Characterization and Regulation of the Rat and Human Ghrelin Promoters
http://www.100md.com
内分泌学杂志 2005年第3期
Department of Surgery, The University of Texas Medical Branch, Galveston, Texas 77555
Address all correspondence and requests for reprints to: George H. Greeley, Jr., Ph.D, Department of Surgery, The University of Texas Medical Branch, 301 University Boulevard, Galveston, Texas 77555-0725. E-mail: ggreeley@utmb.edu
Abstract
Ghrelin is a recently discovered stomach hormone and endogenous ligand for the GH secretagogue receptor. The aim of these studies is to elucidate molecular mechanisms underlying regulation of the ghrelin gene. Distal and proximal transcription initiation sites are present. A short transcript, a product of the proximal site, showed a more widespread distribution. Two sets of 5'-upstream segments of the rat and human ghrelin genes were cloned and sequenced. Rat promoter segments upstream of the distal site showed highest activity in kidney (COS-7) and stomach (AGS) cells, whereas human promoter segments upstream of the proximal site showed highest activity in AGS and pituitary (GH3) cells in transient transfection assays. For the human, the core promoter spanned –667 to –468 bp, including the noncoding exon 1 and a short 5' sequence of intron 1. For the rat, the core promoter spanned –581 to –469 bp, and inclusion of exon 1 and a short 5'-sequence of intron 1 reduced activity by 67%. Mutation of initiator-like elements in the rat lowered activity by 20–50%, whereas in the human, all activity was abolished. Overexpression of upstream stimulatory factors increased ghrelin core promoter activity. Fasting increases stomach ghrelin expression, glucagon-a fasting-induced hormone, increased ghrelin expression in vivo in rats, and promoter activity by approximately 25–50%. Together, these findings indicate that structural differences between the rat and human ghrelin core promoters may account in part for the differences in their transcriptional regulation. Nonetheless, upstream stimulatory factor and glucagon exert similar effects on regulation of rat and human ghrelin promoters.
Introduction
GHRELIN IS A recently discovered 28-amino-acid peptide that is the endogenous ligand for the GH secretagogue receptor (1, 2). Ghrelin was isolated from rat stomach extracts by means of the orphan receptor cloning methodology-crude tissue extracts were screened for their abilities to activate changes in [Ca2+]i in CHO cells transfected with GH secretagogue receptor. Although the highest tissue concentration of ghrelin is found in the mucosal epithelium of the stomach (1), ghrelin is expressed widely in the body. Either the ghrelin transcript or peptide immunoreactivity have been identified in the intestine, pancreas, pituitary, hypothalamus, testes, kidney, heart, and placenta as well as in a variety of tumors (1, 3, 4, 5, 6, 7, 8, 9).
Ghrelin exerts a variety of metabolic effects including stimulation of GH secretion, food intake, body growth, adiposity, gastric emptying, and acid secretion (1, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20). Ghrelin also influences insulin secretion (21, 22). Together, the expression and activity profiles of ghrelin strongly support the notion that ghrelin is an important metabolic hormone.
Ghrelin levels in the bloodstream show a pulsatile secretion (23) with higher plasma levels during food restriction and lower levels after food intake (22, 24). Stomach ghrelin expression also increases with food restriction (22, 25). Furthermore, our laboratory has reported that high-fat and low-protein diets influence stomach ghrelin expression and secretion (22).
The molecular mechanisms underlying regulation of ghrelin gene transcription are not known. The purpose of the present studies, therefore, was to clone the 5'-upstream regulatory regions of the rat and human ghrelin genes and to identify their core promoters by deletional analyses in multiple cell types. Two transcription initiation sites are present; therefore, we screened various cell lines and rat tissues for expression of these two transcripts. The importance of predicted TATA boxes and putative initiator sites in regulation of rat and human ghrelin promoter activity was examined by mutational analyses. Scanning mutational analyses and EMSAs were used to examine for cis-elements in core promoters. Because potential upstream stimulatory factor (USF) binding sites were identified in the rat and human core promoters, the influence of overexpression of either USF1 or USF2 on ghrelin core promoter activity was examined. Additionally, the hypothesis that a fasting hormone, glucagon (26, 27), increases gastric ghrelin expression in vivo was tested. The influence of glucagon on ghrelin promoter activity was also examined.
Materials and Methods
Reagents
Buffers, chemicals, and oligonucleotides were from Sigma (St. Louis, MO) and Invitrogen (Carlsbad, CA). Cell culture media were purchased from Mediatech, Inc. (Herndon, VA) and Life Technologies, Inc. (Carlsbad, CA). Site-directed mutagenesis kits were purchased from Stratagene (La Jolla, CA), and the nuclear protein extraction kit was purchased from Active Motif (Carlsbad, CA). An RNAqueous RNA kit (Ambion, Austin, TX) was used to prepare total cellular RNA from cells. RT-PCR kits were purchased from BD Biosciences (Palo Alto, CA).
Cell culture, transient transfection, and luciferase assays
All cells were maintained in a humidified incubator at 37 C in an atmosphere of 5% CO2-95% air. Human stomach adenocarcinoma cells (AGS cells) (28) were grown in Ham’s F10 media containing 10% fetal bovine serum (FBS). Human placental BeWo cells (29), and human medullary thyroid TT cells (30) were grown in Ham’s F12K medium containing 10% FBS. Rat heart H9C2 cells were cultured in DMEM with 10% FBS (31). Rat pituitary GH3 cells were grown in Ham’s F-12K containing 15% horse serum and 2.5% FBS. PC-12 rat pheochromocytoma cells were cultured in RPMI 1640 containing 5% FBS and 5% horse serum. Rat stomach RGM1 cells were cultured in DMEM/F-12 medium (1:1) containing 20% FBS (32). African green monkey kidney fibroblast cells (COS-7), mouse embryonic fibroblast cells (3T3), and mouse hypothalamic cells (GT1–7) were maintained in DMEM with 4.5 g/liter glucose and 10% FBS. MCF10A cells (33, 34) were maintained in DMEM/F-12 medium (1:1) supplemented with 5% FBS, 10 μg/ml insulin, 100 ng/ml cholera toxin, 0.5 μg/ml hydrocortisone, 20 ng/ml recombinant human epidermal growth factor, and 1.05 mM CaCl2. All media were supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin.
For transient transfection studies, cells were plated onto six-well tissue culture plates at their optimal densities. Approximately 24 h after cell plating, cells were cotransfected with ghrelin promoter-luciferase reporter gene constructs (2 μg) and an internal control vector pRL-TK (80 ng), pRL-null (80 ng), or pSV-?-galactosidase (100 ng) per well using Lipofectamine reagent (Invitrogen) following the manufacturer’s instructions. For the USF experiments, MCF10A cells were cotransfected with USF1 or USF2 expression plasmids (pN3 and pN4, kindly provided by M. Sawadogo, Houston, TX) (35) and ghrelin core promoter constructs. The influence of the empty vector (pSG424) was also tested, and pRL-SV40 (20 ng) was used as an internal control vector. Plasmids were diluted with Opti-MEM (Life Technologies, Inc.), mixed, and then transfected in triplicate. Five hours later, media were changed to media-containing serum. Cells were either harvested 40 h later or grew overnight and then serum-starved for 8 h before treatment with chemicals. Treatments were for 16 h unless specified. All cells were rinsed with PBS and lysed in 200 μl Passive Lysis Buffer (Promega, Madison, WI). Cell lysates were analyzed for firefly and renilla luciferase activities using the Dual-Luciferase Reporter Assay System (Promega) in a luminometer (Monolight 2010, Analytical Luminescence Laboratory, San Diego, CA) as recommended by the manufacturer. Chemiluminescence measurements were made at 10-sec intervals. ?-Galactosidase activity was measured according to the manufacturer’s protocol (Promega). All plasmids and constructs were purified with the Qiagen plasmid purification kit.
In the glucagon treatment experiment, stomach AGS cells were transfected with a glucagon receptor (PMT5/GR) expression vector (1 μg) (36),1 μg ghrelin promoter-reporter gene constructs, and 80 ng control plasmid.
RNA preparation and RT-PCR
Poly [A]+ RNA was prepared as described previously (37). RT-PCR was used to define expression of the short and long ghrelin transcripts in RNA extracts of either various cell lines or tissues harvested from adult male and female Sprague Dawley rats as described previously (37). All procedures on animals were performed by properly qualified personnel in accordance with the mandated standards of human care and were approved by the University of Texas Medical Branch Institutional Animal Care and Use Committee (Galveston, TX). RT-PCR products were subjected to electrophoresis on a 2% agarose gel, and DNA was visualized by ethidium bromide staining. Forward primers for the mouse short and long ghrelin gene transcripts were 5'-TGCTGTCTTCAGGCACCATCT-3' and 5'-ACATCCCCAGGCATTCCAG-3', respectively, and the reverse primer for both transcripts was 5'-TTACTTGTCAGCTGGCGCCT-3'. Forward primers for the rat short and long ghrelin gene transcripts were 5'-ATGGTGTCTTCAGCGACTATCTGC-3' and 5'-ACATCCCCAGGCATTCCAG-3', respectively, and the reverse primer for both transcripts was 5'-GTCAGTGGTTACTTGTTAGCTGGC-3'. Forward primers for the human ghrelin gene short and long transcripts were 5'-TGAACACCAGAGAGTCCAGCAGAGA-3' and 5'-ACCTCCGCCAGGAACTGCAG-3', respectively, and the reverse primer for both transcripts was 5'-CTGAGCTTGTACAACAGTCGTGGGA-3'.
Cloning of the rat and human ghrelin promoters and construction of luciferase expression vectors containing rat and human ghrelin 5'-upstream sequences
The rat ghrelin 5'-upstream regulatory sequence was obtained by screening a rat bacterial artificial chromosome library obtained from Incyte (Palo Alto, CA). An approximately 1.4-kbp fragment was cloned into pCRII-TOPO vector using rat genomic DNA (forward primer, 5'-TTCGGGGTACCAGCAGTACCTCCTGCCT-3'; reverse primer, 5'- CTGGAATGCCTCGAGGATGTGGTGCCTGCT-3'). A 3143-bp of the 5'-upstream regulatory region of the human ghrelin gene (–3161/–18) was cloned by PCR amplification using human genomic DNA (BD Biosciences) and inserted into pCRII-TOPO vector (Invitrogen) (forward primer, 5'-AGGTGGCAGGCCCTCAAGAGACCCGCAACAGA-3'; reverse primer, 5'-AGACAGGTGGGCCTGGGGGAGAGAGGGTCT-3'). 5'-Upstream fragments of various sizes were then obtained by PCR using the TOPO vector containing the cloned rat and human promoters and subcloned into the KpnI and XhoI sites of the promoterless luciferase expression vector pGL3-Basic (Promega). The primers used were designed to include an artificial restriction endonuclease site for KpnI and XhoI. For both species, sequences of the 5'-upstream clones were verified by primer walking.
Site-directed mutagenesis of the rat and human ghrelin 5'-upstream sequences
The –581/–469-bp rat ghrelin promoter construct and the –667/–468-bp human ghrelin promoter construct (the translation start site AUG was designated as +1) were mutated by PCR using the QuikChange Site-Directed Mutagenesis Kit (Stratagene). After 18 cycles of PCR using plaque-forming unit Turbo DNA polymerase at 95 C for 30 sec, 55 C for 1 min, and 68 C for 10 min, the parental, double-stranded DNA was digested with DpnI. Mutant sequences were confirmed by DNA sequencing.
Preparation of nuclear extracts and EMSAs
Nuclear extracts were prepared according to the nuclear protein extraction kit (Active Motif). Cells were prewashed with ice-cold PBS containing phosphatase inhibitors, scraped into fresh ice-cold PBS/phosphatase inhibitors, and centrifuged at 1000 rpm at 4 C for 5 min. Pellets were resuspended in cold 1x hypotonic buffer, lysed by gentle pipetting, and incubated on ice for 15 min. Detergent was added (1/20 volume) and vortexed for 10 sec. After centrifugation at 14,000 x g for 30 sec, nuclei were resuspended in Complete Lysis Buffer by pipetting, vortexed for 10 sec, and then incubated on a shaker for 30 min at 4 C. After centrifugation at 14,000 x g for 10 min at 4 C, supernatants were aliquoted into chilled microcentrifuge tubes and stored at –80 C. Protein concentrations were assayed by the Bradford assay (Bio-Rad Laboratories, Inc., Richmond, CA).
Rat and human ghrelin promoter fragments and corresponding mutated fragments were end-labeled with [-32P]ATP (3000 Ci/mmol) using T4 polynucleotide kinase. Approximately 50,000 cpm of 32P-labeled DNA was added to the nuclear extract, and preparations were incubated with 50 ng/μl polydeoxyinosinic deoxycytidylic acid · polydeoxyinosinic deoxycytidylic acid and 5 ng/μl poly L-lysine in binding buffer [20 mM HEPES (pH 7.6), 1 mM dithiothreitol, 30 mM KCI, 10 mM (NH4)2SO4, 0.2% Tween 20, and 1 mM EDTA] at room temperature for 20 min. DNA-protein complexes were resolved on a 4% polyacrylamide gel in 0.5x Tris-borate, EDTA buffer. Gels were dried and autoradiographed with intensifying screens at –70 C. For competition experiments, nuclear extracts were incubated with a 50-fold molar excess of double-stranded, nonradiolabeled oligonucleotides at room temperature for 10 min before adding radiolabeled oligonucleotides.
Glucagon treatment of rats, measurement of stomach ghrelin mRNA levels
Adult male Sprague Dawley rats were housed in an air-conditioned room with regulated temperature and lighting. All procedures on animals were performed by properly qualified personnel in accordance with the mandated standards of human care and were approved by the University of Texas Medical Branch Institutional Animal Care and Use Committee. Rats were treated with glucagon (200 μg/kg, 3 times a day sc for 3 d). The stomach was harvested, and RNA was prepared as described previously (38). In brief, stomach tissue samples were removed quickly after animals were killed and immediately homogenized in 4 M guanidinium isothiocyanate containing 25 mM sodium citrate (pH 7.0), 0.5% sodium lauroylsarcosine, and 0.1 M ?-mercaptoethanol. Extracts were frozen at –80 C until purification by ultracentrifugation over a cesium chloride cushion (2 ml, 5.7 M). Total cellular RNA samples were then separated on a 1% agarose gel (10 μg/lane) in a 20 mM 3-[N-morpholino] propanesulfonic acid running buffer system and then transferred to a nylon membrane and subjected to Northern hybridization. 32-P-labeled riboprobes prepared from Strip-EZ RNA kits (Ambion Inc.) were used for Northern hybridizations. The rat ghrelin riboprobe was supplied by Kojima (1). Expression levels of ghrelin or the 18S ribosomal genes were quantitated by densitometry.
Statistical analyses
Results are expressed as the mean ± SEM. Data were analyzed by t test or one-way ANOVA and subsequently with Newman-Keuls test when appropriate. P < 0.05 was considered significant.
Results
Cloning and deletional analyses of the 5'-flanking segments of rat and human ghrelin genes
The rat ghrelin promoter was obtained by screening a rat bacterial artificial chromosome library (Incyte) using PCR. A positive clone with an approximately 1.4-kbp DNA insert was obtained and sequenced. The GenBank accession number for the rat sequence is AY701847. Using the information provided by the human genome database on a National Center for Biotechnology Information web site, the 5'-upstream regulatory region of the human ghrelin gene was cloned and sequenced. Its accession number is AY701846. Schematic representations of the 5'-flanking segments of both rat and human ghrelin genes are shown in Fig. 1. The proximal segments show a high sequence identity (65%); however, outside the proximal segments, identity is not detected. Potential TATA boxes are identified –504 to –448 bp and –585 to –579 bp from the respective translation start sites in the rat and human ghrelin genes. CAAT and GGGCG boxes were not identified in the proximal promoter segments. Potential initiator sites were identified –476 to –469 bp and –557 to –550 bp from the respective translation start sites in the rat and human ghrelin genes.
FIG. 1. A, Schematic representation of the 5'-upstream segments of the rat and human ghrelin genes. Locations of exons 1 and 2 are shown. The nucleotide position +1 corresponds to the translation start codon ATG. D and P, Distal and proximal transcription initiation sites, respectively. B, Comparison of the nucleotide sequences of the core promoter regions of the rat and human ghrelin genes. Putative TATA boxes (underlined) and initiators (bold) are identified. Exon 1 is boxed.
In the original report describing the rat and human ghrelin cDNAs (1), the transcription initiation site was identified 30 and 32 bp upstream of the translation start sites, respectively. However, a subsequent report (39) identified an untranslated exon of 19 and 20 bp upstream of the reported translation start sites in the rat and human sequences. Therefore, there are two putative transcription initiation sites, with the proximal site located at –30 bp for the rat, and –32 bp for the human, and the distal site located at –474 bp for the rat and –555 bp for the human (the translation start site was defined as +1). Therefore, two different rat and human 5'-upstream fragments, –1886/–469 bp and –1886/–6 bp for the rat and –3161/–555 bp and –3161/–18 bp for the human, were cloned into the pGL3-Basic plasmid and their activities compared in transient transfection assays. In human stomach AGS cells, the rat ghrelin proximal promoter fragment, –1886/–6 bp, and its deletions showed no activity when compared with the promoterless pGL3-Basic vector, whereas the distal promoter fragment, –1886/–469 bp, showed an 8- to 12-fold greater activity (Fig. 2A). Moreover, a high activity was obtained from a series of deletion fragments of –1886/–469 bp when compared with pGL3-Basic vector and deletion fragments of –1886/–6 bp. For the human, the distal promoter fragment, –3161/–554 bp, showed no activity when compared with the promoterless pGL3-Basic vector, whereas the proximal promoter fragment, –3161/–18 bp, showed a 4-fold greater activity (Fig. 2B). The activity of the –3161/–554-bp fragment was approximately 8-fold less than that of the –3161/–18-bp fragment. Furthermore, deletion fragments of –3161/–554 bp showed no activity. In contrast, deletion analysis of the –3161/–18-bp fragment showed a high activity for all fragments with the exception of the –197/–18-bp fragment (Fig. 2B).
FIG. 2. Deletion analyses of the 5'-upstream regions of the rat and human ghrelin genes with different transcription initiation sites. Selection of two transcription initiation sites was based on published sequences and an unrecognized exon 1 (46 ). A, Rat, promoter activity of the regions upstream of the proximal (–1886/–6 bp) and distal (–1886/–469 bp) transcription initiation sites were analyzed by deletion scanning in transient transfection experiments using AGS cells. The upstream region of the distal transcription initiation site was active, whereas the region upstream of the proximal transcription initiation sites was inactive. B, Human, promoter activity of the regions upstream of the proximal (–3161/–18 bp) and distal (–3161/–554 bp) transcription initiation sites was analyzed by deletion scanning in transient transfection experiments using AGS cells. In contrast to the rat, the upstream region of the distal transcription initiation site was inactive, whereas the region upstream of the proximal transcription initiation site was active. Activities were normalized to the activity of the promoterless vector pGL3-Basic, whose activity was set as 1. Bars, Means ± SEM of two independent experiments, each performed in triplicate.
Analysis of promoter activity in different cell types
To characterize the promoter activity of rat and human ghrelin genes, constructs containing various 5'-flanking sequences fused to a luciferase reporter gene were transiently transfected into cell lines originating from different tissues in the body (Fig. 3). 5'-Upstream rat sequences showed the highest activity in stomach AGS cells and kidney fibroblast COS-7 cells when compared with other cell types. In AGS cells, rat ghrelin promoter activity was approximately 6- to 8-fold higher when compared with the activity of the promoterless vector. In AGS cells, deletion from –1886 to –581 bp did not influence activity; nearly 100% of the maximal promoter activity was retained within the –580 bp of the 5'-flanking sequence. In kidney COS-7 cells, a nonendocrine cell line, ghrelin promoter activity was approximately 3- to 12-fold higher when compared with the activity of the promoterless plasmid. Deletion from –1886 to –642 bp increased activity 3-fold, and deletion from –641 to –581 bp increased activity marginally. In mouse fibroblasts (3T3 cells), a nonendocrine cell line, promoter activity was marginal. In PC-12 and hypothalamic GT1-7 cells, deletion from –1886 to –581 bp resulted in 7- and 3-fold elevations in activity, respectively, over that of the –1886/–469-bp construct.
FIG. 3. Deletion analyses of the rat (A) and human (B) ghrelin promoters in different cell types. Deletion constructs were generated as described in Materials and Methods. AGS (human stomach), PC-12 (rat pheochromocytoma), GH3 (rat pituitary GH cells), COS-7 (monkey kidney), GT1-7 (mouse hypothalamic), and 3T3 (mouse skin fibroblasts) cells were transiently transfected with constructs of either the rat or human ghrelin promoter fragments fused to a luciferase reporter gene (pGL3-Basic) and pRL-TK plasmid as an internal control. Luciferase activities were measured 48 h after transfection. Data are presented as mean ± SEM.
For the human, the full-length fragment –3161/–18 bp and its deletion mutants, with the exception of –197/–18 bp, were most active and unchanged in AGS, GH3, and COS-7 cells (Fig. 3B). Activities were highest in the AGS and GH3 cells when compared with other cell types. Activities of the fragments in PC-12 and 3T3 cells were low, with the exception of the –667/–18-bp fragment.
Expression of short and long ghrelin transcripts in various cell lines and rat tissues
Two transcription initiation sites are present in the ghrelin gene (Fig. 1); therefore, we screened various rat tissues and rat, mouse, and human cell lines for expression of these transcripts using RT-PCR. The short and long transcripts are products of the proximal and distal transcription initiation sites, respectively. For each species, two sets of primers were used. One set amplified both transcripts (Fig. 4, top), and another set amplified only the long transcript (Fig. 4, middle). Therefore, following densitometric analysis of the RT-PCR products, the amount of short transcript was derived by subtraction of the long transcript reading from the short plus long transcripts reading (Fig. 4, bottom). RT-PCR showed that the short transcript was present in all rat tissues examined (stomach, pituitary, adrenal, hypothalamus, kidney, heart, lung, muscle, testes, placenta) and in some cell lines (mouse hypothalamic-GT1-7, mouse fibroblasts-3T3, rat heart-H9C2, rat pituitary-GH3, rat stomach-RGM1, human placenta-BeWo, human stomach-AGS, human medullary TT cells). The short transcript was most abundant in rat GH3 and human placenta-BeWo cells and in rat pituitary and testes, followed by slightly lower levels in mouse 3T3 cells. Intermediate levels were measured in GT1-7 and AGS cells and in rat hypothalamus and placental tissues. Low levels of the short transcript were found in H9C2, RGM1, and TT cells and in rat stomach, adrenal, kidney, heart, lung, and muscle tissues. The long transcript was most abundant in human medullary TT cells and in rat stomach and placenta tissues. Intermediate levels were measured in GT1-7, 3T3, and GH3 cells. Lower levels of the long transcript were found in BeWo and AGS cells and in rat pituitary, hypothalamic and testes tissues. Although the hypothalamic and testes bands cannot be seen in Fig. 4, they are visible in the original gel. The long transcript was not detected in H9C2 and RGM1 cells nor in rat adrenal, kidney, heart, lung, and muscle tissues. Neither the short nor the long transcripts were present in rat PC12 cells.
FIG. 4. RT-PCR screening for short and long ghrelin transcripts in cells and rat tissues. Top, RT-PCR products of the short plus long transcripts. Middle, RT-PCR products of the long transcript only. Bottom, Densitometric quantitation of short and long ghrelin transcripts. The primers used to detect the short and long transcripts are described in Materials and Methods. For each species, one set of primers detects the long transcript, whereas another set of primers detects the short plus long transcripts. Levels of the short and long transcripts are derived by densitometric analysis of the short + long and long transcripts; subtraction of the desitometric readings for the long transcript from the short plus long transcripts yields short transcript levels. Three separate densitometric readings gave identical data. One microgram poly [A]+ RNA was used for the RT-PCR. The predicted sizes of the long transcript for the human, rat, and mouse are 507, 411, and 416 bp, respectively. The predicted sizes of the short plus long transcript for the human, rat, and mouse are 364, 361, and 352 bp, respectively. The products matched the predicted sizes. For the short plus long transcript, the rat hypothalamic and testes bands are not visible (middle left panel); however, they were identified in the original gel. RT(–) (no RT added) showed no reaction products. m, Mouse; r, rat; h, human.
Determination of the minimally active fragment
An earlier experiment showed that fragments upstream of the distal transcription initiation site (i.e. –555 bp) in the human 5'-flanking region have no activity, whereas fragments upstream of the proximal transcription initiation site (i.e. –32 bp) were active (Fig. 2), indicating that activity was dependent on the fragment between the two transcription initiation sites (–555 to –32 bp). Therefore, to identify the minimal fragment required for activity, 5' and 3' deletions of the –667/–18-bp fragment were done. Deletions of 100 and 200 bp (–667/–198 bp, –667/–282 bp) at the 3' end decreased activity by approximately 25% when compared with the –667/–18-bp fragment (P < 0.05) (Fig. 5A). Further 100- and 200-bp deletions (–667/–368 bp, –667/–468 bp) increased activity significantly. A 100-bp upstream fragment lacking the noncoding exon 1 (–667/–554 bp) showed no activity. A 100-bp upstream fragment plus exon 1 (–667/–534 bp) showed activity similar to the –667/–18-bp fragment. Because the –667/–468-bp fragment showed the highest activity, the influence of 5' deletions (–647/–468 bp, –587/–468 bp) on activity was tested. 5' Deletions lowered activity significantly. Therefore, in the human ghrelin promoter, the minimal fragment for activity was –667/–468 bp, which included the noncoding exon 1 and a proximal part of intron 1.
FIG. 5. A, Further deletion analysis of the upstream region (–667/–18 bp) of the human ghrelin gene in human stomach AGS cells. Sequential deletions of –667/–18 bp at the 3' end revealed a shorter fragment containing exon 1 (closed box) and a proximal fragment of intron 1 (–667/–468 bp) with significantly greater activity. 5' Deletions of –667/–468 bp lowered activity. B, Influence of exon 1 and a proximal piece of intron 1 on core promoter activity of rat ghrelin. Inclusion of exon 1 (closed box) alone did not affect activity, whereas exon 1 plus a proximal fragment of intron 1 decreased activity.
In the rat, fragments upstream of the distal transcription initiation site (i.e. –474 bp) were active, whereas fragments upstream of the proximal transcription initiation site (i.e. –30 bp) have no activity (Fig. 2). To determine whether the region downstream of the distal transcription initiation site affected activity, two fragments, –580/–454 bp, which included exon 1 only, and –580/–387 bp, which included exon 1 and the proximal part of intron 1, were tested (Fig. 5B). In contrast to the human, inclusion of exon 1 did not enhance activity, and inclusion of exon 1 and the proximal part of intron 1 decreased activity by 70%, indicating that inhibitory elements exist in this region (Fig. 5B). Therefore, the minimally active fragment for the rat is –580/–469 bp.
Effect of TATA box mutations on activity
Potential TATA box-like elements were identified in the rat and human promoter fragments (–585/–579 bp in the human 5'-upstream sequence; –504/–498 bp in the rat). Mutation of the potential TATA box in the rat promoter fragment reduced activity by 60%, whereas mutation of the potential TATA box in the human promoter reduced activity by 30% (Fig. 6). In an additional experiment, deletion of an 80-bp sequence that included the TATA box in the human promoter resulted in a similar reduction in activity (data not shown).
FIG. 6. Mutational analyses of putative TATA boxes in the rat and human ghrelin core promoters in human stomach AGS cells. TATATAA in the rat (A) and human (B) core promoters were mutated to CTGCAGA, and activity was analyzed in transient transfection experiments.
Importance of an initiator-like element on promoter activity
Initiator-like elements were identified in the rat and human promoters (rat, –476/–469; human, –557/–550). The sequences of the rat and human initiators are CCACATCC and CCACCTCC. Four-base pair mutations were done. In the rat, mutation of the first four base pairs of the initiator (Inr-M1) reduced activity by 50%, whereas mutation of the second four base pairs (Inr-M2) reduced activity by 20% (Fig. 7A). In the human, mutation of the first 4 bp (Inr-M1) of the initiator abolished activity completely, whereas mutation of the second 4 bp (Inr-M2) reduced activity by 60% (Fig. 7B). EMSAs were also done to characterize protein-DNA complex formation at the human initiator site. Oligonucleotide fragments that contained either the wild-type (WT) or mutant initiator site were used as probes or competitors (Fig. 7C). With WT probe, two specific protein-DNA complexes are identified (Fig. 7C, lanes 1 and 2). With Inr-M1 and Inr-M2 as probes, formation of protein-DNA complexes was abolished (lanes 3, 4). Use of Inr-M1 and Inr-M2 as competitors with WT probe did not affect formation of the two specific protein-DNA complexes (lanes 5, 6). Together, the EMSA and luciferase assay findings indicate the importance of the initiator sites for promoter activity in the human.
FIG. 7. Mutational analyses of initiator sites in the rat (A) and human (B) ghrelin core promoters in human stomach AGS cells. Selective 4-bp mutations were done to putative 8-bp initiator sequences in the rat and human ghrelin core promotors, and activities were analyzed in transient transfection experiments. Luciferase assays indicate that the initiator site is more important in the human ghrelin promoter than in the rat. C, EMSA analyses show the influence of mutations on the formation of protein-DNA complexes at the human initiator site. Oligonucleotide fragments (–570/–536 bp) containing the WT or mutated initiator sites (Inr-M1, Inr-M2) were used in EMSA experiments. With WT, oligonucleotide fragment formation of two specific protein-DNA complexes was observed (lanes 1 and 2). Protein-DNA complexes did not form with oligonucleotide fragments containing mutated initiator sites (lanes 3 and 4). Excess Inr-M1 and Inr-M2 did not decrease formation of protein-DNA complexes (lanes 5 and 6).
Site-directed mutagenesis of rat and human ghrelin core promoters: influence on transcriptional activity and binding
Transient transfection experiments were done initially to identify regions important for basal promoter activity in the rat ghrelin gene. Experiments were done using a series of 12 overlapping, site-directed mutations, consisting either of 5 or 6 bp, of the –580- to –526-bp region. Luciferase assays identified several important regions: M1 (–571 to –566 bp), M2 (–568 to –563 bp), M7 (–548 to –544 bp), M11 (–534 to –529 bp), and M12 (–531 to –526 bp). In general, luciferase activities decreased by 40–85% (Fig. 8A). Mutations that did not affect activity are not shown. To further pinpoint specific bases underlying the reduced activities, selective 3-bp mutations were done. Three-base pair mutations of M2, i.e. M14 and M15, and 3-bp mutations of M12, i.e. M17 and M18, reduced activities by 35–80% (Fig. 8B). Three-base pair mutations, i.e. M13 and M16 of the M1 and M11 regions, did not affect activity (data not shown). The influence of mutations on protein-DNA complex formation in the M14–M15, M17–M18, and M7 regions were examined by EMSAs. Either WT oligonucleotide fragments or mutant oligonucleotide fragments were used as probes and competitors. The WT probes for M14-15, M7 and M17-18, were called A, B, and C, respectively. With WT-A probe, two specific protein-DNA complexes that are competed by excess nonradiolabeled WT-A were observed (Fig. 8C, lanes 1 and 2, arrows). With the M14 probe, two protein-DNA complexes that migrate slightly faster than the slow moving complex (Fig. 8C, lane 3, asterisks) were observed, and the slower moving complex disappeared (Fig. 8C, lane 3). With the M15 probe, the slower moving protein-DNA complex disappeared (Fig. 8C, lane 4). With the WT-A probe, neither M14 nor M15 competed (i.e. formation of the slower protein-DNA complex was unaffected), indicating that these regions are important for protein binding (Fig. 8D, lanes 3 and 4). The EMSA findings agree with the reduced activity observed with the M14 and M15 mutations. With the WT-B probe, no protein-DNA complexes were observed (Fig. 8C, lane 5), whereas with the M7 probe, a single protein-DNA complex was seen (Fig. 8C, lane 7), indicating that the reduced luciferase activity for M7 is a result of formation of a novel protein-DNA complex. With the WT-C probe, two specific protein-DNA complexes that were competed by excess nonradioactive WT-C were observed (Fig. 8C, lanes 8 and 9, arrows). With the M17 probe, the slower moving protein-DNA complex disappeared, and three protein-DNA complexes having mobilities different from WT-C were observed (Fig. 8C, lane 10, asterisks). With the M18 probe, the slower moving protein-DNA complex disappeared (Fig. 8C, lane 11). With WT-C probe, the M17 and M18 competitors partially competed for formation of the slower moving protein-DNA complexes, indicating that these regions are important for protein binding (Fig. 8D, lanes 7 and 8). The EMSA findings agree with the reduced activity observed with the M17 and M18 mutations.
FIG. 8. Effect of site-directed mutagenesis on rat ghrelin core promoter activity in human stomach AGS cells and EMSA using nuclear extract prepared from human stomach AGS cells. A series of 12 site-directed mutations, consisting either of 5 or 6 bp, of the –580- to –526-bp region were done to identify specific sequences underlying regulation of basal activity. WT nucleotides were mutated to Ts. Mutations that reduced activity are shown in A. Based on the results of the 5- or 6-bp mutations, a series of 3-bp mutations were then done, and mutations that reduced activity are shown (B). WT ghrelin (–580/–469 bp) and mutated (M) constructs were transfected into stomach AGS cells and luciferase activities measured 48 h after transfection. Activities were normalized to ?-galactosidase activity. Each construct was tested in triplicate and data are given as the mean ± SEM. *, P < 0.05 vs. WT (–580/–469 bp). C, EMSA of the core promoter of the rat ghrelin gene. WT and mutated (M) oligonucleotide fragments were used. Three and 5-bp mutations were introduced into three regions: A (–571 to –563 bp), B (–548 to –544 bp), and C (–534 to –526 bp). EMSA studies were done using either WT oligonucleotide fragments or corresponding mutated oligonucleotide fragments (M14, M15, M7, M17, M18) and nuclear extract from AGS cells. Double-stranded competitor DNA fragments were added in 50-fold molar excess. Specific nuclear protein-DNA complexes are identified by the arrows. Novel protein-DNA complexes are identified by asterisks. D, EMSAs were performed with WT oligonucleotides as probes and mutated fragments as competitors (M14, M15, M17, M18). Arrows indicate specific DNA-protein complexes that are competed by excess mutated fragments.
To identify important regions underlying regulation of basal promoter activity for the human ghrelin gene, transient transfection experiments were done using a series of 16 site-directed 6-bp mutations of the –570- to –468-bp regions, which include exon 1 and part of intron 1. Mutations M1 (–567 to –562 bp) and M14 (–483 to –478 bp) lowered activity 60 and 20%, respectively (Fig. 9A). EMSA experiments were then done to examine the influence of these mutations on protein-DNA complex formation. Either WT oligonucleotide fragments of the core promoter region (identified as A and B) or mutant oligonucleotide fragments (M1 and M14) were used as probes and competitors. With a WT-A probe, four protein-DNA complexes formed that were decreased by excess WT-A competitor (Fig. 9B, lanes 1 and 2, arrows). With the M1 probe, the two fast-moving complexes disappeared (Fig. 9B, lane 3). With the WT-A probe, the M1 competitor failed to compete for formation of the faster moving complexes, whereas M1 decreased formation of the slower moving complexes (Fig. 9B, lane 4), indicating that this region is important for binding. These findings agree with the reduced luciferase activity with the M1 mutation (Fig. 9A). Using the WT-B probe, no protein-DNA complex formed (Fig. 9B, lane 5), whereas using M14 as a probe, a novel protein-DNA complex was observed (Fig. 9B, lane 7, asterisk) that may be associated with the decreased luciferase activity with M14 mutation.
FIG. 9. Effect of site-directed mutagenesis on human ghrelin core promoter activity in human stomach AGS cells and EMSA using nuclear extract prepared from human stomach AGS cells. A series of 16 site-directed mutations, consisting of 6 bp, of the –570- to –468-bp region were done to identify specific sequences underlying regulation of core promoter activity. WT nucleotides were mutated to Ts. A, Mutations that reduced activity are shown. Each construct was tested in triplicate and data are given as the mean ± SEM. *, P < 0.05 vs. WT. B, EMSA of the core promoter region of the human ghrelin gene. Both WT and mutated (M) oligonucleotide fragments were used as probes and competitors. Six-base pair mutations were introduced into two regions: A (–570 to –536 bp) and B (–498 to –468 bp). EMSA studies were done using either WT oligonucleotide fragments (A, B) or corresponding mutated oligonucleotide fragments (M1, M14) and nuclear extract from AGS cells. Double-stranded competitor DNA fragments were added in 50-fold molar excess. Specific nuclear protein-DNA complexes are identified by the arrows. Novel protein-DNA complexes are identified by asterisks.
Role of USF1 and USF2 in regulation of rat and human ghrelin genes
USF binding sites were identified in the rat and human ghrelin promoter regions (rat, –516/–511 bp; human, –566/–561 bp). The sequences were: rat, CAGCAG; and human, CACCAG. Overexpression of USF1 in MCF10A cells transfected with the rat and human ghrelin core promoters increased activities 1- and 4-fold, respectively (Fig. 10). MCF10A cells (normal breast epithelium) were used because USF1 and USF2 activities are optimal in MCF10A cells (33, 34). Overexpression of USF2 in MCF10A cells increased activities 5- and 9-fold of the rat and human ghrelin core promoters, respectively.
FIG. 10. Influence of USF1 and USF2 overexpression on rat (A) and human (B) ghrelin promoter activities. MCF10A cells were cotransfected either with rat or human ghrelin core promoter plus USF1 or USF2 expression vectors. Cells were also cotransfected with an empty vector (pSG424). Luciferase activities were normalized to the empty vector readings. pRLSV40 served as an internal control.
Glucagon increases stomach ghrelin mRNA expression and promoter activity
Because systemic glucagon levels increase with food restriction (26, 27), the effect of glucagon on gastric ghrelin expression was examined in vivo. Glucagon administration (200 μg, 3x/d, for 3 d, SC) increased steady-state stomach ghrelin mRNA levels significantly (Fig. 11). Ghrelin mRNA levels increased 1.3-fold in glucagon-treated rats when compared with ghrelin mRNA levels in vehicle-treated controls. Glucagon stimulation of ghrelin promoter activity is one possible mechanism behind the glucagon-induced elevation in stomach ghrelin expression. Therefore, the effect of glucagon on ghrelin promoter activity was tested. In the rat and human, glucagon treatment at 1 x 10–6 M increased activity by approximately 25–50% (Fig. 11).
FIG. 11. Glucagon increases rat ghrelin mRNA levels and rat and human ghrelin promoter activity. A, Northern analysis showed that glucagon treatment (200 μg, 3x/d, x1 d, sc) increased stomach ghrelin expression in rats. B, Densitometric analysis of hybridization signals. Data are shown as mean ± SEM of ghrelin mRNA readings over 18S ribosomal RNA densitometric readings. n = 10 rats/group. *, P < 0.05 vs. control-treated rats. C, –543/–469 bp and –3161/–18 bp of the rat and human 5'-flanking regions of the ghrelin genes fused to the luciferase reporter gene and a glucagon receptor expression vector (PMT5/GR) were transfected into stomach AGS cells and treated with 1 x 10–6 M glucagon. Luciferase activities were measured 16 h after the start of glucagon treatment and are expressed as mean ± SEM. Glucagon treatment increased activity significantly. *, P < 0.05 vs. control treatment.
Discussion
Ghrelin is a recently discovered stomach hormone that influences a variety of metabolic activities, including GH and insulin secretion, body growth, food intake, and adiposity (1, 22, 40, 41, 42). The aim of the present work was to clone and characterize the 5'-upstream regulatory segments of the rat and human ghrelin genes to elucidate molecular mechanisms underlying regulation of ghrelin gene transcription.
5'-Flanking segments for rat and human ghrelin genes were cloned and sequenced. Earlier reports identified two putative transcription initiation sites in the ghrelin promoter (1, 39). A proximal initiation site is located at position –30 for the rat and at position –32 for the human, and the distal site is located at position –474 for the rat and at position –555 for the human (the translation start site was defined as +1). Interestingly, the present study showed no transcriptional activity in a fragment upstream of the proximal transcription initiation site in the rat, whereas a similar fragment in the human showed activity. Here, we found that basal activity of the human core promoter requires a sequence downstream of the distal transcription initiation site, i.e. noncoding exon 1 and a proximal sequence of intron 1, whereas in the rat, the noncoding exon 1 is not essential for activity. In fact, 5'- and 3'-deletional scanning analyses showed that a 90-bp fragment upstream of the distal transcription initiation site conferred core promoter activity (8-fold increase in activity when compared with promoterless vector) and that inclusion of exon 1 plus a proximal piece of intron 1 inhibited basal promoter activity in the rat. Interestingly, these regulatory differences exist despite a 65% sequence identity between the core promoter regions. Outside the core promoter region, no sequence identity was found. Together, these regulatory differences in the rat and human ghrelin promoters indicate that transcriptional regulation of the human ghrelin gene has diverged from the rat.
Our findings identify an essential role for exon 1 and a proximal piece of intron 1 for human ghrelin promoter activity. Several earlier reports have described the involvement of exon 1 and intron 1 in regulating transcription activity. For instance, the AP-2 element in the untranslated exon 1 of the human hematopoietic prostaglandin D synthase gene is essential for transcriptional activation (43), and multiple cis-acting elements (E-box binding sites and Ets-like element) in the untranslated exon 1 of the human GnRH-II gene function cooperatively to stimulate transcription (44). In contrast, a functional element was not identified in the untranslated exon 1 of the human ghrelin gene by means of mutagenesis assays. Exon 1 alone exerts neither promoter nor enhancer activities (data not shown) but seems to function in a cooperative manner with other elements in the core promoter region.
Other major regulatory differences in the 5'-upstream regulatory regions of the rat and human ghrelin genes include the influence of the putative initiator sites and TATA box elements on basal promoter activity. In both the rat and human, the TATA box and initiator appear to function together to define transcription start point; however, in the human ghrelin promoter, the initiator site appears to be more important because mutation of the initiator abolished transcriptional activity completely. In contrast, in the rat, mutation of the initiator diminished activity only moderately. In the rat ghrelin promoter, the putative TATA box appears more important because mutation of the TATA box decreased basal activity to a greater extent than in the human ghrelin promoter. Interestingly, an earlier report indicated that the TATA box is not functional in the human ghrelin promoter (45).
This is the first report describing the distribution of two ghrelin gene transcripts in various cell types and rat tissues. In the cells and tissues examined, the short ghrelin gene transcript shows a wider distribution pattern when compared with the distribution of the long transcript. Additionally, the distribution patterns of the transcripts confirm and extend the reports on ghrelin expression in various cell types and rat tissues (1, 6, 8, 9, 29, 46, 47, 48, 49, 50, 51, 52).
The present study demonstrates a cell type-specific activity for the rat and human ghrelin promoters. Overall, our findings indicate that many cell types support ghrelin expression. Deletion analyses in various cell types showed rat promoter activity is highest in stomach AGS and kidney COS-7 cells, whereas activity of human promoter was highest in AGS and pituitary GH3 cells. The human ghrelin promoter was also active in COS-7 cells but at a significantly lower level when compared with AGS and GH3 cells. However, an earlier report showed that a proximal fragment (–605/–1 bp) of the human ghrelin promoter was inactive in GH3 and COS-7 cells (45). In contrast to our expectations and the findings with the human ghrelin promoter fragments, rat ghrelin promoter fragments showed no activity in GH3 cells. Ghrelin is expressed in the pituitary (1, 8, 47, 53), and both ghrelin transcripts are detected in GH3 cells; therefore, we would anticipate the rat ghrelin promoter to be active in GH3 cells. Furthermore, the larger rat (–1886/–6 bp) construct is not active in rat GH3 cells (data not shown). At this point, we cannot offer an explanation for the absence of rat ghrelin promoter activity in GH3 cells. Activities of rat distal constructs (–1886/–469 bp; –1116/–469 bp) were lower than a proximal construct (–641/–469 bp) in COS-7 cells and less so in hypothalamic GT1-7 and adrenal PC-12 cells, suggesting that negative elements in the distal promoter region inhibit transcription to varying degrees in these cells. In AGS cells, activities of the rat promoter segments are similar, suggesting that the negative element is not functional in AGS cells or that a transcription factor having inhibitory activity is not present in AGS cells. Interestingly, activity of a proximal human promoter fragment (–667/–18 bp) is greatly increased in PC-12, GTI-7, and 3T3 cells, implying that negative elements in the distal human ghrelin promoter are differentially activated. With a few exceptions, rat and human ghrelin promoter segments showed marginal activity in rat hypothalamic GTI-7 and adrenal PC-12 cells and in mouse fibroblasts (3T3). The low promoter activity in PC-12 cells is expected based upon the absence of ghrelin expression in these cells. However, the low ghrelin promoter activity in 3T3 and GT1-7 cells does not correlate with detection of ghrelin transcripts in these cells. 3T3 and GT1-7 cells are murine derived and may not support rat and human ghrelin promoters.
Elements essential for basal activity in the rat and human ghrelin core promoters were identified using mutational analyses and EMSAs. Transient transfection experiments identified three regions in the rat, –568/–563, –548/–544, and –531/–526 bp, and two regions in the human, –567/–562 and –483/–478 bp, that are important for basal activity. In general, EMSAs showed that reductions in transcriptional activity are associated with a decrease in formation of specific protein-DNA complexes. Although exon-1 and a proximal piece of intron 1 were essential for core promoter activity in the human, mutational analyses and EMSAs failed to identify a single cis-element underlying promoter activity, suggesting that activity of the human ghrelin promoter requires interactions of multiple regulatory elements. In two instances, a reduction in activity was associated with formation of novel protein-DNA complexes. It should be mentioned that we used scanning mutations to identify regions important in the core promoter for activity. A potential shortcoming with this approach is that mutations may create cis-acting elements.
USF1 and USF2 are ubiquitously expressed basic helix-loop-helix-leucine zipper transcription factors that bind as homodimers and heterodimers to DNA sequences centered on CAC(G/A)TG (54, 55). Sequence analyses of both human and rat core promoters identified putative USF sites, CACCAG and CAGCAG, which are not perfect matches to consensus sequences. Our results show that USF may play a stimulatory role in the activation of the rat and human ghrelin promoters. Interestingly, USF levels increase during protein deprivation (56). Our laboratory has reported that stomach ghrelin expression increases in rats fed low-protein diets (22). The increased stomach ghrelin gene expression may be due to increased binding of USF to the ghrelin promoter.
Expression and secretion of gastric ghrelin increase with food restriction and decrease with food intake (22, 24, 40, 41, 57, 58). The molecular mechanism underlying this increased gastric ghrelin production with fasting is not known but may reflect actions of other enteric or pancreatic hormones that are altered by changes in nutritional status. The influence of glucagon on stomach ghrelin expression was explored because glucagon secretion increases with fasting (26, 27). In the present study, glucagon treatment up-regulated ghrelin expression levels in the rat, and we confirmed an earlier report showing that glucagon apparently increased steady-state ghrelin mRNA levels by a stimulation of ghrelin transcription because glucagon stimulates rat and human ghrelin promoter activity (45).
Abundant reports indicate that ghrelin is an important new hormone (42, 46). Ghrelin is widely expressed in the body and exerts a diverse array of metabolic activities. In this study, we have characterized the 5'-upstream regulatory regions of the rat and human ghrelin genes and shown that major differences exist in regulation of their core promoter activities and in regulation of ghrelin promoter activity at the tissue and cellular levels. The pancreatic hormone, glucagon, and the transcription factors, USFs, may be two metabolically sensitive factors that regulate ghrelin gene expression.
References
Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K 1999 Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402:656–660
Tomasetto C, Karam SM, Ribieras S, Masson R, Lefebvre O, Staub A, Alexander G, Chenard MP, Rio MC 2000 Identification and characterization of a novel gastric peptide hormone: the motilin-related peptide. Gastroenterology 119:395–405
Cassoni P, Papotti M, Ghe C, Catapano F, Sapino A, Graziani A, Deghenghi R, Reissmann T, Ghigo E, Muccioli G 2001 Identification, characterization, and biological activity of specific receptors for natural (ghrelin) and synthetic growth hormone secretagogues and analogs in human breast carcinomas and cell lines. J Clin Endocrinol Metab 86:1738–1745
Date Y, Nakazato M, Hashiguchi S, Dezaki K, Mondal MS, Hosoda H, Kojima M, Kangawa K, Arima T, Matsuo H, Yada T, Matsukura S 2002 Ghrelin is present in pancreatic -cells of humans and rats and stimulates insulin secretion. Diabetes 51:124–129
Guan XM, Yu H, Palyha OC, McKee KK, Feighner SD, Sirinathsinghji DJ, Smith RG, Van der Ploeg LH, Howard AD 1997 Distribution of mRNA encoding the growth hormone secretagogue receptor in brain and peripheral tissues. Brain Res Mol Brain Res 48:23–29
Kim K, Arai K, Sanno N, Osamura RY, Teramoto A, Shibasaki T 2001 Ghrelin and growth hormone (GH) secretagogue receptor (GHSR) mRNA expression in human pituitary adenomas. Clin Endocrinol (Oxf) 54:759–768
McKee KK, Tan CP, Palyha OC, Liu J, Feighner SD, Hreniuk DL, Smith RG, Howard AD, Van der Ploeg LH 1997 Cloning and characterization of two human G protein-coupled receptor genes (GPR38 and GPR39) related to the growth hormone secretagogue and neurotensin receptors. Genomics 46:426–434
Muccioli G, Papotti M, Locatelli V, Ghigo E, Deghenghi R 2001 Binding of 125I-labeled ghrelin to membranes from human hypothalamus and pituitary gland. J Endocrinol Invest 24:RC7–RC9
Shuto Y, Shibasaki T, Wada K, Parhar I, Kamegai J, Sugihara H, Oikawa S, Wakabayashi I 2001 Generation of polyclonal antiserum against the growth hormone secretagogue receptor (GHS-R): evidence that the GHS-R exists in the hypothalamus, pituitary and stomach of rats. Life Sci 68:991–996
Arvat E, Di Vito L, Broglio F, Papotti M, Muccioli G, Dieguez C, Casanueva FF, Deghenghi R, Camanni F, Ghigo E 2000 Preliminary evidence that Ghrelin, the natural GH secretagogue (GHS)-receptor ligand, strongly stimulates GH secretion in humans. J Endocrinol Invest 23:493–495
Date Y, Kojima M, Hosoda H, Sawaguchi A, Mondal MS, Suganuma T, Matsukura S, Kangawa K, Nakazato M 2000 Ghrelin, a novel growth hormone-releasing acylated peptide, is synthesized in a distinct endocrine cell type in the gastrointestinal tracts of rats and humans. Endocrinology 141:4255–4261
Kamegai J, Tamura H, Shimizu T, Ishii S, Sugihara H, Wakabayashi I 2000 Central effect of ghrelin, an endogenous growth hormone secretagogue, on hypothalamic peptide gene expression. Endocrinology 141:4797–4800
Masuda Y, Tanaka T, Inomata N, Ohnuma N, Tanaka S, Itoh Z, Hosoda H, Kojima M, Kangawa K 2000 Ghrelin stimulates gastric acid secretion and motility in rats. Biochem Biophys Res Commun 276:905–908
Nakazato M, Murakami N, Date Y, Kojima M, Matsuo H, Kangawa K, Matsukura S 2001 A role for ghrelin in the central regulation of feeding. Nature 409:194–198
Peino R, Baldelli R, Rodriguez-Garcia J, Rodriguez-Segade S, Kojima M, Kangawa K, Arvat E, Ghigo E, Dieguez C, Casanueva FF 2000 Ghrelin-induced growth hormone secretion in humans. Eur J Endocrinol 143:R11–R14
Seoane LM, Tovar S, Baldelli R, Arvat E, Ghigo E, Casanueva FF, Dieguez C 2000 Ghrelin elicits a marked stimulatory effect on GH secretion in freely-moving rats. Eur J Endocrinol 143:R7–R9
Takaya K, Ariyasu H, Kanamoto N, Iwakura H, Yoshimoto A, Harada M, Mori K, Komatsu Y, Usui T, Shimatsu A, Ogawa Y, Hosoda K, Akamizu T, Kojima M, Kangawa K, Nakao K 2000 Ghrelin strongly stimulates growth hormone release in humans. J Clin Endocrinol Metab 85:4908–4911
Tschop M, Smiley DL, Heiman ML 2000 Ghrelin induces adiposity in rodents. Nature 407:908–913
Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, Dhillo WS, Ghatei MA, Bloom SR 2001 Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 86:5992
Wren AM, Small CJ, Ward HL, Murphy KG, Dakin CL, Taheri S, Kennedy AR, Roberts GH, Morgan DG, Ghatei MA, Bloom SR 2000 The novel hypothalamic peptide ghrelin stimulates food intake and growth hormone secretion. Endocrinology 141:4325–4328
Reimer MK, Pacini G, Ahren B 2003 Dose-dependent inhibition by ghrelin of insulin secretion in the mouse. Endocrinology 144:916–921
Lee HM, Wang G, Englander EW, Kojima M, Greeley Jr GH 2002 Ghrelin, a new gastrointestinal endocrine peptide that stimulates insulin secretion: enteric distribution, ontogeny, influence of endocrine, and dietary manipulations. Endocrinology 143:185–190
Tannenbaum GS, Epelbaum J, Bowers CY 2003 Interrelationship between the novel peptide ghrelin and somatostatin/growth hormone-releasing hormone in regulation of pulsatile growth hormone secretion. Endocrinology 144:967–974
Gomez G, Englander EW, Greeley Jr GH 2004 Nutrient inhibition of ghrelin secretion in the fasted rat. Regul Pept 117:33–36
Asakawa A, Inui A, Kaga T, Yuzuriha H, Nagata T, Fujimiya M, Katsuura G, Makino S, Fujino MA, Kasuga M 2001 A role of ghrelin in neuroendocrine and behavioral responses to stress in mice. Neuroendocrinology 74:143–147
Seino S, Sakurai H, Seino Y, Tsuda K, Tanigawa K, Kuzuya H, Goto Y, Imura H 1980 Starvation-induced changes of somatostatin, glucagon, and insulin secretion from the isolated perfused rat pancreas. Diabetes 29:323–325
Unger RH 1971 Glucagon physiology and pathophysiology. N Engl J Med 285:443–449
Hocker M, Zhang Z, Merchant JL, Wang TC 1997 Gastrin regulates the human histidine decarboxylase promoter through an AP-1-dependent mechanism. Am J Physiol 272:G822–G830
Gualillo O, Caminos J, Blanco M, Garcia-Caballero T, Kojima M, Kangawa K, Dieguez C, Casanueva F 2001 Ghrelin, a novel placental-derived hormone. Endocrinology 142:788–794
Kanamoto N, Akamizu T, Hosoda H, Hataya Y, Ariyasu H, Takaya K, Hosoda K, Saijo M, Moriyama K, Shimatsu A, Kojima M, Kangawa K, Nakao K 2001 Substantial production of ghrelin by a human medullary thyroid carcinoma cell line. J Clin Endocrinol Metab 86:4984–4990
Kashour T, Burton T, Dibrov A, Amara F 2003 Myogenic signaling by lithium in cardiomyoblasts is Akt independent but requires activation of the beta-catenin-Tcf/Lef pathway. J Mol Cell Cardiol 35:937–951
Sasaki E, Pai R, Halter F, Komurasaki T, Arakawa T, Kobayashi K, Kuroki T, Tarnawski AS 1998 Induction of cyclooxygenase-2 in a rat gastric epithelial cell line by epiregulin and basic fibroblast growth factor. J Clin Gastroenterol 27(Suppl 1):S21–S27
Ismail PM, Lu T, Sawadogo M 1999 Loss of USF transcriptional activity in breast cancer cell lines. Oncogene 18:5582–5591
Szentirmay MN, Yang HX, Pawar SA, Vinson C, Sawadogo M 2003 The IGF2 receptor is a USF2-specific target in nontumorigenic mammary epithelial cells but not in breast cancer cells. J Biol Chem 278:37231–37240
Meier JL, Luo X, Sawadogo M, Straus SE 1994 The cellular transcription factor USF cooperates with varicella-zoster virus immediate-early protein 62 to symmetrically activate a bidirectional viral promoter. Mol Cell Biol 14:6896–6906
Carruthers CJ, Unson CG, Kim HN, Sakmar TP 1994 Synthesis and expression of a gene for the rat glucagon receptor. Replacement of an aspartic acid in the extracellular domain prevents glucagon binding. J Biol Chem 269:29321–29328
Wang G, Anini Y, Wei W, Qi X, AM OC, Mochizuki T, Wang HQ, Hellmich MR, Englander EW, Greeley Jr GH 2004 Apelin, a new enteric peptide: localization in the gastrointestinal tract, ontogeny, and stimulation of gastric cell proliferation and of cholecystokinin secretion. Endocrinology 145:1342–1348
Gomez G, Padilla L, Udupi V, Tarasova N, Sundler F, Townsend Jr CM, Thompson JC, Greeley Jr GH 1996 Regulation of peptide YY homeostasis by gastric acid and gastrin. Endocrinology 137:1365–1369
Tanaka M, Hayashida Y, Iguchi T, Nakao N, Nakai N, Nakashima K 2001 Organization of the mouse ghrelin gene and promoter: occurrence of a short noncoding first exon. Endocrinology 142:3697–3700
Asakawa A, Inui A, Kaga T, Yuzuriha H, Nagata T, Ueno N, Makino S, Fujimiya M, Niijima A, Fujino MA, Kasuga M 2001 Ghrelin is an appetite-stimulatory signal from stomach with structural resemblance to motilin. Gastroenterology 120:337–345
Broglio F, Arvat E, Benso A, Gottero C, Muccioli G, Papotti M, van der Lely AJ, Deghenghi R, Ghigo E 2001 Ghrelin, a natural GH secretagogue produced by the stomach, induces hyperglycemia and reduces insulin secretion in humans. J Clin Endocrinol Metab 86:5083–5086
Broglio F, Arvat E, Benso A, Papotti M, Muccioli G, Deghenghi R, Ghigo E 2002 Ghrelin: endocrine and non-endocrine actions. J Pediatr Endocrinol Metab 15(Suppl 5):1219–1227
Fujimori K, Kanaoka Y, Sakaguchi Y, Urade Y 2000 Transcriptional activation of the human hematopoietic prostaglandin D synthase gene in megakaryoblastic cells. Roles of the oct-1 element in the 5'-flanking region and the AP-2 element in the untranslated exon 1. J Biol Chem 275:40511–40516
Cheng CK, Hoo RL, Chow BK, Leung PC 2003 Functional cooperation between multiple regulatory elements in the untranslated exon 1 stimulates the basal transcription of the human GnRH-II gene. Mol Endocrinol 17:1175–1191
Kishimoto M, Okimura Y, Nakata H, Kudo T, Iguchi G, Takahashi Y, Kaji H, Chihara K 2003 Cloning and characterization of the 5(')-flanking region of the human ghrelin gene. Biochem Biophys Res Commun 305:186–192
Kojima M, Hosoda H, Matsuo H, Kangawa K 2001 Ghrelin: discovery of the natural endogenous ligand for the growth hormone secretagogue receptor. Trends Endocrinol Metab 12:118–122
Korbonits M, Kojima M, Kangawa K, Grossman AB 2001 Presence of ghrelin in normal and adenomatous human pituitary. Endocrine 14:101–104
Mori K, Yoshimoto A, Takaya K, Hosoda K, Ariyasu H, Yahata K, Mukoyama M, Sugawara A, Hosoda H, Kojima M, Kangawa K, Nakao K 2000 Kidney produces a novel acylated peptide, ghrelin. FEBS Lett 486:213–216
Rindi G, Necchi V, Savio A, Torsello A, Zoli M, Locatelli V, Raimondo F, Cocchi D, Solcia E 2002 Characterisation of gastric ghrelin cells in man and other mammals: studies in adult and fetal tissues. Histochem Cell Biol 117:511–519
Rindi G, Savio A, Torsello A, Zoli M, Locatelli V, Cocchi D, Paolotti D, Solcia E 2002 Ghrelin expression in gut endocrine growths. Histochem Cell Biol 117:521–525
Volante M, Allia E, Gugliotta P, Funaro A, Broglio F, Deghenghi R, Muccioli G, Ghigo E, Papotti M 2002 Expression of ghrelin and of the GH secretagogue receptor by pancreatic islet cells and related endocrine tumors. J Clin Endocrinol Metab 87:1300–1308
Tena-Sempere M, Barreiro ML, Gonzalez LC, Gaytan F, Zhang FP, Caminos JE, Pinilla L, Casanueva FF, Dieguez C, Aguilar E 2002 Novel expression and functional role of ghrelin in rat testis. Endocrinology 143:717–725
Muccioli G, Tschop M, Papotti M, Deghenghi R, Heiman M, Ghigo E 2002 Neuroendocrine and peripheral activities of ghrelin: implications in metabolism and obesity. Eur J Pharmacol 440:235–254
Gregor PD, Sawadogo M, Roeder RG 1990 The adenovirus major late transcription factor USF is a member of the helix-loop-helix group of regulatory proteins and binds to DNA as a dimer. Genes Dev 4:1730–1740
Murre C, McCaw PS, Baltimore D 1989 A new DNA binding and dimerization motif in immunoglobulin enhancer binding, daughterless, MyoD, and myc proteins. Cell 56:777–783
Matsukawa T, Inoue Y, Oishi Y, Kato H, Noguchi T 2001 Up-regulation of upstream stimulatory factors by protein malnutrition and its possible role in regulation of the IGF-binding protein-1 gene. Endocrinology 142:4643–4651
Gualillo O, Caminos JE, Nogueiras R, Seoane LM, Arvat E, Ghigo E, Casanueva FF, Dieguez C 2002 Effect of food restriction on ghrelin in normal-cycling female rats and in pregnancy. Obes Res 10:682–687
Hansen TK, Dall R, Hosoda H, Kojima M, Kangawa K, Christiansen JS, Jorgensen JO 2002 Weight loss increases circulating levels of ghrelin in human obesity. Clin Endocrinol (Oxf) 56:203–206(Wei Wei, Guiyun Wang, Xia)
Address all correspondence and requests for reprints to: George H. Greeley, Jr., Ph.D, Department of Surgery, The University of Texas Medical Branch, 301 University Boulevard, Galveston, Texas 77555-0725. E-mail: ggreeley@utmb.edu
Abstract
Ghrelin is a recently discovered stomach hormone and endogenous ligand for the GH secretagogue receptor. The aim of these studies is to elucidate molecular mechanisms underlying regulation of the ghrelin gene. Distal and proximal transcription initiation sites are present. A short transcript, a product of the proximal site, showed a more widespread distribution. Two sets of 5'-upstream segments of the rat and human ghrelin genes were cloned and sequenced. Rat promoter segments upstream of the distal site showed highest activity in kidney (COS-7) and stomach (AGS) cells, whereas human promoter segments upstream of the proximal site showed highest activity in AGS and pituitary (GH3) cells in transient transfection assays. For the human, the core promoter spanned –667 to –468 bp, including the noncoding exon 1 and a short 5' sequence of intron 1. For the rat, the core promoter spanned –581 to –469 bp, and inclusion of exon 1 and a short 5'-sequence of intron 1 reduced activity by 67%. Mutation of initiator-like elements in the rat lowered activity by 20–50%, whereas in the human, all activity was abolished. Overexpression of upstream stimulatory factors increased ghrelin core promoter activity. Fasting increases stomach ghrelin expression, glucagon-a fasting-induced hormone, increased ghrelin expression in vivo in rats, and promoter activity by approximately 25–50%. Together, these findings indicate that structural differences between the rat and human ghrelin core promoters may account in part for the differences in their transcriptional regulation. Nonetheless, upstream stimulatory factor and glucagon exert similar effects on regulation of rat and human ghrelin promoters.
Introduction
GHRELIN IS A recently discovered 28-amino-acid peptide that is the endogenous ligand for the GH secretagogue receptor (1, 2). Ghrelin was isolated from rat stomach extracts by means of the orphan receptor cloning methodology-crude tissue extracts were screened for their abilities to activate changes in [Ca2+]i in CHO cells transfected with GH secretagogue receptor. Although the highest tissue concentration of ghrelin is found in the mucosal epithelium of the stomach (1), ghrelin is expressed widely in the body. Either the ghrelin transcript or peptide immunoreactivity have been identified in the intestine, pancreas, pituitary, hypothalamus, testes, kidney, heart, and placenta as well as in a variety of tumors (1, 3, 4, 5, 6, 7, 8, 9).
Ghrelin exerts a variety of metabolic effects including stimulation of GH secretion, food intake, body growth, adiposity, gastric emptying, and acid secretion (1, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20). Ghrelin also influences insulin secretion (21, 22). Together, the expression and activity profiles of ghrelin strongly support the notion that ghrelin is an important metabolic hormone.
Ghrelin levels in the bloodstream show a pulsatile secretion (23) with higher plasma levels during food restriction and lower levels after food intake (22, 24). Stomach ghrelin expression also increases with food restriction (22, 25). Furthermore, our laboratory has reported that high-fat and low-protein diets influence stomach ghrelin expression and secretion (22).
The molecular mechanisms underlying regulation of ghrelin gene transcription are not known. The purpose of the present studies, therefore, was to clone the 5'-upstream regulatory regions of the rat and human ghrelin genes and to identify their core promoters by deletional analyses in multiple cell types. Two transcription initiation sites are present; therefore, we screened various cell lines and rat tissues for expression of these two transcripts. The importance of predicted TATA boxes and putative initiator sites in regulation of rat and human ghrelin promoter activity was examined by mutational analyses. Scanning mutational analyses and EMSAs were used to examine for cis-elements in core promoters. Because potential upstream stimulatory factor (USF) binding sites were identified in the rat and human core promoters, the influence of overexpression of either USF1 or USF2 on ghrelin core promoter activity was examined. Additionally, the hypothesis that a fasting hormone, glucagon (26, 27), increases gastric ghrelin expression in vivo was tested. The influence of glucagon on ghrelin promoter activity was also examined.
Materials and Methods
Reagents
Buffers, chemicals, and oligonucleotides were from Sigma (St. Louis, MO) and Invitrogen (Carlsbad, CA). Cell culture media were purchased from Mediatech, Inc. (Herndon, VA) and Life Technologies, Inc. (Carlsbad, CA). Site-directed mutagenesis kits were purchased from Stratagene (La Jolla, CA), and the nuclear protein extraction kit was purchased from Active Motif (Carlsbad, CA). An RNAqueous RNA kit (Ambion, Austin, TX) was used to prepare total cellular RNA from cells. RT-PCR kits were purchased from BD Biosciences (Palo Alto, CA).
Cell culture, transient transfection, and luciferase assays
All cells were maintained in a humidified incubator at 37 C in an atmosphere of 5% CO2-95% air. Human stomach adenocarcinoma cells (AGS cells) (28) were grown in Ham’s F10 media containing 10% fetal bovine serum (FBS). Human placental BeWo cells (29), and human medullary thyroid TT cells (30) were grown in Ham’s F12K medium containing 10% FBS. Rat heart H9C2 cells were cultured in DMEM with 10% FBS (31). Rat pituitary GH3 cells were grown in Ham’s F-12K containing 15% horse serum and 2.5% FBS. PC-12 rat pheochromocytoma cells were cultured in RPMI 1640 containing 5% FBS and 5% horse serum. Rat stomach RGM1 cells were cultured in DMEM/F-12 medium (1:1) containing 20% FBS (32). African green monkey kidney fibroblast cells (COS-7), mouse embryonic fibroblast cells (3T3), and mouse hypothalamic cells (GT1–7) were maintained in DMEM with 4.5 g/liter glucose and 10% FBS. MCF10A cells (33, 34) were maintained in DMEM/F-12 medium (1:1) supplemented with 5% FBS, 10 μg/ml insulin, 100 ng/ml cholera toxin, 0.5 μg/ml hydrocortisone, 20 ng/ml recombinant human epidermal growth factor, and 1.05 mM CaCl2. All media were supplemented with 100 U/ml penicillin and 100 μg/ml streptomycin.
For transient transfection studies, cells were plated onto six-well tissue culture plates at their optimal densities. Approximately 24 h after cell plating, cells were cotransfected with ghrelin promoter-luciferase reporter gene constructs (2 μg) and an internal control vector pRL-TK (80 ng), pRL-null (80 ng), or pSV-?-galactosidase (100 ng) per well using Lipofectamine reagent (Invitrogen) following the manufacturer’s instructions. For the USF experiments, MCF10A cells were cotransfected with USF1 or USF2 expression plasmids (pN3 and pN4, kindly provided by M. Sawadogo, Houston, TX) (35) and ghrelin core promoter constructs. The influence of the empty vector (pSG424) was also tested, and pRL-SV40 (20 ng) was used as an internal control vector. Plasmids were diluted with Opti-MEM (Life Technologies, Inc.), mixed, and then transfected in triplicate. Five hours later, media were changed to media-containing serum. Cells were either harvested 40 h later or grew overnight and then serum-starved for 8 h before treatment with chemicals. Treatments were for 16 h unless specified. All cells were rinsed with PBS and lysed in 200 μl Passive Lysis Buffer (Promega, Madison, WI). Cell lysates were analyzed for firefly and renilla luciferase activities using the Dual-Luciferase Reporter Assay System (Promega) in a luminometer (Monolight 2010, Analytical Luminescence Laboratory, San Diego, CA) as recommended by the manufacturer. Chemiluminescence measurements were made at 10-sec intervals. ?-Galactosidase activity was measured according to the manufacturer’s protocol (Promega). All plasmids and constructs were purified with the Qiagen plasmid purification kit.
In the glucagon treatment experiment, stomach AGS cells were transfected with a glucagon receptor (PMT5/GR) expression vector (1 μg) (36),1 μg ghrelin promoter-reporter gene constructs, and 80 ng control plasmid.
RNA preparation and RT-PCR
Poly [A]+ RNA was prepared as described previously (37). RT-PCR was used to define expression of the short and long ghrelin transcripts in RNA extracts of either various cell lines or tissues harvested from adult male and female Sprague Dawley rats as described previously (37). All procedures on animals were performed by properly qualified personnel in accordance with the mandated standards of human care and were approved by the University of Texas Medical Branch Institutional Animal Care and Use Committee (Galveston, TX). RT-PCR products were subjected to electrophoresis on a 2% agarose gel, and DNA was visualized by ethidium bromide staining. Forward primers for the mouse short and long ghrelin gene transcripts were 5'-TGCTGTCTTCAGGCACCATCT-3' and 5'-ACATCCCCAGGCATTCCAG-3', respectively, and the reverse primer for both transcripts was 5'-TTACTTGTCAGCTGGCGCCT-3'. Forward primers for the rat short and long ghrelin gene transcripts were 5'-ATGGTGTCTTCAGCGACTATCTGC-3' and 5'-ACATCCCCAGGCATTCCAG-3', respectively, and the reverse primer for both transcripts was 5'-GTCAGTGGTTACTTGTTAGCTGGC-3'. Forward primers for the human ghrelin gene short and long transcripts were 5'-TGAACACCAGAGAGTCCAGCAGAGA-3' and 5'-ACCTCCGCCAGGAACTGCAG-3', respectively, and the reverse primer for both transcripts was 5'-CTGAGCTTGTACAACAGTCGTGGGA-3'.
Cloning of the rat and human ghrelin promoters and construction of luciferase expression vectors containing rat and human ghrelin 5'-upstream sequences
The rat ghrelin 5'-upstream regulatory sequence was obtained by screening a rat bacterial artificial chromosome library obtained from Incyte (Palo Alto, CA). An approximately 1.4-kbp fragment was cloned into pCRII-TOPO vector using rat genomic DNA (forward primer, 5'-TTCGGGGTACCAGCAGTACCTCCTGCCT-3'; reverse primer, 5'- CTGGAATGCCTCGAGGATGTGGTGCCTGCT-3'). A 3143-bp of the 5'-upstream regulatory region of the human ghrelin gene (–3161/–18) was cloned by PCR amplification using human genomic DNA (BD Biosciences) and inserted into pCRII-TOPO vector (Invitrogen) (forward primer, 5'-AGGTGGCAGGCCCTCAAGAGACCCGCAACAGA-3'; reverse primer, 5'-AGACAGGTGGGCCTGGGGGAGAGAGGGTCT-3'). 5'-Upstream fragments of various sizes were then obtained by PCR using the TOPO vector containing the cloned rat and human promoters and subcloned into the KpnI and XhoI sites of the promoterless luciferase expression vector pGL3-Basic (Promega). The primers used were designed to include an artificial restriction endonuclease site for KpnI and XhoI. For both species, sequences of the 5'-upstream clones were verified by primer walking.
Site-directed mutagenesis of the rat and human ghrelin 5'-upstream sequences
The –581/–469-bp rat ghrelin promoter construct and the –667/–468-bp human ghrelin promoter construct (the translation start site AUG was designated as +1) were mutated by PCR using the QuikChange Site-Directed Mutagenesis Kit (Stratagene). After 18 cycles of PCR using plaque-forming unit Turbo DNA polymerase at 95 C for 30 sec, 55 C for 1 min, and 68 C for 10 min, the parental, double-stranded DNA was digested with DpnI. Mutant sequences were confirmed by DNA sequencing.
Preparation of nuclear extracts and EMSAs
Nuclear extracts were prepared according to the nuclear protein extraction kit (Active Motif). Cells were prewashed with ice-cold PBS containing phosphatase inhibitors, scraped into fresh ice-cold PBS/phosphatase inhibitors, and centrifuged at 1000 rpm at 4 C for 5 min. Pellets were resuspended in cold 1x hypotonic buffer, lysed by gentle pipetting, and incubated on ice for 15 min. Detergent was added (1/20 volume) and vortexed for 10 sec. After centrifugation at 14,000 x g for 30 sec, nuclei were resuspended in Complete Lysis Buffer by pipetting, vortexed for 10 sec, and then incubated on a shaker for 30 min at 4 C. After centrifugation at 14,000 x g for 10 min at 4 C, supernatants were aliquoted into chilled microcentrifuge tubes and stored at –80 C. Protein concentrations were assayed by the Bradford assay (Bio-Rad Laboratories, Inc., Richmond, CA).
Rat and human ghrelin promoter fragments and corresponding mutated fragments were end-labeled with [-32P]ATP (3000 Ci/mmol) using T4 polynucleotide kinase. Approximately 50,000 cpm of 32P-labeled DNA was added to the nuclear extract, and preparations were incubated with 50 ng/μl polydeoxyinosinic deoxycytidylic acid · polydeoxyinosinic deoxycytidylic acid and 5 ng/μl poly L-lysine in binding buffer [20 mM HEPES (pH 7.6), 1 mM dithiothreitol, 30 mM KCI, 10 mM (NH4)2SO4, 0.2% Tween 20, and 1 mM EDTA] at room temperature for 20 min. DNA-protein complexes were resolved on a 4% polyacrylamide gel in 0.5x Tris-borate, EDTA buffer. Gels were dried and autoradiographed with intensifying screens at –70 C. For competition experiments, nuclear extracts were incubated with a 50-fold molar excess of double-stranded, nonradiolabeled oligonucleotides at room temperature for 10 min before adding radiolabeled oligonucleotides.
Glucagon treatment of rats, measurement of stomach ghrelin mRNA levels
Adult male Sprague Dawley rats were housed in an air-conditioned room with regulated temperature and lighting. All procedures on animals were performed by properly qualified personnel in accordance with the mandated standards of human care and were approved by the University of Texas Medical Branch Institutional Animal Care and Use Committee. Rats were treated with glucagon (200 μg/kg, 3 times a day sc for 3 d). The stomach was harvested, and RNA was prepared as described previously (38). In brief, stomach tissue samples were removed quickly after animals were killed and immediately homogenized in 4 M guanidinium isothiocyanate containing 25 mM sodium citrate (pH 7.0), 0.5% sodium lauroylsarcosine, and 0.1 M ?-mercaptoethanol. Extracts were frozen at –80 C until purification by ultracentrifugation over a cesium chloride cushion (2 ml, 5.7 M). Total cellular RNA samples were then separated on a 1% agarose gel (10 μg/lane) in a 20 mM 3-[N-morpholino] propanesulfonic acid running buffer system and then transferred to a nylon membrane and subjected to Northern hybridization. 32-P-labeled riboprobes prepared from Strip-EZ RNA kits (Ambion Inc.) were used for Northern hybridizations. The rat ghrelin riboprobe was supplied by Kojima (1). Expression levels of ghrelin or the 18S ribosomal genes were quantitated by densitometry.
Statistical analyses
Results are expressed as the mean ± SEM. Data were analyzed by t test or one-way ANOVA and subsequently with Newman-Keuls test when appropriate. P < 0.05 was considered significant.
Results
Cloning and deletional analyses of the 5'-flanking segments of rat and human ghrelin genes
The rat ghrelin promoter was obtained by screening a rat bacterial artificial chromosome library (Incyte) using PCR. A positive clone with an approximately 1.4-kbp DNA insert was obtained and sequenced. The GenBank accession number for the rat sequence is AY701847. Using the information provided by the human genome database on a National Center for Biotechnology Information web site, the 5'-upstream regulatory region of the human ghrelin gene was cloned and sequenced. Its accession number is AY701846. Schematic representations of the 5'-flanking segments of both rat and human ghrelin genes are shown in Fig. 1. The proximal segments show a high sequence identity (65%); however, outside the proximal segments, identity is not detected. Potential TATA boxes are identified –504 to –448 bp and –585 to –579 bp from the respective translation start sites in the rat and human ghrelin genes. CAAT and GGGCG boxes were not identified in the proximal promoter segments. Potential initiator sites were identified –476 to –469 bp and –557 to –550 bp from the respective translation start sites in the rat and human ghrelin genes.
FIG. 1. A, Schematic representation of the 5'-upstream segments of the rat and human ghrelin genes. Locations of exons 1 and 2 are shown. The nucleotide position +1 corresponds to the translation start codon ATG. D and P, Distal and proximal transcription initiation sites, respectively. B, Comparison of the nucleotide sequences of the core promoter regions of the rat and human ghrelin genes. Putative TATA boxes (underlined) and initiators (bold) are identified. Exon 1 is boxed.
In the original report describing the rat and human ghrelin cDNAs (1), the transcription initiation site was identified 30 and 32 bp upstream of the translation start sites, respectively. However, a subsequent report (39) identified an untranslated exon of 19 and 20 bp upstream of the reported translation start sites in the rat and human sequences. Therefore, there are two putative transcription initiation sites, with the proximal site located at –30 bp for the rat, and –32 bp for the human, and the distal site located at –474 bp for the rat and –555 bp for the human (the translation start site was defined as +1). Therefore, two different rat and human 5'-upstream fragments, –1886/–469 bp and –1886/–6 bp for the rat and –3161/–555 bp and –3161/–18 bp for the human, were cloned into the pGL3-Basic plasmid and their activities compared in transient transfection assays. In human stomach AGS cells, the rat ghrelin proximal promoter fragment, –1886/–6 bp, and its deletions showed no activity when compared with the promoterless pGL3-Basic vector, whereas the distal promoter fragment, –1886/–469 bp, showed an 8- to 12-fold greater activity (Fig. 2A). Moreover, a high activity was obtained from a series of deletion fragments of –1886/–469 bp when compared with pGL3-Basic vector and deletion fragments of –1886/–6 bp. For the human, the distal promoter fragment, –3161/–554 bp, showed no activity when compared with the promoterless pGL3-Basic vector, whereas the proximal promoter fragment, –3161/–18 bp, showed a 4-fold greater activity (Fig. 2B). The activity of the –3161/–554-bp fragment was approximately 8-fold less than that of the –3161/–18-bp fragment. Furthermore, deletion fragments of –3161/–554 bp showed no activity. In contrast, deletion analysis of the –3161/–18-bp fragment showed a high activity for all fragments with the exception of the –197/–18-bp fragment (Fig. 2B).
FIG. 2. Deletion analyses of the 5'-upstream regions of the rat and human ghrelin genes with different transcription initiation sites. Selection of two transcription initiation sites was based on published sequences and an unrecognized exon 1 (46 ). A, Rat, promoter activity of the regions upstream of the proximal (–1886/–6 bp) and distal (–1886/–469 bp) transcription initiation sites were analyzed by deletion scanning in transient transfection experiments using AGS cells. The upstream region of the distal transcription initiation site was active, whereas the region upstream of the proximal transcription initiation sites was inactive. B, Human, promoter activity of the regions upstream of the proximal (–3161/–18 bp) and distal (–3161/–554 bp) transcription initiation sites was analyzed by deletion scanning in transient transfection experiments using AGS cells. In contrast to the rat, the upstream region of the distal transcription initiation site was inactive, whereas the region upstream of the proximal transcription initiation site was active. Activities were normalized to the activity of the promoterless vector pGL3-Basic, whose activity was set as 1. Bars, Means ± SEM of two independent experiments, each performed in triplicate.
Analysis of promoter activity in different cell types
To characterize the promoter activity of rat and human ghrelin genes, constructs containing various 5'-flanking sequences fused to a luciferase reporter gene were transiently transfected into cell lines originating from different tissues in the body (Fig. 3). 5'-Upstream rat sequences showed the highest activity in stomach AGS cells and kidney fibroblast COS-7 cells when compared with other cell types. In AGS cells, rat ghrelin promoter activity was approximately 6- to 8-fold higher when compared with the activity of the promoterless vector. In AGS cells, deletion from –1886 to –581 bp did not influence activity; nearly 100% of the maximal promoter activity was retained within the –580 bp of the 5'-flanking sequence. In kidney COS-7 cells, a nonendocrine cell line, ghrelin promoter activity was approximately 3- to 12-fold higher when compared with the activity of the promoterless plasmid. Deletion from –1886 to –642 bp increased activity 3-fold, and deletion from –641 to –581 bp increased activity marginally. In mouse fibroblasts (3T3 cells), a nonendocrine cell line, promoter activity was marginal. In PC-12 and hypothalamic GT1-7 cells, deletion from –1886 to –581 bp resulted in 7- and 3-fold elevations in activity, respectively, over that of the –1886/–469-bp construct.
FIG. 3. Deletion analyses of the rat (A) and human (B) ghrelin promoters in different cell types. Deletion constructs were generated as described in Materials and Methods. AGS (human stomach), PC-12 (rat pheochromocytoma), GH3 (rat pituitary GH cells), COS-7 (monkey kidney), GT1-7 (mouse hypothalamic), and 3T3 (mouse skin fibroblasts) cells were transiently transfected with constructs of either the rat or human ghrelin promoter fragments fused to a luciferase reporter gene (pGL3-Basic) and pRL-TK plasmid as an internal control. Luciferase activities were measured 48 h after transfection. Data are presented as mean ± SEM.
For the human, the full-length fragment –3161/–18 bp and its deletion mutants, with the exception of –197/–18 bp, were most active and unchanged in AGS, GH3, and COS-7 cells (Fig. 3B). Activities were highest in the AGS and GH3 cells when compared with other cell types. Activities of the fragments in PC-12 and 3T3 cells were low, with the exception of the –667/–18-bp fragment.
Expression of short and long ghrelin transcripts in various cell lines and rat tissues
Two transcription initiation sites are present in the ghrelin gene (Fig. 1); therefore, we screened various rat tissues and rat, mouse, and human cell lines for expression of these transcripts using RT-PCR. The short and long transcripts are products of the proximal and distal transcription initiation sites, respectively. For each species, two sets of primers were used. One set amplified both transcripts (Fig. 4, top), and another set amplified only the long transcript (Fig. 4, middle). Therefore, following densitometric analysis of the RT-PCR products, the amount of short transcript was derived by subtraction of the long transcript reading from the short plus long transcripts reading (Fig. 4, bottom). RT-PCR showed that the short transcript was present in all rat tissues examined (stomach, pituitary, adrenal, hypothalamus, kidney, heart, lung, muscle, testes, placenta) and in some cell lines (mouse hypothalamic-GT1-7, mouse fibroblasts-3T3, rat heart-H9C2, rat pituitary-GH3, rat stomach-RGM1, human placenta-BeWo, human stomach-AGS, human medullary TT cells). The short transcript was most abundant in rat GH3 and human placenta-BeWo cells and in rat pituitary and testes, followed by slightly lower levels in mouse 3T3 cells. Intermediate levels were measured in GT1-7 and AGS cells and in rat hypothalamus and placental tissues. Low levels of the short transcript were found in H9C2, RGM1, and TT cells and in rat stomach, adrenal, kidney, heart, lung, and muscle tissues. The long transcript was most abundant in human medullary TT cells and in rat stomach and placenta tissues. Intermediate levels were measured in GT1-7, 3T3, and GH3 cells. Lower levels of the long transcript were found in BeWo and AGS cells and in rat pituitary, hypothalamic and testes tissues. Although the hypothalamic and testes bands cannot be seen in Fig. 4, they are visible in the original gel. The long transcript was not detected in H9C2 and RGM1 cells nor in rat adrenal, kidney, heart, lung, and muscle tissues. Neither the short nor the long transcripts were present in rat PC12 cells.
FIG. 4. RT-PCR screening for short and long ghrelin transcripts in cells and rat tissues. Top, RT-PCR products of the short plus long transcripts. Middle, RT-PCR products of the long transcript only. Bottom, Densitometric quantitation of short and long ghrelin transcripts. The primers used to detect the short and long transcripts are described in Materials and Methods. For each species, one set of primers detects the long transcript, whereas another set of primers detects the short plus long transcripts. Levels of the short and long transcripts are derived by densitometric analysis of the short + long and long transcripts; subtraction of the desitometric readings for the long transcript from the short plus long transcripts yields short transcript levels. Three separate densitometric readings gave identical data. One microgram poly [A]+ RNA was used for the RT-PCR. The predicted sizes of the long transcript for the human, rat, and mouse are 507, 411, and 416 bp, respectively. The predicted sizes of the short plus long transcript for the human, rat, and mouse are 364, 361, and 352 bp, respectively. The products matched the predicted sizes. For the short plus long transcript, the rat hypothalamic and testes bands are not visible (middle left panel); however, they were identified in the original gel. RT(–) (no RT added) showed no reaction products. m, Mouse; r, rat; h, human.
Determination of the minimally active fragment
An earlier experiment showed that fragments upstream of the distal transcription initiation site (i.e. –555 bp) in the human 5'-flanking region have no activity, whereas fragments upstream of the proximal transcription initiation site (i.e. –32 bp) were active (Fig. 2), indicating that activity was dependent on the fragment between the two transcription initiation sites (–555 to –32 bp). Therefore, to identify the minimal fragment required for activity, 5' and 3' deletions of the –667/–18-bp fragment were done. Deletions of 100 and 200 bp (–667/–198 bp, –667/–282 bp) at the 3' end decreased activity by approximately 25% when compared with the –667/–18-bp fragment (P < 0.05) (Fig. 5A). Further 100- and 200-bp deletions (–667/–368 bp, –667/–468 bp) increased activity significantly. A 100-bp upstream fragment lacking the noncoding exon 1 (–667/–554 bp) showed no activity. A 100-bp upstream fragment plus exon 1 (–667/–534 bp) showed activity similar to the –667/–18-bp fragment. Because the –667/–468-bp fragment showed the highest activity, the influence of 5' deletions (–647/–468 bp, –587/–468 bp) on activity was tested. 5' Deletions lowered activity significantly. Therefore, in the human ghrelin promoter, the minimal fragment for activity was –667/–468 bp, which included the noncoding exon 1 and a proximal part of intron 1.
FIG. 5. A, Further deletion analysis of the upstream region (–667/–18 bp) of the human ghrelin gene in human stomach AGS cells. Sequential deletions of –667/–18 bp at the 3' end revealed a shorter fragment containing exon 1 (closed box) and a proximal fragment of intron 1 (–667/–468 bp) with significantly greater activity. 5' Deletions of –667/–468 bp lowered activity. B, Influence of exon 1 and a proximal piece of intron 1 on core promoter activity of rat ghrelin. Inclusion of exon 1 (closed box) alone did not affect activity, whereas exon 1 plus a proximal fragment of intron 1 decreased activity.
In the rat, fragments upstream of the distal transcription initiation site (i.e. –474 bp) were active, whereas fragments upstream of the proximal transcription initiation site (i.e. –30 bp) have no activity (Fig. 2). To determine whether the region downstream of the distal transcription initiation site affected activity, two fragments, –580/–454 bp, which included exon 1 only, and –580/–387 bp, which included exon 1 and the proximal part of intron 1, were tested (Fig. 5B). In contrast to the human, inclusion of exon 1 did not enhance activity, and inclusion of exon 1 and the proximal part of intron 1 decreased activity by 70%, indicating that inhibitory elements exist in this region (Fig. 5B). Therefore, the minimally active fragment for the rat is –580/–469 bp.
Effect of TATA box mutations on activity
Potential TATA box-like elements were identified in the rat and human promoter fragments (–585/–579 bp in the human 5'-upstream sequence; –504/–498 bp in the rat). Mutation of the potential TATA box in the rat promoter fragment reduced activity by 60%, whereas mutation of the potential TATA box in the human promoter reduced activity by 30% (Fig. 6). In an additional experiment, deletion of an 80-bp sequence that included the TATA box in the human promoter resulted in a similar reduction in activity (data not shown).
FIG. 6. Mutational analyses of putative TATA boxes in the rat and human ghrelin core promoters in human stomach AGS cells. TATATAA in the rat (A) and human (B) core promoters were mutated to CTGCAGA, and activity was analyzed in transient transfection experiments.
Importance of an initiator-like element on promoter activity
Initiator-like elements were identified in the rat and human promoters (rat, –476/–469; human, –557/–550). The sequences of the rat and human initiators are CCACATCC and CCACCTCC. Four-base pair mutations were done. In the rat, mutation of the first four base pairs of the initiator (Inr-M1) reduced activity by 50%, whereas mutation of the second four base pairs (Inr-M2) reduced activity by 20% (Fig. 7A). In the human, mutation of the first 4 bp (Inr-M1) of the initiator abolished activity completely, whereas mutation of the second 4 bp (Inr-M2) reduced activity by 60% (Fig. 7B). EMSAs were also done to characterize protein-DNA complex formation at the human initiator site. Oligonucleotide fragments that contained either the wild-type (WT) or mutant initiator site were used as probes or competitors (Fig. 7C). With WT probe, two specific protein-DNA complexes are identified (Fig. 7C, lanes 1 and 2). With Inr-M1 and Inr-M2 as probes, formation of protein-DNA complexes was abolished (lanes 3, 4). Use of Inr-M1 and Inr-M2 as competitors with WT probe did not affect formation of the two specific protein-DNA complexes (lanes 5, 6). Together, the EMSA and luciferase assay findings indicate the importance of the initiator sites for promoter activity in the human.
FIG. 7. Mutational analyses of initiator sites in the rat (A) and human (B) ghrelin core promoters in human stomach AGS cells. Selective 4-bp mutations were done to putative 8-bp initiator sequences in the rat and human ghrelin core promotors, and activities were analyzed in transient transfection experiments. Luciferase assays indicate that the initiator site is more important in the human ghrelin promoter than in the rat. C, EMSA analyses show the influence of mutations on the formation of protein-DNA complexes at the human initiator site. Oligonucleotide fragments (–570/–536 bp) containing the WT or mutated initiator sites (Inr-M1, Inr-M2) were used in EMSA experiments. With WT, oligonucleotide fragment formation of two specific protein-DNA complexes was observed (lanes 1 and 2). Protein-DNA complexes did not form with oligonucleotide fragments containing mutated initiator sites (lanes 3 and 4). Excess Inr-M1 and Inr-M2 did not decrease formation of protein-DNA complexes (lanes 5 and 6).
Site-directed mutagenesis of rat and human ghrelin core promoters: influence on transcriptional activity and binding
Transient transfection experiments were done initially to identify regions important for basal promoter activity in the rat ghrelin gene. Experiments were done using a series of 12 overlapping, site-directed mutations, consisting either of 5 or 6 bp, of the –580- to –526-bp region. Luciferase assays identified several important regions: M1 (–571 to –566 bp), M2 (–568 to –563 bp), M7 (–548 to –544 bp), M11 (–534 to –529 bp), and M12 (–531 to –526 bp). In general, luciferase activities decreased by 40–85% (Fig. 8A). Mutations that did not affect activity are not shown. To further pinpoint specific bases underlying the reduced activities, selective 3-bp mutations were done. Three-base pair mutations of M2, i.e. M14 and M15, and 3-bp mutations of M12, i.e. M17 and M18, reduced activities by 35–80% (Fig. 8B). Three-base pair mutations, i.e. M13 and M16 of the M1 and M11 regions, did not affect activity (data not shown). The influence of mutations on protein-DNA complex formation in the M14–M15, M17–M18, and M7 regions were examined by EMSAs. Either WT oligonucleotide fragments or mutant oligonucleotide fragments were used as probes and competitors. The WT probes for M14-15, M7 and M17-18, were called A, B, and C, respectively. With WT-A probe, two specific protein-DNA complexes that are competed by excess nonradiolabeled WT-A were observed (Fig. 8C, lanes 1 and 2, arrows). With the M14 probe, two protein-DNA complexes that migrate slightly faster than the slow moving complex (Fig. 8C, lane 3, asterisks) were observed, and the slower moving complex disappeared (Fig. 8C, lane 3). With the M15 probe, the slower moving protein-DNA complex disappeared (Fig. 8C, lane 4). With the WT-A probe, neither M14 nor M15 competed (i.e. formation of the slower protein-DNA complex was unaffected), indicating that these regions are important for protein binding (Fig. 8D, lanes 3 and 4). The EMSA findings agree with the reduced activity observed with the M14 and M15 mutations. With the WT-B probe, no protein-DNA complexes were observed (Fig. 8C, lane 5), whereas with the M7 probe, a single protein-DNA complex was seen (Fig. 8C, lane 7), indicating that the reduced luciferase activity for M7 is a result of formation of a novel protein-DNA complex. With the WT-C probe, two specific protein-DNA complexes that were competed by excess nonradioactive WT-C were observed (Fig. 8C, lanes 8 and 9, arrows). With the M17 probe, the slower moving protein-DNA complex disappeared, and three protein-DNA complexes having mobilities different from WT-C were observed (Fig. 8C, lane 10, asterisks). With the M18 probe, the slower moving protein-DNA complex disappeared (Fig. 8C, lane 11). With WT-C probe, the M17 and M18 competitors partially competed for formation of the slower moving protein-DNA complexes, indicating that these regions are important for protein binding (Fig. 8D, lanes 7 and 8). The EMSA findings agree with the reduced activity observed with the M17 and M18 mutations.
FIG. 8. Effect of site-directed mutagenesis on rat ghrelin core promoter activity in human stomach AGS cells and EMSA using nuclear extract prepared from human stomach AGS cells. A series of 12 site-directed mutations, consisting either of 5 or 6 bp, of the –580- to –526-bp region were done to identify specific sequences underlying regulation of basal activity. WT nucleotides were mutated to Ts. Mutations that reduced activity are shown in A. Based on the results of the 5- or 6-bp mutations, a series of 3-bp mutations were then done, and mutations that reduced activity are shown (B). WT ghrelin (–580/–469 bp) and mutated (M) constructs were transfected into stomach AGS cells and luciferase activities measured 48 h after transfection. Activities were normalized to ?-galactosidase activity. Each construct was tested in triplicate and data are given as the mean ± SEM. *, P < 0.05 vs. WT (–580/–469 bp). C, EMSA of the core promoter of the rat ghrelin gene. WT and mutated (M) oligonucleotide fragments were used. Three and 5-bp mutations were introduced into three regions: A (–571 to –563 bp), B (–548 to –544 bp), and C (–534 to –526 bp). EMSA studies were done using either WT oligonucleotide fragments or corresponding mutated oligonucleotide fragments (M14, M15, M7, M17, M18) and nuclear extract from AGS cells. Double-stranded competitor DNA fragments were added in 50-fold molar excess. Specific nuclear protein-DNA complexes are identified by the arrows. Novel protein-DNA complexes are identified by asterisks. D, EMSAs were performed with WT oligonucleotides as probes and mutated fragments as competitors (M14, M15, M17, M18). Arrows indicate specific DNA-protein complexes that are competed by excess mutated fragments.
To identify important regions underlying regulation of basal promoter activity for the human ghrelin gene, transient transfection experiments were done using a series of 16 site-directed 6-bp mutations of the –570- to –468-bp regions, which include exon 1 and part of intron 1. Mutations M1 (–567 to –562 bp) and M14 (–483 to –478 bp) lowered activity 60 and 20%, respectively (Fig. 9A). EMSA experiments were then done to examine the influence of these mutations on protein-DNA complex formation. Either WT oligonucleotide fragments of the core promoter region (identified as A and B) or mutant oligonucleotide fragments (M1 and M14) were used as probes and competitors. With a WT-A probe, four protein-DNA complexes formed that were decreased by excess WT-A competitor (Fig. 9B, lanes 1 and 2, arrows). With the M1 probe, the two fast-moving complexes disappeared (Fig. 9B, lane 3). With the WT-A probe, the M1 competitor failed to compete for formation of the faster moving complexes, whereas M1 decreased formation of the slower moving complexes (Fig. 9B, lane 4), indicating that this region is important for binding. These findings agree with the reduced luciferase activity with the M1 mutation (Fig. 9A). Using the WT-B probe, no protein-DNA complex formed (Fig. 9B, lane 5), whereas using M14 as a probe, a novel protein-DNA complex was observed (Fig. 9B, lane 7, asterisk) that may be associated with the decreased luciferase activity with M14 mutation.
FIG. 9. Effect of site-directed mutagenesis on human ghrelin core promoter activity in human stomach AGS cells and EMSA using nuclear extract prepared from human stomach AGS cells. A series of 16 site-directed mutations, consisting of 6 bp, of the –570- to –468-bp region were done to identify specific sequences underlying regulation of core promoter activity. WT nucleotides were mutated to Ts. A, Mutations that reduced activity are shown. Each construct was tested in triplicate and data are given as the mean ± SEM. *, P < 0.05 vs. WT. B, EMSA of the core promoter region of the human ghrelin gene. Both WT and mutated (M) oligonucleotide fragments were used as probes and competitors. Six-base pair mutations were introduced into two regions: A (–570 to –536 bp) and B (–498 to –468 bp). EMSA studies were done using either WT oligonucleotide fragments (A, B) or corresponding mutated oligonucleotide fragments (M1, M14) and nuclear extract from AGS cells. Double-stranded competitor DNA fragments were added in 50-fold molar excess. Specific nuclear protein-DNA complexes are identified by the arrows. Novel protein-DNA complexes are identified by asterisks.
Role of USF1 and USF2 in regulation of rat and human ghrelin genes
USF binding sites were identified in the rat and human ghrelin promoter regions (rat, –516/–511 bp; human, –566/–561 bp). The sequences were: rat, CAGCAG; and human, CACCAG. Overexpression of USF1 in MCF10A cells transfected with the rat and human ghrelin core promoters increased activities 1- and 4-fold, respectively (Fig. 10). MCF10A cells (normal breast epithelium) were used because USF1 and USF2 activities are optimal in MCF10A cells (33, 34). Overexpression of USF2 in MCF10A cells increased activities 5- and 9-fold of the rat and human ghrelin core promoters, respectively.
FIG. 10. Influence of USF1 and USF2 overexpression on rat (A) and human (B) ghrelin promoter activities. MCF10A cells were cotransfected either with rat or human ghrelin core promoter plus USF1 or USF2 expression vectors. Cells were also cotransfected with an empty vector (pSG424). Luciferase activities were normalized to the empty vector readings. pRLSV40 served as an internal control.
Glucagon increases stomach ghrelin mRNA expression and promoter activity
Because systemic glucagon levels increase with food restriction (26, 27), the effect of glucagon on gastric ghrelin expression was examined in vivo. Glucagon administration (200 μg, 3x/d, for 3 d, SC) increased steady-state stomach ghrelin mRNA levels significantly (Fig. 11). Ghrelin mRNA levels increased 1.3-fold in glucagon-treated rats when compared with ghrelin mRNA levels in vehicle-treated controls. Glucagon stimulation of ghrelin promoter activity is one possible mechanism behind the glucagon-induced elevation in stomach ghrelin expression. Therefore, the effect of glucagon on ghrelin promoter activity was tested. In the rat and human, glucagon treatment at 1 x 10–6 M increased activity by approximately 25–50% (Fig. 11).
FIG. 11. Glucagon increases rat ghrelin mRNA levels and rat and human ghrelin promoter activity. A, Northern analysis showed that glucagon treatment (200 μg, 3x/d, x1 d, sc) increased stomach ghrelin expression in rats. B, Densitometric analysis of hybridization signals. Data are shown as mean ± SEM of ghrelin mRNA readings over 18S ribosomal RNA densitometric readings. n = 10 rats/group. *, P < 0.05 vs. control-treated rats. C, –543/–469 bp and –3161/–18 bp of the rat and human 5'-flanking regions of the ghrelin genes fused to the luciferase reporter gene and a glucagon receptor expression vector (PMT5/GR) were transfected into stomach AGS cells and treated with 1 x 10–6 M glucagon. Luciferase activities were measured 16 h after the start of glucagon treatment and are expressed as mean ± SEM. Glucagon treatment increased activity significantly. *, P < 0.05 vs. control treatment.
Discussion
Ghrelin is a recently discovered stomach hormone that influences a variety of metabolic activities, including GH and insulin secretion, body growth, food intake, and adiposity (1, 22, 40, 41, 42). The aim of the present work was to clone and characterize the 5'-upstream regulatory segments of the rat and human ghrelin genes to elucidate molecular mechanisms underlying regulation of ghrelin gene transcription.
5'-Flanking segments for rat and human ghrelin genes were cloned and sequenced. Earlier reports identified two putative transcription initiation sites in the ghrelin promoter (1, 39). A proximal initiation site is located at position –30 for the rat and at position –32 for the human, and the distal site is located at position –474 for the rat and at position –555 for the human (the translation start site was defined as +1). Interestingly, the present study showed no transcriptional activity in a fragment upstream of the proximal transcription initiation site in the rat, whereas a similar fragment in the human showed activity. Here, we found that basal activity of the human core promoter requires a sequence downstream of the distal transcription initiation site, i.e. noncoding exon 1 and a proximal sequence of intron 1, whereas in the rat, the noncoding exon 1 is not essential for activity. In fact, 5'- and 3'-deletional scanning analyses showed that a 90-bp fragment upstream of the distal transcription initiation site conferred core promoter activity (8-fold increase in activity when compared with promoterless vector) and that inclusion of exon 1 plus a proximal piece of intron 1 inhibited basal promoter activity in the rat. Interestingly, these regulatory differences exist despite a 65% sequence identity between the core promoter regions. Outside the core promoter region, no sequence identity was found. Together, these regulatory differences in the rat and human ghrelin promoters indicate that transcriptional regulation of the human ghrelin gene has diverged from the rat.
Our findings identify an essential role for exon 1 and a proximal piece of intron 1 for human ghrelin promoter activity. Several earlier reports have described the involvement of exon 1 and intron 1 in regulating transcription activity. For instance, the AP-2 element in the untranslated exon 1 of the human hematopoietic prostaglandin D synthase gene is essential for transcriptional activation (43), and multiple cis-acting elements (E-box binding sites and Ets-like element) in the untranslated exon 1 of the human GnRH-II gene function cooperatively to stimulate transcription (44). In contrast, a functional element was not identified in the untranslated exon 1 of the human ghrelin gene by means of mutagenesis assays. Exon 1 alone exerts neither promoter nor enhancer activities (data not shown) but seems to function in a cooperative manner with other elements in the core promoter region.
Other major regulatory differences in the 5'-upstream regulatory regions of the rat and human ghrelin genes include the influence of the putative initiator sites and TATA box elements on basal promoter activity. In both the rat and human, the TATA box and initiator appear to function together to define transcription start point; however, in the human ghrelin promoter, the initiator site appears to be more important because mutation of the initiator abolished transcriptional activity completely. In contrast, in the rat, mutation of the initiator diminished activity only moderately. In the rat ghrelin promoter, the putative TATA box appears more important because mutation of the TATA box decreased basal activity to a greater extent than in the human ghrelin promoter. Interestingly, an earlier report indicated that the TATA box is not functional in the human ghrelin promoter (45).
This is the first report describing the distribution of two ghrelin gene transcripts in various cell types and rat tissues. In the cells and tissues examined, the short ghrelin gene transcript shows a wider distribution pattern when compared with the distribution of the long transcript. Additionally, the distribution patterns of the transcripts confirm and extend the reports on ghrelin expression in various cell types and rat tissues (1, 6, 8, 9, 29, 46, 47, 48, 49, 50, 51, 52).
The present study demonstrates a cell type-specific activity for the rat and human ghrelin promoters. Overall, our findings indicate that many cell types support ghrelin expression. Deletion analyses in various cell types showed rat promoter activity is highest in stomach AGS and kidney COS-7 cells, whereas activity of human promoter was highest in AGS and pituitary GH3 cells. The human ghrelin promoter was also active in COS-7 cells but at a significantly lower level when compared with AGS and GH3 cells. However, an earlier report showed that a proximal fragment (–605/–1 bp) of the human ghrelin promoter was inactive in GH3 and COS-7 cells (45). In contrast to our expectations and the findings with the human ghrelin promoter fragments, rat ghrelin promoter fragments showed no activity in GH3 cells. Ghrelin is expressed in the pituitary (1, 8, 47, 53), and both ghrelin transcripts are detected in GH3 cells; therefore, we would anticipate the rat ghrelin promoter to be active in GH3 cells. Furthermore, the larger rat (–1886/–6 bp) construct is not active in rat GH3 cells (data not shown). At this point, we cannot offer an explanation for the absence of rat ghrelin promoter activity in GH3 cells. Activities of rat distal constructs (–1886/–469 bp; –1116/–469 bp) were lower than a proximal construct (–641/–469 bp) in COS-7 cells and less so in hypothalamic GT1-7 and adrenal PC-12 cells, suggesting that negative elements in the distal promoter region inhibit transcription to varying degrees in these cells. In AGS cells, activities of the rat promoter segments are similar, suggesting that the negative element is not functional in AGS cells or that a transcription factor having inhibitory activity is not present in AGS cells. Interestingly, activity of a proximal human promoter fragment (–667/–18 bp) is greatly increased in PC-12, GTI-7, and 3T3 cells, implying that negative elements in the distal human ghrelin promoter are differentially activated. With a few exceptions, rat and human ghrelin promoter segments showed marginal activity in rat hypothalamic GTI-7 and adrenal PC-12 cells and in mouse fibroblasts (3T3). The low promoter activity in PC-12 cells is expected based upon the absence of ghrelin expression in these cells. However, the low ghrelin promoter activity in 3T3 and GT1-7 cells does not correlate with detection of ghrelin transcripts in these cells. 3T3 and GT1-7 cells are murine derived and may not support rat and human ghrelin promoters.
Elements essential for basal activity in the rat and human ghrelin core promoters were identified using mutational analyses and EMSAs. Transient transfection experiments identified three regions in the rat, –568/–563, –548/–544, and –531/–526 bp, and two regions in the human, –567/–562 and –483/–478 bp, that are important for basal activity. In general, EMSAs showed that reductions in transcriptional activity are associated with a decrease in formation of specific protein-DNA complexes. Although exon-1 and a proximal piece of intron 1 were essential for core promoter activity in the human, mutational analyses and EMSAs failed to identify a single cis-element underlying promoter activity, suggesting that activity of the human ghrelin promoter requires interactions of multiple regulatory elements. In two instances, a reduction in activity was associated with formation of novel protein-DNA complexes. It should be mentioned that we used scanning mutations to identify regions important in the core promoter for activity. A potential shortcoming with this approach is that mutations may create cis-acting elements.
USF1 and USF2 are ubiquitously expressed basic helix-loop-helix-leucine zipper transcription factors that bind as homodimers and heterodimers to DNA sequences centered on CAC(G/A)TG (54, 55). Sequence analyses of both human and rat core promoters identified putative USF sites, CACCAG and CAGCAG, which are not perfect matches to consensus sequences. Our results show that USF may play a stimulatory role in the activation of the rat and human ghrelin promoters. Interestingly, USF levels increase during protein deprivation (56). Our laboratory has reported that stomach ghrelin expression increases in rats fed low-protein diets (22). The increased stomach ghrelin gene expression may be due to increased binding of USF to the ghrelin promoter.
Expression and secretion of gastric ghrelin increase with food restriction and decrease with food intake (22, 24, 40, 41, 57, 58). The molecular mechanism underlying this increased gastric ghrelin production with fasting is not known but may reflect actions of other enteric or pancreatic hormones that are altered by changes in nutritional status. The influence of glucagon on stomach ghrelin expression was explored because glucagon secretion increases with fasting (26, 27). In the present study, glucagon treatment up-regulated ghrelin expression levels in the rat, and we confirmed an earlier report showing that glucagon apparently increased steady-state ghrelin mRNA levels by a stimulation of ghrelin transcription because glucagon stimulates rat and human ghrelin promoter activity (45).
Abundant reports indicate that ghrelin is an important new hormone (42, 46). Ghrelin is widely expressed in the body and exerts a diverse array of metabolic activities. In this study, we have characterized the 5'-upstream regulatory regions of the rat and human ghrelin genes and shown that major differences exist in regulation of their core promoter activities and in regulation of ghrelin promoter activity at the tissue and cellular levels. The pancreatic hormone, glucagon, and the transcription factors, USFs, may be two metabolically sensitive factors that regulate ghrelin gene expression.
References
Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K 1999 Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402:656–660
Tomasetto C, Karam SM, Ribieras S, Masson R, Lefebvre O, Staub A, Alexander G, Chenard MP, Rio MC 2000 Identification and characterization of a novel gastric peptide hormone: the motilin-related peptide. Gastroenterology 119:395–405
Cassoni P, Papotti M, Ghe C, Catapano F, Sapino A, Graziani A, Deghenghi R, Reissmann T, Ghigo E, Muccioli G 2001 Identification, characterization, and biological activity of specific receptors for natural (ghrelin) and synthetic growth hormone secretagogues and analogs in human breast carcinomas and cell lines. J Clin Endocrinol Metab 86:1738–1745
Date Y, Nakazato M, Hashiguchi S, Dezaki K, Mondal MS, Hosoda H, Kojima M, Kangawa K, Arima T, Matsuo H, Yada T, Matsukura S 2002 Ghrelin is present in pancreatic -cells of humans and rats and stimulates insulin secretion. Diabetes 51:124–129
Guan XM, Yu H, Palyha OC, McKee KK, Feighner SD, Sirinathsinghji DJ, Smith RG, Van der Ploeg LH, Howard AD 1997 Distribution of mRNA encoding the growth hormone secretagogue receptor in brain and peripheral tissues. Brain Res Mol Brain Res 48:23–29
Kim K, Arai K, Sanno N, Osamura RY, Teramoto A, Shibasaki T 2001 Ghrelin and growth hormone (GH) secretagogue receptor (GHSR) mRNA expression in human pituitary adenomas. Clin Endocrinol (Oxf) 54:759–768
McKee KK, Tan CP, Palyha OC, Liu J, Feighner SD, Hreniuk DL, Smith RG, Howard AD, Van der Ploeg LH 1997 Cloning and characterization of two human G protein-coupled receptor genes (GPR38 and GPR39) related to the growth hormone secretagogue and neurotensin receptors. Genomics 46:426–434
Muccioli G, Papotti M, Locatelli V, Ghigo E, Deghenghi R 2001 Binding of 125I-labeled ghrelin to membranes from human hypothalamus and pituitary gland. J Endocrinol Invest 24:RC7–RC9
Shuto Y, Shibasaki T, Wada K, Parhar I, Kamegai J, Sugihara H, Oikawa S, Wakabayashi I 2001 Generation of polyclonal antiserum against the growth hormone secretagogue receptor (GHS-R): evidence that the GHS-R exists in the hypothalamus, pituitary and stomach of rats. Life Sci 68:991–996
Arvat E, Di Vito L, Broglio F, Papotti M, Muccioli G, Dieguez C, Casanueva FF, Deghenghi R, Camanni F, Ghigo E 2000 Preliminary evidence that Ghrelin, the natural GH secretagogue (GHS)-receptor ligand, strongly stimulates GH secretion in humans. J Endocrinol Invest 23:493–495
Date Y, Kojima M, Hosoda H, Sawaguchi A, Mondal MS, Suganuma T, Matsukura S, Kangawa K, Nakazato M 2000 Ghrelin, a novel growth hormone-releasing acylated peptide, is synthesized in a distinct endocrine cell type in the gastrointestinal tracts of rats and humans. Endocrinology 141:4255–4261
Kamegai J, Tamura H, Shimizu T, Ishii S, Sugihara H, Wakabayashi I 2000 Central effect of ghrelin, an endogenous growth hormone secretagogue, on hypothalamic peptide gene expression. Endocrinology 141:4797–4800
Masuda Y, Tanaka T, Inomata N, Ohnuma N, Tanaka S, Itoh Z, Hosoda H, Kojima M, Kangawa K 2000 Ghrelin stimulates gastric acid secretion and motility in rats. Biochem Biophys Res Commun 276:905–908
Nakazato M, Murakami N, Date Y, Kojima M, Matsuo H, Kangawa K, Matsukura S 2001 A role for ghrelin in the central regulation of feeding. Nature 409:194–198
Peino R, Baldelli R, Rodriguez-Garcia J, Rodriguez-Segade S, Kojima M, Kangawa K, Arvat E, Ghigo E, Dieguez C, Casanueva FF 2000 Ghrelin-induced growth hormone secretion in humans. Eur J Endocrinol 143:R11–R14
Seoane LM, Tovar S, Baldelli R, Arvat E, Ghigo E, Casanueva FF, Dieguez C 2000 Ghrelin elicits a marked stimulatory effect on GH secretion in freely-moving rats. Eur J Endocrinol 143:R7–R9
Takaya K, Ariyasu H, Kanamoto N, Iwakura H, Yoshimoto A, Harada M, Mori K, Komatsu Y, Usui T, Shimatsu A, Ogawa Y, Hosoda K, Akamizu T, Kojima M, Kangawa K, Nakao K 2000 Ghrelin strongly stimulates growth hormone release in humans. J Clin Endocrinol Metab 85:4908–4911
Tschop M, Smiley DL, Heiman ML 2000 Ghrelin induces adiposity in rodents. Nature 407:908–913
Wren AM, Seal LJ, Cohen MA, Brynes AE, Frost GS, Murphy KG, Dhillo WS, Ghatei MA, Bloom SR 2001 Ghrelin enhances appetite and increases food intake in humans. J Clin Endocrinol Metab 86:5992
Wren AM, Small CJ, Ward HL, Murphy KG, Dakin CL, Taheri S, Kennedy AR, Roberts GH, Morgan DG, Ghatei MA, Bloom SR 2000 The novel hypothalamic peptide ghrelin stimulates food intake and growth hormone secretion. Endocrinology 141:4325–4328
Reimer MK, Pacini G, Ahren B 2003 Dose-dependent inhibition by ghrelin of insulin secretion in the mouse. Endocrinology 144:916–921
Lee HM, Wang G, Englander EW, Kojima M, Greeley Jr GH 2002 Ghrelin, a new gastrointestinal endocrine peptide that stimulates insulin secretion: enteric distribution, ontogeny, influence of endocrine, and dietary manipulations. Endocrinology 143:185–190
Tannenbaum GS, Epelbaum J, Bowers CY 2003 Interrelationship between the novel peptide ghrelin and somatostatin/growth hormone-releasing hormone in regulation of pulsatile growth hormone secretion. Endocrinology 144:967–974
Gomez G, Englander EW, Greeley Jr GH 2004 Nutrient inhibition of ghrelin secretion in the fasted rat. Regul Pept 117:33–36
Asakawa A, Inui A, Kaga T, Yuzuriha H, Nagata T, Fujimiya M, Katsuura G, Makino S, Fujino MA, Kasuga M 2001 A role of ghrelin in neuroendocrine and behavioral responses to stress in mice. Neuroendocrinology 74:143–147
Seino S, Sakurai H, Seino Y, Tsuda K, Tanigawa K, Kuzuya H, Goto Y, Imura H 1980 Starvation-induced changes of somatostatin, glucagon, and insulin secretion from the isolated perfused rat pancreas. Diabetes 29:323–325
Unger RH 1971 Glucagon physiology and pathophysiology. N Engl J Med 285:443–449
Hocker M, Zhang Z, Merchant JL, Wang TC 1997 Gastrin regulates the human histidine decarboxylase promoter through an AP-1-dependent mechanism. Am J Physiol 272:G822–G830
Gualillo O, Caminos J, Blanco M, Garcia-Caballero T, Kojima M, Kangawa K, Dieguez C, Casanueva F 2001 Ghrelin, a novel placental-derived hormone. Endocrinology 142:788–794
Kanamoto N, Akamizu T, Hosoda H, Hataya Y, Ariyasu H, Takaya K, Hosoda K, Saijo M, Moriyama K, Shimatsu A, Kojima M, Kangawa K, Nakao K 2001 Substantial production of ghrelin by a human medullary thyroid carcinoma cell line. J Clin Endocrinol Metab 86:4984–4990
Kashour T, Burton T, Dibrov A, Amara F 2003 Myogenic signaling by lithium in cardiomyoblasts is Akt independent but requires activation of the beta-catenin-Tcf/Lef pathway. J Mol Cell Cardiol 35:937–951
Sasaki E, Pai R, Halter F, Komurasaki T, Arakawa T, Kobayashi K, Kuroki T, Tarnawski AS 1998 Induction of cyclooxygenase-2 in a rat gastric epithelial cell line by epiregulin and basic fibroblast growth factor. J Clin Gastroenterol 27(Suppl 1):S21–S27
Ismail PM, Lu T, Sawadogo M 1999 Loss of USF transcriptional activity in breast cancer cell lines. Oncogene 18:5582–5591
Szentirmay MN, Yang HX, Pawar SA, Vinson C, Sawadogo M 2003 The IGF2 receptor is a USF2-specific target in nontumorigenic mammary epithelial cells but not in breast cancer cells. J Biol Chem 278:37231–37240
Meier JL, Luo X, Sawadogo M, Straus SE 1994 The cellular transcription factor USF cooperates with varicella-zoster virus immediate-early protein 62 to symmetrically activate a bidirectional viral promoter. Mol Cell Biol 14:6896–6906
Carruthers CJ, Unson CG, Kim HN, Sakmar TP 1994 Synthesis and expression of a gene for the rat glucagon receptor. Replacement of an aspartic acid in the extracellular domain prevents glucagon binding. J Biol Chem 269:29321–29328
Wang G, Anini Y, Wei W, Qi X, AM OC, Mochizuki T, Wang HQ, Hellmich MR, Englander EW, Greeley Jr GH 2004 Apelin, a new enteric peptide: localization in the gastrointestinal tract, ontogeny, and stimulation of gastric cell proliferation and of cholecystokinin secretion. Endocrinology 145:1342–1348
Gomez G, Padilla L, Udupi V, Tarasova N, Sundler F, Townsend Jr CM, Thompson JC, Greeley Jr GH 1996 Regulation of peptide YY homeostasis by gastric acid and gastrin. Endocrinology 137:1365–1369
Tanaka M, Hayashida Y, Iguchi T, Nakao N, Nakai N, Nakashima K 2001 Organization of the mouse ghrelin gene and promoter: occurrence of a short noncoding first exon. Endocrinology 142:3697–3700
Asakawa A, Inui A, Kaga T, Yuzuriha H, Nagata T, Ueno N, Makino S, Fujimiya M, Niijima A, Fujino MA, Kasuga M 2001 Ghrelin is an appetite-stimulatory signal from stomach with structural resemblance to motilin. Gastroenterology 120:337–345
Broglio F, Arvat E, Benso A, Gottero C, Muccioli G, Papotti M, van der Lely AJ, Deghenghi R, Ghigo E 2001 Ghrelin, a natural GH secretagogue produced by the stomach, induces hyperglycemia and reduces insulin secretion in humans. J Clin Endocrinol Metab 86:5083–5086
Broglio F, Arvat E, Benso A, Papotti M, Muccioli G, Deghenghi R, Ghigo E 2002 Ghrelin: endocrine and non-endocrine actions. J Pediatr Endocrinol Metab 15(Suppl 5):1219–1227
Fujimori K, Kanaoka Y, Sakaguchi Y, Urade Y 2000 Transcriptional activation of the human hematopoietic prostaglandin D synthase gene in megakaryoblastic cells. Roles of the oct-1 element in the 5'-flanking region and the AP-2 element in the untranslated exon 1. J Biol Chem 275:40511–40516
Cheng CK, Hoo RL, Chow BK, Leung PC 2003 Functional cooperation between multiple regulatory elements in the untranslated exon 1 stimulates the basal transcription of the human GnRH-II gene. Mol Endocrinol 17:1175–1191
Kishimoto M, Okimura Y, Nakata H, Kudo T, Iguchi G, Takahashi Y, Kaji H, Chihara K 2003 Cloning and characterization of the 5(')-flanking region of the human ghrelin gene. Biochem Biophys Res Commun 305:186–192
Kojima M, Hosoda H, Matsuo H, Kangawa K 2001 Ghrelin: discovery of the natural endogenous ligand for the growth hormone secretagogue receptor. Trends Endocrinol Metab 12:118–122
Korbonits M, Kojima M, Kangawa K, Grossman AB 2001 Presence of ghrelin in normal and adenomatous human pituitary. Endocrine 14:101–104
Mori K, Yoshimoto A, Takaya K, Hosoda K, Ariyasu H, Yahata K, Mukoyama M, Sugawara A, Hosoda H, Kojima M, Kangawa K, Nakao K 2000 Kidney produces a novel acylated peptide, ghrelin. FEBS Lett 486:213–216
Rindi G, Necchi V, Savio A, Torsello A, Zoli M, Locatelli V, Raimondo F, Cocchi D, Solcia E 2002 Characterisation of gastric ghrelin cells in man and other mammals: studies in adult and fetal tissues. Histochem Cell Biol 117:511–519
Rindi G, Savio A, Torsello A, Zoli M, Locatelli V, Cocchi D, Paolotti D, Solcia E 2002 Ghrelin expression in gut endocrine growths. Histochem Cell Biol 117:521–525
Volante M, Allia E, Gugliotta P, Funaro A, Broglio F, Deghenghi R, Muccioli G, Ghigo E, Papotti M 2002 Expression of ghrelin and of the GH secretagogue receptor by pancreatic islet cells and related endocrine tumors. J Clin Endocrinol Metab 87:1300–1308
Tena-Sempere M, Barreiro ML, Gonzalez LC, Gaytan F, Zhang FP, Caminos JE, Pinilla L, Casanueva FF, Dieguez C, Aguilar E 2002 Novel expression and functional role of ghrelin in rat testis. Endocrinology 143:717–725
Muccioli G, Tschop M, Papotti M, Deghenghi R, Heiman M, Ghigo E 2002 Neuroendocrine and peripheral activities of ghrelin: implications in metabolism and obesity. Eur J Pharmacol 440:235–254
Gregor PD, Sawadogo M, Roeder RG 1990 The adenovirus major late transcription factor USF is a member of the helix-loop-helix group of regulatory proteins and binds to DNA as a dimer. Genes Dev 4:1730–1740
Murre C, McCaw PS, Baltimore D 1989 A new DNA binding and dimerization motif in immunoglobulin enhancer binding, daughterless, MyoD, and myc proteins. Cell 56:777–783
Matsukawa T, Inoue Y, Oishi Y, Kato H, Noguchi T 2001 Up-regulation of upstream stimulatory factors by protein malnutrition and its possible role in regulation of the IGF-binding protein-1 gene. Endocrinology 142:4643–4651
Gualillo O, Caminos JE, Nogueiras R, Seoane LM, Arvat E, Ghigo E, Casanueva FF, Dieguez C 2002 Effect of food restriction on ghrelin in normal-cycling female rats and in pregnancy. Obes Res 10:682–687
Hansen TK, Dall R, Hosoda H, Kojima M, Kangawa K, Christiansen JS, Jorgensen JO 2002 Weight loss increases circulating levels of ghrelin in human obesity. Clin Endocrinol (Oxf) 56:203–206(Wei Wei, Guiyun Wang, Xia)