sign in sign up
<<
Home , Pancreatic islets

PATHOPHYSIOLOGY

GLP-1 Effects on Islets: Hormonal, Neuronal, or Paracrine?

1 MARC Y. DONATH, MD control of glycemia through the activation 2 RÉMY BURCELIN, PHD of the gut brain axis. Furthermore, the di- rect administration of the DPP-4 inhibitor into the portal significantly in- creased portal (but not peripheral) GLP-1 ccording to the classical con- (5) within minutes from the absorption and levels and decreased Acept, -like (GLP)-1 of glucose and lipids. The final aim of concentrations (14). However, despite the is viewed as a produced this axis is to anticipate the breakthrough large amount of experimental evidence de- in the intestinal L cells and acting via of the nutrients into the blood and their scribed above showing the important role the circulation on satiety in the brain, better handling. Indeed, GLP-1 secreted of GLP-1 on the gut-to-brain axis, a recent gut motility, and insulin and glucagon from L cells can influence brain neuronal observation in mice suggests that the circu- secretion in the pancreatic islet. How- activities via an alternative neural path- lating GLP-1 could also directly access the ever, in contrast to typical , way initiated by sensors in the hepatic brain and the b-cells and induce insulin plasma levels of GLP-1 are relatively low portal region (6–8). Thereby, the vagus secretion (15). Transgenic mice that ex- with a very short half-life. Furthermore, transmits the metabolic infor- pressed the GLP-1 receptor in islets GLP-1 is rapidly inactivated by dipep- mation to the nucleus tractus solitarii and in pancreatic ductal cells within the tidyl peptidase-4 (DPP-4) in the vicinity in the brain stem, which relays the glu- background of the GLP-1 receptor knock- of L cells within ,1 min from the secre- cose signal to hypothalamic nuclei (9). out mice were characterized by increased tion of the gut peptide (1,2). This rapid This process is called the gut-to-brain- glucose-induced insulin secretion that was metabolism of GLP-1 raises questions to-periphery axis. Seminal studies from sufficient to normalize glucose tolerance, about how its effects are mediated on our group showed that the direct infusion whereas no effect on food intake, hindbrain target organs such as pancreatic b-cells. of glucose into the portal vein of mice at a c-fos expression, or gastric emptying was In this review, we will discuss possible low rate increased muscle glucose utiliza- observed (15). This new set of data suggests alternative pathways for the incretin ef- tion through an insulin-independent that part of the gut-released GLP-1 may re- fect on pancreatic islet. These involve mechanism (10–12). This process re- semble some of the incretin effect through a Lcell–derived GLP-1 via neuronal acti- quired the activation of the hepatoportal process not involving the gut-brain axis. vation and a-cell–derived GLP-1 via vein glucose sensor (6). The blockage of Alternatively or additionally, the release of auto/paracrine effects. Of note, GLP-1 is the portal GLP-1 receptor by exendin 9 GLP-1 from an intraislet processing may not acting alone, and its effect can be directly into the hepatoportal vein or in contribute to triggering glucose-induced modulated by other factors including GLP-1 receptor knockout mice (8) pre- insulin secretion (see below). GIP.Sincenodataareavailableyetin vented the portal glucose sensor activa- A further demonstration of the role the present context, we have limited this tion for the control of muscle glucose played by the GLP-1–dependent gut- review to GLP-1. utilization (8) or insulin secretion (8,13). brain axis is the recent analysis of the ther- Furthermore, the inhibition of the enteric apeutic role of GLP-1 receptor agonists on GLP-1 in the gut-to-brain-to- DPP-4 by small doses of DPP-4 inhibitor neuropathy in mice with diabetes owing periphery axis for the control of improved glucose tolerance without in- to (16). The authors glucose metabolism creasing the blood concentration of GLP-1 – showed the presence of the GLP-1 recep- Recent rodent data show that GLP-1 can through a GLP-1 receptor dependent tor on the lumbar dorsal root ganglion by induce its metabolic actions by interact- manner (3). In such conditions, the vagus immunohistochemical analyses and ing with its receptors in extrapancreatic nerve activity was increased in response to further demonstrated that exendin-4 in- locations such as the gut to activate the oral DPP-4 inhibitors, whereas the intrave- creases the neurite outgrowth. Importantly, submucosal and the myenteric nervous nous administration of the drug had no the delayed current perception threshold plexi (3,4) and the brain, which then therapeutic effect, further suggesting the and motor and sensory nerve conduction transmit the signal to peripheral tissues important role of enteric GLP-1 on the velocity impaired by , was ccccccccccccccccccccccccccccccccccccccccccccccccc improved by the GLP-1 agonist (16). From the 1Clinic of , Diabetes & Metabolism, University Hospital Basel, Basel, Switzerland; and Hence, gut-released hormone would fur- 2INSERM U1048, Institut des Maladies Métaboliques et Cardiovasculaires, Toulouse, France. ther favor the gut-brain axis by control- Corresponding author: Marc Y. Donath, [email protected]. ling the enteric neural development. This publication is based on the presentations from the 4th World Congress on Controversies to Consensus in Diabetes, Obesity and Hypertension (CODHy). The Congress and the publication of this supplement were Similarly, a therapeutic role of the gut- made possible in part by unrestricted educational grants from Abbott, AstraZeneca, Boehringer Ingelheim, brain axis has been proposed regarding Bristol-Myers Squibb, Eli Lilly, Ethicon Endo-Surgery, Janssen, Medtronic, Novo Nordisk, Sanofi, and the therapeutic efficacy of gastric bypass. Takeda. Obese and diabetic patients who under- DOI: 10.2337/dcS13-2015 © 2013 by the American Diabetes Association. Readers may use this article as long as the work is properly went this type of bariatric surgery lose cited, the use is educational and not for profit, and the work is not altered. See http://creativecommons.org/ weight within months and reverse their licenses/by-nc-nd/3.0/ for details. diabetes status within weeks from the care.diabetesjournals.org DIABETES CARE, VOLUME 36, SUPPLEMENT 2, AUGUST 2013 S145 GLP-1 effects on islets surgery. It has been proposed that a hor- The next step was to uncover the and obesity are characterized by an acti- monal characteristic is that these patients molecular trigger of PC1/3 with subse- vation of the innate secrete large amounts of GLP-1 (17) that quent GLP-1 production in a-cells. Two reflected by an increase in various in- could even lead to uncontrolled insulin important studies point to a direct effect flammatory markers including elevated secretion and hypoglycemic episodes of elevated glucose concentrations. Thus, plasma levels of IL-6 (33). It was shown in (18), although this still needs to be con- in the Psammomy obesus, high-energy diet mice that elevated IL-6 levels in response firmed. However, numerous other pepti- induces diabetes along with a significant to exercise (34), as well as acute and des such as peptide YY, oxyntomodulin, increase of GLP-1 in the portal vein and chronic IL-6 administration in mice, can and GLP-2 may be responsible for the ac- extract (31). Isolated islets from stimulate GLP-1 secretion from intestinal tivation of the gut-brain axis in patients hyperglycemic animals released more L cells and pancreatic a-cells leading to with bypass surgery (19). GLP-1 than in controls. This ex vivo effect improved glucose (35). In vi- Altogether, the role of GLP-1 on the was confirmed in vitro by culture of nor- tro, IL-6 increased GLP-1 synthesis and gut-to-brain-to-periphery axis is now mal P. obesus islets with high glucose, secretion from entero-endocrine L cells considered a major mechanism of action which induced GLP-1 release. Similarly, both acutely and chronically. IL-6 also in- of the gut hormone for the control of high glucose concentrations increased creased GLP-1 synthesis and secretion glycemia. Many controversial debates are PC1/3 in rat islets, while in an a-cell line from human pancreatic a-cells in asso- ongoing regarding its role in the control of glucose increased GLP-1 and decreased ciation with increased proglucagon food intake. Similarly, the production of glucagon secretion (32). and PC1/3 transcription. Similarly, the GLP-1 into the brain and the colocalized An additional trigger of GLP-1 is beneficial effects of elevated IL-6 were receptors have also been important fields interleukin (IL)-6. Indeed, showninanimalmodelsoftype2 of investigation in explaining the meta- bolic effect of this insulinotropic peptide. a-Cell as a source of GLP-1 GLP-1 is spliced from its precursor pro- glucagon in intestinal L cells through processing by the enzyme PC1/3 (20– 23). In the pancreatic a-cell, proglucagon is processed by PC2 to yield glucagon (24,25). Adult a-cells are thought to pro- duce little GLP-1. However, a switch of PC2 to PC1/3 is sufficient to convert the a-cell from a hyperglycemia-promoting cell to one that lowers blood glucose lev- els and promotes islet survival (26). Fur- thermore, several animal studies show that islet-derived GLP-1 may be stimu- lated by increased demand of insulin se- cretion, suggesting that locally produced GLP-1 may play a role in the long-term b-cell adaptation. Indeed, b-cell destruc- tion with streptozotocin leads to hyper- glycemia followed by b-cell regeneration associated with a-cell hyperplasia and pancreatic PC1/3 upregulation along with increased pancreatic and circulating levels of GLP-1 and GLP-1–to–glucagon ratio (27,28). Interestingly, blocking GLP-1 reduced b-cell regeneration. Simi- larly, a-cell hyperplasia along with PC1/3 upregulation and GLP-1 production was observed in mouse models of insulin re- sistance including pregnant, ob/ob, db/db, and prediabetic NOD mice (29). When considered together, these studies strongly suggest that a-cell hyperplasia and release of GLP-1 play an integral role in b-cell regeneration. In line with this concept, a-cell hyperplasia occurs early in response to high-fat diet and pre- ceding b-cell expansion and is required for b-cell adaptation (30). Figure 1dHypothetical model for the acute and chronic effects of GLP-1 on pancreatic islets.

S146 DIABETES CARE, VOLUME 36, SUPPLEMENT 2, AUGUST 2013 care.diabetesjournals.org Donath and Burcelin diabetes and obesity. Furthermore, IL-6 Summary and proposed hypothesis of the gastrointestinal neutralization deteriorated glycemic In summary, we propose the following tract and the dorsal vagal complex of the control and reduced pancreatic GLP-1 working hypothesis for the GLP-1 effect hindbrain in the rat by a GLP-1 receptor. – content. Hence, IL-6 mediates cross- on the pancreatic islet (Fig. 1). The acute Brain Res 2010;1344:124 133 talk between insulin-sensitive tissues, postprandial-mediated incretin effect 5. Cabou C, Burcelin R. GLP-1, the gut- brain, and brain-periphery axes. Rev Di- L cells, and pancreatic islets to adapt to would be mediated mainly by GLP-1 re- abet Stud 2011;8:418–431 changes in insulin demand by increas- leased from L cells acting locally and via 6. Nakagawa A, Satake H, Nakabayashi H, ing L-cell GLP-1 secretion and reprog- portal release into the on neuronal et al. Receptor gene expression of glucagon- ramming a-cells to process proglucagon activation. Chronic trophic effects of like peptide-1, but not glucose-dependent to GLP-1. GLP-1 on b-cells promoting survival insulinotropic polypeptide, in rat nodose The question remains of the impor- and insulin production would be medi- ganglion cells. Auton Neurosci 2004;110: tance of a-cell–derived compared with ated via reprograming of the a-cell by 36-43 systemic GLP-1. An argument against a increasing PC1/3. This would switch pro- 7. Baraboi ED, St-Pierre DH, Shooner J, significant contribution by the a-cell is glucagon processing from glucagon to Timofeeva E, Richard D. Brain activation the islet architecture. Indeed, in most ro- GLP-1 allowing for auto/paracrine effects. following peripheral administration of the GLP-1 receptor agonist exendin-4. Am J dents, b-cells compose the core of the is- The trigger for this long-term adaptation a Physiol Regul Integr Comp Physiol 2011; lets while the -cells form the mantle could be IL-6 released by insulin-resistant 301:R1011–R1024 region. Furthermore, it is believed that fat tissues by contracting muscles or by 8. Burcelin R, Da Costa A, Drucker D, the blood flow goes from the center to as well as hyperglycemia. Ac- Thorens B. Glucose competence of the the periphery, making it difficult to imag- cordingly, the main action of DPP-4 inhi- hepatoportal vein sensor requires the ine that GLP-1 released from a-cells could bition may occur at the tissue rather than presence of an activated glucagon-like act on b-cells. However, in animal models at the plasma level. peptide-1 receptor. Diabetes 2001;50:1720– of diabetes in which insulin secretion is We do realize that several aspects of 1728 decreased, normal organization of islet this hypothesis are controversial or not 9. Shimizu N, Oomura Y, Novin D, Grijalva cells was found to be perturbed so that sufficiently supported by experimental CV, Cooper PH. Functional correlations b-cells were intermingled with non- data. In particular, the relative contribu- between lateral hypothalamic glucose- b sensitive neurons and hepatic portal -cells (36). Moreover, a study describing tion of circulating GLP-1 versus neuronal glucose-sensitive units in rat. Brain Res the architecture in human islets reported and auto/paracrine remains to be clari- 1983;265:49–54 direct intercellular contacts between fied. Finally, most of the findings de- 10. Burcelin R, Dolci W, Thorens B. Portal a-andb-cells, supporting the notion scribed were obtained using rodents, glucose infusion in the mouse induces hy- that a-cell products can act in a paracrine isolated human islets, or animal models; poglycemia: evidence that the hepatoportal manner to regulate the b-cell (37). In sup- therefore, the in vivo relevance for hu- glucose sensor stimulates glucose utiliza- port of this, acetylcholine secreted by mans remains to be demonstrated. tion. Diabetes 2000;49:1635–1642 a-cells acts in a paracrine manner to 11. Burcelin R, Dolci W, Thorens B. Glucose prime the b-cell to respond optimally to sensing by the hepatoportal sensor is subsequent increases in glucose (38). d fl GLUT2-dependent: in vivo analysis in Acknowledgments No potential con icts of GLUT2-null mice. Diabetes 2000;49:1643– Hence, release of GLP-1 directly into the interest relevant to this article were reported. b 1648 islets at the vicinity of the -cells might M.Y.D. and R.B. wrote the manuscript. M.Y.D. 12. Burcelin R, Crivelli V, Perrin C, et al. be a potent way to trigger cell survival and is the guarantor of this work and, as such, had GLUT4, AMP kinase, but not the insulin insulin production. Another issue is the full access to all the data in the study and takes receptor, are required for hepatoportal activity of the GLP-1 species released by responsibility for the integrity of the data and the glucose sensor-stimulated muscle glucose a-cells. Indeed, under normal conditions accuracy of the data analysis. utilization. J Clin Invest 2003;111:1555– a-cells do not process proglucagon to ac- 1562 tive GLP-1 (39). However, metabolic 13. Preitner F, Ibberson M, Franklin I, et al. a References Gluco- control insulin secretion stress appears to reprogram the -cells, 1. Hansen L, Deacon CF, Orskov C, Holst JJ. allowing processing of active GLP-1. In- at multiple levels as revealed in mice Glucagon-like peptide-1-(7-36)amide is lacking GLP-1 and GIP receptors. J Clin deed, islet-derived GLP-1 activity was transformed to glucagon-like peptide-1- Invest 2004;113:635–645 demonstrated in human islets using a bio- (9-36)amide by dipeptidyl peptidase IV in 14. Fujiwara K, Gotoh K, Chiba S, et al. In- assay and for mouse islet experiments us- the capillaries supplying the L cells of the traportal administration of DPP-IV in- ing GLP-1 receptor antagonists and porcine intestine. Endocrinology 1999; hibitor regulates insulin secretion and 2/2 Glp1r mice (35). Furthermore, GLP-1 140:5356–5363 food intake mediated by the hepatic vagal derived from human islets has been de- 2. Deacon CF. Circulation and degradation afferent nerve in . J Neurochem 2012; scribed ex vivo from patients with type 2 of GIP and GLP-1. Horm Metab Res 2004; 121:66–76 – diabetes (31,40). Importantly, the pres- 36:761 765 15. Lamont BJ, Li Y, Kwan E, Brown TJ, 3. Waget A, Cabou C, Masseboeuf M, et al. Gaisano H, Drucker DJ. Pancreatic GLP-1 ence of a functionally competent GLP-1 fi system was recently demonstrated (40). Physiological and pharmacological mecha- receptor activation is suf cient for in- nisms through which the DPP-4 in- Finally, in human islets, after IL-6 stimula- cretin control of glucose metabolism hibitor sitagliptin regulates glycemia in in mice. J Clin Invest 2012;122:388– tion, the GLP-1–to–glucagon molar ratio is – . mice. Endocrinology 2011;152:3018 402 1.5 (35); therefore, it can be postulated 3029 16. Himeno T, Kamiya H, Naruse K, et al. Ben- that this is a meaningful phenomenon. 4. Washington MC, Raboin SJ, Thompson eficial effects of exendin-4 on experimental Clearly, this hypothesis remains to be dem- W, Larsen CJ, Sayegh AI. Exenatide re- polyneuropathy in diabetic mice. Diabetes onstrated by direct evidences. duces food intake and activates the enteric 2011;60:2397–2406 care.diabetesjournals.org DIABETES CARE, VOLUME 36, SUPPLEMENT 2, AUGUST 2013 S147 GLP-1 effects on islets

17. Van der Schueren BJ, Homel P, Alam M, or glucagon-like peptide. J Biol Chem 32. Whalley NM, Pritchard LE, Smith DM, et al. Magnitude and variability of the 1995;270:26488–26496 White A. Processing of proglucagon to glucagon-like peptide-1 response in pa- 25. Furuta M, Zhou A, Webb G, et al. Severe GLP-1 in pancreatic a-cells:isthisa tients with type 2 diabetes up to 2 years defect in proglucagon processing in islet paracrine mechanism enabling GLP-1 to following gastric bypass surgery. Diabetes A-cells of prohormone convertase 2 null act on b-cells? J Endocrinol 2011;211:99– Care 2012;35:42–46 mice. J Biol Chem 2001;276:27197– 106 18. Myint KS, Greenfield JR, Farooqi IS, 27202 33. Donath MY, Shoelson SE. Type 2 diabetes Henning E, Holst JJ, Finer N. Prolonged 26. Wideman RD, Covey SD, Webb GC, as an inflammatory disease. Nat Rev Im- successful therapy for hyperinsulinaemic Drucker DJ, Kieffer TJ. A switch from munol 2011;11:98–107 hypoglycaemia after gastric bypass: the prohormone convertase (PC)-2 to PC1/3 34. Febbraio MA, Pedersen BK. Muscle- pathophysiological role of GLP1 and its re- expression in transplanted alpha-cells is derived interleukin-6: mechanisms for ac- sponse to a analogue. Eur J accompanied by differential processing of tivation and possible biological roles. – Endocrinol 2012;166:951–955 proglucagon and improved glucose ho- FASEB J 2002;16:1335 1347 19. Lee WJ, Chen CY, Chong K, Lee YC, Chen meostasis in mice. Diabetes 2007;56: 35. Ellingsgaard H, Hauselmann I, Schuler B, SC, Lee SD. Changes in postprandial gut 2744–2752 et al. Interleukin-6 enhances insulin secre- hormones after metabolic surgery: a com- 27. Nie Y, Nakashima M, Brubaker PL, et al. tion by increasing glucagon-like peptide-1 parison of gastric bypass and sleeve gastrec- Regulation of pancreatic PC1 and PC2 secretion from L cells and alpha cells. Nat Med 2011;17:1481–1489 tomy. Surg Obes Relat Dis 2011;7:683–690 associated with increased glucagon-like 36. Gannon M, Ray MK, Van Zee K, Rausa F, 20. Lund PK, Goodman RH, Dee PC, Habener peptide 1 in diabetic rats. J Clin Invest Costa RH, Wright CV. Persistent expres- JF. Pancreatic preproglucagon cDNA 2000;105:955–965 sion of HNF6 in islet endocrine cells contains two glucagon-related coding se- 28. Thyssen S, Arany E, Hill DJ. Ontogeny of causes disrupted islet architecture and regeneration of beta-cells in the neonatal quences arranged in tandem. Proc Natl loss of function. Development Acad Sci USA 1982;79:345–349 rat after treatment with streptozotocin. – – 2000;127:2883 2895 21. Bell GI, Santerre RF, Mullenbach GT. Endocrinology 2006;147:2346 2356 37. Bosco D, Armanet M, Morel P, et al. Hamster preproglucagon contains the se- 29. Kilimnik G, Kim A, Steiner DF, Friedman Unique arrangement of alpha- and beta- quence of glucagon and two related pep- TC, Hara M. Intraislet production of cells in human islets of Langerhans. Di- – tides. Nature 1983;302:716 718 GLP-1 by activation of prohormone con- abetes 2010;59:1202–1210 a 22. Rouillé Y, Bianchi M, Irminger JC, Halban vertase 1/3 in pancreatic -cells in mouse 38. Rodriguez-Diaz R, Dando R, Jacques-Silva b PA. Role of the prohormone convertase models of -cell regeneration. Islets 2010; MC, et al. Alpha cells secrete acetylcholine – PC2 in the processing of proglucagon to 2:149 155 as a non-neuronal paracrine signal prim- glucagon. FEBS Lett 1997;413:119–123 30. Ellingsgaard H, Ehses JA, Hammar EB, ing beta cell function in . Nat Med 23. Zhu X, Zhou A, Dey A, et al. Disruption of et al. Interleukin-6 regulates pancreatic 2011;17:888–892 PC1/3 expression in mice causes dwarfism alpha-cell mass expansion. Proc Natl Acad 39. Holst JJ, Bersani M, Johnsen AH, Kofod H, and multiple neuroendocrine peptide pro- Sci USA 2008;105:13163–13168 Hartmann B, Orskov C. Proglucagon pro- cessing defects. Proc Natl Acad Sci USA 31. Hansen AM, Bödvarsdottir TB, Nordestgaard cessing in porcine and human pancreas. 2002;99:10293–10298 DN, et al. Upregulation of glucagon- J Biol Chem 1994;269:18827–18833 24. Rouillé Y, Martin S, Steiner DF. Differen- like peptide 1 (GLP-1) in Psammomys 40. Marchetti P, Lupi R, Bugliani M, et al. A tial processing of proglucagon by the obesusdan adaptive response to hyper- local glucagon-like peptide 1 (GLP-1) subtilisin-like prohormone convertases glycaemia? Diabetologia 2011;54:1379– system in human pancreatic islets. Dia- PC2 and PC3 to generate either glucagon 1387 betologia 2012;55:3262–3272

S148 DIABETES CARE, VOLUME 36, SUPPLEMENT 2, AUGUST 2013 care.diabetesjournals.org