Regulation of ATP-sensitive potassium channel subunit Kir6.2 expression in rat intestinal insulin-producing progenitor cells.

We have reported that the combined expression of Pdx-1 (pancreatic duodenal homeobox 1) and Isl-1 (islet 1) enables immature rat enterocytes (IEC-6) to produce and release insulin. A key component regulating the release of insulin is the ATP-sensitive potassium channel subunit Kir6.2. To investigate the regulation of Kir6.2 gene expression, we assessed Kir6.2 expression in IEC-6 cells expressing Pdx-1 and/or Isl-1. We observed that Kir6.2 protein was expressed de novo in IEC-6 cells expressing both Pdx-1 and Isl-1 but not in cells expressing Pdx-1 alone. Next, we analyzed the regions of the Kir6.2 promoter (-1677/-45) by performing a luciferase assay and electrophoretic mobility shift assay. The results have demonstrated that Kir6.2 promoter possesses two regions regulating the promoter activity: a Foxa2-binding site (-1364 to -1210) and an Sp1/Sp3-binding site (-1035 to -939). The additional expression of Isl-1 in IEC-6 cells expressing Pdx-1 attenuated overexpression of Foxa2 protein and enhanced Kir6.2 expression. Finally, knockdown of Isl-1 using the iRNA technique resulted in decreased expression of Kir6.2 protein in a rat pancreatic beta-cell line (RIN-5F cells). These results indicate that expression of Kir6.2 in the rat intestine is moderated by Isl-1.

A key component regulating the release of insulin is Kir6.2 (11,12), which is expressed at particularly high levels in pancreatic ␤-cells (13). Briefly, an increase in ATP and a decrease in ADP stimulated by glucose metabolism depolarize the ␤-cells by closing Kir6.2. Membrane depolarization results in the opening of voltage-dependent Ca 2ϩ channels and the influx of Ca 2ϩ is the main trigger for insulin secretion (14,15). Thus, insulin secretion in pancreatic ␤-cells is controlled through metabolic regulation of the electrical activity by Kir6.2 channels that control membrane potential. In this study, we observed no gene expression of Kir6.2 in IEC-6 cells transfected with Pdx-1 gene alone. However, additional expression of Isl-1 in IEC-6 cells expressing Pdx-1 induced Kir6.2 expression. These results indicate that Isl-1 plays an important role in expression of Kir6.2.
The Kir6.2 promoter contains several A-box motifs and forkhead/winged helix binding motifs, which are functionally important. Whereas both Pdx-1 and Isl-1 bind to TAAT sites in A-box, forkhead/winged helix proteins, which include the Foxa (forkhead box A) family of transcription factors encoded by the three genes Foxa1 (Hnf3␣), Foxa2 (Hnf3␤), and Foxa3 (Hnf3␥), bind to the core sequence (AAATA) and regulate hepatic and/or pancreatic gene expression (16 -22). Thus, we hypothesized that these transcription factors might induce Kir6.2 expression. To prove our hypothesis, we used cell lines that are parental IEC-6 cells believed to be intestinal progenitor cells differentiating into insulin-producing cells (10) and RIN-5F cells as a pancreatic ␤-cell line. Through making a comparison between these cell lines, we discovered the transcriptional regulation of Kir6.2 gene expression by Isl-1 in these cells.
Preparation of Recombinant Adenoviruses-cDNAs encoding for wild type mouse Isl-1 were subcloned into Adeno-X viral DNA vector (BD Biosciences Clontech) and cotransfected into 293 cells. Successful homologous recombination resulted in recombinant viruses encoding Isl-1 (Ad-Isl-1) and the control (Ad-LacZ). Viruses were amplified in 293 cells and used with centrifugation to avoid cell pellet formation. Plaqueforming units (pfu) were assayed using the Adeno-X rapid titer kit (BD Biosciences Clontech).
Isl-1 RNA Interference Preparation-The double-stranded RNA nucleotides for Isl-1 (Isl-1-iRNA) were obtained from iGENE Therapeutics (Tsukuba, Japan) as follows: Isl-1-iRNA-1, sense (5Ј-GCAACUGGUC-AAUUUUUCAGAAGGA-AG-3Ј) and antisense (5Ј-UCCUUCUGAAAA-* This work was supported, in part, by Grant-in-aid for Scientific Research (C) 16590143 from the Ministry of Education, Science and Culture of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Transfection was performed as described elsewhere (23). In brief, RIN-5F cells were transfected with 50 nM Isl-1-iRNAs using the Tran-sIT-TKO Transfection Reagent (Mirus, Madison, WI). After a 48-h incubation, the cells were washed two times with phosphate-buffered saline and subjected to molecular biological analysis.
The thermal cycle profile was as follows: A single 1-min denaturing step at 94°C was followed by 35 cycles of 30 s at 94°C, 30 s at 55°C, and 1 min at 72°C. The identities of the PCR products were confirmed by agarose gel electrophoresis and nucleotide sequence analysis.
Western Blot Analysis-Relative amounts of Isl-1 and Kir6.2 proteins in the cells were assessed using Western blot analysis, as described elsewhere (23). Specifically, 20 g of cell lysate was separated by standard SDS-PAGE and then transferred to a polyvinylidine difluoride membrane. Specific antibodies against Isl-1 (25), Foxa2 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA), and Kir6.2 (Chemicom International, Temecula, CA) were used to probe the blot. Specific antibody against Nucleoporin P62 (BD Biosciences) was also used as a control. The antibody-protein complex was detected using an enhanced chemiluminescence kit (PerkinElmer Life Sciences).
Nuclear Extracts and Electrophoretic Mobility Shift Assay (EMSA)-After preparation of nuclear extracts (23), we performed EMSA (24). In brief, the double-stranded probes were end-labeled with [␥-32 P]ATP (Amersham Biosciences) using T4 kinase (TAKARA, Shiga, Japan). One l of radiolabeled probe (activity 50,000 -100,000 cpm/l) was added to the mixture, and the samples were incubated for 20 min at room temperature. Three g of nuclear protein was used for each reaction. For the cold competition experiments, a 50-fold molar excess of unlabeled oligonucleotides was added to the reaction mixture. DNA-protein complexes were separated from unbound probe on native 4 -6% polyacrylamide gels in 0.25ϫ TBE (45 mM Tris borate, 1 mM EDTA) at 150 V for 1.5 h. The gels were transferred to Whatman MM filter paper, dried at 80°C for 2 h, and exposed to film (Eastman Kodak Co.) for 12-15 h.
The cells were seeded onto 12-well cell culture plates. After 24 h of incubation, the cells were cotransfected with 0.6 g of pGL3-Basic or pGL3-Kir6.2 plasmid, 0.2 g of pRL-TK plasmid (Promega) as normalization reference for transfection efficiency and/or the indicated amount of Sp1, Sp3, or Foxa2 expression vector using LipofectAmine 2000 (Invitrogen) following the instructions for the reagent. After 48 h of transfection, the cells were harvested. The firefly and Renilla luciferase activities were determined by using a dual luciferase reporter assay kit (Promega) with a signal detection duration of 30 s by a luminometer (Auto LUMIcounter Nu1422ES; Microtec, Tokyo, Japan). Site-directed mutagenesis was performed on the putative GC-box sites in the distal enhancer region of the Kir6.2-1.6-kb plasmid using the QuikChange site-directed mutagenesis kit (Stratagene, Inc., La Jolla, CA) according to the manufacturer's instructions. The mutated oligonucleotides are as follows: 5Ј-GCGGGAC-GAGAATCCTCTGcaaatGGGGCTTGCGTCTAGGCCC-3Ј (lowercase letters represent mutated nucleotides).
Statistical Analysis-Results are expressed as mean Ϯ S.E., unless otherwise stated. Scheffe's multiple comparison test was used to determine the significance of any differences among three or more groups, and the unpaired Student's t test was used to determine the significance of any differences between two groups. p Ͻ 0.05 was considered significance.

Isl-1 Induces Expression of Kir6.2-
We have already reported that immature intestinal stem cells can differentiate into insulin-producing cells upon expression of the transcription factors Pdx-1 and Isl-1 (10). The process of differentiation leading to the formation of pancreatic ␤-cells requires activation of specific genes for insulin secretion. We examined whether expression of these genes is observed in IEC-6 cells expressing Pdx-1 and/or Isl-1 by RT-PCR. IEC-6 cells expressing both Pdx-1 and Isl-1 (IEC-IP cells) clearly expressed de novo mRNAs of both Kir6.2 and Glut2 (Fig. 1A), whereas Gck was faintly expressed in these cells. mRNA expression of sulfonylurea receptor 1 (SUR1) was much stronger in these cells than in IEC cells expressing Pdx-1 alone (IEC-Pd) (Fig. 1A). RT-PCR product was verified as corresponding to each mRNA by nucleotide sequencing. To confirm expression of Kir6.2 protein, Western blot analysis was performed. As in RIN-5F cells, we observed a 45-kDa band of Kir6.2 (26) in IEC-IP cells but not in other cells (Fig. 1B). Next, to study the effect of Isl-1 on the expression of the Kir6.2 gene, IEC-Pd cells were transduced with Ad-Isl-1 (12.5 ϫ 10 8 pfu or 25 ϫ 10 8 pfu/ml), and expression of Kir6.2 was studied after 7-day infection. Consistent with our previous results (10), Isl-1 (43-kDa molecular mass) was identified. In this experiment, de novo expression of the Kir6.2 gene was found, whereas no signals were identified with Ad-LacZ (Fig. 1C). Identical results were observed in IEC-6 cells transduced with Ad-Isl-1 (data not shown). These studies clearly show that additional expression of Isl-1 in IEC-6 and IEC-IP cells is associated with de novo production of Kir6.2 protein. To investigate the effects of endogenous Isl-1 on expression of Kir6.2, we knocked down Isl-1 expression in RIN-5F cells, using a 25-nucleotide RNA duplex (Isl-1-iRNAs), and analyzed protein expression of Kir6.2. These cells were treated with Isl-1-iRNAs for 2 days, and whole-cell lysate was analyzed by Western blot analysis using both anti-Isl-1 and anti-Kir6.2 antibodies. As shown in Fig. 1D, Isl-1-iRNAs reduced Isl-1 expression to 30% of the control, and the expression of Kir6.2 protein content was also reduced to at least 50% (Fig. 1D, lanes   2 and 3). In contrast, no change was observed in nucleoporin P62 in the presence of Isl-1-iRNAs (Fig. 1D, lane 1). The decreased expression of Isl-1 thus causes a reduction of Kir6.2 protein content.
Luciferase Reporter Assay Using Kir6.2 Promoter-We cloned the 5Ј-flanking region of the mouse Kir6.2 gene (Ϫ1677/ Ϫ43) and fused it to luciferase reporter gene to investigate the mechanism of the new expression of Kir6.2 in IEC-IP cells. First, we compared the promoter activity of IEC-Pd and IEC-IP cells. When we analyzed the promoter activities, the basal luciferase activities were adjusted by measurement of the activity obtained using pGL3-Basic vector alone in each cell line.
Upon analysis of the regulatory regions of Kir6.2 promoter activity, the luciferase activity was significantly increased with a truncation from Ϫ1364 to Ϫ1210 in IEC-Pd cells, whereas no significant change was observed in IEC-IP cells ( Fig. 2A). We also identified another regulatory element of Kir6.2 promoter, from Ϫ1035 to Ϫ939, in both IEC-Pd and IEC-IP cells, since promoter activities were significantly decreased with the truncation from Ϫ1035 to Ϫ939 in these cells. Thus, we found that Kir6.2 promoter contains two regions that regulate Kir6.2 promoter activity, one extending from Ϫ1364 to Ϫ1210 and the other from Ϫ1035 to Ϫ939 (Fig. 2A).
To confirm whether identical findings would be observed in pancreatic ␤-cells, we measured the promoter activity in RIN-5F cells, one of the rat pancreatic ␤-cell lines. As expected, the promoter activities in RIN-5F cells were similar to that in IEC-IP cells (Fig. 2B). Taken together, these results suggest that the 5Ј-flanking region of the mouse Kir6.2 gene (Ϫ1364/ Ϫ1210) may work as a suppressive element in the absence of  To test specificity, lanes 2, 4, 6, 8, and 10 include a 50-fold molar excess of each oligonucleotide. C, gel shift competition analysis using fragment 4 wild type (4-WT) and its mutants was performed. Radiolabeled probe spanning the Ϫ985/Ϫ956 region of the Kir6.2 promoter was incubated with 3 g of nuclear extract of RIN-5F. Unlabeled mutant probes (m1, m2, and m3) were used as competitors in a 50-fold molar excess. D, nuclear extracts of RIN-5F were incubated with 2 g of polyclonal Sp1, Sp3, and Foxa2 antibodies before the addition of the fragment 4 wild type radiolabeled probes.

Isl-1 Induces Kir6.2 Gene Expression
Isl-1 expression. On the other hand, unknown transcription factor(s) that functions as an activator(s) may bind to the region (Ϫ1035 to Ϫ939) of the Kir6.2 promoter.
Identification of Nuclear Transcription Factors Binding to the Kir6.2 Gene Promoter GC-box (Ϫ985 to Ϫ956)-In order to identify which transcription factor(s) binds to the promoter region from Ϫ1035 to Ϫ939 of Kir6.2 promoter, EMSA was performed. The Kir6.2 promoter region from Ϫ1035 to Ϫ939 was divided into five fragments (Fig. 3A), and EMSA was performed using nuclear extract from RIN-5F cells. Two specific DNA-protein complexes (Fig. 3B, lane 7) were identified using oligonucleotides from Ϫ985 to Ϫ956 (probe 4). Corresponding cold oligonucleotides completely abrogated this complex formation (Fig. 3B, lane 8). Next, to confirm whether nuclear proteins from RIN-5F cells specifically bind to the Kir6.2 promoter from Ϫ985 to Ϫ956, we constructed three mutations (mut1, mut2, and mut3) of the core sequences of GC-box (Ϫ985 to Ϫ956) as a competitor (Fig. 3C). [␥-32 P]ATPlabeled oligonucleotides of the Kir6.2 promoter from Ϫ985 to Ϫ956 formed no DNA-protein complexes in the presence of wild type, mut1, or mut2 oligonucleotide as a competitor (Fig. 3C,  lanes 3 and 4), whereas mut3 oligonucleotide was ineffective for use as a competitor (Fig. 3C, lane 5). Since this sequence contains some regions recognized by Maf-A, Pax4, and Pax6, which are all key transcription factors in the development pancreatic ␤-cells (4, 27), we performed EMSA with anti-MafA, Pax4, and Pax6 antibodies. However, we failed to find any interaction (data not shown). It is well known that Sp family members contain conserved zinc finger DNA binding domains and bind to the GC-box of several genes (28). Next, we performed a supershift assay using anti-Sp1 and anti-Sp3 anti-bodies (Fig. 3D). In the supershift assay, the addition of anti-Sp1 antibody reduced the intensity of the Sp1 band and supershifted its complex (Fig. 3D, lane 3). When anti-Sp3 antibody was used, the Sp3 band completely disappeared, and its complex supershifted (Fig. 3D, lane 4). On the other hand, the addition of anti-Foxa2 antibody had no effect (Fig. 3D, lane 5). Identical results were obtained using nuclear extracts from IEC-6, IEC-Pd, and IEC-IP cells (data not shown).
To study the effects of Sp1 and Sp3 on the promoter activity of Kir6.2, we overexpressed each of these proteins in RIN-5F cells and measured the luciferase activities (Fig. 4A). We found that the promoter activity paralleled the expression level of Sp1, whereas overexpression of Sp3 decreased this activity in RIN-5F cells. Thus, Sp1 increases the promoter activities of Kir6.2, and Sp3 decreases it. Next, we measured luciferase activity using reporter vectors containing mutated regions of Kir6.2 promoter (Fig. 4B). In both IEC-IP and RIN-5F cells, the luciferase activity was 50% compared with the wild type promoter when investigated using mutant promoter. Taken together, these results demonstrate that both Sp1 and Sp3 bind to Kir6.2 promoter from Ϫ985 to Ϫ956 and regulate its activity.
Identification of Nuclear Transcription Factors Binding to the Kir6.2 Gene Promoter (Ϫ1364 to Ϫ1210)-Since both Sp1 and Sp3 are ubiquitously expressed in IEC-Pd, IEC-IP, and RIN-5F cells, there were no differences in Kir6.2 promoter activity (Ϫ985 to Ϫ956) among these cells (data not shown). Therefore, this element is not a region specifically associated with the effect of Isl-1. Thus, we hypothesized that the promoter region (Ϫ1364 to Ϫ1210) of the Kir6.2 gene is a crucial regulatory element affected by Isl-1 expression. To test this hypothesis, the Kir6.2 promoter (Ϫ1364 to Ϫ1210) was divided
Effects of Foxa2 on Kir6.2 Promoter (Ϫ1364 to Ϫ1210)-We demonstrated that Kir6.2 promoter activity does not change with the truncation of the promoter (Ϫ1364/Ϫ1210) in cells expressing Isl-1, such as IEC-IP and RIN-5F cells. Indeed, the activity increases with this truncation in IEC-Pd cells. To investigate its effect on Kir6.2 promoter activity, we measured the effect of Foxa2 on luciferase activity in RIN-5F cells (Fig. 6). We transfected RIN-5F cells with pGL3-Basic or pGL3-Kir6.2 reporter construct along with Foxa2 expression vector. We found that Foxa2 increased Kir6.2 promoter activity in a dosedependent manner, in accordance with a previous report (29). However, Foxa2 similarly increased the basic promoter activity in a dose-dependent manner. Thus, to adjust the effect of Foxa2 on Kir6.2 promoter activity, we calculated the ratio between the luciferase activities of Kir6.2 promoter and promoterless reporter construct in the presence of the same dose of Foxa2 expression vector (Fig. 6A). The relative luciferase activity reciprocally decreased, to 50, 32, and 28% of basal level, in the presence of 0.4, 0.8, and 1.6 g of Foxa2 vector, respectively.
Foxa2 Protein Levels in the Presence of Isl-1-How does Isl-1 influence the expression of Foxa2? To study this issue, we measured the amount of Foxa2 protein by Western blot analysis in IEC-6, IEC-Pd, and IEC-IP cells. Our data showed that  To test specificity, lanes 2, 4, 6, 8, and 10 include a 50-fold molar excess of each oligonucleotide. C, 3 g of nuclear extract of IEC-Pd cells was incubated with 2 g of polyclonal Foxa2 antibodies before the addition of the indicated radiolabeled probes. the amount of Foxa2 protein was lower in IEC-IP cells than in parent IEC-6 and IEC-Pd cells (Fig. 6B). Next, to identify the effect of Isl-1 on Foxa2 protein level, we transduced IEC-Pd cells with the Ad-Isl-1 virus. The cells were transduced with 12.5, 25, or 50 ϫ 10 8 pfu/ml of Ad-Isl-1 and harvested after 48 h of infection. The amount of Foxa2 protein was decreased to 38, 23, and 18% of basal level, respectively (Fig. 6C). These results indicate that Isl-1 attenuates overexpression of Foxa2 in intestinal progenitor cells.

DISCUSSION
Differentiation of pancreatic cells from progenitor cells requires the actions of both intrinsic and extrinsic influences, such as transcription and growth factors. In this study, we used the cell line IEC-6, derived from rat immature intestinal crypt cells (9), to test whether these cells can be induced to express Kir6.2. Our previous studies showed that when IEC-6 cells acquire the ability to express the transcription factor Pdx-1, they undergo differentiation into entero-endocrine cells (31). In IEC-Pd cells, NeuroD, Pax6, and Nkx6.1 were expressed but not Isl-1. After transfection with Isl-1 to IEC-6 cells expressing Pdx-1 (IEC-Pd cells), insulin release into the culture medium was observed. These previous data prompted us to investigate whether the specific regulatory factors for insulin secretion, such as Gck, Glut2, SUR1, and Kir6.2, would also be observed in IEC-IP cells. We detected the de novo expression of mRNAs of Kir6.2, SUR1, and Glut2. These results suggest that the additional expression of Isl-1 enables IEC-Pd cells to express Kir6.2, SUR1, and Glut2 through the regulation of their promoter activities.
In this study, focusing on the regulation of Kir6.2 gene expression, we found that additional expression of Isl-1 enabled de novo Kir6.2 gene expression through attenuating Foxa2 in IEC cells overexpressing Pdx-1. Both Pdx-1 and Isl-1 are homeobox proteins, which bind TAAT sites in the A-box, whereas Isl-1 is also a class of LIM/homeodomain transcription factors with important roles in determining cell lineage and pattern formation during differentiation to pancreas or brain cells (32). Since the LIM domain has been demonstrated to be a proteinprotein interaction motif that is crucially involved in these processes, we suspect that the LIM domain of Isl-1 may regulate Foxa2 gene expression in both IEC-IP and RIN-5F cells.
Foxa2 encodes a transcription factor (forkhead box A2) that has been postulated to play a central role in ␤-cell development due to its ability to bind to the Kir6.2 promoter (29). Lantz et al. (29) demonstrated that Foxa2 deficiency resulted in excessive insulin release in response to amino acids and complete loss of glucose-stimulated insulin secretion. They also showed that co-transfection of Foxa2 expression vector resulted in stimulation of luciferase activity from the Kir6.2 promoter construct containing a 1.73-kb promoter sequence (Ϫ551/ϩ1186) and that Foxa2 acted as a transcriptional activator of Kir6.2 in ␤-cells. However, our results revealed that Foxa2 can serve as a suppressor in the Kir6.2 promoter from Ϫ1324 to Ϫ1149 (Fig.  3). To assess this discrepancy, we transfected pancreatic ␤-cells (RIN-5F cells) with Kir6.2 promoter/luciferase (pGL3-Kir6.2) or reporter constructs without the promoter (pGL3-Basic) along with a Foxa2 expression plasmid (Fig. 6A). Interestingly, more than 15-fold activation was observed with the transfection of both pGL3-Basic and Foxa2 vector (1.6 g/plate), whereas only 7-fold activation was observed with the transfection of both pGL3-Kir6.2 and Foxa2. No significant activity was observed in these cells transfected with both pGL3-Basic and Pdx-1 expression vector (data not shown) (i.e. Foxa2 enabled pGL3-Basic vector to activate luciferase activity by itself). Thus, the difference between the luciferase activities of pGL3-Basic and pGL3-Kir6.2 revealed that Foxa2 overexpression suppresses Kir6.2 luciferase activity. Lantz et al. (29) did not report the luciferase activity of promoterless vector with Foxa2 expression. Another possibility is that the discrepancy may arise from the nature of the cells studied. We used intestinal cells, such as IEC-6, IEC-Pd, and IEC-IP cells, which express intrinsic Foxa2 (Fig. 6B), whereas Lantz et al. transfected baby hamster kidney cells, which might not express intrinsic Foxa2, with Foxa2 expression vector; thus, the transfection with Foxa2 might have enabled induction of Kir6.2 gene activity in these cells. We need to study this discrepancy in more detail. Nevertheless, it is known that mRNA levels of both SUR1 and Kir6.2 are reduced by ϳ75% in Foxa2 knockout mice (30). Furthermore, the requirement of Foxa2 for maintenance of Kir6.2 expression has already been confirmed by Northern blot analysis in insulinoma-1 cells overexpressing a dominant negative mutant Foxa2 (33). These results indicate that Foxa2 is an essential transcriptional regulator of Kir6.2 gene. Although the expression of Foxa2 in intestinal cells (IEC-IP cells) newly expressing Kir6.2 was lower than that in IEC-Pd or IEC-6 cells (Fig. 6B), this expression may be sufficient to enhance Kir6.2 gene expression in intestinal cells. Moreover, we showed that IEC-Pd cells transduced with Ad-Isl-1 for 7 days newly expressed Kir6.2 protein by Western blot analysis (Fig. 2). At the same time, the overexpression of Isl-1 significantly reduced Foxa2 protein level (Fig. 6C), suggesting that weak Foxa2 expression may be enough to regulate Kir6.2 gene expression.
Kir6.2 promoter does not contain any putative TATA or CCAAT box, but it has a GC-box that is a putative binding site for the Sp family of transcription factors (34,35). Sp1 is a sequence-specific ubiquitously expressed zinc finger transcription factor that supports constitutive basal expression of a variety of eukaryotic genes that lack a functional TATA box. The Sp family is one of the most studied transcription factors. Five other, Sp1-related proteins, Sp2-Sp6, have been described (36). Consistent with the previous data (37), EMSA showed that full-length and truncated forms of Sp3 and Sp1 can interact specifically with the GC-box in the Kir6.2 promoter. Sp3 slightly suppresses Kir6.2 promoter activity, whereas Sp1 stimulates that activity in a dose-dependent manner. Additionally, we also confirmed that there was no difference in Sp1 and Sp3 concentrations among these cells, suggesting that Isl-1 did not influence Sp1 and Sp3 expression (data not shown). Our results suggest that Sp1 is an important mediator of the basal transcriptional activity of the Kir6.2 gene. This is the first demonstration that the expression of Isl-1 activates Kir6.2 gene expression in rat intestinal progenitor cell lines. Our data demonstrate that the expression of Pdx-1 is not sufficient for activation of Kir6.2 gene expression in these cells, although it may activate insulin, Gck, and pancreatic polypeptide promoter (38). The gut constitutes a potentially accessible target for oral delivery of vectors, and several studies have demonstrated the utility of targeting the gut for gene therapy in vivo (39,40). To our knowledge, Isl-1 is the first transcription factor confirmed to enhance Kir6.2 gene expression in vitro. Future studies should be directed at examining the utility of adenovirus-delivered Pdx-1 and Isl-1 targeted to the gut epithelium to enhance production of both insulin and Kir6.2 for potential application in new therapies for diabetes.