoenolpyruvate carboxykinase (GTP) (PEPCK-C) (EC 4.1.1.32) in birds occurs prior to birth and decreases to negligible levels before hatching, whereas in mammals the gene for PEPCK-C in the liver is expressed at birth and is active throughout the life of the animal. The administration of cyclic AMP to adult chickens results in the induction of transcription of the gene for PEPCK-C and the transient accumulation of PEPCK-C mRNA in the liver. DNase I footprint analysis of 330 bp of the avian PEPCK-C promoter immediately 5' of the start-site of transcription indicated the presence of several protein binding domains, purified CAAT/enhancer binding protein alpha , cAMP regulatory element binding protein and nuclear factor-1 bound to these regions of the promoter. Sequences corresponding to an hepatic nuclear factor-1 binding domain and to the insulin response sequence, previously identified in the rat PEPCK-C promoter, were also found in the chicken PEPCK-C promoter. Co-transfection of an expression vector for CAAT/enhancer binding protein alpha  or CAAT/enhancer binding protein beta  markedly stimulated transcription from both the chicken and rat PEPCK-C promoters in human hepatoma cells. Sequences involved in the regulation of gene transcription by cyclic AMP and insulin were found to reside between -210 and +1 of the avian PEPCK-C promoter. In general, transcription from the avian promoter was more sensitive to inhibition by insulin than was noted for the rat PEPCK-C promoter, which may explain in part the lack of expression of the gene for PEPCK-C in the livers of adult birds.

Key words: phosphoenolpyruvate carboxykinase, birds, transcription factors, insulin, cAMP.

INTRODUCTION

There are major differences in the regulation of gluconeogenesis in birds compared with mammals, which are due in part to the unique features of P-enolpyruvate carboxykinase (PEPCK)5 gene expression in birds (Hod et al. 1986, Shen and Mistry 1978a and 1978b, Watford et al. 1981). Because the gene for the cytosolic form of PEPCK (PEPCK-C) in mammals is subjected to a high degree of regulation (Hanson and Patel 1994, Lamers et al. 1982, Magnunson et al. 1987, McGrane et al. 1988), we have studied the transcription of the gene for PEPCK-C from the chicken to understand more clearly the mechanistic basis for the differential patterns of gene expression in birds. This is of special interest in the case of PEPCK-C, in view of its very complex arrangement of hormone-responsive elements (Hanson and Patel 1994). It is probable that the gene for PEPCK-C in birds represents a functional variant of its counterpart in mammals and that key differential regulatory features may highlight important aspects of promoter architecture involved in gene regulation. An insight into the regulation of PEPCK-C gene expression in the chicken may also provide a greater understanding of the species-specific variation in the regulation of hepatic gluconeogenesis.

Organisms contain varying amounts of the two isozyme forms of PEPCK, the cytosolic PEPCK-C and a mitochondrial PEPCK-M. These two forms of PEPCK are encoded by distinct nuclear genes (Hod et al. 1986) which have different patterns of regulation. The gene for PEPCK-C in all species studied to date is acutely regulated by diet and hormones, whereas the gene for PEPCK-M is largely constitutive in its pattern of expression [see (Hanson and Patel 1994) for a review]. Most mammalian species display almost equal activity of the two forms of PEPCK in their livers. However, there are metabolically important variations in this pattern of expression of the two isozyme forms of PEPCK which occur in several animal species. The rat, mouse and hamster express almost entirely (>90%) the gene for PEPCK-C (Hanson and Garber 1972, Nordlie and Lardy 1963), whereas animals such as chicken, pigeon and rabbit predominately express the gene for PEPCK-M (Soling et al. 1973). The gene for PEPCK-C is hormonally responsive, so that starvation or acute diabetes significantly induces the activity of PEPCK-C in rabbit liver (Johnson et al. 1970).

Birds are unique among animals studied to date in that the intracellular distribution of PEPCK varies among tissues, with the liver containing only PEPCK-M, whereas both isozymic forms are present in the kidney. This tissue-specific pattern of distribution of PEPCK may contribute to the differing abilities of chicken liver and kidney to support gluconeogenesis from a variety of precursors (Watford et al. 1981). In tissues such as avian liver in which PEPCK-M is the major isozymic form, gluconeogenesis occurs predominantly from lactate, whereas glucose formation from more oxidized substrates such as pyruvate and alanine occurs in tissues which contain PEPCK-C, such as the avian kidney (Watford et al. 1981). These differences in the regulation of gluconeogenesis may be related to the different ways of generating cytosolic NADH when specific metabolic precursors are used in gluconeogenesis. If PEPCK-C is present, malate is transported across the mitochondrial membrane, generating NADH in the cytosol where it is oxidized to oxaloacetate. Because the requisite cytosolic NADH is being generated after the formation of oxaloacetate, a variety of gluconeogenic precursors can be used (alanine, pyruvate and gluconeogenic amino acids), without respect to their initial relative redox states. On the other hand, if only PEPCK-M is present in a tissue, gluconeogenesis occurs from lactate because oxidation to pyruvate produces NADH directly in the cytosol (Hanson and Garber 1972, Soling et al. 1973).

This mode of expression of PEPCK in these two major gluconeogenic organs may be the result of specific adaptations in avian carbohydrate metabolism to meet the novel energy demands faced by birds. Avian flight muscle goes through periods of prolonged use in which the cycling of lactate to glucose via the Cori cycle is essential. Avian liver supports this critical metabolic function by expressing high levels of PEPCK-M. In contrast, the net glucose formation from amino acids that occurs during starvation, when muscle protein itself is catabolized, requires the PEPCK-C activity found in the kidney. In addition, birds have a less pronounced need for glucose because they have a relatively small brain and nucleated red blood cells with mitochondria and prefer ketone bodies as a metabolic fuel (Sturkie 1965). Birds can also use a diet in which fat replaces carbohydrate as a source of energy (Renner and Elcombe 1964).

Despite the negligible level of PEPCK-C in chicken liver, expression of its gene can be induced in the livers of adult chickens by the administration of dibutryl cyclic AMP (Bt2cAMP) (Hod et al. 1984). There is also a high level of expression of PEPCK-C in the embryonic liver of the chicken, which decreases to negligible levels 2 d before hatching (Savon et al. 1993). This contrasts with the pattern of development of mammalian PEPCK-C in which the gene is not expressed until immediately after birth (Ballard and Hanson 1968). Thus, differences in the organization of regulatory elements in the promoter of the avian and mammalian PEPCK-C genes are likely to account for this variation in transcriptional capacity in fully differentiated liver from rats and chickens. In rat liver, a complex interaction between multiple hormones, including glucagon, glucocorticoids thyroid hormone and insulin, has been shown to control the level of PEPCK-C gene expression (Hanson and Patel 1994). In this report, we present an analysis of the hormonal regulatory elements present in the promoter of the gene for PEPCK-C from the chicken and provide a comparison of the organization and function of these elements with the better-studied transcriptional control of the PEPCK-C gene from mammals.

MATERIALS AND METHODS

Materials. All chemicals used were of reagent grade or the highest quality available commercially. Tissue culture media were DMEM F-12 (GIBCO, Grand Island, NY), supplemented with calf serum (GIBCO), and fetal calf serum (Hyclone, Logan, UT), unless noted otherwise. Radiolabeled nucleotides [alpha 32P]dCTP (111 TBq/mmol), [gamma 32P]ATP (222 TBq/mmol) and [35S]dATP (37 TBq/mmol) were obtained from New England Nuclear, Wilmington, DE. Restriction enzymes were purchased from Boehringer Mannheim, Indianapolis IN or New England Biolabs, Beverly, MA and used with the buffers recommended by the manufacturer. Other enzymes were obtained from the sources indicated. DNA ligase, Klenow fragment of DNA polymerase I, calf intestinal phosphatase, S1 nuclease, DNase I and polynucleotide kinase were from Boehringer Mannheim; RNase A was from Sigma, St. Louis, MO. Radiolabeled, double-stranded DNA probes were routinely made using the Random Primer DNA labeling kit from Boehringer Mannheim. The Sequenase Version 2.0 DNA Sequencing kit from U.S. Biochemical, Cleveland, OH, was used to obtain promoter sequence. G418 was purchased from GIBCO.

Fig. 1. A comparison of the DNA sequence of the promoters for the phosphoenolpyruvate carboxykinase (PEPCK-C) genes from the chicken and rat. The DNA sequence of the coding strand extending from base position -1865 to +34 relative to the start site of transcription. A comparison of the primary structure of the promoter regulatory regions for PEPCK-C genes from the chicken (upper line) and the rat (lower line). Alignment was established by orienting the respective start sites of transcription. Homologous sequences in the two genes are shown in bold and the protein binding domains, cP1 through cP6, in the chicken PEPCK-C promoter are underlined.
[View Larger Versions of these Images (51 + 74K GIF file)]

Vectors and oligonucleotides were supplied by commercial and private sources. pSCAN was obtained from Stratagene, La Jolla, CA and the plasmids pH 4 and pH 6 were constructed as described previously (Hod et al. 1984). Primers complementary to the T3 and T7 promoter regions of pSCAN were synthesized or purchased from Stratagene, respectively. Oligonucleotides were chemically synthesized using an Applied Biosystems 380A DNA Synthesizer (Foster City, CA). The following proteins (with their source) were used for DNase I footprinting analysis of the chicken PEPCK-C promoter: CAAT enhancer binding protein alpha , (C/EBPalpha ) Steven McKnight, Carnegie Institute, Baltimore, MD; cAMP regulatory element binding protein (CREB), Mark Montminy, Salk Institute, La Jolla, CA; and nuclear factor-1 (NF-1) CTF, N. Mermod and R. Tijan, Department of Biochemistry, University of California, Berkeley, CA. C/EBPalpha and CREB were purified after over-expression in Esherichia coli. Expression vectors for C/EBPalpha and C/EBPbeta were a gift from Steven McKnight.

Animals. Male, Sprague-Dawley rats, weighing 150-200 g, purchased from Zivic Miller Company (Zelienople, PA), were used in these studies. The animals were freely fed a nonpurified diet (Formulab Chow, Ralston Mills, St. Louis, MO), a high energy, high protein diet for rodents, and were provided with free access to water. Male Leghorn chickens (4-6 weeks old) weighing 250-400 g were hatched in the Case Western Reserve Animal Resource Facility and were also fed a nonpurified diet (Lab Chick Chow S-G, Madina Farmers Exchange, Madina, OH). The chickens were used as a source of RNA for Northern analysis and for nuclear proteins for the DNase I footprinting studies. All animals used for research purposes at Case Western Reserve University were housed in a modern, AALAC certified animal facility under the supervision of board-certified veterinarians. The animals were handled in accordance with an animal use protocol approved by the Institutional Animal Care and Use Committee. The chickens were injected intraperitoneally with Bt2cAMP and theophylline (30 µg/g body weight of each) or with saline (controls); after 2 h the animals were killed by cervical dislocation, the livers removed, and nuclear proteins and RNA isolated as described below.

Cell culture and stable transfection. LMH cells were derived from a hepatocellular carcinoma which was induced in a male Leghorn chicken by long-term treatment with diethynitrosamine (Kawaguchi et al. 1987). The LMH cells were kindly provided by Alan G. Goodridge, University of Iowa, Iowa City, IA. The LMH cells maintain several differentiated morphological and biochemical features, including the expression of the genes for glucose-6-phosphatase and ATPase. The cells were cultured in Weymouths MB medium, containing 10% fetal calf serum. The H4 and HepG2 cells used in this study were maintained in Dulbecco's modified Eagle's medium, supplemented with 5% calf serum and 5% fetal calf serum. During stable transfection of DNA into the LMH or H4 cells, the cells were grown to confluence in 10-cm dishes and then transfected with the PEPCK-neo gene. The cells were then selected in medium containing 500 mg/L of the antibiotic G418.

Preparation of nuclear extracts. Nuclear extracts were prepared from rat or chicken liver by the method of Gorski et al. (1986). The purification of the cAMP regulatory element (CRE) binding protein(s) from rat liver nuclear extracts was described previously (Roesler, et al. 1989).

DNase I footprinting. The DNA probes used in the footprinting assays were prepared by end-labeling one strand of DNA using T4 polynucleotide kinase and [gamma -32P]ATP. The DNase I footprinting conditions were described previously (Roesler et al. 1989).

Construction of recombinant plasmids. The initial chimeric gene, pSCP/Pf, was engineered by using the Klenow fragment of DNA polymerase to fill in the ends of the 400 bp PvuII fragment of pH 4, followed by ligation to pSCAN through ends created by digestion with EcoRI. An additional plasmid, pSC-216del, containing a shortened promoter sequence, was produced by first isolating the 250 bp SmaI-ClaI fragment of pSCP/Pf. Compatible ends for ligation were formed in pSCAN by EcoRI digestion, filled in using Klenow enzyme, and subsequent restriction with ClaI. These two PEPCK-C promoter segments were subsequently subcloned in front of the CAT structural gene in the plasmid pXSV1CAT (Bokar et al. 1988). The larger two promoter fragments (1.8 kb and 700 bp) were also ligated into pXSV1CAT. Corresponding plasmids containing these two larger promoter regions ligated to the neomycin resistance gene were made by XbaI and ClaI digestion of pH 4 followed by ligation to the filled in EcoRI site of pSCAN. The block mutation was created from pSCPEPC-366 using the approach described previously (Liu and Hanson 1991).

Cell transfection and isolation of RNA. Transfection of DNA into eukaryotic cells was accomplished using calcium phosphate precipitation (Ausubel et al. 1987) on dishes containing 80% confluent cultures. After an overnight exposure to the DNA, the cells were shocked with glycerol and then incubated with normal complete media for 2 d. Hormonal treatments were performed as described in detail previously (Liu et al. 1991). Final amounts of each hormone added to an individual plate were as follows: Bt2cAMP (2.5 mg in water), theophylline (1.8 mg in PBS) and insulin (30 µg/plate). The cells were incubated for 4 h in 10 mL of serum-free media, to which the individual hormones were added.

Total RNA was extracted from intact tissues and tissue culture cells by homogenization in 4 mol/L guanidinium thiocyanate followed by centrifugation through a CsCl cushion (Chomczynski and Sacchi 1987). Pelleted RNA samples were resuspended, separated by electrophoresis, transferred and hybridized under established conditions (Ausubel et al. 1987).

S1 nuclease analysis. RNA analysis by S1 nuclease digestion was performed as described previously (Hod et al. 1984). An end-labeled DNA probe was made by digesting pSCP/Pf with NarI, followed by treatment with calf alkaline phosphatase. Subsequent restriction with EcoRI resulted in the formation of a 646 bp fragment which was gel isolated and end-labeled with polynucleotide kinase. Precipitated RNA was resuspended in a formamide-containing hybridization solution to which 1.75 kBq of probe was added. Denaturation of the DNA was achieved by a 15-min incubation at 75°C, followed immediately by overnight treatment at 52°C to optimize DNA/RNA hybrid formation. The next day, samples were cooled to 37°C, digested with S1 nuclease and prepared for electrophoresis in a 6% denaturing sequencing gel.

Primer extension analysis. A synthetic oligodeoxynucleotide (SS-2) was end-labeled and used to prime DNA synthesis, with total RNA from chicken kidney serving as a template. The elongated, single-stranded DNA species thus formed was subjected to electrophoresis on a 6% polyacrylamide gel containing urea. The extension reaction and electrophoresis were as described previously (Klemm et al. 1990).

Transcriptional activity. The transcriptional activity of specific segments of the PEPCK-C promoter was determined by stably transfecting chimeric genes into either LMH cells (chicken) or H4 cells (rat) or by determining the level of transcription after transient transfection into HepG2 cells (human) or LMH cells (chicken). The PEPCK-C promoter was linked to the structural gene for chloramphenicol acetyltransferase (CAT) and transiently transfected into LMH or HepG2 cells by asiloglycoprotein receptor-mediated endocytosis (Crawford and Hanson 1993) of 10 µg of the PEPCK-CAT vector together with 5 µg of a vector containing the beta -galactosidase gene driven by the Rous sarcoma virus promoter. The effect of C/EBPalpha and C/EBPbeta on transcription from the PEPCK-C promoter was determined by co-transfection of 10 µg of the chimeric PEPCK-CAT gene, together with 5 µg of the expression vector containing the cDNA for the transcription factors, into HepG2 cells. Approximately 4 × 106 trypsinized cells were added to the precipitated DNA and the mixture was plated out. Two days after transfection, the cells were harvested and lysed by freeze-thawing. CAT assays were performed on equal amounts of protein from each extract as described by Gorman et al. (1982). The activity of CAT in the transfected cells is expressed relative to the activity of beta -galactosidase to correct for the efficiency of transfection. Each value is the average of three separate transfections experiments or as noted in the figure legends.

Statistical analysis. The values derived for both the stable and transient transfection of genes containing the PEPCK-C promoter into chicken and rat hepatocytes were analyzed statistically using Student's t distribution.

RESULTS

Analysis of the sequence of the PEPCK-C promoter region from the chicken. To analyze the regulation of transcription from the promoter for the gene for PEPCK-C from the chicken, four recombinant plasmids were created: pSCchickPEPC-1.8, pSCchickPEPC-700, pSCchickPEC-366 and pSCchickPEPC-216. These plasmids contain 1.8 kb, 700 bp, 366 bp and 216 bp, respectively, of the 5'-flanking region upstream of the transcriptional start site in the PEPCK-C gene and were made using the pSCAN vector which has a convenient multiple cloning region immediately 5' to the neo structural gene. The multiple cloning site is also flanked by oppositely oriented T7 and T3 promoters which allow for sequencing of newly introduced DNA. The nucleotide sequence of the PvuII-PvuII fragment of the PEPCK-C promoter had been previously determined (Hod et al. 1984). Sequencing of this region, initially performed to confirm the structure of the recombinant plasmid, indicated several errors in the earlier sequence analysis. Repeated sequencing of this 400 bp promoter fragment in both directions, combined with complete sequencing of the 1.8 kb fragment of pSCchickPEPC-1.8, permitted determination of the primary nucleotide structure as indicated in Figure 1. When aligned through the respective transcriptional start sites, the promoter regulatory regions of PEPCK-C from the chicken and rat have an overall sequence identity of 25%; this compares with a 95% sequence identity between the rat and mouse for the same region of promoter of the PEPCK-C gene (S. Hakimi, M. Johnson and R. Hanson, personal communication, Case Western Reserve University). This large sequence divergence is evenly distributed throughout the region of the promoters being compared, and sizeable local regions of sequence identity are conspicuously absent. The most extensive correlation occurs within the sequence which spans from the presumed TATA box to the beginning of the first exon (-35 to +1) which has 46% identity. A second area of relatedness (48%) occurs upstream and extends from promoter positions -1075 to -1050. This region of the PEPCK-C gene from the rat has been shown to contain a peroxisome proliferator-activated receptor gamma 2 (PPARgamma 2) binding site, which is required for the expression of the gene in adipose tissue (Tontonoz et al. 1995). Additional stretches of sequence identity between the rat and chicken PEPCK-C promoters are short (usually 4 base pairs or less) and dispersed throughout the entire promoter.

The start site of transcription of the PEPCK-C promoter was determined by S1 nuclease analysis (Fig. 2A) and by primer extension analysis (Fig. 2B). S1 nuclease analysis was performed using neo mRNA produced by H4 cells which had been transfected with pSCchickPEC-366. The probe was end-labeled on the 5' end of the coding strand at the NarI site and extended to the EcoRI site in the pSCAN poly linker. The undigested probe was 646 bases long (see lane 1), whereas the protected fragment was 272 bases in length. This corresponds to a position at -34 bp in the promoter fragment, which is identical to the more 5'-positioned site determined by primer extension. For primer extension analysis a 20-base, single-stranded oligonucleotide, made complementary to the last 20 bases of the subcloned promoter region, was end-labeled and hybridized to total RNA from avian kidney. There was the expected increase in the level of PEPCK-C mRNA in the kidneys of chickens treated with Bt2cAMP, and extension products 30 and 34 bases long were detected.


Fig. 2. Identification of the start-site of transcription of the chicken phosphoenolpyruvate carboxykinase (PEPCK-C) gene by S1 analysis of neo mRNA. The position at which transcription initiates in a PEPCK-C neo mRNA was identified by S1 nuclease analysis (bottom panel) of total RNA samples from H4 cells which were transfected with pSCP/Pf. Primer extension analysis was performed (top panel) in which a single-stranded oligonucleotide primer corresponding to the 20 bp at the 3'-end of the PEPCK-C promoter was hybridized to endogenous mRNA from avian kidney.
[View Larger Versions of these Images (25 + 45K GIF file)]

Mapping of putative protein binding sites in the promoter regulatory region of the PEPCK-C gene from the chicken. To aid in the identification of regulatory elements present in the PEPCK-C promoter from the chicken, DNase I footprint analysis was performed using a 366 bp end-labeled fragment of the chicken PEPCK-C as a probe. Digestion of the DNA (Fig. 3, lane 2) generated a nested series of fragments that are the result of specific DNase I cleavages at various positions along the promoter fragment. Six protein-binding domains, cP1 through cP6, were detected when nuclear proteins from chicken liver were used in the footprinting assay; a similar pattern of binding was noted when a protein extract from rat liver nuclei was used (data not shown). The distal half of cP1 contains the recognition sequence for nuclear factor-1 (NF-1). When the PEPCK-C promoter was footprinted with NF-1 purified from HeLa cells, this portion of cP1 was uniquely protected from DNase I digestion. The rat PEPCK-C promoter contains four binding sites for members of the CCAAT/enhancer binding protein (C/EBP) family of transcription factors. Footprinting of the chicken PEPCK-C promoter with the 11 kDa recombinant C/EBPalpha resulted in protection of the proximal half of cP1, as well as sites cP3, cP4, cP5 and cP6. The cAMP regulatory element-binding protein (CREB) mediates the stimulation by cAMP of transcription of several eukaryotic genes and also may play a role in establishing basal promoter strength for other genes (Montminy et al. 1990). Incubation of the probe with bacterially expressed CREB resulted in weak binding to the 3' half of cP1 and stronger binding at the cP4 position.
Fig. 3. Identification of protein binding domains in the chicken phosphoenolpyruvate carboxykinase (PEPCK-C) promoter by DNase I footprinting. Proteins extracted from liver nuclei from the chicken were used for DNase I footprint analysis of the chicken PEPCK-C promoter sequence extending from -366 to +34. G represents a G sequencing reaction; no protein was added in lane -; lane 1 contained nuclear proteins isolated from chicken liver. Footprint patterns in lanes 2, 3 and 4 are the result of binding of purified nuclear factor-1 (NF-1), CAAT/enhancer binding protein (C/EBP) and cAMP regulatory element-binding protein (CREB), respectively. cP1 through cP6 represent regions of the PEPCK-C promoter which bind protein and are underlined in the sequence of the promoter in Figure 1.
[View Larger Version of this Image (37K GIF file)]

Hormonal responsiveness of chimeric genes containing the avian PEPCK-C promoter. Next, we tested the regulation of transcription from the PEPCK-C promoter from the chicken using four recombinant plasmids containing segments of the promoter (-1.8 kb, -700 bp, -366 bp and -216 bp) linked to the neo structural gene. These chimeric genes were introduced into H4 cells, a rat hepatoma cell line, by calcium phosphate precipitation. Several independently isolated cell lines (n = 5) were selected to study hormonal regulation of transcription from the PEPCK-C promoter by determining the alterations in the levels of neo mRNA by Northern analysis (Fig. 4). There was an ~2- to 1.5-fold induction of transcription from the four PEPCK promoter constructs used in this study within 2 h after Bt2cAMP was added, a finding which is comparable to that noted for the induction of PEPCK-C mRNA by these hormones in chickens (Hod et al. 1984). The addition of insulin to the culture medium blocked Bt2cAMP-induced transcription from the PEPCK-C promoter by an average of about 50% for all of the promoter constructs tested. The insulin inhibition of transcription from the PEPCK-C promoter in the presence of Bt2cAMP was most pronounced on the segment deleted to -216 (P < 0.005). However, the inhibitory effect of insulin was significant at the 95% confidence level for all of the segments of the PEPCK-C promoter with the exception of -1.8 kb, where there was considerable variability in the response to insulin. The addition of insulin also decreased the basal level of transcription of the PEPCK-neo gene in all four of the PEPCK-C promoter constructs studied but was significant (P < 0.05) only for the promoter deleted to -216 and -700. By way of comparison, the expression of the analogous rat PEPCK-C-neo gene, while retaining both cAMP and dexamethasone responsiveness, does not show the expected inhibition when treated with insulin (Wynshaw-Boris et al. 1986). As a control for the effectiveness of the hormonal treatments, the Northern blot with a cDNA to PEPCK-C from the rat was rehybridized with RNA from the H4 cells, and the stimulatory response of the endogenous rat PEPCK-C gene to hormones was as expected (data not shown).
Fig. 4. Identification of regions of the chicken phosphoenolpyruvate carboxykinase (PEPCK-C) promoter responsive to dibutryl cyclic AMP and insulin. Northern analysis of neo mRNA transcribed from a variety of chimeric PEPCK-C-neo genes consisting of -1.8 kb, -700 bp, -366 bp and -216 bp of the chicken PEPCK-C promoter driving the neo structural gene. Rat H4 hepatoma cells, which were selected in G418, were treated with saline (control cells), Bt2cAMP, Bt2cAMP plus insulin, or insulin. Total RNA was isolated and neo mRNA was determined by Northern analysis as described in Materials and Methods. The effect of hormones on the level of PEPCK-C mRNA in these cells was expressed relative to the control value. The results of a densitometric analysis of five separate transfection experiments are shown as the mean ± the standard error of the mean.
[View Larger Version of this Image (24K GIF file)]

Next, we compared the effects of Bt2cAMP and insulin on transcription from the PEPCK-C promoter from the chicken and the rat in chicken (LMH) and mammalian (HepG2) hepatoma cells by transient transfection. The transcriptional response of the avian PEPCK-C promoter linked to the CAT structural gene was determined in LMH cells using the same series of deletions used in the experiments reported in Figure 4 (-1.8 kb, -700 bp, -366 bp and -216 bp) (Fig. 5A). The 700 bp segment of the PEPCK-C promoter gave the best response to the addition of Bt2cAMP; it was induced threefold by the cyclic nucleotide from the control value (P < 0.001). Insulin blocked the induction by Bt2cAMP (P < 0.05) but did not lower the basal level of transcription from the avian PEPCK-C promoter in LMH cells. The other segments of the PEPCK-C promoter tested for their transcriptional responsiveness were less markedly induced by Bt2cAMP. The addition of insulin to the LMH cells did not cause a significant decrease (at the 95% confidence level) in the ability of Bt2cAMP to induce transcription from segments of the avian PEPCK-C promoter shorter than -700, an effect which may be due to the low level of transcriptional induction by Bt2cAMP on these segments of the promoter. The reason for this difference in response between the regions of the chicken PEPCK-C promoter is not clear, but it is possible that there are critical transcriptional regulatory elements between -700 and -366 which are required for the full response of the PEPCK-C promoter in the LMH cell. Finally, insulin had no significant effect (95% confidence level) on basal transcription from the PEPCK-C promoter in LMH cells.

Fig. 5. A comparison of transcriptional regulation from the chicken and rat phosphoenolpyruvate carboxykinase (PEPCK-C) promoter by dibutryl cyclic AMP and insulin in chicken and mammalian hepatoma cells. (A) LMH cells were transfected with chimeric genes containing the truncated chicken PEPCK-C promoter (-1.8 kb to -216 bp), which was ligated to the structural gene for chloramphenicol acetyltransferase (CAT) or (insert) the rat PEPCK-C promoter (-490 to +73), which was also ligated to the CAT structural gene. The activity of CAT was determined 4 h after the addition of saline (control cells) dibutryl cyclic AMP (Bt2cAMP), Bt2cAMP plus insulin, or insulin alone, as indicated in Materials and Methods. (B) HepG2 cells were transfected with a chimeric gene containing -1.8 kb of the promoter from the chicken PEPCK-C gene linked to the structural gene for CAT or (C) a chimeric gene containing -490 to +73 of the rat PEPCK-C gene linked to the structural gene for CAT. The activity of CAT was determined 4 h after treatment with saline (control cells), Bt2cAMP, Bt2cAMP plus insulin, or insulin alone, as indicated in Materials and Methods. The values are expressed as fold induction relative to the control (saline addition) cells and were corrected for the relative activity of beta -galactosidase. The values presented in this figure are the average of at least three separate transfection experiments and are expressed as the mean ± SEM of the CAT activity in the cells, except for the values for the rat PEPCK-C-CAT gene which represent the average of two separate transfection experiments.
[View Larger Version of this Image (38K GIF file)]

The rat PEPCK-C promoter (from -490 to +73), when introduced into LMH cells, responded to Bt2cAMP with a threefold induction in transcription, which was only partially inhibited by insulin (significant at the 95% confidence level) (Fig. 5A, insert). As with the avian PEPCK-C promoter, there was no effect of insulin on the basal level of transcription from the rat PEPCK-C promoter in LMH cells. We next tested the effect of the hormones on a chimeric PEPCK-C-CAT gene containing the avian or rat PEPCK-C promoters, which were transfected into HepG2 cells, a mammalian hepatoma cell line. Transcription from the avian PEPCK-C promoter was not stimulated by Bt2cAMP in these cells, and the basal level of transcription was not inhibited by insulin; in addition, the overall level of transcription from the chicken PEPCK-C promoter was very low in HepG2 cells (Fig. 5B). This contrasts with the chimeric PEPCK-C-CAT gene containing the PEPCK-C promoter from the rat which was induced 10-fold by Bt2cAMP but was not responsive to insulin (Fig. 5C).

A comparison of the 5'-flanking PEPCK-C sequences from the chicken and the rat suggested that the segment extending from -140 to -49 bp in the promoter sequence from the chicken was highly homologous to an insulin response sequence (IRS) identified (O'Brien and Granner 1991; O'Brien et al. 1990) in the PEPCK-C gene from the rat. Consequently, a pSCAN-based chimeric gene containing a block mutation made in this location of the chicken PEPCK-C promoter was created and introduced into H4 cells by transfection, followed by selection with G418. An analysis of five individual clones of cells expressing this recombinant plasmid (Fig. 6) indicated no loss of transcriptional responsiveness of the modified PEPCK-C promoter to insulin; we noted an average reduction of neo mRNA levels caused by insulin of at least 60% in the five clonal cell lines tested. One of the clonal cell lines (line 2) did not respond to any treatment, and others (cell lines 1 and 4) demonstrated only a marginal stimulation by Bt2cAMP. In addition, Bt2cAMP-stimulated transcription from the PEPCK-C promoter in all of the clonal cell lines, except line 2, was markedly inhibited by insulin addition. Clonal cell line 6 served as a control cell line and was transfected with a chimeric gene containing the unmodified segment of the avian PEPCK-C promoter linked to the neo structural gene. In these cells, we noted the expected increase in transcription from the PEPCK-C promoter after the addition of Bt2cAMP, which was blocked by the addition of insulin. We conclude that the putative IRS in the avian PEPCK-C promoter is not involved in the negative effect of insulin on PEPCK-C gene transcription.

Fig. 6. The effect of mutating the putative insulin response sequence (IRS) on the action of dibutryl cyclic AMP and insulin on transcription from the chicken phosphoenolpyruvate carboxykinase (PEPCK-C) promoter in clonal cell lines isolated from transfected H4 hepatoma cells. Stable transfection experiments were performed using a chimeric gene in which the chicken PEPCK-C promoter sequence was altered by introducing a block mutation in the putative IRS site, which maps -95 to -105 in the promoter. Total RNA was isolated from five individual G418-selected cell lines which had been treated for 4 h with saline (control cells), dibutryl cyclic AMP (Bt2cAMP) Bt2cAMP plus insulin, or insulin alone, as indicated in Materials and Methods. Clonal cell line 6 was transfected with a chimeric PEPCK-C-neo gene which contained the intact chicken PEPCK-C promoter, without a block mutation in the putative IRS.
[View Larger Version of this Image (30K GIF file)]

Both the rat and chicken PEPCK-C genes have a number of C/EBP binding domains; C/EBP has been shown to be involved in the control of liver-specific expression of the PEPCK-C gene in the rat (Park et al. 1992, Patel et al. 1994). We therefore tested the effect of C/EBPalpha and C/EBPbeta on transcription from the avian PEPCK-C promoter by transient transfection of a chimeric PEPCK-C-CAT gene into HepG2 cells and compared the effect with that noted with the rat PEPCK-C promoter. As shown in Figure 7, there was a 10- to 15-fold induction of transcription from both the chicken (-1.8 kb) and the rat (-490 to +73) PEPCK-C promoters caused by the co-transfection of C/EBPalpha or C/EBPbeta (significant at greater than the 95% confidence level), suggesting that these transcription factors also play a key role in the regulation of hepatic PEPCK-C gene expression in birds.

Fig. 7. The effect of CAAT/enhancer binding protein (C/EBP)alpha and C/EBPbeta on transcription from the rat and chicken phosphoenolpyruvate carboxykinase (PEPCK-C) promoters in HepG2 cells. Ten micrograms of a chimeric PEPCK-C chloramphenicol acetyltransferase (CAT) gene containing -490 to +73 of the rat PEPCK-C promoter and 10 µg of a PEPCK-C-CAT gene containing -1,800 to +1 of the chicken PEPCK-C promoter were transfected into HepG2 cells, in the presence and absence of cotransfected C/EBPalpha or C/EBPbeta (5 µg each). Experimental details are presented in Materials and Methods. The level of CAT activity in the cells was determined 48 h later. The values are expressed relative to the basal level of CAT activity in the absence of C/EBP and represent the means ± SEM of three separate transfection experiments.
[View Larger Version of this Image (26K GIF file)]

DISCUSSION

The gene for PEPCK-C in chickens is hormonally regulated, but with developmental and tissue-specific patterns of expression that differ from those observed with the same gene from the rat. The avian embryo has an absolute need for gluconeogenesis from relatively "oxidized" substrates such as amino acids, and we have elsewhere demonstrated (Savon et al. 1993) high levels of expression of PEPCK-C in liver during embryonic development. Chicks, on the other hand, appear to have only a moderate need for glucose from any source and can grow on a "carbohydrate free" diet in which soybean oil is the only nonprotein constituent (Renner and Elcombe 1964). In general, birds have only marginal levels of hepatic gluconeogenesis from precursors other than lactate. Thus, the expression of the PEPCK-C gene is negligible in the livers of chickens, even after 48 h of starvation, and occurs primarily in the kidney (Watford et al. 1981). Paradoxically, PEPCK-C mRNA can be induced by administration of Bt2cAMP to normal, fed chickens (Hod et al. 1984), so that the gene for PEPCK-C in the chicken has the capacity to be expressed in the livers of birds throughout life but it is clearly subject to an alternate mode of hormonal regulation compared with the analogous gene from rats. Here, we show that the promoters from the genes for avian and rat PEPCK-C have some similarities in the organization of their transcriptional regulatory elements. However, there are major differences in both the sequences of segments of the PEPCK-C promoter from the rat as compared with the chicken and the arrangement of these regulatory elements within the promoter. The functional importance of these variations is exemplified by the differences in hormonal regulation of the expression of chimeric genes which contain portions of the 5'-flanking region of PEPCK-C from the chicken. Our findings indicate that the avian PEPCK-C gene is more sensitive to insulin and is somewhat less responsive to cAMP compared with the same gene in the rat.

When the promoter sequences of the chicken and rat PEPCK-C genes are aligned at their start sites of transcription and then adjusted slightly to maximize relatedness, there is an overall nucleotide identity of 42% within the first 366 bp. A segment of the rat PEPCK-C promoter from -460 to the start site of transcription has been shown to contain most of the regulatory elements required to confer physiologically relevant control of transcription of chimeric genes introduced into transgenic mice (McGrane et al. 1988 and 1990, Patel et al. 1994). Expression of a PEPCK-C-bovine growth hormone gene containing a segment of the rat PEPCK-C promoter from -460 to +73, introduced into the germ line of mice, occurs in a tissue-specific manner, is induced appropriately by diet and hormones and develops with the same pattern as the endogenous PEPCK-C gene (McGrane et al. 1990). The pattern of transcriptional regulatory elements in the PEPCK-C promoter from the rat has been well established (see Hanson and Patel 1994 for a review); the relative positions of the putative regulatory elements in the PEPCK-C promoter from the chicken differ significantly from the locations of these same sequences in the PEPCK-C promoter from the rat (Fig. 8). Interestingly, the promoters for the mouse and the human PEPCK genes have recently been sequenced and shown to be virtually identical to the sequence for the rat, suggesting that the significant divergence in the chicken PEPCK gene relates to the specific metabolic function of PEPCK in birds and to the observed differences in hormonal regulation noted in this study.

Fig. 8. A comparison of transcriptional regulatory elements in the chicken and the rat phosphoenolpyruvate carboxykinase (PEPCK-C) genes. Putative transcriptional regulatory elements and transcription factor binding sites, identified by protein binding and/or sequence conservation are compared in the rat (P1 through P4) and chicken PEPCK-C genes (cP1 through cP4). The abbreviations for the elements and factors are: CREB, cAMP regulatory element-binding protein; C/EBP, CAAT/enhancer binding protein; FOS/JUN, FOS/JUN binding domain; NF1, nuclear factor-1; HNF-1, hepatic nuclear factor-1; IRE, insulin responsive element; GRE, glucocorticoid responsive element; T3, thyroid hormone responsive element.
[View Larger Version of this Image (16K GIF file)]

The rat PEPCK-C promoter contains three regions which have clusters of regulatory elements. Region 1 contains the CRE and adjacent NF-1 site (P-1). A mutation in the CRE markedly reduces the cAMP responsiveness of the PEPCK-C promoter in the livers of transgenic mice, whereas a block mutation in the NF-1 binding site causes a marked increase in the level of basal gene transcription (Patel et al. 1994). The same type of response was noted when the PEPCK promoter was transfected into hepatoma cells in culture (Liu et al. 1991). In addition, unpublished work from our laboratory has demonstrated that NF-1 has a powerful negative effect on transcription from the PEPCK-C promoter in the same cell system (D. Crawford, P. Lehey, R. Gronostajski, and R. Hanson, personal communication, Case Western Reserve University). The CRE from the PEPCK promoter will bind members of the leucine zipper family of transcription factors, including CREB (Yamamoto et al. 1988), C/EBPalpha (Roesler et al. 1989), C/EBPbeta (Park et al. 1992), D-binding protein (Roesler et al. 1992) and fos/jun heterodimers (Gurney et al. 1992).

The region of the PEPCK-C promoter from the chicken which contains the most sequence identity with the PEPCK-C promoter from the rat also has a CRE adjacent to an NF-1 binding site. The cP1 region in the chicken PEPCK-C promoter maps between -78 and -85 bp and contains an octamer which differs from the consensus CRE sequence in two positions. Both the CREB and C/EBPalpha can bind within or near the cP1 region of the chicken PEPCK-C promoter, although CREB binds only weakly to this region of the PEPCK promoter (Fig. 3). Interestingly, the cP1 and cP3 sites have an almost perfect consensus C/EBP binding domain and there is strong binding of C/EBPalpha to the cP3 site, suggesting that C/EBP may be involved in cAMP regulation of PEPCK-C gene transcription. The role of C/EBPalpha in mediating the action of cAMP on transcription of the gene for PEPCK from the rat has been suggested from previous studies (Park et al. 1990) and directly demonstrated more recently (Roesler et al. 1996). Figure 7 demonstrates that both C/EBPalpha and C/EBPbeta can stimulate transcription from the avian PEPCK promoter. In addition, preliminary work in our laboratory has demonstrated that the deletion of the gene for C/EBPalpha greatly diminishes the induction of PEPCK-C gene expression by cAMP in the livers of fetal rats (C. Croniger, V. Poli, G. Darlington and R. Hanson, personal communication, Case Western Reserve University).

Region 2 of the PEPCK promoter from the rat contains an HNF-1 site which has been shown to be critical for transcription of the PEPCK gene in the kidney (Patel et al. 1994). A similar element lies upstream the CRE/NF-1 site in the PEPCK-C promoter from the chicken. Whether the putative hepatic nuclear factor (HNF)-1 site in the avian PEPCK-C gene is involved in the regulation of PEPCK-C gene transcription in the kidney remains to be determined. In addition, Region 2 of the rat PEPCK-C promoter contains a C/EBP binding domain (Park et al. 1990 and 1992) and a thyroid hormone regulatory element (Giralt et al. 1991). In the rat, the C/EBP binding sites have been shown to be required for the full transcriptional responsiveness of the promoter to cAMP (Liu et al. 1991). The PEPCK-C gene from the chicken has a cluster of C/EBP binding sites in a similar region of the promoter, suggesting a key role for this transcription factor in the control of PEPCK-C in the avian liver. It has recently been shown that C/EBPalpha is necessary for the expression of PEPCK in the livers of mice in the perinatal period (Wang et al. 1995).

Region 3 of the PEPCK promoter from the rat is a complex domain with several regulatory elements. Granner and colleagues (Imai et al. 1990 and 1993) have described a glucocorticoid regulatory unit (GRU) in this area of the PEPCK-C promoter, within which there is a glucocorticoid regulatory element (GRE), an insulin regulatory element (IRS) and two accessory protein binding domains. The analogous region of the PEPCK-C promoter in the chicken, which extends from -359 to -345, contains considerable homology with the GRE consensus sequence; in addition, its position within the promoter is very similar to the comparable position of the GRE in the promoter of the rat PEPCK-C gene. In preliminary studies in which the gene was stably transfected into H4 hepatoma cells, we noted that the removal of the putative GRE from the chicken PEPCK-C promoter by truncation to -216 resulted in a loss of the normal twofold induction of transcription from the promoter by glucocorticoids (S. Savon, P. Hakimi and R. Hanson, personal communication, Case Western Reserve University, Cleveland, OH).

The physiological role of insulin as a negative regulator of PEPCK-C gene transcription is most likely a critical factor in the control of glucose homeostasis in birds. The regulation of PEPCK gene transcription by insulin is clearly very complex; there is no single regulatory element which has been implicated in the powerful negative effect of this hormone on the transcription of the PEPCK-C gene in the liver. The IRS found within the GRU of the rat PEPCK-C promoter, which maps between -416 to -407, accounts for only half of the inhibitory effect of insulin of transcription from the PEPCK-C promoter in hepatoma cells (O'Brien et al. 1990). Because the effect of insulin on PEPCK-C gene transcription most likely involves an interaction between this hormone and the CRE of the promoter (which maps between -87 to -74) a more proximal binding site for transcription factors involved in this process has been suggested. Most of the work done to date has involved the distal protein binding site; the proximal site has not yet been identified. The only available data on the protein(s) which bind to the IRS and the CRE arise from a recent study by O'Brien et al. (1994) which demonstrated that C/EBPalpha , either alone or complexed with an IRS binding protein, termed p20-C/IBP, can gel shift oligonucleotides corresponding to both the CRE from the PEPCK promoter and the distal IRS.

It is of interest to note that the sequence between -149 and -140 in the PEPCK-C promoter from the chicken contains a close match (9/10) with the consensus IRS found at positions -411 to -402 in the rat PEPCK-C promoter (O'Brien and Granner 1991). This sequence could be a candidate for the proximal IRS discussed above. Based on the possibility that insulin sensitivity of the avian PEPCK-C gene is dependent on this sequence, a recombinant chimeric gene containing a 10 bp block mutation corresponding to this area was tested for its transcriptional responsiveness to insulin (Fig. 6). It was found, however, to remain insulin responsive, indicating that the relative position of responsive elements both with respect to each other and with the TATA box has functional significance in regulating gene transcription (Hanson and Patel 1994).

Insulin inhibits transcription from the chicken PEPCK-C promoter; this contrasts with the absence of such an effect on transcription from PEPCK-C-neo fusion genes made with rat sequences observed previously in our laboratory (Wynshaw-Boris et al. 1986). To demonstrate the effect of insulin on transcription from the rat PEPCK promoter in transformed cells in culture, it has been necessary to carry out more elaborate types of selection procedures. For example, it has not been possible to date to demonstrate a marked negative effect of insulin on transcription when the PEPCK promoter is introduced by transient transfection (see Fig. 5C for an example). The procedure used to identify the IRS in the rat PEPCK-promoter (stable transfection in H4 cells) failed to demonstrate the dramatic inhibitory effects of insulin noted for the endogenous PEPCK-C gene in the same cells (O'Brien et al. 1990). In the present study, insulin decreased Bt2cAMP stimulated transcription from the avian PEPCK-C promoter and characteristically reduced basal gene expression by at least 50%, suggesting a substantial difference in the transcriptional response to insulin by the PEPCK-C gene from the rat and chicken.

A comparison of the hormonal responsiveness of PEPCK-C-neo chimeric genes transfected into rat hepatoma cell lines demonstrates both the similarities and differences between the promoter regulatory regions of the chicken and rat PEPCK-C genes. Although cAMP induces transcription from the PEPCK-C promoter from the rat four- to sixfold (Liu et al. 1991, Wynshaw-Boris et al. 1984), transcription from the chicken PEPCK-C promoter is stimulated only twofold. This reduced sensitivity to cAMP may be related to the blunted response of the PEPCK-C gene to the high circulating levels of glucagon in the blood of chickens. Larger concentrations of Bt2cAMP are required to induce the expression of hepatic PEPCK-C in the chicken, which may explain why there is an observed lack of induction of PEPCK-C gene expression in the livers of starved chickens (Hod et al. 1984). Interestingly, when a chimeric gene containing the avian PEPCK-C promoter was transiently transfected into HepG2 cells, there was a high level of basal transcription but there was no regulation by hormones. However, a similar chimeric gene containing the promoter for PEPCK-C from the rat was regulated in the expected manner by Bt2cAMP and insulin (Fig. 5). Thus, it is probable that factors required for the control of avian PEPCK-C gene transcription are missing from HepG2 cells.

Segments of the PEPCK-C promoter from -1.8 kb to -216 kb respond almost identically to the administration of hormones when stably transfected into H4 cells (Fig. 4). This suggests that the regulation of the PEPCK-C gene by cAMP and insulin resides within the first 210 bp from the start-site of gene transcription. This region of the promoter contains a CRE adjacent to an NF-1 binding domain, a putative IRS and an hepatic nuclear factor-1 (HNF-1) binding site. However, when the same chimeric genes containing the PEPCK-C promoter deletions were transiently transfected into chicken LMH cells, the segment of the promoter from -700 to +1 demonstrated the most marked activity and was most responsive to hormonal treatment. The reason for this difference in response is not immediately apparent, but may reflect inherent differences between transient and stable DNA transfection techniques.

Comparison of hormonal responsiveness of PEPCK-C-CAT chimeric genes comprised of promoter sequences from the chicken or rat, and transiently transfected into a chicken hepatocyte cell line (LMH cells), further suggests that differences in regulation are at least partially due to differences in transcription response elements present in the two promoter regulatory regions (Fig. 5). When introduced into chicken cells, transcription of chimeric genes containing promoter sequences from the chicken is inhibited by insulin. Chimeric genes containing segments of the PEPCK-C promoter from the rat when transfected into chicken cells (Fig. 5), on the other hand, elicit hormonal responses quantitatively more typical of this gene when studied in rat cells. In fact, the PEPCK-C promoter from the chicken did not respond to hormonal treatment when introduced transiently into HepG2 cells, further emphasizing the presence of functionally important differences in transcriptional regulatory elements between these two promoters, differences which may result from an altered dependence on chromosomal orientation within the nucleus.

Patel et al. (1994) used transgenic mice which contained chimeric genes in which the promoter for PEPCK-C gene from the rat contained block mutations in regulatory elements, to determine their role in the tissue-specific expression and hormonal regulation of PEPCK gene expression. When the CRE was removed by mutation, the level of expression of the transgene increased four- to fivefold, whereas transcriptional stimulation by cAMP was completely abolished. This finding suggests that protein binding to the CRE acts to maintain low levels of PEPCK gene transcription in the absence of cAMP. It is probable that the CRE site of the PEPCK promoter is always occupied by proteins in the cell because in vivo footprinting of the promoter using hepatoma cells has indicated the presence of a protected region at the CRE, which does not require added cAMP for protein binding to the promoter (Faber et al. 1993). In addition, NF-1 strongly inhibits transcription from the PEPCK promoter when the genes for NF-1 and PEPCK-CAT are co-transfected into HepG2 hepatoma cells (D. Crawford, P. Lehey, R. Gronostajski and R. Hanson, personal communication, Case Western Reserve University). It is possible that transcription of the gene for PEPCK-C in the chicken is maintained at a low level due to the arrangement of the NF-1 site relative to the CREB binding site(s), which markedly differs from the distribution of these elements found in the promoter for PEPCK in the rat. Therefore, only an elevation in the concentration of cAMP above the normal physiological levels will result in the transcription of the PEPCK-C gene in the livers of chickens. Why the gene is expressed in the livers of birds before hatching is an interesting question which remains to be answered. Further investigation of the differences in the hormonal response regions between these species will not only provide insight into the functional alterations operating in the PEPCK-C gene from the chicken, but also increase understanding of how a convergence of influences provided by discrete responsive elements results in specifically regulated transcriptional activity.

FOOTNOTES

1   Supported by grants DK 21859 and DK24451 (to R.W.H.) and by the Metabolism Training Program grant DK 07319 (to D.J.K. and D.R.C.) from the National Institutes of Health. The authors are indebted to Kulwant Singh Aulak for his critical help in the preparation of this manuscript.
2   The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.
3   Deborah Crawford and Parvin Hakimi made equal contributions to this work.
4   To whom correspondence should be addressed
5   Abbreviations used: Bt2cAMP, dibutryl cyclic AMP; CAT, chloramphenicol acetyltransferase; C/EBP, CAAT/enhancer binding protein; CRE, cAMP regulatory element; CREB, cAMP regulatory element-binding protein; GRE, glucorticoid response element; GRU, glucocorticoid regulatory unit; HNF-1, hepatic nuclear factor-1; IRS, insulin response sequence; NF-1, nuclear factor-1; PEPCK, phosphoenolpyruvate carboxykinase; PPARgamma 2, peroxisome proliferator-activated receptor gamma 2.

Manuscript received 20 November 1995. Initial reviews completed 14 December 1995. Revision accepted 28 October 1996.

LITERATURE CITED

0022-3166/97 $3.00 ©1997 American Society for Nutritional Sciences



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
C. Croniger, P. Leahy, L. Reshef, and R. W. Hanson
C/EBP and the Control of Phosphoenolpyruvate Carboxykinase Gene Transcription in the Liver
J. Biol. Chem., November 27, 1998; 273(48): 31629 - 31632.
[Full Text] [PDF]

This Article