INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

ADRENOCORTICOTROPIN (ACTH), a key regulator of the hypothalamo-pituitary-adrenal axis, is expressed in the corticotroph cells of the anterior pituitary, the release of which is stimulated by two hypothalamic factors, corticotropin-releasing hormone (CRH) and vasopressin (9). Recently, various subtypes of receptors for CRH and vasopressin have been identified, with subsequent cloning of the genes being almost completed. Among the receptor subtypes, CRH-R₁ and V_1b vasopressin receptors are known to be expressed in the corticotroph cells of the pituitary (14, 19, 25). It has also been shown that both hormones are involved in the synthesis and secretion of ACTH through specific receptors (17).

The expression of ACTH, however, is not solely determined by the extracellular levels of CRH and/or vasopressin. It is well known that chronic stimulation of a hormone, in general, alters the expression of its own or related receptors, and indeed CRH has been shown to decrease the expression of CRH-R₁ receptor mRNA (26). In this sense, it is also important to clarify the regulation of the V_1b vasopressin receptor for complete understanding of the regulation of ACTH at the corticotroph level.

For this purpose, in this study we cloned, sequenced, and characterized the 5'-flanking region of the rat V_1b vasopressin receptor gene. The DNA we cloned showed substantial transcriptional activity in the AtT-20 cell, indicating that the region is the 5'-promoter of the V_1b receptor gene. More interestingly, multiple upstream open reading frames (uORFs) were found to be located in the 5'-untranslated region (5'-UTR) of the gene, and some of these were shown to have potent inhibitory effects on V_1b receptor expression.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cloning and sequencing. An ~5 kb of the upstream region of the rat V_1b vasopressin receptor gene (25) was subcloned into the pBluescript plasmid vector (Stratagene, La Jolla, CA) and sequenced by an ABI PRISM 310 genetic analyzer (Perkin-Elmer, Foster City, CA) with a BigDye Terminator Cycle Sequencing Reaction kit (Perkin-Elmer).

5'-Amplification of cDNA ends. The transcription start site(s) of the rat V_1b vasopressin receptor gene was determined by 5'-amplification of cDNA ends (RACE) by use of a Marathon cDNA amplification kit (Clontech, Palo Alto, CA). Total RNA was extracted from the Sprague-Dawley rat pituitary gland with TRIzol reagent (Life Technologies, Grand Island, NY), and poly(A)⁺ RNA was isolated using the poly(A) quick mRNA isolation kit (Stratagene). Then double-stranded cDNA synthesis and adapter ligation were performed, and the product was used as a template for the RACE analysis. PCR was performed using the AP1 primer (provided by the supplier) and the gene-specific antisense primer (5'-CAGTAGCTAGGATACCGATCTCC-3') complementary to the published nucleotide sequence of the rat V_1b receptor cDNA (25). Finally, the PCR product was cloned into pCR2.1 using the TA Cloning kit (Invitrogen, Carlsbad, CA), and >10 clones were isolated for nucleotide sequencing. If multiple clones had the same sequence starting at the most upstream region of the 5'-UTR, the 5'-end was recognized as the transcription start site.

Plasmid construction. Various lengths of the promoter region of the V_1b vasopressin receptor gene were obtained by digestion with appropriate restriction enzymes and subcloned into the plasmid containing luciferase reporter gene, pA3Luc (15). Inactivation of one or more translation start site(s) of the uORFs was performed using PCR-based site-directed mutagenesis (7), in which ATG was changed to TTG, ACG, ATT, and CTG for uORF-1, -2, -3, and -4, respectively. To see the translational activity of each uORF, the coding sequence of the luciferase protein was directly linked to the initiation codon of each uORF. For testing the function of each uORF in the heterologous context, the region containing an individual uORF (nucleotide positions between 472 and 345 for uORF-1 and -2, between 364 and 213 for uORF-3, and between 230 and 46 for uORF-4) was isolated and inserted between the Raus sarcoma virus (RSV) long terminal repeat promoter and luciferase reporter gene in pA3Luc. To check the sequence dependency of the function of uORFs, a construct in which the nucleotide sequences of the uORF-3 were replaced with the partial sequence of -lactamase was made and used as a representative one. Finally, constructs containing 5'-UTR with uORF-1 to -4 (between 472 and 1), in which all or all but one (uORF-3) of the ATGs were inactivated, were used in the mRNA expression experiment.

Cell culture and transfection. The murine corticotroph tumor cell line AtT-20/D16v (AtT-20) was maintained in a T₇₅ culture flask with Dulbecco's modified Eagle's medium (DMEM; Life Technologies) supplemented with 10% fetal bovine serum (FBS; Life Technologies) and antibiotics (50 µU/ml penicillin and 50 µg/ml streptomycin; Life Technologies) under 5% CO₂-95% atmosphere at 37°C. Culture medium was changed twice a week, and the cells were subcultured once a week.

Northern blot analysis. Northern blot analysis was performed essentially as previously described (27). Briefly, total RNA was isolated from the cells transfected with V_1b-luciferase (RSVORF or RSVORF3⁺) and pSEAP2-control plasmids by use of the RNeasy RNA extraction kit (Qiagen, Hilden, Germany). Fifteen micrograms of the RNA were denatured and electrophoresed with 1% agarose gel and then transferred to a nylon membrane (GeneScreen Plus, NEN, Boston, MA), followed by prehybridization at 42°C for 3 h. Antisense oligonucleotide probes complementary to the coding sequences of the luciferase or of the rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA were labeled with [-³²P]dCTP by use of the Nick Translation System (Promega, Madison, WI). Hybridization was performed at 42°C for 20 h in a rotating bottle of the hybridization oven. Membranes were then washed, and the signals were densitometrically quantified using a Bioimage analyzer (BAS-2000, Fuji Photo Film, Tokyo, Japan), with subsequent normalization against a corresponding relative amount of GAPDH mRNA in each sample. The values were also normalized with transfection efficiency determined by the SEAP activity.

Reporter assays. Luciferase assay was performed as previously described (1). SEAP activity was determined using the SEAP reporter assay kit (Toyobo, Osaka, Japan). The -gal activity was determined by Galacto-Light Plus -gal assay kit (Tropix, Bedford, MA).

Data analysis. Samples in each group of the experiments were in triplicate or quadruplicate. All data were expressed as means ± SE. When statistical analyses were performed, data were compared by one-way ANOVA with Fisher's protected least significant difference test, and P values <0.05 were considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Determination of the nucleotide sequences and gene structure of the 5'-flanking region and 5'-UTR of the rat V_1b vasopressin receptor gene. Approximately 5 kb of the upstream region of the rat V_1b vasopressin receptor gene were sequenced and deposited in the GenBank database (accession number AB042197). Figure 1 shows the nucleotide sequences of the downstream region (1000 to +3; +1 designates the translation start site of the open reading frame encoding the V_1b receptor protein). 5'-RACE identified the major transcriptional start site 801 bp upstream from the authentic translation start site. There was no TATA box, but stretches of CT or CA repeats were recognized near the transcription start site. In addition, comparison of the sequences with the sequence of cDNA revealed a new intron within the 5'-UTR (769 to 608). Furthermore, analysis of the sequences of the 5'-UTR showed five uORFs that precede the major ORF encoding the V_1b receptor protein.

View larger version (80K): [in this window] [in a new window]   Fig. 1.   Nucleotide sequences of the rat V1b vasopressin receptor gene 5'-promoter and 5'-untranslated region (UTR). Italic, bold, and lower case letters designate the 5'-promoter, 5'-UTR, and intron, respectively. The major transcription start site is circled. Boxes and underlines indicate the initiation and termination codon, respectively. +1 designates the first nucleotide of the translation start site of the major open reading frame (ORF) for V1b protein.

Serial deletion analysis of the transcriptional activity of the rat vasopressin V_1b receptor gene 5'-promoter. The transcriptional activities of the various lengths of the 5'-promoter region of the cloned gene were examined using the AtT-20 mouse corticotroph cell line. As shown in Fig. 2, substantial luciferase activity was obtained in all of the constructs compared with the promoterless vector. Serial deletion revealed that the promoter activity was highest in the second shortest construct examined (1641-444) and was still maintained in the shortest construct examined (1173-444). These results indicate that the region examined is indeed a 5'-promoter of the V_1b vasopressin receptor gene and that the area between 1173 and 444 appears to be enough for the basal promoter activity.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effects of serial deletion of the rat V1b vasopressin receptor 5'-promoter (4894, 3366, 2663, 1641, 1173) on basal transcriptional activity in corticotroph cells. AtT-20 cells were transfected transiently with various lengths of the rat V1b vasopressin receptor 5'-promoter luciferase (Luc) fusion gene. Values were normalized by -galactosidase (-gal) activity used as an internal control.

Characterization of the function of the uORFs in the 5'-UTR of the V_1b vasopressin receptor mRNA. As mentioned above, nucleotide sequences of the 5'-UTR showed the presence of multiple uORFs (designated as uORF-0 to -5), and thus we tried to see whether any of the uORFs would influence the expression of the V_1b receptor. We focused on uORFs-1 to -4, because the first one (uORF-0) is very close to the transcription start site. When the 5'-flanking region with or without the uORFs was subcloned into the luciferase vector and expressed in AtT-20 cells, luciferase activity was much lower in the construct with uORFs (WT) than in that without them (ORFs) (Fig. 3), indicating the inhibitory nature of some of the uORF(s).

View larger version (22K): [in this window] [in a new window]   Fig. 3.   A: schematic representation of the upstream ORFs (uORFs) in the 5'-UTR of the V1b vasopressin receptor gene. Wild-type (WT) construct contains the uORFs (uORF-1 to -4), whereas the ORF construct does not. B: luciferase expression of the WT and ORFs in AtT-20 cells. Values were normalized by -gal activity. *P < 0.05 vs. ORFs.

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Effects of inactivation of the initiation codon of each uORF on luciferase expression. WT or mutated (ORFs, M1, M2, M3, M4) constructs were transfected into the AtT-20 cells, and luciferase activity was determined. Values were normalized by -gal activity. Circles and crosses designate the intact and inactivated initiation codons, respectively. *P < 0.05 vs. ORFs.

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Effect of individual uORFs on luciferase expression in AtT-20 cells. A: comparison of luciferase expression among constructs in which none (1/2), only uORF-1 (ORF1), or only uORF-2 (ORF2) is intact. uORF-3 and -4 are deleted in all constructs. B: comparison of luciferase expression among constructs in which none (ORFs), only uORF-3 (ORF3), or only uORF-4 (ORF4) is intact. C: translational activity of the uORF-3 and -4. The initiation codon of either uORF-3 or -4 was directly linked to luciferase protein (ORF3AUG, ORF4AUG), and expression of luciferase protein was compared with that of ORFs. *P < 0.05 vs.ORFs.

Effect of the uORFs in the context of a heterologous promoter. To see whether the negative effect of the uORFs is exhibited only in the V_1b vasopressin receptor gene or is still functional in a heterologous context, a series of constructs in which the DNA fragments containing each uORF were individually located downstream of the RSV promoter were expressed in AtT-20 cells. The results showed a pattern virtually similar to that of the V_1b vasopressin gene: mild suppressive effect with uORF-2, and potent inhibition with both uORF-3 and -4 (Fig. 6, A and B). Thus the suppression is valid in a heterologous context as well, suggesting a common mechanism in the inhibitory effect of the uORFs.

View larger version (36K): [in this window] [in a new window]   Fig. 6.   Effect of an individual uORF in a heterologous promoter on luciferase expression in AtT-20 cells. A: the 5'-UTR of the V1b vasopressin receptor gene containing uORF-1 and -2 was isolated and inserted between the Raus sarcoma virus (RSV) promoter and luciferase reporter gene. Because uORF-1 and -2 overlap, the insert section contains both uORFs, but the initiation codon of either uORF-1 (RSVORF1) or uORF-2 (RSVORF2) is made intact. The luciferase expression of the constructs was compared with that without uORF (RSVORFs). B: the 5'-UTR of the V1b vasopressin receptor containing uORF-3 or -4 was isolated and inserted between the RSV promoter and the luciferase reporter gene. Luciferase expression of constructs was compared with that without uORF (RSVORFs). C: the coding sequence of uORF-3 except initiation and stop codons was replaced with an unrelated one (partial coding sequence of -lactamase) (RSVORF3R), and the suppressive effect was compared with the authentic one (RSVORF3A) or the construct without uORFs (RSVORFs). *P < 0.05 vs. RSVORFs.

View larger version (40K): [in this window] [in a new window]   Fig. 7.   Effect of uORF-3 on the expression of luciferase mRNA and protein. Constructs containing no uORF (RSVORF) or only uORF-3 (RSVORF3+) were transfected into AtT-20 cells, and both mRNA and protein levels were determined by Northern blot analysis and luciferase assay, respectively. Photograph shows representative Northern blot of luciferase mRNA. *P < 0.05 vs. RSVORF.

Possible regulation of V_1b receptor expression by intracellular calcium through the uORFs. Finally, to see whether the uORF-mediated inhibition is associated with a specific regulatory process, we examined the responsiveness of V_1b promoter-luciferase protein expression to increased intracellular calcium by thapsigargin, which mimics the stimulation of corticotroph with vasopressin through the V_1b receptor. In the construct with uORFs (WT), treatment of the cells with thapsigargin significantly increased the V_1b-luciferase expression (Fig. 8), whereas no response was found in that without uORFs (ORFs). This suggests that an intracellular calcium-mediated signaling pathway might regulate the expression of the V_1b receptor posttranscriptionally through the uORFs.

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Effect of increase in intracellular calcium on expression of luciferase protein. Constructs with (WT) or without uORFs (ORFs) were treated with thapsigargin (2 µM) or vehicle for 5 h. Open and closed bars represent cells treated with vehicle or thapsigargin, respectively. *P < 0.05 vs. vehicle-treated group.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

In this paper, we cloned the 5'-promoter and 5'-UTR of the rat V_1b vasopressin receptor gene and determined the nucleotide sequences up to 5 kb from the translation start site. Using the cloned gene, we confirmed that the 5'-promoter region actually has transcriptional activity in the corticotroph cell line AtT-20. Interestingly, the 5'-UTR of the gene has multiple uORFs, some of which are found to exert inhibitory effects on protein synthesis. Our results raise the possibility that the expression of the V_1b receptor is regulated not only at the transcriptional but also at the posttranscriptional level.

The vasopressin receptor consists of three subtypes, i.e., V_1a, V_1b, and V₂ (28). The V₂ receptor plays a key role in the regulation of water reabsorption at the collecting duct of the kidney (10), whereas the V_1a receptor is thought to be involved in cardiovascular regulation (6). The V_1b receptor, on the other hand, is expressed mainly in the corticotroph of the anterior pituitary, where it mediates the effect of vasopressin on ACTH secretion (21). In this study, our data clearly indicate by use of the luciferase reporter system that the area we examined indeed has promoter activity in corticotroph cells. It is a TATA-less promoter, and ~0.35 kb from the transcription start site (between 1173 and 801) seems to be enough to maintain the basal promoter activity. Moreover, repressor and enhancer elements were suggested to be located in the regions 2663-1641 and 1641-1173, respectively. Interestingly, Rabadan-Diehl et al. (22) failed to find any transcriptional activity of the same promoter in AtT-20 cells.

The most interesting feature of the 5'-UTR sequences of the V_1b vasopressin receptor gene is the location of five uORFs preceding the initiation codon of the major ORF, which is relatively uncommon in mRNA of vertebrate cells. Initially, we tried to characterize the transcriptional regulation of the V_1b promoter region by using the constructs with the uORFs, and we found very low promoter activity. Subsequently, however, when the area containing uORFs (uORF-1 to uORF-4) was eliminated from the construct, high basal activity was consistently obtained, suggesting that some of the uORFs have an inhibitory effect on protein expression. Extensive examinations by mutating the translation start sites of each uORF revealed that the two downstream uORFs (uORF-3 and -4) seemed to mediate the suppressive effect in both homologous and heterologous contexts. In fact, when the sequences around the initiation codon of each uORF were compared with Kozak's consensus (11), those of the two downstream uORFs were in favorable context (Table 1). Furthermore, the expression of the protein product (luciferase) in the construct with intact uORF-3 was much decreased compared with that with the inactivated one despite the comparable levels of mRNAs, suggesting strongly that the suppression by the uORF is occurring at the posttranscriptional level.

The mechanism(s) whereby these uORFs are involved in the suppressive effect is not entirely clear. In general, one mRNA harbors a unique ORF, and <10% of the vertebrate mRNA has uORF(s) (12). However, it is reported that uORFs are found with a relatively high frequency in the 5'-UTR of the G protein-coupled receptor (GPCR) genes such as CRH, angiotensin II, and ₂-adrenergic receptors (3, 13, 16, 18, 29). If an mRNA has uORF(s), it is expected that the uORF(s) somehow interferes with the function of the major ORF, because re-initiation of translation is not common in eukaryocytes. Two major possible mechanisms regarding the function of uORF(s) are proposed. The first possibility is that the product of the uORF(s) (a small peptide) interferes with the initiation of the authentic ORF (4, 24). The second one is that the initiation at the uORF(s) by itself inhibits the re-initiation at downstream ORF (2, 3, 8). Our data show that the inhibition occurred mainly at the uORFs with Kozak's consensus (11), and actually uORF-3 and -4 were functional in terms of translation. We also found that replacement of the coding sequence of uORF3 with a completely different one (RSVORF3R) still had a significant suppressive effect, supporting the second possibility (sequence-independent event). However, the sequence-dependent mechanism cannot completely be ruled out, because the degree of inhibition was significantly less in RSVORF3R than in RSVORF3A.

Another interesting point is whether any kind of regulatory mechanism(s) involves the function of uORF(s). In some cases, such as carbamoyl-phosphate synthetase or S-adenosylmethionine decarboxylase mRNAs, the inhibition through uORF(s) is shown to be regulated by the metabolic milieu within the cell (4, 24). Because uORFs are sometimes found in the GPCR or other types of receptor genes (13), including our case, it may be possible that the preceding uORF(s) is regulating the expression of the major ORF by switching its translation on and off. In our study, the expression of the V_1b vasopressin receptor constructs containing the uORFs was potently suppressed at the posttranscriptional level in AtT-20 cells. However, because the V_1b receptor mRNA containing the same uORFs is normally expressed in the corticotroph cells in vivo, there should be some mechanism(s) for releasing the inhibition by uORFs. In fact, a discrepancy between the levels of mRNA and binding capacity in vivo has been suggested, further indicating the additional posttranscriptional regulatory step(s) (20, 23). Our data showed the increase in V_1b-luciferase expression by thapsigargin, suggesting that vasopressin-induced rise in intracellular calcium through the V_1b receptor might regulate the expression of the receptor itself through an uORF-mediated posttranscriptional mechanism. We could not confirm this hypothesis in this study, because AtT20 cells did not express a functional V_1b vasopressin receptor (data not shown).

CRH and vasopressin, the two major positive regulators of ACTH secretion, exert their effects through different intracellular signaling pathways, i.e., the cAMP/protein kinase A and the phosphoinositide/Ca²⁺ pathways, respectively. Furthermore, the cross talk between the two pathways has also been shown (5). In this sense, it is possible that one ligand influences the expression of the receptor of the other ligand(s), and indeed the administration of CRH is shown to decrease the V_1b receptor mRNA as well as that of the CRH-R₁ receptor (26). Interestingly, the mRNA of the CRH-R₁ receptor also has uORFs (29), and therefore we suppose that this kind of cross talk may be occurring through the uORFs. Investigation of the regulation of the CRH-R₁ and V_1b vasopressin receptors at both transcriptional and posttranscriptional levels will be necessary for the complete understanding of the molecular mechanism of the CRH-vasopressin interaction.