Pancreatic islets rich in β cells and microdissected from noninbred ob/ob mice contained about 15 pmol sialic acid per μg dry weight as determined by a microversion of the Warren thiobarbituric acid method. Removal of half of the sialic acid by treatment with Clostridium perfringens neuraminidase had no effect on the islet content of insulin, moderately inhibited glucose oxidation, and resulted in decreased insulin secretory responses to 20 mM D glucose, 5 mM D glyceraldehyde, 20 mM L leucine, 10 mM L arginine, and 5 μg of glucagon per ml. In contrast, the insulin releasing action of 20 mM glucose in combination with 5 mM theophylline, 1 mM N 6, O 2' dibutyryl adenosine 3':5' monophosphate, or 0.1 mM iodoacetamide, and insulin release in response to 0.1 mM glibenclamide, 0.05 mM chloromercuribenzene p sulfonic acid, 1 mM 6,6' dithiodinicotinic acid, and 1 mM 4 acetamido 4' isothiocyanate stilbene 2,2' disulfonic acid were not significantly inhibited. In addition to the above previously known stimulators of insulin release, cholera toxin was studied. It did not stimulate insulin release in the absence of glucose but potentiated the insulin releasing action of 20 mM glucose. The potentiating action of 5 μg of cholera toxin per ml was abolished by neuraminidase treatment of the islets. In control islets incubated for 60 min with 1 or 5 μg of cholera toxin per ml or 5 mM theophylline, the content of cyclic adenosine 3':5' monophosphate (cyclic AMP) was significantly increased and was correlated with the potentiating actions of cholera toxin and theophylline on glucose initiated insulin release. Treatment with neuraminidase abolished the capacity of cholera toxin, but not that of theophylline, to raise the islet content of cyclic AMP. The cyclic AMP content did not differ between islets incubated for 60 min with 3 or 20 mM glucose alone, although insulin release was markedly stimulated by the higher concentration. Conversely, cholera toxin doubled the content of cyclic AMP in control islets incubated without glucose although insulin release was not stimulated. It is suggested that sialic acid is essential for the activity of adenylate cyclase in pancreatic β cells and hence for the maintenance in these cells of cyclic AMP at a concentration that is requisite for adequate insulin release in response to glucose and certain other stimuli. However, the results fail to support the idea that the capacity of glucose to initiate insulin release is mediated by stimulation of the adenylate cyclase. (Less)