- Email: [email protected]
Sympathetic pathways mediate GLP-1 actions in the gastrointestinal tract of the rat
Regulatory Peptides 74 (1998) 19–25
Sympathetic pathways mediate GLP-1 actions in the gastrointestinal tract of the rat M. Giralt, P. Vergara* ` de Barcelona, 08193 Bellaterra, Spain Department of Cell Biology and Physiology, Veterinary School, Universitat Autonoma Received 26 July 1997; received in revised form 23 January 1998; accepted 23 January 1998
Abstract The aim of this study was to establish the actions of GLP-1 (7-37) on gastrointestinal motility in rats. We prepared anaesthetized Sprague-Dawley rats with strain-gauges in the antrum, duodenum and the proximal jejunum and a catheter in the aorta close to the coeliac artery for close infusion of substances. Intraarterial GLP-1 infusions (3 3 10 210 and 3 3 10 29 moles / kg per 10 min) (n 5 8) induced inhibition of spontaneous motor activity in the antrum, duodenum and proximal jejunum. Inhibition induced by GLP-1 was reversed by i.v. infusion of GLP-1 receptor antagonist, Exendin (9-39) (3 3 10 28 moles / kg per 10 min) (n 5 6). Neither the presence of L-NNA (10 25 moles / kg) (n 5 9) nor the VIP receptor antagonist [4-Cl-D-Phe 6 , Leu 17 ]-VIP (3 3 10 28 moles / kg per 10 min) (n 5 8) modified responses to GLP-1. However, a combination of the adrenergic blockers phentolamine and propranolol (1 mg / kg each) (n 5 8) completely blocked motor actions of GLP-1 in all the organs studied. Moreover, inhibition of gastrointestinal motor activity by GLP-1 was blocked by previous infusion of hexamethonium (10 mg / kg) (n 5 4). This study demonstrates that GLP-1 inhibits gastrointestinal motor activity of the rat acting on specific GLP-1 receptors and via stimulation of adrenergic pathways. 1998 Elsevier Science B.V. Keywords: Intestinal motor patterns; Gastric emptying; Adrenoceptors; Exendin; Ileus; Intestinal inhibition
1. Introduction Glucagon-like peptide 1 (GLP-1) has recently attracted considerable interest because it acts as an incretin, making an important contribution to insulin secretion via the enteroinsular axis [1,2] and because it may be the most potent inhibitor of feeding yet identified in the rat [3]. Furthermore, GLP-1 can act as an enterogastrone, inhibiting gastric acid secretion in man [4,5] as well as decreasing gastric emptying in parallel with an inhibition of pancreatic secretion [6]. Mammalian proglucagon consists of 160 amino acid residues synthesized in the alpha cells of the islets of Langerhans, in the L cells of the ileum and colonic mucosa as well as in the brain [7,3]. The intestinal L cell produces two elongated forms of glucagon, glicentin and oxy*Corresponding author. Tel.: 1 34 3 581 1848; fax: 1 34 3 581 2006; e-mail: [email protected] 0167-0115 / 98 / $19.00 1998 Elsevier Science B.V. All rights reserved. PII S0167-0115( 98 )00010-X
ntomodulin, from the amino-terminal region, and another biologically active peptide, GLP-1 (proglucagon 78-107amide), from the carboxyl-terminal portion of the proglucagon molecule. Furthermore, the mammalian glucagon amino acid sequence is highly homologous to the anglerfish glucagon and is identical in various species of mammals such as the rat, hamster, guinea pig, bovine and human. This homology indicates evolutionary pressure to preserve this sequence and is consistent with a specific biological role [8]. The most important circulating intestinal glucagon gene product appears to be GLP-1 [9]. Luminal glucose seems to be the main GLP-1-releaser of meal constituents. The levels of this peptide peaked and reached statistical significance compared with basal levels after 30 min of ingestion of carbohydrate [1]. Additionally, cholinergic agonists, peptides of the bombesin family, CGRP and GIP induce GLP-1 secretion [10]. Although inhibitory actions of glucagon have been well
20
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
known for a long time [11], to our knowledge there have been no studies on the mechanism of action of glucagon peptides in the small intestine. However, recently exogenous GLP-1 has been shown to markedly produce an inhibition of migrating motor complexes (MMC) [12] and to inhibit contractile activity in the antro-pyloro-duodenal segment [13]. The purpose of this study was to further investigate the effects and the mechanism of action of GLP-1 (7-37) (proglucagon 78-108) on gastrointestinal motility in rats.
2. Methods
2.1. Animal preparation Male Sprague-Dawley rats, 8 to 10 weeks old and weighing 300–350 g, fasted for 12–16 h, were anaesthetized by inhalation of halothane to allow cannulation of the right jugular vein with a polyethylene tubing. Level III of anaesthesia was induced with i.v. infusion of thiopental sodium (20 mg / kg) and maintained with bolus infusions of anaesthetic as required. Body temperature was kept constant by placing the animals on a heating pad maintained at 378C. Rats were tracheotomized to facilitate spontaneous breathing. A polyethylene tube was inserted through the carotid artery to the aorta and placed at the bifurcation of coeliac artery for close infusion of drugs. The intestine was exposed through a midline abdominal incision. Three 3 3 5-mm strain-gauges (Hugo Sachs Elektronik, Germany) were sutured to the antrum, the duodenum (1.5 cm from the pylorus) and the jejunum (5 cm from Treitz’s ligament) oriented to record circular muscle activity. Strain-gauges were connected to high-gain amplifiers (Lectromed MTP8) and amplified signals were recorded in a polygraph.
2.2. Drugs Drugs used in this study were: Glucagon-like peptide 1 (7-37) (GLP-1) (Peptide Institute, Osaka, Japan), GLP-1 antagonist Exendin (9-39) (Bachem, Cardiff, UK), VIP receptor antagonist [4-Cl-D-Phe 6 , Leu 17 ]-VIP (Peninsula Laboratories, Belmont, CA), propranolol hydrochloride and phentolamine mesylate (Research Biochemical, Natick, MA), hexamethonium, L-nitroarginine ( L-NNA), yohimbine and carbachol (Sigma Chemical, St Louis, MO).
min before the start of GLP-1 infusion until the end of peptide infusion (n 5 6); (b) a bolus of hexamethonium (10 mg / kg, i.v.) was given. Then, in order to maintain mechanical activity after hexamethonium, carbachol (1.5 3 10 28 moles / kg) was continuously infused during GLP-1 infusion (n 5 4); (c) VIP receptor antagonist [4-Cl-D-Phe 6 , Leu 17 ]-VIP (3 3 10 28 moles / kg per 10 min, i.a.) was infused 2 min before the start and during GLP-1 infusion (n 5 8); (d) L-NNA (10 25 moles / kg, i.a.) was infused 20 min before the beginning of GLP-1 infusion (n 5 9); (e) a bolus of a combination of phentolamine and propranolol (1 mg / kg, each, i.v.) was infused 10 min before GLP-1 infusion (n 5 8); (f) a bolus of yohimbine (0.3 mg / kg, i.v.) was infused 10 min before GLP-1 infusion (n 5 5); (g) propranolol (1 mg / kg, i.v.) was infused 10 min before GLP-1 (n 5 4).
2.4. Analysis of data Mechanical activity was evaluated by measuring the area under the curve (AUC) of the mechanical activity tracing for a period of 10 min before and during peptide infusion. Data are expressed as mean6SEM. Statistical analysis was performed by means of either ANOVA, nonparametric repeated measure test or paired nonparametric test when applicable. Post-tests were Bonferroni and Dunnett’s test accordingly. Differences were considered significant when p , 0.05.
3. Results
3.1. Effects of GLP-1 on gastrointestinal mechanical activity During the preinfusion period, all the organs studied i.e. antrum, duodenum and jejunum, exhibited spontaneous activity consisting of contractions with an approximate frequency of 4 contractions every 10 min which propagated from the antrum to the jejunum (Fig. 1). Intraarterial infusion of GLP-1 at 3 3 10 210 moles / kg or higher, induced inhibition of the spontaneous motor activity in the antrum, duodenum and proximal jejunum (Fig. 1a). Smaller doses of GLP-1 did not modify spontaneous mechanical activity.
2.3. Experimental procedures
3.2. Mechanism of action of GLP-1 on gastrointestinal mechanical activity
GLP-1 (3 3 10 210 moles / kg per 10 min, i.a.) was infused for 10 min at a rate of 100 ml / min (n 5 8) in order to evaluate GLP-1 actions on studied organs. To study the mechanism of action of GLP-1, the following protocols were followed: (a) GLP-1 receptor antagonist, Exendin (9-39) (3 3 10 28 moles / kg per 10 min, i.v.) was infused 2
Intravenous infusion of GLP-1 receptor antagonist, Exendin (9-39) completely blocked inhibitory responses to GLP-1 (Fig. 1b). Hexamethonium completely blocked spontaneous activity (Fig. 2). In consequence, carbachol was infused continuously to maintain phasic activity. In these conditions, GLP-1 infusion at a dose of 3 3 10 210
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
21
Fig. 1. Mechanical activity of the antrum (A), duodenum (D) and jejunum (J). (a) Effect of GLP-1 (3 3 10 210 moles / kg per 10 min) close i.a. infusion. (b) Lack of effect of GLP-1 in the presence of antagonist exendin (3 3 10 28 moles / kg per 10 min).
Fig. 2. Mechanical activity of the antrum, duodenum and jejunum. Effects of hexamethonium, carbachol and the lack of effect of GLP-1. Abbreviations as in Fig. 1.
moles / kg did not modify phasic activity in any of the studied organs (Fig. 2). Neither L-NNA nor VIP antagonist [4-Cl-D-Phe 6 , Leu 17 ]-VIP modified GLP-1 response. Whereas VIP antagonist did not modify basal motility, L-NNA significantly increased motor activity in the jejunum, but not in the
antrum or duodenum (Fig. 3). Finally, previous infusion of the adrenergic blockers phentolamine and propranolol did not modify basal activity but completely blocked the inhibitory response to GLP-1 in all the studied organs (Fig. 4). The next step was to study the type(s) of adrenoceptor
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
22
Fig. 3. Lack of effect of L-NNA (a) and VIP antagonist [4-Cl-D-Phe 6 , Leu 17 ]-VIP (b) on GLP-1 response. *, Significant differences between basal and GLP-1 at each organ.
involved in GLP-1 response. However, this task became rather difficult. Infusion of the a 2 -adrenoceptor antagonist yohimbine in 3 out of 5 experiments completely blocked the spontaneous mechanical activity in the duodenum and jejunum, while in the antrum it induced a different pattern characterized by regular mechanical contractions in a 3–5 cycles per min frequency. In these experiments it was not possible to test the GLP-1 effects. In the other two experiments yohimbine did not block GLP-1-induced inhibition. Finally, to study the role of b-adrenergic receptors in GLP-1 response, a bolus of propranolol (1 mg / kg, i.v.) was infused (n 5 4). Propranolol did not modify basal activity. In the duodenum and jejunum, GLP-1 infusion still induced an inhibition of mechanical activity in 3 out of 4 cases, while in the antrum, propranolol gave contradictory results. Response to GLP-1 was blocked in two out of 4 cases (Fig. 5) and remained unchanged in the other two.
4. Discussion This study demonstrates that GLP-1 inhibits sponta-
neous activity of the proximal gastrointestinal tract of the rat by acting on specific GLP-1 receptors. In the mechanism of action, sympathetic pathways are involved. Glycine-extended GLP-1 (7-37) rather than carboxyamidated form corresponding to GLP-1 (7-36 amide) was used. It is known that for some amidated peptides, amide is crucial for biological activity. But this is not the case with GLP-1 molecules [7] because both GLP-1 (7-36) amide and GLP-1 (7-37) stimulate insulin secretion with the same potency [14]. Both peptide forms have been suggested as a possible new treatment for noninsulin-dependent Diabetes mellitus [15]. Glucagon-like peptide-1 induced an inhibitory action in all the studied areas. An inhibition of motor activity by other glucagon family peptides had already been reported [11]. However, the fact that the inhibitory action observed in our study was specifically blocked by the GLP-1 receptor antagonist exendin, indicates a specific action on GLP-1 receptors. This specificity together with the presence of GLP-1-containing cells in the distal areas of the intestine [16] and the increase of GLP-1 levels postprandially [1], suggests a role of this peptide in the control of gastrointestinal motor activity. Our results also agree with the inhibitory function reported for this peptide in the gastroduodenal area in humans [13]. With the specificity of GLP-1 actions established, our focus was to study the mechanism of action of GLP-1. We started by studying the involvement of either NO or VIP on GLP-1 actions. These two neurotransmitters seem to be responsible for the tonic inhibition of the intestine in the rat [17,18] and both have been suggested as responsible for NANC transmission [19,20]. However, neither the blockade of NO synthesis nor the VIP antagonist infusion modified inhibitory response to GLP-1. On the contrary, the simultaneous infusion of phentolamine and propranolol completely blocked response to GLP-1, demonstrating an involvement of adrenergic pathways in GLP-1 actions. To our knowledge this is the first evidence of an intestinal peptide modulating gastrointestinal motility through adrenergic pathways. In agreement with our results, a b-adrenergic mechanism has been reported in GIP- and GLP-1induced insulin secretion [21]. Furthermore, a nonspecific effect of adrenergic antagonists can be discarded since they did not modify either basal activity or the inhibitory response to infused VIP (data not shown). Stimulation of adrenoceptors in the gastrointestinal tract induces an inhibition of all the intestinal patterns including phase III of the MMC where acetylcholine is involved [22,23]. The main mechanism of action is the presynaptic inhibition of cholinergic neurons [24,25]. Furthermore, it is well established that cholinergic neurons possess a-adrenergic receptors that inhibit release of acetylcholine [26]. However, several authors have also demonstrated that a 2 receptors inhibit NO synthesis in the canine ileocolonic junction [27–29]. That also seems to be the case in the rat gastrointestinal tract where yohimbine causes inhibition of
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
23
Fig. 4. Effect of adrenergic blockers, phentolamine and propranolol, on GLP-1 response. (a) Mechanical activity; (b) summary of effects. Abbreviations as in Fig. 1. *, Significant differences between basal and GLP-1; 1 , significant differences between GLP-1 in the presence and absence of antagonists.
Fig. 5. Tracing of one experiment showing the effect of propranolol on GLP-1 response. Abbreviations as in Fig. 1.
24
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
spontaneous motor activity. This inhibition seems to be a consequence of an NO increase because it was blocked by L-NNA (data not shown). Spontaneous motility in our model was also blocked either by hexamethonium or atropine, a similar response to that observed in GLP-1. The lack of effect of GLP-1 in the presence of hexamethonium suggests that GLP-1 action is due to presynaptic inhibition of cholinergic neurons. Determining type(s) of receptor involved in the response turned out to be difficult. The reason being that neither antagonist given alone was able to consistently block the effect of GLP-1. This indicates that both alfa and beta receptor stimulation occurs after GLP-1 infusion. It is also of interest that in some of the experiments performed with propranolol, there was a blockade of GLP-1 response indicating the variability of the response as well as the involvement of beta receptors. Although a beta adrenoceptor with atypical characteristics has been described in the rat intestine [30], our results could also indicate that GLP-1 could induce a long reflex involving a large number of sympathetic pathways or, as has been suggested in humans, a central action of the peptide [31]. The fact that GLP-1 cells are located in the mucosa of distal parts of the intestine could imply that the GLP-1 release might affect sympathetic afferents. In this sense, colonic sympathetic afferents controlling the gastroduodenal area have been described [32]. In consequence, a GLP-1 action on these fibres which could activate an ileal brake mechanism is also likely. It has also been reported that GLP-1 induces insulin release [33]. Although insulin-induced changes to motility have been described, in the rat these actions are excitatory [34] and not inhibitory as we always saw with GLP-1. Therefore these peptide actions seem to be direct and nonmediated by insulin release. In conclusion, GLP-1 induces inhibition of gastrointestinal motor activity. Stimulation of adrenergic pathways mediates this inhibition. The mechanism of action is independent of both NO and VIP, suggesting that there is no inhibitory motoneuron stimulation. Although a central action of this peptide has been reported to mediate satiety [33], inhibitory motor actions of GLP-1 might also contribute to a satiety feeling. In the same way, GLP-1-induced motor inhibition could indirectly decrease plasma glucose concentration by slowing glucose absorption.
Acknowledgements ´ This work has been financially supported by Direccion ´ en Ciencia y Tecnologıa ´ General de Investigacion (DGICYT PB92-0638 and PM95-0121. The authors are grateful to Mr Antonio Acosta for his skilful assistance in the care of animals. They are also in debt to Prof. E.E. Daniel for his encouragement on the study of GLP-1 and
to Mr. A.C. Hudson for the editorial revision of the manuscript.
References [1] Elliott RM, Morgan LM, Tredger JA, Deacon S, Wright J, Marks V. 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 1993;138:159–66. [2] Parker JC, Hargrove DM, Shepherd KL, Nardone NA, Andrews KM, Persson LM. Glucagon-like peptide-1 (7-37) and glibenclamide stimulate insulin secretion and biosynthesis by different glucosedependent mechanisms: comparison in vivo and in vitro. Digestion 1993;54:357–8. [3] Turton MD, O’Shea D, Gunn I, Beak SA, Edwards CMB, Meeran K, Choi SJ, Taylor GM, Heath MM, Lambert PD, Wilding JPH, Smith DM, Ghatei MA, Herbert J, Bloom SR. A role for glucagonlike peptide-1 in the central regulation of feeding. Nature 1996;379:69–72. [4] Schjoldager BTG, Mortensen PE, Christiansen J, Orskov C, Holst JJ. GLP-1 (glucagon-like peptide 1) and truncated GLP-1, fragments of human proglucagon, inhibit gastric acid secretion in humans. Digest Dis Sci 1989;34:703–8. [5] O’Halloran SJ, Nikou GC, Kreymann B, Ghatei MA, Bloom SR. Glucagon-like peptide-1 (7-36)-NH 2 : a physiological inhibitor of gastric acid secretion in man. J Endocrinol 1990;126:169–73. [6] Wettergren A, Schjoldager B, Mortensen PE, Myhre J, Christiansen J, Holst JJ. Truncated GLP-1 (proglucagon 78-107-amide) inhibits gastric and pancreatic functions in man. Digest Dis Sci 1993;38:665–73. [7] Kreymann B, Yiangou Y, Kanse S, Williams G, Ghatei MA, Bloom SR. Isolation and charaterization of GLP-1 7-36 amide from rat intestine. FEBS Lett 1988;242:167–70. [8] Lund PK, Goodman RH, Montminy MR, Dee PC, Habener JF. Anglerfish islet preproglucagon II. Nucleotide and corresponding amino acid sequence of the cDNA. J Biol Chem 1983;258:3280–4. [9] Orskov C, Bersani M, Johnsen AH, Hojrup P, Holst JJ. Complete sequences of glucagon-like peptide-1 from human and pig small intestine. J Biol Chem 1989;264:12826–9. ¨ A, Goke ¨ B. Regulation of glucagon-like peptide [10] Herrmann C, Voge release from the isolated perfused rat ileum by nutrients, peptides and neuromediators. Digestion 1993;54:367. [11] Whalen GE. Glucagon and the small gut. Gastroenterology 1974;67:1284–6. ¨ PM. Direct inhibitory [12] Tolessa T, Gutniak M, Efendic S, Hellstrom action of glucagon-like peptide-1 on migrating myoelectric complex independently of insulin or somatostatin release. Neurogastroent Motil 1996;8:194 (Abstract). ¨ [13] Schirra J, Wank U, Houck P, Arnold P, Goke B, Katschinski M. Effects of GLP-1 on human antro-pyloro-duodenal motility. Neurogastroent Motil 1996;8:190 (Abstract). [14] Orskov C, Wettergren A, Holst JJ. Biological effects and metabolic rates of glucagon-like peptide-1 7-36 amide and glucagon-like peptide 1 7-37 in healthy subjects are indistinguishable. Diabetes 1993;42:658–61. [15] Habener JF, Mcmanus K, Nathan DM. Insulinotropic actions of glucagon-like peptide-1 (7-37) in diabetic and nondiabetic subjects. Digestion 1993;54:376–7. ¨ R, Willemer S, Harthus H-P, Vermeer H, Arnold [16] Eisselein R, Goke ¨ B. Glucagon-like peptide-1 cells in the gastrointestinal tract R, Goke and pancreas of rat, pig and man. Eur J Clin Invest 1992;22:283–91. ´ ´ ´ ˜ [17] Rodrıguez-Membrilla A, Martınez V, Jimenez M, Gonalons E, Vergara P. Is the nitric oxide the final mediator regulating the
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
[18]
[19]
[20]
[21]
[22] [23]
[24]
[25]
[26]
migrating myoelectric complex cycle?. Am J Physiol 1995;268:G207–14. ¨ PM, Ljung T. Nitrergic inhibition of migrating myoelecHellstrom tric complex in the rat is mediated by vasoactive intestinal peptide. Neurogastroenterol Mot 1996;8:299–306. Grider JR, Cable MB, Said SI, Makhlouf GM. Vasoactive intestinal peptide as a neural mediator of gastric relaxation. Am J Physiol 1985;248:G73–8. Daniel EE, Haugh C, Woshowska Z, Cipris S, Jury J, Fox-Threlheld JET. Role of nitric oxide-related inhibition in intestinal function: relation to vasoactive intestinal polypeptide. Am J Physiol 1994;266:G31–9. Ikeda T, Ochi H, Ohtani I, Fujiyama K, Hoshino T, Tanaka Y, Takeuchi T, Mashiba H. Possible role of the adrenergic mechanism in gastric inhibitory polypeptide- and glucagon-like peptide 1 (7-36) amide-induced insulin release in the rat. Metabolism 1993;42:209– 13. Kosterlitz HW, Lees GM. Pharmacological analysis of intrinsic intestinal reflexes. Pharmacol Rev 1964;16:301–39. Code CF, Marlett JA. The interdigestive myoelectric complex of the stomach and small bowel of dogs. J Physiol London 1975;246:289– 309. Jansson G, Martinson J. Studies on the ganglionic site of action of sympathetic outflow to the stomach. Acta Physiol Scand 1966;68:184–92. Sakai Y, Daniel EE, Jury J, Fox JET. Neurotensin inhibition of canine intestinal motility in vivo via a-adrenoceptors. Can J Physiol Pharmacol 1984;62:403–11. Wikberg J. Localization of adrenergic receptors in guinea pig ileum
[27]
[28]
[29]
[30]
[31] [32]
[33]
[34]
25
and rabbit jejunum to cholinergic neurons and to smooth muscle cells. Acta Physiol Scand 1977;99:190–207. Boeckxstaens GE, De man JG, Pelckmans PA, Herman AG, Van Maercke YM. a 2 -adrenoceptor-mediated modulaton of the nitrergic innervation of the canine isolated ileocolonic junction. J Pharmacol 1993;109:1079–84. De man JG, Boeckxstaens GE, Herman AG, Pelckmans PA. Effect of potassium channel blockade and a 2 -adrenoceptor activation on the release if nitric oxide from nonadrenergic noncholinergic nerves. Br J Pharmacol 1994;112:341–5. De man JG, Boeckxstaens GE, Herman AG, Pelckmans PA. Effect of potassium channel blockade and a 2 -adrenoceptor activation on the release if nitric oxide from nonadrenergic noncholinergic nerves. Can J Physiol Pharmacol 1984;62:403–11. Vliet A, Rademaker B, Bast A. A beta adrenoceptor with atypical characteristics is involved in the relaxation of the rat small intestine. J Pharmacol Exp Ther 1990;255:218–26. Holst JJ. Glucagon-like peptide 1: a newly discovered gastrointestinal hormone. Gastroenterology 1994;107:1848–55. Messenger JP, Furness JB. Calbindin-immunoreactive nerve terminals in the guinea pig celiac ganglion originate from colonic nerve cells. J Auton Nerv Syst 1991;35:133–42. Holst JJ, Orskov C, Vagn nielsen O, Schwartz TW. Truncated glucagon-like peptide is, an insulin-releasing hormone from the distal gut. FEBS Lett 1987;211:169–74. Allescher HD, Willis S, Schusdziarra V, Classen M. Modulatory effect of insulin on rat small intestinal motility and peptide release in vitro. Digestion 1991;48:192–201.
Sympathetic pathways mediate GLP-1 actions in the gastrointestinal tract of the rat M. Giralt, P. Vergara* ` de Barcelona, 08193 Bellaterra, Spain Department of Cell Biology and Physiology, Veterinary School, Universitat Autonoma Received 26 July 1997; received in revised form 23 January 1998; accepted 23 January 1998
Abstract The aim of this study was to establish the actions of GLP-1 (7-37) on gastrointestinal motility in rats. We prepared anaesthetized Sprague-Dawley rats with strain-gauges in the antrum, duodenum and the proximal jejunum and a catheter in the aorta close to the coeliac artery for close infusion of substances. Intraarterial GLP-1 infusions (3 3 10 210 and 3 3 10 29 moles / kg per 10 min) (n 5 8) induced inhibition of spontaneous motor activity in the antrum, duodenum and proximal jejunum. Inhibition induced by GLP-1 was reversed by i.v. infusion of GLP-1 receptor antagonist, Exendin (9-39) (3 3 10 28 moles / kg per 10 min) (n 5 6). Neither the presence of L-NNA (10 25 moles / kg) (n 5 9) nor the VIP receptor antagonist [4-Cl-D-Phe 6 , Leu 17 ]-VIP (3 3 10 28 moles / kg per 10 min) (n 5 8) modified responses to GLP-1. However, a combination of the adrenergic blockers phentolamine and propranolol (1 mg / kg each) (n 5 8) completely blocked motor actions of GLP-1 in all the organs studied. Moreover, inhibition of gastrointestinal motor activity by GLP-1 was blocked by previous infusion of hexamethonium (10 mg / kg) (n 5 4). This study demonstrates that GLP-1 inhibits gastrointestinal motor activity of the rat acting on specific GLP-1 receptors and via stimulation of adrenergic pathways. 1998 Elsevier Science B.V. Keywords: Intestinal motor patterns; Gastric emptying; Adrenoceptors; Exendin; Ileus; Intestinal inhibition
1. Introduction Glucagon-like peptide 1 (GLP-1) has recently attracted considerable interest because it acts as an incretin, making an important contribution to insulin secretion via the enteroinsular axis [1,2] and because it may be the most potent inhibitor of feeding yet identified in the rat [3]. Furthermore, GLP-1 can act as an enterogastrone, inhibiting gastric acid secretion in man [4,5] as well as decreasing gastric emptying in parallel with an inhibition of pancreatic secretion [6]. Mammalian proglucagon consists of 160 amino acid residues synthesized in the alpha cells of the islets of Langerhans, in the L cells of the ileum and colonic mucosa as well as in the brain [7,3]. The intestinal L cell produces two elongated forms of glucagon, glicentin and oxy*Corresponding author. Tel.: 1 34 3 581 1848; fax: 1 34 3 581 2006; e-mail: [email protected] 0167-0115 / 98 / $19.00 1998 Elsevier Science B.V. All rights reserved. PII S0167-0115( 98 )00010-X
ntomodulin, from the amino-terminal region, and another biologically active peptide, GLP-1 (proglucagon 78-107amide), from the carboxyl-terminal portion of the proglucagon molecule. Furthermore, the mammalian glucagon amino acid sequence is highly homologous to the anglerfish glucagon and is identical in various species of mammals such as the rat, hamster, guinea pig, bovine and human. This homology indicates evolutionary pressure to preserve this sequence and is consistent with a specific biological role [8]. The most important circulating intestinal glucagon gene product appears to be GLP-1 [9]. Luminal glucose seems to be the main GLP-1-releaser of meal constituents. The levels of this peptide peaked and reached statistical significance compared with basal levels after 30 min of ingestion of carbohydrate [1]. Additionally, cholinergic agonists, peptides of the bombesin family, CGRP and GIP induce GLP-1 secretion [10]. Although inhibitory actions of glucagon have been well
20
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
known for a long time [11], to our knowledge there have been no studies on the mechanism of action of glucagon peptides in the small intestine. However, recently exogenous GLP-1 has been shown to markedly produce an inhibition of migrating motor complexes (MMC) [12] and to inhibit contractile activity in the antro-pyloro-duodenal segment [13]. The purpose of this study was to further investigate the effects and the mechanism of action of GLP-1 (7-37) (proglucagon 78-108) on gastrointestinal motility in rats.
2. Methods
2.1. Animal preparation Male Sprague-Dawley rats, 8 to 10 weeks old and weighing 300–350 g, fasted for 12–16 h, were anaesthetized by inhalation of halothane to allow cannulation of the right jugular vein with a polyethylene tubing. Level III of anaesthesia was induced with i.v. infusion of thiopental sodium (20 mg / kg) and maintained with bolus infusions of anaesthetic as required. Body temperature was kept constant by placing the animals on a heating pad maintained at 378C. Rats were tracheotomized to facilitate spontaneous breathing. A polyethylene tube was inserted through the carotid artery to the aorta and placed at the bifurcation of coeliac artery for close infusion of drugs. The intestine was exposed through a midline abdominal incision. Three 3 3 5-mm strain-gauges (Hugo Sachs Elektronik, Germany) were sutured to the antrum, the duodenum (1.5 cm from the pylorus) and the jejunum (5 cm from Treitz’s ligament) oriented to record circular muscle activity. Strain-gauges were connected to high-gain amplifiers (Lectromed MTP8) and amplified signals were recorded in a polygraph.
2.2. Drugs Drugs used in this study were: Glucagon-like peptide 1 (7-37) (GLP-1) (Peptide Institute, Osaka, Japan), GLP-1 antagonist Exendin (9-39) (Bachem, Cardiff, UK), VIP receptor antagonist [4-Cl-D-Phe 6 , Leu 17 ]-VIP (Peninsula Laboratories, Belmont, CA), propranolol hydrochloride and phentolamine mesylate (Research Biochemical, Natick, MA), hexamethonium, L-nitroarginine ( L-NNA), yohimbine and carbachol (Sigma Chemical, St Louis, MO).
min before the start of GLP-1 infusion until the end of peptide infusion (n 5 6); (b) a bolus of hexamethonium (10 mg / kg, i.v.) was given. Then, in order to maintain mechanical activity after hexamethonium, carbachol (1.5 3 10 28 moles / kg) was continuously infused during GLP-1 infusion (n 5 4); (c) VIP receptor antagonist [4-Cl-D-Phe 6 , Leu 17 ]-VIP (3 3 10 28 moles / kg per 10 min, i.a.) was infused 2 min before the start and during GLP-1 infusion (n 5 8); (d) L-NNA (10 25 moles / kg, i.a.) was infused 20 min before the beginning of GLP-1 infusion (n 5 9); (e) a bolus of a combination of phentolamine and propranolol (1 mg / kg, each, i.v.) was infused 10 min before GLP-1 infusion (n 5 8); (f) a bolus of yohimbine (0.3 mg / kg, i.v.) was infused 10 min before GLP-1 infusion (n 5 5); (g) propranolol (1 mg / kg, i.v.) was infused 10 min before GLP-1 (n 5 4).
2.4. Analysis of data Mechanical activity was evaluated by measuring the area under the curve (AUC) of the mechanical activity tracing for a period of 10 min before and during peptide infusion. Data are expressed as mean6SEM. Statistical analysis was performed by means of either ANOVA, nonparametric repeated measure test or paired nonparametric test when applicable. Post-tests were Bonferroni and Dunnett’s test accordingly. Differences were considered significant when p , 0.05.
3. Results
3.1. Effects of GLP-1 on gastrointestinal mechanical activity During the preinfusion period, all the organs studied i.e. antrum, duodenum and jejunum, exhibited spontaneous activity consisting of contractions with an approximate frequency of 4 contractions every 10 min which propagated from the antrum to the jejunum (Fig. 1). Intraarterial infusion of GLP-1 at 3 3 10 210 moles / kg or higher, induced inhibition of the spontaneous motor activity in the antrum, duodenum and proximal jejunum (Fig. 1a). Smaller doses of GLP-1 did not modify spontaneous mechanical activity.
2.3. Experimental procedures
3.2. Mechanism of action of GLP-1 on gastrointestinal mechanical activity
GLP-1 (3 3 10 210 moles / kg per 10 min, i.a.) was infused for 10 min at a rate of 100 ml / min (n 5 8) in order to evaluate GLP-1 actions on studied organs. To study the mechanism of action of GLP-1, the following protocols were followed: (a) GLP-1 receptor antagonist, Exendin (9-39) (3 3 10 28 moles / kg per 10 min, i.v.) was infused 2
Intravenous infusion of GLP-1 receptor antagonist, Exendin (9-39) completely blocked inhibitory responses to GLP-1 (Fig. 1b). Hexamethonium completely blocked spontaneous activity (Fig. 2). In consequence, carbachol was infused continuously to maintain phasic activity. In these conditions, GLP-1 infusion at a dose of 3 3 10 210
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
21
Fig. 1. Mechanical activity of the antrum (A), duodenum (D) and jejunum (J). (a) Effect of GLP-1 (3 3 10 210 moles / kg per 10 min) close i.a. infusion. (b) Lack of effect of GLP-1 in the presence of antagonist exendin (3 3 10 28 moles / kg per 10 min).
Fig. 2. Mechanical activity of the antrum, duodenum and jejunum. Effects of hexamethonium, carbachol and the lack of effect of GLP-1. Abbreviations as in Fig. 1.
moles / kg did not modify phasic activity in any of the studied organs (Fig. 2). Neither L-NNA nor VIP antagonist [4-Cl-D-Phe 6 , Leu 17 ]-VIP modified GLP-1 response. Whereas VIP antagonist did not modify basal motility, L-NNA significantly increased motor activity in the jejunum, but not in the
antrum or duodenum (Fig. 3). Finally, previous infusion of the adrenergic blockers phentolamine and propranolol did not modify basal activity but completely blocked the inhibitory response to GLP-1 in all the studied organs (Fig. 4). The next step was to study the type(s) of adrenoceptor
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
22
Fig. 3. Lack of effect of L-NNA (a) and VIP antagonist [4-Cl-D-Phe 6 , Leu 17 ]-VIP (b) on GLP-1 response. *, Significant differences between basal and GLP-1 at each organ.
involved in GLP-1 response. However, this task became rather difficult. Infusion of the a 2 -adrenoceptor antagonist yohimbine in 3 out of 5 experiments completely blocked the spontaneous mechanical activity in the duodenum and jejunum, while in the antrum it induced a different pattern characterized by regular mechanical contractions in a 3–5 cycles per min frequency. In these experiments it was not possible to test the GLP-1 effects. In the other two experiments yohimbine did not block GLP-1-induced inhibition. Finally, to study the role of b-adrenergic receptors in GLP-1 response, a bolus of propranolol (1 mg / kg, i.v.) was infused (n 5 4). Propranolol did not modify basal activity. In the duodenum and jejunum, GLP-1 infusion still induced an inhibition of mechanical activity in 3 out of 4 cases, while in the antrum, propranolol gave contradictory results. Response to GLP-1 was blocked in two out of 4 cases (Fig. 5) and remained unchanged in the other two.
4. Discussion This study demonstrates that GLP-1 inhibits sponta-
neous activity of the proximal gastrointestinal tract of the rat by acting on specific GLP-1 receptors. In the mechanism of action, sympathetic pathways are involved. Glycine-extended GLP-1 (7-37) rather than carboxyamidated form corresponding to GLP-1 (7-36 amide) was used. It is known that for some amidated peptides, amide is crucial for biological activity. But this is not the case with GLP-1 molecules [7] because both GLP-1 (7-36) amide and GLP-1 (7-37) stimulate insulin secretion with the same potency [14]. Both peptide forms have been suggested as a possible new treatment for noninsulin-dependent Diabetes mellitus [15]. Glucagon-like peptide-1 induced an inhibitory action in all the studied areas. An inhibition of motor activity by other glucagon family peptides had already been reported [11]. However, the fact that the inhibitory action observed in our study was specifically blocked by the GLP-1 receptor antagonist exendin, indicates a specific action on GLP-1 receptors. This specificity together with the presence of GLP-1-containing cells in the distal areas of the intestine [16] and the increase of GLP-1 levels postprandially [1], suggests a role of this peptide in the control of gastrointestinal motor activity. Our results also agree with the inhibitory function reported for this peptide in the gastroduodenal area in humans [13]. With the specificity of GLP-1 actions established, our focus was to study the mechanism of action of GLP-1. We started by studying the involvement of either NO or VIP on GLP-1 actions. These two neurotransmitters seem to be responsible for the tonic inhibition of the intestine in the rat [17,18] and both have been suggested as responsible for NANC transmission [19,20]. However, neither the blockade of NO synthesis nor the VIP antagonist infusion modified inhibitory response to GLP-1. On the contrary, the simultaneous infusion of phentolamine and propranolol completely blocked response to GLP-1, demonstrating an involvement of adrenergic pathways in GLP-1 actions. To our knowledge this is the first evidence of an intestinal peptide modulating gastrointestinal motility through adrenergic pathways. In agreement with our results, a b-adrenergic mechanism has been reported in GIP- and GLP-1induced insulin secretion [21]. Furthermore, a nonspecific effect of adrenergic antagonists can be discarded since they did not modify either basal activity or the inhibitory response to infused VIP (data not shown). Stimulation of adrenoceptors in the gastrointestinal tract induces an inhibition of all the intestinal patterns including phase III of the MMC where acetylcholine is involved [22,23]. The main mechanism of action is the presynaptic inhibition of cholinergic neurons [24,25]. Furthermore, it is well established that cholinergic neurons possess a-adrenergic receptors that inhibit release of acetylcholine [26]. However, several authors have also demonstrated that a 2 receptors inhibit NO synthesis in the canine ileocolonic junction [27–29]. That also seems to be the case in the rat gastrointestinal tract where yohimbine causes inhibition of
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
23
Fig. 4. Effect of adrenergic blockers, phentolamine and propranolol, on GLP-1 response. (a) Mechanical activity; (b) summary of effects. Abbreviations as in Fig. 1. *, Significant differences between basal and GLP-1; 1 , significant differences between GLP-1 in the presence and absence of antagonists.
Fig. 5. Tracing of one experiment showing the effect of propranolol on GLP-1 response. Abbreviations as in Fig. 1.
24
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
spontaneous motor activity. This inhibition seems to be a consequence of an NO increase because it was blocked by L-NNA (data not shown). Spontaneous motility in our model was also blocked either by hexamethonium or atropine, a similar response to that observed in GLP-1. The lack of effect of GLP-1 in the presence of hexamethonium suggests that GLP-1 action is due to presynaptic inhibition of cholinergic neurons. Determining type(s) of receptor involved in the response turned out to be difficult. The reason being that neither antagonist given alone was able to consistently block the effect of GLP-1. This indicates that both alfa and beta receptor stimulation occurs after GLP-1 infusion. It is also of interest that in some of the experiments performed with propranolol, there was a blockade of GLP-1 response indicating the variability of the response as well as the involvement of beta receptors. Although a beta adrenoceptor with atypical characteristics has been described in the rat intestine [30], our results could also indicate that GLP-1 could induce a long reflex involving a large number of sympathetic pathways or, as has been suggested in humans, a central action of the peptide [31]. The fact that GLP-1 cells are located in the mucosa of distal parts of the intestine could imply that the GLP-1 release might affect sympathetic afferents. In this sense, colonic sympathetic afferents controlling the gastroduodenal area have been described [32]. In consequence, a GLP-1 action on these fibres which could activate an ileal brake mechanism is also likely. It has also been reported that GLP-1 induces insulin release [33]. Although insulin-induced changes to motility have been described, in the rat these actions are excitatory [34] and not inhibitory as we always saw with GLP-1. Therefore these peptide actions seem to be direct and nonmediated by insulin release. In conclusion, GLP-1 induces inhibition of gastrointestinal motor activity. Stimulation of adrenergic pathways mediates this inhibition. The mechanism of action is independent of both NO and VIP, suggesting that there is no inhibitory motoneuron stimulation. Although a central action of this peptide has been reported to mediate satiety [33], inhibitory motor actions of GLP-1 might also contribute to a satiety feeling. In the same way, GLP-1-induced motor inhibition could indirectly decrease plasma glucose concentration by slowing glucose absorption.
Acknowledgements ´ This work has been financially supported by Direccion ´ en Ciencia y Tecnologıa ´ General de Investigacion (DGICYT PB92-0638 and PM95-0121. The authors are grateful to Mr Antonio Acosta for his skilful assistance in the care of animals. They are also in debt to Prof. E.E. Daniel for his encouragement on the study of GLP-1 and
to Mr. A.C. Hudson for the editorial revision of the manuscript.
References [1] Elliott RM, Morgan LM, Tredger JA, Deacon S, Wright J, Marks V. 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 1993;138:159–66. [2] Parker JC, Hargrove DM, Shepherd KL, Nardone NA, Andrews KM, Persson LM. Glucagon-like peptide-1 (7-37) and glibenclamide stimulate insulin secretion and biosynthesis by different glucosedependent mechanisms: comparison in vivo and in vitro. Digestion 1993;54:357–8. [3] Turton MD, O’Shea D, Gunn I, Beak SA, Edwards CMB, Meeran K, Choi SJ, Taylor GM, Heath MM, Lambert PD, Wilding JPH, Smith DM, Ghatei MA, Herbert J, Bloom SR. A role for glucagonlike peptide-1 in the central regulation of feeding. Nature 1996;379:69–72. [4] Schjoldager BTG, Mortensen PE, Christiansen J, Orskov C, Holst JJ. GLP-1 (glucagon-like peptide 1) and truncated GLP-1, fragments of human proglucagon, inhibit gastric acid secretion in humans. Digest Dis Sci 1989;34:703–8. [5] O’Halloran SJ, Nikou GC, Kreymann B, Ghatei MA, Bloom SR. Glucagon-like peptide-1 (7-36)-NH 2 : a physiological inhibitor of gastric acid secretion in man. J Endocrinol 1990;126:169–73. [6] Wettergren A, Schjoldager B, Mortensen PE, Myhre J, Christiansen J, Holst JJ. Truncated GLP-1 (proglucagon 78-107-amide) inhibits gastric and pancreatic functions in man. Digest Dis Sci 1993;38:665–73. [7] Kreymann B, Yiangou Y, Kanse S, Williams G, Ghatei MA, Bloom SR. Isolation and charaterization of GLP-1 7-36 amide from rat intestine. FEBS Lett 1988;242:167–70. [8] Lund PK, Goodman RH, Montminy MR, Dee PC, Habener JF. Anglerfish islet preproglucagon II. Nucleotide and corresponding amino acid sequence of the cDNA. J Biol Chem 1983;258:3280–4. [9] Orskov C, Bersani M, Johnsen AH, Hojrup P, Holst JJ. Complete sequences of glucagon-like peptide-1 from human and pig small intestine. J Biol Chem 1989;264:12826–9. ¨ A, Goke ¨ B. Regulation of glucagon-like peptide [10] Herrmann C, Voge release from the isolated perfused rat ileum by nutrients, peptides and neuromediators. Digestion 1993;54:367. [11] Whalen GE. Glucagon and the small gut. Gastroenterology 1974;67:1284–6. ¨ PM. Direct inhibitory [12] Tolessa T, Gutniak M, Efendic S, Hellstrom action of glucagon-like peptide-1 on migrating myoelectric complex independently of insulin or somatostatin release. Neurogastroent Motil 1996;8:194 (Abstract). ¨ [13] Schirra J, Wank U, Houck P, Arnold P, Goke B, Katschinski M. Effects of GLP-1 on human antro-pyloro-duodenal motility. Neurogastroent Motil 1996;8:190 (Abstract). [14] Orskov C, Wettergren A, Holst JJ. Biological effects and metabolic rates of glucagon-like peptide-1 7-36 amide and glucagon-like peptide 1 7-37 in healthy subjects are indistinguishable. Diabetes 1993;42:658–61. [15] Habener JF, Mcmanus K, Nathan DM. Insulinotropic actions of glucagon-like peptide-1 (7-37) in diabetic and nondiabetic subjects. Digestion 1993;54:376–7. ¨ R, Willemer S, Harthus H-P, Vermeer H, Arnold [16] Eisselein R, Goke ¨ B. Glucagon-like peptide-1 cells in the gastrointestinal tract R, Goke and pancreas of rat, pig and man. Eur J Clin Invest 1992;22:283–91. ´ ´ ´ ˜ [17] Rodrıguez-Membrilla A, Martınez V, Jimenez M, Gonalons E, Vergara P. Is the nitric oxide the final mediator regulating the
M. Giralt, P. Vergara / Regulatory Peptides 74 (1998) 19 – 25
[18]
[19]
[20]
[21]
[22] [23]
[24]
[25]
[26]
migrating myoelectric complex cycle?. Am J Physiol 1995;268:G207–14. ¨ PM, Ljung T. Nitrergic inhibition of migrating myoelecHellstrom tric complex in the rat is mediated by vasoactive intestinal peptide. Neurogastroenterol Mot 1996;8:299–306. Grider JR, Cable MB, Said SI, Makhlouf GM. Vasoactive intestinal peptide as a neural mediator of gastric relaxation. Am J Physiol 1985;248:G73–8. Daniel EE, Haugh C, Woshowska Z, Cipris S, Jury J, Fox-Threlheld JET. Role of nitric oxide-related inhibition in intestinal function: relation to vasoactive intestinal polypeptide. Am J Physiol 1994;266:G31–9. Ikeda T, Ochi H, Ohtani I, Fujiyama K, Hoshino T, Tanaka Y, Takeuchi T, Mashiba H. Possible role of the adrenergic mechanism in gastric inhibitory polypeptide- and glucagon-like peptide 1 (7-36) amide-induced insulin release in the rat. Metabolism 1993;42:209– 13. Kosterlitz HW, Lees GM. Pharmacological analysis of intrinsic intestinal reflexes. Pharmacol Rev 1964;16:301–39. Code CF, Marlett JA. The interdigestive myoelectric complex of the stomach and small bowel of dogs. J Physiol London 1975;246:289– 309. Jansson G, Martinson J. Studies on the ganglionic site of action of sympathetic outflow to the stomach. Acta Physiol Scand 1966;68:184–92. Sakai Y, Daniel EE, Jury J, Fox JET. Neurotensin inhibition of canine intestinal motility in vivo via a-adrenoceptors. Can J Physiol Pharmacol 1984;62:403–11. Wikberg J. Localization of adrenergic receptors in guinea pig ileum
[27]
[28]
[29]
[30]
[31] [32]
[33]
[34]
25
and rabbit jejunum to cholinergic neurons and to smooth muscle cells. Acta Physiol Scand 1977;99:190–207. Boeckxstaens GE, De man JG, Pelckmans PA, Herman AG, Van Maercke YM. a 2 -adrenoceptor-mediated modulaton of the nitrergic innervation of the canine isolated ileocolonic junction. J Pharmacol 1993;109:1079–84. De man JG, Boeckxstaens GE, Herman AG, Pelckmans PA. Effect of potassium channel blockade and a 2 -adrenoceptor activation on the release if nitric oxide from nonadrenergic noncholinergic nerves. Br J Pharmacol 1994;112:341–5. De man JG, Boeckxstaens GE, Herman AG, Pelckmans PA. Effect of potassium channel blockade and a 2 -adrenoceptor activation on the release if nitric oxide from nonadrenergic noncholinergic nerves. Can J Physiol Pharmacol 1984;62:403–11. Vliet A, Rademaker B, Bast A. A beta adrenoceptor with atypical characteristics is involved in the relaxation of the rat small intestine. J Pharmacol Exp Ther 1990;255:218–26. Holst JJ. Glucagon-like peptide 1: a newly discovered gastrointestinal hormone. Gastroenterology 1994;107:1848–55. Messenger JP, Furness JB. Calbindin-immunoreactive nerve terminals in the guinea pig celiac ganglion originate from colonic nerve cells. J Auton Nerv Syst 1991;35:133–42. Holst JJ, Orskov C, Vagn nielsen O, Schwartz TW. Truncated glucagon-like peptide is, an insulin-releasing hormone from the distal gut. FEBS Lett 1987;211:169–74. Allescher HD, Willis S, Schusdziarra V, Classen M. Modulatory effect of insulin on rat small intestinal motility and peptide release in vitro. Digestion 1991;48:192–201.