A New Phase Of Insulin Secretion: How Will It Contribute to Our Understanding of β-Cell Function?

Although initially described two decades ago, biphasic insulin secretion has gradually been understood to reflect (β-cell rate sensitivity, be important in minimizing overinsulinization in normal individuals, be defective in non-insulin-dependent diabetes mellitus (NIDDM), and be useful as an early predictor in prediabetic individuals. Recently, a third phase of insulin secretion has been observed in fully in vitro islets or pancreatic preparations. This phase is characterized as a spontaneous decline of secretion (desensitization) during 24 h of sustained exposure to glucose or other secretagogues and does not appear to be simply an artifact of in vitro preparations. The impaired secretion is localized to the final release process in that neither glucose-stimulated proinsulin synthesis nor its conversion to insulin is affected. The mechanisms responsible for the third phase of reduced secretion are unknown. Kinetic evidence suggests it is not caused by emptying of a single finite insulin storage compartment but does not exclude the possibility that the decreased release reflects depletion of threshold-sensitive β-cells recruited at a given secretagogue level. Alternatively, the third phase may reflect inhibition of a priming or terminal insulin-release process by metabolic feedback. Because several secretagogues cause similar third-phase impaired release, even in the absence of glucose, desensitization probably occurs at a common fundamental site in the secretory site (e.g., calcium metabolism). Preliminary studies indicate the third phase is not the result of a paracrine effect by other islet hormones or of a change in muscarinic regulation. Whether other neurologic effectors are involved requires further investigation. Experiments examining the effect of glucose-induced third-phase desensitization on other secretagogues emphasize that priming (potentiation) and desensitization of insulin secretion are occurring simultaneously and that interpretations of results can be highly dependent on the design of the experiment. The possibility that the third phase of insulin secretion may involve mechanisms (waning of priming; desensitization) that relate to glucose-induced β-cell desensitization in NIDDM is discussed. Ultimate appreciation of the significance of the third phase of insulin secretion may develop, as was true for biphasic secretion, with our increasing understanding of the underlying processes in human diabetes.