ORIGINAL ARTICLE Endocrine Research Brief Report Effect of Glucagon-Like Peptide-1 on - and -Cell Function in C-Peptide-Negative Type 1 Diabetic Patients Urd Kielgast, Meena Asmar, Sten Madsbad, and Jens J. Holst Department of Biomedical Sciences (U.K., M.A., J.J.H.), the Panum Institute, University of Copenhagen, 2200 Copenhagen, Denmark; and Department of Endocrinology (U.K., M.A., S.M.), Hvidovre Hospital and University of Copenhagen, 250 Hvidovre, Denmark Context: The mechanism by which glucagon-like peptide-1 (GLP-1) suppresses glucagon secretion is uncertain, and it is not determined whether endogenous insulin is a necessary factor for this effect. Objective: To characterize the - and -cell responses to GLP-1 in type 1 diabetic patients without residual -cell function. Methods: Nine type 1 diabetic patients, classified as C-peptide negative by a glucagon test, were clamped at plasma glucose of 20 mmol/liter for 90 min with arginine infusion at time 5 min and concomitant infusion of GLP-1 (1.2 pmol/kg min) or saline. Results: Infusion with GLP-1 increased C-peptide concentration just above the detection limit of 33 pmol/liter in one patient, but C-peptide remained immeasurable in all other patients. In the eight remaining patients, total area under the curve of glucagon was significantly decreased with GLP-1 compared with saline: 5 72 vs. 70 97 pmol/liter min (P 0.001). In addition, GLP-1 decreased the arginine-stimulated glucagon release (incremental AUC of 3 21 and 137 pmol/liter min, with GLP-1 and saline, respectively, P 0.05). Conclusions: In type 1 diabetic patients without endogenous insulin secretion, GLP-1 decreases the glucagon secretion as well as the arginine-induced glucagon response during hyperglycemia. GLP-1 induced endogenous insulin secretion in one of nine type 1 diabetic patients previously classified as being without endogenous insulin secretion. (J Clin Endocrinol Metab 95: 292 29, 20) Glucagon-like peptide-1 (GLP-1) stimulates insulin biosynthesis and insulin secretion (1), whereas glucagon secretion is inhibited (2). GLP-1 increases insulin release and biosynthesis through direct interaction with GLP-1 receptors present at the cell membrane of -cells (2), but the mechanism of GLP-1-induced decrement of glucagon release is not fully understood. GLP-1 may exert its glucagonostatic effect via direct interaction with GLP-1 receptors on -cells, but this is controversial because some studies have provided evidence of the presence of GLP-1 receptors on a subset of -cells (3, ), whereas others have ISSN Print 0021-972X ISSN Online 195-7197 Printed in U.S.A. Copyright 20 by The Endocrine Society doi:./jc.2009-20 Received November, 2009. Accepted February 17, 20. First Published Online March 5, 20 not (5 7). However, GLP-1 receptor activation has been reported to stimulate exocytosis in isolated rat -cells (3), making a direct mechanism of the in vivo glucagonostatic action of GLP-1 less likely. Insulin and somatostatin inhibit glucagon secretion (), and it is suggested that GLP-1 affects -cell secretion indirectly through paracrine or local mechanisms via changes of somatostatin (3, 9), insulin, or other -cell-derived mediators (, 11). It is well established that secretion of somatostatin is stimulated by GLP-1 () and that islet somatostatin inhibits glucagon release (13), and there is evidence that somatostatin me- Abbreviations: GAD, Glutamic acid decarboxylase; GIR, glucose infusion rates; GLP-1, glucagon-like peptide-1. 292 jcem.endojournals.org J Clin Endocrinol Metab, May 20, 95(5):292 29
J Clin Endocrinol Metab, May 20, 95(5):292 29 jcem.endojournals.org 293 diates the glucagonostatic action of GLP-1 (9). In a previous study, GLP-1 inhibited glucagon secretion in type 1 diabetic patients (), suggesting that endogenous insulin is not a necessary factor; however, this study was designed to study typical type 1 diabetic patients and therefore included eight (of 11) subjects with measurable basal C- peptide levels, and the group showed a slight but significant increase in C-peptide response to GLP-1. Interestingly, an autopsy study revealed residual -cells in % of patients with type 1 diabetes regardless of disease duration (15), consistent with ongoing -cell formation even in longstanding type 1 diabetes. Because of the possible existence of secretory silent or very low (and therefore clinically undetectable) numbers of -cells in longstanding type 1 diabetes, and because GLP-1 enhances the secretory performance of -cells (), we speculated that patients with type 1 diabetes, classified as being without residual -cell function, might be able to secrete insulin in response to a maximal -cell function test if GLP-1 was added. If insulin secretion remained absent, we would be able to elucidate whether the glucagonostatic effect of GLP-1 in humans is dependent of endogenous insulin secretion. Subjects and Methods Nine patients with type 1 diabetes without C-peptide response to a glucagon test underwent a -cell stimulation test using a hyperglycemic clamp technique including arginine stimulation with and without concomitant infusion of GLP-1. Participants agreed to participate after oral and written information, and the study was conducted according to the Helsinki 2 Declaration. Patients were diagnosed between age 5 and 37, had lost weight up to diagnosis, and were treated with insulin immediately after diagnosis. The male to female ratio was :1, mean age was 33.9 2 yr, glycosylated hemoglobin was 7.2 0.3%, BMI was 22.7 0.5 kg/m 2, and duration of diabetes was 13 1. yr. Two patients were positive for glutamic acid decarboxylase (GAD)-5 only, and one was positive for both GAD-5 and islet cell antibodies. Patients received their normal long-acting insulin the night before the study (seven were treated with insulin detemir, one with neutral protamine Hagedorn insulin and one with insulin glargine). No insulin was given in the morning of the study except in one patient who received her normal dose of, i.e. insulin detemir. Doses and injection sites were identical on the two study days. After an overnight fast, plasma glucose was raised to 20 mmol/liter with an iv bolus of glucose according to the following formula: (20 fasting plasma glucose) 35 weight (kilograms) grams glucose to be infused. Plasma glucose was maintained at 20 mmol/liter for 90 min with infusion of glucose, and after 5 min, a bolus of 5 g arginine was infused. Thirty minutes before the clamp, a randomized and singleblinded infusion of GLP-1 (1.2 pmol/kg min) or saline was started. Blood samples for analyses of plasma glucose, insulin, amylin, and C-peptide were taken as indicated in Fig. 1A and total GLP-1 and glucagon as indicated in Fig. 2. A glucose (mmo ol/l) B Glucose infu usion rate (mg g/kg/min) 2 22 20 1 time (minutes) -50-0 -30-20 - 0 20 30 0 50 0 70 0 90 1 2 0 GLP-1/NaCl hyperglycemic clamp 0 20 30 0 50 0 70 0 90 time(minutes) FIG. 1. A, Plasma glucose concentrations during the hyperglycemic clamp with arginine stimulation during saline (E) or GLP-1 infusion (F) in type 1 diabetic patients without endogenous insulin secretion. B, GIR during the hyperglycemic clamp with arginine stimulation during saline (E) or GLP-1 infusion (F) in type 1 diabetic patients without endogenous insulin secretion. GLP-1 infusions GLP-1 (7 3) amide was dissolved in saline containing 5 ml human serum albumin (human albumin 5%; ZLB Bering, Copenhagen, Denmark) and infused immediately after preparation. Analyses Plasma glucose was analyzed bedside using a blood gas analyzer (ABL 00 Flex; Radiometer Medical, Brønshøj, Denmark). C-peptide and insulin were analyzed with solid-phase two-site chemiluminescent immunometric assay (Immulite 2000; Diagnostic Products Corp., Los Angeles, CA) with reportable range of 33.1 95 and. 25 pmol/liter for C-peptide and insulin, respectively. Amylin was measured with an ELISA kit with immunofluorescent detection. Glucagon was measured by RIA with antibody code 305 directed against the C terminus of the glucagon molecule. Total GLP-1 was measured with anti-
29 Kielgast et al. GLP-1 and - and -Cell Function J Clin Endocrinol Metab, May 20, 95(5):292 29 plasma a glucagon (pm mol/l) 22 20 1 2 body code 9390 directed against the C terminus of the GLP-1 molecule. Detection limit was 1 pmol/liter for GLP-1, glucagon, and amylin. Statistics Results are presented as mean SEM. Comparison of normally distributed data was carried out using paired two-tailed t test and for nonnormally distributed data with a Wilcoxon signed rank test. Total and incremental area under the plasma concentration curves were calculated using the trapezoidal rule. P values 0.05 were considered statistically significant. Results 0-50 -0-30 -20-0 20 30 0 50 0 70 0 90 time (minutes) GLP-1/NaCl Glucose levels tended to be lower with GLP-1 between the two study days from time zero to min, but hereafter no difference was seen (Fig. 1A). GLP-1 slightly increased mean glucose infusion rates (GIR) (.3 0. vs. 5. 0. mg/kg min, P 0.03; Fig. 1B) and the amount of glucose used in the clamp (2 3. vs. 37 3. g, P 0.03). With saline, plasma concentration of total GLP-1 was maintained at basal levels of approximately 17 pmol/liter, whereas GLP-1 increased the plasma concentration of total GLP-1 to moderately supraphysiological levels with steady-state concentration of approximately 200 pmol/ liter 20 min after start of infusion. -Cell function One patient secreted detectable amounts of C-peptide after glucose and arginine stimulation plus GLP-1 infusion. This patient s plasma C-peptide remained very low, hyperglycemic clamp FIG. 2. Plasma glucagon concentrations during the hyperglycemic clamp with arginine stimulation at 5 min with infusion of GLP-1 (F) or saline (E) started 30 min before the clamp in type 1 diabetic patients without endogenous insulin secretion. To test for significance at specific time points, we performed a two-way ANOVA with Bonferroni correction for multiple comparisons., P 0.05. reaching a maximum of 55 pmol/liter after 90 min. When analyzing the effect of GLP-1 on glucagon secretion, data from this patient were excluded. In all other subjects, plasma C-peptide remained below the lower detection limit of 33 pmol/liter at all times on both study days. In a subgroup of four patients, we also measured amylin (which is co-secreted with insulin and present in the same granules) and found mean basal and stimulated amylin levels of 1. 0.3 and 1.9 0.3 pmol/liter as well as of 2. 0.3 and 2.3 0. pmol/liter with GLP-1 and saline, respectively (P value not significant), which are below normal fasting and stimulated values (17). Mean plasma insulin concentrations were 5 170 and 93 9 pmol/ liter (P 0.) with GLP-1 and saline, respectively. Values are higher than usual fasting levels because insulin detemir is albumin bound, which causes higher plasma concentrations of free and bound insulin detemir than equivalent doses of neutral protamine Hagedorn insulin (1). -Cell function Basal glucagon levels were.0 0.7 and.5 1.0 pmol/liter with GLP-1 and saline infusions, respectively (P 0.5). Immediately after start of GLP-1-infusion, plasma levels of glucagon were suppressed and remained lower compared with saline until after arginine stimulation (Fig. 2). Total area under plasma glucagon concentration curve from 30 to 90 min was significantly lower with GLP-1 compared with saline (5 72 vs. 70 97 pmol/liter min, P 0.001), and decremental area from 30 to 0 min (from start of infusion to arginine bolus) was also greater with GLP-1 (19 1 vs. 75 23 pmol/ liter min, P 0.03). Incremental area in response to arginine stimulation (0 75 min) was significantly smaller with GLP-1 than with saline (3 21 vs. 137 pmol/liter min, P 0.05). Discussion Our main finding is that GLP-1-induced glucagon suppression during hyperglycemia is intact in type 1 diabetic patients without endogenous insulin secretion, indicating that GLP-1 is capable of inhibiting -cell secretion whether endogenous insulin is present or not. Interestingly, the hyperglycemic clamp had no additional influence on glucagon suppression. The decrease in glucagon levels was accompanied by a slight but significant increase in mean GIR as well as of actual amount of glucose used in the clamp. We also conclude that arginine stimulation of glucagon release is independent of endogenous insulin but sensitive to the inhibitory action of GLP-1. Addition of pharmacological levels of GLP-1 on top of maximal -cell stimulation generally could
J Clin Endocrinol Metab, May 20, 95(5):292 29 jcem.endojournals.org 295 not reveal functional -cells in C-peptide-negative type 1 diabetic patients in the present study, although a single patient displayed a minimal C-peptide response after maximal -cell stimulation and GLP-1. This patient had 5 yr diabetes duration; he was diagnosed at age 37 and was negative for GAD-5 and islet cell antibodies. The limit of detection of C-peptide in our assay is 33 pmol/liter, so we cannot completely exclude the possibility that minimal amounts of insulin in the other patients may have escaped our analyses. However, in four patients, we also measured amylin and found very low basal levels and no evidence of a GLP-1 or glucose-induced increase. It is therefore unlikely that insulin or other -cell factors are responsible for the glucagonostatic effect of GLP-1 in our study. Several factors inhibit glucagon secretion, including glucose, insulin, and somatostatin (). In the present study, insulin and changes in glucose have been eliminated as possible mechanisms. However, GLP-1 may stimulate somatostatin secretion, which in turn may suppress glucagon release by paracrine interaction. In support of this, the inhibitory effect of GLP-1 on glucagon secretion was reduced during somatostatin-antibody administration and completely abolished during co-infusion of somatostatin receptor subtype 2 antagonist in the perfused rat pancreas (9), making somatostatin a possible mediator of the GLP-1-induced glucagon-lowering effect (at least in rodents). GLP-1 stimulates pancreatic somatostatin secretion in animal models and cell lines through interaction with GLP-1 receptors present on -cells (), and islet somatostatin inhibits glucagon secretion (13). However, a single study () failed to detect GLP-1 receptors in human -cells. Our result are consistent with earlier studies of the effects of GLP-1 on glucagon secretion in type 1 diabetic patients with some residual -cell function (), but in that study, a paracrine effect of residual insulin secretion in most of the patients could not be excluded. Interestingly, in another group of patients, it was demonstrated that after adequate insulinization and normoglycemia, which strongly suppressed glucagon secretion, GLP-1 did not have any further suppressive effect (19), consistent with powerful inhibitory effects of peripheral insulin. Other published studies involved meal stimulation where the GLP-1-induced inhibitory effect on gastric emptying complicates the interpretation of a GLP-1-induced effect on -cell function (20, 21). In conclusion, our results indicate that GLP-1 exerts its suppressive effect on glucagon secretion in the absence of endogenous insulin. However, this does not exclude that GLP-1 still may modulate -cell activity through intra-islet insulin, although endogenously secreted insulin did not seem to be a necessary factor in this study. During maximal -cell stimulation, GLP-1 induced endogenous insulin secretion in one of nine type 1 diabetic patients previously classified as C- peptide negative. Acknowledgments Address all correspondence and requests for reprints to: Urd Kielgast, M.D., Department of Endocrinology, Hvidovre Hospital, Kettegaard Alle 30, 250 Hvidovre, Denmark. E-mail: urd.kielgast@hvh.regionh.dk. Disclosure Summary: None of the authors listed has any disclosures in relation to this article. References 1. Fehmann HC, Habener JF 1992 Insulinotropic hormone glucagonlike peptide-i(7 37) stimulation of proinsulin gene expression and proinsulin biosynthesis in insulinoma TC-1 cells. Endocrinology 130:159 2. Holst JJ 2007 The physiology of glucagon-like peptide 1. Physiol Rev 7:09 39 3. Ding WG, Renström E, Rorsman P, Buschard K, Gromada J 1997 Glucagon-like peptide I and glucose-dependent insulinotropic polypeptide stimulate Ca 2 -induced secretion in rat -cells by a protein kinase A-mediated mechanism. Diabetes :792 00. Heller RS, Kieffer TJ, Habener JF 1997 Insulinotropic glucagon-like peptide I receptor expression in glucagon-producing -cells of the rat endocrine pancreas. Diabetes :75 791 5. Moens K, Heimberg H, Flamez D, Huypens P, Quartier E, Ling Z, Pipeleers D, Gremlich S, Thorens B, Schuit F 199 Expression and functional activity of glucagon, glucagon-like peptide I, and glucosedependent insulinotropic peptide receptors in rat pancreatic islet cells. Diabetes 5:257 21. Tornehave D, Kristensen P, Rømer J, Knudsen LB, Heller RS 200 Expression of the GLP-1 receptor in mouse, rat, and human pancreas. J Histochem Cytochem 5:1 51 7. Franklin I, Gromada J, Gjinovci A, Theander S, Wollheim CB 2005 -Cell secretory products activate -cell ATP-dependent potassium channels to inhibit glucagon release. Diabetes 5: 115. Dunning BE, Foley JE, Ahrén B 2005 -Cell function in health and disease: influence of glucagon-like peptide-1. Diabetologia :1700 1713 9. de Heer J, Rasmussen C, Coy DH, Holst JJ 200 Glucagon-like peptide-1, but not glucose-dependent insulinotropic peptide, inhibits glucagon secretion via somatostatin (receptor subtype 2) in the perfused rat pancreas. Diabetologia 51:223 2270. Ishihara H, Maechler P, Gjinovci A, Herrera PL, Wollheim CB 2003 Islet -cell secretion determines glucagon release from neighbouring -cells. Nat Cell Biol 5:330 335 11. Kim W, Egan JM 200 The role of incretins in glucose homeostasis and diabetes treatment. Pharmacol Rev 0:70 5. Gromada J, Holst JJ, Rorsman P 199 Cellular regulation of islet hormone secretion by the incretin hormone glucagon-like peptide 1. Pflugers Arch 35:53 59 13. Cejvan K, Coy DH, Efendic S 2003 Intra-islet somatostatin regulates glucagon release via type 2 somatostatin receptors in rats. Diabetes 52:117 111. Creutzfeldt WO, Kleine N, Willms B, Orskov C, Holst JJ, Nauck MA 199 Glucagonostatic actions and reduction of fasting hyperglycemia by exogenous glucagon-like peptide I(7-3) amide in type I diabetic patients. Diabetes Care 19:50 5
29 Kielgast et al. GLP-1 and - and -Cell Function J Clin Endocrinol Metab, May 20, 95(5):292 29 15. Meier JJ, Bhushan A, Butler AE, Rizza RA, Butler PC 2005 Sustained -cell apoptosis in patients with long-standing type 1 diabetes: indirect evidence for islet regeneration? Diabetologia :2221 222. Vilsbøll T, Toft-Nielsen MB, Krarup T, Madsbad S, Dinesen B, Holst JJ 2000 Evaluation of -cell secretory capacity using glucagon-like peptide 1. Diabetes Care 23:07 17. Mitsukawa T, Takemura J, Asai J, Nakazato M, Kangawa K, Matsuo H, Matsukura S 1990 Islet amyloid polypeptide response to glucose, insulin, and somatostatin analogue administration. Diabetes 39:39 2 1. Smeeton F, Shojaee Moradie F, Jones RH, Westergaard L, Haahr H, Umpleby AM, Russell-Jones DL 2009 Differential effects of insulin detemir and neutral protamine Hagedorn (NPH) insulin on hepatic glucose production and peripheral glucose uptake during hypoglycaemia in type 1 diabetes. Diabetologia 52:2317 2323 19. Meier JJ, Nauck MA 2005 Glucagon-like peptide 1(GLP-1) in biology and pathology. Diabetes Metab Res Rev 21:91 117 20. Behme MT, Dupré J, McDonald TJ 2003 Glucagon-like peptide 1 improved glycemic control in type 1 diabetes. BMC Endocr Disord 3:3 21. Dupre J, Behme MT, Hramiak IM, McFarlane P, Williamson MP, Zabel P, McDonald TJ 1995 Glucagon-like peptide I reduces postprandial glycemic excursions in IDDM. Diabetes :2 30 Visit the Online Store for the latest information in endocrinology! Check out CME Self-Assessment & MOC products. www.endo-society.org