The idea that adenosine-Y, 5'-cyclic monophosphate (cyclic AMP) may act in the pancreatic B-cell as a second messenger was probably first raised when Samols et al. 72 reported that glucagon stimulates insulin release in man. The inhibitory effect of catecholamines on insulin secretion, first described by Coore and Randle 17, was also soon considered within the framework of this idea. Cerasi and Luft" even postulated that glucose, in addition to serving as a metabolizable substrate, acted in the B-cell on a specific membrane receptor leading to activation of adenylate cyclase, increased cyclic AMP production and stimulation of insulin release. In the latter model, cyclic AMP was considered to represent the signal for insulin release, whereas the metabolism of glucose would merely modulate this signal function, for example by increasing ATP availability to adenylate cyclase. At variance with the latter view, we had defended the concept that cyclic AMP should not be considered as a signal for insulin release but, instead, as a modulator of the metabolic and secretory response to nutrient secretagogues 53.

Almost 20 years have elapsed since cyclic AMP entered the field of insulin release, and more than a dozen years have passed since the contrasting views expressed above were introduced in relevant textbooks. In retrospect, it appears that the role of cyclic AMP in insulin release merits reevaluation. In the present report, the enzymes involved in the regulation of cyclic AMP synthesis and breakdown, the target systems responsive to this nucleotide and the role of cyclic AMP in the regulation of insulin release will be successively taken into consideration.