Injections and infusions of oxytocin into conscious dogs caused an increase in plasma concentrations of glucose, insulin and glucagon. When blood glucose was clamped at a raised level the injection of oxytocin still increased insulin and glucagon concentrations in plasma. Infusion of somatostatin suppressed plasma concentrations of glucagon and insulin but did not prevent oxytocin-induced increments in blood glucose. Injection of oxytocin still caused a marked release of glucagon, whereas the insulin response was greatly diminished. When endogenous insulin and glucagon secretion was suppressed by infusion of somatostatin and glucose levels were stabilized by concomitant infusions of glucagon and insulin, injections of oxytocin did not alter blood glucose concentrations. It is concluded that the increase in blood glucose following the administration of oxytocin is secondary to the release of glucagon and that oxytocin exerts a direct stimulatory effect on glucagon and possibly insulin secretion.

Many years ago, it was suggested that the duodenum contains a factor which controls the internal (i.e. endocrine) secretion of the pancreas (1). After the isolation of insulin this suggestion was refined and it was proposed that an intestinal factor called incretin controlled glucose-induced insulin release (2).After many years neglect, interest in this concept was revived by the observation that an oral glucose load caused higher levels of circulating immunoreactive insulin (IRI) than an intravenous glucose load giving the same peripheral glucose concentrations (3). Attempts were, therefore, made in the early 1970s to isolate and identify insulin releasing factors from the intestinal mucosa (4, 5).At much the same time, a peptide called gastric inhibitory polypeptide (GIP) was isolated from porcine intestine on the basis of its ability to inhibit histamine-induced gastric acid secretion (6). GIP was shown to have insulin releasing activity in vivo (7) and it was thought that the search for the elusive "incretin" of Zunz & LaBarre (2) was at an end. GIP is synthesized in and released by the K cells which are mainly found in the mucosa of the duodenum and jejunum (8). Porcine GIP contains 42 amino acid residues in a sequence which has marked similarities to glucagon, VIP and secretin (9, 10). Human GIP differs from porcine GIP in two residues (1 1) and bovine GIP differs from human GIP in three residues (12) (Fig. 1).Studies of the levels of imrnunoreactive GIP (IRGIP) in human plasma showed that the upper intestine released GIP after the ingestion of carbohydrates and fats (13). In man, mimicking these post-prandial levels of IRGIP by infusing purified porcine GIP caused insulin release, provided that the peripheral glucose concentration was approximately 2 mmoW above fasting (14).In brief, porcine GIP is a peptide known to release insulin in vitro (see, for example, 15) and in man in the presence of moderate hyperglycaemia (7, 14). There is some controversy as to the role of endogenous GIP in the control of post-prandial insulin release (16).This peptide, therefore, warrants investigation as a potential insulin releasing agent in diabetic man, especially since in the early phase of insulin-dependent diabetes GIP repsonse to a meal is below normal (17).A study of the effects of infused natural porcine GIP on insulin release and glucose clearance in normal and diabetic man is now in progress. This paper briefly presents initial findings after the infusion of porcine GIP in man during a glucose clamp and together with an infusion of glucose. METHODS AND RESULTSThe porcine GIP used in this study was isolated from porcine small intestines, fully sequenced (10) and quantified by measuring the amino acid content of the stock solution. This batch of GIP w a s shown to be free of GIP fragments, especially of Des-Tyrl-Ala2 GIP, a biologically inactive peptide (10) known to be present in other batches of GIP (9) infused into man (14, 16). Des-Tyrl-Alaz GIP has full molar reactivity in all GIP RIAs