A Highly Ca2+-sensitive Pool of Granules Is Regulated by Glucose and Protein Kinases in Insulin-secreting INS-1 Cells

We have used membrane capacitance measurements and carbon-fiber amperometry to assay exocytosis triggered by photorelease of caged Ca(2+) to directly measure the Ca(2+) sensitivity of exocytosis from the INS-1 insulin-secreting cell line. We find heterogeneity of the Ca(2+) sensitivity of release in that a small proportion of granules makes up a highly Ca(2+)-sensitive pool (HCSP), whereas the bulk of granules have a lower sensitivity to Ca(2+). A substantial HCSP remains after brief membrane depolarization, suggesting that the majority of granules with high sensitivity to Ca(2+) are not located close to Ca(2+) channels. The HCSP is enhanced in size by glucose, cAMP, and a phorbol ester, whereas the Ca(2+)-sensitive rate constant of exocytosis from the HCSP is unaffected by cAMP and phorbol ester. The effects of cAMP and phorbol ester on the HCSP are mediated by PKA and PKC, respectively, because they can be blocked with specific protein kinase inhibitors. The size of the HCSP can be enhanced by glucose even in the presence of high concentrations of phorbol ester or cAMP, suggesting that glucose can increase granule pool sizes independently of activation of PKA or PKC. The effects of PKA and PKC on the size of the HCSP are not additive, suggesting they converge on a common mechanism. Carbon-fiber amperometry was used to assay quantal exocytosis of serotonin (5-HT) from insulin-containing granules following preincubation of INS-1 cells with 5-HT and a precursor. The amount or kinetics of release of 5-HT from each granule is not significantly different between granules with higher or lower sensitivity to Ca(2+), suggesting that granules in these two pools do not differ in morphology or fusion kinetics. We conclude that glucose and second messengers can modulate insulin release triggered by a high-affinity Ca(2+) sensor that is poised to respond to modest, global elevations of [Ca(2+)](i).