Rapid Publications Expression and Functional Activity of Glucagon, Glucagon-Like Peptide I, and Glucose-Dependent Insulinotropic Peptide Receptors in Rat Pancreatic Islet Cells Karen Moens, Harry Heimberg, Daisy Flamez, Peter Huypens, Erik Quartier, Zhidong Ling, Daniel Pipeleers, Sandrine Gremlich, Bernard Thorens, and Frans Schuit

Rat pancreatic a- and P-cells are critically dependent on hormonal signals generating cyclic AMP (cAMP) as a synergistic messenger for nutrient-induced hormone re- he intracellular messenger cAMP is known to lease. Several peptides of the glucagon-secretin family affect glucose-induced insulin secretion at two have been proposed as physiological ligands for cAMP levels: 1) a minimal level of the nucleotide is production in P-cells, but their relative importance for required to exert a permissive effect on insulin islet function is still unknown. The present study shows Trelease (1); 2~) cAMP levels above this basal threshold can expression at the RNA level in p-cells of receptors for glucagon, glucose-dependent insulinotropic polypeptide further modulate the amplitude of the secretory response by (GIP), and glucagon-like peptide 1(7-36) amide (GLP-I), changing the sensitivity of the exocytotic machinery to while RNA from islet a-cells hybridized only with GIP cytosolic calcium (2). Because of its importance as signal receptor cDNA. Western blots confirmed that GLP-I re- mediator for P-cell function, it is a major goal to identify the ceptors were expressed in P-cells and not in a-cells. physiological ligands that regulate basal and elevated cAMP Receptor activity, measured as cellular cAMP production levels in p-cells. In isolated rat islets, the paracrine effect of after exposing islet P-cells for 15 min to a range of glucagon was identified as a permissive factor (1,3). Insulin peptide concentrations, was already detected using 10 release can be amplified both in vivo and in vitro by the pmol/1 GLP-I and 50 pmol/1 GIP but required 1 nmol/1 incretin hormones glucagon-like peptide (GLP)-I and glu- glucagon. EC50 values of GLP-I- and GIP-induced cAMP cose-dependent insulinotropic peptide (GIP) (4,5) and by formation were comparable (0.2 nmol/1) and 45-fold glucagon (6). It is still uncertain, however, whether glucagon lower than the ECg0 of glucagon (9 nmol/1). Maximal plays the same permissive role in vivo. First, directionality of stimulation of cAMP production was comparable for the islet flow is such that p-cells in the islet center are perfused three peptides. In purified a-cells, 1 nmol/1 GLP-I failed to by afferent arterioles and not via capillaries, which irrigate increase cAMP levels, while 10 pmol/1 to 10 nmol/1 GIP blood that has accumulated glucagon from a-cells in the islet exerted similar stimulatory effects as in P-cells. In con- clusion, these data show that stimulation of glucagon, periphery (7). This static picture of islet blood flow indicates GLP-I, and GIP receptors in rat p-cells causes cAMP that a paracrine a- to p-cell axis may only exist for p-cells in the islet periphery. Second, high-affinity binding sites for production required for insulin release, while adenylate 125 cyclase in a-cells is positively regulated by GIP. Diabetes I-labeled glucagon were demonstrated in pure 3-cells (8). 45:257-261, 1996 Kofod et al. (9) described, however, competition between glucagon and GLP-I for the same high-affinity binding sites. The idea that glucagon exerts its effects on insulin release via binding to GLP-I receptors (9) has been challenged (6) by demonstrating that GLP-I-induced insulin release was dis- turbed in pancreases perfused with exendin(9-39)amide, while glucagon-induced insulin release remained undis- turbed. Also, the presence of glucagon receptor mRNA in rat From the Diabetes Research Center, Vrrje Universiteit Brussel, Brussels, Belgium; pancreatic islets has been recently documented (10). The and the Institut de Pharmacologie, Universite de Lausanne, Lausanne, Switzerland. Address correspondence and reprint requests to Dr. Frans Schuit, Department cloning of cDNAs encoding receptors for glucagon (11), of Biochemistry, Diabetes Research Center, Faculty of Medicine, Vrye Universiteit GLP-I (12), and GIP (13; S.G., B.T., unpublished observa- Brussel, Laarbeeklaan 103, B-1090 Brussels, Belgium. E-mail: [email protected]. tions) allowed us to assess the extent to which these ac. Received for publication 27 October 1995 and accepted in revised form 7 receptors are expressed in flow-sorted rat a- and P-cells. November 1995. Second, since p-cells with different responsiveness to GLP-I BSA, bovine serum albumin; GIP, glucose-dependent insulinotropic peptide; GLP-I, glucagon-like peptide 1(7-36) amide; IBMX, 3-isobutyl-l-methylxanthine; and glucose were described (14), GLP-I receptor expression RT-PCR, reverse transcription-polymerase chain reaction. was tested in p-cell subsets sorted on the basis of differential

DIABETES, VOL. 45, FEBRUARY 1996 257 RECEPTOR EXPRESSION IN RAT ISLET CELLS responsiveness to glucose (15). Our data indicate that in O O (3-cells, the three G-protein-coupled receptors are expressed OX) both at the mRNA level and at the level of receptor activity, s suggesting that both local release of glucagon and distal a 3 release of incretins contribute to the complex regulation of cellular cAMP levels. GLP-1- RESEARCH DESIGN AND METHODS receptor II Chemicals. Synthetic glucagon, GLP-I, GIP, and somatostatin were all purchased from Sigma (St. Louis, MO), and 3-isobutyl-l-methylxanthine (IBMX) was from Aldrich (Janssen Chimica, Beerse, Belgium). All other chemicals were analytical grade and purchased at Merck (Darmstadt, Germany) or Sigma. GIP- Purification and culture of islet cells. Islet cells were prepared from male adult Wistar rats using previously published methods (16). Islet receptor non-(3-cells (<10% p-cells), pure a-cells (>80% a-cells, <5% (B-cells), and pure 0-cells (>90% purity) were cultured for 16 h in Ham's F10 (Gibco BRL, Gaithersburg, MD), supplemented with 2 mmol/1 L-glu- tamine, 1% bovine serum albumin (BSA; fraction V, radioimmunoassay grade), 0.075 mg/ml penicillin (Sigma), 0.1 mg/ml streptomycin (Sigma), glucagon and 10 mmol/1 glucose (P-cells) or 6 mmol/1 glucose (a-cells). Before -receptor culturing, the cells were briefly reaggregated using basal medium t (a-cells) or medium supplemented with 50 (xmol/1 IBMX and 2 mmol/1 Ca2+ ((3-cells). In some experiments, the total (3-cell preparation was further separated via flow sorting into cells with basal NADPH fluores- cence (low cells) and cells with elevated NADPH autofluorescence (high cells) after 15-min incubation at 7.5 mmol/1 glucose (15). B-actin Analysis of receptor mRNA. Total RNA was isolated via guanidinium t- isothyocyanate-cesium chloride and separated according to size on 1.5% agarose gels (10 |xg per lane). Blots were UV-cross-linked and FIG. 1. Glucagon, GLP-I, and GIP receptor mRNA abundance in rat 32 pancreatic islets and islet cells. Total RNA (10 jig) from rat control hybridized with P-labeled cDNA probes recognizing mRNA encoding tissues (lung, stomach, intestine, liver, and brain) as well as from receptors for glucagon (11), GLP-I (12), and GIP (13; S.G., B.T., insulinoma, glucagonoma, islets, purified non-P-cells (>80% a-cells), and unpublished observations), requiring autoradiographic exposure times purified p-cells was blotted and hybridized with receptor-specific cDNA of 14, 8, and 7 days, respectively. The finalstringen t washing step was 30 probes. P-actin transcript was taken as an internal control. Exposure min at 60°C in 2.5% SDS and 0.2x (glucagon receptor) or 0.5X sodium times to obtain autoradiographic signals were 14 days (glucagon chloride-sodium citrate (GLP-I receptor and GIP receptor). receptor), 8 days (GLP-I receptor), 7 days (GIP receptor), and 6 h (P-actin). The blots shown here are representative of three different For reverse transcription-polymerase chain reaction (RT-PCR), total experiments. RNA (0.4 |xg) from pancreatic (3- and non-P-cells and control tissues (insulinoma, brain, liver, lung, stomach, and intestine) was reverse transcribed and amplified according to an adapted protocol of the RESULTS GeneAmp RNA-PCR kit (Perkin-Elmer/Cetus, Emeryville, CA). Specific mRNA analysis. The GLP-I receptor cDNA probe hybridized primers were designed to anneal with codon 126-132/201-206 for the with at least two abundant transcripts present in total RNA GLP-I receptor, codon 80-85/257-263 for the GIP receptor, codon from lung, islets, and (3-cells (Fig. 1). Signal ratio in non-(3- 85-91/167-173 for the glucagon receptor, and codon 249-255/338-344 for (3-actin, yielding PCR products of 243, 550, 265, and 288 bp, cells was one order of magnitude lower than in 3-cells. GLP-I respectively. The cycling profile for GLP-I, GIP, and (J-actin was 2.5 min receptor mRNA signal was also abundant in total RNA at 95°C, followed by 1 min at 94°C, 1.5 min at 55°C, and 1.5 min at 72°C extracted from MSL-G2-IN, a transplantable insulinoma (19), for the first 10 cycles and 0.5 min at 94°C, 1 min at 50°C, and 1.5 min at but not from MSL-GAN (19), a glucagonoma. On the con- 72°C for 25 cycles. The cycling profile for the glucagon receptor was the trary, GIP receptor mRNA was at least 10-fold more abun- same except for the annealing temperature, which was 68°C. PCR products were controlled for their length on a 2% agarose gel and dant in glucagonoma than in insulinoma. Compatible with sequenced using dideoxy nucleotides and 35S-labeling. this observation, the GIP receptor appears to be present in Protein analysis. Western blotting for the analysis of GLP-I receptor islets and in purified (3- and non-(3-cells. Total RNA from expression was performed using a standard protocol (17). An affinity- liver, islets, and (3-cells hybridized with the glucagon recep- purified polyclonal antibody prepared against a fusion protein consisting of the glutathione S-transferase and the C-tail of the rat GLP-I receptor tor cDNA probe, while the other control tissues remained and prepared as described (18) was used at a final concentration of 5 negative after 14 days of exposure (Fig. 1). Transcript length |xg/ml. Bound primary antibody was detected with a goat anti-rabbit was comparable in liver and (3-cells. The signal ratio of immunoglobulin antibody coupled to horseradish peroxidase and the glucagon receptor mRNA over (3-actin mRNA was again at enhanced chemiluminescence detection system of Amersham. least one order of magnitude lower in islet non-(3-cells (80% Analysis of cAMP production in a- and p-cells. Receptor-induced cAMP accumulation in intact cells was assessed as described previously a-cells) than in purified (3-cells, which is compatible with the (3). Islet cells were incubated for 15 min at 37°C in Earle's HEPES contamination of <10% (3-cells in this preparation (16). In medium supplemented with 0.5% BSA, 250 |xmol/l IBMX, and the peptide addition to Northern blot analysis, expression of glucagon, hormones as indicated. In addition, incubation medium for p-cells GLP-I, and GIP receptor mRNA in islets and (3- and non-(3- contained 20 mmol/1 glucose, while that for a-cells was supplemented cells was assessed via RT-PCR analysis and sequencing of with 2.8 mmol/1 glucose; 2 mmol/1 each of alanine, arginine, and glutamine; and 1 mmol/1 ascorbic acid. cAMP content of acetylated the amplified cDNA fragments. Using these techniques, PCR samples was measured using a commercially available 125I cAMP fragments encoding the GLP-I, GIP, and glucagon receptors radioimmunoassay kit (Amersham, Little Chalfont, U.K.). When [3H]- of the right lengths and nucleotide sequences were reproduc- cAMP was added to the samples before extraction, >90% of the label ibly obtained from p-cells (data not shown). Furthermore, was recovered at the end of this procedure. Statistical analysis. Statistical significance was assessed using the the abundance of the GIP and GLP-I receptor fragments was unpaired two-tailed Student's t test. similar in (3-cell subsets with high and low responsiveness to

258 DIABETES, VOL. 45, FEBRUARY 1996 K. MOENS AND ASSOCIATES B-cells a-cells islets 3xlO5 lxlO5 3xlO5 100 kDa — 94 — 67

— 43 FIG. 2. GLP-1 receptor protein in rat islets and purified a- and p-cells. Total protein homogenates from purified P-cells, purified a-cells, and islets were examined for the presence of immunoreactive GLP-I receptor. Numbers of islet cells or islets loaded on the gel are indicated. Arrows show fully glycosylated (~67-kDa) and immature — 30 (~43-kDa) forms of the receptor protein. glucose. Amplification of the same cDNA fragments obtained However, when the peptide was used at concentrations >10 in P-cells was also positive from most conditions using islet nmol/1, a marked decline in cAMP content of a-cells was non-p-cells that contained up to 10% p-cells. observed (Fig. 3A). Glucagon, GLP-I, and GIP increased GLP-I receptor protein. The functional integrity of the cAMP levels in purified p-cells in a dose-dependent manner GLP-I receptor mRNA was examined via Western blots of (Fig. 3B). Minimal concentrations required for significant total protein from islets and purified a- and p-cells. The effects were 1 nmol/1 (glucagon), 10 pmol/1 (GLP-I), and 50 polyclonal antibody recognized in P-cell extracts both the pmol/1 (GIP). The higher sensitivity of P-cells to the incretin mature glycosylated protein (molecular weight —67 kDa) hormones versus glucagon is also reflected in the respective and an immature ~43-kDa GLP-I receptor precursor (arrows EC50 values that were observed (0.2 nmol/1 for GLP-I and GIP in Fig. 2). Bands migrating with higher and lower mobility versus 9 nmol/1 for glucagon). Maximal effects at the level of represent nonspecific cross-reacting proteins (18). The weak cAMP accumulation were comparable for the three peptides. immunoreactivity detected in islet non-p-cells (data not Such effects were observed with 10 nmol/1 GIP, 1 (xmol/1 shown) could be ascribed to p-cell contamination, since a GLP-I, and 1 |xmol/l glucagon. At 1 |xmol/l GIP, cAMP content parallel experiment with pure a-cells gave no detectable decreased nonsignificantly as compared with that in cells signal (Fig. 2). Immunoblotting of protein from low and high exposed to 10 nmol/1 of the peptide. P-cells, which differ in their acute responsiveness to 7.5 mmol/1 glucose (15), exhibited the same GLP-I receptor protein abundance (data not shown). DISCUSSION Hormone-regulated cAMP content in purified a- and The present study contributes to our understanding of the P-cells. Exposure of purified islet a-cells to 1 nmol/1 GLP-I complex physiological regulation of cAMP levels in the had no effect on cAMP content (basal 3 ± 1 vs. GLP-I 2 ± 1 endocrine pancreas, a process that is of major importance fmol/103 cells; means ± SD; n = 4), while glucagon could not for the fuel sensor function of islet a- and p-cells. Our data be tested in these cells. Incubating the cells with 1 nmol/1 GIP demonstrate directly that G-protein-coupled receptor mRNA is induced a fourfold rise in cAMP (13 ± 3 fmol/103 cells; n = present in flow-sorted rat islet cells. They indicate that cAMP 4; P < 0.01 vs. basal). This effect could be prevented for 80% levels in the insulin-producing p-cell population are regu- by simultaneous exposure to 1 nmol/1 somatostatin-14 (5.3 ± lated by at least three receptors from the secretin-glucagon 0.3 fmol/103 cells; n = 4), a potent inhibitor of adenylate family, since mRNA encoding receptors for glucagon, GLP-I, cyclase in these cells (20). GIP-stimulated cAMP production and GIP was detected in purified p-cells. This search for was dose dependent between 10 pmol/1 and 10 nmol/1. members of the secretin-glucagon receptor family was not

30- - 100 " A B • basal • basal • GIP j [ D GLP-1 1 nff*4

/ \ A GIP jJ^I O glucagon iC—kl 8 I15 50 o

FIG. 3. Regulation of cAMP content in purified a- (A ) and P-cells (B ). Data represent means ± SE of five i ij i i i i i i • // i . i i (a-cells) or three (p-cells) independent experiments. 10 6 oo" 12 10 Cellular incubations were for 15 min at 37°C in the presence of 250 |jimol/l IBMX. \J ^-log[Peptide] (M)

DIABETES, VOL. 45, FEBRUARY 1996 259 RECEPTOR EXPRESSION IN RAT ISLET CELLS exhaustive. Recent in vivo experiments with transgenic effects of GLP-I-stimulated insulin release. Alternatively, vasoactive intestinal peptide (VIP) mice (21) and in vitro since GLP-I receptors are present on a somatostatin-secret- effects using pituitary adenylate cyclase-activating polypep- ing cell line (30), paracrine effects including somatostatin- tide (22) indicate that additional receptor types modulate inhibited glucagon release (20) can be involved. Fourth, the insulin release via the stimulation of (3-cell adenylate cyclase. reported in vivo stimulation of glucagon release by GIP It is not known what part of this plethora is essential for observed in the perfused pancreas in the presence of low (3-cell physiology. One question to be answered is whether all glucose concentrations (31,32) seems attributable to direct different receptor types are expressed on the same p-cells or interactions of this peptide with pancreatic a-cells. In the on (3-cell subsets. Since rat (3-cells behave functionally as a picomolar to low nanomolar concentration range, the pep- heterogenous population of cells (15,23), it is conceivable tide markedly increased cAMP content in a-cells to levels that (3-cell subsets differ in their cell-surface G-protein- that are required for nutrient-induced glucagon release (33). coupled receptor expression. Such heterogeneity may be a The physiological relevance of this stimulation is not clear. result of the anatomic organization of the islets of Langer- The peptide may act as a paracrine amplifier, since the hans, juxtaposing centrally located islet (3-cells without released glucagon may act in concert with GIP itself on the paracrine exposure to glucagon and somatostatin and (3-cells P-cells. Alternatively, GIP could act on a-cells in concert in the islet mantle. Likewise, it is possible that individual 3-cells with (nor)adrenergic stimulation (34) during episodes of differ in their capacity to detect incretin hormones, explain- stress. Remarkably, the effects of GIP on cAMP production ing the heterogenous response of isolated cells to GLP-I (14). were severely and reproducibly blunted at supraphysiologi- In line with this idea, we have compared GLP-I receptor cal concentrations, for which we have no explanation. Rapid mRNA and protein abundance in flow-sorted (3-cell subsets homologous desensitization of the receptor at a high ligand (15) and have observed no major differences in GLP-I recep- concentration is one possible mechanism, but this does not tor expression levels in two cell preparations. We have not explain why the process was much more important in a-cells excluded that postreceptor defects exist in some (3-cells. It is than in p-cells. also possible that sensitivity or resistance to GLP-I can be In conclusion, these data show that stimulation of gluca- related to other factors, since glucose competence depends gon, GLP-I, and GIP receptors in rat (3-cells causes cAMP not only on variables related to cAMP-generating signal production required for insulin release, while adenylate transduction but also on the expression of enzymes critical cyclase in a-cells is positively regulated by GIP. It remains to to glucose metabolism (24). It is, for instance, possible that be investigated to what extent molecular defects in this flow sorting of low and high P-cells (15) does not distinguish complex regulation can cause cAMP-deficient (3-cells with between p-cell populations with variable GLP-I receptor defective insulin secretion to glucose, such as are frequently numbers per cell, in the case that these existed. observed in subjects with NIDDM. The demonstration of direct effects of glucagon, GLP-I, and GIP on rat (3-cells may alter the interpretation of some ACKNOWLEDGMENTS previous experiments. First, experiments with the GLP-I This study was supported by grants from the Belgian Pro- receptor antagonist exendin(9-39)amide, which blocks at gramme on Interuniversity Poles of Attraction initiated by least 60% of the incretin response to glucose in rats (25), the Belgian State (IUAP 15), the Nationaal Fonds voor suggested that GLP-I mediates most of the incretin effect of Wetenschappelijk Onderzoek (NFWO; grant 3.3132.91 to glucose. In parallel, it was proposed that the incretin effect of D.P., Krediet aan Navorsers to F.S., and Aspirant Fellowship GIP occurs via stimulation of intestinal GLP-I release, which to P.H.), and the Flemish Community (Concerted Action acts on (B-cells (26). Our data demonstrating GIP receptor GOA 92/97-1807). B.T. is the recipient of a Career Develop- expression on fJ-cells and functional responsiveness to GIP ment Award from the Swiss National Science Foundation levels as low as 50 pmol/1 indicate that direct effects of GIP and is supported by Swiss National Science Foundation on p-cells can also explain the insulinotropic properties of Grant 31-30313.90. this peptide. In fact, the present results support data re- We wish to thank the personnel of the Department of ported after immunoneutralization of GIP in the rat (27) and Metabolism and Endocrinology for rat islet cell preparation the observed additive effect of GLP-I and GIP on glucose- and purification and Dr. O. Madsen for the insulinoma and induced insulin release (28). Second, the idea that glucagon glucagonoma. The rat glucagon receptor cDNA was a gift stimulates glucose-induced insulin release via its binding to from Dr. W. Kindsvogel, Zymogenetics, Seattle, WA. GLP-I receptors on p-cells (9) is not compatible with our present demonstration of glucagon receptor mRNA in pure p-cells. Our data complement the earlier demonstration of REFERENCES 1. Pipeleers DG, Schuit FC, in't Veld PA, Maes E, Hooghe-Peters EL, Van De high-affinity glucagon binding sites present on p-cells (8) and Winkel M, Gepts W: Interplay of nutrients and hormones in the regulation the recent description of glucagon receptor mRNA in iso- of insulin release. Endocrinology 117:824-833, 1985 lated rat islets (10). As such, they may explain the fact that 2. Prentki M, Matschinsky FM: Ca2+, cAMP, and phospholipid-derived messengers in coupling mechanisms of insulin secretion. Physiol Rev exendin(9-39)amide blocks GLP-I-induced rather than glu- 67:1185-1248, 1987 cagon-induced insulin secretion from the perfused rat pan- 3. Schuit FC, Pipeleers DG: Regulation of adenosine 3', 5'-monophosphate levels in the pancreatic p cell. Endocrinology 117:834-840, 1985 creas (6). Third, the mechanism via which GLP-I can inhibit 4. Weir GC, Mojsov S, Hendrick GK, Habener JF: Glucagon-like peptide I glucagon release in vivo (29) seems to involve indirect effects (7-37) actions on endocrine pancreas. Diabetes 38:338-342, 1989 on the a-cells, since GLP-I receptor protein could not be 5. Creutzfeldt W: The incretin concept today. Diabetologia 16:15-85, 1979 6. Kawai K, Yokota C, Ohashi S, Watanabe Y, Yamashita K: Evidence that demonstrated in purified a-cells and 1 nmol/1 GLP-I did not glucagon stimulates insulin secretion through its own receptor in rats. affect cAMP levels in these cells. It is conceivable that the Diabetologia 38:274-276, 1995 GLP-I inhibition of glucagon release, an effect that is of 7. Bonner-Weir S, Orci L: New perspective on the microvasculature of the islets of Langerhans in the rat. Diabetes 31:883-889, 1982 therapeutical value in NIDDM, is mediated via the paracrine 8. Van Schravendyk CFH, Foriers A, Hooghe-Peters EL, Rogiers V, De

260 DIABETES, VOL. 45, FEBRUARY 1996 K. MOENS AND ASSOCIATES

Meyts P, Sodoyez JC, Pipeleers DG: Pancreatic hormone receptors on S, Yoshimoto T, Okamoto H: Transgenic mice overexpressing human islet cells. Endocrinology 117:841-848, 1985 vasoactive intestinal peptide (VIP) gene in pancreatic p cells. J Biol 9. Kofod H, Adelhorst K, Christensen LB, Fuhlendorff J, Godtfredsen SE, Chem 269:21223-21228, 1994 Kirk 0, Lundgren K: Beta-cells do not have glucagon receptors but GLP-1 22. Yada T, Sakurada M, Ihida K, Nakata M, Murata F, Arimura A, Kikuchi M: receptors! Digestion 54:346-347, 1993 Pituitary adenylate cyclase activating polypeptide is an extraordinarily 10. Abrahamsen N, Nishimura E: Regulation of glucagon and glucagon-like potent intra-pancreatic regulator of insulin secretion from islet fl-cells. J peptide-1 receptor messenger ribonucleic acid expression in cultured rat Biol Chem 269:1290-1293, 1994 pancreatic islets by glucose, cyclic adenosine 3',5'-monophosphate, and 23. Schuit FC, in't Veld PA, Pipeleers DG: Glucose stimulates insulin biosyn- glucocorticoids. Endocrinology 136:1572-1578, 1995 thesis by a dose-dependent recruitment of pancreatic beta cells. Proc 11. Jelinek LJ, Lok S, Rosenberg GB, Smith RA, Grant FJ, Biggs S, Bensch Natl Acad Sci USA 85:3865-3869, 1988 PA, Kuyper JL, Sheppard PO, Sprecher CA, O'Hara PJ, Foster D, Walker 24. Heimberg H, De Vos A, Vandercammen A, Van Schaftingen E, Pipeleers KM, Chen LHJ, McKernan PA, Kindsvogel W: Expression cloning and D, Schuit F: Heterogeneity in glucose sensitivity among pancreatic signaling properties of the rat glucagon receptor. Science 259:1614-1616, B-cells is correlated to differences in glucose phosphorylation rather than 1993 glucose transport. EMBO J 12:2873-2879, 1993 12. Thorens B: Expression cloning of the pancreatic B-cell receptor for the 25. Kolligs F, Fehmann H-C, Goke R, Goke B: Reduction of the incretin effect gluco-incretin hormone glucagon-like peptide 1. Proc NatlAcad Sci USA in rats by the glucagon-like peptide 1 receptor antagonist exendin(9- 89:8641-8645, 1992 39)amide. Diabetes 44:16-19, 1995 13. Usdin TB, Mezey E, Button DC, Brownstein MJ, Bonner TI: Gastric 26. Roberge JN, Brubaker PL: Regulation of intestinal proglucagon-derived inhibitory polypeptide receptor, a member of the secretin-vasoactive peptide secretion by glucose-dependent insulinotropic peptide in a novel intestinal peptide receptor family, is widely distributed in peripheral enteroendocrine loop. Endocrinology 133:233-240, 1993 organs and the brain. Endocrinology 133:2861-2870, 1993 27. Ebert R, Unger H, Creutzfeldt W: Preservation of incretin activity after 14. Holz IV GG, Kiihtreiber WM, Habener JF: Pancreatic beta-cells are removal of gastric inhibitory polypeptide (GIP) from rat gut extracts by rendered glucose-competent by insulinotropic hormone glucagon-like immunoadsorption. Diabetologia 24:449-454, 1993 peptide-l(7-37). Nature 361:362-365, 1993 28. Fehmann HC, Goke B, Goke R, Trautmann ME, Arnold R: Synergistic 15. Kiekens R, in't Veld P, Mahler T, Schuit F, Van de Winkel M, Pipeleers D: stimulatory effect of glucagon-like peptide-l(7-36) amide and glucose- Differences in glucose recognition by individual rat pancreatic B cells are dependent insulin-releasing polypeptide on the endocrine rat pancreas. associated with intercellular differences in glucose-induced biosynthetic FEBS Lett 252:109-112, 1989 activity. J Clin Invest 89:117-125, 1992 29. Gutniak M, Orskov C, Hoist JJ, Ahr£n B, Efendic S: Antidiabetogenic 16. Pipeleers DG, in't Veld P, Van De Winkel M, Maes E, Schuit FC, Gepts W: effects of glucagon-like peptide-1 (7-36)amide in normal subjects and A new in vitro model for the study of pancreatic A and B cells. patients with diabetes mellitus. N EnglJ Med 326:1316-1322, 1992 Endocrinology 117:806-816, 1985 30. Fehmann HC, Habener JF: Functional receptors for the insulinotropic 17. Thorens B, Sarkar HK, Lodish HR: Cloning and functional expression in hormone glucagon-like peptide-I(7-37) on a somatostatin secreting cell bacteria of a novel glucose transporter present in liver, intestine, kidney, line. FEBS Lett 279:335-340, 1991 and (3-pancreatic islet cells. Cell 55:281-290, 1988 31. Pederson RA, Brown JC: Interaction of gastric inhibitory polypeptide, 18. Widmann C, Dolci W, Thorens B: Agonist-induced internalization and glucose, and arginine on insulin and glucagon secretion from the per- recycling of the glucagon-like peptide-1 receptor in transfected fibro- fused rat pancreas. Endocrinology 103: 610-615, 1978 blasts and in insulinomas. Biochem J 310:203-214, 1995 32. Suzuki S, Kawai K, Ohashi S, Mukai H, Murayama Y, Yamashita K: 19. Madsen OD, Karlsen C, Nielsen E, Lund K, Kofod H, Welinder B, Rehfeld Reduced insulinotropic effects of glucagon-like peptide-l-(7-36)amide JF, Larsson LI, Steiner DF, Hoist JJ, Michelsen BK: The dissociation of and gastric inhibitory polypeptide in isolated perfused diabetic rat tumor-induced weight loss from hypoglycemia in a transplantable pluri- pancreas. Diabetes 38:741-744, 1990 potent rat islet tumor results in the segregation of stable a- and (3-cell 33. Pipeleers DG, Schuit FC, Van Schravendyk CFH, Van De Winkel M: tumor phenotypes. Endocrinology 133:2022-2030, 1993 Interplay of nutrients and hormones in the regulation of glucagon release. 20. Schuit FC, Derde MP, Pipeleers DG: Sensitivity of pancreatic A and B Endocrinology 117:817-823, 1985 cells to somatostatin. Diabetologia 32:207-212, 1989 34. Schuit FC, Pipeleers DG: Differences in adrenergic recognition by pan- 21. Kato I, Suzuki Y, Akabane A, Yonekura H, Tanaka O, Kondo H, Takasawa creatic A and B cells. Science 232:875-877, 1986

DIABETES, VOL. 45, FEBRUARY 1996 261