This study was designed to determine the possible role of brain glucagon-like peptide-1 (GLP-1) receptors in feeding behavior. In situ hybridization showed colocalization of the mRNAs for GLP-1 receptors, glucokinase, and GLUT-2 in the third ventricle wall and adjacent arcuate nucleus, median eminence, and supraoptic nucleus. These brain areas are considered to contain glucose-sensitive neurons mediating feeding behavior. Because GLP-1 receptors, GLUT-2, and glucokinase are proteins involved in the multistep process of glucose sensing in pancreatic beta cells, the colocalization of specific GLP-1 receptors and glucose sensing-related proteins in hypothalamic neurons supports a role of this peptide in the hypothalamic regulation of macronutrient and water intake. This hypothesis was confirmed by analyzing the effects of both systemic and central administration of GLP-1 receptor ligands. Acute or subchronic intraperitoneal administration of GLP-1 (7-36) amide did not modify food and water intake, although a dose-dependent loss of body weight gain was observed 24 h after acute administration of the higher dose of the peptide. By contrast, the intracerebroventricular (i.c.v.) administration of GLP-1 (7-36) amide produced a biphasic effect on food intake characterized by an increase in the amount of food intake after acute i.c.v. delivery of 100 ng of the peptide. There was a marked reduction of food ingestion with the 1,000 and 2,000 ng doses of the peptide, which also produced a significant decrease of water intake. These effects seemed to be specific because i.c.v. administration of GLP-1 (1-37), a peptide with lower biological activity than GLP-1 (7-36) amide, did not change feeding behavior in food-deprived animals. Exendin-4, when given by i.c.v. administration in a broad range of doses (0.2, 1, 5, 25, 100, and 500 ng), proved to be a potent agonist of GLP-1 (7-36) amide. It decreased, in a dose-dependent manner, both food and water intake, starting at the dose of 25 ng per injection. Pretreatment with an i.c.v. dose of a GLP-1 receptor antagonist [exendin (9-39); 2,500 ng] reversed the inhibitory effects of GLP-1 (7-36) amide (1,000 ng dose) and exendin-4 (25 ng dose) on food and water ingestion. These findings suggest that GLP-1 (7-36) amide may modulate both food and drink intake in the rat through a central mechanism.