Mary E. Rabaglia

A gene expression network model of type 2 diabetes links cell cycle regulation in islets with diabetes susceptibility

Insulin resistance is necessary but not sufficient for the development of type 2 diabetes. Diabetes results when pancreatic beta-cells fail to compensate for insulin resistance by increasing insulin production through an expansion of beta-cell mass or increased insulin secretion. Communication between insulin target tissues and beta-cells may initiate this compensatory response. Correlated changes in gene expression between tissues can provide evidence for such intercellular communication. We profiled gene expression in six tissues of mice from an obesity-induced diabetes-resistant and a diabetes-susceptible strain before and after the onset of diabetes. We studied the correlation structure of mRNA abundance and identified 105 co-expression gene modules. We provide an interactive gene network model showing the correlation structure between the expression modules within and among the six tissues. This resource also provides a searchable database of gene expression profiles for all genes in six tissues in lean and obese diabetes-resistant and diabetes-susceptible mice, at 4 and 10 wk of age. A cell cycle regulatory module in islets predicts diabetes susceptibility. The module predicts islet replication; we found a strong correlation between (2)H(2)O incorporation into islet DNA in vivo and the expression pattern of the cell cycle module. This pattern is highly correlated with that of several individual genes in insulin target tissues, including Igf2, which has been shown to promote beta-cell proliferation, suggesting that these genes may provide a link between insulin resistance and beta-cell proliferation.

Positional cloning of Sorcs1, a type 2 diabetes quantitative trait locus.

We previously mapped the type 2 diabetes mellitus-2 locus (T2dm2), which affects fasting insulin levels, to distal chromosome 19 in a leptin-deficient obese F2 intercross derived from C57BL/6 (B6) and BTBR T+ tf/J (BTBR) mice. Introgression of a 7-Mb segment of the B6 chromosome 19 into the BTBR background (strain 1339A) replicated the reduced insulin linked to T2dm2. The 1339A mice have markedly impaired insulin secretion in vivo and disrupted islet morphology. We used subcongenic strains derived from 1339A to localize the T2dm2 quantitative trait locus (QTL) to a 242-kb segment comprising the promoter, first exon and most of the first intron of the Sorcs1 gene. This was the only gene in the 1339A strain for which we detected amino acid substitutions and expression level differences between mice carrying B6 and BTBR alleles of this insert, thereby identifying variation within the Sorcs1 gene as underlying the phenotype associated with the T2dm2 locus. SorCS1 binds platelet-derived growth factor, a growth factor crucial for pericyte recruitment to the microvasculature, and may thus have a role in expanding or maintaining the islet vasculature. Our identification of the Sorcs1 gene provides insight into the pathway underlying the pathophysiology of obesity-induced type 2 diabetes mellitus.

Loss of Stearoyl-CoA Desaturase-1 Improves Insulin Sensitivity in Lean Mice but Worsens Diabetes in Leptin-Deficient Obese Mice

The lipogenic gene stearoyl-CoA desaturase (SCD)1 appears to be a promising new target for obesity-related diabetes, as mice deficient in this enzyme are resistant to diet- and leptin deficiency–induced obesity. The BTBR mouse strain replicates many features of insulin resistance found in humans with excess visceral adiposity. Using the hyperinsulinemic-euglycemic clamp technique, we determined that insulin sensitivity was improved in heart, soleus muscle, adipose tissue, and liver of BTBR SCD1-deficient mice. We next determined whether SCD1 deficiency could prevent diabetes in leptin-deficient BTBR mice. Loss of SCD1 in leptinob/ob mice unexpectedly accelerated the progression to severe diabetes; 6-week fasting glucose increased ∼70%. In response to a glucose challenge, Scd1−/− leptinob/ob mice had insufficient insulin secretion, resulting in glucose intolerance. A morphologically distinct class of islets isolated from the Scd1−/− leptinob/ob mice had reduced insulin content and increased triglycerides, free fatty acids, esterified cholesterol, and free cholesterol and also a much higher content of saturated fatty acids. We believe the accumulation of lipid is due to an upregulation of lipoprotein lipase (20-fold) and Cd36 (167-fold) and downregulation of lipid oxidation genes in this class of islets. Therefore, although loss of Scd1 has beneficial effects on adiposity, this benefit may come at the expense of β-cells, resulting in an increased risk of diabetes.

Glucose- and GTP-dependent stimulation of the carboxyl methylation of CDC42 in rodent and human pancreatic islets and pure beta cells. Evidence for an essential role of GTP-binding proteins in nutrient-induced insulin secretion.

Several GTP-binding proteins (G-proteins) undergo post-translational modifications (isoprenylation and carboxyl methylation) in pancreatic beta cells. Herein, two of these were identified as CDC42 and rap 1, using Western blotting and immunoprecipitation. Confocal microscopic data indicated that CDC42 is localized only in islet endocrine cells but not in acinar cells of the pancreas. CDC42 undergoes a guanine nucleotide-specific membrane association and carboxyl methylation in normal rat islets, human islets, and pure beta (HIT or INS-1) cells. GTPgammaS-dependent carboxyl methylation of a 23-kD protein was also demonstrable in secretory granule fractions from normal islets or beta cells. AFC (a specific inhibitor of prenyl-cysteine carboxyl methyl transferases) blocked the carboxyl methylation of CDC42 in five types of insulin-secreting cells, without blocking GTPgammaS-induced translocation, implying that methylation is a consequence (not a cause) of transfer to membrane sites. High glucose (but not a depolarizing concentration of K+) induced the carboxyl methylation of CDC42 in intact cells, as assessed after specific immunoprecipitation. This effect was abrogated by GTP depletion using mycophenolic acid and was restored upon GTP repletion by coprovision of guanosine. In contrast, although rap 1 was also carboxyl methylated, it was not translocated to the particulate fraction by GTPgammaS; furthermore, its methylation was also stimulated by 40 mM K+ (suggesting a role which is not specific to nutrient stimulation). AFC also impeded nutrient-induced (but not K+-induced) insulin secretion from islets and beta cells under static or perifusion conditions, whereas an inactive structural analogue of AFC failed to inhibit insulin release. These effects were reproduced not only by S-adenosylhomocysteine (another methylation inhibitor), but also by GTP depletion. Thus, the glucose- and GTP-dependent carboxyl methylation of G-proteins such as CDC42 is an obligate step in the stimulus-secretion coupling of nutrient-induced insulin secretion, but not in the exocytotic event itself. Furthermore, AFC blocked glucose-activated phosphoinositide turnover, which may provide a partial biochemical explanation for its effect on secretion, and implies that certain G-proteins must be carboxyl methylated for their interaction with signaling effector molecules, a step which can be regulated by intracellular availability of GTP.

BTBR Ob/Ob Mutant Mice Model Progressive Diabetic Nephropathy

There remains a need for robust mouse models of diabetic nephropathy (DN) that mimic key features of advanced human DN. The recently developed mouse strain BTBR with the ob/ob leptin-deficiency mutation develops severe type 2 diabetes, hypercholesterolemia, elevated triglycerides, and insulin resistance, but the renal phenotype has not been characterized. Here, we show that these obese, diabetic mice rapidly develop morphologic renal lesions characteristic of both early and advanced human DN. BTBR ob/ob mice developed progressive proteinuria beginning at 4 weeks. Glomerular hypertrophy and accumulation of mesangial matrix, characteristic of early DN, were present by 8 weeks, and glomerular lesions similar to those of advanced human DN were present by 20 weeks. By 22 weeks, we observed an approximately 20% increase in basement membrane thickness and a >50% increase in mesangial matrix. Diffuse mesangial sclerosis (focally approaching nodular glomerulosclerosis), focal arteriolar hyalinosis, mesangiolysis, and focal mild interstitial fibrosis were present. Loss of podocytes was present early and persisted. In summary, BTBR ob/ob mice develop a constellation of abnormalities that closely resemble advanced human DN more rapidly than most other murine models, making this strain particularly attractive for testing therapeutic interventions.

Obesity and genetics regulate microRNAs in islets, liver, and adipose of diabetic mice

Type 2 diabetes results from severe insulin resistance coupled with a failure of β cells to compensate by secreting sufficient insulin. Multiple genetic loci are involved in the development of diabetes, although the effect of each gene on diabetes susceptibility is thought to be small. MicroRNAs (miRNAs) are noncoding 19–22-nucleotide RNA molecules that potentially regulate the expression of thousands of genes. To understand the relationship between miRNA regulation and obesity-induced diabetes, we quantitatively profiled approximately 220 miRNAs in pancreatic islets, adipose tissue, and liver from diabetes-resistant (B6) and diabetes-susceptible (BTBR) mice. More than half of the miRNAs profiled were expressed in all three tissues, with many miRNAs in each tissue showing significant changes in response to genetic obesity. Furthermore, several miRNAs in each tissue were differentially responsive to obesity in B6 versus BTBR mice, suggesting that they may be involved in the pathogenesis of diabetes. In liver there were approximately 40 miRNAs that were downregulated in response to obesity in B6 but not BTBR mice, indicating that genetic differences between the mouse strains play a critical role in miRNA regulation. In order to elucidate the genetic architecture of hepatic miRNA expression, we measured the expression of miRNAs in genetically obese F2 mice. Approximately 10% of the miRNAs measured showed significant linkage (miR-eQTLs), identifying loci that control miRNA abundance. Understanding the influence that obesity and genetics exert on the regulation of miRNA expression will reveal the role miRNAs play in the context of obesity-induced type 2 diabetes.

Evidence for Differential Roles of the Rho Subfamily of GTP-Binding Proteins in Glucose- and Calcium-Induced Insulin Secretion from Pancreatic β Cells

We utilized clostridial toxins (with known specificities for inhibition of GTPases) to ascertain the contribution of candidate GTPases in physiologic insulin secretion from beta cells. Exposure of normal rat islets or isolated beta (HIT-T15) cells to Clostridium difficile toxins A and B catalyzed the glucosylation (and thereby the inactivation) of Rac, Cdc42, and Rho endogenous to beta cells; concomitantly, either toxin reduced glucose- or potassium-induced insulin secretion from rat islets and HIT cells. Treatment of beta cells with Clostridium sordellii lethal toxin (LT; which modified only Ras, Rap, and Rac) also reduced glucose- or potassium-induced secretion. However, clostridial toxin C3-exoenzyme (which ADP-ribosylates and inactivates only Rho) was without any effect on either glucose- or potassium-induced insulin secretion. These data suggest that Cdc42, Rac, Ras, and/or Rap (but not Rho) may be needed for glucose- or potassium-mediated secretion. The effects of these toxins appear to be specific on stimulus-secretion coupling, since no difference in metabolic viability (assessed colorimetrically by quantitating the conversion of the tetrazolium salt into a formazan in a reduction reaction driven by nutrient metabolism) was demonstrable between control and toxin (A or LT)-treated beta cells. Toxin (A or LT) treatment also did not alter glucose- or potassium-mediated rises in cytosolic free calcium concentrations ([Ca2+]i), suggesting that these GTPases are involved in steps distal to elevations in [Ca2+]i. Recent findings indicate that the carboxyl methylation of Cdc42 is stimulated by only glucose, whereas that of Rap (Kowluru et al., J Clin Invest 98: 540-555, 1996) and Rac (present study) are regulated by glucose or potassium. Together, these findings provide direct evidence, for the first time, that the Rho subfamily of GTPases plays a key regulatory role(s) in insulin secretion, and they suggest that Cdc42 may be required for early steps in glucose stimulation of insulin release, whereas Rap and/or Rac may be required for a later step(s) in the stimulus-secretion coupling cascade (i.e. Ca2+-induced exocytosis of insulin).

Positional Cloning of a Type 2 Diabetes Quantitative Trait Locus; Tomosyn-2, a Negative Regulator of Insulin Secretion

We previously mapped a type 2 diabetes (T2D) locus on chromosome 16 (Chr 16) in an F2 intercross from the BTBR T (+) tf (BTBR) Lepob/ob and C57BL/6 (B6) Lepob/ob mouse strains. Introgression of BTBR Chr 16 into B6 mice resulted in a consomic mouse with reduced fasting plasma insulin and elevated glucose levels. We derived a panel of sub-congenic mice and narrowed the diabetes susceptibility locus to a 1.6 Mb region. Introgression of this 1.6 Mb fragment of the BTBR Chr 16 into lean B6 mice (B6.16BT36–38) replicated the phenotypes of the consomic mice. Pancreatic islets from the B6.16BT36–38 mice were defective in the second phase of the insulin secretion, suggesting that the 1.6 Mb region encodes a regulator of insulin secretion. Within this region, syntaxin-binding protein 5-like (Stxbp5l) or tomosyn-2 was the only gene with an expression difference and a non-synonymous coding single nucleotide polymorphism (SNP) between the B6 and BTBR alleles. Overexpression of the b-tomosyn-2 isoform in the pancreatic β-cell line, INS1 (832/13), resulted in an inhibition of insulin secretion in response to 3 mM 8-bromo cAMP at 7 mM glucose. In vitro binding experiments showed that tomosyn-2 binds recombinant syntaxin-1A and syntaxin-4, key proteins that are involved in insulin secretion via formation of the SNARE complex. The B6 form of tomosyn-2 is more susceptible to proteasomal degradation than the BTBR form, establishing a functional role for the coding SNP in tomosyn-2. We conclude that tomosyn-2 is the major gene responsible for the T2D Chr 16 quantitative trait locus (QTL) we mapped in our mouse cross. Our findings suggest that tomosyn-2 is a key negative regulator of insulin secretion.

A Potent α/β-Peptide Analogue of GLP-1 with Prolonged Action in Vivo

Glucagon-like peptide-1 (GLP-1) is a natural agonist for GLP-1R, a G protein-coupled receptor (GPCR) on the surface of pancreatic β cells. GLP-1R agoinsts are attractive for treatment of type 2 diabetes, but GLP-1 itself is rapidly degraded by peptidases in vivo. We describe a design strategy for retaining GLP-1-like activity while engendering prolonged activity in vivo, based on strategic replacement of native α residues with conformationally constrained β-amino acid residues. This backbone-modification approach may be useful for developing stabilized analogues of other peptide hormones.

Subcellular localization and kinetic characterization of guanine nucleotide binding proteins in normal rat and human pancreatic islets and transformed β cells

The subcellular localization and the kinetics of the GTPase activities of monomeric and heterotrimeric GTP-binding proteins were investigated in normal rat and human pancreatic islets and were compared to those obtained using a transformed hamster beta cell line (HIT cells). The [alpha-32P]GTP overlay technique revealed the presence of at least four low-molecular-mass proteins (approx. 20-27 kDa) in normal rat islets, which were enriched in the secretory granule fraction compared to the membrane fraction (with little abundance of these proteins in the cytosolic fraction). In contrast, in HIT cells, these proteins (at least six) were predominantly cytosolic. Three of these proteins were immunologically identified as rab3A, rac2, and CDC42Hs in islets as well as in HIT cells. In addition, pertussis toxin augmented the ribosylation of at least one heterotrimeric G-protein of about 39 kDa (probably G(i) and/or G(o)) in the membrane and secretory granule fractions of normal rat and human islets, whereas at least three such Ptx substrates (36-39 kDa) were found in HIT cell membranes. Kinetic activities revealed the presence of at least three such activities (Km for GTP of 372 nM, 2.2 microM, and 724 microM) in islet homogenates which were differentially distributed in various subcellular fractions; similar activities were also demonstrable in HIT cell homogenates. Thus, these studies demonstrate the presence of both monomeric G-proteins intrinsic to the secretory granules of normal rat islets which can be ascribed to beta cells; since these G-proteins are regulated by insulinotropic lipids (as described in the accompanying article), such proteins may couple the activation of phospholipases (endogenous to islets) to the exocytotic secretion of insulin. These findings also suggest that caution is necessary in extrapolating data concerning G-proteins from cultured, transformed beta cell lines to the physiology of normal islets, in view of both qualitative and quantitative differences between the two preparations.