Reduction in Pancreatic Transcription Factor PDX-1 Impairs Glucose-stimulated Insulin Secretion
2002; Elsevier BV; Volume: 277; Issue: 13 Linguagem: Inglês
10.1074/jbc.m111272200
ISSN1083-351X
AutoresMarcela Briššová, Masakazu Shiota, Wendell E. Nicholson, Maureen Gannon, Susan M. Knobel, David W. Piston, Christopher V.E. Wright, Alvin C. Powers,
Tópico(s)Metabolism, Diabetes, and Cancer
ResumoComplete lack of transcription factor PDX-1 leads to pancreatic agenesis, whereas heterozygosity for PDX-1 mutations has been recently noted in some individuals with maturity-onset diabetes of the young (MODY) and in some individuals with type 2 diabetes. To determine how alterations in PDX-1 affect islet function, we examined insulin secretion and islet physiology in mice with one PDX-1 allele inactivated. PDX-1+/− mice had a normal fasting blood glucose and pancreatic insulin content but had impaired glucose tolerance and secreted less insulin during glucose tolerance testing. The expression of PDX-1 and glucose transporter 2 in islets from PDX-1+/−mice was reduced to 68 and 55%, respectively, whereas glucokinase expression was not significantly altered. NAD(P)H generation in response to glucose was reduced by 30% in PDX-1+/− mice. The in situ perfused pancreas of PDX-1+/− mice secreted about 45% less insulin when stimulated with 16.7 mm glucose. The Km for insulin release was similar in wild type and PDX-1+/− mice. Insulin secretion in response to 20 mm arginine was unchanged; the response to 10 nm glucagon-like peptide-1 was slightly increased. However, insulin secretory responses to 10 mm 2-ketoisocaproate and 20 mm KCl were significantly reduced (by 61 and 66%, respectively). These results indicate that a modest reduction in PDX-1 impairs several events in glucose-stimulated insulin secretion (such as NAD(P)H generation, mitochondrial function, and/or mobilization of intracellular Ca2+) and that PDX-1 is important for normal function of adult pancreatic islets. Complete lack of transcription factor PDX-1 leads to pancreatic agenesis, whereas heterozygosity for PDX-1 mutations has been recently noted in some individuals with maturity-onset diabetes of the young (MODY) and in some individuals with type 2 diabetes. To determine how alterations in PDX-1 affect islet function, we examined insulin secretion and islet physiology in mice with one PDX-1 allele inactivated. PDX-1+/− mice had a normal fasting blood glucose and pancreatic insulin content but had impaired glucose tolerance and secreted less insulin during glucose tolerance testing. The expression of PDX-1 and glucose transporter 2 in islets from PDX-1+/−mice was reduced to 68 and 55%, respectively, whereas glucokinase expression was not significantly altered. NAD(P)H generation in response to glucose was reduced by 30% in PDX-1+/− mice. The in situ perfused pancreas of PDX-1+/− mice secreted about 45% less insulin when stimulated with 16.7 mm glucose. The Km for insulin release was similar in wild type and PDX-1+/− mice. Insulin secretion in response to 20 mm arginine was unchanged; the response to 10 nm glucagon-like peptide-1 was slightly increased. However, insulin secretory responses to 10 mm 2-ketoisocaproate and 20 mm KCl were significantly reduced (by 61 and 66%, respectively). These results indicate that a modest reduction in PDX-1 impairs several events in glucose-stimulated insulin secretion (such as NAD(P)H generation, mitochondrial function, and/or mobilization of intracellular Ca2+) and that PDX-1 is important for normal function of adult pancreatic islets. The islet transcription factor PDX-1 (also known as IPF-1, IDX-1, and STF-1) was originally discovered as an activator of the insulin and somatostatin genes and from developmental work in the frog (1.Habener J.F. Stoffers D.A. Proc. Assoc. Am. Physicians. 1998; 110: 12-21PubMed Google Scholar, 2.Edlund H. Diabetes. 1998; 47: 1817-1823Crossref PubMed Scopus (276) Google Scholar, 3.Jonsson J. Ahlgren U. Edlund T. Edlund H. Int. J. Dev. Biol. 1995; 39: 789-798PubMed Google Scholar, 4.Guz Y. Montminy M.R. Stein R. Leonard J. Gamer L.W. Wright C.V. Teitelman G. Development. 1995; 121: 11-18Crossref PubMed Google Scholar, 5.Leonard J. Peers B. Johnson T. Ferreri K. Lee S. Montminy M.R. Mol. Endocrinol. 1993; 7: 1275-1283Crossref PubMed Google Scholar, 6.Ahlgren U. Jonsson J. Edlund H. Development. 1996; 122: 1409-1416Crossref PubMed Google Scholar, 7.Offield M.F. Jetton T.L. Labosky P.A. Ray M. Stein R.W. Magnuson M.A. Hogan B.L.M. Wright C.V.E. Development. 1996; 122: 983-995Crossref PubMed Google Scholar, 8.Ahlgren U. Jonsson J. Jonsson L. Simu K. Edlund H. Genes Dev. 1998; 12: 1763-1768Crossref PubMed Scopus (792) Google Scholar, 9.Stoffers D.A. Thomas M.K. Habener J.F. Trends Endocrinol. Metab. 1997; 8: 145-151Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar). Moreover, several laboratories have demonstrated that PDX-1 plays a key role in pancreatic development and that animals and humans lacking PDX-1 have pancreatic agenesis (1.Habener J.F. Stoffers D.A. Proc. Assoc. Am. Physicians. 1998; 110: 12-21PubMed Google Scholar, 2.Edlund H. Diabetes. 1998; 47: 1817-1823Crossref PubMed Scopus (276) Google Scholar, 3.Jonsson J. Ahlgren U. Edlund T. Edlund H. Int. J. Dev. Biol. 1995; 39: 789-798PubMed Google Scholar, 6.Ahlgren U. Jonsson J. Edlund H. Development. 1996; 122: 1409-1416Crossref PubMed Google Scholar, 7.Offield M.F. Jetton T.L. Labosky P.A. Ray M. Stein R.W. Magnuson M.A. Hogan B.L.M. Wright C.V.E. Development. 1996; 122: 983-995Crossref PubMed Google Scholar, 8.Ahlgren U. Jonsson J. Jonsson L. Simu K. Edlund H. Genes Dev. 1998; 12: 1763-1768Crossref PubMed Scopus (792) Google Scholar, 9.Stoffers D.A. Thomas M.K. Habener J.F. Trends Endocrinol. Metab. 1997; 8: 145-151Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 10.Stoffers D.A. Zinkin N.T. Stanojevic V. Clarke W.L. Habener J.F. Nat. Genet. 1997; 15: 106-110Crossref PubMed Scopus (956) Google Scholar). PDX-1 also regulates transcription of other islet genes such as GLUT2, glucokinase (GK), 1The abbreviations used are: GKglucokinaseMODYmaturity- onset diabetes of the youngGSISglucose-stimulated insulin secretionIAPPislet amyloid polypeptideGLUT2glucose transporter 2RIAradioimmunoassayGLP-1glucagon-like peptide-1TPEMtwo-photon excitation microscopyKIC2-ketoisocaproateINSinsulinSOMsomatostatinGLUglucagon and islet amyloid polypeptide (IAPP) (1.Habener J.F. Stoffers D.A. Proc. Assoc. Am. Physicians. 1998; 110: 12-21PubMed Google Scholar, 2.Edlund H. Diabetes. 1998; 47: 1817-1823Crossref PubMed Scopus (276) Google Scholar, 9.Stoffers D.A. Thomas M.K. Habener J.F. Trends Endocrinol. Metab. 1997; 8: 145-151Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 11.Carty M.D. Lillquist J.S. Peshavaria M. Stein R. Soeller W.C. J. Biol. Chem. 1997; 272: 11986-11993Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar, 12.Watada H. Kajimoto Y. Umayahara Y. Matsuoka T. Kaneto H. Fujitani Y. Kamada T. Kawamori R. Yamasaki Y. Diabetes. 1996; 45: 1478-1488Crossref PubMed Google Scholar, 13.Waeber G. Thompson N. Nicod P. Bonny C. Mol. Endocrinol. 1996; 10: 1327-1334Crossref PubMed Scopus (326) Google Scholar, 14.Watada H. Kajimoto Y. Kaneto H. Matsuoka T. Fujitani Y. Miyazaki Ji Yamasaki Y. Biochem. Biophys. Res. Commun. 1996; 229: 746-751Crossref PubMed Scopus (87) Google Scholar). How PDX-1 functions in concert with other islet transcription factors and leads to normal pancreatic exocrine and islet development is unknown. glucokinase maturity- onset diabetes of the young glucose-stimulated insulin secretion islet amyloid polypeptide glucose transporter 2 radioimmunoassay glucagon-like peptide-1 two-photon excitation microscopy 2-ketoisocaproate insulin somatostatin glucagon Mutations in PDX-1 lead to abnormalities in islet function and diabetes in humans and mice (1.Habener J.F. Stoffers D.A. Proc. Assoc. Am. Physicians. 1998; 110: 12-21PubMed Google Scholar, 9.Stoffers D.A. Thomas M.K. Habener J.F. Trends Endocrinol. Metab. 1997; 8: 145-151Abstract Full Text Full Text PDF PubMed Scopus (96) Google Scholar, 15.Stoffers D.A. Ferrer J. Clarke W.L. Habener J.F. Nat. Genet. 1997; 17: 138-139Crossref PubMed Scopus (8) Google Scholar, 16.Hani E.H. Stoffers D.A. Chevre J.C. Durand E. Stanojevic V. Dina C. Habener J.F. Froguel P. J. Clin. Invest. 1999; 104: R41-R48Crossref PubMed Scopus (275) Google Scholar, 17.Macfarlane W.M. Frayling T.M. Ellard S. Evans J.C. Allen L.I. Bulman M.P. Ayres S. Shepherd M. Clark P. Millward A. et al.J. Clin. Invest. 1999; 104: R33-R39Crossref PubMed Scopus (226) Google Scholar, 18.Dutta S. Bonner-Weir S. Montminy M.R. Wright C.V. Nature. 1998; 392: 560Crossref PubMed Scopus (118) Google Scholar, 19.Clocquet A.R. Egan J.M. Stoffers D.A. Muller D.C. Wideman L Chin G.A. Clarke W.L. Hanks J.B. Habener J.F. Elahi D. Diabetes. 2000; 49: 1856-1864Crossref PubMed Scopus (61) Google Scholar). Individuals who are heterozygous for the Pro63fsdelC mutation develop maturity-onset diabetes of the young (MODY4) and have impaired insulin secretion in response to glucose (19.Clocquet A.R. Egan J.M. Stoffers D.A. Muller D.C. Wideman L Chin G.A. Clarke W.L. Hanks J.B. Habener J.F. Elahi D. Diabetes. 2000; 49: 1856-1864Crossref PubMed Scopus (61) Google Scholar). This mutation is thought to reduce PDX-1 activity by creating an alternate internal translation start site that produces a dominant negative isoform of PDX-1 (20.Stoffers D.A. Stanojevic V. Habener J.F. J. Clin. Invest. 1998; 102: 232-241Crossref PubMed Scopus (162) Google Scholar). Some recent reports indicate that mutations in PDX-1 may predispose individuals to late onset type 2 diabetes mellitus (15.Stoffers D.A. Ferrer J. Clarke W.L. Habener J.F. Nat. Genet. 1997; 17: 138-139Crossref PubMed Scopus (8) Google Scholar, 16.Hani E.H. Stoffers D.A. Chevre J.C. Durand E. Stanojevic V. Dina C. Habener J.F. Froguel P. J. Clin. Invest. 1999; 104: R41-R48Crossref PubMed Scopus (275) Google Scholar, 17.Macfarlane W.M. Frayling T.M. Ellard S. Evans J.C. Allen L.I. Bulman M.P. Ayres S. Shepherd M. Clark P. Millward A. et al.J. Clin. Invest. 1999; 104: R33-R39Crossref PubMed Scopus (226) Google Scholar, 21.Hansen L. Urioste S. Petersen H.V. Jensen J.N. Eiberg H. Barbetti F. Serup P. Hansen T. Pedersen O. J. Clin. Endocrinol. Metab. 2000; 85: 1323-1326PubMed Google Scholar). Similarly in murine models both conventional and postnatal, cell-specific inactivation of one PDX-1 allele leads to impaired glucose tolerance (2.Edlund H. Diabetes. 1998; 47: 1817-1823Crossref PubMed Scopus (276) Google Scholar, 6.Ahlgren U. Jonsson J. Edlund H. Development. 1996; 122: 1409-1416Crossref PubMed Google Scholar, 8.Ahlgren U. Jonsson J. Jonsson L. Simu K. Edlund H. Genes Dev. 1998; 12: 1763-1768Crossref PubMed Scopus (792) Google Scholar, 18.Dutta S. Bonner-Weir S. Montminy M.R. Wright C.V. Nature. 1998; 392: 560Crossref PubMed Scopus (118) Google Scholar). In addition, Edlund and co-workers (2.Edlund H. Diabetes. 1998; 47: 1817-1823Crossref PubMed Scopus (276) Google Scholar, 8.Ahlgren U. Jonsson J. Jonsson L. Simu K. Edlund H. Genes Dev. 1998; 12: 1763-1768Crossref PubMed Scopus (792) Google Scholar) demonstrated a striking reduction in GLUT2 expression in the islets of PDX-1 heterozygote mice and suggested that impaired expression of GLUT2 might be a general cause of hyperglycemia and type 2 diabetes. Whereas PDX-1 is crucial for pancreatic development and islet formation during the fetal period, the role of PDX-1 in islet function in the adult is incompletely defined and characterized. To better understand the role of PDX-1 in the postnatal period, we have studied glucose tolerance, insulin secretion, and islet protein expression in mice with one allele of PDX-1 inactivated. The results indicate that PDX-1 is required for normal islet function in the adult and that a modest reduction in PDX-1 impairs normal glucose sensing and insulin secretion. PDX-1-deficient mice were generated as described earlier by fusing the lacZ reporter gene with the N terminus of the PDX-1 coding region (7.Offield M.F. Jetton T.L. Labosky P.A. Ray M. Stein R.W. Magnuson M.A. Hogan B.L.M. Wright C.V.E. Development. 1996; 122: 983-995Crossref PubMed Google Scholar). Studies were performed in 8–40-week-old PDX-1+/− mice and their age-matched, wild type littermates. Glucose tolerance tests and in situpancreas perfusion studies were performed on mice fasted for 14–16 h before testing. Islets were isolated from PDX-1+/+ and PDX-1+/− mice by dissection of the splenic portion of the pancreas followed by digestion with collagenase P (Roche Molecular Biochemicals) (22.Stefan Y. Meda P. Neufeld M. Orci L. J. Clin. Invest. 1987; 80: 175-183Crossref PubMed Scopus (141) Google Scholar). Groups of two pancreases were digested in 2 mg of collagenase/pancreas in Hanks buffered saline for 6–9 min at 37 °C using a wrist-action shaker. Islets for immunoblotting and further analysis were handpicked under microscopic guidance. The dissected pancreas was quickly rinsed in ice-cold phosphate-buffered saline, blotted with filter paper, weighed, and homogenized (Polytron PT 10/35, Brinkmann Instruments) in 1 ml of acid alcohol (23.Davalli A.M. Ogawa Y. Scaglia L. Wu Y.J. Hollister J. Bonner-Weir S. Weir G.C. Diabetes. 1995; 44: 104-111Crossref PubMed Google Scholar). The homogenate was then extracted with an additional 5 ml of acid alcohol for 48 h at 4 °C and centrifuged at 2500 rpm for 30 min. Then, the supernatant was stored at −20 °C until it was assayed for peptides by radioimmunoassay or protein content (Bio-Rad protein assay, Bio-Rad). For Western blot detection of PDX-1, GLUT2, and glucokinase, 60–100 freshly isolated islets were lysed in a buffer containing 5% SDS, 80 mm Tris/HCl, pH 6.8, 5 mm EDTA, and 0.5 mm phenylmethylsulfonyl fluoride (24.Thorens B. Wu Y.J. Leahy J.L. Weir G.C. J. Clin. Invest. 1992; 90: 77-85Crossref PubMed Scopus (146) Google Scholar) by sonication with three 5-s pulses at 20% power (Sonifier II 250, Branson Ultrasonics). Protein concentrations were determined by Bio-Rad DC protein assay. The lysates (5–10 μg of protein/lane) were resolved on SDS-10% polyacrylamide gels and electroblotted onto an Immobilon-P membrane (Millipore). The membranes were incubated with an in-house (from C. V. E. W.) rabbit anti-mouse PDX-1 serum (1:3000), rabbit anti-rat GLUT2 IgG (1:200, Alpha Diagnostic International), or sheep anti-GK IgG (1:5000, gift of M. Magnuson, Vanderbilt University) overnight at 4 °C. Appropriate second antibodies were from Jackson ImmunoResearch Laboratories. Signal detection was performed using an ECL chemiluminescent system (Amersham Biosciences). The signals for each protein on autoradiographs were quantified by densitometry. Densitometry measurements from 4 to 6 separate islet isolations were averaged. Pancreatic tissue was rinsed in ice-cold phosphate-buffered saline and then fixed in 4.0% paraformaldehyde/0.1 m sodium phosphate buffer for 1.5–2 h at room temperature. After fixation, 5-μm cryosections were mounted on silanized slides. Primary and secondary antibodies were diluted in phosphate-buffered saline containing 1% bovine serum albumin and 0.1% Triton X-100. The cryosections were blocked with 5% normal donkey serum (Jackson ImmunoResearch Laboratories) and then incubated with the PDX-1 antiserum (1:5000), the GLUT2 antibody (1:500), guinea pig anti-bovine insulin serum (1:500, INS, Linco Research), sheep anti-somatostatin IgG (1:2000, SOM, American Research Products, Inc.) or guinea pig anti-glucagon serum (1:500, GLU, Linco Research) overnight at 4 °C. PDX-1 and GLUT2 signals were visualized with CY3-conjugated donkey anti-rabbit IgG (1:1000); insulin and glucagon with CY2-conjugated donkey anti-guinea pig IgG (1:1000); and somatostatin with CY5-conjugated donkey anti-sheep IgG (1:1000), all from Jackson ImmunoResearch Laboratories. Digital images of the samples were acquired with a Zeiss LSM410 confocal laser scanning microscope. Populations of α, β, and δ cells in the pancreatic islets were quantified by integrated morphometric analysis. Adjacent sections from 3 to 6 mouse pancreases were labeled for INS, and double-labeled for GLU and SOM as described above. In digital images of individual islets, the area of INS+, GLU+, and SOM+cells was computed using MetaMorph 4.6 software (Universal Imaging). The population of each cell type in the islet was then expressed as a percentage of the total area corresponding to INS+, GLU+, and SOM+ cells. Mouse pancreas was perfused in situ in a humidified and temperature-controlled chamber (37 °C) according to the method of Bonnevie-Nielsen et al. (25.Bonnevie-Nielsen V. Steffes M.W. Lernmark A. Diabetes. 1981; 30: 424-429Crossref PubMed Google Scholar) with some modifications. After ligating the superior mesenteric artery, the hepatic artery, the splenic artery, the right and left renal arteries, and the aorta just below the diaphragm, the celiac trunk was perfused through a catheter placed in the aorta. The effluents were collected through a catheter placed in the portal vein. The perfusion was maintained at 0.5 ml/min with a peristaltic pump (ISM758, Ismatech). The filtered perfusate (0.45 μm) consisted of oxygenated Krebs-Ringer bicarbonate buffer containing 1% bovine serum albumin, 3% Dextran T70 (Amersham Biosciences), and a variable glucose concentration. Arginine·HCl, 2-ketoisocaproate (KIC), and glibenclamide were from Sigma-Aldrich. The glucagon-like peptide GLP-1-(7–36) amide was from Peninsula Laboratories. Mouse plasma insulin was determined by a heterospecies-specific RIA operated with guinea pig anti-rat insulin serum, rat insulin reference standard, normal guinea pig serum, goat anti-guinea pig gammaglobulin serum (Linco Research), and 125I-human insulin (Diagnostic Products). The RIA has a sensitivity (ED90) of 7.5 pg of insulin/ml, which converts to 150 pg of insulin/ml when using 5 μl of mouse plasma/assay tube. Insulin released during perfusion of the mouse pancreas or extracted from the mouse pancreas with acid alcohol was determined by a second heterospecies-specific RIA using insulin antibody-coated tubes (INC Pharmaceuticals, Inc.) in place of the guinea pig anti-rat insulin serum. This RIA has a sensitivity (ED90) of 0.25 ng of insulin/ml, which converts to 0.25 ng of insulin/ml when using 100 μl of pancreatic perfusate/assay tube and 25 ng/ml when using 1 μl of pancreatic extract/assay tube. Glucagon released during perfusion of mouse pancreas or extracted from the mouse pancreas with acid alcohol was measured by RIA with guinea pig anti-glucagon serum, 125I-glucagon, normal guinea pig serum, and goat anti-rabbit gammaglobulin serum (Linco Research). An equilibrium assay with a sensitivity (ED90) of 15 pg of glucagon/ml, which converts to 1.5 ng of glucagon/ml when using a 1 μl of extract/assay tube, was used to determine pancreatic glucagon content. A more sensitive disequilibrium glucagon RIA with an ED90 of 5 pg of glucagon/ml, which converts to 25 pg of glucagon/ml when using 20 μl of perfusate/assay tube, was employed to measure glucagon in the perfusate. IAPP or amylin content in mouse pancreatic extracts was determined using a RIA kit from Phoenix Pharmaceuticals according to the manufacturer's instructions. NAD(P)H imaging was performed by two-photon excitation microscopy (TPEM) as described previously (26.Piston D.W. Knobel S.M. Postic C. Shelton K.D. Magnuson M.A. J. Biol. Chem. 1999; 274: 1000-1004Abstract Full Text Full Text PDF PubMed Scopus (58) Google Scholar). The glucose response of NAD(P)H autoflourescence was quantified by digital image analysis using NIH Image 1.61. Results are expressed as mean ± S.E. The unpaired Student's t test was used for comparisons of wild type mice with PDX-1 heterozygote mice, with the exception of perfusion with glucose gradient where data were compared by repeated measures of the two-way analysis of variance. Differences were considered to be significant at p < 0.05. Mice with one inactivated allele of PDX-1 have impaired glucose clearance during glucose tolerance testing (Fig. 1). At 8 weeks of age, the fasting blood glucose levels were not different in the PDX-1+/− and wild type littermates (PDX-1+/+ males 62 ± 6 mg/dl, n = 5 versus PDX-1+/− males 73 ± 6 mg/dl, n = 5 and PDX-1+/+females 50 ± 7 mg/dl, n = 5 versusPDX-1+/− females 63 ± 4 mg/dl, n = 9). In older mice (16–25 weeks), the fasting glucose levels increased, but they were still statistically insignificant in the male group of mice (PDX-1+/+ males 78 ± 9 mg/dl, n = 6 versus PDX-1+/− males 92 ± 6 mg/dl, n = 13, p = 0.22; and PDX-1+/+ females 63 ± 7, n = 4versus PDX-1+/− females 81 ± 4, n = 10, p < 0.05). However, following glucose administration, PDX-1+/− mice had a pronounced and prolonged rise in their blood glucose compared with PDX-1+/+ mice, and this abnormality became even greater as mice aged (8-week-old mice in Fig. 1, A and D versus 16–25-week-old mice in Fig. 1, B and E). There was a gender-related difference in the glucose clearance, but both males and females had impaired glucose tolerance. The data from IPGTT showed that 30% of PDX-1+/−males (7/23) and 20% of PDX-1+/− females (4/20) had normal glucose tolerance; however, their plasma insulin levels were reduced to levels similar to those measured in the PDX-1+/− littermates with impaired glucose tolerance (Fig. 1C). This finding is somewhat similar to data from human subjects who are heterozygous for mutations in the PDX-1 gene (D76N variant) and who have normal plasma glucose levels, but their plasma insulin levels in response to glucose load are significantly lower (16.Hani E.H. Stoffers D.A. Chevre J.C. Durand E. Stanojevic V. Dina C. Habener J.F. Froguel P. J. Clin. Invest. 1999; 104: R41-R48Crossref PubMed Scopus (275) Google Scholar). More recent data from the study of MODY4 patients (heterozygous for the Pro63fsdelC mutation) also suggests that increased peripheral tissue sensitivity to insulin is a mechanism for maintaining glucose homeostasis in MODY4 individuals (19.Clocquet A.R. Egan J.M. Stoffers D.A. Muller D.C. Wideman L Chin G.A. Clarke W.L. Hanks J.B. Habener J.F. Elahi D. Diabetes. 2000; 49: 1856-1864Crossref PubMed Scopus (61) Google Scholar). After PDX-1+/− mice were phenotyped by IPGTT, the mice with normal glucose tolerance were not included in the studies described below. To investigate the mechanism by which PDX-1+/− mice had lower plasma insulin levels than their wild type littermates, we examined expression of insulin and other islet genes known to be regulated in vitro by PDX-1. As shown in Table I, the decrease in insulin secretion in PDX-1+/− mice is not caused by reduced pancreatic insulin. Rather, the decrease suggests an inability of the PDX-1 heterozygotes to respond to extracellular glucose. Ahlgren et al. (8.Ahlgren U. Jonsson J. Jonsson L. Simu K. Edlund H. Genes Dev. 1998; 12: 1763-1768Crossref PubMed Scopus (792) Google Scholar) reported similar results for pancreatic insulin content of PDX-1+/− mice. In contrast, the pancreatic content of glucagon was increased by 29%, whereas the content of IAPP was reduced by 32% in PDX-1-deficient mice (Table I).Table IPancreatic peptide contentAgeBody wt.Pancreatic wt.Insulin Content/Pancreatic ProteinGlucagon Content/Pancreatic ProteinIAPP Content/Pancreatic Proteinwkgmgμg/mgng/mgng/mgPDX-125–4036.1 ± 1.8220 ± 1117.5 ± 1.6365 ± 31411 ± 46+/+(n = 14)(n = 14)(n = 9)(n = 13)(n = 14)PDX-125–4035.8 ± 2.2199 ± 1518.0 ± 3.5470 ± 40a278 ± 33a+/−(n = 16)(n = 16)(n = 12)(n = 10)(n = 15)a p < 0.05 compared with PDX-1 +/+ mice. Open table in a new tab a p < 0.05 compared with PDX-1 +/+ mice. Immunohistochemical evaluation of pancreases from 16–18-week-old PDX-1+/− mice revealed normal islet morphology (Fig. 2, a–f) and a normal ratio of β to α or δ cells. However, the expression of glucose transporter GLUT2 was reduced quite dramatically compared with wild type islets (Fig. 2, c and d). Similar changes in GLUT2 expression were seen in 7-week-old mice (data not shown). Ahlgren et al. (8.Ahlgren U. Jonsson J. Jonsson L. Simu K. Edlund H. Genes Dev. 1998; 12: 1763-1768Crossref PubMed Scopus (792) Google Scholar) reported similar decreases in GLUT2 expression in PDX-1-deficient mice. The reduction in GLUT2 expression was somewhat variable from islet to islet, and in some islets GLUT2 expression was only slightly reduced. The expression of PDX-1, GLUT2, and GK in isolated islets of PDX-1+/− mice was further evaluated by immunoblot analysis (Fig. 2g). Expression of PDX-1 and GLUT2 was reduced to about 68 and 55%, respectively, compared with wild type mice, whereas GK expression was relatively unchanged. To define the molecular mechanisms by which PDX-1 deficiency alters glucose-stimulated insulin secretion (GSIS) in pancreatic β cells, the pancreas of PDX-1+/+ and PDX-1+/− mice were challenged with a linear glucose gradient (Fig. 3) using an in situ pancreas perfusion technique. This technique uses the native vasculature to deliver changes in glucose to the pancreatic islet and avoids the need for enzymatic digestion during islet isolation. Whereas the amount of insulin released after glucose stimulation was less in PDX-1+/− animals, the Km for insulin release in response to the glucose gradient during pancreas perfusion was similar in wild type (12.0 ± 1.3 mm) and heterozygote (8.0 ± 0.4 mm) mice. To further examine molecular events in the GSIS pathway, we examined glucose-induced NAD(P)H responses of isolated islets using TPEM. Islets isolated from PDX-1+/+ and PDX-1+/− mice showed a very uniform autofluorescent signal during exposure to 1 mm glucose (Fig. 4a). In the case of wild type islets, this signal was greatly enhanced when glucose was increased from 1 mm to 20 mm (Fig. 4b). However, islets of PDX-1 heterozygotes generated about 30% less NAD(P)H in response to 20 mm glucose compared with islets of wild type mice (Fig. 4c).Figure 4NAD(P)H response from islets measured by TPEM. a, NAD(P)H autofluorescent images of islets (PDX-1+/+ islet (left) and PDX-1+/−islet (right)) at 1 mm glucose. b, same two islets after 5-min challenge with 20 mm glucose. The bar at the bottom corresponds to 50 μm. c, response of PDX-1+/+ (n = 35) and PDX-1+/− (n = 35) islets to 20 mm glucose (**, p < 0.01 compared with controls). Data for males and females were not significantly different and were combined.View Large Image Figure ViewerDownload Hi-res image Download (PPT) To examine the molecular events in insulin secretion, such as cytoplasmic and mitochondrial metabolism of glucose, closure of KATP+ channels, opening of voltage-dependent calcium channels, Ca2+influx, mobilization of Ca2+ from intracellular stores (ryanodine receptor and inositol 1,4,5-trisphosphate receptor-mediated Ca2+ release), and exocytosis of insulin from secretory granules, we measured insulin secretion in response to glucose and non-glucose secretagogues from the in situ perfused pancreas. Changes in insulin output with respect to the time course of pancreas stimulation are plotted in Figs.5A,6, and 7.Figure 6Insulin secretory response to glucagon-like peptide 1 and glibenclamide in the pancreas perfused in situ. Insulin secretion from the pancreas of PDX-1 heterozygotes (closed circles) and their wild type littermates (open circles) was analyzed in the male and female group of mice. A, insulin secretion in males. The integrated response to 10 nm GLP-1 in the presence of 5.6 mm glucose was 32.3 ± 7.5 ng of insulin in PDX-1+/+ mice (n = 4) versus68.5 ± 11.1 ng of insulin in PDX-1+/− mice (n = 5), p = 0.038. The integrated response to 10 nm glibenclamide in the presence of 5.6 mm glucose was 26.1 ± 3.5 ng of insulin in PDX-1+/+ mice (n = 4) versus30.7 ± 6.7 ng of insulin in PDX-1+/− mice (n = 5), p = 0.593. B, insulin secretion in females. The integrated response to 20 nm GLP-1 in the presence of 5.6 mm glucose was 4.9 ± 1.8 ng of insulin in PDX-1+/+(n = 3) mice versus 13.3 ± 2.5 ng of insulin in PDX-1+/− mice (n = 3), p = 0.052. The integrated response to 10 nmglibenclamide in the presence of 5.6 mm glucose was 14.8 ± 3.2 ng of insulin in PDX-1+/+ mice (n = 3) versus 15.7 ± 2.1 ng of insulin in PDX-1+/− mice (n = 3), p = 0.831.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Figure 7Insulin secretory response to 2-ketoisocaproate and potassium chloride in the pancreas perfused in situ. Insulin secretion from the pancreas of PDX-1 heterozygotes (closed circles, n = 5) and their wild type littermates (open circles, n = 8) was analyzed in response to glucose, KIC, and KCl. Data for males and females were not significantly different and were combined. The integrated response to 16.7 mm glucose was 55.4 ± 5.6 ng of insulin in PDX-1+/+ mice versus 29.0 ± 2.9 ng of insulin in PDX-1+/− mice, p = 0.005. Integrated response to 10 mm KIC in the presence of 2.8 mmglucose was 120 ± 19 ng of insulin in PDX-1+/+ mice versus 46.8 ± 8.2 ng of insulin in PDX-1+/− mice, p = 0.016. Integrated response to 20 mm KCl in the presence of 2.8 mmglucose was 134 ± 20 ng of insulin in PDX-1+/+ mice versus 45.4 ± 9.1 ng of insulin in PDX-1+/− mice, p = 0.006.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Similar to in vivo results, the pancreas of PDX-1+/− mice released about 45% less insulin than the pancreas of wild type littermates in response to 16.7 mmglucose (Figs. 5A and 7). To localize which steps in the insulin secretion pathway are PDX-1-regulated, we examined insulin secretion in response to several secretagogues, either coupled to glucose or acting independently, such as arginine, GLP-1, glibenclamide, KIC, and KCl. The magnitude of
Referência(s)