The role of glucagon-like peptide 1 in the regulation of glucose homeostasis and satiety.

Glucagon-like peptide 1 (GLP-1) is an insulinotropic peptide secreted by intestinal L-cells (mucosal cells belonging to the APUD (amine precursor uptake and decarboxylation) system in response to oral food intake. GLP-1 is co-encoded in the proglucagon gene and while post-translational processing of proglucagon in the pancreas gives rise to glucagon, GLP-1 (7–36 amide) is produced in the intestine and to a lesser extent in the hypothalamus. The postprandial secretion of an intestinal factor that lowers plasma glucose levels (incretin) had already been hypothesized a century ago. However, only 10 years ago, when GLP-1 was discovered as a second incretin besides gastric inhibitory peptide (GIP), this concept of an entero–insular axis became established. Since then the implication of GLP-1 in glucose metabolism and the potential use of this incretin in diabetes have been studied in different models including clinical studies. Recently, GLP-1 has been reported to act centrally as a satiety factor in rodents and this hormone is now also thought of as a ‘gut–brain peptide’. Finally, a GLP-1 receptor knock-out mouse could be created, confirming the importance of GLP-1 in glucose homeostasis, but casting doubts on its role in appetite regulation. These recent findings help to clarify the physiological importance of GLP-1 and are discussed in the present highlight. GLP-1 concentrations in the picomolar range induce insulin secretion from pancreatic b-cells in vitro and in vivo when elevated glucose concentrations (> 5 mmol/l) are present. In patients with non-insulin-dependent diabetes mellitus (NIDDM) parenteral (i.v. and s.c.) administration of GLP-1 led to reconstitution of the early phase insulin secretion and reduction of postprandial glucose excursions. Even in insulin-deficient type-1 diabetic patients GLP-1 reduced the insulin requirements, suggesting additional peripheral activities (1, 2). In contrast to sulfonylurea drugs, GLP-1 not only stimulates insulin secretion, but also proinsulin gene transcription via activation of the transcription factor CREB (cyclic AMP-responsive element binding protein), potentially counteracting the problem of insulin depletion of the b-cell (1). In patients with NIDDM, overnight infusion of GLP-1 improved basal and stimulated b-cell function to the level of non-diabetic controls (3). GLP-1 plasma levels are elevated in patients with NIDDM and even pharmacological doses of GLP-1 have attenuated glucose-lowering effects compared with healthy subjects. This partial resistance to GLP-1 in NIDDM is not due to a receptor defect (4) but is speculated to be caused by GLP-1 induced desensitization of pancreatic GLP-1 receptors. The GLP-1 receptor, a 63 kDa protein belonging to the group of trimeric G-protein linked receptors signaling via cyclic AMP, is found on pancreatic islets (b and d cells), adipose tissue, skeletal muscle, liver, intestine, stomach, lung and brain (hypothalamus) (1). The presence of GLP-1 receptors in hypothalamic nuclei implicated in appetite regulation raised the question whether GLP-1 might be involved in feeding behavior (5). In addition, intestinal GLP-1 secretion seems to be attenuated in obese subjects (6). In fasted rats, intracerebroventricular (i.c.v.) injection of GLP-1 inhibited food intake in a dose-dependent manner, whereas the peptide exendin (9–39), a GLP-1 receptor antagonist, increased food intake of fed rats as did the well-known appetite stimulant NPY (neuropeptide Y) (5). Intracerebroventricular injection of the GLP-1 receptor agonist exendin-4 to rats resulted in a decrease of food intake, while the biologically inactive GLP-1 isoform, GLP-1 (1–37) amide, had no effect and exendin (9–39) reversed the inhibitory effect of GLP-1 on feeding. As peripheral administration had little or no effect on appetite a central mechanism was suggested (5, 7, 8). Even though it is tempting to speculate that the incretin GLP-1 secreted postprandially is conferring a feedback mechanism by inducing satiety, it is not clear whether this effect is specific or physiological. As both the GLP-1 receptor and GLP-1-like immunoreactivity are found in the hypothalamus (8) a paracrine effect can be hypothesized. However, there is no evidence of increased hypothalamic postprandial GLP-1 secretion. In addition, it is controversial whether i.c.v. GLP-1 induces a conditioned taste aversion and thereby reduces feeding. While Thiele and colleagues (9, 10) describe an emetic effect of GLP-1, Sheikh’s group (7) did not find any GLP-1induced taste aversion in rats. In order to assess the physiological relevance of GLP-1 Drucker’s group (11) constructed a mouse model harboring a null mutation of the GLP-1 receptor thereby eliminating successfully all GLP-1 binding sites and effects. Interestingly, these GLP-1 receptor knock-out mice had no obvious phenotype and showed no difference in body weight or food intake compared with wild-type mice. Intracerebroventricular GLP-1 administration to wildtype mice resulted in a similar inhibition of feeding as in rats, whereas no effect was seen in the mutant mice. Thus, the inhibitory effect of i.c.v. GLP-1 on feeding seems European Journal of Endocrinology (1997) 137 220–221 ISSN 0804-4643