Insulin Resistance in Tetracycline-Repressible Munc18c Transgenic Mice

  1. Beth A. Spurlin1,
  2. Rhonda M. Thomas1,
  3. Angela K. Nevins1,
  4. Hyo-Jeong Kim2,
  5. Yoon-Jung Kim2,
  6. Hye-Lim Noh2,
  7. Gerald I. Shulman23,
  8. Jason K. Kim2 and
  9. Debbie C. Thurmond1
  1. 1Department of Biochemistry and Molecular Biology, Center for Diabetes Research, Indiana University School of Medicine, Indianapolis, Indiana
  2. 2Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut
  3. 3Howard Hughes Medical Institute, Yale University School of Medicine, New Haven, Connecticut
  1. Address correspondence and reprint requests to Debbie C. Thurmond, Department of Biochemistry and Molecular Biology, Center for Diabetes Research, Indiana University School of Medicine, Indianapolis, IN 46202. E-mail: dthurmon{at}iupui.edu

Abstract

To investigate the physiological effects of modulating the abundance of Munc18c or syntaxin 4 (Syn4) proteins on the regulation of glucose homeostasis in vivo, we generated tetracycline-repressible transgenic mice that overexpress either Munc18c or Syn4 proteins in skeletal muscle, pancreas and adipose tissue seven-, five-, and threefold over endogenous protein, respectively. Munc18c transgenic mice displayed whole-body insulin resistance during hyperinsulinemic-euglycemic clamp resulting from >41% reductions in skeletal muscle and white adipose tissue glucose uptake, but without alteration of hepatic insulin action. Munc18c transgenic mice exhibited ∼40% decreases in whole-body glycogen/lipid synthesis, skeletal muscle glycogen synthesis, and glycolysis. Glucose intolerance in Munc18c transgenic mice was reversed by repression of transgene expression using tetracycline or by simultaneous overexpression of Syn4 protein. In addition, Munc18c transgenic mice had depressed serum insulin levels, reflecting a threefold reduction in insulin secretion from islets isolated therefrom, thus uncovering roles for Munc18c and/or Syn4 in insulin granule exocytosis. Taken together, these results indicate that balance, more than absolute abundance, of Munc18c and Syn4 proteins directly affects whole-body glucose homeostasis through alterations in insulin secretion and insulin action.

Insulin resistance, in large part, results from an inability to recruit adequate quantities of GLUT4 protein to the cell surface. Skeletal muscle and adipose tissue clearance of circulating blood glucose accounts for the majority of insulin-stimulated glucose uptake, as these tissues express the insulin-responsive glucose transporter GLUT4 (1). In the basal non-insulin-stimulated state, GLUT4 localizes to tubulovesicular elements and small intracellular vesicles throughout the cell cytoplasm (2). Upon stimulation with insulin, these GLUT4-containing compartments translocate to and fuse with the plasma membrane (37). This ultimately results in a large increase in the number of functional glucose transporters on the cell surface, which accounts for nearly all of the insulin-stimulated increase in glucose uptake.

The insulin-stimulated translocation of GLUT4-containing vesicles is a complex multistep process necessary for normal maintenance of glucose homeostasis (4,6,7). Insulin binding to its receptor activates the intrinsic protein kinase of the receptor β subunit, resulting in its autophosphorylation and tyrosine phosphorylation of the family of insulin receptor substrate (IRS) proteins (8) and Cbl (9). The phosphorylation of IRS results in the association, activation, and targeting of the phosphatidylinositol 3-kinase (PI 3-kinase) (1012). The active PI 3-kinase generates phosphatidylinositol-3,4,5-trisphosphate, which is necessary for the stimulation of both protein kinase B and atypical protein kinase C isoforms (13,14). This insulin signaling cascade leads to vectorial movement, or “trafficking,” of the GLUT4 vesicle toward the plasma membrane. Once near the inner leaflet of the plasma membrane, GLUT4 vesicles associate with t-SNARE (SNAP [soluble NSF (N-ethylmaleimide sensitive factor) attachment protein] receptor) proteins syntaxin 4 (Syn4) and SNAP-23 present in the plasma membrane (1520) via the GLUT4 vesicle cargo v-SNARE protein VAMP2 (17,2124). VAMP2, Syn4, and SNAP-23 proteins form the heterotrimeric SNARE core complex essential for the eventual incorporation of GLUT4 into the plasma membrane to facilitate glucose uptake.

SNARE core complex interactions in Saccharomyces cerevisiae, Caenorhabditis elegans, and Drosophila melanogaster are regulated by the Sec1-type proteins, through specific and high-affinity binding to their cognate syntaxins (2527). Three Sec1 homologs have been identified in mammalian plasma membranes: Munc18a, Munc18b, and Munc18c. Munc18a is predominantly expressed in neurons, whereas Munc18b and Munc18c are expressed ubiquitously, and only Munc18c binds Syn4 with high affinity (2832). Null mutations in the genes for the SM proteins (Sec1/Munc18) cause dramatic reductions in vesicle exocytosis, suggesting that these proteins are essential for normal SNARE function (27,33,34). Similar to other SM proteins, disruption of Munc18c binding to Syn4 impairs vesicle fusion, indicating that it is required for the insulin-stimulated GLUT4 translocation process (35,36).

Overexpression of Munc18c results in inhibition of insulin-stimulated vesicle fusion in vitro, but its effect on glucose uptake in vivo is unknown (29,3739). Munc18c binds to Syn4 in a manner mutually exclusive of either Syn4 binding proteins (SNAP-23 and VAMP2) in adipocytes, and competes for Syn4 when overexpressed (36). Overexpression of Munc18c in 3T3L1 adipocytes inhibits insulin-stimulated GLUT4 translocation (38,39). This inhibitory effect was fully reversed by increased expression of Syn4, indicating a functional requirement for excess Syn4 over Munc18c protein (36). Consistent with this, increased expression in skeletal muscle of GLUT4-EGFP mice by adenoviral-Munc18c resulted in inhibition of insulin-stimulated GLUT4 translocation, specifically in the transverse tubules and not in the sarcolemma (40). Similarly, sarcolemmal membranes contained much higher levels of Syn4 protein than transverse tubules, suggesting that overexpression of Munc18c was sufficient to saturate the available endogenous Syn4 binding sites in transverse tubules yet not in the sarcolemmal membrane (40). Evidence favoring this model stems from a growing number of cellular systems reporting an unequal stoichiometric abundance of Munc18 protein relative to syntaxin protein (36,41,42).

The physiological relevance of Syn4 and Munc18c protein abundance with respect to insulin-stimulated GLUT4 translocation and glucose homeostasis has not been investigated in vivo. Therefore, we generated transgenic mice that overexpress Munc18c and/or Syn4 proteins under the control of a tetracycline-repressible promoter in peripheral insulin-sensitive tissues (skeletal muscle and fat) and pancreas. In this report, we demonstrate that Munc18c but not Syn4 protein overexpression alone results in insulin resistance and impaired insulin secretion. Hyperinsulinemic-euglycemic clamp analyses showed defective peripheral glucose disposal, consistent with a defect in skeletal muscle insulin-stimulated GLUT4 translocation. Oral administration of tetracycline resulted in normalization of glucose intolerance and serum insulin levels coordinate with the downregulation of transgene expression, indicating that the deleterious results of Munc18c overexpression are reversible by restoration of endogenous Munc18c protein levels. Furthermore, glucose intolerance is also normalized by coordinate overexpression of Syn4 with Munc18c, suggesting that Munc18c protein must be balanced by an overabundance of Syn4 to maintain normoglycemia in vivo.

RESEARCH DESIGN AND METHODS

Materials.

The rabbit polyclonal GLUT4, Syn4, VAMP2, and Munc18c antibodies were obtained as described (17,39). The SNAP-23, GLUT1, transferrin receptor, phosphotyrosine-horseradish peroxidase (PY20-HRP), and phosphoserine-specific Akt antibodies were purchased from Affinity Bioreagents (Golden, CO), Chemicon (Temecula, CA), Zymed Laboratories (South San Francisco, CA), Transduction Laboratories (Lexington, KY), and New England Biolabs (Beverley, MA), respectively. Goat anti-rabbit HRP and anti-digoxigenin HRP-conjugated secondary antibodies were purchased from Bio-Rad (Hercules, CA) and Roche (Indianapolis, IN), respectively. The GST-Syn4 ΔTM, GST-SNAP-23, and GST-Munc18cCT plasmids were gifts of Dr. Jeffrey Pessin (SUNY Stony Brook, NY). The rat insulin radioimmunoassay kit was acquired from Linco Research (St. Charles, MO).

Generation of pCOMBICMV-Munc18c transgenic mice.

All studies involving mice followed the Guidelines for the Use and Care of Laboratory Animals. The pUC-COMBICMV plasmid for generation of tetracycline-repressible transgenic (Tg) mice was a gift from Dr. Ulli Certa (Hoffman la Roche, Switzerland) (43). Flag-tagged Munc18c (39) was inserted into the PmeI site pCombi-CMV vector. The construct was linearized by digestion with NotI and microinjected into the nucleus of preimplantation embryos and transferred into the oviduct of pseudo-pregnant C57Bl6J female mice by staff at the the University of Iowa Transgenic Animal Facility (Iowa City, IA). Pups were screened for the presence of the transgene using PCR of genomic DNA, and the positive animals were bred with C57Bl6J stock.

Tissue homogenization and immunoblotting.

Tissues were homogenized in a 1% Igepal detergent buffer (25 mmol/l Tris, pH 7.4, 1% Igepal, 10% glycerol, 50 mmol/l sodium fluoride, 10 mmol/l sodium pyrophosphate, 137 mmol/l sodium chloride, 1 mmol/l sodium vanadate, 1 mmol/l phenylmethylsulfonyl fluoride, 10 μg/ml aprotinin, 1 μg/ml pepstatin, and 5 μg/ml leupeptin) for 15 s and centrifuged at 2000g for 5 min; subsequent supernatants were microcentrifuged at 13,500g for 20 min at 4°C. Proteins were separated on 10% or 12% SDS-PAGE followed by transfer to polyvinylidene difluoride (PVDF) or nitrocellulose membrane for immunoblotting.

Intraperitoneal glucose tolerance test (IPGTT).

Male Munc18c Tg and wild-type mice (4–6 months old) were fasted overnight for 18 h. Blood was collected from the tail vein and blood glucose monitored (Hemocue). After sample collection of fasted blood, mice were administered glucose (2 g/kg) by intraperitoneal injection, and subsequent blood glucose readings were taken at 30-min intervals over 120 min.

Hyperinsulemic-euglycemic clamp.

Surgery was performed to chronically cannulate the left jugular vein 5 days before the in vivo experiment. After an overnight fast, a 120-min hyperinsulinemic-euglycemic clamp experiment in awake mice with a primed continuous infusion of insulin (Humulin R; Eli Lilly, Indianapolis, IN), high-performance liquid chromatography-purified [3-3H]glucose, and 2-deoxy-d-[1-14C]glucose was conducted as previously described (44).

Isolation, culture, and stimulation of insulin secretion of mouse islets.

Pancreatic mouse islets were isolated using a modification of the previously described method (45). Briefly, pancreata from 8- to 12-week-old male mice were digested with collagenase and purified using a Ficoll density gradient. After isolation, islets were cultured overnight in CMRL-1066 medium. Fresh islets were hand-picked into groups of 10, preincubated in Krebs-Ringer bicarbonate buffer (10 mmol/l HEPES pH 7.4, 134 mmol/l NaCl, 5 mmol/l NaHCO3, 4.8 mmol/l KCl, 1 mmol/l CaCl2, 1.2 mmol/l MgSO4, 1.2 mmol/l KH2PO4) containing 2 mmol/l glucose and 0.1% BSA for 2 h, followed by stimulation with 20 mmol/l glucose for 2 h. Media was collected to measure insulin secretion, and islets were harvested in NP-40 lysis buffer to determine cellular insulin content by radioimmunoassay.

Quantitation of SNARE proteins in tissue homogenates.

GST-Munc18CT (13 residues of the far COOH-terminus of Munc18c), GST-Syn4ΔTM, and GST-SNAP-23 were expressed in DH5α strain of Escherichia coli using isopropylthiogalactoside induction (36). Skeletal muscle homogenate proteins were subjected to electrophoresis alongside known quantities of GST-Munc18c, GST-Syn4, or GST-SNAP-23 recombinant proteins on 10% SDS-PAGE followed by transfer to PVDF membranes and immunoblotting with anti-Munc18c, anti-Syn4, or anti-SNAP-23, respectively. Proteins were detected using enhanced chemiluminescence, using exposures well within the linear range of the film, and quantitated using the BioRad Quantity One software package (Hercules, CA).

RESULTS

Protein expression in Munc18c Tg mice.

To determine the importance of Munc18c protein in insulin-stimulated GLUT4 translocation in vivo, we generated transgenic mice with overexpression of Munc18c in skeletal muscle and adipose tissue. Of the six founders, four lines transmitted the transgene and two lines, 14131/2 and 14140/2, overexpressed the Munc18c protein in skeletal muscle by threefold and sevenfold, respectively. Mice overexpressing Munc18c protein in skeletal muscle by threefold showed no significant glucose intolerance, whereas the line expressing sevenfold was glucose intolerant and insulin resistant. Subsequent studies were performed using the 14140/2 line of Munc18c Tg mice.

The initial use of this particular vector described differential levels of transgenic protein expression among various tissues, with skeletal muscle exhibiting the largest increase in expression and no changes in brain tissue (43). The overexpression of Munc18c protein in the Tg male and female mice was directly compared to endogenous levels in a wild-type littermate expressed in heart, lung, liver, kidney, spleen, pancreas, skeletal muscle, and epididymal fat by immunoblot analysis. Of these tissues, only pancreas (Fig. 1A, lanes 11–12), fat (Fig. 1A, lanes 13–14), and skeletal muscle (Fig. 1A, lanes 15–16) showed a repeated significant increase in Munc18c protein (five-, three-, and sevenfold, respectively), without effecting Syn4, SNAP-23, or VAMP2 abundance. No significant differences in overall body weight (28 ± 1 g) or tissue weight between wild-type and Munc18c Tg mice were detected (data not shown) with the exception of heart weight—Munc18c Tg heart was 14% heavier than that of wild-type (0.21 ± 0.01 vs. 0.18 ± 0.01 g; P < 0.05). No changes in GLUT4 or the GLUT4 vesicle cargo protein IRAP (insulin responsive aminopeptidase) protein were detected in insulin-responsive tissues, and characteristic tissue-specific expression was displayed in heart, skeletal muscle, and fat but not liver (Fig. 1B), with no differences in GLUT1 protein abundance (data not shown). Syn4 protein was similarly overexpressed in only pancreas, skeletal muscle, and adipose tissues of the Syn4 Tg mice compared with wild-type mice (Fig. 1C), without alterations in VAMP2 or SNAP-23 abundance (data not shown). Thus the transgenic mouse lines overexpressed the Munc18c or Syn4 protein in insulin-secreting and insulin-responsive tissues.

Munc18c Tg mice exhibit glucose intolerance and insulin resistance.

To determine the effects of Munc18c overexpression in vivo, male Munc18c Tg and wild-type littermate male mice were fasted overnight and subjected to an intraperitoneal injection of glucose (2 g/kg), and disposal was monitored over a 2-h period (Fig. 2A). Peak blood glucose was reached by 30 min after injection in the wild-type mice and was cleared over the remaining 90 min. Although fasting glucose levels were similar, the blood glucose levels of Munc18c Tg mice peaked higher at 30 min and remained significantly higher for the duration of the glucose challenge. In contrast to the Munc18c Tg mice, the Syn4 transgenic mice demonstrated the ability to clear glucose to the same level as wild-type mice throughout the experiment (Fig. 2B). These data clearly demonstrated that the Munc18c Tg mice had impaired ability to clear glucose from the peripheral circulation.

To investigate the mechanism of glucose intolerance, hyperinsulinemic-euglycemic clamps were conducted in the Munc18c Tg mice. The plasma glucose concentrations were maintained at 7 mmol/l, and plasma insulin concentrations were raised from 16 to 77 pmol/l during the clamps. The glucose infusion rate necessary to maintain euglycemia under conditions of constant infusion of insulin (2.5 mU · kg−1 · min−1) was reduced by 32% in Munc18c Tg mice compared with wild-type mice (Fig. 3A). Consistent with this, Munc18c Tg mice exhibited 31, 24, and 39% reductions in insulin-stimulated whole-body glucose turnover, glycolysis, and glycogen/lipid synthesis, respectively, compared with the wild-type mice (Fig. 3B). Conversely, neither basal hepatic glucose production nor insulin’s ability to suppress hepatic glucose production was affected in the Munc18c Tg mice compared with wild-type mice (wild-type: basal, 92 ± 13; insulin, 0.0 ± 6.2; Munc18c Tg: basal, 84 ± 14; insulin, 0.0 ± 7.8 μmol · kg−1 · min−1). Taken together, these data indicate that glucose intolerance in the Munc18c Tg mice was due to reduced peripheral glucose disposal in response to insulin and not to alteration in hepatic insulin action.

Impaired glucose uptake in skeletal muscle and adipose tissue of the Munc18c Tg mice.

Because glucose uptake into skeletal muscle and adipose tissue accounts for the majority of peripheral glucose disposal, we examined tissue-specific glucose uptake in vivo during hyperinsulinemic-euglycemic clamps. Insulin-stimulated glucose uptake in skeletal muscle (gastrocnemius) was decreased by 42% in the Munc18c Tg mice compared with the wild-type littermate mice (Fig. 4A). Similarly, skeletal muscle glycolysis and glycogen synthesis rates were reduced, by 41 and 47%, respectively, in the Munc18c Tg mice. In white adipose tissue, glucose uptake was significantly decreased by 43% in the Munc18c Tg mice compared with wild-type littermates (Fig. 4B). Glucose uptake in brown adipose tissue of the Munc18c Tg was also reduced by 25% compared with wild-type mice (Fig. 4C). In contrast, glucose uptake in cardiac muscle of Munc18c Tg mice did not differ from that of wild-type mice, consistent with a lack of Munc18c protein overexpression in cardiac muscle (Fig. 4D). Overall, whole-body insulin resistance in the Munc18c Tg mice was accounted for by the 41% decrease in skeletal muscle and 43% decrease in white adipose tissue glucose uptake in these mice and was without systemic effects in tissues that do not overexpress Munc18c.

The mechanism for glucose uptake into peripheral tissues involves initiation of the insulin signaling cascade, which leads to the translocation of GLUT4-containing vesicles to the cell surface membranes. We have previously shown that overexpression of Munc18c by adenoviral infection of skeletal muscle of GLUT4-EGFP mice resulted in inhibition of insulin-stimulated GLUT4 translocation (40), as assessed by fractionation of hindquarter muscles into cell surface and intracellular membrane compartments (46). Consistent with these data, the Munc18c Tg mice failed to translocate GLUT4 to the cell surface fractions in response to insulin (compared with wild-type mice), whereas all mice showed similar quantities of GLUT4 protein in the cell surface fractions under basal conditions (data not shown). By contrast, insulin signaling was unaffected in the overexpression of Munc18c in skeletal muscle tissue homogenates, since transgenic and wild-type mice injected with insulin showed equivalently elevated levels of tyrosine phosphorylation of insulin receptor (IR) and IRS-1 proteins as well as increased serine phosphorylation of PKB/Akt (data not shown). Taken together, these data are consistent with the concept that Munc18c overexpression impairs insulin-stimulated GLUT4 translocation and is independent of effects on proximal insulin signaling events.

Munc18c overexpression reduces insulin secretion.

To examine the metabolic characteristics of Munc18c Tg and wild-type mice with and without tetracycline administration, fasting levels of serum glucose, insulin, triglycerides, cholesterol, and nonesterified fatty acid (NEFA) levels were quantitated (Table 1). Munc18c Tg mice showed no significant increase in fasting blood glucose compared with wild-type mice, and no statistically significant differences were observed for serum triglycerides, cholesterol or NEFAs, although Munc18c Tg mice tended to have slightly elevated triglycerides and cholesterol levels. In contrast, Munc18c Tg mice exhibited significantly reduced fasting insulin levels compared with wild-type, and after tetracycline treatment these levels rose to the wild-type levels. This alteration in insulin levels might reflect the potential effect of fivefold overexpression of Munc18c on pancreatic functions, especially since it was normalized by tetracycline administration.

To investigate the possibility that Munc18c overexpression reduced insulin secretion, islets were isolated from Munc18c Tg or wild-type mice. Glucose stimulation (20 mmol/l) resulted in a 27-fold increase in insulin release compared with unstimulated islets of wild-type mice (Fig. 5). By contrast, glucose stimulation resulted in only a sixfold increase in insulin release from Munc18c Tg islets. Data shown are corrected for total insulin content, although no significant alterations of total insulin content of wild-type or Munc18c Tg islets incubated with or without glucose were detected. These data demonstrate that overexpression of Munc18c in pancreatic islet cells reduced glucose-stimulated insulin secretion.

Restoration of glucose tolerance in Munc18c Tg mice by tetracycline or by coexpressing Syn4 protein.

The Munc18c transgene is under the regulation of the Tet operator, such that oral administration of tetracycline causes downregulation of the transgene (43). To evaluate the effectiveness of the Tet-repressible system, 4- to 6-month-old Munc18c Tg and wild-type littermate mice were pair-fed tetracycline in drinking water (1 mg/ml) for 7 days, after which their tissues were homogenized and immunoblotted for protein expression. Similar to the results from non-tetracycline-fed mice, Munc18c protein content was similar in heart, lung, liver, kidney, and spleen (Fig. 6A, lanes 1–10). However, tetracycline administration downregulated Munc18c protein overexpression to levels near those observed in wild-type mice in pancreas, skeletal muscle, and fat (Fig. 6A, lanes 11–16). Moreover, tetracycline feeding had no apparent adverse effects on expression of Syn4, SNAP-23, or VAMP2, indicating its specific repressive effect on the tetracycline-sensitive Munc18c transgene.

The advantage of the tetracycline-repressible transgene system is the flexibility of using a mouse as its own control in metabolic studies. To determine whether the profound metabolic phenotype (i.e., peripheral insulin resistance) in the Munc18c Tg mice was directly caused by the increased Munc18c protein, we administered tetracycline to the Munc18c Tg and wild-type mice used in the data set for Fig. 2A for 1 week and performed the IPGTT. Consistent with the immunoblotting results, normalization of Munc18c protein content in the transgenic mice resulted in normalization of glucose tolerance to levels indistinguishable from those of wild-type mice fed tetracycline (Fig. 6B). With the exception of insulin levels, which were normalized by tetracycline administration, tetracycline exerted no changes in metabolic parameters, in either wild-type or Munc18c Tg mice.

To determine if glucose intolerance in the Munc18c Tg mice could be normalized by simultaneous overexpression of Syn4 in the same tissues, Munc18c Tg and Syn4 transgenic mice were cross-bred to generate Munc18c/Syn4 double-Tg mice (SM Tg). The SM Tg mice were as glucose tolerant as the Syn4 transgenic mice (Fig. 6C). In addition, the SM Tg mice displayed insulin levels (0.3 ± 0.09) not significantly different from wild-type. Taken together, these data demonstrate that simultaneous increase of Syn4 with Munc18c in vivo reestablished normoglycemia by restoring the necessary stoichiometric balance between these proteins.

DISCUSSION

In this report, we document the generation and characterization of transgenic mouse models in which the level of Munc18c or Syn4 protein can be modulated by tetracycline. Our data clearly showed that Munc18c overexpression in skeletal muscle, adipose tissue, and pancreas led to insulin resistance and impaired insulin secretion, which were reversible upon tetracycline administration or by the simultaneous overexpression of Syn4 in vivo. Glucose intolerance resulted from significant 40% reductions in whole-body glucose turnover and glycogen/lipid synthesis, and similar decreases in skeletal muscle glucose uptake, glycolysis, and glycogen synthesis, in addition to a 43% decrease in white adipose glucose uptake. Importantly, insulin resistance in the Munc18c Tg mice was independent of any effect on hepatic insulin action, indicating that the observed whole-body insulin resistance was the result of defects in peripheral glucose uptake. Moreover, our Munc18c Tg mice showed no differences from littermate wild-type mice in body weight or adipose mass, similar to the adipose-specific GLUT4 knockout mice (47). Intriguingly, Munc18c Tg mice showed defective insulin secretion that was reversible by tetracycline—the first report on the effects of Munc18c on insulin secretion. In all, these alterations in Munc18c protein abundance culminated in insulin resistance and impaired insulin secretion in vivo.

The glucose intolerance of the Munc18c Tg mice was fully reversed after 1 week of oral tetracycline administration to ablate transgene expression, using the same mice initially used to characterize the glucose intolerant phenotype. Quantitative immunoblotting using total homogenates of skeletal muscle (gastrocnemius) from male wild-type and male Munc18c Tg mice revealed that Syn4 and SNAP-23 were present at ∼50 nmol/l and 150 nmol/l, respectively. By contrast, Munc18c was present in wild-type mice at 2 nmol/l, elevated to 15 nmol/l in Munc18c Tg mice. These quantitative results were consistent with a previously reported threefold overabundance of SNAP-23 relative to Syn4 in 3T3L1 adipocytes (42), with Syn4 and Munc18 abundances reported in rat liver and kidney tissues (41), and also with our previous findings using 3T3L1 adipocytes (36). Interestingly, even with a more than sevenfold increase in Munc18c protein, there remains a greater than threefold excess of Syn4, and yet this still results in the insulin-resistant phenotype and dramatically impaired skeletal muscle GLUT4 translocation. Because the Munc18-syntaxin complexes form in a 1:1 stoichiometry (30), further studies using variable dosage and duration of tetracycline treatment in these mice are underway to relate this phenomenon to glucose homeostasis.

Unexpectedly, Munc18c Tg mice exhibited significant reductions in fasting serum insulin levels, which were reversed by tetracycline administration to downregulate Munc18c transgene expression, overall suggesting that Munc18c protein affected islet function. Munc18c was overexpressed by fivefold in pancreas tissue from Munc18c Tg mice; however, there were no apparent differences in pancreas tissue weight or islet size compared with those of wild-type mice. Although not believed to regulate glucose uptake in β-cells, Munc18 and SNARE proteins have been shown to mediate glucose-stimulated exocytosis of insulin-containing granules (48). Consistent with our finding here with Munc18c overexpression, a recent Munc18a overexpression study implicated Munc18a (Munc18–1/n-Sec1) as a negative regulator of the insulin secretory machinery via a mechanism involving its binding partner syntaxin 1 (49), and Munc18c overexpression in islets may inhibit insulin granule exocytosis via a mechanism involving Syn4. Alternatively, Munc18c may bind and sequester a different protein required for insulin secretion. It has also been reported that the distribution of Munc18c in pancreatic acinar cells can affect exocytosis (50). Thus, more studies will be necessary to resolve the mechanism underlying the inhibition of insulin secretion in Munc18c Tg islets, and ultimately techniques other than overexpression will be required to determine the true function of the Munc18 proteins in insulin secretion.

FIG. 1.

Trangenic protein expression in pancreas, skeletal muscle, and fat of Munc18c Tg and Syn4 transgenic mice. Heart, lung, liver, kidney (Kid), spleen, pancreas (Panc), gastrocnemius skeletal muscle (Musc), and epididymal adipose (Fat) tissues were isolated from Munc18c Tg (Tg) and wild-type littermate (Wt) mice and homogenized as described in research design and methods. Samples (50 μg per lane) were resolved by SDS-PAGE and immunoblotted with antibodies for Munc18c, Syn4 (Syn 4), SNAP-23 or VAMP2 (A) or IRAP and GLUT4 (B) (5 μg per lane). C: Pancreas, gastrocnemius skeletal muscle, and epididymal adipose tissue homogenates were prepared from Syn4 transgenic and wild-type littermate mice and immunoblotted for the presence of Syn4 protein as described above. Data are representative of at least three independent sets of tissues.

FIG. 2.

Munc18c Tg mice are glucose intolerant. Glucose tolerance testing (GTT) of Munc18c Tg and Syn4 transgenic mice was performed by intraperitoneal injection of d-glucose (2 g/kg) into 4- to 6-month-old male mice fasted for 18 h. Blood glucose was monitored at 0, 30, 60, 90, and 120 min postinjection as described in research design and methods. Data shown are the averages ± SE from 11 Munc18c Tg (Tg) and 10 wild-type (Wt) mice (A) or 8 Syn4 Tg and 8 Wt mice (B). P < 0.05 vs. wild-type mice, unpaired Student’s t test.

FIG. 3.

Munc18c Tg mice exhibit impaired whole-body glucose homeostasis as assessed by hyperinsulinemic-euglycemic clamp. A: Steady-state glucose infusion rate obtained from averaged rates during 90–120 min of clamp procedure. B: Insulin-stimulated glucose uptake, glycolysis, and glycogen/lipid synthesis. Data shown are the averages ± SE from seven wild-type (□) or Munc18c Tg mice (▪). P < 0.05 vs. wild-type mice, unpaired Student’s t test.

FIG. 4.

Munc18c Tg mice show defects in skeletal muscle glucose uptake, glycolysis, and glycogen synthesis in hyperinsulinemic-euglycemic clamp. A: Insulin-stimulated glucose uptake, glycolysis, and glycogen synthesis in gastrocnemius muscles. B: Glucose uptake into white adipose tissue. C: Glucose uptake into brown adipose tissue. D: Glucose uptake into cardiac muscle. Data shown are the averages ± SE from seven wild-type (□) or Munc18c Tg mice (▪). P < 0.05 vs. wild-type mice, unpaired Student’s t test.

FIG. 5.

Insulin secretion is impaired in islets from Munc18c Tg mice. Islets were isolated as described in research design and methods and placed in culture for 15 h at 37°C in CMRL medium. Islet cells were preincubated for 2 h in low glucose Krebs-Ringer buffer (2.8 mmol/l) followed by a 2-h incubation under basal (2.8 mmol/l glucose) or stimulated (20 mmol/l glucose) conditions. Insulin secretion was measured by radioimmunoassay and normalized for insulin content, and data subsequently normalized to basal to obtain fold secretion. Data shown are the averages ± SE; n = 3 replicates for one of three independent experiments. P < 0.01 vs. basal or glucose-stimulated Tg islets.

FIG. 6.

Downregulation of the Munc18c transgene or coexpression of the Syn4 transgene results in normalized glucose tolerance. A: Munc18c Tg and wild-type (Wt) male mice 4 to 6 months old were fed tetracycline (1 mg/ml) in drinking water for 7 days. Tissue extracts from heart, lung, liver, kidney (Kid), spleen, pancreas (Panc), skeletal muscle (Musc), and epididymal adipose (Fat) were prepared and protein (50 μg) separated by 10% SDS-PAGE and immunoblotted as described in Fig. 1. Data are representative of at least three independent sets of tissues. B: Before being killed for tissue analysis, mice were fasted for 18 h and glucose tolerance tested as described in Fig. 2 (data represent the averages ± SE of seven Tg and seven Wt mice). C: Glucose intolerance is normalized in mice overexpressing Syn4 with Munc18c. Munc18c Tg mice were crossed with Syn4 Tg mice and 4- to 6-month-old male offspring carrying both Munc18c and Syn4 transgenes (SM Tg, open stars) and compared with Syn4 Tg mice (•) in glucose tolerance tests as described in Fig. 1. The averages ± SE are shown for six SM Tg and six Syn4 Tg mice.

TABLE 1

Metabolic characteristics of wild-type and Munc18c Tg mice

Acknowledgments

Supported by a predoctoral fellowship from the Indiana University Diabetes Graduate Training Program (B.A.S.), a Career Development Award from the American Diabetes Association (D.C.T.), and a research grant from the Indiana University School of Medicine Showalter Research Trust Fund (D.C.T.). The clamp studies were conducted at the Yale Mouse Metabolic Phenotyping Center and supported by grants from the U.S. Public Health Service (U24 DK-59635, J.K.K. and G.I.S.) and the American Diabetes Association (7-01-JF-05, J.K.K.). Gerald I. Shulman is an investigator of the Howard Hughes Medical Institute.

We are very grateful to Dr. Jeffrey E. Pessin for generating the transgenic mice at the University of Iowa Transgenic Animal Facility. We would also like to thank Drs. Ulli Certa, Richard Scheller, and Steve Waters for the pCOMBI vector, Syn4 cDNA, and IRAP antibody, respectively. The Indiana University School of Medicine Analyte Core Facility was invaluable for their assistance with metabolic measurements of serum samples.

Footnotes

    • Accepted April 21, 2003.
    • Received December 13, 2002.

REFERENCES

| Table of Contents