THE INCRETIN hormones gastrin inhibitory polypeptide

Transcription

1 X/97/$03.00/0 Vol. 82, No. 3 Journal of Clinical Endocrinology and Metabolism Printed in U.S.A. Copyright 1997 by The Endocrine Society Rapid Oscillations in Plasma Glucagon-Like Peptide-1 (GLP-1) in Humans: Cholinergic Control of GLP-1 Secretion via Muscarinic Receptors* H. J. BALKS, J. J. HOLST, A. VON ZUR MÜHLEN, AND G. BRABANT Zentrum Innere Medizin, Abteilung Klinische Endokrinologie, Medizinische Hochschule Hannover, Hannover, Germany; and Institute of Medical Physiology C, The Panum Institute, University of Copenhagen (J.J.H.), Copenhagen, Denmark ABSTRACT The mechanisms involved in the rapid glucagon-like peptide-1 (GLP-1) release following glucose ingestion are poorly defined. Besides a direct intestinal stimulation of L cells, humoral and neuronal mechanisms have been discussed. We investigated the temporal pattern of GLP-1 release in five healthy men (aged yr; body mass index, kg/m 2 ) after an overnight fast for 60 min under basal conditions and for 60 min after an oral glucose load (OGL; 100 g) in both the presence and absence of atropine (80 ng/kg min, iv). Blood was sampled every 2 min, and data were evaluated for the temporal pattern of GLP-1 secretion by several computer-assisted programs (deconvolution, Pulsar analysis, and Fourier transformation). With all methods a pulsatile pattern of plasma GLP-1 levels with a frequency of five to seven per h was detected; this remained unchanged THE INCRETIN hormones gastrin inhibitory polypeptide (GIP) (1 4) and glucagon-like peptide-1-(7 36) amide [GLP-1-(7 36) amide] (5, 6) are secreted in response to ingestion of mixed meals, potently stimulating glucoseinduced insulin secretion from pancreatic -cells. In contrast to GIP, the insulinogenic property of GLP-1 is still conserved in type 2 diabetic patients (7). These metabolic effects have stimulated research on the potential of GLP-1 as a therapeutic agent for patients with type 2 diabetes mellitus (8, 10 15), whereas the mechanisms involved in the release of GLP-1 have not been fully elucidated as yet. Previous studies demonstrated a rapid release of GLP-1 after glucose ingestion (8, 16 18). As GLP-1-positive L- cells are dispersed in increasing number in the distal jejunum, ileum, and throughout the large bowel (19, 20), luminal stimulation of GLP-1 release by nutrients has been questioned. The increment in plasma glucose, however, is unlikely to be the cause of this early GLP-1 stimulation as iv application of glucose mimicking the increment following an oral glucose load did not alter GLP-1 secretion (8, 11, 21). Several compounds, including GIP, the cholinergic agonist betanecol, bombesin, and calcitonin gene-related Received August 8, Revision received November 15, Accepted November 26, Address all correspondence and requests for reprints to: H. J. Balks, M.D., Department of Internal Medicine, Division of Clinical Endocrinology, Medizinische Hochschule Hannover, Konstanty-Gutschow-Strasse 8, D Hannover, Germany @compuserve.com. * This work was supported in part by grants from the Danish Medical Research Council. in the different metabolic states and during atropine treatment. Glucose and GLP-1 plasma levels showed a parallel increase after OGL (OGL without atropine control: and min, respectively). Atropine infusion delayed this increase significantly ( and min, respectively; P 0.02). In contrast to plasma glucose concentrations ( % of control; P 0.05), atropine infusion reduced the integrated GLP-1 pulse amplitude to % of the control levels (P 0.05). In conclusion, GLP-1 is secreted in a pulsatile manner with a frequency comparable to that of pancreatic hormones. Mean GLP-1 plasma concentrations increase after OGL due to augmented GLP-1 pulse amplitudes but not frequency. The differential effect of atropine on glucose and GLP-1 plasma levels suggest a direct cholinergic muscarinic control of L cells. (J Clin Endocrinol Metab 82: , 1997) peptide, are stimulatory on GLP-1 secretion in vitro in rats (22, 23), opening the possibility for a regulatory enteroendocrine and neuroneuroendocrine loop between the proximal and the distal small intestine (22, 23). The cholinergic stimulation is dependent on muscarinic M3-subtype receptors, as indicated by a recent study using a murine intestinal cell line, STC1 (24). In contrast to the animal studies, infusion of synthetic GIP in humans has no effect on GLP-1 secretion (8). To better define the mechanisms involved in GLP-1 release in vivo, we investigated the temporal pattern of GLP-1 release after glucose ingestion by high frequency blood sampling in healthy young volunteers. Subjects and Methods Subjects and experimental protocol Five healthy men, aged yr, with a body mass index of kg/m 2, volunteered for the study. No subject had a family history of diabetes mellitus. They all were healthy at the time of investigation, had no history of gastrointestinal dysfunction, and had an unremarkable physical and biochemical examination. The investigations had been approved by the committee on medical ethics of the Medizinische Hochschule Hannover, and all subjects gave written consent. The secretory pattern of GLP-1 was investigated after an overnight fast over 60 min under basal conditions and after an oral glucose load (OGL; 100 g glucose) for another 60 min by sampling blood every 2 min via an indwelling central venous catheter (25). Under identical conditions a second OGL was performed, but a bolus of 1 mg atropine was given iv 30 min before the OGL followed by a constant infusion of 80 ng atropine/kg BW min. 786

2 PULSATILE SECRETION OF GLP-1, VAGAL CONTROL 787 Biochemical analysis Plasma glucose levels were determined enzymatically (25). Plasma GLP-1 concentrations were measured after ethanol extraction of plasma samples (26) as previously described (27), using an antiserum (89890) specific for the C-terminal region of GLP-1-(7 36) amide with a crossreactivity of less than 0.01% with the C-terminally truncated peptides GLP-1-(7 34) and GLP-1-(7 35) and 100% with GLP-1-(9 36) amide. Assay sensitivity was 1 pmol/l, intra- and interassay coefficients of variation were less than 5% and less than 11%, respectively. Analysis of the temporal secretion pattern and statistical evaluation The temporal pattern of GLP-1 secretion was analyzed by discrete deconvolution (DESADE) using a one-compartment model (28) with a plasma half-time of GLP-1 calculated from the original time series as previously described (25). The half-time of min (mean sd) found is comparable to data obtained in infusion and bolus injection studies using either a side-viewing or a C-terminal-specific GLP-1 antiserum (12, 29). For comparison, the Pulsar program (30) was used as an alternative pulse detection program, with G(1) 3.80, G(2) 2.26, G(3) 1.56, G(4) 1.13, and G(5) Finally, spectral analysis was performed using the Matlab program (Math Works, Natick, MA). To determine the OGL-induced increase in plasma glucose and hormone concentrations, the slopes of two succeeding regression lines (first regression line consisting of 12 points; second line of 8 points) were computed iteratively for succeeding points as previously described (25). Statistical analyses were performed using the SPSS/PC 3.1 program. If not stated otherwise, data are given as the mean sd. ANOVA was used for analysis of changes in parameters within the time series, and P 0.05 or lower was considered significant. Cross-correlation was calculated using the BMDP statistical package (31). Results Basal and glucose-stimulated GLP-1 plasma levels In all subjects plasma GLP-1 concentrations showed a pulsatile pattern both under basal conditions and after glucose ingestion (Fig. 1) as analyzed by the DESADE or Pulsar program. Spectral analysis revealed a dominant frequency of min comparable to the rate of min detected by DESADE and min by the Pulsar program. OGL did not significantly alter the number of GLP-1 pulses, but GLP-1 pulse amplitudes were significantly increased by glucose ingestion (Table 1). Scanning the individual time series for the first significant increase in plasma concentrations of glucose and GLP-1, plasma glucose and GLP-1 levels started to increase and min after OGL, respectively. Plasma levels of glucose and GLP-1 increased in parallel and reached peak values at 32 or 34 min, respectively, (difference not significant; Fig. 2). No cross-correlation of fasting glucose and GLP-1 plasma FIG. 1. Individual patterns of glucose and GLP-1 during fasting and after an oral glucose challenge (100 g) in three healthy volunteers (A C). Blood sampling was performed every 2 min starting at 0700 h ( 0 min), glucose load was given at 0800 h (60 min; closed circles). Under identical experimental conditions, a bolus of 1 mg atropine at 0730 h (30 min) was administered iv, followed by a constant infusion of 80 ng atropine/kg BW min (open circles). The upper panel shows plasma glucose levels, and the middle panels pulses obtained by DESADE and Pulsar analysis. In the bottom panel, plasma GLP-1 concentrations are shown.

3 788 BALKS ET AL. JCE&M 1997 Vol 82 No 3 TABLE 1. Analysis of GLP-1 secretion as the mean SD of five healthy volunteers during 1hoffasting (0 60 min) and 1 h ( min) after an oral glucose load (100 g) with and without iv administered atropine (80 ng/kg BW min) starting after 30 min of the fasting period Time segments during fasting and after glucose stimulation (min) Control Atropine Control Atropine Pulses (h 1 ) DESADE Pulsar Area of pulse amplitudes (pmol) DESADE a b Pulsar a b Frequency (min) of GLP-1 pulses estimated over 120 min DESADE Pulsar FT Pulse analysis was performed by deconvolution with DESADE, Pulsar analysis, and Fourier transformation (FT) of the original time series. Significant differences between the time segments were determined by ANOVA. a P 0.05 between basal and glucose-stimulated period within the control and atropine series. b P 0.05 between control and atropine. 30-min segment without and with iv atropine within and between comparable time segments of the two test situations, indicating no influence of atropine on the basal dynamics of GLP-1 secretion. However, atropine significantly attenuated both the increase in plasma glucose and GLP-1 concentrations in all subjects (Figs. 1 and 2) and reduced the area under the GLP-1 curve to % of controls. In contrast, the area under the curve for glucose was only reduced by approximately % (Table 2). Again, the frequency of GLP-1 pulses after glucose stimulation remained unaltered, but the lower plasma GLP-1 levels resulted from a selective effect on the amplitude of GLP-1 pulses (Table 1). Atropine infusion induced a significant (P 0.02) delay of min for the increase in plasma glucose concentrations and min for the increase in GLP-1 concentrations. However, individual analysis revealed a close temporal coupling of glucose and GLP-1 stimulation after OGL, with a significant cross-correlation (r ; P 0.02). Correlation of plasma glucose concentration to plasma GLP-1 concentrations resulted in a shift of the regression line, indicating that under the influence of atropine, plasma glucose levels were associated with lower GLP-1 levels. FIG. 2. Mean sd plasma levels of glucose and GLP-1 during an oral glucose load (100 g at 0800 h; 60 min) in five healthy volunteers. For methodology, see Fig. 1 (closed circles, control situation; open circles, atropine treatment). concentrations was found (r ; P NS), but after OGL, both parameters were significantly cross-correlated (r ; P 0.02). Atropine treatment Intravenously administered atropine did not affect basal plasma glucose or GLP-1 levels. The frequency of pulses during the fasting period was almost identical between the Discussion The secretion of pancreatic hormones is pulsatile, with a frequency of one pulse approximately every min (25, 32). The temporal pattern of GLP-1 secretion has not been elucidated as yet. In the present study we used three different and independent analytical methods to analyze the secretion dynamics of GLP-1. All three independent techniques used for pulse analysis revealed a pulsatile pattern of hormone release. One of these methods, the deconvolution technique DESADE, is based on the plasma half-life of the hormone, which we estimated from the original time series. The plasma half-life of GLP-1 may vary with the specificity of the antibody used, but as all three techniques independently revealed a comparable frequency of GLP-1 pulses, such technical problems are unlikely (29, 33). The mechanisms underlying GLP-1 pulse generation and the coordinated release from dispersed L cells in the distal

4 PULSATILE SECRETION OF GLP-1, VAGAL CONTROL 789 TABLE 2. Integrated concentrations of glucose and GLP-1 in time segments of 30 min during fasting and after an oral glucose challenge (100 g) with infusion of atropine (80 ng/kg BW min) starting after 30 min of the fasting period Integrated values Time interval (min) Controls (n 5) Atropine (n 5) P value Glucose (mmol/l min) a NS b b NS GLP-1 (pmol/l min) a NS b b NS Data are the mean SEM of five healthy volunteers. a Zero to 30 min basal situation without atropine infusion. b Difference between basal 30 min without and with atropine infusion, P (by ANOVA). parts of the intestinal tract (19, 20) are unclear. Stimulation of GLP-1 secretion by circulating glucose concentrations, luminal contact of L cells to ingested glucose, or neuronal modulations are possible explanations. Direct stimulation of basal GLP-1 secretion by circulating plasma glucose levels is an unlikely explanation, as the threshold to induce GLP-1 release in cell culture experiments (34) is higher than the circulating glucose concentrations. Moreover, iv glucose infusion in man failed to increase plasma GLP-1 concentrations even when mimicking the glucose increment achieved by OGL (18, 21). L Cells are located in the distal jejunum, with increasing density throughout the ileum and colon (19, 20). Rapid filling of the distal small intestine is a prerequisite for GLP-1 stimulation by intestinal luminal contact with glucose. Therefore, an extremely rapid gastric emptying and intestinal glucose transport have to be postulated to explain the parallel increase in plasma glucose and GLP-1 only 8 min after glucose ingestion. Using a -scintillation technique, we investigated in a preliminary study the gastric emptying of a liquid glucose load (100 g) containing a liquid tracer (40 millibecquerels 99m Tc-DTPA) into the small intestine. The proximal duodenum was reached only 6 8 min after oral ingestion of the glucose load (Balks H. J. and Mihlau E., unpublished results), closely resembling the time interval observed for the initial increase in either plasma glucose or GLP-1 in the present study. This fits with other findings questioning such rapid intestinal transport (18, 20). The autonomic nervous system is involved in the modulation of glucose homeostasis, coordinating gastric emptying, intestinal transport of ingested nutrients, and coupling of glucoregulatory hormones from the gut (e.g. GIP) and pancreas (22, 23, 35 38). In rats GIP is directly stimulating GLP-1 secretion in vitro (22, 23, 39), an effect that has not been confirmed in humans in vivo (21). However, cholinergic agonists potently stimulated GLP-1 secretion in vitro in the rat (23, 24, 39) through acetycholine M3 subtype receptors as reported recently (24). Thus, the cholinergic system may be a good candidate for the physiological regulation of GLP-1. This is supported by our finding of an attenuated increase in GLP-1 after glucose administration in the presence of atropine, suggesting a direct inhibition beyond the effects on intestinal motility, as GLP-1 pulse amplitude was disproportionally lower for a given glucose plasma level. Neuronal integration of GLP-1 release from L cells would help in understanding the pulsatile release mechanism in a dispersed cellular system such as L cells. This, furthermore, may explain the pulsatile nature of GLP-1 release and fits observations on the release pattern of duodenal K cells (40, 41) and pancreatic -cells (36 38), where the endogenous dynamic of single cells appears to be integrated by intrinsic and extrinsic humoral and neuronal factors (42 44). In summary, we have shown that GLP-1 secretion is pulsatile and that the mechanisms involved in controlling and coordinating L cells and in the formation of a pulsatile secretion pattern of GLP-1 depend on the parasympathetic nervous system in vivo in man. Acknowledgments We thank Mrs. D. Becker, Mrs. S. Baars, and Mrs. L. Albaek for their excellent assistance with the analytical procedures. References 1. Cataland S, Crockett SE, Brown JC, Mazzaferri EL Gastric inhibitory polypeptide (GIP) stimulation by oral glucose in man. J Clin Endocrinol Metab. 39: Pederson RA, Schubert HE, Brown JC Gastric inhibitory polypeptide its physiologic release and insulinotropic action in the dog. Diabetes. 24: Creutzfeldt W, Ebert R Release of gastric inhibitory polypeptide (GIP) to a test meal under normal and pathological conditions in man. In: Bajaj JS, ed. Proceedings of the Ninth International Congress of International Diabetes Federation. Amsterdam: Excerpta Medica; Fried M, Mayer EA, Bloom SR, Taylor IL, Meyer JH GIP and insulin release in relation to gastric emptying of a mixed meal in man. Regul Peptides. 26: Holst JJ, Ørskov C, Nielsen VO, Schwartz TW Truncated glucagon-like peptide-1, an insulin releasing hormone from the distal gut. FEBS Lett. 211: Kreyman B, Williams G, Ghatei MA, Bloom SR Glucagon-like peptide- 1(7 36): a physiological incretin in man. Lancet. 2: Komatsu R, Matsuyama T, Namba M, et al Glucagonostatic and insulinotropic action of glucagon-like polypeptide I(7 36)amide. Diabetes. 38: Nauck M, Heimesaat MM, Ørskov C, Holst JJ, Ebert R, Creutzfeldt W Preserved incretin activity of glucagon-like peptide-1 [7 36 amide] but not of synthetic human gastric inhibitory polypeptide in patients with type-2 diabetes mellitus. J Clin Invest. 91: Hvidberg A, Nielsen M-BT, Hilsted J, Ørskov C, Holst JJ Effect of glucagon-like peptide-1 (proglucagon amide) on hepatic glucose production in healthy man. Metabolism. 43: Nauck MA, Kleine N, Ørskov C, Holst JJ, Willms B, Creutzfeldt W Normalization of fasting hyperglycemia by exogenous glucagon-like peptide 1(7 36)amide in type 2 (non-insulin-dependent) diabetic patients. Diabetologia. 36: Gutniak M, Ørskov C, Holst JJ, Ahren B, Efendic S Antidiabetogenic effect of glucagon-like peptide-1(7 36)amide in normal subjects and patients with diabetes mellitus. N Engl J Med. 326: Ørskov C, Wettergren A, Holst JJ Biological effects and metabolic rates

5 790 BALKS ET AL. JCE&M 1997 Vol 82 No 3 of glucagon-like peptide amide and glucagon-like peptide in healthy subjects are indistinguishable. Diabetes. 42: Nathan DM, Schreiber E, Fogel H, Mojsov S, Habener J Preliminary studies of insulinotropic actions of glucagon-like peptide-i(7 37) administered to diabetic and nondiabetic human subjects. Diabetes Care. 15: Gutniak MK, Linde B, Efendic S Subcutaneous injection of glucagonlike insulinotropic peptide reduces postprandial glycemia in non-insulin-dependent diabetes. Digestion. 54: Elahi D, McAloon-Dyke M, Fukagawa NK, et al The insulinotropic actions of glucose-dependnet insulinotropic polypeptide (GIP) and glucagonlike peptide-1 (7 37) in normal and diabetic subjects. Regul Pept. 51: Ørskov C, Jeppesen J, Madsbad S, Holst JJ Proglucagon products in plasma of non insulin-dependent diabetics and non diabetic controls in the fasting state and after oral glucose and intravenous arginin. J Clin Invest. 87: Elliot RM, Morgan LM, Tredger JA, Deacon S, Wright J Glucagon-like peptide-1(7 36)amide and glucose-dependent insulinotropic polypeptide secretion in response to nutrient ingestion in man: acute post-prandial and 24-h secretion patterns. J Endocrinol. 138: Herrmann C, Göke R, Richter G, Fehmann HC, Arnold R, Göke B Glucagon-like peptide-1 and glucose-dependent insulin releasing polypeptide plasma levels in response to nutrients. Digestion. 56: Larsson LI, Holst JJ, Hakanson R, Sundler F Distribution and properties of glucagon immunreactivity in the digestive tract of various mammals: an immunohistochemical and immunochemical study. Histochemistry. 44: Eissele R, Göke R, Willemer S, Harthus HP, Vermeer H, Arnold R, Göke B Glucagon-like peptide-1 cells in the gastrointestinal tract and pancreas of rat, pig and man. Eur J Clin Invest. 22: Nauck MA, Bartels E, Orskov C, Ebert R, Creutzfeldt W Additive insulinotropic effects of exogenous synthetic human gastric inhibitory polypeptide and glucagon-like peptide-1-(7 36) amide infused at near-physiological insulinotropic hormone and glucose concentrations. J Clin Endocrinol Metab. 76: Roberge JN, Brubaker PL Regulation of intestinal proglucagon-drived peptide secretion by glucose-dependent insulinotropic peptide in a novel enteroendocrine loop. Endocrinology. 133: Brubaker PL Regulation of intestinal proglucagon-derived peptide secretion by intestinal regulatory peptides. Endocrinology. 128: Abello J, Ye F, Bosshard A, Bernard C, Cuber JC, Chayvialle JA Stimulation of glucagon-like peptide-1 secretion by muscarinic agonist in a murine intestinal endocrine cell line. Endocrinology. 134: Balks HJ, Schmidt A, Prank K, Hemmer F, Von zur Mühlen A, Brabant G Temporal pattern of pancreatic insulin and C-peptide secretion and of plasma glucose levels after nutritional stimulation. J Clin Endocrinol Metab. 75: Ørskov C, Holst JJ Radio-immunoassays for glucagon-like peptides 1 and 2 (GLP-1 and GLP-2). Scand J Clin Lab Invest. 47: Ørskov C, Rabenhøj L, Wettergren A, Kofod H, Holst JJ Tissue and plasma concentrations of amidated and glycine extended glucagon-like peptide-1 in humans. Diabetes. 43: Brabant G, Prank K, Ranft U, Schuermeyer T, Wagner TOF, Hauser H, Kummer B, Feistner H, Hesch RD, Von zur Mühlen A Physiological regulation of circadian and pulsatile thyrotropin secretion in normal men and woman. J Clin Endocrinol Metab. 70: Deacon CF, Johnson AH, Holst JJ Degradation of glucagon-like peptide-1 by human plasma in vitro yields an N-terminally truncated peptide that is a major endogenous metabolite in vivo. J Clin Endocrinol Metab. 80: Merriam GR, Wachter KW Algorythms for the study of episodic hormone secretion. Am J Physiol. 243:E310 E Dixon WJ, ed BMDP statistical software; program IT. Berkley: University of California Press. 32. Jaspan JB, Lever E, Polonsky KS, Van Cauter E In vivo pulsatility of pancreatic islet peptides. Am J Physiol. 251:E215 E Deacon CF, Nauck M, Toft-Nielsen M, Pridal L, Willms B, Holst JJ Both subcutaneously and intravenously administered glucagon-like peptide I are rapidly degraded from the NH 2 -terminus in type II diabetic patients and in healthy subjects. Diabetes. 44: D Alessio DA, Fujimoto WY, Ensinck JW Effects of glucagon-like polypeptide I(7 36) on the release of insulin, glucagon and somatostatin by rat pancreatic islet cell monolayer cultures. Diabetes. 38: Cooke AR Control of gastric emptying and motility. Gastroenterology. 68: Frohman LA, Bernardis LL Effect of hypothalamic stimulation on plasma glucose, insulin and glucagon levels. Am J Physiol. 221: Gerich JE, Lorenzi M The role of the autonomic nervous system and somatostatin in the control of insulin and glucagon secretion. Front Neuroendocrinol. 5: Hedo JA, Villanueva ML, Marco J Stimulation of pancreatic polypeptide and glucagon secretion by 2-deoxy-d-glucose in man: evidence for cholinergic mediation. J Clin Endocrinol Metab. 47: Herrmann-Rinke C, Vöge A, Hess M, Göke B Regulation of glucagonlike peptide-1 secretion from rat ileum by neurotransmitters and peptides. J Endocrinol. 147: Creutzfeldt W The incretin concept today. Diabetologia. 16: Larrimer JN, Mazzaferri EL, Cataland S, Mekhjian HS The effect of atropine on glucose-stimulated gastric inhibitory polypeptide. Diabetes. 27: Stagner JE, Samols W, Weir GC Sustained oscillations of insulin, glucagon and somatostatin from the isolated canine pancreas during exposure to a constant glucose concentration. J Clin Invest. 65: O Meara NM, Sturis J, Blackman JD, et al Oscillatory insulin secretion after pancreas transplant. Diabetes. 42: Porksen N, Munn S, Ferguson D, O Brien T, Veldhuis J, Butler P Coordinate pulsatile insulin secretion by chronic intraportally transplanted islets in the isolated perfused rat liver. J Clin Invest. 94: