Portal de Periódicos da CAPES

Glial Cell Line-Derived Neurotrophic Factor Increases β-Cell Mass and Improves Glucose Tolerance

2008; Elsevier BV; Volume: 134; Issue: 3 Linguagem: Inglês

10.1053/j.gastro.2007.12.033

1528-0012

Simon M. Mwangi, Mallappa Anitha, Chaithanya Mallikarjun, Xiaokun Ding, Manami Hara, Alexander Parsadanian, Christian P. Larsen, Peter M. Thulé, Shanthi V. Sitaraman, Frank A. Anania, Shanthi Srinivasan,

Diet, Metabolism, and Disease

Background & Aims: Pancreatic β-cell mass increases in response to increased demand for insulin, but the factors involved are largely unknown. Glial cell line-derived neurotrophic factor (GDNF) is a growth factor that plays a role in the development and survival of the enteric nervous system. We investigated the role of GDNF in regulating β-cell survival. Methods: Studies were performed using the β-TC-6 pancreatic β-cell line, isolated mouse pancreatic β cells, and in vivo in transgenic mice that overexpress GDNF in pancreatic glia. GDNF receptor family α1 and c-Ret receptor expression were assessed by reverse-transcription polymerase chain reaction and immunofluorescence microscopy. Apoptosis was evaluated by assessing caspase-3 cleavage. Phosphoinositol-3-kinase signaling pathway was analyzed by Akt phosphorylation. Glucose homeostasis was assessed by performing intraperitoneal glucose tolerance tests. Insulin sensitivity was assessed using intraperitoneal injection of insulin. Results: We demonstrate the presence of receptors for GDNF, GFRα1, and c-Ret on β cells. GDNF promoted β-cell survival and proliferation and protected them from thapsigargin-induced apoptosis (P<.0001) in vitro. Exposure of β-cells to GDNF also resulted in phosphorylation of Akt and GSK3β. Transgenic mice that overexpress GDNF in glia exhibit increased β-cell mass, proliferation, and insulin content. No differences in insulin sensitivity and c-peptide levels were noted. Compared with wild-type mice, GDNF-transgenic mice have significantly lower blood glucose levels and improved glucose tolerance (P<.01). GDNF-transgenic mice are resistant to streptozotocin-induced β-cell loss (P<.001) and subsequent hyperglycemia. Conclusions: We demonstrate that over expression of GDNF in pancreatic glia improves glucose tolerance and that GDNF may be a therapeutic target for improving β-cell mass. Background & Aims: Pancreatic β-cell mass increases in response to increased demand for insulin, but the factors involved are largely unknown. Glial cell line-derived neurotrophic factor (GDNF) is a growth factor that plays a role in the development and survival of the enteric nervous system. We investigated the role of GDNF in regulating β-cell survival. Methods: Studies were performed using the β-TC-6 pancreatic β-cell line, isolated mouse pancreatic β cells, and in vivo in transgenic mice that overexpress GDNF in pancreatic glia. GDNF receptor family α1 and c-Ret receptor expression were assessed by reverse-transcription polymerase chain reaction and immunofluorescence microscopy. Apoptosis was evaluated by assessing caspase-3 cleavage. Phosphoinositol-3-kinase signaling pathway was analyzed by Akt phosphorylation. Glucose homeostasis was assessed by performing intraperitoneal glucose tolerance tests. Insulin sensitivity was assessed using intraperitoneal injection of insulin. Results: We demonstrate the presence of receptors for GDNF, GFRα1, and c-Ret on β cells. GDNF promoted β-cell survival and proliferation and protected them from thapsigargin-induced apoptosis (P<.0001) in vitro. Exposure of β-cells to GDNF also resulted in phosphorylation of Akt and GSK3β. Transgenic mice that overexpress GDNF in glia exhibit increased β-cell mass, proliferation, and insulin content. No differences in insulin sensitivity and c-peptide levels were noted. Compared with wild-type mice, GDNF-transgenic mice have significantly lower blood glucose levels and improved glucose tolerance (P<.01). GDNF-transgenic mice are resistant to streptozotocin-induced β-cell loss (P<.001) and subsequent hyperglycemia. Conclusions: We demonstrate that over expression of GDNF in pancreatic glia improves glucose tolerance and that GDNF may be a therapeutic target for improving β-cell mass. Type 1 and type 2 diabetes are characterized by loss or dysfunction of β cells. The rate of β-cell proliferation, hyperplasia, neogenesis, and death combine to determine β-cell mass1Bonner-Weir S. Life and death of the pancreatic beta cells.Trends Endocrinol Metab. 2000; 11: 375-378Abstract Full Text Full Text PDF PubMed Scopus (247) Google Scholar and is influenced by exposure to insulin, insulin-like growth factor I, glucose, glucagon-like peptide-1, growth hormone, and several other growth factors.2Nielsen J.H. Galsgaard E.D. Moldrup A. et al.Regulation of β-cell mass by hormones and growth factors.Diabetes. 2001; 50: S25-S29Crossref PubMed Google Scholar Among signaling cascades, the significant role of the phosphoinositide 3-kinase (PI-3-K)/Akt pathway in enhancing β-cell proliferation and survival has been recognized.3Bernal-Mizrachi E. Wen W. Stahlhut S. et al.Islet β-cell expression of constitutively active Akt1/PKB α induces striking hypertrophy, hyperplasia, and hyperinsulinemia.J Clin Invest. 2001; 108: 1631-1638Crossref PubMed Scopus (345) Google Scholar, 4Tuttle R.L. Gill N.S. Pugh W. et al.Regulation of pancreatic β-cell growth and survival by the serine/threonine protein kinase Akt1/PKBα.Nat Med. 2001; 7: 1133-1137Crossref PubMed Scopus (438) Google Scholar Pancreatic β cells share several biologic properties with neuronal cells including expression of the neuronal differentiation factor neuroD/β2 and dependence on nerve growth factor in vitro.5Pierucci D. Cicconi S. Bonini P. et al.NGF-withdrawal induces apoptosis in pancreatic β cells in vitro.Diabetologia. 2001; 44: 1281-1295Crossref PubMed Scopus (60) Google Scholar Moreover, multipotent precursors that generate both neural and pancreatic lineages have been isolated from pancreatic ductal cells and islets,6Seaberg R.M. Smukler S.R. Kieffer T.J. et al.Clonal identification of multipotent precursors from adult mouse pancreas that generate neural and pancreatic lineages.Nat Biotechnol. 2004; 22: 1115-1124Crossref PubMed Scopus (475) Google Scholar suggesting that neurotrophic factors such as glial cell line-derived neurotrophic factor (GDNF) might play a role in β-cell growth and survival. GDNF is a neurotrophic factor that plays a critical role in the development and survival of the enteric nervous system.7Heuckeroth R.O. Lampe P.A. Johnson E.M. et al.Neurturin and GDNF promote proliferation and survival of enteric neuron and glial progenitors in vitro.Dev Biol. 1998; 200: 116-129Crossref PubMed Scopus (190) Google Scholar GDNF mediates its functions through binding to a multicomponent receptor complex consisting of the Ret receptor tyrosine kinase and the glycosylphosphoinositol-anchored coreceptor GFRα1.8Airaksinen M.S. Saarma M. The GDNF family: signalling, biological functions and therapeutic value.Nat Rev Neurosci. 2002; 3: 383-394Crossref PubMed Scopus (1450) Google Scholar Although the role of GDNF in pancreatic β-cell growth and survival is not known, reports of increased expression of GDNF and other neurotrophic factors in the proximity of pancreatic β cells following islet injury9Teitelman G. Guz Y. Ivkovic S. et al.Islet injury induces neurotrophin expression in pancreatic cells and reactive gliosis of peri-islet Schwann cells.J Neurobiol. 1998; 34: 304-318Crossref PubMed Scopus (56) Google Scholar raises the possibility that GDNF and other neurotrophic factors might be involved in islet repair. In the present study, we examined the role of GDNF in the regulation of β-cell proliferation and survival and the physiologic consequence on glucose homeostasis. We show that pancreatic β cells express receptors for GDNF and that GDNF supports the survival and proliferation of β cells in vitro. Using transgenic (tg) mice that over express GDNF in glial cells, we show that GDNF promotes increased β-cell mass, improves glycemic control, and protects against β-cell destruction in vivo. The following antibodies were used for immunocytochemistry: rabbit polyclonal antibodies to GDNF (D-20) and GFRα1 (H-70) (Santa Cruz Biotechnology, Santa Cruz, CA), human c-Ret (R787: Immuno-Biological Laboratories, Takasaki-shi, Gunma, Japan), S-100β (BD Pharmingen, San Diego, CA), cleaved caspase-3 (Cell Signaling Technologies, Danvers, MA), and guinea pig anti-insulin (Zymed Laboratories, San Francisco, CA) were used at 1:50 dilution. Biotin-conjugated donkey anti-rabbit IgG secondary antibody and peroxidase-conjugated streptavidin (Jackson ImmunoResearch Laboratories, Westgrove, PA), Alexa Fluor 488 and 594 donkey anti-guinea pig, and anti-rabbit IgG (Molecular Probes, Eugene, OR) and Ki67 (Novocastra, Newcastle, United Kingdom) were used at a 1:500 dilution. For Western blotting, rabbit polyclonal antibodies to phospho-Akt (ser473), Akt, and phospho-GSK3β, GSK3β, and cleaved caspase-3 (Cell Signaling Technologies) were used at a 1:1000 dilution; streptavidin-horseradish peroxidase-conjugated anti-rabbit and anti-mouse IgG (Cell Signaling Technologies) at 1:2500 dilution; and mouse monoclonal antibody to β-actin (Sigma, St. Louis, MO) at 1:5000 dilution. In vivo studies were conducted in littermates obtained from crossing CF1 wild-type (WT) mice with GDNF-transgenic (GDNF-tg) mice on a CF1 background generated at Washington University, St. Louis, MO. GDNF-tg mice are engineered to overexpress GDNF in glial cells under the control of the glial fibrillary acidic protein promoter.10Zhao Z. Alam S. Oppenheim R.W. et al.Overexpression of glial cell line-derived neurotrophic factor in the CNS rescues motoneurons from programmed cell death and promotes their long-term survival following axotomy.Exp Neurol. 2004; 190: 356-372Crossref PubMed Scopus (94) Google Scholar The genotypes of the mice were determined by polymerase chain reaction (PCR) using DNA extracted from mouse tail using the REDExtract-N-Amp Tissue PCR Kit (Sigma-Aldrich CO, St. Louis, MO) according to recommended procedure. Forward (5′-AGACGCATCACCTCCGCT-3′) and reverse (5′-TGACGTCATCAAACTGGTCAGG-3′) primers designed to only amplify the transgene sequence were used. Mouse insulin I promoter (MIP)-green fluorescent protein (GFP) tg mice11Hara M. Wang X. Kawamura T. et al.Transgenic mice with green fluorescent protein-labeled pancreatic β cells.Am J Physiol Endocrinol Metab. 2003; 284: E177-E183PubMed Google Scholar were used to isolate primary β cells for in vitro studies. These mice express GFP in pancreatic β cells under the control of the MIP. The mice were used at 8–10 weeks of age. All animal studies were approved by the Emory University Animal Care and Use Committee. Islets were isolated using a modification of the procedure described by Bernal-Mizrachi et al.3Bernal-Mizrachi E. Wen W. Stahlhut S. et al.Islet β-cell expression of constitutively active Akt1/PKB α induces striking hypertrophy, hyperplasia, and hyperinsulinemia.J Clin Invest. 2001; 108: 1631-1638Crossref PubMed Scopus (345) Google Scholar Pancreata were digested for 40 minutes at 37°C with 2 mg/mL collagenase (Type IV; Gibco BRL, Grand Island, NY) and islets isolated by density gradient centrifugation using 1.108, 1.096, and 1.037 islet gradient (Mediatech Inc., Herndon, VA). To obtain single cells, islets were digested with 0.01% trypsin in modified Eagle’s medium and dispersed by gently pipetting up and down.3Bernal-Mizrachi E. Wen W. Stahlhut S. et al.Islet β-cell expression of constitutively active Akt1/PKB α induces striking hypertrophy, hyperplasia, and hyperinsulinemia.J Clin Invest. 2001; 108: 1631-1638Crossref PubMed Scopus (345) Google Scholar β-TC-6 cells (ATCC, Manassas, VA, U.S.A.) were cultured in Dulbecco’s modified Eagle medium (DMEM) (ATCC) supplemented with 15% fetal bovine serum. For GDNF stimulation studies, cells were serum deprived for 48 hours, followed by incubation with DMEM alone or DMEM supplemented with different concentrations of GDNF. First-strand complementary DNA (cDNA) synthesized using the Omniscript reverse transcription kit (Qiagen GmbH, Hilden, Germany) from RNA isolated using the RNeasy Mini kit (Qiagen) was subjected to 40 cycles of PCR amplification. The primers used included mouse GFRα1 forward (5′-ATGAAGAACGAGAGAGGCCCAA-3′) and reverse (5′-ACTCTGGCTGGCAGTTGGTAAA-3′) primers, Ret forward (5′-TACCGTACACGGCTGCATGAGAAT-3′) and reverse (5′-ATGTGGAAGTGGTAGAAGGTGCCA-3′) primers, and GDNF forward (5′-TCGATATTGCAGCGGTTCCTGT-3′) and reverse (5′-ACATCCACACCGTTTAGCGGAA-3′) primers. All primers used were designed to span at least an intron and thus are able to discriminate between products amplified from messenger RNA (mRNA) and genomic DNA. β-TC-6 cells, isolated pancreatic β cells, and pancreatic sections from frozen and paraffin-embedded tissues were stained for immunofluorescence microscopy as previously described12Mwangi S. Anitha M. Fu H. et al.Glial cell line-derived neurotrophic factor-mediated enteric neuronal survival involves glycogen synthase kinase-3β phosphorylation and coupling with 14-3-3.Neuroscience. 2006; 143: 241-251Crossref PubMed Scopus (20) Google Scholar and analyzed on a Zeiss Axioskop 2 plus fluorescent microscope (Carl Zeiss Werk, Gottingen, Germany) mounted with an AxioCam MRc 5 camera. Images were taken with the aid of the Axiovision (Rel 4.5) software (Carl Zeiss Imaging System). Cells were also scanned with an LSM 510 laser scanning confocal microscope (Zeiss, Heidelberg, Germany). Apoptosis was assessed either by Western blotting for cleaved caspase-3 or cleaved caspase-3 with immunofluorescence microscopy.12Mwangi S. Anitha M. Fu H. et al.Glial cell line-derived neurotrophic factor-mediated enteric neuronal survival involves glycogen synthase kinase-3β phosphorylation and coupling with 14-3-3.Neuroscience. 2006; 143: 241-251Crossref PubMed Scopus (20) Google Scholar, 13Anitha M. Chandrasekharan B. Salgado J.R. et al.Glial-derived neurotrophic factor modulates enteric neuronal survival and proliferation through neuropeptide y.Gastroenterology. 2006; 131: 1164-1178Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar Cell proliferation in cultured cells and pancreatic tissue was assessed by immunofluorescence microscopy13Anitha M. Chandrasekharan B. Salgado J.R. et al.Glial-derived neurotrophic factor modulates enteric neuronal survival and proliferation through neuropeptide y.Gastroenterology. 2006; 131: 1164-1178Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar using an anti-Ki67 polyclonal antibody.14Teta M. Long S.Y. Wartschow L.M. et al.Very slow turnover of β cells in aged adult mice.Diabetes. 2005; 54: 2557-2567Crossref PubMed Scopus (393) Google Scholar Western blotting was performed as previously described.12Mwangi S. Anitha M. Fu H. et al.Glial cell line-derived neurotrophic factor-mediated enteric neuronal survival involves glycogen synthase kinase-3β phosphorylation and coupling with 14-3-3.Neuroscience. 2006; 143: 241-251Crossref PubMed Scopus (20) Google Scholar A semiquantitative measurement of band density was performed using Scion Image for Windows software (Scion Corp, MD). Pancreata were frozen in Tissue-Tek OCT compound (Sakura Finetek, Torrance, CA) or fixed in 10% formalin solution and embedded in paraffin using standard techniques. Frozen sections were fixed in 4% paraformaldehyde and paraffin sections processed according to suggested protocols (Cell Signaling Technologies). Staining was performed according to standard protocols using the Histomouse-SP (AEC broad spectrum) kit (Zymed Laboratories). Four, 5-μm pancreas sections (separated by 200 μm)/per mouse were used to assess β-cell mass as previously described.3Bernal-Mizrachi E. Wen W. Stahlhut S. et al.Islet β-cell expression of constitutively active Akt1/PKB α induces striking hypertrophy, hyperplasia, and hyperinsulinemia.J Clin Invest. 2001; 108: 1631-1638Crossref PubMed Scopus (345) Google Scholar Images were taken after insulin staining, and islet size, islet number, β-cell size, and total areas of the sections were determined using the Image-Pro Plus 5.0 software (Media Cybernetics, Silver Spring, MD). The percentage of β-cell area in each pancreas was then determined. Pancreata were homogenized using acid-alcohol as previously described15Montana E. Bonner-Weir S. Weir G.C. Beta cell mass and growth after syngeneic islet cell transplantation in normal and streptozocin diabetic C57BL/6 mice.J Clin Invest. 1993; 91: 780-787Crossref PubMed Scopus (180) Google Scholar and insulin levels measured by radioimmunoassay at Linco Diagnostic Services (St. Louis, MO). Age-matched WT and GDNF-tg mice were fasted for 6 hours and baseline blood glucose levels measured with the aid of an Accu-Check Advantage blood glucose meter (Roche, Mannheim, Germany) using blood collected from the tail vein. To test for glucose tolerance, the mice were injected intraperitoneally with 2 mg glucose/g body weight in sterile phosphate-buffered saline, and blood glucose levels were measured 30, 60, and 120 minutes after injection.3Bernal-Mizrachi E. Wen W. Stahlhut S. et al.Islet β-cell expression of constitutively active Akt1/PKB α induces striking hypertrophy, hyperplasia, and hyperinsulinemia.J Clin Invest. 2001; 108: 1631-1638Crossref PubMed Scopus (345) Google Scholar For the insulin sensitivity test, fasted mice were injected intraperitoneally with 0.75 U/kg human rapid insulin (Eli Lilly Co, Indianapolis, IN), and blood glucose levels were measured 15, 30, 60, 90, and 120 minutes after injection.16Lauro D. Kido Y. Castle A.L. et al.Impaired glucose tolerance in mice with a targeted impairment of insulin action in muscle and adipose tissue.Nat Genet. 1998; 20: 294-298Crossref PubMed Scopus (118) Google Scholar For in vivo insulin secretion, 6-hour fasted mice were injected with 3 mg glucose/kg body weight, and plasma insulin levels were measured 0 and 2.5 minutes postinjection using a rat/mouse insulin ELISA kit (Linco Diagnostic Services). For in vitro insulin secretion analysis, isolated islets were cultured overnight in Ham’s F10 medium, and 40 handpicked islets were cultured per well for 2 hours in Kreb’s Ringer bicarbonate buffer containing 1.67 mmol/L glucose followed by 1 hour in buffer containing 1.67 mmol/L or 20 mmol/L glucose. The culture media were collected, the islets lysed in acid-alcohol, and insulin concentrations measured by insulin ELISA (Linco Diagnostic Services). Values are expressed as percent of islet content relative to basal secretion. WT and GDNF-tg littermates were injected intraperitoneally with 75 mg/kg streptozotocin (STZ) followed by another 75 mg/kg STZ after 12 hours and their blood glucose levels measured once daily to monitor onset of hyperglycemia. Hyperglycemia was defined as postprandial blood glucose greater than 145 mg/dL. At the end of the experiments, the mice were killed; pancreata were embedded in paraffin, sectioned, and stained for insulin; and islet images were obtained. The outlines of islets were marked, the images were thresholded based on insulin staining intensity, and threshold area relative to total islet area was calculated with the aid of the MetaMorph Offline version 7.0r3 software (Molecular Devices Corp., Downingtown, PA). To assess apoptosis, pancreas sections from STZ-treated mice were made, and apoptosis was assessed by cleaved caspase-3 immunofluorescence microscopy with TUNEL staining.12Mwangi S. Anitha M. Fu H. et al.Glial cell line-derived neurotrophic factor-mediated enteric neuronal survival involves glycogen synthase kinase-3β phosphorylation and coupling with 14-3-3.Neuroscience. 2006; 143: 241-251Crossref PubMed Scopus (20) Google Scholar All statistical analyses were conducted using the GraphPad Prism software version 3.00 for Windows (GraphPad Software, San Diego, CA). Data were tested for normality and subjected to t tests or 1-way ANOVA with Tukey posttest. To understand the role of GDNF in β-cell growth and survival, we first analyzed by RT-PCR the expression of its receptors in cells of the insulin-secreting mouse pancreatic β-TC-6 cell line and mouse pancreatic β cells. As seen in Figure 1A, both cell types express significant amounts of GFRα1 and Ret receptor mRNA. The expression of the receptors was further analyzed by immunofluorescence microscopy using receptor-specific antibodies. High expression of both receptors was observed in cultured β-TC-6 cells and isolated mouse pancreatic β cells (Figure 1B and 1C). The localization of the receptors on the surface of the cells was confirmed by laser confocal microscopy (Figure 1D). GDNF is known to be a trophic factor for neurons, but its role in β-cell growth and survival is not known. We thus tested whether GDNF (10–500 ng/mL) could prevent apoptosis and promote proliferation of primary β cells and β-TC-6 cells. Apoptosis was assessed by blotting for cleaved caspase-3 and cleaved caspase-3 immunocytochemistry. GDNF treatment for 72 hours suppressed caspase-3 cleavage in β-TC-6 cells in a dose-dependent fashion (Figure 2A and 2B). To investigate further the ability of GDNF to promote β-cell survival, we assessed the ability of GDNF to block the effects of thapsigargin, a proapoptotic stimulus for β cells. GDNF significantly reduced thapsigargin-induced apoptosis in β-TC-6 cells compared with vehicle-treated cells (Figure 2C). Similarly, isolated mouse pancreatic β cells from MIP-GFP mice revealed a significant reduction in apoptosis when cultured for 48 hours in the presence of GDNF (100–500 ng/mL) compared with vehicle (Figure 2D). We next assessed whether GDNF could promote the proliferation of β cells in vitro. β-TC-6 cells were cultured in serum-free medium for 48 hours followed by 24 hours in the presence or absence of GDNF and assessed for proliferation by staining for the proliferation marker Ki67.17Akerblom B. Anneren C. Welsh M. A role of FRK in regulation of embryonal pancreatic β-cell formation.Mol Cell Endocrinol. 2007; 270: 73-78Crossref PubMed Scopus (13) Google Scholar GDNF increased the number of Ki67-positive cells in a dose-dependent fashion (Figure 2E). Taken together, these data suggest an important survival and proliferation role for GDNF in β cells in vitro. Because GDNF signals through the PI-3-K/Akt pathway in neurons to promote cell survival, we examined the possible activation of this pathway in β cells by GDNF. β-TC-6 cells precultured in serum-free medium for 48 hours were stimulated with vehicle only (no GDNF), GDNF (100 ng/mL), or serum (15%) for 30 minutes and analyzed by Western blotting for Akt phosphorylated at ser473. Significant amounts of phospho-Akt were detected in cells cultured in the presence of GDNF, but were absent in cells cultured in vehicle only (P < .001; Figure 3A). We also investigated the ability of GDNF to stimulate the phosphorylation of glycogen synthase kinase-3β (GSK3β), a downstream target of Akt, in these cells. More than 2.7-fold more phospho-GSK3β (ser9) was detected by Western blotting in β-TC-6 cells cultured for 30 minutes with GDNF than in cells cultured in vehicle only (P < .001; Figure 3B). This increase was lost when the cells were cultured with GDNF in the presence of the PI3-K inhibitor LY294002. These data thus demonstrate that GDNF activates the PI-3-K/Akt-signaling pathway in β cells in vitro. Having demonstrated the effects of GDNF in vitro, we next examined the effects of GDNF in vivo using a GDNF-tg mouse in which the overexpression of GDNF has been demonstrated in astrocytes in the brain and spinal cord,10Zhao Z. Alam S. Oppenheim R.W. et al.Overexpression of glial cell line-derived neurotrophic factor in the CNS rescues motoneurons from programmed cell death and promotes their long-term survival following axotomy.Exp Neurol. 2004; 190: 356-372Crossref PubMed Scopus (94) Google Scholar and in glia the peripheral nervous system. Experiments were performed on GDNF-tg mice and their WT littermates. Both male and female GDNF-tg mice have approximately 20% lower body weight compared with their WT littermates (weight in grams at 8 weeks: WT male, 35.1 ± 1.4, GDNF-tg male, 30.8 ± 0.7, P < .05; WT-female, 29.5 ± 0.8, GDNF-tg female, 23.7 ± 1.2, P < .01). Despite their lower body weights, GDNF-tg mice have normal eye opening, fur growth, and weaning and reproductive capacity, similar to WT mice. GDNF-tg mice have a slight tremor at birth that disappears by 2–3 weeks of age. To confirm the overexpression of GDNF in the pancreas of GDNF-tg mice, we compared by RT-PCR the levels of GDNF mRNA in pancreas from WT and GDNF-tg mice. GDNF-tg mice had more than 2.2-fold more GDNF mRNA than WT mice (P < .01; Figure 4A). To identify the specific areas of the pancreas in which GDNF is expressed, sections of pancreas from both WT and GDNF-tg mice were immunofluorescently stained for GDNF. As seen in Figure 4B, GDNF expression was localized to glial cells identified by the glial specific marker S-100β and was enhanced in GDNF-tg mice compared with WT mice. To assess the effects of increased GDNF expression in the pancreas, successive sections taken 200 μm apart from pancreata from 8 week-old WT and GDNF-tg mice were stained for insulin. Examination of these sections revealed significantly more islets in GDNF-tg mice than their WT littermates (Figure 5A). A more detailed morphometric analysis to assess β-cell mass showed a 2.5-fold increase in β-cell area/pancreas in GDNF-tg mice compared with WT mice (P < .001; Figure 5A). The number of islets per unit pancreatic tissue was higher in GDNF-tg mice compared with WT mice (P < .05; Figure 5B). The size of individual islets was higher in GDNF-tg mice compared with WT mice (P < .05; Figure 5C). However, individual β-cell size appeared similar (P > .05; n = 26, Figure 5D), and no change in the islet architecture was noted, with a similar distribution of β and α cells. Assessment of pancreatic insulin content also revealed a 3.2-fold higher insulin content in GDNF-tg mice than in WT mice (P < .05; Figure 5E). To understand the processes accounting for the increased β-cell mass in GDNF-tg mice, WT and GDNF-tg mice were stained for insulin and Ki67 to assess β-cell proliferation (Figure 5F). A significantly higher number of Ki67+/insulin+ cells were observed in islets of GDNF-tg mice compared with WT mice (P < .0001, n = 3). To assess the functional effect of increased β-cell mass on glucose homeostasis, glucose tolerance tests were conducted in male and female WT and GDNF-tg mice. Fasting blood glucose levels of GDNF-tg mice of both sexes were significantly (P < .05) lower than those of WT mice (Figure 6A and 6B). Following intraperitoneal glucose administration, blood glucose levels remained significantly lower in GDNF-tg mice than in WT mice at all time points, except at the 120-minute time point in females (Figure 6A and 6B). Intraperitoneal glucose tolerance curves were compared in weight-matched WT and GDNF-tg mice, and the impairment in glucose tolerance persisted (blood glucose level 30 minutes postinjection of glucose [mg/dL]), WT: 252 ± 34, GDNF-tg: 175 ± 12, P < .01; WT, weight (grams): 33.73 ± 0.9, GDNF-tg weight: 32.94 ± 0.36, n = 5, P >.05. We also assessed whether there was a difference in insulin sensitivity between WT and GDNF-tg mice using an intraperitoneal insulin sensitivity test and found no difference (Figure 6C). To investigate the factors contributing to the improved glucose tolerance, c-peptide and insulin levels were assessed in plasma samples collected following 6 hours of fasting. GDNF-tg mice had similar plasma c-peptide and insulin levels to WT mice (P > .05; Figure 6D and 6E). We then assessed glucose-stimulated insulin release in vivo. We found that 2.5 minutes after intraperitoneal glucose administration, plasma insulin increase relative to baseline was 144.7% ± 28.75% in GDNF-tg mice compared with 30.85% ± 15.52% in WT mice (P < .01; Figure 6F). These data suggest that the increased β-cell mass and associated increased insulin release in GDNF-tg mice results in improved glucose tolerance. We also tested glucose-stimulated insulin secretion in vitro in islets isolated from WT and GDNF-tg mice. Glucose-stimulated insulin secretion was assessed in response to 20 mmol/L glucose compared with baseline 1.67 mmol/L glucose. We found no difference in in vitro insulin release in islets isolated from WT vs GDNF-tg mice (percent insulin release, WT: 109.2 ± 27.68, GDNF-tg: 89.3 ± 52.61, n = 4, P > .05). Administration of multiple low doses of STZ produces diabetes by the selective loss of β cells.18O’Brien B.A. Harmon B.V. Cameron D.P. et al.Beta-cell apoptosis is responsible for the development of IDDM in the multiple low-dose streptozotocin model.J Pathol. 1996; 178: 176-181Crossref PubMed Scopus (171) Google Scholar The rate of conversion to hyperglycemia following a low-dose STZ protocol was used to assess diabetes susceptibility. Whereas WT mice developed diabetes as early as day 4 post-STZ injection, GDNF-tg mice remained normoglycemic. At day 14 post-STZ injection, the number of hyperglycemic mice was significantly higher among WT mice than GDNF-tg mice (Figure 7A). The effect of STZ on β-cell mass was assessed by insulin staining. STZ treatment resulted in an over 2.8-fold (P < .001) reduction in insulin staining in WT mice than in GDNF-tg mice, which would suggest a resistance to STZ-induced destruction of β cells in GDNF-tg mice (Figure 7B and 7C). To determine the mechanisms involved in insulin loss following STZ administration, β-cell apoptosis was assessed by a combination of TUNEL staining19Srinivasan S. Stevens M. Wiley J.W. Diabetic peripheral neuropathy: evidence for apoptosis and associated mitochondrial dysfunction.Diabetes. 2000; 49: 1932-1938Crossref PubMed Scopus (235) Google Scholar with cleaved caspase-3 immunocytochemistry. GDNF-tg mice showed 4-fold less β-cell apoptosis than WT mice 12 days after injection with STZ as evidenced by the smaller number of TUNEL and cleaved caspase-3-positive β cells in these mice (P < .05; Figure 7D). GDNF-tg mice also had enhanced proliferation compared with WT mice (Figure 7E). Because WT mice received more STZ compared with GDNF-tg mice because of their higher body weights, the studies were repeated in weight-matched male WT and GDNF-tg mice. The effect of STZ on induction of hyperglycemia was independent of weight because we found a difference in STZ-induced hyperglycemia in weight-matched WT compared with GDNF-tg mice (blood glucose level [mg/dL]) post-STZ, WT: 265 ± 39, GDNF-tg: 179 ± 31, P < .05; weight (grams): WT: 33.93 ± 1.38, GDNF-tg: 31.79 ± 0.37, n = 3 in each group P = .21). The role of the enteric nervous system in influencing β-cell survival or mass has not been studied. In the current series of experiments, we document the presence of GDNF receptors on β cells and demonstrate that GDNF can influence β-cell mass by promoting survival and increasing proliferation. Similar to its effect on neurons, GDNF stimulates the PI-3-K pathway in β cells. We demonstrated GDNF effects on β cells in vivo using the GFAP-GDNF-tg mouse that overexpresses GDNF in glial cells. In addition to exhibiting increased rates of β-cell proliferation and β-cell mass, these mice demonstrated significantly lower fasting glucose levels and an enhanced response to an intraperitoneal glucose load. They demonstrated no difference in insulin resistance. Consistent with this finding, fasting plasma insulin levels were comparable in both mice. However, GDNF-tg mice demonstrated higher insulin release in response to intraperitoneal glucose administration compared with WT mice, which explains the improved glucose tolerance seen in these mice. The ratio of plasma insulin to blood glucose was higher in the GDNF-tg mice compared with WT mice. This is similar to what was observed in tg mice that overexpress placental lactogen in β cells that also have increased β-cell mass.20Vasavada R.C. Garcia-Ocana A. Zawalich W.S. et al.Targeted expression of placental lactogen in the β cells of transgenic mice results in β-cell proliferation, islet mass augmentation, and hypoglycemia.J Biol Chem. 2000; 275: 15399-15406Crossref PubMed Scopus (166) Google Scholar The capacity of GDNF to diminish chemically induced β-cell destruction was also demonstrated in vivo. Intraperitoneal STZ injection induced less β-cell loss in GDNF-tg mice than in WT mice. This β-cell sparing effect was reflected in the maintenance of normal blood glucose levels in GDNF-tg mice injected with STZ in contrast to STZ-injected WT mice. The results of this study demonstrate novel GDNF effects that may prove useful in sustaining β-cell mass or promoting survival. Tissue-specific overexpression of Akt in mice increases β-cell mass.3Bernal-Mizrachi E. Wen W. Stahlhut S. et al.Islet β-cell expression of constitutively active Akt1/PKB α induces striking hypertrophy, hyperplasia, and hyperinsulinemia.J Clin Invest. 2001; 108: 1631-1638Crossref PubMed Scopus (345) Google Scholar Our in vitro results demonstrate that GDNF can activate the PI-3-K/Akt pathway. This may be one of the mechanisms through which GDNF enhances β-cell mass. GDNF is expressed in pancreatic enteric glia. The close proximity of these ganglia to β cells suggests a paracrine effect of GDNF on β cells. Further studies are necessary to examine the mechanism by which GDNF overexpression in enteric glia activates its receptors on β cells and to evaluate the contribution of potential indirect effects. Our studies demonstrate a 3-fold increase in β-cell mass in GDNF-tg mice compared with WT mice. This degree of enhancement is similar to the 3-fold increase in β-cell mass observed in other growth factor overexpression studies, including parathyroid hormone-related protein and placental lactogen.20Vasavada R.C. Garcia-Ocana A. Zawalich W.S. et al.Targeted expression of placental lactogen in the β cells of transgenic mice results in β-cell proliferation, islet mass augmentation, and hypoglycemia.J Biol Chem. 2000; 275: 15399-15406Crossref PubMed Scopus (166) Google Scholar, 21Porter S.E. Sorenson R.L. Dann P. et al.Progressive pancreatic islet hyperplasia in the islet-targeted, parathyroid hormone-related protein-overexpressing mouse.Endocrinology. 1998; 139: 3743-3751Crossref PubMed Scopus (50) Google Scholar It is important to note that the effects of GDNF on β cells is not a systemic effect because serum GDNF levels have been measured in GDNF-tg mice and are no different from littermate controls.10Zhao Z. Alam S. Oppenheim R.W. et al.Overexpression of glial cell line-derived neurotrophic factor in the CNS rescues motoneurons from programmed cell death and promotes their long-term survival following axotomy.Exp Neurol. 2004; 190: 356-372Crossref PubMed Scopus (94) Google Scholar Islet β-cell mass is influenced by changes in proliferation, apoptosis, and neogenesis.1Bonner-Weir S. Life and death of the pancreatic beta cells.Trends Endocrinol Metab. 2000; 11: 375-378Abstract Full Text Full Text PDF PubMed Scopus (247) Google Scholar In our study, we found that GDNF can reduce thapsigargin-induced apoptosis in β cells in vitro as well as STZ-induced apoptosis in vivo. GDNF induces phosphorylation of Akt and its downstream target GSK3β. Other downstream targets of Akt may be involved in this survival pathway and will be examined in future studies. GDNF has been shown to increase neuropeptide-Y expressing neurons. One of the possible mechanisms by which GDNF could influence β-cell mass could be through neuropeptide-Y.13Anitha M. Chandrasekharan B. Salgado J.R. et al.Glial-derived neurotrophic factor modulates enteric neuronal survival and proliferation through neuropeptide y.Gastroenterology. 2006; 131: 1164-1178Abstract Full Text Full Text PDF PubMed Scopus (44) Google Scholar Future studies will examine this possibility. A recent study that traced the genetic lineage of newly forming adult β cells in mice concluded that preexisting β cells, and not adult stem cells, are the major source of β-cell neogenesis in adults following pancreatectomy.22Dor Y. Brown J. Martinez O.I. et al.Adult pancreatic β cells are formed by self-duplication rather than stem-cell differentiation.Nature. 2004; 429: 41-46Crossref PubMed Scopus (1914) Google Scholar Several other studies, however, report the presence of a pool, in both rodents and humans, of endocrine and nonendocrine pancreatic precursors that differentiate into cells of the β-cell lineage.6Seaberg R.M. Smukler S.R. Kieffer T.J. et al.Clonal identification of multipotent precursors from adult mouse pancreas that generate neural and pancreatic lineages.Nat Biotechnol. 2004; 22: 1115-1124Crossref PubMed Scopus (475) Google Scholar, 23Hao E. Tyrberg B. Itkin-Ansari P. et al.Beta-cell differentiation from nonendocrine epithelial cells of the adult human pancreas.Nat Med. 2006; 12: 310-316Crossref PubMed Scopus (206) Google Scholar Future studies will examine the possibility that GDNF induces β-cell neogenesis and the involvement of notch signaling.24Darville M.I. Eizirik D.L. Notch signaling: a mediator of β-cell de-differentiation in diabetes?.Biochem Biophys Res Commun. 2006; 339: 1063-1068Crossref PubMed Scopus (25) Google Scholar The protection from STZ-induced β-cell loss in the transgenic model suggests a potential mechanism for the prevention of diabetes. We found that GDNF-tg mice have lower rates of STZ-induced β-cell destruction. Aside from a reduced apoptotic rate, other potential reasons for the increased β-cell mass in the STZ-treated transgenic mice include a greater initial β-cell mass or an enhanced ability to rapidly replenish β-cells by neogenesis or replication. Crossbreeding of GDNF-tg mice with genetic models of β-cell failure may help elucidate potential mechanisms underlying the resistance to the development of diabetes. In summary, the current study demonstrates a novel mechanism of a neurotrophic factor expressed in the enteric nervous system that clearly influences β-cell mass. The enteric nervous system can affect both functional and structural effects of β cells by increasing cellular proliferation, reducing apoptosis, and improving glycemic control in vivo. The current study provides new therapeutic targets for improving β-cell mass. The authors thank Dr. Mark Rigby, Emory Transplantation Center, for help with the insulin staining; Dr. Bindu Chandrasekharan and Irene Joseph for their help with the experiments; and Dr. Mauricio Rocca, Edillson Torres, and Jianguo Xu, Division of Pulmonary Diseases, for help with histologic tissue processing.