help button home button Endocrine Society Molecular Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Endocrinology, doi:10.1210/me.2005-0127
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
20/1/167    most recent
Author Manuscript (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Volinic, J. L.
Right arrow Articles by Thomas, M. K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Volinic, J. L.
Right arrow Articles by Thomas, M. K.
Molecular Endocrinology 20 (1): 167-182
Copyright © 2006 by The Endocrine Society

Overexpression of the Coactivator Bridge-1 Results in Insulin Deficiency and Diabetes

Jamie L. Volinic, Jee H. Lee, Kazuhiro Eto, Varinderpal Kaur and Melissa K. Thomas

Laboratory of Molecular Endocrinology and Diabetes Unit, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114

Address all correspondence and requests for reprints to: Melissa K. Thomas, M.D., Ph.D., Laboratory of Molecular Endocrinology and Diabetes Unit, Massachusetts General Hospital, Thier 340, 50 Blossom Street, Boston, Massachusetts 02114. E-mail: mthomas1{at}partners.org.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Multiple forms of heritable diabetes are associated with mutations in transcription factors that regulate insulin gene transcription and the development and maintenance of pancreatic ß-cell mass. The coactivator Bridge-1 (PSMD9) regulates the transcriptional activation of glucose-responsive enhancers in the insulin gene in a dose-dependent manner via PDZ domain-mediated interactions with E2A transcription factors. Here we report that the pancreatic overexpression of Bridge-1 in transgenic mice reduces insulin gene expression and results in insulin deficiency and severe diabetes. Dysregulation of Bridge-1 signaling increases pancreatic apoptosis with a reduction in the number of insulin-expressing pancreatic ß-cells and an expansion of the complement of glucagon-expressing pancreatic {alpha}-cells in pancreatic islets. Increased expression of Bridge-1 alters pancreatic islet, acinar, and ductal architecture and disrupts the boundaries between endocrine and exocrine cellular compartments in young adult but not neonatal mice, suggesting that signals transduced through this coactivator may influence postnatal pancreatic islet morphogenesis. Signals mediated through the coactivator Bridge-1 may regulate both glucose homeostasis and pancreatic ß-cell survival. We propose that coactivator dysfunction in pancreatic ß-cells can limit insulin production and contribute to the pathogenesis of diabetes.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
DIABETES MELLITUS results from an absolute or relative deficiency of the hormone insulin. Although absolute deficits in the numbers of insulin-expressing pancreatic ß-cells are seen in type 1 diabetes, type 2 diabetes often is accompanied by relative deficits in pancreatic ß-cell mass and insulin production in the setting of insulin resistance (1, 2, 3). Monogenic heritable forms of insulin-deficient, early-onset diabetes are associated with mutations in genes encoding factors essential for the transcriptional regulation of insulin production in pancreatic ß-cells (4). Five of the first six genes linked to maturity-onset diabetes of the young (MODY) in humans, hnf-4{alpha}/MODY1, hnf-1{alpha}/MODY3, ipf-1(pdx-1)/MODY4, hnf-1ß/MODY5, and neuroD1/MODY6, are transcription factors known to regulate endocrine cell development and/or the glucose-responsive expression of the insulin gene in pancreatic ß-cells. Thus, transcriptional regulators of pancreatic ß-cell mass and insulin production represent an important pool of candidate diabetes genes.

Glucose-responsive regulatory regions that are highly conserved in the insulin promoters of multiple species serve as convergence points for ß-cell transcription factors of the homeodomain and basic helix-loop-helix (bHLH) classes including several of the designated MODY genes (5, 6). The physiologic regulation of insulin gene expression is dependent on the precise assembly of transcription factors and coactivators within specific concentration ranges. Multiple protein-protein interactions generate functional regulatory complexes that couple with the coactivators cAMP response element binding protein-binding protein (CBP) or p300 in the transcriptional activation of the insulin gene (7, 8, 9, 10).

By screening for novel interaction partners for the bHLH transcription factor E12 in a clonal ß-cell line, we identified Bridge-1 as a PDZ domain coactivator of the insulin gene (11). Within the pancreas, Bridge-1 is expressed predominantly in the insulin-expressing ß-cells of the endocrine compartment (11). Bridge-1 contains an atypical PDZ protein interaction domain through which it interacts with E12 and E47 to coactivate glucose-responsive enhancers within the insulin promoter (11). Protein interactions between Bridge-1 and the homeodomain transcription factor and maturity-onset diabetes of the young (MODY4) gene product PDX-1 increase transcriptional activation by PDX-1 (12). Bridge-1 antisense constructs substantially reduce insulin promoter activation in insulin-producing cells in vitro (11), suggesting that endogenous Bridge-1 signaling regulates insulin production in pancreatic ß-cells.

To further investigate the function of the coactivator Bridge-1 in the regulation of glucose homeostasis in vivo, we generated a transgenic mouse model in which Bridge-1 is overexpressed in the pancreas. We envisioned that the overexpression of Bridge-1 might disrupt the regulation of Bridge-1 target genes. Here we report that increasing Bridge-1 expression represses insulin gene expression, increases pancreatic islet cell apoptosis, and reduces the mass of insulin-expressing ß-cells to result in severe, insulin-deficient diabetes.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Generation of Bridge-1 Transgenic Mice
To develop a transgenic mouse model of pancreatic Bridge-1 overexpression, we used a –4.6-kb segment of the mouse pdx-1 promoter that is known to confer expression in the developing pancreas and adult pancreatic ß-cells (13, 14) to express rat Bridge-1 cDNA in transgenic mice (Fig. 1AGo). We identified multiple independent founder lines by Southern blots of genomic DNA (Fig. 1BGo) and confirmed Bridge-1 mRNA overexpression by RT-PCR of pancreatic RNA derived from transgenic and control mice (Fig. 1CGo). The relative pancreatic expression of the Bridge-1 protein in male and female transgenic mice was estimated to range between 5- and 10-fold higher than control mice as assessed by Western blots of total pancreatic extracts (data not shown). No differences were observed in Bridge-1 protein expression levels on Western blots of liver extracts from transgenic or control mice. Bridge-1 protein expression patterns in the transgenic mice were heterogeneous as observed by immunostaining, with overexpression in the endocrine, exocrine, and ductal compartments of the pancreas (Fig. 1DGo). These protein expression patterns were consistent with those expected from the in vivo regulatory properties of the pdx-1 promoter. In mouse models in which the –4.6-kb pdx-1 promoter directed a ß-galactosidase reporter, the expression of ß-galactosidase was seen primarily in pancreatic islets and, at lower levels, in the pancreatic ducts and exocrine pancreas (13, 14).



View larger version (78K):
[in this window]
[in a new window]
  		Fig. 1. Transgenic Mouse Model for the Pancreatic Overexpression of Bridge-1

A, Schematic model of the transgene consisting of a 4.6-kb segment of the mouse pdx-1 promoter, a 900-bp segment of rat Bridge-1 cDNA coding sequence, and a 550-bp segment of rabbit ß-globin polyadenylation sequence. B, Bridge-1 transgene presence in genomic DNA. An autoradiogram from a Southern blot of genomic DNA derived from five distinct lines of Bridge-1 TG mice probed with a radiolabeled probe spanning the indicated region (*) of the transgene diagram in panel A is shown. C, Pancreatic Bridge-1 RNA expression in two distinct lines (TG2 and TG5) of Bridge-1 TG mice. cDNA was synthesized from total pancreatic RNA isolated from transgenic (+) or control (–) mice. A 402-bp segment of amplified Bridge-1 cDNA representing both endogenous and transgenic Bridge-1 expression was identified by PCR and subjected to agarose gel electrophoresis. An image of the ethidium bromide-stained gel is shown with the 500-bp marker migration position indicated. Relative overexpression of pancreatic Bridge-1 mRNA in transgenic as compared with control mice ranged from 17- to 28-fold. D, Images are shown of paraffin sections of pancreas samples from adult male Bridge-1 TG or strain-matched WT mice immunostained with anti-Bridge-1 antiserum (in brown) and counterstained with hematoxylin. Pancreatic islets are indicated with arrows.

 
Overexpression of Bridge-1 Disrupts Pancreatic Architecture in Adult Mice
On postnatal d 1 (P1), the histologic appearance of the pancreas and pancreatic islets in male transgenic and control mice was comparable, with scattered small and large islets (Fig. 2Go, G and I) and distinctly separate endocrine and exocrine compartments (Fig. 2Go, H and J). In contrast, we observed a spectrum of phenotypes of altered pancreatic architecture on hematoxylin- and eosin-stained sections derived from the adult Bridge-1 transgenic mice (Fig. 2Go, A–F). A phenotype of pancreatic duct enlargement occurred with variable penetrance in adult male and female transgenic (Fig. 2BGo) in comparison to age-, strain-, and gender-matched control mice (Fig. 2EGo). We did not observe any evidence of ductal metaplasia or loss of epithelial cell polarity in association with the enlarged ducts. We also observed scattered areas of disorganization of exocrine cells with marked variability in the size of cells and nuclei in the acinar compartment of the adult male and female Bridge-1 transgenic mice (Fig. 2CGo) as compared with the more orderly arrangement of the exocrine pancreas in control mice (Fig. 2FGo).



View larger version (93K):
[in this window]
[in a new window]
  		Fig. 2. Pancreatic Islet, Ductal, and Acinar Architecture Are Modified in Adult But Not Neonatal Bridge-1 Transgenic Mice

A–F, Hematoxylin- and eosin-stained paraffin sections of adult male Bridge-1 TG (A–C) or WT (D–F) mouse pancreas. Pancreatic endocrine tissue (ISLET) in Bridge-1 transgenic mice (arrows, A) shows a loss of defined endocrine/exocrine boundaries and increased cellular and nuclear heterogeneity as compared with control islets (arrow, D). Enlarged and dilated pancreatic ducts (DUCT) are noted in transgenic (arrow, B) as compared with control pancreas (E). Patchy regions of acinar cell disorganization (ACINAR) also are seen in transgenic pancreas (C) as compared with control exocrine tissue (F). G–J, Hematoxylin- and eosin-stained paraffin sections of postnatal d 1 (NEWBORN) male Bridge-1 TG (G and H) or WT (I and J) mouse pancreas. Arrows indicate small and large islets in transgenic and control P1 pancreas (PANCREAS, G and I). A clear separation of endocrine and exocrine compartments is indicated (arrows, ISLET, H and J) in images of transgenic and control pancreatic islets.

 
We noted the most striking morphological changes in the arrangement and distribution of pancreatic endocrine cells. In adult male Bridge-1 transgenic mice, the pancreatic islet morphology ranged from nearly normal to streak-like nests of cells. In contrast to wild-type adult pancreatic islets in which endocrine cells were assembled in well-organized spherical structures (Fig. 2DGo), in the pancreas of adult transgenic mice strands of endocrine cells were interwoven among exocrine cells in disordered patterns (Figs. 2AGo and 3AGo) with loss of the normal tissue boundaries between the endocrine and exocrine compartments. Disordered groups of cells were identified as endocrine cell clusters by their expression of the hormones insulin (Fig. 3BGo) or glucagon (Fig. 3CGo) and correlated with observed patterns of Bridge-1 overexpression (Fig. 3DGo). The endocrine cells of the transgenic mice had marked heterogeneity of nuclear size and shape. The loss of both intracellular and extracellular organization was illustrated by marked changes in the intracellular expression patterns for ß-catenin in endocrine cell clusters derived from adult Bridge-1 transgenic (Fig. 3EGo) relative to control mice (Fig. 3FGo).



View larger version (91K):
[in this window]
[in a new window]
  		Fig. 3. Pancreatic Endocrine Cells Are Disordered with Loss of Endocrine and Exocrine Compartment Boundaries in Bridge-1 Transgenic Mice

A–E, Paraffin sections of mouse pancreas samples derived from multiple male Bridge-1 TG mice were stained with hematoxylin and eosin (A; H & E) or immunostained for insulin (B), glucagon (C), or Bridge-1 (D) (in brown) and counterstained. The immunostaining pattern indicated for ß-catenin (in brown) is markedly different in Bridge-1 TG (E) as compared with WT (F) mice.

 
Endocrine Cell Distribution Is Altered in Bridge-1 Transgenic Mice
When we examined pancreatic endocrine hormone protein expression patterns of severely affected adult male Bridge-1 transgenic mice by immunohistochemistry, we observed reductions in the extent and intensity of insulin expression in transgenic as compared with wild-type control mice. We also found a relative increase in glucagon expression within pancreatic endocrine cell clusters of transgenic mice. Only a small fraction of endocrine cells expressed somatostatin in transgenic or control mice (data not shown).

The spatial organization of the glucagon-expressing cells relative to the insulin-expressing cells also was disrupted in the pancreatic endocrine cell clusters of adult male and female Bridge-1 transgenic mice. In pancreatic islets of adult control mice, we observed the typical endocrine cell arrangement of glucagon-expressing {alpha}-cells in an outer rim (Fig. 4EGo) surrounding a central core of insulin-expressing ß-cells (Fig. 4Go, D and F). However, in adult transgenic endocrine cell clusters, a disorganized pattern of ß-cells (Fig. 4AGo) with a relative increased complement of {alpha}-cells (Fig. 4BGo) was seen in the context of poorly defined endocrine and exocrine compartment boundaries. No coexpression of insulin and glucagon was observed (Fig. 4CGo) to suggest a population of multipotential endocrine cells. In contrast, we did not observe differences between the expression patterns (Fig. 4Go, G and J) or the distribution (Fig. 4Go, G–L) of {alpha} and ß-cells in pancreatic islets of P1 newborn transgenic as compared with control mice. Insulin-expressing cells were consistently and abundantly represented in pancreatic islets from litters of newborn Bridge-1 transgenic mice.



View larger version (35K):
[in this window]
[in a new window]
  		Fig. 4. The Spatial Relationship of Insulin- and Glucagon-Expressing Cells Is Altered in Adult But Not Neonatal Bridge-1 Transgenic Mice

A–F, Indirect immunofluorescence of pancreas sections from adult TG (A–C) and WT (D–F) male mice stained for insulin (INS) in red (A and D) and glucagon (GLU) in green (B and E) are shown with the merged images as indicated (C and F). Pancreatic islet architecture is abnormal in the adult transgenic mice with diminished numbers of insulin-expressing cells (arrow, A) and a disordered arrangement of glucagon-expressing cells (arrow, B). G–L, Indirect immunofluorescence of pancreas sections from postnatal d 1 (NEWBORN) TG (G–I) and WT (J–L) male mice stained for insulin in red and glucagon in green is shown in merged images as indicated.

 
Next we examined the expression patterns of the transcription factors PDX-1 and Nkx 6.1 as markers of pancreatic ß-cells in adult mice (15, 16). As compared with the well-organized expression patterns in ß-cells within the central core of control islets (Fig. 5Go, E–H), the expression levels of both transcription factors were reduced in disordered patterns within endocrine cell clusters from adult Bridge-1 transgenic mice (Fig. 5Go, A–D). These ß-cell marker expression patterns resembled that of insulin, consistent with a reduction in the functioning mass of pancreatic ß-cells. We did not observe any coexpression of glucagon with either PDX-1 (Fig. 5CGo) or Nkx 6.1 (Fig. 5DGo) that would support a model of ß-cell transdifferentiation to an intermediate {alpha}-cell phenotype. The nuclear expression patterns of both PDX-1 and Nkx 6.1 illustrated the heterogeneous range of nuclear sizes observed in the transgenic ß-cells.



View larger version (47K):
[in this window]
[in a new window]
  		Fig. 5. Expression Levels of Pancreatic ß-Cell-Specific Transcription Factors Are Reduced in the Pancreatic Endocrine Cells of Bridge-1 Transgenic Mice

A–H, Protein expression levels of PDX-1 and Nkx 6.1 are reduced in Bridge-1 transgenic mice. Representative images from pancreatic sections from adult male TG (A–D) and WT (E–H) mice stained with antiserum directed against PDX-1 (A and E) or Nkx 6.1 (B and F) (in brown) and counterstained with hematoxylin are shown (A, B, E, and F). Within the disordered endocrine cell clusters of Bridge-1 transgenic mice (arrows, A and B) differences in nuclear sizes are seen. PDX-1 (E) and Nkx 6.1 (F) are expressed in the cytoplasm and nuclei of pancreatic ß-cells comprising the central core of the wild type pancreatic islets. Indirect immunofluorescence of pancreatic sections costained for PDX-1 in red and glucagon in green (C and G) or for Nkx 6.1 in red and glucagon in green (D and H) are shown. I, Pancreatic mRNA expression levels of PDX-1, insulin, and glucagon gene regulators are altered in Bridge-1 transgenic mice. Pancreatic RNA was prepared from adult male Bridge-1TG and strain- and age-matched WT mice and real-time RT-PCR was performed for each sample in triplicate with primers and probes to detect PDX-1, Foxa2, Nkx 6.1, Maf A, NeuroD1, neurogenin-3 (NGN-3), Brain-4, Arx, Maf B, and HPRT, as indicated. Results shown are the means ± SEM of the relative expression levels normalized to cyclophilin expression with the mean expression level for the WT mice set at 1 (n = 3–5 mice per genotype; *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, no significant difference by Student’s t test). J, Scattered neurogenin-3-expressing cells (NGN-3, arrows, in brown) are seen near ducts in pancreatic sections derived from adult Bridge-1 transgenic mice with diabetes.

 
To complement the protein expression patterns of the ß-cell-specific transcription factors, we isolated pancreatic RNA from adult male Bridge-1 transgenic and age- and strain-matched control mice for quantitative real-time RT-PCR analysis of differences in transcription factor gene expression. In concordance with the observed protein expression patterns, we observed substantial and significant reductions in PDX-1 (64% reduction) and Nkx 6.1 (72% reduction) mRNA expression levels in the transgenic relative to control mice (Fig. 5IGo). We found a 70% reduction in the expression of Foxa2, a key activator of pdx-1 gene expression (17, 18, 19), but no significant differences in the expression of Foxo1, an inhibitor of PDX-1 expression (20) or of E2A (data not shown). The expression of the insulin promoter regulator Maf A (21, 22) was substantially decreased by 78%; however, Maf B expression was increased by 89% in the transgenic as compared with control RNA. In the same series of experiments we found an 85% increase in the mRNA expression levels of the {alpha}-cell-specific transcription factor Brain-4 (23) and a trend (P = 0.06) suggesting a 45% increase in mRNA expression of the {alpha}-cell regulator Arx (24) in the transgenic relative to control mice.

We looked for evidence of any reactivation of developmental transcription programs for the generation of new endocrine cells in adult Bridge-1 transgenic mice. Notably, we observed a 3-fold increase in the pancreatic mRNA expression levels of the endocrine progenitor cell marker neurogenin-3 (25, 26) in adult transgenic as compared with control mice (Fig. 5IGo). Although we did not observe convincing evidence of neurogenin-3 protein expression in the adult control pancreas sections, we found occasional scattered neurogenin-3-expressing cells in pancreas sections derived from adult Bridge-1 transgenic mice with diabetes (Fig. 5JGo). Neurogenin-3 protein expression was found in isolated cells near ducts but not within endocrine cell clusters. NeuroD1/Beta-2 mRNA expression levels also were elevated by 2.5-fold in transgenic as compared with control mice (Fig. 5IGo), consistent with the known role of neurogenin-3 as a potent activator of the expression of NeuroD1 (26, 27). The mRNA expression levels of Hes-1, Pax 4, and p48/Ptf1a were not significantly different between transgenic and control mice, but Pax 6 expression levels were significantly reduced in transgenic mice by 54% (data not shown). We did not observe any significant differences between transgenic and control mice in pancreatic mRNA expression levels of the ubiquitously expressed hypoxanthine phosphoribosyltransferase (HPRT) gene.

Pancreatic ß-Cell Complement Is Reduced and {alpha}-Cell Complement Is Increased in Endocrine Cell Clusters of Bridge-1 Transgenic Mice
To determine whether the observed altered patterns of insulin and glucagon expression were indicative of differences in cell numbers, we estimated the relative numbers of pancreatic {alpha} and ß-cells in adult control and transgenic mouse pancreas by counting stained and unstained endocrine cells in paraffin-embedded pancreatic sections stained for insulin (Fig. 6Go, A–C) or for glucagon (Fig. 6Go, D–F). Pancreatic sections derived from male Bridge-1 transgenic mice (TG) had an average of 12 insulin-expressing cells per endocrine cell cluster as compared with an average of 54 insulin-expressing cells in those derived from male control [wild-type (WT)] mice (Fig. 6AGo). On average only 25% of the cells in an endocrine cell cluster expressed insulin in transgenic mice, whereas 80% expressed insulin in control islets (Fig. 6BGo). These data suggest that ß-cell mass in severely affected adult Bridge-1 transgenic mice is reduced by approximately 70%. The reductions in estimated pancreatic ß-cell mass were similar in magnitude to the decrements in pancreatic mRNA expression levels observed for the ß-cell-specific transcription factors PDX-1 and Nkx 6.1.



View larger version (54K):
[in this window]
[in a new window]
  		Fig. 6. Bridge-1 Transgenic Mice Have Reduced Pancreatic ß-Cell Mass and Increased Pancreatic Apoptosis

A–C, Paraffin sections of pancreas derived from adult male Bridge-1 TG or WT mice were immunostained for insulin and counterstained with hematoxylin. For each endocrine cell cluster or islet, insulin-expressing (ß-cells) and nonexpressing (non-ß-cells) cells were counted. Data presented are the mean ± SEM (n = 3 mice per genotype; 2400–4294 endocrine cells per genotype; **, P < 0.01; ***, P < 0.001). D–F, Paraffin sections of pancreas derived from adult male Bridge-1 TG or WT mice were immunostained for glucagon and counterstained with hematoxylin. For each endocrine cell cluster or islet, glucagon-expressing ({alpha}-cells) and nonexpressing (non-{alpha}-cells) cells were counted. Data presented are the mean ± SEM (n = 3 mice per genotype; 5076–8028 endocrine cells per genotype; *, P < 0.10; **, P < 0.01). G, Peroxidase-based TUNEL assays were performed on paraffin sections of pancreas derived from adult male Bridge-1 TG or WT mice. Representative images of TUNEL-positive cells (in brown) are shown. H, We prepared pancreatic protein extracts from adult male Bridge-1 TG and WT mice and conducted Western blots first with anti-cleaved caspase-3 antiserum and then with anti-Stat-3 antiserum as a loading control. Representative images for Stat-3 and the 17-kDa activated, cleaved form of caspase-3 are shown (at left). Densitometric scanning of the Western blots was used to quantify expression of the 17-kDa form of caspase-3 and of Stat-3. Data shown (at right) are the average ± range of relative expression levels normalized to the average expression level from control mice set at 1 (n = 2 mice per genotype).

 
Both increased numbers and relative proportions of pancreatic {alpha}-cells were seen in endocrine cell clusters in Bridge-1 transgenic mice (Fig. 6Go, D–F). On average, 12 glucagon-expressing cells per islet were found in WT mouse pancreas sections, whereas 23 glucagon-expressing cells per endocrine cell cluster were seen in TG mouse pancreas sections (Fig. 6DGo). Similarly, the average proportion of glucagon-expressing cells per islet or endocrine cell cluster was higher at 52% for transgenic mice than the 20% observed for control mice (Fig. 6EGo). These {alpha}-cell counts were consistent with the observed increase in the pancreatic mRNA expression levels of the {alpha}-cell-specific transcription factor Brain-4.

Pancreatic Apoptosis Is Increased in Bridge-1 Transgenic Mice
To identify potential mechanisms by which pancreatic ß-cell mass is reduced in Bridge-1 transgenic mice, we performed terminal deoxynucleotidyl transferase-mediated deoxyuridine 5'-triphosphate nick end labeling (TUNEL) assays to determine the extent of apoptosis in pancreatic sections derived from adult transgenic and age-, strain-, and gender-matched control mice. In WT mice, we rarely observed TUNEL-positive cells within pancreatic islets (Fig. 6GGo, left panel). However, in Bridge-1 TG mice, TUNEL-positive cells were frequently found within endocrine cell clusters (Fig. 6GGo, middle panel) and in wandering strands of cells interspersed between acinar cells of the exocrine pancreas (Fig. 6GGo, right panel). The TUNEL-positive cells often had morphologic features of nuclear fragmentation or condensation. To confirm this finding we measured the protein levels of the activated, cleaved form of the protease caspase-3 that is central to the apoptotic protease cascade. By Western blot analysis of total pancreatic protein extracts, we observed an increase of greater than 3-fold in the levels of the 17-kDa cleaved form of caspase-3 in contrast to unchanged levels of pancreatic signal transducer and activator of transcription (Stat)-3 protein expression in adult Bridge-1 transgenic as compared with control mice (Fig. 6HGo). These data indicate that increased pancreatic apoptosis occurs in Bridge-1 transgenic mice.

Overexpression of Bridge-1 Results in Insulin Deficiency and Hyperglycemia
We observed hyperglycemic phenotypes in multiple lines of Bridge-1 transgenic mice. The metabolic phenotypes in adult Bridge-1 transgenic mice ranged from mild hyperglycemia to severe diabetes. Female Bridge-1 transgenic mice had modest fasting hyperglycemia and insulin deficiency demonstrated by ip glucose tolerance testing (Fig. 7AGo). Fasting insulin levels were reduced and serum insulin levels rose minimally in response to a glucose challenge in the female transgenic mice.



View larger version (34K):
[in this window]
[in a new window]
  		Fig. 7. Bridge-1 Transgenic Mice Have Insulin Deficiency and Diabetes

A, Intraperitoneal (IP) glucose tolerance testing of adult female transgenic mice (closed circles) demonstrates hyperglycemia, reduced glucose-stimulated insulin production and markedly diminished insulin/glucose ratios (n = 6 mice per genotype; *, P = 0.056; **, P < 0.01; ***, P < 0.001) as compared with age-, strain-, and gender-matched control mice (open circles). B, Fasting glucose and insulin levels and insulin/glucose ratios are shown for adult male TG and age-, strain-, and gender-matched WT mice (n = 5–7 mice per genotype; ***, P < 0.001). C, Bridge-1 transgenic mice with severe diabetes (TG) have marked deficits in pancreatic insulin mRNA expression as compared with WT mice. Autoradiograms are shown of Northern blots of total pancreatic RNA from two TG and two WT adult male mice probed for insulin and actin mRNA. D, We prepared pancreatic RNA from adult male Bridge-1 TG and strain- and age-matched WT mice and performed real-time RT-PCR for each sample in triplicate with primers and probes to detect insulin, glucagon, glucokinase, Glut-2, elastase, and somatostatin, as indicated. Results shown are the means ± SEM of the relative expression levels normalized to cyclophilin expression with the mean expression level for the WT mice set at 1 (n = 3–5 mice per genotype; *, P = 0.058; **, P < 0.01; ns, no significant difference by Student’s t test). E, Fasting glucagon levels are shown for adult male TG and age-, strain-, and gender-matched WT mice (n = 6–8 mice per genotype; P = 0.04).

 
In male transgenic mice, with marked abnormalities in pancreatic histology we observed the most severe diabetes. Male transgenic mice had fasting blood glucose levels as high as 438 mg/dl and random fed plasma glucose levels as high as 855 mg/dl associated with nearly undetectable fasting serum insulin levels (Fig. 7BGo). Elevated random-fed triglyceride levels accompanied the hyperglycemia with an average triglyceride level of 620 mg/dl (n = 6) in Bridge-1 transgenic mice with severe diabetes as compared with 158 mg/dl (n = 4) in nontransgenic control mice. The severity of the observed metabolic phenotypes partially correlated with transgene copy number by Southern blot or real-time quantitative PCR analyses with a spectrum of higher levels of hyperglycemia seen with increased gene dosage.

Insulin deficiency in hyperglycemic Bridge-1 transgenic mice was associated with markedly reduced levels of pancreatic preproinsulin mRNA that was scarcely detectable by Northern blot analyses (Fig. 7CGo). Insulin mRNA levels could be detected by real-time quantitative RT-PCR of total pancreatic RNA in another series of adult Bridge-1 transgenic and control mice. In this cohort, insulin mRNA levels were reduced by 76% in Bridge-1 TG relative to WT mice (Fig. 7DGo). The most severe insulin deficiency that we observed in a Bridge-1 TG mouse was accompanied by ketonuria detected by urine dipstick testing, a clinical finding most commonly associated with type 1 diabetes and an absolute insulin deficit.

Pancreatic glucagon mRNA levels were increased by 28% and somatostatin mRNA levels were decreased by 53% in Bridge-1 transgenic mice, but elastase mRNA levels were not significantly different in the same experimental series (Fig. 7DGo). The relative increase of glucagon mRNA expression in the transgenic mouse pancreas was modest in comparison to the observed increases in {alpha}-cell complement in endocrine cell clusters or in levels of Brain-4 mRNA expression. Pancreatic mRNA expression levels of the ß-cell-specific glucose sensors glucokinase and glucose transporter-2 (Glut-2) were not significantly different between Bridge-1 transgenic and control mice. Fasting glucagon levels were not increased in adult Bridge-1 transgenic as compared with control mice (Fig. 7EGo).

Bridge-1 Regulation of Insulin Gene Expression Is Biphasic
We previously demonstrated that endogenous levels of the coactivator Bridge-1 function to augment insulin promoter activity (11). In this experimental model, we found that increased pancreatic Bridge-1 expression reduced insulin mRNA levels. The overexpression of other insulin gene activators, including PDX-1 and E47, is known to suppress the transcriptional activation of the insulin promoter (7, 28, 29, 30). Therefore, we suspected that varying the levels of Bridge-1 expression also might result in a dose-dependent, biphasic pattern of insulin promoter activation, by first promoting and then uncoupling the optimal stoichiometric assembly of transcriptional regulatory complexes containing Bridge-1 protein interaction partners. In transient transfections in HeLa cells, we observed that increasing the amounts of exogenous Bridge-1 expression resulted in a shift from activation of a rat insulin I enhancer-reporter construct to repression in a pattern reminiscent of those observed for other transcriptional activators of insulin gene expression (7, 28, 29, 30) (Fig. 8AGo).



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 8. Overexpression of Bridge-1 in Vitro Regulates Insulin Gene Expression in a Biphasic Pattern

A, Dose-dependent activation of insulin promoter enhancer elements by Bridge-1 is biphasic. HeLa cells were transiently transfected in triplicate with a multimerized FarFlat enhancer-reporter plasmid and pcDNA3 or pcDNA3-Bridge-1. For each condition, the total amount of pcDNA3 plasmids was constant. Results shown are the mean ± SEM normalized to the basal activation of the reporter in the presence of pcDNA3 alone. B, Inducible overexpression of Bridge-1 protein in double stable CMV-rtTA/Tet-Bridge-1 INS-1 cells. CMV-rtTA/Tet-Bridge-1 INS-1 cells were treated with 1000 µg/ml doxycycline (dox, +) or vehicle (–) for 72 h before harvest. Whole cell extracts were analyzed by Western blotting with anti-Bridge-1 antiserum. Bridge-1 protein was visualized by enhanced chemiluminescence and a representative film is shown with the Bridge-1 protein indicated (arrow). In control INS-1 cells (not shown), doxycycline did not induce Bridge-1 protein expression. C, Dose-dependent inducible Bridge-1 mRNA expression. CMV-rtTA/Tet-Bridge-1 INS-1 cells were treated with 0, 50, 100, 250, 500, or 1000 µg/ml doxycycline, as indicated, for 72 h before harvest and RNA isolation. Relative Bridge-1 mRNA expression levels were determined by real-time RT-PCR conducted in duplicate for each of two independent samples per treatment condition. Results shown are the means ± SD of the relative expression levels normalized to cyclophilin expression with the mean expression level for control cells (0) set at 1 [*, P < 0.05; **, P < 0.01 as compared with control cells (0)]. D, Increasing Bridge-1 expression in vitro regulates insulin mRNA expression levels in a biphasic pattern. CMV-rtTA/Tet-Bridge-1 INS-1 cells were treated with 0, 50, 100, 250, 500, or 1000 µg/ml doxycycline, as indicated, for 72 h before harvest and RNA isolation. Relative insulin mRNA expression levels were determined by real-time RT-PCR conducted in duplicate for each of two independent samples per treatment condition. Results shown are the means ±SD of the relative expression levels normalized to cyclophilin expression with the mean expression level for control cells (0) set at 1 [*, P < 0.05; **, P < 0.01 as compared with control cells (0)].

 
To extend our studies of the regulation of insulin gene expression by Bridge-1 in vitro, we generated double stable CMV-rtTA/Tet-Bridge-1 INS-1 cell lines in which the overexpression of Bridge-1 could be induced in a dose-dependent manner by the administration of the tetracycline analog doxycycline. In CMV-rtTA/Tet-Bridge-1 INS-1 cells, the administration of doxycycline significantly increased Bridge-1 protein (Fig. 8BGo) and mRNA (Fig. 8CGo) expression. In contrast, the administration of doxycycline in a range of concentrations from 50–1000 µg/ml did not significantly change Bridge-1 protein or mRNA expression levels in control INS-1 cells as assessed by Western blots and real-time quantitative RT-PCR. Notably, the overexpression of Bridge-1 in insulin-producing cells in vitro altered the levels of insulin gene expression in a biphasic pattern. Increasing Bridge-1 mRNA expression levels by approximately 6-fold significantly increased insulin mRNA expression by 36% (P = 0.011), whereas a 10-fold increase in Bridge-1 expression significantly inhibited insulin gene expression by 50% (P = 0.007). Doxycycline administration did not significantly change insulin mRNA expression levels in control INS-1 cells.

Under conditions in which the overexpression of Bridge-1 reduced insulin gene expression by 50%, we found no evidence for significant reductions in the protein or mRNA expression levels for the Bridge-1-interacting partners PDX-1 or E2A proteins E12 and E47 as evaluated by Western blots of nuclear and whole cell extracts and by real-time quantitative RT-PCR. Similarly, under these experimental conditions we did not observe changes in activated caspase-3 protein expression levels on Western blots. To assess the function of PDX-1 and E2A proteins in the setting of Bridge-1 overexpression and reduced insulin gene expression, we conducted DNA-binding assays with nuclear extracts derived from double stable CMV-rtTA/Tet-Bridge-1 INS-1 cells treated with 1000 µg/ml doxycycline or vehicle for 72 h. We found no evidence that the overexpression of Bridge-1 disrupted the binding of PDX-1 to the rat insulin I promoter elements FarFlat or P1 or the binding of E2A proteins to the rat insulin I promoter element Nir (data not shown). These data support a model of Bridge-1 function in which increases of Bridge-1 expression levels within a distinct range augment the transcriptional activation of the insulin gene. Beyond this range, larger increases in Bridge-1 expression levels likely shift transcriptional regulation to limit insulin production.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
In this transgenic mouse model, the pancreatic overexpression of the coactivator Bridge-1 results in insulin deficiency and diabetes. We overexpressed Bridge-1 to uncouple supramolecular protein complexes that regulate Bridge-1 target genes in the pancreas. With these in vivo studies, we now suggest potential functions for Bridge-1 signaling in the regulation of glucose homeostasis and of pancreatic ß-cell survival.

Hyperglycemia and insulin deficiency in Bridge-1 transgenic mice likely result from reductions both in ß-cell mass and in insulin gene expression. In contrast to neonatal transgenic mice with normal insulin expression patterns, adult transgenic mice had a marked deficit in pancreatic ß-cell mass associated with increased caspase-3-dependent apoptosis. A similar phenotype of increased pancreatic ß-cell apoptosis and decreased ß-cell mass is found in humans with type 2 diabetes (1). The increased pancreatic expression of neurogenin-3 and NeuroD1/Beta-2 in adult Bridge-1 transgenic mice with diabetes is consistent with a limited compensatory reactivation of a transcriptional program to generate new endocrine cells (31); however, this response was insufficient to restore ß-cell mass, insulin production, or glucose homeostasis.

The precise stoichiometric assembly of transcription factors and coactivators is essential for appropriately regulated gene expression. The pancreatic overexpression of the coactivator Bridge-1 likely altered the transcriptional regulatory complexes of Bridge-1 target genes in vivo, effectively interrupting normal Bridge-1 signaling and suppressing insulin gene expression. The deficit in insulin gene expression correlated both with reduced pancreatic ß-cell mass and with diminished protein expression levels of insulin and key transcriptional regulators of the insulin promoter within ß-cells. Thus, even the surviving ß-cells in this model provided evidence of an impaired capacity to produce insulin.

We were surprised to find unchanged pancreatic expression levels of glucokinase or Glut-2 in the setting of reduced numbers of pancreatic ß-cells. These two proteins are rate-limiting components for glucose-sensing in pancreatic ß-cells and downstream regulatory targets of PDX-1 (32). Our data raise the possibility that the surviving pancreatic ß-cells in Bridge-1 transgenic mice with diabetes up-regulated the expression of Glut-2 and glucokinase as an adaptive response to the reduction in ß-cell mass. It is unclear whether increased expression of glucose sensors in combination with hyperglycemia would be protective or deleterious to pancreatic ß-cell survival. In contrast, both glucokinase and Glut-2 expression levels are decreased in pancreatic islets with the overexpression of the coactivator PGC-1{alpha} (peroxisome proliferator-activated receptor-{gamma} coactivator 1) (33).

This mechanism for insulin gene regulation is consistent with reports that the overexpression of other insulin gene activators, such as PDX-1 or E47, represses insulin promoter activity (7, 28, 29, 30). Increasing concentrations of E47 first promote and then disrupt the synergistic activation of glucose-responsive enhancers of the insulin promoter in conjunction with PDX-1 (7). Similarly, interference with endogenous Bridge-1 expression in insulin-producing cells reduces the transcriptional activation of the insulin promoter (11), and increasing Bridge-1 expression levels in vitro result in a dose-dependent, biphasic activation of insulin gene enhancers and insulin gene expression (Fig. 8Go, A and D). Dose-dependent expression of PDX-1 is essential to maintain normal glucose homeostasis, and relatively small changes in PDX-1 expression levels result in metabolic dysfunction (34). Our data suggest that changes in the levels of Bridge-1 expression may have a similar capacity to augment or impair insulin gene expression.

We considered the possibility that the overexpression of this PDZ domain coactivator might result in nonspecific dysregulation of pancreatic ß-cell function. We believe that this is not the case for several reasons. We have consistently observed specificity in our studies of Bridge-1 interactions with other proteins. In mammalian two-hybrid assays, Bridge-1 interacted with E12 but not with the related bHLH transcription factor NeuroD1/Beta-2 (11). Similarly, in yeast interaction trap assays, Bridge-1 specifically interacted with E12, E47, and PDX-1 but not with negative control proteins (11, 12). Specificity of Bridge-1 protein interactions is partially attributable to the unique primary structure of the Bridge-1 PDZ protein interaction domain that is atypical both in length and in amino acid sequence as compared with other PDZ domain-containing proteins (11). Only a few amino acid substitutions are sufficient to markedly change the binding specificities of PDZ domains (35). Furthermore, the carboxy-terminal amino acid sequence of E12, the known PDZ domain ligand for Bridge-1 (11), does not fall within the three major classes of consensus binding sequences for PDZ domains at the critical P2 position (36, 37). To date, few existing examples in the literature implicate other PDZ domain proteins as candidate regulators of apoptosis (38, 39). Finally, temporal specificity is observed in the phenotype of reductions in ß-cell mass that are not present at birth but develop as postnatal deficits, despite the activity of the pdx-1 promoter in the embryonic pancreas. Thus, we propose that the phenotypes observed in Bridge-1 transgenic mice likely reflect the specific capacity of Bridge-1 to interact with and modulate the function of proteins that regulate glucose homeostasis, insulin gene expression, and pancreatic ß-cell survival.

Metabolic dysfunction in this model ranged from mild to severe hyperglycemia. Milder metabolic phenotypes in female Bridge-1 transgenic mice correlated with smaller deficits in fasting serum insulin levels and in pancreatic insulin protein expression levels. Male transgenic mice had severe diabetes accompanied by extremely low serum insulin levels. Defective insulin production similarly is observed in humans with MODY or early type 1 diabetes. In many mouse models of type 2 diabetes, including models with defective insulin signaling or insulin production, hyperglycemia is more prominent in males than females (34, 40), as we noted in Bridge-1 transgenic mice. The mechanisms of these gender-related differences in glucose metabolism have not been established, although estrogen levels may modulate insulin signaling in mice (41). In human studies, females have lower fasting glucose levels than males (42), possibly related to gender-specific differences in glucagon secretion, epinephrine release, or hepatic glucose mobilization (43, 44).

Because we identified an increased proportion of glucagon-expressing {alpha}-cells relative to ß-cells in Bridge-1 transgenic mice, it is possible that abnormal regulation of glucagon production contributed to the phenotype of hyperglycemia. Although we did not find increased fasting serum glucagon levels in Bridge-1 transgenic mice, we also did not observe a substantial suppression in glucagon levels in the setting of hyperglycemia. Increased proportions of {alpha}-cells relative to ß-cells are found in pancreatic islets from many mouse models of ß-cell dysfunction and diabetes as well as in islets from humans with type 2 diabetes (3). The reduction of somatostatin mRNA expression in Bridge-1 transgenic mice likely was a result of increased Bridge-1 expression from the transgene because the pdx-1 promoter is transcriptionally active in subpopulations of somatostatin-expressing {delta}-cells.

The overexpression of Bridge-1 disrupted several aspects of pancreatic architecture. The loss of spatial organization of the pancreatic endocrine cells in adult Bridge-1 transgenic mice is a striking phenotype. Indistinct endocrine and exocrine cellular compartments also are seen in a small number of pancreatic diseases in humans, including nesidioblastosis and selected pancreatic cancers. We consider the altered patterns of intracellular ß-catenin expression observed in Bridge-1 transgenic mice to be a reflection of disrupted endocrine cell adhesion (45). The disordered endocrine and exocrine pancreatic compartments within the transgenic mice suggest that signals transduced through Bridge-1 may regulate intercellular communication.

These studies raise the possibility that multiple intracellular targets exist for Bridge-1 signaling in addition to interactions with E2A and PDX-1 proteins. Our analyses of the suppression of insulin gene expression by Bridge-1 overexpression in vitro suggest that Bridge-1 signals likely converge on additional transcriptional regulators of insulin gene expression. Given the large number of transcriptional regulatory proteins implicated in insulin gene regulation, future studies will be needed to identify the full scope of Bridge-1 regulatory targets in the endocrine pancreas.

The human Bridge-1 homolog, PSMD9, was identified as a potential proteasomal modulator (11, 46). PSMD9 was isolated in a protein complex with the proteasomal subunit HIV-1 transactivator protein (Tat)-binding protein (TBP-1)/regulatory particle triphosphatase (Rpt5) (46, 47) that is implicated in transcriptional regulation in yeast and mammals (48, 49). The processes of transcriptional activation and protein degradation often are closely coupled (50, 51, 52, 53). The potential dual capacity of Bridge-1 to regulate gene expression and to modulate the functions of proteasomal components could be similar to dual functions ascribed to a small group of proteasomal subunits, coactivators, and ubiquitin ligases.

We envision that the coactivator Bridge-1 may serve as a rheostat to receive extracellular signals and translate them into graded levels of insulin gene expression. Bridge-1 may have a similar modulatory function to regulate pancreatic ß-cell mass via additional target genes and protein interactions. Additional studies will be needed to identify the signaling pathways and protein interaction partners that converge upon Bridge-1 to mediate these functions and to establish the potential importance of signaling through Bridge-1 in the regulation of glucose homeostasis.

Coactivator dysfunction is implicated in the pathogenesis of many diseases, including malignancies, neurodegenerative disorders, and mental retardation (54). Accumulating evidence indicates that coactivators regulate insulin action and nutrient metabolism in extrapancreatic tissues. CBP alters insulin sensitivity (55), and steroid receptor coactivator 1 and transcription intermediary factor 2 adjust rates of energy expenditure (56). PGC-1{alpha} regulates glucose uptake and gluconeogenesis (57) and its overexpression inhibits glucose-stimulated insulin secretion (33). Genetic variations in the pgc-1 gene may increase susceptibility to insulin resistance and diabetes in humans (58). Furthermore, mutations in the MODY genes hnf-4{alpha}, neuroD1, and ipf-1(pdx-1) that are associated with heritable forms of diabetes in humans also alter their interactions with the coactivators CBP or p300 (10, 59, 60).

We propose that coactivator dysfunction in pancreatic ß-cells can limit insulin production and contribute to the pathogenesis of diabetes in humans. The locus of the human Bridge-1 gene, chr 12q24.31–32 (46), lies within a region identified by multiple genome-wide scans for candidate type 2 or MODY diabetes genes that is distinct from the adjacent MODY3 locus (61, 62, 63, 64, 65, 66). We envision that mutations or polymorphisms in the Bridge-1 gene could conceivably contribute to metabolic dysfunction by decreasing insulin production or reducing pancreatic ß-cell mass. Thus, we consider Bridge-1 to represent a promising candidate diabetes gene that may warrant investigation in genetic studies in humans.

The therapeutic effectiveness of selective estrogen receptor modulators demonstrates the feasibility of targeting distinct tissue-specific and promoter-specific recruitment of coactivators in the treatment of disease (67). We predict that the regulation of pancreatic ß-cell function by coactivators like Bridge-1 may provide new therapeutic opportunities to restore insulin production in individuals with diabetes.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Generation of Transgenic Mice
The 6-kb Bridge-1 transgene was constructed by cloning the full-length rat Bridge-1 cDNA coding sequence (11) (nucleotides 492-1392) downstream of a previously characterized –4.6-kb segment of the mouse pdx-1 promoter (13, 14) and upstream of rabbit ß-globin polyadenylation tail sequences. Bridge-1 overexpression mice were generated in the FVB strain at the Beth Israel Hospital Transgenic Facility using standard procedures (68). Transgenic mice were compared with age-, gender-, and strain-matched control mice for all analyses. Transgenic mice were identified by PCR amplification of a 400-bp fragment of transgenic genomic DNA that extends from nucleotide –92 in the mouse pdx-1 promoter to nucleotide 744 within rat Bridge-1 cDNA. Male gender was assigned to neonatal transgenic or control mice by PCR amplification of a 333-bp fragment of mouse Sry genomic DNA. Animal studies were approved by and conducted according to the policies of the Massachusetts General Hospital Institutional Animal Care and Use Committee.

Southern Blots
Transgene incorporation was confirmed by Southern blots of genomic DNA according to published methods (69) with a radiolabeled 223-bp fragment of the transgene that spans the junction of the pdx-1 promoter and rat Bridge-1 cDNA.

Western Blots
Total pancreatic and liver protein extracts were prepared and Western blots were conducted using enhanced chemiluminescence as described (70). Protein concentrations were determined by Micro BCA (Pierce, Rockford, IL) and Bradford assays for sample normalization. Antisera used included rabbit polyclonal anti-Bridge-1, rabbit polyclonal anti-E2A (E2A.E12, V18) (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit polyclonal anti-PDX-1, rabbit polyclonal anti-Stat-3 (K-15) (Santa Cruz Biotechnology) and human anti-cleaved caspase-3 that reacts with both the inactive 32-kDa and the active cleaved 17-kDa forms of caspase-3 (Biocarta, San Diego, CA). Relative expression levels were estimated by densitometric scanning of gels or Western blots conducted with an Image Station 440CF and Image Analysis software (Eastman Kodak, Rochester, NY).

RNA Expression Analyses
Total pancreatic RNA isolation, RT-PCR, and Northern blots for insulin and actin were conducted with previously reported methods (70, 71). Quantitative real-time RT-PCR was conducted for each sample in triplicate on an ABI Prism 7900HT sequence detection system using the manufacturer’s reagents and methods (Applied Biosystems, Foster City, CA). We designed primer and MGB probe sets for Arx, Brain-1, Bridge-1, cyclophilin, E2A, elastase, Foxo1, glucagon, glucokinase, glucose transporter-2, Hes-1, Foxa2, HPRT, insulin, Maf A, Maf B, NeuroD1, neurogenin-3, Nkx 6.1, P48/Ptf1a, Pax 4, Pax 6, PDX-1, and somatostatin. Sequences of primers and probes are available upon request.

Histologic Analyses
Mouse pancreas samples were fixed in 10% saline-buffered formalin, embedded in paraffin, and sectioned at 4-µm intervals. Immunostaining was conducted according to standard methods (11, 72). Sections were incubated for 1 h with primary antiserum including rabbit polyclonal anti-ß-catenin (Chemicon International, Temecula, CA), rabbit polyclonal anti-Bridge-1, mouse monoclonal antiglucagon (K79bB10; Sigma, St. Louis, MO), guinea pig antihuman insulin (IgG fraction; Linco Research, St. Charles, MO), mouse monoclonal antineurogenin-3 (BD Biosciences Pharmingen, San Diego, CA), rabbit polyclonal anti-GST-Nkx 6.1 (gift from P. Serup and R. S. Heller, Hagedorn Research Institute, Gentofte, Denmark), rabbit polyclonal anti-PDX-1, or rabbit polyclonal antihuman somatostatin (A0566; Dako Corp., Carpinteria, CA) as indicated. Biotinylated species-specific secondary antiserum and avidin-biotinylated horseradish peroxidase complexes (Vector Laboratories, Burlingame, CA) were used for peroxidase-based staining. Hematoxylin and/or eosin counterstaining was conducted as indicated. Alternatively, Cy-3 or fluorescein isothiocyanate-conjugated species-specific anti-IgG secondary antiserum (Jackson ImmunoResearch Laboratories, West Grove, PA) was employed for indirect immunofluorescence studies as described (11). TUNEL assays were conducted with peroxidase-based In Situ Cell Death Detection kits (Roche Diagnostics Corp., Indianapolis, IN) according to the manufacturer’s instructions with modifications including the addition of a 20-min incubation in a blocking solution consisting of 3% BSA in PBS before peroxidase treatment and of signal conversion in a 0.5x converter-POD solution diluted in PBS. Digital images were acquired with a SPOT-RT Slider color camera (Diagnostic Instruments, Sterling Heights, MI) and a Nikon epifluorescence microscope interfaced with a Macintosh G4 computer and processed with Adobe Photoshop software (Adobe Systems Inc., San Jose, CA).

Metabolic Studies
Metabolic studies were conducted with transgenic and age-, gender-, and strain-matched control mice. Glucose levels were determined with a YSI 2300 STAT glucose analyzer (Yellow Springs Instrument Co., Yellow Springs, OH). Serum insulin levels were measured in duplicate by a rat insulin ELISA kit with mouse insulin standards (Crystal Chem, Chicago, IL). Glucose tolerance tests were conducted with ip injection of 1.5 g glucose/kg body weight after an 8-h fast as previously reported (34). Serum triglycerides were measured with an Infinity Triglyceride assay according to the manufacturer’s instructions (Sigma). Urine ketone and glucose levels were detected by Chemstrip uGK (Roche, Basel, Switzerland). Serum glucagon levels were measured in duplicate by RIA (Linco Research, St. Charles, MO) from blood samples collected after an 8-h fast in tubes with EDTA, trasylol, and diprotin A according to the manufacturer’s instructions (73).

Cell Culture and Transfection Studies
HeLa cells were cultured in DMEM supplemented with 10% fetal bovine serum, 100 U of penicillin G, and 100 µg of streptomycin sulfate/ml. Transfections and chloramphenicol acetyltransferase (CAT) reporter assays were conducted as described (11). The pcDNA3-Bridge-1 expression vector encoding full-length rat Bridge-1 cDNA was previously reported (11), and the multimerized FarFlat enhancer-CAT reporter plasmid (5FF1CAT) was a gift from L. G. Moss (Duke University Medical Center, Durham, NC).

Tetracycline-regulated double stable CMV-rtTA/Tet-Bridge-1 INS-1 cell lines were generated using previously established methods (34). INS-1 cells were transfected with CMV-rtTA (pUHG 17-1) (74) and a neomycin-resistance plasmid pSV2-neo (BD Biosciences Clontech, Palo Alto, CA) followed by selection with G418 (Calbiochem, La Jolla, CA) to generate CMV-rtTA INS-1 cell lines. CMV-rtTA cells were transfected with a Tet-Bridge-1 plasmid consisting of full-length rat Bridge-1 cDNA (11) cloned in pUHD 10–3 (75) and with the hygromycin resistance plasmid pTK-Hyg (BD Biosciences Clontech). Double stable CMV-rtTA/Tet-Bridge-1 INS-1 cells were grown for passage under selection with G418 and hygromycin B. Selection agents were removed from the culture medium before conducting experiments with doxycycline. Doxycycline or vehicle was administered to CMV-rtTA/Tet-Bridge-1 INS-1 cell lines and to negative control INS-1 cell lines at time zero and again after 48 h with a change in culture medium. Cells were harvested after 72 h of doxycycline or vehicle treatment.


   ACKNOWLEDGMENTS
 
We thank J. Habener and colleagues in the Molecular Endocrinology Unit for insightful discussions, support, and reagents. We appreciate the expert assistance of C. Banz, D. Borger, and M. Tenser, of K. McManus and the Boston Area Diabetes Endocrinology Research Center (P30) RIA Core Laboratory, and of the staff in the Beth Israel Hospital Transgenic Facility. We acknowledge the gifts of Nkx 6.1 antiserum from P. Serup and R. Heller (Hagedorn Research Institute, Gentofte, Denmark); the pdx-1 promoter construct from C. Miller; HeLa cells from R. Stein (Vanderbilt University School of Medicine, Nashville, TN); INS-1 cells from C. Wollheim (University Medical Center, Geneva, Switzerland); pUHD 10–3 and pUHG 17–1 plasmids from H. Bujard (Universitat Heidelberg, Heidelberg, Germany), and the rat insulin I promoter enhancer-reporter construct 5FF1CAT from L. Moss (Duke University Medical Center, Durham, NC). We thank K. MacDonald for her assistance in the preparation of this manuscript.


   FOOTNOTES
 
These studies were supported by grants from the Public Health Service (DK59419 and DK58783 to M.K.T.) from the National Institute of Diabetes and Digestive and Kidney Diseases. M.K.T. also receives support from a Career Development Award from the American Diabetes Association.

First Published Online August 11, 2005

Abbreviations: bHLH, Basic helix-loop-helix; CAT, chloramphenicol acetyltransferase; CBP, cAMP response element binding protein-binding protein; Glut-2, glucose transporter-2; HPRT, hypoxanthine phosphoribosyltransferase; MODY, maturity-onset diabetes of the young; P1, postnatal d 1; PGC, peroxisome proliferator-activated receptor-{gamma} coactivator; PSMD9, proteasome 26S subunit, non-ATPase, 9; Stat, signal transducer and activator of transcription; TG, transgenic; TUNEL, terminal deoxynucleotidyl transferase-mediated deoxyuridine 5'-triphosphate nick end labeling; WT, wild type (control).

Received for publication March 17, 2005. Accepted for publication August 4, 2005.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Butler AE, Janson J, Bonner-Weir S, Ritzel R, Rizza RA, Butler PC 2003 ß-Cell deficit and increased ß-cell apoptosis in humans with type 2 diabetes. Diabetes 52:102–110[Abstract/Free Full Text]
  2. Gerich JE 2003 Contributions of insulin-resistance and insulin-secretory defects to the pathogenesis of type 2 diabetes mellitus. Mayo Clin Proc 78:447–456[Medline]
  3. Yoon KH, Ko SH, Cho JH, Lee JM, Ahn YB, Song KH, Yoo SJ, Kang MI, Cha BY, Lee KW, Son HY, Kang SK, Kim HS, Lee IK, Bonner-Weir S 2003 Selective ß-cell loss and {alpha}-cell expansion in patients with type 2 diabetes mellitus in Korea. J Clin Endocrinol Metab 88:2300–2308[Abstract/Free Full Text]
  4. Fajans SS, Bell GI, Polonsky KS 2001 Molecular mechanisms and clinical pathophysiology of maturity-onset diabetes of the young. N Engl J Med 345:971–980[Free Full Text]
  5. Steiner DF, Chan SJ, Welsh JM, Kwok SC 1985 Structure and evolution of the insulin gene. Annu Rev Genet 19:463–484[CrossRef][Medline]
  6. German MS, Wang J, Chadwick RB, Rutter WJ 1992 Synergistic activation of the insulin gene by a LIM-homeodomain protein and a basic helix-loop-helix protein: building a functional insulin minienhancer complex. Genes Dev 6:2165–2176[Abstract/Free Full Text]
  7. Ohneda K, Mirmira RG, Wang J, Johnson JD, German MS 2000 The homeodomain of PDX-1 mediates multiple protein-protein interactions in the formation of a transcriptional activation complex on the insulin promoter. Mol Cell Biol 20:900–911[Abstract/Free Full Text]
  8. Asahara H, Dutta S, Kao HY, Evans RM, Montminy M 1999 Pbx-Hox heterodimers recruit coactivator-corepressor complexes in an isoform-specific manner. Mol Cell Biol 19:8219–8225[Abstract/Free Full Text]
  9. Qiu Y, Guo M, Huang S, Stein R 2002 Insulin gene transcription is mediated by interactions between the p300 coactivator and PDX-1, BETA2, and E47. Mol Cell Biol 22:412–420[Abstract/Free Full Text]
  10. Stanojevic V, Habener JF, Thomas MK 2004 Pancreas duodenum homeobox-1 transcriptional activation requires interactions with p300. Endocrinology 145:2918–2928[Abstract/Free Full Text]
  11. Thomas MK, Yao KM, Tenser MS, Wong GG, Habener JF 1999 Bridge-1, a novel PDZ-domain coactivator of E2A-mediated regulation of insulin gene transcription. Mol Cell Biol 19:8492–8404[Abstract/Free Full Text]
  12. Stanojevic V, Yao KM, Thomas MK 2005 The coactivator Bridge-1 increases transcriptional activation by pancreas duodenum homeobox-1 (PDX-1). Mol Cell Endocrinol 237:67–74[CrossRef][Medline]
  13. Stoffers DA, Heller RS, Miller CP, Habener JF 1999 Developmental expression of the homeodomain protein IDX-1 in mice transgenic for an IDX-1 promoter/lacZ transcriptional reporter. Endocrinology 140:5374–5381[Abstract/Free Full Text]
  14. Gannon M, Gamer LW, Wright CV 2001 Regulatory regions driving developmental and tissue-specific expression of the essential pancreatic gene pdx1. Dev Biol 238:185–201[CrossRef][Medline]
  15. Sander M, Sussel L, Conners J, Scheel D, Kalamaras J, Dela Cruz F, Schwitzgebel V, Hayes-Jordan A, German M 2000 Homeobox gene Nkx6.1 lies downstream of Nkx2.2 in the major pathway of ß-cell formation in the pancreas. Development 127:5533–5540[Abstract]
  16. Thomas MK, Habener JF 2004 IDX-1: pancreatic agenesis and type 2 diabetes. In: Erikson P, Epstein CJ, Wynshaw-Boris A, eds. Molecular basis of inborn errors of development. Oxford, UK: Oxford University Press; 552–556
  17. Sharma S, Jhala US, Johnson T, Ferreri K, Leonard J, Montminy M 1997 Hormonal regulation of an islet-specific enhancer in the pancreatic homeobox gene STF-1. Mol Cell Biol 17:2598–2604[Abstract]
  18. Wu KL, Gannon M, Peshavaria M, Offield MF, Henderson E, Ray M, Marks A, Gamer LW, Wright CV, Stein R 1997 Hepatocyte nuclear factor 3ß is involved in pancreatic ß-cell-specific transcription of the pdx-1 gene. Mol Cell Biol 17:6002–6013[Abstract]
  19. Lee CS, Sund NJ, Vatamaniuk MZ, Matschinsky FM, Stoffers DA, Kaestner KH 2002 Foxa2 controls Pdx1 gene expression in pancreatic ß-cells in vivo. Diabetes 51:2546–2551[Abstract/Free Full Text]
  20. Kitamura T, Nakae J, Kitamura Y, Kido Y, Biggs 3rd WH, Wright CV, White MF, Arden KC, Accili D 2002 The forkhead transcription factor Foxo1 links insulin signaling to Pdx1 regulation of pancreatic ß cell growth. J Clin Invest 110:1839–847[CrossRef][Medline]
  21. Olbrot M, Rud J, Moss LG, Sharma A 2002 Identification of ß-cell-specific insulin gene transcription factor RIPE3b1 as mammalian MafA. Proc Natl Acad Sci USA 99:6737–6742[Abstract/Free Full Text]
  22. Matsuoka TA, Zhao L, Artner I, Jarrett HW, Friedman D, Means A, Stein R 2003 Members of the large Maf transcription family regulate insulin gene transcription in islet ß cells. Mol Cell Biol 23:6049–6062[Abstract/Free Full Text]
  23. Hussain MA, Lee J, Miller CP, Habener JF 1997 POU domain transcription factor brain 4 confers pancreatic {alpha}-cell-specific expression of the proglucagon gene through interaction with a novel proximal promoter G1 element. Mol Cell Biol 17:7186–7194[Abstract]
  24. Collombat P, Mansouri A, Hecksher-Sorensen J, Serup P, Krull J, Gradwohl G, Gruss P 2003 Opposing actions of Arx and Pax4 in endocrine pancreas development. Genes Dev 17:2591–2603[Abstract/Free Full Text]
  25. Apelqvist A, Li H, Sommer L, Beatus P, Anderson DJ, Honjo T, Hrabe de Angelis M, Lendahl U, Edlund H 1999 Notch signalling controls pancreatic cell differentiation. Nature 400:877–881[CrossRef][Medline]
  26. Gradwohl G, Dierich A, LeMeur M, Guillemot F 2000 Neurogenin3 is required for the development of the four endocrine cell lineages of the pancreas. Proc Natl Acad Sci USA 97:1607–1611[Abstract/Free Full Text]
  27. Huang HP, Liu M, El-Hodiri HM, Chu K, Jamrich M, Tsai MJ 2000 Regulation of the pancreatic islet-specific gene BETA2 (neuroD) by neurogenin 3. Mol Cell Biol 20:3292–3307[Abstract/Free Full Text]
  28. German MS, Wang J 1994 The insulin gene contains multiple transcriptional elements that respond to glucose. Mol Cell Biol 14:4067–4075[Abstract/Free Full Text]
  29. Marshak S, Totary H, Cerasi E, Melloul D 1996 Purification of the ß-cell glucose-sensitive factor that transactivates the insulin gene differentially in normal and transformed islet cells. Proc Natl Acad Sci USA 93:15057–15062[Abstract/Free Full Text]
  30. Seijffers R, Ben-David O, Cohen Y, Karasik A, Berezin M, Newgard CB, Ferber S 1999 Increase in PDX-1 levels suppresses insulin gene expression in RIN 1046–38 cells. Endocrinology 140:3311–3317[Abstract/Free Full Text]
  31. Heremans Y, Van De Casteele M, in’t Veld P, Gradwohl G, Serup P, Madsen O, Pipeleers D, Heimberg H 2002 Recapitulation of embryonic neuroendocrine differentiation in adult human pancreatic duct cells expressing neurogenin 3. J Cell Biol 159:303–312[Abstract/Free Full Text]
  32. Ahlgren U, Jonsson J, Jonsson L, Simu K, Edlund H 1998 ß-Cell-specific inactivation of the mouse Ipf1/Pdx1 gene results in loss of the ß-cell phenotype and maturity onset diabetes. Genes Dev 12:1763–1768[Abstract/Free Full Text]
  33. Yoon JC, Xu G, Deeney JT, Yang SN, Rhee J, Puigserver P, Levens AR, Yang R, Zhang CY, Lowell BB, Berggren PO, Newgard CB, Bonner-Weir S, Weir G, Spiegelman BM 2003 Suppression of ß cell energy metabolism and insulin release by PGC-1{alpha}. Dev Cell 5:73–83[CrossRef][Medline]
  34. Thomas MK, Devon ON, Lee JH, Peter A, Schlosser DA, Tenser MS, Habener JF 2001 Development of diabetes mellitus in aging transgenic mice following suppression of pancreatic homeoprotein IDX-1. J Clin Invest 108:319–329[CrossRef][Medline]
  35. Schneider S, Buchert M, Georgiev O, Catimel B, Halford M, Stacker SA, Baechi T, Moelling K, Hovens CM 1999 Mutagenesis and selection of PDZ domains that bind new protein targets. Nat Biotechnol 17:170–175[CrossRef][Medline]
  36. Songyang Z, Fanning AS, Fu C, Xu J, Marfatia SM, Chishti AH, Crompton A, Chan AC, Anderson JM, Cantley LC 1997 Recognition of unique carboxyl-terminal motifs by distinct PDZ domains. Science 275:73–77[Abstract/Free Full Text]
  37. Harris BZ, Lim WA 2001 Mechanism and role of PDZ domains in signaling complex assembly. J Cell Sci 114:3219–3231[Medline]
  38. Sato T, Irie S, Kitada S, Reed JC 1995 FAP-1: a protein tyrosine phosphatase that associates with Fas. Science 268:411–415[Abstract/Free Full Text]
  39. Mao X, Seidlitz E, Truant R, Hitt M, Ghosh HP 2004 Re-expression of TSLC1 in a non-small-cell lung cancer cell line induces apoptosis and inhibits tumor growth. Oncogene 23:5632–5642[CrossRef][Medline]
  40. Bruning JC, Winnay J, Bonner-Weir S, Taylor SI, Accili D, Kahn CR 1997 Development of a novel polygenic model of NIDDM in mice heterozygous for IR and IRS-1 null alleles. Cell 88:561–572[CrossRef][Medline]
  41. Baba T, Shimizu T, Suzuki Y, Ogawara M, Isono K, Koseki H, Kurosawa H, Shirasawa T 2005 Estrogen, insulin, and dietary signals cooperatively regulate longevity signals to enhance resistance to oxidative stress in mice. J Biol Chem 280:16417–16426[Abstract/Free Full Text]
  42. Merimee TJ, Tyson JE 1974 Stabilization of plasma glucose during fasting; Normal variations in two separate studies. N Engl J Med 291:1275–1278[Medline]
  43. Amiel SA, Maran A, Powrie JK, Umpleby AM, Macdonald IA 1993 Gender differences in counterregulation to hypoglycaemia. Diabetologia 36:460–464[CrossRef][Medline]
  44. Davis SN, Galassetti P, Wasserman DH, Tate D 2000 Effects of gender on neuroendocrine and metabolic counterregulatory responses to exercise in normal man. J Clin Endocrinol Metab 85:224–230[Abstract/Free Full Text]
  45. Nelson WJ, Nusse R 2004 Convergence of Wnt, ß-catenin, and cadherin pathways. Science 303:1483–1487