The islet in non-insulin-dependent diabetes mellitus (NIDDM) is characterized by loss of ,B cells and large local deposits of amyloid derived from the 37-amino acid protein, islet amyloid polypeptide (IAPP Non-insulin-dependent diabetes mellitus (NIDDM) is characterized by (3-cell destruction and islet amyloid derived from islet amyloid polypeptide (IAPP) (1, 2). IAPP is a 37-amino acid protein that possesses amyloidogenic properties in species that spontaneously develop NIDDM (humans, monkeys, cats), but is non-amyloidogenic in mice that do not develop NIDDM (3, 4). Overexpression of human IAPP (h-IAPP), but not rat IAPP, in COS cells resulted in intracellular IAPP amyloidosis that was associated with cell death (5). Thus far, hemizygous transgenic mice for h-IAPP have not been reported to develop islet amyloid or diabetes mellitus spontaneously (6-8). Induction of marked insulin resistance in hemizygous mice transgenic for h-IAPP provokes intra-and extracellular IAPP amyloid formation, which is associated with 13-cell death and hyperglycemia (9).Based on these observations, we hypothesized that sufficiently increased rates of h-IAPP expression and synthesis results in intracellular IAPP amyloidosis and (3-cell death, which results in diabetes mellitus (10). To examine this further, we developed a homozygous line of mice transgenic for h-IAPP, thereby doubling the h-IAPP gene copy number. We report here that these mice spontaneously developed diabetes mellitus due to (3-cell death, which was associated with abnormal intra-and extracellular aggregates of h-IAPP. We conclude that overproduction of IAPP in vulnerable species (humans, monkeys, cats) may cause (3-cell destruction and diabetes mellitus. MATERIALS AND METHODSPreparation of Transgenic Construct. The RIPHAT transgene (2395 bp) is described elsewhere (9). It consists of a PCR-generated cDNA encompassing the h-IAPP coding sequence (270 bp) under the regulation of the rat insulin II promoter/5' untranslated region and followed by intron I (728 bp) from the human albumin gene and the polyadenylylation site/RNA termination region (525 bp) from the human glyceraldehyde-3-phosphate gene (GAPDH). Use of the albumin intron I and GAPDH polyadenylylation site in transgenic constructs has been described (11).Generation of Transgenic Mice. Hemizygotes of the RHF line described in Couce et al. (9) were self-crossed to generate Fl offspring. Transgenic offspring were identified by PCR amplification of RIPHAT from tail DNA. Hemizygotes were distinguished from homozygotes by backcross breeding to nontransgenic FVB/N mice. Homozygotes were defined as those mice that generated more than 20 transgenic and no nontransgenic offspring. Four such homozygotes were used to establish the core RHF breeding colony.Northern Blot Analysis. Total RNA was prepared from whole pancreata of FVB/N, RHF hemizygote, and RHF homozygote males and females. Gels and blots were prepared and hybridized as described (9)

Nondiabetic obese humans adapt to insulin resistance by increasing ␤-cell mass. In contrast, obese humans with type 2 diabetes have an ϳ60% deficit in ␤-cell mass. Recent studies in rodents reveal that ␤-cell mass is regulated, increasing in response to insulin resistance through increased ␤-cell supply (islet neogenesis and ␤-cell replication) and/or decreased ␤-cell loss (␤-cell apoptosis). Prospective studies of islet turnover are not possible in humans. In an attempt to establish the mechanism for the deficit in ␤-cell mass in type 2 diabetes, we used an obese versus lean murine transgenic model for human islet amyloid polypeptide (IAPP) that develops islet pathology comparable to that in humans with type 2 diabetes. By 40 weeks of age, obese nontransgenic mice did not develop diabetes and adapted to insulin resistance by a 9-fold increase (P < 0.001) in ␤-cell mass accomplished by a 1.7-fold increase in islet neogenesis (P < 0.05) and a 5-fold increase in ␤-cell replication per islet (P < 0.001). Obese transgenic mice developed midlife diabetes with islet amyloid and an 80% (P < 0.001) deficit in ␤-cell mass that was due to failure to adaptively increase ␤-cell mass. The mechanism subserving this failed expansion was a 10-fold increase in ␤-cell apoptosis (P < 0.001). There was no relationship between the extent of islet amyloid or the blood glucose concentration and the frequency of ␤-cell apoptosis. However, the frequency of ␤-cell apoptosis was related to the rate of increase of islet amyloid. These prospective studies suggest that the formation of islet amyloid rather than the islet amyloid per se is related to increased ␤-cell apoptosis in this murine model of type 2 diabetes. This finding is consistent with the hypothesis that soluble IAPP oligomers but not islet amyloid are responsible for increased ␤-cell apoptosis. The current studies also support the concept that replicating ␤-cells are more vulnerable to apoptosis, possibly accounting for the failure of ␤-cell mass to expand appropriately in response to obesity in type 2 diabetes. Diabetes 52:2304 -2314, 2003

The islet in type 2 diabetes is characterized by a deficit in ␤-cell mass, increased ␤-cell apoptosis, and impaired insulin secretion. Also, islets in type 2 diabetes often contain deposits of islet amyloid derived from islet amyloid polypeptide (IAPP), a 37-amino acid protein cosecreted with insulin by ␤-cells. Several lines of evidence suggest that proteins with a capacity to develop amyloid fibrils may also form small toxic oligomers that can initiate apoptosis. The amino acid sequence of IAPP in rats and mice is identical and differs from that in humans by substitution of proline residues in the amyloidogenic sequence so that the protein no longer forms amyloid fibrils or is cytotoxic. In the present study, we report a novel rat model for type 2 diabetes: rats transgenic for human IAPP (the HIP rat). HIP rats develop diabetes between 5 and 10 months of age, characterized by an ϳ60% deficit in ␤-cell mass that is due to an increased frequency of ␤-cell apoptosis. HIP rats develop islet amyloid, but the extent of amyloid was not related to the frequency of ␤-cell apoptosis (r ‫؍‬ 0.10, P ‫؍‬ 0.65), whereas the fasting blood glucose was (r ‫؍‬ 0.77, P < 0.001). The frequency of ␤-cell apoptosis was related to the frequency of ␤-cell replication (r ‫؍‬ 0.97, P < 0.001) in support of the hypothesis that replicating cells are more vulnerable to apoptosis than nondividing cells. The HIP rat provides additional evidence in support of the potential role of IAPP oligomer formation toward the increased frequency of apoptosis in type 2 diabetes, a process that appears to be compounded by glucose toxicity when hyperglycemia supervenes. Diabetes 53: 1509 -1516, 2004

An enzyme system that accurately initiates transcription of the engrailed gene has been prepared from Drosophila embryos. The system has been separated chromatographically into two fractions, both of which are required for specific engrailed transcription. DNase footprint and competition analysis detected at least two sequence-specific DNA-binding proteins in one of these two fractions. Together, these proteins bind to eight regions within 400 bp of the transcription initiation sites. Most of the regions containing these binding sites are required for manimal engrailed transcription in vitro. In addition, a region downstream from the initiation sites and within the first 40 residues of the transcription unit is essential for transcription. Transient in vivo expression assays indicated that these same upstream and downstream sequences are required for transcription in Drosophila tissue culture cells.

Facilitative glucose transporters exhibit variable hexose affinity and tissue-specific expression. These characteristics contribute to specialized metabolic properties of cells. Here we describe the characterization of a novel glucose transporter-like molecule, GLUT-12. GLUT-12 was identified in MCF-7 breast cancer cells by homology to the insulin-regulatable glucose transporter GLUT-4. The GLUT-12 cDNA encodes 617 amino acids, which possess features essential for sugar transport. Di-leucine motifs are present in NH(2) and COOH termini at positions similar to the GLUT-4 FQQI and LL targeting motifs. GLUT-12 exhibits 29% amino acid identity with GLUT-4 and 40% to the recently described GLUT-10. Like GLUT-10, a large extracellular domain is predicted between transmembrane domains 9 and 10. Genomic organization of GLUT-12 is highly conserved with GLUT-10 but distinct from GLUTs 1-5. Immunofluorescence showed that, in the absence of insulin, GLUT-12 is localized to the perinuclear region in MCF-7 cells. Immunoblotting demonstrated GLUT-12 expression in skeletal muscle, adipose tissue, and small intestine. Thus GLUT-12 is potentially part of a second insulin-responsive glucose transport system.