Glucagon suppression by somatostatin reduces or abolishes hyperglycemia in dogs made insulin-deficient by somatostatin, alloxan, or total pancreatectomy. This suggests that the development of severe diabetic hyperglycemia requires the presence of glucagon, whether secreted by pancreatic or newly identified gastrointestinal A cells, as well as a lack of insulin. Glucagon suppression could improve therapeutic glucoregulation in diabetes.

No abstract

No abstract

A B S T R A C T The effects of glucose, amino acids, pancreozymin-cholecystokinin, and tolbutamide upon the release of immunoreactive somatostatin (IRS) from the isolated perfused pancreas were studied. In seven experiments in which glucose was perfused either at a concentration of 100 or 350 mg/dl or at 25 mg/dl, IRS levels were significantly greater at-the higher glucose concentrations. In three doseresponse experiments in which the perfusing glucose concentration was increased at 30-min intervals from an initial concentration of 25 mg/dl to a final concentration of 300 mg/dl, progressive increases in IRS release were noted at glucose concentrations of 100 mg/dl and above. Perfusion of a 20 mM mixture of 10 amino acids also elicited a prompt and significant biphasic IRS rise in each of six experiments. In five experiments, 20 mM leucine evoked a similar response in mean IRS. Perfusion with 0.075 Ivy U/ml of pancreozymin-cholecystokinin, with or without the presence of a 1 mM 10-amino acid mixture, elicited a prompt rise in IRS with a pattern resembling that of insulin in a total of six experiments. Tolbutamide (0.75 mg/min) also stimulated IRS release in five of six challenges. The IRS responses to nutrients and to pancreozymin and their similarity to the insulin responses raise the possibility that, like insulin, pancreatic somatostatin may have an endocrine role related to nutrient homeostasis.

The location of the somatostatin-containing D-cells of the pancreatic islets between the A-and B-cells suggests that their function might be to inhibit insulin and/or glucagon secretion by these neighboring cells. To Somatostatin immunoreactivity has been demonstrated by immunocytochemical technics (1, 2), by radioimmunoassay (3), and by bioassay (4) to be present in the islets of Langerhans and more recently has been identified in the secretion granules of D-cells (5-7). The fact that D-cells are situated between A-and B-cells, cells whose secretory function somatostatin so profoundly inhibits (8-10), has suggested the possibility of local hormone-cell interactions between the secretory product of the D-cells and one or both of the neighboring cell types, interactions that might influence glucagon and/or insulin secretion (11).To test this hypothesis, we measured immunoreactive somatostatin (IRS) in the effluent of the isolated dog pancreas during perfusion with high concentrations of glucagon and insulin intended to simulate those which might be present in the extracellular space surrounding the D-cells. In additional experiments, arginine, which stimulates endogenous glucagon and insulin secretion, was also perfused. MATERIALS AND METHODSPancreases were isolated from fasting 17-to 20-kg mongrel male dogs by the technic of Iversen and Miles (12). Lymph nodes in the pancreatic region were ligated and removed. Blood vessels and organ attachments were left intact until 1-3 min before perfusion was initiatiated. The effluent cannula, secured in the portal vein immediately proximal to the point of bifurcation, drained the entire organ. The duodenal lumen was intubated and drained by gravity flow.The perfusion buffer was Krebs-Ringer bicarbonate solution as modified by Iversen and Miles (12), except that the glucose concentration was 1 g/liter, fumarate, glutamate, and pyruvate were not added, and the albumin concentration was 0.6% (wt/vol). Liter aliquots of perfusion fluid with a 95% 02/5% CO2 mixture bubbling constantly at a rate of 3 liters/min were maintained in a water bath at 370, and perfused by means of a Harvard peristaltic pump at a constant flow of 17 ml/min. In the first three experiments, collections were made for 30-45 sec and in all subsequent experiments for 1-min periods. Samples were collected in chilled tubes maintained in an ice bath and then stored at -20°until assayed.Insulin and glucagon were perfused at a concentration of 10 milliunits (mU)/ml and 10 ,g/ml, respectively. In other experiments arginine was perfused at a 20 mM concentration. An experiment was considered technically unacceptable if the appearance of the pancreas changed during the experiment as a consequence of edema, hemorrhage, or cyanosis, if leaks developed in the perfusion system, if the flow rate declined, if levels of all hormones were unmeasurable throughout an experiment, and if the viability test of insulin or glucagon response to 10 mM arginine after perfusion was impaired. IRS was measured by a modification...