Cloning and Characterization of the 5′-Flanking Region of the Human Growth Hormone-releasing Hormone Receptor Gene*

We cloned the 5′-flanking region of the human growth hormone-releasing hormone receptor (GHRH-R) gene and determined the nucleotide sequence of 2.7 kilobases upstream from the translation start site. RNase protection analysis showed the major transcription start site is 122 base pairs upstream from the translation start site. The 5′-end of the longest product of 5′-rapid amplification of cDNA ends was close to the site. There were no typical TATA homologies but several putative regulatory elements including Pit-1-binding site-like element. Transient transfection studies using a luciferase reporter gene demonstrated that 5′-flanking region had promoter activity in GH3 cells (derived from rat pituitary tumor) but not in nonpituitary cells, BeWo and HeLa cells. However, co-transfection of Pit-1 expression vector increased luciferase activity in BeWo cells. Deletion study showed that the regions from −310 to −130 and from −130 to −120 were important for the GHRH-R gene expression in GH3 cells, although the latter contributed less to the gene expression. In BeWo cells co-transfected with Pit-1 expression vector, the region from −310 to −130 was essential for the Pit-1-dependent expression of GHRH-R gene. The region from −310 to −120 has two putative Pit-1-binding sites, P1 and P2, located from −129 to −123 and from −171 to −160, respectively. Both mobility shift assay and DNase-I footprint analysis showed that P2 had much higher Pit-1 binding affinity than P1. Mutation of P2 decreased GHRH-R gene expression in GH3 cells. These findings were consistent with the results that the region from −310 to −130 is an important element for Pit-1-dependent expression of GHRH-R gene.

human (1,2), mouse (3), rat (1), and pig (4). The GHRH-R is a member of G protein-coupled receptor family and transduces GHRH-dependent increase in intracellular cAMP via G s activation for stimulating somatotroph proliferation and GH gene expression (1)(2)(3)(4). Lin et al. (5) demonstrated that one amino acid substitution of the GHRH-R in the little mouse, showing genetically transmitted dwarfism, caused GH deficiency and somatotroph hypoplasia. In addition, an amber-type mutation (Glu72Stop) of the GHRH-R in humans was demonstrated to cause profound GH deficiency (6). These genetic disorders suggest the physiological significance of GHRH-R in hypothalamopituitary GH axis.
The relationship between the dynamics of GHRH-R expression and GH secretion remains to be clarified. Not only the functional defects of GHRH-R but also the amount of GHRH-R should affect GH synthesis and secretion in the pituitary. Hypothalamic hormone, neurotransmitters, various hormonal states, and nutrition all could modulate the activity of the GHRH-GH axis. For instance, glucocorticoids potentiate GHRH action and enhance GH secretion in rats (7)(8)(9), whereas they have a biphasic effect in humans (10,11). A recent study showed that GHRH-R gene expression was increased by glucocorticoids in rats (12). The result suggests that glucocorticoids stimulate GH synthesis and secretion, at least in part, via up-regulation of GHRH-R gene expression. Interestingly, Horikawa et al. (13) observed a marked reduction of GHRH-R mRNA in GHRH-deprived neonatal rats. This observation suggests that GHRH up-regulates GHRH-R expression in addition to direct action to stimulate GH synthesis and secretion. Thus, we need to take the alteration of GHRH-R expression into account to understand the regulatory mechanism of GH secretion.
The GHRH-R transcripts have a highly specific distribution in the anterior pituitary, even though the GHRH-R gene is also expressed in the rat hypothalamus (14). During the development of the anterior pituitary gland, the GHRH-R transcripts were detected at embryonic day 16 in somatotrophs following the gene activation of Pit-1, a pituitary-specific POU domain transcription factor (15). The GHRH-R are not expressed in the pituitary of Pit-1-defective Snell/Jackson dwarf mice (16,17).
Thus, it appears that Pit-1 is required for the GHRH-R gene expression. However, it remains unknown whether the Pit-1binding site is present in the GHRH-R gene and whether it functions for the GHRH-R expression.
In the present study, we have cloned and characterized the 5Ј-flanking region of the human GHRH-R gene to elucidate the regulatory molecular mechanism of the human GHRH-R gene expression. Furthermore, we demonstrated that the 5Ј-flanking region contained a functional promoter and that its promoter activity was dependent on Pit-1.

EXPERIMENTAL PROCEDURES
Cloning of the 5Ј-Flanking Region of the Human GHRH-R Gene by PCR Amplification Method-We cloned 5Ј-flanking region of the human GHRH-R gene using a PCR-based method for DNA walking in uncloned genomic DNA. (PromoterFinder DNA walking kit (CLONTECH)) (18). Adaptor-ligated genomic DNA was amplified with two consecutive PCRs using two oligonucleotide primer pairs. One set of primers used in the primary PCR was the adaptor-specific primer ASP1 (5Ј-CCATCC-TAATACGACTCACTATAGGGC-3Ј) corresponding to the synthetic adaptor and the gene-specific antisense primer GSP2 (5Ј-TCCATGGT-GAGCCCAGCAGTGGCT-3Ј) corresponding to the 5Ј-end of the human GHRH-R cDNA (nucleotides position ϩ5 to Ϫ19 according to Mayo (1)). Another set of primers for the secondary PCR was the adaptor specific primer ASP2 (5Ј-CTATAGGGCACGCGCGTGGT-3Ј), nested to ASP1, and the nested gene-specific antisense primer GSP1 (5Ј-GGGAAGCT-TCCCTCCACCAGCCTCAGTAAGCCTT-3Ј, nucleotides position Ϫ19 to Ϫ44). The underlined HindIII site was added to the 5Ј-end of GSP1 to facilitate subcloning. PCR was performed according to the manufacturer's instructions. The PCR conditions were 32 cycles of 94°C for 25 s and 67°C for 4 min following 7 cycles of 94°C for 25 s and 72°C for 4 min. PCR products were analyzed by electrophoresis on a 1% agarose gel.
Rapid Amplification of 5Ј-cDNA Ends-5Ј-RACE was performed using Marathon-Ready cDNA library (CLONTECH) derived from human pituitary glands according to the manufacturer's instructions. In the first experiment, PCR was performed using ASP1 (5Ј-CCATCCTAAT-ACGACTCACTATAGGGC-3Ј) specific to the adaptor ligated to cDNA ends and GSP5 (5Ј-TTGTACAGGCACTCTCATCCTCTCTCA-3Ј) corresponding to the coding region from ϩ101 to ϩ127 of the human GHRH-R cDNA (1). Cycling parameters used in this 5Ј-RACE protocol were 35 cycles of 94°C for 30 s, 64°C for 30 s and 68°C for 2 min 30 s. PCR products were analyzed by electrophoresis on an ethidium bromide-stained 1.2% agarose gel. The PCR products were subcloned into pT7Blue vector and sequenced with a DNA sequencer.
RNase Protection Analysis-RNase protection assays were performed, based on the method of Gilman (19). A DNA fragment corresponding to nucleotides Ϫ2207 to Ϫ19 was inserted into pBluescript SK(ϩ). The plasmid containing the DNA fragment from Ϫ2207 to Ϫ19 was linearized with BalI and used as a template for antisense cRNA probe. The antisense cRNA probe was synthesized by T 3 RNA polymerase and [␣-32 P]CTP. The labeled probe (10 5 cpm) was hybridized to 4 g of normal human pituitary gland poly(A) ϩ RNA (CLONTECH) in 30 l of hybridization buffer containing 80% formamide, 40 mM PIPES, 400 mM NaCl, and 1 mM EDTA overnight at 50°C. Following the hybridization step, RNases A and T1 were added to the hybridization mixture, and the mix was incubated for 1 h at 37°C. The protected fragments were analyzed by electrophoresis on a 6% polyacrylamide/ 50% urea sequencing gel. A bacteriophage M13 mp18 single strand DNA sequencing ladder was prepared using primer (Ϫ40) to size the protected fragment.
Cell Culture-Cells from the rat pituitary-derived cell line, GH3, were cultured in Ham's F-10 medium supplemented with 15% (v/v) horse serum and 2.5% (v/v) fetal bovine serum. HeLa and BeWo cells were maintained in Dulbecco's modified Eagle's medium with 10% (v/v) calf serum. All culture media contained penicillin (100 units/ml) and kanamycin (100 g/ml).
These plasmids were introduced into GH3, HeLa, or BeWo cells using LipofectAce (Life Technologies, Inc.). 2 g of GHRH-R-luciferase reporter construct, pGL3 basic vector alone, or rat PRL-luciferase reporter construct, with or without RSV-Pit-1 expression vector, was transfected to the cells in 60-mm dishes. Cells were harvested 48 h after transfection, and luciferase activity was measured using Luciferase Assay System and Luminometer TD-20/20 (Promega). Protein concentration was measured by Bio-Rad protein assay based on the method of Bradford (29). DNase-I Footprint Analysis-For footprint studies, recombinant Histagged Pit-1 was produced in Escherichia coli and partially purified with nickel-nitrilotriacetic acid Resin (Qiagen) according to the manufacturer's instructions. His-tagged Pit-1 appeared to have same function compared with recombinant Pit-1 previously described (20). The DNA fragment used for DNase-I footprint analysis of the promoter and enhancer regions of GHRH-R gene was prepared by digestion of GHRH-R 5Ј-flanking region with Bsu36I and HindIII and uniquely radiolabeled at the upstream end by filling recessed 3Ј-termini with [␣-32 P]deoxy-ATP using DNA polymerase-I. The radiolabeled DNA fragment was first incubated with purified His-tagged Pit-1 in a 25-l reaction containing 10 mM Tris-HCl (pH 7.5), 50 mM NaCl, 1 mM dithiothreitol, 1 mM EDTA, 5% glycerol, and 2 g of salmon sperm DNA. After incubation for 30 min at room temperature, 10 l of DNase-I (0.05 unit; RQ-1 DNase-I, Promega) in 100 mM Tris-HCl (pH 7.5) and 35 mM MgCl 2 were added, and the sample was incubated for 8 min at room temperature. The reaction was terminated by the addition of 85 l of 30 mM EDTA, and then the sample was extracted with and equal volume of 1:1 (v/v) phenol-chloroform mixture. After precipitation of the aqueous fraction with ethanol, the samples were analyzed by denaturing polyacrylamide gel electrophoresis and autoradiography.

Cloning of 5Ј-Flanking Region of the Human GHRH-R Gene by PCR-based
Method-First, we cloned 5Ј-flanking region of the human GHRH-R gene as described under "Experimental Procedures." We obtained three genomic fragments, which were about 6, 2.7, and 0.9 kb in size, respectively. The nucleotide sequence of the 0.9-kb fragment was identical to that of 3Ј-side of the 2.7-kb fragment. In addition, restriction mapping of those fragments indicated that the 6-kb fragment contained the 2.7-and 0.9-kb fragments. The complete sequence of the 2.7-kb genomic fragment is shown in Fig. 1.
According to the analysis of the 5Ј-flanking sequence of the human GHRH-R gene by Mac DNASIS Pro software (Hitachi), no typical mammalian TATA boxes were present, but a CAAT box was present at nucleotides Ϫ399 to Ϫ393. In addition, there were several putative regulatory elements similar to consensus elements, EREs (Ϫ662 to Ϫ648 and Ϫ1887 to Ϫ1873), and eleven putative Pit-1-binding elements. The locations of one nucleotide mismatched Pit-1-binding site consensus are as follows: nucleotides Ϫ129 to Ϫ123, Ϫ166 to Ϫ160, Ϫ171 to Ϫ165, Ϫ434 to Ϫ428, Ϫ839 to Ϫ833, Ϫ1049 to Ϫ1043, Ϫ1138 to Ϫ1132, Ϫ1336 to Ϫ1330, Ϫ1631 to Ϫ1625, Ϫ1816 to Ϫ1810, and Ϫ1881 to Ϫ1875.
Analysis of the 5Ј-Flanking Region of the Human GHRH-R Gene-To analyze the 5Ј-end of the human GHRH-R cDNA, we performed 5Ј-RACE experiments. The 5Ј-RACE analysis using ASP1 and GSP5 as primers showed that about 300-nucleotide band was amplified ( Fig. 2A). From the sequence analysis of these products, the 5Ј-end of the longest clone was found to extend to 137 nucleotides upstream from the translation start site. The 5Ј-ends of all the RACE products are indicated with solid circles in Fig. 2B. To confirm the results of 5Ј-RACE experiments, RNase protection assays were performed. RNase mapping with a 381-nucleotide labeled cRNA probe comple- RNase protection assay was performed as described under "Experimental Procedures." A uniformly labeled probe corresponding to bases Ϫ399 to Ϫ19 was hybridized to 4 g of human pituitary poly(A) ϩ RNA or 20 g of yeast tRNA and digested with RNases A and T1. The protected fragments were analyzed by electrophoresis on a 6% polyacrylamide/ 50% urea sequencing gel. Lane 1, human pituitary poly (A) ϩ RNA. Lane 2, tRNA. Lanes M, bacteriophage M13 mp18 single strand DNA sequence when primer (Ϫ40) was used. The major protected band is indicated by an arrow. bp, base pairs. B, schematic representation of the probe used for RNase protection assay and protected fragment. The major transcription start site is indicated by a bent arrow. Numbers are relative to the adenosine of the initiation codon. mentary to Ϫ399 to Ϫ19 gave a major protected band of 104 nucleotides with several faint protected bands (Fig. 3). The result indicated that the major transcription start site was 122 nucleotides upstream from the translation start site. In addition, the several faint bands suggested the existence of possible minor start sites.
Promoter Activity of the Human GHRH-R Gene-We performed transient transfection studies to determine whether our cloned 5Ј-flanking region is functional. Plasmids containing cloned 5Ј-flanking sequences of the human GHRH-R gene fused to the luciferase gene were transfected into rat pituitary tumor cell line GH3 and heterologous cell lines HeLa and BeWo. As illustrated in Fig. 4, a significant increase of luciferase activity was observed in GH3 cells transfected with hGHRHR-Luc(Ϫ2207,Ϫ19), but not in HeLa cells and BeWo cells. The result indicated that the 5Ј-flanking region has pituitary-specific promoter activity.
Next, to define DNA regions important for gene expression, deletion mutants of the 5Ј-flanking sequences were constructed and used to transfect GH3 cells. Deletion of upstream between Ϫ2207 and Ϫ310 resulted in a moderate increase in luciferase activity. Further deletion to Ϫ130 decreased activity remarkably. Deletion to Ϫ120 or Ϫ100 resulted in a level of activity that was comparable with that of the pGL3 basic vector (Fig.  5A). This finding suggested that the regions between Ϫ310 and Ϫ130 and between Ϫ130 to Ϫ120 were important for the expression in GH3 cells. The regions from Ϫ130 to Ϫ120 and Ϫ310 to Ϫ130 have putative Pit-1-binding sites P1 and P2, respectively. P1 contains one Pit-1-binding element (Ϫ129 to Ϫ123), and P2 contains two Pit-1-binding elements, one from Ϫ166 to Ϫ160 (P2 upper) and one from Ϫ171 to Ϫ165 in the opposite orientation (P2 lower). hGHRHR-Luc(Ϫ310,Ϫ19) containing mutations in either P2 upper or P1 showed sig-

FIG. 4. Promoter activity of the 5-flanking region of the human GHRH-R gene in heterologous cells.
The human GHRH-R 5Ј-flanking 2.2-kb fragment (Ϫ2207 to Ϫ19) was subcloned into the promoterless luciferase plasmids, pGL3 basic vector. Two g of these expression plasmids were transfected into GH3, HeLa, or BeWo cells, and luciferase activity was measured as described under "Experimental Procedures." rPRL-Luc, a reporter plasmid containing rPRL gene, was used as a positive control. Each value was normalized for total protein. Results represent the means Ϯ S.E. for three transfections from a representative experiment.

FIG. 5. Expression analysis of the 5-flanking region of the human GHRH-R gene in GH3 cells.
A, 2 g of reporter plasmids were transfected into GH3 cells by lipofection method and harvested 2 days after transfection. rPRL-Luc, a reporter plasmid containing rPRL gene, was used as a positive control. Luciferase activity was measured as described under "Experimental Procedures." Each value was normalized for total protein. Results represent the means Ϯ S.E. for three transfections from a representative experiment. Deletion of the 5Ј-flanking sequences from the 5Ј-end showed a moderate increase in luciferase activity as the deletion progressed from Ϫ2207 to Ϫ310. The deletion mutant, progressed to Ϫ130, showed a remarkable decrease in activity. The deletion mutant to Ϫ120 further decreased its activity. B, the regions from Ϫ130 to Ϫ120 and Ϫ310 to Ϫ130 have putative Pit-1-binding sites P1 and P2, respectively. P1 contains one Pit-1-binding element from Ϫ129 to Ϫ123, and P2 contains two Pit-1-binding elements from Ϫ166 to Ϫ160 in upper strand (P2 upper) and from Ϫ171 to Ϫ165 in lower strand (P2 lower). To clarify their role in stimulating GHRH-R gene expression, reporter plasmid containing mutation of P1 (P1m; TGTACGA, mutated sequences are underlined), P2 upper (P2m; TAGTAAG), or both P2 upper and lower (dP2m; GTCAGTCGTAAG) was made and transfected to GH3 cells. Also, pGL3 basic vector (negative control) or rPRL-Luc (positive control) was transfected into GH3 cells, and luciferase activity was measured. hGHRHR-Luc(Ϫ310,Ϫ19) containing P2 upper mutation (P2 m) or P1 mutation (P1m) showed significantly lower luciferase activity than wild hGHRHR-Luc(Ϫ310,Ϫ19), and hGHRHR-Luc(Ϫ310,Ϫ19) containing both P2 upper and lower mutations (dP2 m) showed marked reduction in the activity, which was almost same as background activity of pGL3 basic vector. On the other hand, when hGHRHR-Luc(Ϫ130,Ϫ19) was used as the reporter plasmid, P1 mutation did not change the luciferase activity, suggesting that P1 may work in the reporter plasmid with a longer 5Ј-region. nificantly lower luciferase activity than wild hGHRHR-Luc-(Ϫ310,Ϫ19), and hGHRHR-Luc(Ϫ310,Ϫ19) containing mutations in both P2 upper and P2 lower showed a marked reduction in activity, which was almost the same as that of the pGL3 basic vector (Fig. 5B). These findings, consistent with the results from the deletion study, suggested that P1, P2 upper, and P2 lower contribute to stimulation of GHRH-R expression.
On the other hand, when hGHRHR-Luc(Ϫ130,Ϫ19) was used as the reporter plasmid, a P1 mutation did not change the luciferase activity, suggesting that P1 may work in the reporter plasmid with a longer 5Ј-region (Fig. 5B). The drop in luciferase activity due to deletion of the region between Ϫ130 to Ϫ120, which was observed in the deletion study, could not be due to the loss of Pit-1 binding because inactivation of the Pit-1 site by mutation has no effect in the context of a short upstream region. In BeWo cells, co-transfection of the deletion mutants down to Ϫ310 and RSV-Pit-1 expression vector increased luciferase activity compared with control. However, deletion to Ϫ130 resulted in a decrease in the activity in BeWo cells even when Pit-1 expression vector was co-transfected (Fig. 6). Results of these transfection studies indicated that the region from Ϫ310 to Ϫ130 was important for Pit-1-dependent GHRH-R gene expression in BeWo cells.
DNase-I Footprint Analysis-The region from Ϫ310 to Ϫ130, which was important for the Pit-1-dependent expression of GHRH-R gene, contains DNA sequence P2 (Ϫ171 to Ϫ160), similar to the Pit-1-binding element. To confirm that Pit-1 binds to the sequence, DNase-I footprint studies were used. Recombinant Pit-1 protected the region from Ϫ176 to Ϫ155, in which P2 was included, in a dose-dependent fashion. However, Pit-1 did not protect P1 (Ϫ129 to Ϫ123), which is also similar to the Pit-1-binding element (Fig. 7).
Mobility Shift Assays for Analysis of Pit-1 DNA Binding-To further examine Pit-1 binding to P1 and P2, a mobility shift assay was used. A radiolabeled oligonucleotide probe spanning Ϫ175 to Ϫ142 (P2 probe) formed two specific DNA-protein complexes when incubated with Pit-1, because P2 has two putative Pit-1-binding elements. The two complexes were specifically competed by excess unlabeled oligonucleotide. When the consensus sequence of Pit-1 was mutated in the P2 probe, the upper DNA protein complex disappeared, confirming that the element was important for binding of Pit-1. Another radiolabeled oligonucleotide probe spanning Ϫ145 to Ϫ112 (P1 probe) formed only one DNA-protein complex when incubated with the same amount of Pit-1. When the consensus sequence of Pit-1 was mutated in the P1 probe, the binding of Pit-1 to P1 was reduced, confirming that the site was a Pit-1-binding element. However, the amount of DNA-protein complex was smaller than that when P2 probe was used, suggesting that the binding affinity of P1 to Pit-1 was weaker than that of P2 (Fig.  8). The finding appeared to be consistent with the result observed in the DNase-I footprint analysis that Pit-1 did not protect P1 (Fig. 7). DISCUSSION In the present study, we cloned the 5Ј-flanking region of the human GHRH-R gene by the PCR-based DNA walking method and determined transcription start site with RNase protection assay and 5Ј-RACE. Furthermore, we analyzed its function and Pit-1 dependence using transient transfection assay and deter- DNase-I footprint analysis of the promoter and enhancer region of GHRH-R gene was prepared by digestion of GHRH-R 5Ј-flanking region with Bsu36I and HindIII and uniquely radiolabeled at the upstream end by filling 3Ј-termini with [␣-32 P]deoxy-ATP using DNA polymerase. The DNA fragment contains two putative Pit-1-binding sites located from Ϫ129 to Ϫ123 (P1) and from Ϫ171 to Ϫ160 (P2). Recombinant His-tagged Pit-1 protected the region from Ϫ176 to Ϫ155, in which P2 is included (indicated by arrow-ended bracket). However, recombinant Pit-1 did not protect P1.
mined Pit-1-binding sites with DNase-I footprint analysis and mobility shift assay. PCR-based methods are available for walking from a known region to an unknown region of the genome. Recent improvement of PCR technique allowed us to obtain longer and more accurate sequences of genomic DNA of interest. We applied the method reported by Siebert et al. (18) to clone the 5Ј-flanking region of the human GHRH-R gene.
Recently Petersenn et al. (21) reported that 5Ј-flanking region of human GHRH-R. They (22) and we (23,24) have been independently trying to clone the 5Ј-upstream sequence of GHRH-R to analysis the mechanism of GHRH-R gene expression. The sequence reported here is almost identical to that reported by Petersenn et al., although minor differences are present. The reason why minor difference exists between theirs and ours is unknown. This difference may be explained by polymorphism.
The result of our RNase protection assay indicated that the major transcripts appeared to start at Ϫ122 nucleotides upstream from the translation start site in the normal pituitary gland (Fig. 3). Also, the 5Ј-RACE products were close to that of the major band obtained in RNase protection assay, supporting the results in RNase protection assay (Fig. 3). However, some shorter bands were detected in 5Ј-RACE experiments and in RNase protection assay. This finding suggests that some transcripts start from the downstream sites of the major transcript start site, which is often observed in TATA-less promoter.
Petersenn et al. (21) reported that the transcription start site of GHRH-R gene was 40 nucleotides upstream from the translation start site. However, the site reported by Petersenn et al. is located in the 5Ј-untranslated region of GHRH-R mRNA already reported by Mayo (1). In addition, our results obtained from RNase protection assay and 5Ј-RACE showed that the site was located in the sequence to be transcribed to mRNA. Thus, it is unlikely that the site reported by Petersenn et al. is a major transcription start site.
No intron was found in the 5Ј-flanking region of the human GHRH-R gene, because the sequence of the 5Ј-RACE products was completely identical to that obtained from the analysis of genomic DNA flanking to GHRH-R coding region. In addition, these results suggest the accuracy of the PCR-based method. The structure of the GHRH-R gene resembles that of recently characterized human vasoactive intestinal peptide receptor gene (25), which belongs to the same subfamily of G proteincoupled receptors, in respect to lack of intron in its 5Ј-untranslated region.
Typical TATA homologies were not present in the 5Ј-flanking sequence upstream from the transcription start site. A number of genes are known to lack obvious TATA boxes. Such TATAless promoters can be divided into two classes. One class consists of GC-rich promoters (26), found in housekeeping genes, which usually contain several transcription start sites spread over a fairly large region and several Sp1-binding sites (27,28). Another class has no apparent TATA boxes and is not GC-rich (27). Many of these promoters are not constitutively active but rather are regulated during differentiation and development (27). In the human GHRH-R gene, GC content appeared not so rich, and neither a CAAT box nor a typical Sp1-binding site was at the site appropriate to the start site. Thus, the human GHRH-R promoter may belong to the latter class.
Despite the absence of typical TATA homologies, analysis of the 2.7-kb nucleotide sequence of the 5Ј-flanking region revealed that it contained a number of putative regulatory cisacting elements. Of particular interest, we identified several sequences corresponding to Pit-1-binding consensus (TAT-NCAT) (Fig. 1). As described previously, the GHRH-R gene expression is considered to be controlled by Pit-1 (15)(16)(17). In fact, we demonstrated that our cloned 5Ј-flanking region of the gene increased luciferase activity in Pit-1-expressing GH3 cells and that the 5Ј-flanking region increased luciferase activity in non-Pit-1-expressing BeWo cells exclusively when RSV-Pit-1 expression vector was co-transfected. The region Ϫ310 to Ϫ130 was required for Pit-1-dependent promoter activity in BeWo cells according to the deletion mutant analysis. In the region, there are two Pit-1-binding elements, P2 upper and P2 lower. The mutation of P2 upper caused 50% reduction in the luciferase activity, and the mutation of both P2 upper and P2 lower suppressed the activity to basal levels in GH3 cells. These findings demonstrated that P2 is a functional site for GHRH-R FIG. 8. Mobility shift assays for analysis of Pit-1 DNA binding. A radiolabeled oligonucleotide probe spanning Ϫ175 to Ϫ142 (P2 probe) formed two specific DNA-protein complexes when incubated with Pit-1, because P2 have two putative Pit-1-binding elements. The two complexes were specifically competed by the excess of unlabeled oligonucleotide. When a consensus sequence of Pit-1 was mutated in P2 probe, upper DNA-protein complex disappeared, confirming that the site was important for binding of Pit-1. Another radiolabeled oligonucleotide probe spanning Ϫ145 to Ϫ122 (P1 probe) formed only one DNA-protein complex when incubated with the same amount of Pit-1. When the consensus sequence of Pit-1 was mutated in P1 probe, the binding of Pit-1 was reduced, confirming that the site was a Pit-1-binding element. However, the amount of DNA-protein complex was smaller than that when P2 probe was used, suggesting that the binding affinity of P1 to Pit-1 was weaker than that of P2. gene expression. Another important region for GHRH-R gene expression was located from Ϫ130 to Ϫ120. Although the Pit-1-binding site (P1) was present in the region, mutation of P1 did not affect the luciferase activity when hGHRHR-Luc(Ϫ130,Ϫ19) was used as a reporter plasmid. This finding was consistent with the results that P1 is a low affinity binding site for Pit-1 compared with P2, which was obtained from mobility shift assay and DNase-I footprint analysis. In addition, this appears to explain why co-transfected RSV-Pit-1 expression vector did not stimulate GHRH-R reporter gene deleted to Ϫ130 in BeWo cells. However, this finding does not exclude the possibility that P1 did not play a role in stimulating GHRH-R gene expression, because P1 mutation decreased luciferase activity by 40% when hGHRHR-Luc(Ϫ310,Ϫ19) was used as a reporter gene. P1 may function in combination with P2 or other cis-elements, located from Ϫ310 to Ϫ130.
On the other hand, the hGHRHR-Luc(Ϫ2207,Ϫ19) exhibited two times lower luciferase activity than serial deletion mutant constructs. The luciferase activity appeared to drop between Ϫ2207 and Ϫ1377 (Fig. 5A). The results might imply that the region Ϫ2207 to Ϫ1377 serves as a significant repressor in GH3 cells.
In summary, we have cloned and characterized the 5Ј-flanking region of the human GHRH-R gene. The 5Ј-flanking region contained a functional promoter, and its promoter activity was dependent on Pit-1. The binding site of Pit-1 was located from Ϫ176 to Ϫ155 judging from DNase-I footprint analysis, and the region containing the Pit-1-binding sites was important for Pit-1-dependent expression of GHRH-R gene. Characterization of this 5Ј-flanking region will further clarify the regulatory mechanism of human GHRH-R gene expression.