The distribution of central neurons displaying somatostatin immunoreactivity was studied using three monoclonal antibodies to cyclic somatostatin. The sensitive ABC immunoperoxidase technique was employed. A large number of positive cell groups including many previously undescribed populations were detected throughout the brain and spinal cord. Telencephalic somatostatin neurons included periglomerular cells in the olfactory bulb, mitral cells in the accessory olfactory bulb, and multipolar cells in the anterior olfactory nuclei, neocortex, amygdala, hippocampus, lateral septum, striatum, and nucleus accumbens. Within the hypothalamus, positive neurons were found in the periventricular, suprachiasmatic, and arcuate nuclei, and throughout the anterior and lateral hypothalamus. The entopeduncular nucleus and zona incerta contained many positive neurons, and the lateral habenula had a dense terminal field suggesting a pallidohabenula somatostatin pathway. Somatostatin neurons were also found in association with many sensory systems. Positive cells were present in the superior and inferior colliculi, the ventral cochlear nuclei, the ventral nucleus of the lateral lemniscus, nucleus cuneatus, nucleus gracilus, and the substantia gelatinosa. Various cerebellar circuits also displayed somatostatin immunoreactivity. Golgi cells throughout the cerebellar cortex were intensely stained, and some Purkinje cells in the paraflocculus also showed a positive reaction. Positive fibers were present in the granular layer and large varicose fibers were present in the inferior cerebellar peduncle. Many nuclei known to project to the cerebellum, including the nucleus reticularis tegmenti pontis, the medial accessory inferior olive, the nucleus prepositus hypoglossi, and many areas of the reticular formation contained positive neurons. These studies demonstrate that these new monoclonal antibodies are of great value for the study of central somatostatin systems. Previously described somatostatin systems are readily detected with these antibodies, and in addition, many otherwise unrecognized somatostatin cell groups have been discovered.

The possible involvement of gastric somatostatinlike immunoreactivity (SLI) in the acid inhibitory action of gastric inhibitory polypeptide (GIP) was studied in an isolated perfused rat stomach. GIP, in a dose of 5 or 50 ng/mL, caused a 4- and 12-fold increase in SLI secretion, respectively. At the higher dose level the stimulated secretory rate declined throughout the perfusion suggesting that secretion exceeded the capacity to synthesize SLI under excessive GIP stimulation. Acetylcholine (10 microM) or vagal stimulation (7 V, 10 Hz, 5 ms) completely inhibited GIP-stimulated SLI secretion. It is therefore proposed that the acid inhibitory activity of GIP is probably mediated via release of gastric SLI and this action is under cholinergic control.

Glucose-dependent insulinotropic polypeptide or gastric inhibitory polypeptide (GIP) is a 42 amino acid intestinal hormone, which exhibits several direct and indirect effects on fat and glucose metabolism. To determine the bioactive region(s) of the molecule, synthetic and proteolytic fragments of the hormone were generated and tested for their ability to induce a biological response in the isolated, perfused rat pancreas and stomach. A synthetic fragment corresponding to porcine GIP residues 1-30 retained strong insulinotropic activity in the isolated, perfused rat pancreas but greatly reduced somatostatinotropic activity in the isolated perfused rat stomach. A synthetic fragment corresponding to porcine GIP residues 15 to 30 was biologically inactive. However, enterokinase treatment of the synthetic 15-30 fragment restored partial insulinotropic activity in the isolated, perfused rat pancreas. The hypothesis that the restoration of biological activity was due to the enzymatic removal of the amino-terminal dipeptide (Asp-Lys) of GIP15-30 was supported by the observation that a synthetic fragment lacking these two residues was also insulinotropic. Further fractionation of the molecule generated a biologically active 19-30 fragment, suggesting that the residues necessary for the insulin response are contained within this region.

Previous studies on the isolated perfused stomach have shown that gastric inhibitory polypeptide (GIP) and glucagon-like peptide-1(7-36) amide (GLP-1(7-36) amide) stimulate release of somatostatin (somatostatin-like immunoreactivity, SLI). GIP produced a paradoxical increase in gastrin secretion, whereas GLP-1(7-36) was inhibitory. In the current study, the actions of synthetic (sp) and native (np) porcine and synthetic human (sh) GIP, GLP-1(7-36), and GLP-1(7-37) on SLI and gastrin secretion were compared using a gradient perfusion of peptide. All peptides increased SLI secretion at a threshold concentration of approximately 50 pmol/L (p < 0.05). The initial rate of increase in response to spGIP (119 +/- 39 pg/min) was greater than with other forms of GIP or GLP-1. Maximal increases obtained with the two porcine peptides did not differ. Gastrin secretion was increased by concentrations of spGIP and npGIP similar to those increasing SLI secretion, but the maximal response to shGIP was lower. In contrast to GIP-induced increases, both GLP-1(7-36) and GLP-1(7-37) suppressed gastrin secretion. It is concluded that human and porcine GIP, GLP-1(7-36), and GLP-1(7-37) all stimulate SLI secretion but with different maximal effects, and GIP stimulates gastrin secretion whereas both forms of GLP-1 inhibit gastrin secretion.

Glucose-dependent insulinotropic polypeptide or gastric inhibitory polypeptide (GIP) is a 42 amino acid intestinal hormone, which exhibits several direct and indirect effects on fat and glucose metabolism. To determine the bioactive region(s) of the molecule, synthetic and proteolytic fragments of the hormone were generated and tested for their ability to induce a biological response in the isolated, perfused rat pancreas and stomach. A synthetic fragment corresponding to porcine GIP residues 1-30 retained strong insulinotropic activity in the isolated, perfused rat pancreas but greatly reduced somatostatinotropic activity in the isolated perfused rat stomach. A synthetic fragment corresponding to porcine GIP residues 15 to 30 was biologically inactive. However, enterokinase treatment of the synthetic 15-30 fragment restored partial insulinotropic activity in the isolated, perfused rat pancreas. The hypothesis that the restoration of biological activity was due to the enzymatic removal of the amino-terminal dipeptide (Asp-Lys) of GIP15-30 was supported by the observation that a synthetic fragment lacking these two residues was also insulinotropic. Further fractionation of the molecule generated a biologically active 19-30 fragment, suggesting that the residues necessary for the insulin response are contained within this region.