No abstract

IT has been reported previously that fasting or fat feeding causes a definite reduction in the insulin content of the pancreas of rats [Haist, Ridout & Best, 1939;Best, Haist & Ridout, 1939]. It was thought that the administration of insulin to fasting animals or to those receiving fat might throw some light on the mechanism by which this effect was brought about. Several interesting facts have emerged. Insulin administration produces a decrease in the insulin content of the pancreas in fed rats and augments the fall in fasting or fat-fed animals. METHODSThe procedures employed for the extraction of insulin from the pancreas and the determination of the potency of the insulin solutions were the same as reported previously [Best et al. 1939]. Male rats of the Wistar strain weighing between 200 and 300 g. were used and the pancreatic tissue from ten rats was taken for each test. RESULTS(1) The effect of insulin in fed rats Several experiments were performed to demonstrate the effect of insulin administration in rats fed a balanced diet. The groups injected with insulin received the same diet and the same caloric intake as the controls. In the first experiment a daily subcutaneous injection of 1 unit of protamine zinc insulin was given to each rat. In the second expeximent each rat received 1 unit of protamine zinc insulin daily for the first 2 days, 2 units daily for the following 2 days, 3 units on the next day and 1 A preliminary account of part of this work appeared in Science (1940), 91, 410.

The effects of a standard glucose tolerance test (GTT) and of pretreatment with aminophylline (AM) were examined in dogs with a reduced reserve of islet function prepared by partial removal of the pancreas and in normal intact dogs. Serum glucose, free fatty acid and insulin levels were measured. No significant differences were found before and after partial pancreatectomy when less than 50 per cent of the pancreas was removed or after sham operation. When AM was infused prior to the GTT, serum insulin levels were elevated to almost twice those obtained during a standard GTT. Dogs with less than 50 per cent of the pancreas remaining showed deterioration in GTT two weeks after operation, although the fasting blood sugar was normal except in one animal. The deterioration in GTT response progressed with time. Maintaining the animal on a high protein diet improved the GTT response. In chow-fed dogs with less than 50 per cent of the pancreas but no glucosuria or fasting hyperglycemia, prior AM infusion significantly improved the GTT and increased insulin secretion. In one dog with fasting hyperglycemia, AM did not cause improvement.

Incorporation of tritium-labeled leucine into proinsulin and insulin was used to indicate proinsulin synthesis and insulin formation. In these experiments the incorporation of labeled leucine into proinsulin and that into insulin were closely related. This incorporation was stimulated by addition of glucose and mannose but not by fructose. The glucose- and mannose-stimulated incorporation was inhibited by mannoheptulose. Inhibition by mannoheptulose was not relieved by addition of pyruvate. Pyruvate, with the addition of small amounts of fumarate and glutamate, failed to stimulate proinsulin and insulin synthesis even though insulin secretion was stimulated by tolbutamide or ribose. It was concluded that neither free glucose, nor pyruvate derived from glucose, acts as a signal to initiate biosynthesis. Possibly some more immediate metabolite of glucose or some enzyme or cofactor associated with its metabolism may be involved. Ribose and xylitol failed to stimulate the biosynthesis of proinsulin, and insulin. This, taken together with the reported stimulating effect of ribose and xylitol on insulin secretion, supports, for the islets, the postulate of Krass and LaBella for posterior pituitary hormones, that the pentose phosphate pathway is concerned with hormone storage and secretion rather than with hormone synthesis.

The experimental findings reported here may have some implications concerning the secretion of insulin under physiological conditions. They would suggest that at a given level of blood glucose, the rate of insulin loss from the beta cells may be larger when the level of insulin in the blood is low. They would suggest too that beta cells lining the proximal portions of the islet sinusoids would secrete to a greater degree than beta cells lining the more distal parts of the sinusoids.In previous experiments reported elsewhere (see reference 3) it could not be demonstrated satisfactorily that large amounts of beef insulin (30 U.) added to solutions of glucose inhibited the rate of degranulation of beta cells of the infused rats. The results of those experiments and the results with insulin-antibody, reported here, need to be reconciled. Excess circulating insulin-antibody probably keeps the concentration of free diffusible insulin at levels close to zero in the sinusoids of the islets. The mechanism facilitating beta cell degranulation may be operating effectively at extremely low concentrations of insulin. Maximal inhibition of the glucose effect may be reached very soon as the concentration of diffusible insulin rises in the sinusoids of the islets. In rats infused with glucose, endogenous insulin in the milieu of the beta cells might quickly reach this level so that the addition of exogenous beef insulin has no demonstrable effect. DIABETES J4:493-500, August 1965.Degranulation of the beta cells of the islets and decrease in the extractable insulin of the pancreas in many species have been described following hyperglycemia induced either by the administration of glucose 1 " 3 or by the injection of anti-insulin antibodies. 4 Although hyperglycemia is common in both experimental conditions,