Modulation of insulin secretion by cyclic AMP and prostaglandin E: The effects of theophylline, sodium salicylate and tolbutamide

Abstract

Cyclic AMP (cAMP) and prostaglandin E (PGE) are acknowledged to be local modulators of insulin secretion in response to physiologic stimuli. We have attempted to define the mechanisms of action of distinct classes of pharmacologic insulin secretagogues (nonsteroidal anti-inflammatory drugs, methylxanthines and sulfonylureas) by examining their dependence on changes in cAMP and PGE accumulation in monolayer cultures of neonatal rat pancreatic cells. As we have previously observed, sodium salicylate (SS, 20 mg/dl) stimulated insulin secretion in proportion to the ambient glucose concentration. SS potently inhibited PGE synthesis and decreased cAMP accumulation; exogenous PGE 1 reversed the effect on insulin secretion. Similar results were found using ibuprofen, a structurally dissimilar inhibitor of prostaglandin synthesis. Theophylline (1.5 mM) also stimulated insulin secretion in a fashion dependent on glucose concentration. However, in contrast to SS, theophylline (THEO) augmented cAMP accumulation without reducing PGE synthesis. Furthermore, the effects of THEO and SS on insulin release were additive. Exogenous PGE 1 did not reverse the effects of THEO on insulin secretion. The effects of sulfonylureas were markedly different from those of both SS and THEO. Stimulation of insulin release by tolbutamide (TOLB, 2 mg/dl) was accompanied by only minor and inconsistent inhibition of PGE synthesis. The effect of TOLB was not reversed by exogenous PGE 1 and was largely independent of ambient glucose concentration. Additionally, the stimulatory effect of SS on insulin secretion was additive to that of a maximally effective concentration of TOLB. Accumulation of cAMP could also not be implicated in TOLB action. Tolbutamide failed to augment cAMP generation, and its stimulatory action on insulin secretion was additive to that of a maximally stimulatory concentration of THEO. Virtually identical results were observed with a second sulfonylurea, chlorpropamide. We conclude that stimulation of insulin secretion by certain distinct classes of pharmacologic agents may be defined by changes in cAMP synthesis in some cases and in PGE availability in others. In the case of the sulfonylureas, other mechanisms must be sought.