Regulation of beta-cell glucose transporter gene expression

Mixed-Lineage Leukaemia Gene Regulates Glucose-Sensitive Gene Expression and Insulin Secretion in Pancreatic Beta Cells

The escalating prevalence of diabetes mellitus underscores the need for a comprehensive understanding of pancreatic beta cell function. Interest in glucose effectiveness has prompted the exploration of novel regulatory factors. The myeloid/lymphoid or mixed-lineage leukaemia gene (MLL) is widely recognised for its role in leukemogenesis and nuclear regulatory mechanisms through its histone methyltransferase activity in active chromatin. However, its function within pancreatic endocrine tissues remains elusive. Herein, we unveil a novel role of MLL in glucose metabolism and insulin secretion. MLL knockdown in βHC-9 pancreatic beta cells diminished insulin secretion in response to glucose loading, paralleled by the downregulation of the glucose-sensitive genes SLC2a1 and SLC2a2. Similar observations were made in MLL heterozygous knockout mice (MLL+/-), which exhibited impaired glucose tolerance and reduced insulin secretion without morphological anomalies in pancreatic endocrine cells. The reduction in insulin secretion was independent of changes in beta cell mass or insulin granule morphology, suggesting the regulatory role of MLL in glucose-sensitive gene expression. The current results suggest that MLL interacts with circadian-related complexes to modulate the expression of glucose transporter genes, thereby regulating glucose sensing and insulin secretion. Our findings shed light on insulin secretion control, providing potential avenues for therapeutics against diabetes.

Isolation of feline islets of Langerhans by selective osmotic shock produces glucose responsive islets

Introduction

Pancreatic islet isolation is essential for studying islet physiology, pathology, and transplantation, and feline islets could be an important model for human type II diabetes mellitus (T2D). Traditional isolation methods utilizing collagenases inflict damage and, in cats, may contribute to the difficulty in generating functional islets, as demonstrated by glucose-stimulated insulin secretion (GSIS). GLUT2 expression in β cells may allow for adaptation to hyperosmolar glucose solutions while exocrine tissue is selectively disrupted.

Methods

Here we developed a protocol for selective osmotic shock (SOS) for feline islet isolation and evaluated the effect of different hyperosmolar glucose concentrations (300 mmol/L and 600 mmol/L) and incubation times (20 min and 40 min) on purity, morphology, yield, and GSIS.

Results

Across protocol treatments, islet yield was moderate and morphology excellent. The treatment of 600 mmol/L glucose solution with 20 min incubation resulted in the highest stimulation index by GSIS.

Discussion

Glucose responsiveness was demonstrated, permitting future in vitro studies. This research opens avenues for understanding feline islet function and transplantation possibilities and enables an additional islet model for T2D.

The mitochondrial pyruvate carrier at the crossroads of intermediary metabolism

Pyruvate metabolism, a central nexus of carbon homeostasis, is an evolutionarily conserved process and aberrant pyruvate metabolism is associated with and contributes to numerous human metabolic disorders including diabetes, cancer, and heart disease. As a product of glycolysis, pyruvate is primarily generated in the cytosol before being transported into the mitochondrion for further metabolism. Pyruvate entry into the mitochondrial matrix is a critical step for efficient generation of reducing equivalents and ATP and for the biosynthesis of glucose, fatty acids, and amino acids from pyruvate. However, for many years, the identity of the carrier protein(s) that transported pyruvate into the mitochondrial matrix remained a mystery. In 2012, the molecular-genetic identification of the mitochondrial pyruvate carrier (MPC), a heterodimeric complex composed of protein subunits MPC1 and MPC2, enabled studies that shed light on the many metabolic and physiological processes regulated by pyruvate metabolism. A better understanding of the mechanisms regulating pyruvate transport and the processes affected by pyruvate metabolism may enable novel therapeutics to modulate mitochondrial pyruvate flux to treat a variety of disorders. Herein, we review our current knowledge of the MPC, discuss recent advances in the understanding of mitochondrial pyruvate metabolism in various tissue and cell types, and address some of the outstanding questions relevant to this field.

Insulinoma induces a hyperinsulinemia-mediated decrease of GLUT2 and GLP1 receptor in normal pancreatic β-cells

There have been several clinical reports of transient postoperative hyperglycemia in patients with insulinoma, but the effect of insulinoma on normal β-cells has not been investigated. We examined the glucose transporter 2 (GLUT2) and glucagon-like peptide 1 receptor (GLP1R) expression in normal pancreatic β-cells of five patients with insulinoma and five patients with normal glucose tolerance (NGT) as controls. The positive rate of GLUT2-or GLP1R-positive islets in the nontumor area was calculated by the ratio with the analyzed islets. For functional in vitro analyses, q-PCR and Western blotting were performed after insulin loading on MIN6 cells. The expression rates of both GLUT2 and GLP1R were significantly lower in nontumor area islets of insulinoma patients than in patients with NGT (GLUT2: 31.6 ± 15.3% vs 95.9 ± 6.7%, p < 0.01, GLP1R: 66.8 ± 15.0% vs 96.7 ± 5.0%, p < 0.01). Exposure of MIN6 cells to high concentrations of insulin resulted in a significant decrease in GLUT2 protein for 12 h and GLP1R protein for 24 h (GLUT2; 1.00 ± 0.079 vs 0.81 ± 0.04. p = 0.02, GLP1R; 1.00 ± 0.10 vs 0.50 ± 0.24, p = 0.03) but not in those mRNAs. Our findings show that insulinoma is associated with the downregulation of GLUT2 and GLP1R expression in nontumor area islets. These phenomena may be caused by high levels of insulin.

Glucose transporters in pancreatic islets

The fine-tuning of glucose uptake mechanisms is rendered by various glucose transporters with distinct transport characteristics. In the pancreatic islet, facilitative diffusion glucose transporters (GLUTs), and sodium-glucose cotransporters (SGLTs) contribute to glucose uptake and represent important components in the glucose-stimulated hormone release from endocrine cells, therefore playing a crucial role in blood glucose homeostasis. This review summarizes the current knowledge about cell type-specific expression profiles as well as proven and putative functions of distinct GLUT and SGLT family members in the human and rodent pancreatic islet and further discusses their possible involvement in onset and progression of diabetes mellitus. In context of GLUTs, we focus on GLUT2, characterizing the main glucose transporter in insulin-secreting β-cells in rodents. In addition, we discuss recent data proposing that other GLUT family members, namely GLUT1 and GLUT3, render this task in humans. Finally, we summarize latest information about SGLT1 and SGLT2 as representatives of the SGLT family that have been reported to be expressed predominantly in the α-cell population with a suggested functional role in the regulation of glucagon release.

Increase of Calcium Sensing Receptor Expression Is Related to Compensatory Insulin Secretion during Aging in Mice

Type 2 diabetes is caused by both insulin resistance and relative insulin deficiency. To investigate age-related changes in glucose metabolism and development of type 2 diabetes, we compared glucose homeostasis in different groups of C57BL/6J mice ranging in age from 4 months to 20 months (4, 8, 12, 16 and 20 months). Interestingly, we observed that non-fasting glucose levels were not significantly changed, but glucose tolerance gradually increased by 20 months of age, whereas insulin sensitivity declined with age. We found that the size of islets and glucose-stimulated insulin secretion increased with aging. However, mRNA expression of pancreatic and duodenal homeobox 1 and granuphilin was decreased in islets of older mice compared with that of 4-month-old mice. Serum calcium (Ca²⁺) levels were significantly decreased at 12, 20 and 28 months of age compared with 4 months and calcium sensing receptor (CaSR) mRNA expression in the islets significantly increased with age. An extracellular calcium depletion agent upregulated CaSR mRNA expression and consequently enhanced insulin secretion in INS-1 cells and mouse islets. In conclusion, we suggest that decreased Ca²⁺ levels and increased CaSR expression might be involved in increased insulin secretion to compensate for insulin resistance in aged mice.

Alternate Immersion in an External Glucose Solution Differentially Affects Blood Sugar Values in Older Versus Younger Zebrafish Adults

Recently, zebrafish have been used to examine hyperglycemia-induced complications (retinopathy and neuropathy), as would occur in individuals with diabetes. Current models to induce hyperglycemia in zebrafish include glucose immersion and streptozotocin injections. Both are effective, although neither is reported to elevate blood sugar values for more than 1 month. In this article, we report differences in hyperglycemia induction and maintenance in young (4-11 months) versus old (1-3 years) zebrafish adults. In particular, older fish immersed in an alternating constant external glucose solution (2%) for 2 months displayed elevated blood glucose levels for the entire experimental duration. In contrast, younger adults displayed only transient hyperglycemia, suggesting the fish were acclimating to the glucose exposure protocol. However, modifying the immersion protocol to include a stepwise increasing glucose concentration (from 1% → 2%→3%) resulted in maintained hyperglycemia in younger zebrafish adults for up to 2 months. Glucose-exposed younger fish collected after 8 weeks of exposure also displayed a significant decrease in wet weight. Taken together, these data suggest different susceptibilities to hyperglycemia in older and younger fish and that stepwise increasing glucose concentrations of 1% are required for maintenance of hyperglycemia in younger adults, with higher concentrations of glucose resulting in greater increases in blood sugar values.

FADD phosphorylation impaired islet morphology and function

Previous studies have indicated that Fas-FasL pathway and its downstream caspase-8 can regulate islet mass and insulin secretion. As a classical adaptor in Fas-FasL signaling, Fas-associated death domain-containing protein (FADD) takes part in many non-apoptosis processes regulated by its phosphorylation. However, its role in islets has not been evaluated to date. Here, through comparative proteomics and bioinformatic analysis on FADD phosphorylated (FADD-D) and wild-type (WT) MEFs, we found three proteins involved in islet differentiation and function were dysregulated due to FADD phosphorylation. The mouse model of FADD-D, which mimics constitutive phosphorylated FADD expression in mice, was further analyzed to address this issue. We confirmed the proteomic results by immunohistological analyses on pancreatic islets. In addition, we found that FADD-D mice displayed decreased islet area, and the glucose stimulated insulin secretion (GSIS) of FADD-D islets was impaired. These data suggest a novel role of FADD in islet development and insulin secretion.

Maternal malnutrition programs pancreatic islet mitochondrial dysfunction in the adult offspring

Accumulating evidence has shown that maternal malnutrition increases the risk of metabolic disease in the progeny. We previously reported that prenatal exposure to a low-protein diet (LP) leads to mitochondrial dysfunction in pancreatic islets from adult rodent offspring that could relate physiological and cellular alterations due to early diet. We aim to determine whether mitochondrial dysfunction could be a common consequence of prenatal nutritional unbalances. Pregnant Wistar rats received either a global food restriction (GFR), consisting in the reduction by 50% of the normal daily food intake, or a high-fat diet (HF) throughout gestation. GFR or HF diet during pregnancy leads to a lack of increase in insulin release and ATP content in response to glucose stimulation in islets from 3-month-old male and female offspring. These similar consequences originated from impairment in either glucose sensing or glucose metabolism, depending on the type of early malnutrition and on the sex of the progeny. Indeed, the glucose transport across β-cell membrane seemed compromised in female HF offspring, since GLUT-2 gene was markedly underexpressed. Additionally, for each progeny, consequences downstream the entry of glucose were also apparent. Expression of genes involved in glycolysis, TCA cycle and oxidative phosphorylations was altered in GFR and HF rats in a sex- and diet-dependent manner. Moreover, prenatal malnutrition affected the regulators of mitochondrial biogenesis, namely, PPAR coactivator 1 alpha (PGC-1α), since its expression was higher in islets from GFR rats. In conclusion, programming of mitochondrial dysfunction is a consequence of maternal malnutrition, which may predispose to glucose intolerance in the adult offspring.

Transcriptional regulation of glucose sensors in pancreatic β-cells and liver: an update

Pancreatic β-cells and the liver play a key role in glucose homeostasis. After a meal or in a state of hyperglycemia, glucose is transported into the β-cells or hepatocytes where it is metabolized. In the β-cells, glucose is metabolized to increase the ATP:ADP ratio, resulting in the secretion of insulin stored in the vesicle. In the hepatocytes, glucose is metabolized to CO(2), fatty acids or stored as glycogen. In these cells, solute carrier family 2 (SLC2A2) and glucokinase play a key role in sensing and uptaking glucose. Dysfunction of these proteins results in the hyperglycemia which is one of the characteristics of type 2 diabetes mellitus (T2DM). Thus, studies on the molecular mechanisms of their transcriptional regulations are important in understanding pathogenesis and combating T2DM. In this paper, we will review a recent update on the progress of gene regulation of glucose sensors in the liver and β-cells.

Glucose represses PPARα gene expression via AMP-activated protein kinase but not via p38 mitogen-activated protein kinase in the pancreatic β-cell

BACKGROUND

Peroxisome proliferator-activated receptor α (PPARα) regulates the expression of fatty acid metabolism genes and is thought to play a role in the regulation of insulin secretion and lipid detoxification. We have examined the mechanism whereby glucose decreases PPARα gene expression in the pancreatic β-cell.

METHODS

INS832/13 β-cell and isolated rat islets were incubated at 3 and 20 mM glucose for 18 h in the absence or presence of adenosine monophosphate (AMP)-activated protein kinase (AMPK) activators and inhibitors, as well as p38 mitogen-activated protein kinase (p38 MAPK) inhibitors. In another set of experiments, INS832/13 were infected with an adenovirus expressing a dominant-negative form of AMPK. PPARα expression levels were measured by reverse transcription polymerase chain reaction and Western blot.

RESULTS

Elevated glucose reduced the abundance of the PPARα transcript and protein, and its target genes acyl-coenzyme A (CoA) oxidase (ACO) and uncoupling protein 2 (UCP-2) in INS832/13 β-cell and isolated rat islets. Glucose reduced AMPK activity, while the AMPK activators 5-amino-4-imidazolecarboxamide riboside and metformin increased PPARα expression and suppressed the action of glucose. By contrast, the AMPK inhibitor compound C mimicked the glucose effect. A dominant negative form of AMPKα reduced the PPARα, ACO and UCP-2 transcripts to the same extent as elevated glucose. Pharmacological evidence indicated that glucose-regulated PPARα expression does not involve p38 MAPK, a target of AMPK in several cell types.

CONCLUSIONS

The results indicate that glucose represses PPARα gene expression via AMPK, but not via p38 MAPK in the β-cell.

Long-term exposure to genistein improves insulin secretory function of pancreatic beta-cells

We recently found that genistein, a plant-derived natural compound, is a novel cAMP signaling agonist in pancreatic beta-cells. In the present study, we further show that exposure of clonal insulin secreting (INS-1E) cells to genistein for 48 h enhanced glucose-stimulated insulin secretion (GSIS), whereas insulin content was not altered, suggesting that genistein-enhanced GSIS is not due to a modulation of insulin synthesis. This genistein effect is protein tyrosine kinase- and K(ATP) channel-independent. In addition, genistein had no effect on glucose transporter-2 expression or cellular ATP production, but similarly augmented pyruvate-stimulated insulin secretion in INS-1E cells, indicating that the improvement of insulin secretory function by long-term genistein exposure is not related to an alternation in glucose uptake or the glycolytic pathway. The enhanced insulin secretion by genistein was associated with elevated intracellular Ca(2+) concentration and dependent on protein kinase A and new protein synthesis as this effect was completely blocked by N-[2-(p-Bromocinnamylamino)ethyl]-5-isoquinolinesulfonamide or cycloheximide. Similarly, 48 h of genistein exposure also enhanced GSIS in freshly isolated mouse and human pancreatic islets, suggesting a non-species-specific and biologically relevant effect. These findings provide evidence that genistein may be a novel bioactive compound that has an anti-diabetic effect by improving insulin secretion from pancreatic beta-cells.

Oscillatory control of insulin secretion

Pancreatic beta-cells possess an inherent ability to generate oscillatory signals that trigger insulin release. Coordination of the secretory activity among beta-cells results in pulsatile insulin secretion from the pancreas, which is considered important for the action of the hormone in the target tissues. This review focuses on the mechanisms underlying oscillatory control of insulin secretion at the level of the individual beta-cell. Recent studies have demonstrated that oscillations of the cytoplasmic Ca(2+) concentration are synchronized with oscillations in beta-cell metabolism, intracellular cAMP concentration, phospholipase C activity and plasma membrane phosphoinositide lipid concentrations. There are complex interdependencies between the different messengers and signalling pathways that contribute to amplitude regulation and shaping of the insulin secretory response to nutrient stimuli and neurohormonal modulators. Several of these pathways may be important pharmacological targets for improving pulsatile insulin secretion in type 2 diabetes.

Short-term food restriction and refeeding alter expression of genes likely involved in brain glucosensing

Several genes involved in glucosensing of the endocrine pancreas have been proposed to serve a similar function in the brain. These genes include the glucose transporter-2 (Glut-2) and glucokinase (GK). In addition, the glucagon-like peptide 1 receptor, which serves as a downstream signal modulator in pancreatic glucosensing and centrally alters feeding, is also of interest. We used quantitative real-time RT-PCR to measure changes in hypothalamic and brainstem Glut-2, GK, and Glp-1R expression of these genes induced by food restriction and refeeding. Sprague-Dawley rats were 50% food restricted for 1 day; one-half of the food-restricted rats were refed with chow for 1 hr before sacrifice. In both hypothalamus and brainstem, gene expression of Glut-2, GK, and Glp-1R was significantly lower in refed rats compared with food-restricted rats. The measures of gene expression in two feeding control groups (ad libitum and voluntarily overfed animals) were intermediate between the food-restricted and refed groups, but were not significantly different from each other. The results indicate that putative glucosensing (GK, Glut-2, and Glp-1R) gene expression in the hypothalamus and brainstem is reduced in response to food intake, depending on prior nutritional status.

Acute hyperglycemia alters the ability of the normal beta-cell to sense and respond to glucose

Impaired glucose tolerance (IGT) and non-insulin-dependent diabetes mellitus (NIDDM) are associated with an impaired ability of the beta-cell to sense and respond to small changes in plasma glucose. The aim of this study was to establish whether acute hyperglycemia per se plays a role in inducing this defect in beta-cell response. Seven healthy volunteers with no family history of NIDDM were studied on two occasions during a 12-h oscillatory glucose infusion with a periodicity of 144 min. Once, low-dose glucose was infused at a mean rate of 6 mg x kg(-1) x min(-1) and amplitude 33% above and below the mean rate, and, once, high-dose glucose was infused at 12 mg x kg(-1) x min(-1) and amplitude 16% above and below the mean rate. Mean glucose levels were significantly higher during the high-dose compared with the low-dose glucose infusion [9.5 +/- 0.8 vs. 6.8 +/- 0.2 mM (P < 0.01)], resulting in increased mean insulin secretion rates [ISRs; 469.1 +/- 43.8 vs. 268.4 +/- 29 pmol/min (P < 0.001)] and mean insulin levels [213.6 +/- 46 vs. 67.9 +/- 10.9 pmol/l (P < 0.008)]. Spectral analysis evaluates the regularity of oscillations in glucose, insulin secretion, and insulin at a predetermined frequency. Spectral power for glucose, ISR, and insulin was reduced during the high-dose glucose infusion [11.8 +/- 1.4 to 7.0 +/- 1.6 (P < 0.02), 7.6 +/- 1.5 to 3.2 +/- 0.5 (P < 0.04), and 10.5 +/- 1.6 to 4.6 +/- 0.7 (P < 0.01), respectively]. In conclusion, short-term infusion of high-dose glucose to obtain glucose levels similar to those previously seen in IGT subjects results in reduced spectral power for glucose, ISR, and insulin. The reduction in spectral power previously observed for ISR in IGT or NIDDM subjects may be due partly to hyperglycemia.

Identification of transacting factors responsible for the tissue-specific expression of human glucose transporter type 2 isoform gene. Cooperative role of hepatocyte nuclear factors 1alpha and 3beta

We investigated transacting factors binding to the cis-element important in tissue-specific expression of the human glucose transporter type 2 isoform (GLUT2) gene. By transient transfection assay, we determined that the 227-base pair fragment upstream of the ATG start site contained promoter activity and that the region from +87 to +132 (site C) was responsible for tissue-specific expression. DNase I footprinting and electrophoretic mobility shift assay indicated that site C contained one binding site for hepatocyte nuclear factor 1 (HNF1) and two binding sites for HNF3. The mutations at positions +101 and +103, which are considered to be critical in binding HNF1 and HNF3, resulted in a 53% decrease in promoter activity, whereas the mutation of the proximal HNF3 binding site (+115 and +117) reduced promoter activity by 28%. The mutations of these four sites resulted in marked decrease (70%) in promoter activity as well as diminished bindings of HNF1 and HNF3. A to G mutation, which causes conversion of the HNF1 and HNF3 binding sequence to the NF-Y binding site, resulted in a 22% decrease in promoter activity. We identified that both HNF1 and HNF3 function as transcriptional activators in GLUT2 gene expression. Coexpression of the pGL+74 (+74 to +301) construct with the HNF1alpha and HNF3beta expression vectors in NIH 3T3 cells showed the synergistic effect on GLUT2 promoter activity compared with the expression of HNF1alpha, HNF3beta, or a combination of HNF1beta and HNF3beta. These data suggest that HNF1alpha and HNF3beta may be the most important players in the tissue-specific expression of the human GLUT2 gene.

Characterization and partial purification of liver glucose transporter GLUT2

GLUT2, the major facilitative glucose transporter isoform expressed in hepatocytes, pancreatic beta-cells, and absorptive epithelial cells, is unique not only with its low affinity and broad substrate specificity as a glucose transporter, but also with its implied function as a glucose-sensor. As a first essential step toward structural and biochemical elucidation of these unique, GLUT2 functions, we describe here the differential solubilization and DEAE-column chromatography of rat hepatocyte GLUT2 protein and its reconstitution into liposomes. The reconstituted GLUT2 bound cytochalasin B in a saturable manner with an apparent dissociation constant (K(d)) of 2.3 x 10(-6) M and a total binding capacity (B(T)) of 8.1 nmol per mg protein. The binding was completely abolished by 2% mercury chloride, but not affected by cytochalasin E. Significantly, the binding was also not affected by 500 mM D-glucose or 3-O-methyl D-glucose (3OMG). The purified GLUT2 catalyzed mercury chloride-sensitive 3OMG uptake, and cytochalasin B inhibited this 3OMG uptake. The inhibition was dose-dependent with respect to cytochalasin B, but was independent of 3OMG concentrations. These findings demonstrate that our solubilized GLUT2 reconstituted in liposomes is at least 60% pure and functional, and that GLUT2 is indeed unique in that its cytochalasin B binding is not affected by its substrate (D-glucose) binding. Our partially purified GLUT2 reconstituted in vesicles will be useful in biochemical and structural elucidation of GLUT2 as a glucose transporter and as a possible glucose sensor.

Glucose transporters: structure, function and consequences of deficiency

There are two mechanisms for glucose transport across cell membranes. In the intestine and renal proximal tubule, glucose is transported against a concentration gradient by a secondary active transport mechanism in which glucose is cotransported with sodium ions. In all other cells, glucose transport is mediated by one or more of the members of the closely related GLUT family of glucose transporters. The pattern of expression of the GLUT transporters in different tissues is related to the different roles of glucose metabolism in different tissues. Primary defects in glucose transport all appear to be extremely rare and not all possible deficiencies have been identified. Deficiency of the secondary active sodium/glucose transporters result in glucose/galactose malabsorption or congenital renal glycosuria. GLUT1 deficiency produces a seizure disorder with low glucose concentration in cerebrospinal fluid and GLUT2 deficiency is the basis of the Fanconi-Bickel syndrome, which resembles type I glycogen storage disease.

The large intracytoplasmic loop of the glucose transporter GLUT2 is involved in glucose signaling in hepatic cells

The hypothesis that the glucose transporter GLUT2 can function as a protein mediating transcriptional glucose signaling was addressed. To divert the putative interacting proteins from a glucose signaling pathway, two intracytoplasmic domains of GLUT2, the C terminus and the large loop located between transmembrane domains 6 and 7, were transfected into mhAT3F hepatoma cells. Glucose-induced accumulation of two hepatic gene mRNAs (GLUT2 and L-pyruvate kinase) was specifically inhibited in cells transfected with the GLUT2 loop and not with the GLUT2 C terminus. The dual effects of glucose were dissociated in cells expressing the GLUT2 loop; in fact a normal glucose metabolism into glycogen occurred concomitantly with the inhibition of the glucose-induced transcription. This inhibition by the GLUT2 loop could be due to competitive binding of a protein that normally interacts with endogenous GLUT2. In addition, the GLUT2 loop, tagged with green fluorescent protein (GFP), was located within the nucleus, whereas the GFP and GFP-GLUT2 C-terminal proteins remained in the cytoplasm. In living cells, a fraction (50%) of the expressed GFP-GLUT2 loop translocated rapidly from the cytoplasm to the nucleus in response to high glucose concentration and conversely in the absence of glucose. We conclude that, via protein interactions with its large loop, GLUT2 may transduce a glucose signal from the plasma membrane to the nucleus.

Regulation of macrophage migration inhibitory factor (MIF) expression by glucose and insulin in adipocytes in vitro

BACKGROUND

It has been reported that macrophage migration inhibitory factor (MIF) stimulated insulin secretion from pancreatic islet beta-cells in an autocrine manner, which suggests its pivotal role in the glucose metabolism. According to this finding, we evaluated MIF expression in cultured adipocytes and epididymal fat pads of obese and diabetic rats to investigate its role in adipose tissue.

MATERIALS AND METHODS

The murine adipocyte cell line 3T3-L1 was used to examine MIF mRNA expression and production of MIF protein in response to various concentrations of glucose and insulin. Epididymal fat pads of Otsuka Long-Evans Tokushima fatty (OLETF) and Wistar fatty rats, animal models of obesity and diabetes, were subjected to Northern blot analysis to determine MIF mRNA levels.

RESULTS

MIF mRNA of 3T3-L1 adipocytes was up-regulated by costimulation with glucose and insulin. Intracellular MIF content was significantly increased by stimulation, whereas its content in the culture medium was decreased. When the cells were treated with cytochalasin B, MIF secretion in the medium was increased. Pioglitazone significantly increased MIF content in the culture medium of 3T3-L1 cells. However, MIF mRNA expression of both epididymal fat pads of OLETF and Wistar fatty rats was down-regulated despite a high plasma glucose level. The plasma MIF level of Wistar fatty rats was significantly increased by treatment with pioglitazone.

CONCLUSION

We show here that the intracellular glucose level is critical to determining the MIF mRNA level as well as its protein content in adipose tissue. MIF is known to play an important role in glucose metabolism as a positive regulator of insulin secretion. In this context, it is conceivable that MIF may affect the pathophysiology of obesity and diabetes.

Overexpression of leptin receptors in pancreatic islets of Zucker diabetic fatty rats restores GLUT-2, glucokinase, and glucose-stimulated insulin secretion

The high-Km glucose transporter, GLUT-2, and the high-Km hexokinase of beta cells, glucokinase (GK), are required for glucose-stimulated insulin secretion (GSIS). GLUT-2 expression in beta cells of Zucker diabetic fatty (ZDF) rats is profoundly reduced at the onset of beta-cell dysfunction of diabetes. Because ZDF rats are homozygous for a mutation in their leptin receptor (OB-R) gene and are therefore leptin-insensitive, we expressed the wild-type OB-R gene in diabetic islets by infusing a recombinant adenovirus (AdCMV-OB-Rb) to determine whether this reversed the abnormalities. Leptin induced a rise in phosphorylated STAT3, indicating that the transferred wild-type OB-R was functional. GLUT-2 protein rose 17-fold in AdCMV-OB-Rb-treated ZDF islets without leptin, and leptin caused no further rise. GK protein rose 7-fold without and 12-fold with leptin. Preproinsulin mRNA increased 64% without leptin and rose no further with leptin, but leptin was required to restore GSIS. Clofibrate and 9-cis-retinoic acid, the partner ligands for binding to peroxisome proliferator-activator receptor alpha (PPARalpha) and retinoid X receptor, up-regulated GLUT-2 expression in islets of normal rats, but not in ZDF rats, in which PPARalpha is very low. Because the fat content of islets of diabetic ZDF rats remains high unless they are treated with leptin, it appears that restoration of GSIS requires normalization of intracellular nutrient homeostasis, whereas up-regulation of GLUT-2 and GK is leptin-independent, requiring only high expression of OB-Rb.

The effect of intrauterine growth restriction upon fetal and postnatal hepatic glucose transporter and glucokinase proteins

Employing immunohistochemical and Western blot analyses, we investigated the cellular localization (22-d fetal and 14-d postnatal animals) and concentrations (22-d fetal to 21-d postnatal animals) of rat hepatic glucose transporters (Glut 1 and Glut 2) and glucokinase in response to development and uteroplacental insufficiency with IUGR. Glut 1, the predominant fetal hematopoietic cellular isoform, persisted in postnatal hematopoietic islands and was noted minimally in fetal hepatic cellular membranes. A approximately 40% extrauterine decline in Glut 1 levels paralleled the decline in hematopoietic cells. IUGR increased the fetal hepatic Glut 1 levels in parallel with an expanded hematopoietic cell mass (p < 0.05). In contrast, IUGR failed to alter the 2-fold increase in extrauterine Glut 2 concentrations (1-7-d postnatal animals), the isoform found in fetal and postnatal hepatocytic cell membranes. Glucokinase, the nuclear enzyme, increased 25% postnatally. IUGR caused a 16% increase in fetal glucokinase levels and a approximately 25% decline at postnatal d 1 (p < 0.05) without a comparable change in the hepatocytic cell number (92 +/- 6 versus 86 +/- 4). We conclude that hepatic Glut 1 concentrations reflect the extramedullary hematopoietic cellular mass, whereas extrauterine Glut 2 changes herald the need for enhanced flexibility in hepatocytic glucose transport with the initiation of food ingestion. The age-related alteration along with the IUGR-induced compensatory changes in the nuclear-mitochondrial glucokinase levels attributes a critical role for this enzyme in perinatal hepatocytic glucose homeostasis.

Regulation of the expression of lipogenic enzyme genes by carbohydrate

Diets high in simple carbohydrates and low in fats lead in the mammalian liver to induction of a set of enzymes involved in lipogenesis. This induction occurs, in part, through transcriptional mechanisms that lead to elevated levels of the mRNA for these enzymes. For most of the lipogenic enzymes, an increase in glucose metabolism is required to trigger the transcriptional response. The intracellular mediator of this signaling pathway is unknown, although evidence suggests either glucose-6-phosphate or xylulose-5-phosphate. Studies to map the regulatory sequences of lipogenic enzyme genes involved in the transcriptional response have been performed for the L-type pyruvate kinase, S14, and acetyl-coenzyme A carboxylase genes. These studies have identified the DNA sequences necessary to link the signal generated by carbohydrate metabolism to specific nuclear transcription factors.

Elevated plasma glucose 2 h postchallenge predicts defects in beta-cell function

Studies were performed in subjects with no known family history of diabetes, normoglycemic subjects who have first-degree relatives with non-insulin-dependent diabetes mellitus (NIDDM), and subjects with nondiagnostic oral glucose tolerance tests (NDX) or impaired glucose tolerance (IGT). Insulin sensitivity index (SI) was similar in all four groups. However, a number of defects in insulin secretion were seen in the NDX and IGT groups, including reduced first-phase insulin secretory responses in intravenous glucose in relation to the degree of insulin resistance, and reduced normalized spectral power of insulin secretion during oscillatory glucose infusion. The latter finding demonstrates a decreased ability of the beta-cell to detect and respond to the successive increases and decreases in glucose and therefore to be entrained by the exogenous glucose infusion. The ability of a low-dose glucose infusion to prime the insulin secretory response to a subsequent glucose stimulus was normal in subjects with IGT but reduced or absent in subjects with overt NIDDM. These studies demonstrate that a number of alterations in beta-cell function are detectable in nondiabetic first-degree relatives of subjects with NIDDM with mild elevations in the 2-h postchallenge glucose level, and these abnormalities antedate the onset of overt hyperglycemia and clinical diabetes.

Glucose transporter expression and functional role of hexokinase in insulin biosynthesis in mouse beta TC3 cells

It was previously reported that insulin biosynthesis in mouse beta TC3 cells was regulated by glucose (Nagamatsu, S., and D. F. Steiner. Endocrinology 130: 748-754, 1992). In the present study, we examined the effect of glucose on the glucose transporter expression and hexokinase activities and determined the relationship between them and glucose-stimulated insulin biosynthesis in beta TC3 cells. Reverse transcriptase-polymerase chain reaction and Northern blot analysis revealed that beta TC3 cells expressed GLUT-1 and GLUT-3 glucose transporter mRNAs, but not GLUT-2. The levels of GLUT-1 and GLUT-3 mRNAs were not affected by glucose (0 or 11 mM glucose) over a period of 48 h. Immunoprecipitation of metabolically labeled beta TC3 cells with specific antibodies against GLUT-1 or GLUT-3 proteins revealed no effect of glucose on the biosynthesis of glucose transporters. Hexokinase [low Michaelis constant (Km) hexokinase] activity from cells incubated in 11 mM glucose for 48 h increased nearly twofold compared with cells maintained in 0 mM glucose, although the amount of cellular hexokinase protein detected by immunoblot analysis was unchanged between 0 and 11 mM glucose conditions. Glucokinase (high Km hexokinase) activity, in contrast, was not affected by glucose. Preincubation of beta TC3 cells with 2-deoxyglucose to inhibit hexokinase, thereby inhibiting all glycolysis, resulted in the decrease of glucose-stimulated insulin biosynthesis. Thus, in mouse beta TC3 cells that do not express GLUT-2, there is a close relationship between hexokinase activity and glucose-stimulated insulin biosynthesis, but not between the glucose transporter and glucose-stimulated insulin biosynthesis.

Mathematical analysis of a proposed mechanism for oscillatory insulin secretion in perifused HIT-15 cells

Oscillatory secretion of insulin has been observed in many different experimental preparations ranging from pancreatic islets to the whole pancreas. Here we examine the mathematical features underlying a possible model for oscillatory secretion from the perifused, insulin-secreting cell line, HIT-15. The model includes the kinetics of uptake of glucose by GLUT transporters, the rate of glucose metabolism within the cell, and the effect of glucose on the rate of insulin secretion. Putative feedback by insulin on the rate of glucose transport into the cells is treated phenomenologically and leads to insulin oscillations similar to those observed experimentally in HIT cells. The resulting set of ordinary differential equations is simplified by time-scale analysis to a two-variable set of ordinary differential equations. Because of this simplification we can explore, in great detail, the characteristics of the oscillations and their sensitivity to parameter variation using phase plane analysis.

Effects of glucose refeeding and glibenclamide treatment on glucokinase and GLUT2 gene expression in pancreatic B-cells and liver from rats

The mutual role of glucose and insulin in the regulation of glucokinase and GLUT2 glucose transporter gene expression in pancreatic B-cells and liver has been studied in vivo in the rat. Glucokinase mRNA was quantified by competitive reverse-transcriptase PCR analysis, and GLUT2 mRNA by Northern-blot analysis in total RNA fractions. As in the liver, glucokinase mRNA decreased by 64% in pancreatic B-cells after starvation for 2 days and was induced 3-fold by short-term treatment (1 h) of the rats with oral glucose (4 g/kg body wt.). In contrast the sulphonylurea compound glibenclamide (0.1 mg/kg body wt.) did not significantly stimulate glucokinase gene expression in pancreatic B-cells. But glibenclamide caused a 4-fold increase of glucokinase mRNA in liver which was abolished by concomitant administration of diazoxide, a drug which antagonizes glibenclamide stimulated insulin secretion. GLUT2 gene expression was decreased by 50% in pancreatic B-cells and liver after starvation of the rats for 2 days. Neither short-term treatment (1 h) with glucose nor glibenclamide resulted in a significant increase of GLUT2 gene expression in pancreatic B-cells and liver. The results suggest that it is glucose which stimulates glucokinase gene expression in pancreatic B-cells whereas the transcriptional regulation of the glucokinase gene in liver is directed by insulin.

Differences in glucose transporter gene expression between rat pancreatic alpha- and beta-cells are correlated to differences in glucose transport but not in glucose utilization

Glucose exerts inverse effects upon the secretory function of islet alpha- and beta-cells, suppressing glucagon release and increasing insulin release. This diverse action may result from differences in glucose transport and metabolism between the two cell types. The present study compares glucose transport in rat alpha- and beta-cells. beta-Cells transcribed GLUT2 and, to a lesser extent, GLUT 1; alpha-cells contained GLUT1 but no GLUT2 mRNA. No other GLUT-like sequences were found among cDNAs from alpha- or beta-cells. Both cell types expressed 43-kDa GLUT1 protein which was enhanced by culture. The 62-kDa beta-cell GLUT2 protein was converted to a 58-kDa protein after trypsin treatment of the cells without detectable consequences upon glucose transport kinetics. In beta-cells, the rates of glucose transport were 10-fold higher than in alpha-cells. In both cell types, glucose uptake exceeded the rates of glucose utilization by a factor of 10 or more. Glycolytic flux, measured as D-[5(3)H]glucose utilization, was comparable in alpha- and beta-cells between 1 and 10 mmol/liter substrate. In conclusion, differences in glucose transporter gene expression between alpha- and beta-cells can be correlated with differences in glucose transport kinetics but not with different glucose utilization rates.

Post-GLUT-2 defects in beta-cells of non-insulin-dependent diabetic obese rats

Zucker diabetic fatty (ZDF) rats develop non-insulin-dependent diabetes mellitus concomitantly with loss of glucose responsiveness and GLUT-2, the high-Michaelis constant glucose transporter of beta-cells. To determine the integrity of beta-cell glucose metabolism distal to the level of glucose transport and phosphorylation, we examined the insulin responses of isolated pancreata to 5, 10, and 20 mM D-glyceraldehyde and monomethylsuccinate, as well as to glucose. The insulin response of diabetic pancreata to glucose was 90% below the response prior to the onset of diabetes, whereas the responses to glyceraldehyde and succinate had declined to 65 and 44%, respectively, below the prediabetic responses. D-[14C]glyceraldehyde oxidation by diabetic islets was 74% below that of islets from lean nondiabetic controls. We conclude that 1) the insulin responses to glyceraldehyde and monomethylsuccinate, as well as to glucose, are impaired in the diabetes of ZDF rats and 2) the impairment of the glucose response was greater than that of the glyceraldehyde response, which was, in turn, greater than that of the monomethylsuccinate response; this decrescendo pattern of impairment is consistent with defects at multiple sites in glucose metabolism; if the defect were entirely due to a postmetabolic signaling defect, the impairment to glucose and its metabolites should be comparable.

The role of reduced glucose transporter content and glucose metabolism in the immature secretory responses of fetal rat pancreatic islets

Isolated fetal islets show an immature or poor secretory response to nutrient secretagogues which may result from impaired mitochondrial oxidative processes. Insulin secretion, glucose metabolism and detection of metabolic enzymes by radiolabelling and immunoprecipitation were compared in islets isolated from neonatal (aged 5 days) and fetal rats (at 20 days gestation). The insulin secretory dynamics of fetal islets were abnormal in response to stimulation by glucose (10 mmol/l); a rapid release of insulin reaching a maximum 6 min after stimulation was observed with no rising second phase release. However, when the data were expressed as percentage of islet insulin content released, fetal islets released significantly more insulin than neonatal islets in response to glucose (4.86 +/- 0.45% vs 1.81 +/- 0.62%, p < 0.01) or 100 nmol/l glibenclamide (2.49 +/- 0.17% vs 0.25 +/- 0.06%, p < 0.001). Fetal islets however, failed to release insulin in response to stimulation by glyceraldehyde (10 mmol/l) unlike neonatal islets. Both glucose utilisation (as measured by the formation of [3H] H2O from 5-[3H] glucose) and glucose oxidation (as measured by the formation of [14C] CO2 from U-[14C] glucose) did not increase significantly in response to increasing the medium glucose concentration to 10 mmol/l whereas in neonatal islets, glucose utilisation and glucose oxidation were significantly increased 2.5- and 2.7-fold, respectively. When islets were incubated with both radiolabelled glucoses simultaneously, the rate of glucose oxidation was shown to be directly proportional to the rate of glucose utilisation. The relationship between glucose utilisation and glucose oxidation was similar in fetal and neonatal islets.(ABSTRACT TRUNCATED AT 250 WORDS)

Facilitative glucose transporters

Facilitative glucose transport is mediated by members of the Glut protein family that belong to a much larger superfamily of 12 transmembrane segment transporters. Six members of the Glut family have been described thus far. These proteins are expressed in a tissue- and cell-specific manner and exhibit distinct kinetic and regulatory properties that reflect their specific functional roles. Glut1 is a widely expressed isoform that provides many cells with their basal glucose requirement. It also plays a special role in transporting glucose across epithelial and endothelial barrier tissues. Glut2 is a high-Km isoform expressed in hepatocytes, pancreatic beta cells, and the basolateral membranes of intestinal and renal epithelial cells. It acts as a high-capacity transport system to allow the uninhibited (non-rate-limiting) flux of glucose into or out of these cell types. Glut3 is a low-Km isoform responsible for glucose uptake into neurons. Glut4 is expressed exclusively in the insulin-sensitive tissues, fat and muscle. It is responsible for increased glucose disposal in these tissues in the postprandial state and is important in whole-body glucose homeostasis. Glut5 is a fructose transporter that is abundant in spermatozoa and the apical membrane of intestinal cells. Glut7 is the transporter present in the endoplasmic reticulum membrane that allows the flux of free glucose out of the lumen of this organelle after the action of glucose-6-phosphatase on glucose 6-phosphate. This review summarizes recent advances concerning the structure, function, and regulation of the Glut proteins.

The glucose transporter family: structure, function and tissue-specific expression

Images

GLUT-2 function in glucose-unresponsive beta cells of dexamethasone-induced diabetes in rats

Spontaneous and dexamethasone-induced noninsulin-dependent diabetes mellitus (NIDDM) in rats is associated with loss of glucose-stimulated insulin secretion (GSIS) and a reduction in both GLUT-2-positive beta cells and high Km glucose transport. To determine if the chronology and correlation of these abnormalities is consistent with a causal relationship, Zucker (fa/fa) rats were studied longitudinally before and during 10 d of dexamethasone-induced (0.4 mg/kg per d i.p.) NIDDM. Within 24 h of dexamethasone treatment blood glucose rose and GSIS declined, becoming paradoxically negative (-87 +/- 12 microU/ml per min) on day 10. Blood glucose was negatively correlated with GSIS (r = -0.92; P < 0.001). 3-0-methyl-D-glucose (3MG) transport was unchanged at 12 h, 23% below normal on day 1, and declined further to a nadir 59% below normal. The GLUT-2-positive beta cell area did not decline until 48 h, reaching a nadir of 35% of normal at 10 d. The area of GLUT-2-positive beta cells was correlated with GSIS (r = 0.77; P < 0.005). We conclude that the chronology and correlation between GSIS loss and hyperglycemia is consistent with a cause-effect relationship, but that the subtotal impairment in glucose transport by itself cannot explain the total loss of GSIS if one assumes that normal beta cells are functionally homogenous.

Interaction of facilitative glucose transporter with glucokinase and its modulation by ADP and glucose-6-phosphate

Bacterial glucokinase (GK) binds to purified, human erythrocyte glucose transporter (GT) reconstituted in vesicles. The binding is largely abolished if GT is predigested with trypsin, indicating that GK binds to the cytoplasmic domain of GT. The binding is a saturable function of GK concentration showing two distinct affinities with apparent KD of 0.33 and 5.1 microM. The binding is stimulated by an increasing concentration of ADP with the 50% maximal effect at 5 mM. Glucose-6-phosphate (G6P) also stimulates the binding with a distinct optimum at 25 mM. The binding is stimulated only slightly by ATP. D-glucose has no affect on the binding. KCl enhances the binding with the maximal effect at physiological intracellular concentrations. The binding is sensitive to changes in pH with an optimum at pH 4. The binding causes no detectable functional change in GT. However, the enzymatic activity of GK measured at nanomolar concentrations of GK is significantly greater in the presence of GT vesicles than in its absence or in the presence of protein-free vesicles, indicating that GK interacts with GT at this low concentration range with an apparent KD of 10 mM. Although its physiological significance is not known, the GK-GT interaction in vitro described here suggests that these two proteins may also interact in the cell and regulate carbohydrate metabolism.

Hypoglycemia but not hyperglycemia induces rapid changes in pancreatic beta-cell gene transcription

The purpose of these studies was to quantify several mRNAs expressed specifically in pancreatic islet cells and known or postulated to be important for insulin release after acute well defined alterations in levels of plasma glucose. Glucose levels were maintained at 50, 120, or 180 mg/dl (2.8, 6.7, or 10 mM) for 3 h in conscious unrestrained rats. Hypoglycemia (for 3 h) caused significant decreases in pancreatic content of mRNAs for insulin 2 and GLUT-2 to 55 and 34% of control values, respectively. There were no significant changes in insulin 1, amylin, glucokinase, or glucagon mRNAs. Unprocessed insulin 1 and 2 mRNA precursors were decreased to 17 and 10% of levels in controls, consistent with effects of short-term hypoglycemia on new mRNA synthesis. Hyperglycemia (for 3 h) caused no increase in pancreatic content of any mRNA measured. To discriminate between effects of hypoglycemia and hyperinsulinemia in the hypoglycemic animals, rats were made hypoglycemic by infusion with etomoxir, a carnitine palmitoyltransferase I inhibitor that lowers glucose in the fasted (glycogen-depleted) state by inhibiting hepatic gluconeogenesis. A single dose of this agent caused a decrease in glucose from 120 mg/dl (6.7 mM) to 80 mg/dl (4.4 mM) and significantly decreased insulin mRNA and pre-mRNA. These results are consistent with the hypothesis that glucose modulates islet cell gene transcription directly. They indicate that the range of glucose concentrations that modulate gene transcription differs from the levels of glucose that alter both insulin biosynthetic and secretion rates.

Skeletal muscle glucose transporter gene expression is not affected by injecting growth-hormone-secreting cells in young rats

To elucidate the diabetogenic effect of growth hormone on glucose metabolism the regulation of glucose transporter (GLUT) gene expression was examined in rat skeletal muscles. Female Wistar-Furth rats were implanted subcutaneously with growth-hormone-producing pituitary tumour (GH3) cells. Animals were killed 4 or 9 weeks after GH3 cell injection. Although body weight, serum growth hormone and insulin-like growth factor I levels were remarkably elevated during the 4-9 week period, serum blood glucose levels were within normal range. Muscles were obtained from the quadriceps muscle, diaphragm and heart, respectively. Northern blot analysis and Western blot analysis were performed using specific cDNA probes and antibodies. During the 4-9 week period, the levels of muscle GLUT1 and 4 mRNA (corrected by beta-actin mRNA level) in each muscle from the rats injected with tumour cells were not significantly different from those of control rats. Chronic elevation of growth hormone in these rats did not cause any change in GLUT 1 and 4 expression compared to the controls during the euglycaemic period. These results provide the first evidence that chronic growth hormone elevation itself does not affect a key gene of in vivo glucose metabolism.

Differential expression of rat pancreatic islet beta-cell glucose transporter (GLUT 2), proinsulin and islet amyloid polypeptide genes after prolonged fasting, insulin-induced hypoglycaemia and dexamethasone treatment

The question posed by these studies was whether chronic adaptive changes in glucose-stimulated insulin secretion are accompanied by comparable changes in islet Beta-cell glucose transporter (GLUT 2) gene expression. Control, fasted (3-day), insulin-injected hypoglycaemic (5-day), and dexamethasone-treated (4-day) rats (n = 5 for each condition), were studied. After fasting significant decrements in proinsulin mRNA/microgram RNA (-32%, p < 0.05) and islet amyloid polypeptide mRNA/microgram RNA (-44%, p < 0.05) were observed, while there was no change in GLUT 2 mRNA/microgram RNA (-13%, p > 0.05). After insulin-induced hypoglycaemia, decrements in proinsulin mRNA/microgram RNA (-49%, p < 0.01) and islet amyloid polypeptide mRNA/microgram RNA (-44%, p < 0.01) were also observed, with no change in islet GLUT 2 mRNA/microgram RNA (-18%, p > 0.05). Dexamethasone treatment resulted in a marked stimulatory effect on proinsulin mRNA/microgram RNA (+236%, p < 0.001) and islet amyloid polypeptide mRNA/microgram RNA (+221%, p < 0.01), while again there was no change in islet GLUT 2 mRNA/microgram RNA (+0.3%, p > 0.05). Quantitative immunoblot analysis with a GLUT 2 specific antibody revealed no change in islet GLUT 2 protein with fasting, but a small decrease (-39 +/- 11%) in islet GLUT 2/microgram protein after insulin-induced hypoglycaemia. These results do not support the hypothesis that chronic changes in glucose-stimulated insulin secretion are accompanied by changes in GLUT 2 expression. In contrast to the lack of correlation with GLUT 2, there was a striking correlation between proinsulin and islet amyloid polypeptide mRNAs for all experimental conditions (r = 0.974, p < 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)

Roles of insulin resistance and beta-cell dysfunction in dexamethasone-induced diabetes

The roles of insulin resistance and beta-cell dysfunction in glucocorticoid-induced diabetes were determined in Wistar and Zucker (fa/fa) rats. All Wistar rats treated with 5 mg/kg per d of dexamethasone for 24 d exhibited increased beta-cell mass and basal and arginine-stimulated insulin secretion, indicating insulin resistance, but only 16% became diabetic. The insulin response to 20 mM glucose was normal in the perfused pancreas of all normoglycemic dexamethasone-treated rats but absent in every diabetic rat. Immunostainable high Km beta-cell transporter, GLUT-2, was present in approximately 100% of beta-cells of normoglycemic rats, but in only 25% of beta cells of diabetic rats. GLUT-2 mRNA was not reduced. All Zucker (fa/fa) rats treated with 0.2-0.4 mg/kg per d of dexamethasone for 24 d became diabetic and glucose-stimulated insulin secretion was absent in all. High Km glucose transport in islets was 50% below nondiabetic controls. Only 25% of beta cells of diabetic rats were GLUT-2-positive compared with approximately 100% in controls. Total pancreatic GLUT-2 mRNA was increased twofold suggesting a posttranscriptional abnormality. We conclude that dexamethasone induces insulin resistance, whether or not it induces hyperglycemia. Whenever hyperglycemia is present, GLUT-2-positive beta cells are reduced, high Km glucose transport into beta cells is attenuated and the insulin response to glucose is absent.