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

J. Biol. Chem., Vol. 275, Issue 37, 28882-28887, September 15, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/37/28882    most recent M002175200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Baumeister, H. Articles by Meyerhof, W. Search for Related Content PubMed PubMed Citation Articles by Baumeister, H. Articles by Meyerhof, W. Social Bookmarking                      What's this?

The POU Domain Transcription Factor Tst-1 Activates Somatostatin Receptor 1 Gene Expression in Pancreatic -Cells^*

Hans Baumeister and Wolfgang Meyerhof

From the Abteilung Molekulare Genetik, Deutsches Institut für Ernährungsforschung und Universität Potsdam, Arthur-Scheunert-Allee 114-116, D-14558 Potsdam-Rehbrücke, Germany

Received for publication, March 15, 2000, and in revised form, June 21, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The peptide hormone somatostatin inhibits the release of insulin. The gene encoding somatostatin receptor 1 is expressed in pancreatic -cells and insulinoma RIN 1046-38 cells. In the present study the mechanisms underlying the regulation of the somatostatin receptor 1 gene in pancreatic -cells were investigated. Transient transfections of RIN 1046-38 cells with promoter/reporter gene constructs and footprint analysis revealed two regions, fp1 and fp2, that were necessary for the observed promoter activity. Mutagenesis of the fp2 region delineated the cis-acting element to the motif 5'-TTAATCATT-3'. The POU domain transcription factor Tst-1 was identified as trans-activator mediating the 5'-TTAATCATT-3' motif-dependent transcription in RIN 1046-38 cells and heterologous CV1 cells. Tst-1, known as a transcriptional regulator in keratinocytes, glial cells, and neurons, has been detected by immunohistochemistry in pancreatic islets. Altogether, we demonstrate Tst-1 as transcriptional regulator in pancreatic neuroendocrine cells.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The major neuroendocrine cell types of pancreatic islets, the -, -, -, and pancreatic polypeptide-cells, secrete the peptide hormones, glucagon, insulin, somatostatin, and pancreatic polypeptide, respectively. These hormones are powerful regulators of whole body metabolism and are released in response to gastrointestinal, nutritional, and neuronal signals (1). One important function of somatostatin is to limit the secretion of insulin, glucagon, pancreatic polypeptide, and somatostatin itself (1, 2). Molecular cloning identified six somatostatin receptors (sst1, sst2A, sst2B, sst3, sst4, and sst5) that bind somatostatin with high affinity, are coupled to G-proteins, and regulate various cellular effectors (for review see Refs. 3 and 4). The sst subtypes are encoded by five genes that are differentially expressed in human and rat pancreatic islets. sst1 has been exclusively detected in human -cells, whereas sst2 is confined to -cells. sst3 and sst4 were found in subpopulations of -cells. sst5 is present in - and -cells (5). Similar results were obtained in rat pancreatic islets for sst2A and sst5 (6). In line with these observations sst2- and sst5-specific agonists inhibited the release of glucagon and insulin from rodent islet preparations, respectively (7-10). An sst1-specific agonist, [des-Ala¹,des-Gly²,des-Asn⁵,D-Trp⁸,IAmp⁹]somatostatin-14 (CH-275), became recently available (11, 12). In rat pancreatic insulinoma RIN 1046-38 cells, which serve as a model to study insulin synthesis and secretion (13), and in pituitary GH₁2C₁ cells stably transfected with sst1, CH-275 inhibited the Ca²⁺ influx through voltage-gated Ca²⁺ channels (14, 15). The inhibition of voltage-gated Ca²⁺ channels may represent a mechanism by which somatostatin inhibits the release of insulin (16). Therefore, it was concluded that, in addition to sst5, sst1 receptors may be involved in the regulation of insulin release as well (5, 15).

The observed differential expression of the sst1 gene in pancreatic islets, the gastrointestinal tract, and distinct regions of the brain and the pituitary suggests that cell type-specific transcription factors regulate the sst1 gene promoter (17-21). A number of such factors are necessary for pancreatic gene expression and cellular differentiation (22). Many of these regulators belong to the large family of homeodomain proteins, such as IPF-1, ISL1, Pax4, Pax6, Nkx2.2, and Nkx6.1. POU domain proteins constitute a subfamily of the homeodomain proteins and are characterized by the additional presence of a POU-specific domain (23). Members of this family, such as Pit-1, Brn-2, and Brn-4, are well known as regulators of neuroendocrine gene expression and cellular differentiation in the brain, pancreatic -cells, and the pituitary (24-26). In the pituitary, the POU domain protein Pit-1 has been identified recently as an essential regulator of the sst1 gene (27). In the present paper we investigated whether other POU domain proteins could be responsible for the activation of the sst1 gene in pancreatic -cells.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Isolation of Pancreatic Islets-- Pancreatic ducts of decapitated adult rats were clamped at the distal end, and 5 ml of 1% bovine serum albumin, 8 mg/ml collagenase (type IV, 273 units/mg, Worthington) in Hanks' buffered saline solution was injected into the gallbladder duct. The pancreas was removed and incubated at 37 °C for 2 h. The digestion was stopped by addition of ice-cold Hanks' buffered saline solution containing 1% bovine serum albumin. Islets released from the pancreatic tissue were picked with a 20-µl pipette under a light microscope and, for RT-PCR¹ analysis, immediately transferred to ice-cold PeqGold solution (PeqLab, Erlangen, Germany).

Reporter Gene Constructs, Cell Culture, Transient Transfections, Luciferase, and Electrophoretic Mobility Shift Assays-- All of the plasmids and experimental techniques have been described in detail recently (27-29). For luciferase assays it should be noted that the luciferase activities were normalized for efficiency of transfection by determination of -galactosidase activities and protein concentrations (Bradford dye kit, Bio-Rad).

DNase I Footprinting Analyses-- p-324sst1Luc DNA was digested either with KpnI (which cleaves in the polylinker 5' to the sst1 gene fragment) and BssHII (position +52) or with MluI (which cuts in the polylinker 5' to the sst1 gene fragment) and PvuII (positions +45). For labeling the coding and noncoding strand, the 5'-overhangs of the BssHII site or of the MluI site were filled in, respectively, using [-³²P]dCTP and [-³²P]dGTP. Fragments were separated on 5% polyacrylamide, purified, digested with DNase I using the Hotfoot kit (Stratagene, Heidelberg, Germany), and analyzed on 8% sequencing gels.

RT-PCR-- Total RNA was isolated from RIN 1046-38 cells, 30 isolated rat pancreatic islets, and rat brain using PeqGold (PeqLab, Erlangen, Germany) and treated with DNase I (2 units/5 µg RNA) for 1 h at 37 °C in the presence of 44 units of RNase Inhibitor (Fermentas, Vilnius, Lithuania), 10 mM MgCl₂, and 1 mM dithiothreitol. 2 µg of RNA was reverse-transcribed using mouse mammary leukemia virus reverse transcriptase (Life Technologies, Inc.) and oligo(dT)₁₈ primers (Amersham Pharmacia Biotech). sst1 and -actin cDNA fragments were amplified as described previously (15, 21). For analysis of tst-1 gene expression the rat tst-1 specific primers (5'-CGCTGCACGAGGACGGCCAC-3'; position 606-625, and 5'-TCGAGCGCGCCTTTGACACC-3'; position 1101-1082, of the rat Tst-1 cDNA (30)) were employed as described (31).

Western Blot Analysis-- Protein extracts were prepared from RIN 1046-38 and CV1 cell nuclei as described previously (32), analyzed on SDS-15% polyacrylamide gels (15 µl, 20 µg of protein), and transferred onto cellulose nitrate membranes (Optitran BA-S 83, Schleicher & Schüll). Immobilized proteins were stained with Amido Black (33) to control the transfer of proteins. Following destaining, membranes were incubated in 5% milk powder, 0.05% Tween 20 and incubated overnight at room temperature with the 1:3000 diluted primary antibody directed against Tst-1 (32). Anti-rabbit IgG conjugated with sheep alkaline phosphatase was employed as secondary antibody (Sigma). A ready-made protein mixture was used as molecular weight marker (Sigma, Deisenhofen, Germany).

Immunohistochemistry-- Animal experimentation was carried out in accordance with the German Animal Protection Act and approved by local authorities (No. 48-3560-0/3). Following transcardial perfusion of adult rats with 4% paraformaldehyde, the isolated pancreata were post-fixed in the same fixative for 3 h and then infiltrated overnight in 20% sucrose at 4 °C. After mounting the tissue onto cryostat chucks, serial sections of 14 µm were cut at 20 °C (Mikrom HM 505 E, Walldorf, Germany) and mounted onto poly-L-lysine-coated slides. For immunohistochemical detection the Tst-1-specific primary antiserum (1:4000) and a secondary anti-rabbit IgG antibody coupled to peroxidase (ABC-Kit, Vector, Burlingame, CA) were used. For control, a rabbit serum raised against Pit-1 was used as primary antiserum in a final dilution of 1:4000. The expression of Pit-1 in adult rats is restricted to pituitary cells (24).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Identification of sst1 Gene Regulatory Elements-- It has been reported that the sst1 gene is expressed in human pancreatic -cells (5), but no such data were available for the rat. Therefore we examined whether sst1 mRNA can be detected in rat pancreatic islets. In fact, Fig. 1 shows sst1 transcripts to be present in this tissue. In line with this observation, sst1 has been identified as major somatostatin receptor in rat insulinoma RIN 1046-38 cells (15). Hence, this cell line represents an appropriate model to study sst1 gene promoter function in -cells. 5'-Flanking sequences of various lengths from the sst1 gene were fused to the luciferase reporter gene, and the resulting constructs were employed in transfection assays of RIN 1046-38 cells (Table I). A construct comprising the region from 1985 to +190 (1985sst1Luc) showed a 14-fold increase of luciferase activity in RIN 1046-38 cells compared with basicLuc lacking a gene promoter. Deletion of the DNA region from 1985 to 165 did not significantly change the activity of the reporter (Table I). However, further reduction of the 5'-flanking sequences by 48 bp decreased the luciferase activity by 62 ± 4%. The residual activity was abolished upon deletion of the region from 48 to +52. Therefore, it is concluded that the regions from 165 to 117 and 48 to +52 are essential for transcription of the sst1 gene in RIN 1046-38 cells. In CHO cells, the reporter activity of 1985sst1Luc is low, only about 2-fold above background. Luciferase activity remained almost unaffected by the 5'-deletions down to 48 and is abolished using the +52sst1Luc construct. These observations suggest that the region between 48 and +52 contains the basal sst1 gene promoter and that the region between 165 and 117 does not contribute to the low promoter activity in this cell line.

View larger version (56K): [in this window] [in a new window]   Fig. 1.   Detection of sst1 transcripts in isolated pancreatic islets by RT-PCR. Total RNA from isolated pancreatic islets was reverse-transcribed and analyzed for the presence of sequences encoding sst1 by amplification of a 318-bp cDNA fragment (arrow) using sst1-specific primers. The quality of the cDNA was controlled by a PCR specific for -actin cDNA. No cDNA fragment was amplified in absence of the reverse transcriptase suggesting that the isolated RNA was not contaminated with genomic DNA. Amplified fragments were analyzed on a 1.5% agarose gel. M, 100-bp ladder.

Identification of cis-Acting Regulatory Elements-- DNase I protection assays were performed to detect binding sites for transcriptional regulators in the promoter region of the sst1 gene. Three regions, fp1, fp2a, and fp2b, were protected from DNase I digestion in the presence of nuclear proteins extracted from RIN 1046-38 cells (Fig. 2). fp1 is located between 3 and +19 around the transcriptional start site (34) (Fig. 2B). Since fp1 is located within the region that mediated low promoter activity in RIN 1046-38 and CHO cells, it likely represents the basal sst1 gene promoter. fp2a covers 71 bp between 126 and 56. Although no functional activity could be assigned to this region it contains several putative binding sites for nuclear factors (Fig. 2B). Directly adjacent to fp2a, fp2b extents from 163 to 134. It contains an A/T-rich sequence element located between 154 and 140. The location of the fp2b site within the region from 165 to 117 suggests that it is of functional importance. This assumption has been assessed by fusing a pentameric fp2b site in both orientations to the heterologous SV40 early promoter. The corresponding constructs, fp2bfSV40Luc and fp2brSV40Luc, were transfected into RIN 1046-38 cells and, for control, in CHO cells. Fig. 3A clearly shows that fp2b DNA enhanced the activity of the heterologous SV40 promoter in an orientation-independent manner in insulinoma cells only. Mutations of the fp2b region (Table II) revealed that the A/T-rich sequence element 5'-TTAATCATT-3' located between 151 and 143 is necessary for full promoter activity (Fig. 3B). Both findings demonstrate that the fp2b region represents a cis-acting positive regulatory element of the sst1 gene promoter in RIN 1046-38 cells.

View larger version (44K): [in this window] [in a new window]   Fig. 2.   DNase I footprinting of the sst1 gene upstream region. A, the radiolabeled coding (left panel) and noncoding strands (right panel) of an sst1 gene fragment between 324 and +52 were treated with DNase I with increasing amounts (0, 10, 20 µg of protein) of nuclear extracts prepared from RIN 1046-38 cells. The G + A sequencing ladder and DNase I reactions were analyzed on an 8% sequencing gel. Three footprints, fp1, fp2a, and fp2b, have been identified. Arrows point to DNase I-hypersensitive sites found at the ends of the protected site fp2b. Note that the G + A sequencing ladder of the noncoding strand corresponds to C and T nucleotides of the coding strand presented in B. B, the sequence of the protected sites and their positions in the sst1 gene are given. An asterisk marks the position of the mRNA start point as described previously (34). Lines above the sequence of fp2a indicate the positions of sequences that are homologous to binding sites of POU domain proteins (OCT (38, 40)), of AP-1 (54), or of the heterodimeric nuclear receptors RxR/T3R (55, 56). In fp2b the binding site for Tst-1 identified in this study is indicated (see below).

View larger version (39K): [in this window] [in a new window]   Fig. 3.   The fp2b DNA constitutes an essential region of the sst1 gene promoter in RIN 1046-38 cells. A, the fp2b DNA has been fused to the SV40 early promoter/luciferase reporter gene (SV40Luc). Five copies of the fp2b DNA were present in forward (fp2bfSV40Luc) or in reverse orientation (fp2brSV40Luc). Following transient transfection of RIN 1046-38 cells, or for control, CHO cells luciferase activities were determined and related to those obtained with SV40Luc. B, mutations of the fp2b region reduced the sst1 gene promoter activity. The four mutations, m1, m2, m3, and m4, presented in Table II were introduced into the 324sst1Luc construct. RIN 1046-38 cells were transiently transfected with the resulting plasmids, 324m1Luc, 324m2Luc, 324m3Luc, 324m4Luc, wild type 324sst1Luc, and basicLuc. Luciferase activities were related to that of basicLuc. Representative experiments performed in triplicate are shown.

Identification of Transcription Factors That Bind to the fp2b Region-- To identify the proteins that bind to the fp2b region, electrophoretic mobility shift assays (EMSAs) were carried out. Four DNA-protein complexes, c1, c2, c3, and c4, were detected with fp2b DNA and RIN 1046-38 cell nuclear proteins (Fig. 4A, left panel, lane 2). The complexes c3 and c4 appeared as double band on the gels. Several POU domain proteins were tested for their ability to bind fp2b DNA. Antisera directed against Brn-1, Brn-2, Tst-1, Oct-1, or Pit-1 were applied in EMSAs. Two fp2b DNA-binding proteins forming complexes c2 and c4 were identified as Oct-1 and Tst-1, respectively (Fig. 4A, left panel). In contrast, no binding of Brn-1, Brn-2, and Pit-1 could be detected. The identity of Tst-1 as an fp2b DNA-binding protein was confirmed by EMSA analysis of nuclear extracts from CV1 cells previously transfected with a Tst-1 expression plasmid (CV1/Tst-1, Fig. 4A, right panel). The major complex co-migrated with complex c4 and was completely abolished by the Tst-1 antiserum (Fig. 4A, right panel, lane 3). These results indicate that Tst-1 and Oct-1 bind to the fp2b DNA. The proteins that formed complexes c1 and c3 remain to be identified.

View larger version (47K): [in this window] [in a new window]   Fig. 4.   Electrophoretic mobility shift assays with radiolabeled fp2b DNA. A, left panel, the radiolabeled fp2b probe (1 ng, for sequence see Table II) was incubated with nuclear extracts from RIN 1046-38 cells (lanes 2-7), or for control, without extract (lane 1). c1, c2, c3, c4, denote the observed DNA-protein complexes. Lanes 3-7, supershift experiments with antisera against various POU domain transcription factors. Arrows that point to c2 and c4 indicate complexes eliminated by -Oct-1 (lane 3) and -Tst-1 antisera (lane 7). Asterisk denotes complexes found with antisera for -Brn-1, -Brn-2, and -Tst-1 in the absence of nuclear extract (data not shown). Right panel, the fp2b probe was incubated with nuclear extracts from RIN 1046-38 or CV1 cells (CV1/Tst-1) previously transfected with a Tst-1 expression plasmid (29) in the absence or presence of the -Tst-1 antiserum. B, mutations within the A/T-rich sequence of the fp2b region results in diminished binding of Tst-1. The radiolabeled fp2b DNA was incubated with the nuclear extract from RIN 1046-38 cells in presence of 10-, 50-, or 100-fold excess of the competitor DNA m1fp2b, m2fp2b, m3fp2b, and m4fp2b, which represent mutant fp2b DNA (for sequences see Table II). The competition with the homologous competitor (fp2b) indicates that the complexes c1-c4 represent specific protein DNA interactions.

Binding of the RIN 1046-38 cell nuclear proteins at the fp2b site was also analyzed by competition studies with mutant fp2b DNA (Table II). Three mutants, m1fp2b, m2fp2b, and m3fp2b, did not compete or only weakly competed with the wild type fp2b DNA for binding of Tst-1, Oct-1, and the unidentified proteins (Fig. 4B). In contrast, m4fp2b showed strong competition. Thus, the sequence 5'-TTAATCATT-3' that has been demonstrated to be essential for full sst1 gene promoter activity (Fig. 3B) appears to be the binding site of Tst-1.

Tst-1 Activates the sst1 Gene Promoter by Binding to the fp2b DNA Region-- CV1 cells were co-transfected with increasing amounts of the expression plasmid pCMVTst-1 and constant amounts of the p-324sst1Luc plasmid or the promoter-less plasmid pbasicLuc. Fig. 5A shows that co-transfections with pbasicLuc resulted, independently of the amount of pCMVTst-1, only in basal luciferase activities. However, co-transfections of p-324sst1Luc with pCMVTst-1 caused a dose-dependent 7-fold increase in luciferase activity. The half-maximal effect was observed at a ratio of 1:2 of the expression plasmid pCMVTst-1 and p-324sst1Luc (Fig. 5A). Therefore, it is concluded that Tst-1 not only binds to the fp2b DNA but also trans-activates the sst1 gene promoter in a heterologous expression system. In order to demonstrate that the fp2b region is necessary for activation by Tst-1, CV1 cells were co-transfected with the mutant constructs p-324m1Luc, p-324m2Luc, p-324- m3Luc, or p-324m4Luc and the expression plasmid CMVTst-1 (Fig. 5B). In line with the previous results, the mutants p-324 m1Luc, p-324m2Luc, and p-324m3Luc did not show promoter activity nor could they be trans-activated by Tst-1. These results indicate that Tst-1 trans-activates the sst1 gene promoter by binding to the 5'-TTAATCATT-3' element of the fp2b DNA.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Dose- and sequence-dependent trans-activation of the sst1 gene by Tst-1. A, CV1 cells were transfected with 10 µg of 324sst1Luc (closed circles) or basicLuc (open circles) in the presence of increasing amounts of the Tst-1 expression vector CMVTst-1, and luciferase activities were determined. B, the A/T-rich sequence within the fp2b region is necessary for Tst-1 to activate the sst1 gene promoter. CV1 cells were transfected in absence (vehicle) or presence of 5 µg of CMVTst-1 expression plasmid with 10 µg of the wild type construct, 324sst1Luc, basicLuc, 324m1Luc, 324m2Luc, 324m3Luc, or 324m4Luc (Fig. 3, Table II). Relative luciferase activities obtained with basicLuc were set to 1. Representative experiments performed in triplicate are shown.

Presence of Tst-1 in RIN 1046-38 Cells and Pancreatic Islets-- If Tst-1 is a physiological regulator of the sst1 gene in vivo, Tst-1 mRNA and protein must be present in the insulinoma cell line and, in particular, in pancreatic islets. Clearly, cDNA fragments of the predicted size of 497 bp were amplified using RNA from insulinoma cells (Fig. 6A) and pancreatic islets (Fig. 6B) by RT-PCR. To exclude cross-reactivities of the primers with cDNA encoding related POU domain transcription factors, the DNA sequence of the amplified fragment was determined and found to be identical to the published Tst-1 cDNA sequence (data not shown). In Western blot experiments the anti-Tst-1 antiserum detected a major protein band of approximately 50 kDa in nuclear extracts of RIN 1046-38 cells (Fig. 6C). A band of identical size was also detected in nuclear extracts of CV1 cells previously transfected with the Tst-1 expression plasmid pCMVTst-1. The observed apparent molecular weight of Tst-1 is in good agreement with that seen in other tissues (35, 36). To extend these observations further, immunohistochemistry was used to localize Tst-1 in islets of adult rat pancreas. All islets were stained with the anti-Tst-1 antiserum (data not shown), and within an islet, all or almost all cells were immunoreactive (Fig. 6D). However, no immunoreactivity was seen in the exocrine pancreas and in control sections (Fig. 6, D and E). These results show unequivocally that, in addition to its presence in the insulinoma cell line, Tst-1 is also expressed in pancreatic islets.

View larger version (60K): [in this window] [in a new window]   Fig. 6.   Presence of Tst-1 in RIN 1046-38 cells and pancreatic islets. A and B, RNA from RIN 1046-38 cells and pancreatic islets and, as positive control, from rat brain was analyzed for the presence of Tst-1 sequences by RT-PCR amplification and agarose gel electrophoresis. The expected 497-bp fragment is indicated by arrows. In the absence of the reverse transcriptase no DNA fragment could be amplified suggesting that the isolated RNA was not contaminated with genomic DNA. M, 100-bp ladder. C, nuclear extracts from RIN 1046-38 cells or CV1 cells transfected with CMVTst-1 (CV1/Tst-1) or CMV (CV1/Gal) were run on an SDS-15% polyacrylamide gel, transferred to a nylon membrane, and incubated with an antiserum directed against Tst-1 and an alkaline phosphatase-conjugated anti-rabbit IgG antibody. The position and molecular weight of standard proteins are indicated on the left. Arrow, 50-kDa protein detected with the anti Tst-1 antiserum. D, immunohistochemical detection of Tst-1 in rat pancreas. E, for control, a rabbit antiserum directed against Pit-1 that is not present in the pancreas was employed.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The fp2b promoter element, located between 158 and 136, plays a dominant role for transcriptional regulation of the sst1 gene. In pituitary GH3 cells the POU domain protein Pit-1 trans-activates the sst1 gene by binding to this site in vitro and in vivo (27). Here, the importance of the fp2b element is emphasized by our findings that it also confers transcriptional activity to the sst1 gene in insulin-producing pancreatic cells. First, deletion of a 48-bp region containing the fp2b element abolished most of the promoter activity in transient transfection experiments. Second, the fp2b element served as an enhancer in RIN 1046-38 cells when fused to a heterologous promoter. Third, mutants of the fp2b region diminished transcription. In pancreatic RIN 1046-38 cells a transcription factor other than Pit-1 must be responsible for the fp2b-dependent activation of sst1 gene transcription since Pit-1 is restricted to pituitary cells. Our data identify the POU domain protein Tst-1 as the activator. Tst-1 binds the fp2b DNA and trans-activates the sst1 gene promoter dose-dependently in heterologous CV1 cells previously transfected with a Tst-1 expression plasmid. Mutations within the fp2b DNA prevented binding of Tst-1, strongly reduced promoter activities in RIN 1046-38 cells, and blocked trans-activation by Tst-1 in CV1 cells completely. Therefore, whereas Pit-1 is necessary for sst1 gene transcription in pituitary cells, another POU domain protein, Tst-1, activates the sst1 gene in the insulinoma cell line. The fp2b region might represent a promoter element that confers cell-type specificity when bound by appropriate POU domain proteins.

Within the fp2b region the exact binding site of Tst-1 has been determined by mutagenesis analysis to be the 5'-TTAATCATT-3' motif located between 151 and 143. This A/T-rich sequence represents a typical bipartite binding site of POU domain proteins that are characterized by two conserved domains, the POU-specific domain and the POU homeodomain (23). The POU-specific domain contacts the 5'-CAT-3' motif and the POU homeodomain the A/T-rich element (37-39). There are differences between the binding specificities among the POU proteins of different subclasses. Class I and II proteins (Pit-1, Oct-1 and Oct-2) prefer the orientation 5'-TAATNCAT-3', whereas binding sites of POU-III and -IV proteins (Brn-1, Brn-2, Brn-4, Tst-1, Brn-3.0, Brn-3.1, and Brn-3.2) mostly represent the opposite orientation 5'-CATNTAAT-3' (40). However, class III POU proteins exhibit remarkable flexibility in DNA site recognition (40). It has been shown that Tst-1 tolerates linkers of 0, 2, or 3 nucleotides between the two motifs and binding sites with both orientations (40-43). This explains the ability of Tst-1 to bind to the element 5'-TTAATCATT-3' within the sst1 gene promoter.

Additional RIN 1046-38 cell nuclear proteins, represented as minor bands in the EMSA analyses, bind to the fp2b region. Analysis of protein binding to the mutant fp2b sequences indicated that all fp2b-binding proteins show the same sequence requirements. CV1 cells that were co-transfected with sst1 promoter/reporter gene constructs and a Tst-1 expression plasmid showed lower promoter activity than RIN 1046-38 cells. It is therefore concluded that the additional fp2b-binding proteins may be necessary for full promoter activity. One may be the ubiquitous POU protein Oct-1 that was shown by EMSA to bind to the fp2b element. In co-transfection experiments of an Oct-1 expression plasmid with p-324sst1Luc, we observed a dose-dependent increase in luciferase activity (results not shown). The increase was similar to that seen with the Tst-1 expression plasmid, and the effects of both were additive. Therefore we conclude that Oct-1 can contribute to the regulation of the sst1 gene. However, since Oct-1 is ubiquitously expressed including endocrine pancreatic cells (25, 29), its presence alone is not sufficient to explain the tissue-specific expression of the sst1 gene. In the pituitary and endocrine pancreas sst1 gene expression depends on Pit-1 and Tst-1. Additional regulators may be represented by the unidentified fp2b DNA-binding proteins.

Tst-1 transcripts have been detected in testis, distinct areas of the brain, skin, and myelinating Schwann cells of the peripheral nervous system (36, 44-47). In an attempt to identify POU domain transcription factors that may be involved in the process of self-renewal of the epithelium, the expression of Tst-1 has been observed in the gastrointestinal tract (48). Yet the low level of Tst-1 expression detected by S1 nuclease protection assays was suggested to reflect the presence of Tst-1 in the Schwann cells of the peripheral nervous system. A hint for a pancreatic expression of Tst-1 was given by Hussain et al. (25) who detected Tst-1 transcripts in the mouse insulinoma TC-1 cell line. However, a functional role for Tst-1 has not been determined. In the present study, RT-PCR, EMSA analysis, Western blotting, and immunohistochemistry demonstrated that Tst-1 is present in clonal insulinoma cells and pancreatic islets. Since Tst-1 is present in almost all islet cells, and since about 80% of rat islet cells represent insulin-producing -cells, it is concluded that Tst-1 must be present in -cells.

POU domain proteins are well known as regulators of development of distinct tissues and cell types (39). Tst-1 appears to be an important regulator of glial (46), epidermal (44), and neuronal differentiation (26). Other POU domain proteins, i.e. Pit-1, Brn-2, and Brn-4, represent essential regulators of neuroendocrine function in the pituitary, hypothalamus, and pancreatic -cells, respectively (24-26). Here it is shown for the first time that Tst-1 is also important for transcriptional regulation in a neuroendocrine cell type, i.e. the activation of the sst1 gene. Tst-1 probably also activates a number of other genes in the endocrine pancreas. This assumption is supported by the observation that sequences homologous to the Tst-1 consensus sequence 5'-GAWTWANA-3' (35) are present within the rat insulin I (552 to 559 (49)) and II gene (535 to 541 (49)), the glucagon gene (525 to 532 and 652 to 659 (50)), and the glucagon receptor gene (134 to 150 (51)).

Tst-1 gene expression is induced upon exposure of cells to forskolin, a compound that increases the intracellular cAMP concentration (30, 52). This observation may be of importance for pancreatic -cells since increased cAMP levels mediate the effect of insulin secretagogues, such as the glucagon-like peptide-1 (53). It is tempting to speculate that the Tst-1 gene is induced by cAMP in -cells. In turn it could activate the sst1 gene and thereby up-regulate the number of sst1 receptors on the plasma membrane. As a result, -cells would respond to somatostatin in a more pronounced way to limit the release of insulin. Analysis of the expression profile of Tst-1 in pancreatic islets and identification of additional target genes of Tst-1 are likely to improve our understanding of the function of Tst-1 in pancreatic islets.

	ACKNOWLEDGEMENTS

We thank Dr. T. Hübschle for help with immunohistochemistry, Dr. M. Strowski (Rahway) for help with the preparation of pancreatic islets, and Dr. M. Wegner (Hamburg) for the generous gift of the expression plasmids CMVTst-1 and CMVOct-1 and of antisera against Brn-1, Brn-2, and Tst-1.

	FOOTNOTES

^* This work was supported by the Fonds der Chemischen Industrie (to W. M.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Abteilung Molekulare Genetik, Deutsches Institut für Ernährungsforschung, Arthur-Scheunert-Allee 114-116, D-14558 Potsdam-Rehbrücke, Germany. Tel.: 49 33200 88 282; Fax: 49 33200 88 444; E-mail: meyerhof@www.dife.de.

Published, JBC Papers in Press, June 23, 2000, DOI 10.1074/jbc.M002175200

	ABBREVIATIONS

The abbreviations used are: RT-PCR, reverse transcriptase-polymerase chain reaction; CHO, Chinese hamster ovary; EMSA, electrophoretic mobility shift assays; bp, base pair.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				W. W. Woodmansee, R. L. Mouser, D. F. Gordon, J. M. Dowding, W. M. Wood, and E. C. Ridgway Mutational Analysis of the Mouse Somatostatin Receptor Type 5 Gene Promoter Endocrinology, June 1, 2002; 143(6): 2268 - 2276. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/37/28882    most recent M002175200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Baumeister, H. Articles by Meyerhof, W. Search for Related Content PubMed PubMed Citation Articles by Baumeister, H. Articles by Meyerhof, W. Social Bookmarking                      What's this?