Large differences in the second phase of insulin secretion by small differences in islet isolation

 

The kinetics of insulin secretion is important for the maintenance of glucose homeostasis in vivo. There is a multitude of mechanistic studies to explain the biphasic secretion, often with contradictory results. Here we show how small variations in the islet isolation procedure result in marked changes of the second phase of glucose-stimulated secretion.

Insulin secretion was measured by ELISA from the fractionated efflux of freshly isolated batch-perifused NMRI mouse islets. Such islets were also used to measure the NAD(P)H- and FAD-autofluorescence, the mitochondrial membrane potential (TMRE-fluorescence) and the cytosolic Ca²⁺ concentration ([Ca²⁺]_i, Fura-fluorescence).

The Krebs-Ringer medium during hand-picking of the islets contained either 0.2% BSA or 0.4% BSA or 10% FCS. The perifusion medium (0.2% BSA) contained 0 mM glucose for 60 min, then 30 mM for another 60 min, then again 0 mM. While the initial responses were qualitatively similar (increase from 5 to about 25 pmol x min^-1 x islet^-1 within 10 min), the subsequent responses progressively diverged. Islets isolated in 0.4% BSA showed a clearly ascending second phase (up to 50 pmol x min^-1 x islet^-1 after 60 min), those isolated in 0.2% BSA a transient decrease, those isolated in FCS a return back to prestimulatory values. These differences in the second phase were neither reflected by [Ca²⁺]_i, nor the NAD(P)H/FAD-ratio, nor the TMRE fluorescence.

Much of what is considered to be typical for the secretion pattern of mouse islets ex vivo depends on the conditions of islet isolation, apparently by affecting the metabolic amplification of secretion.