Abstract
Since the pioneering work of Dean P. M. and Matthews E. K. (1970), more than five decades have elapsed without any consensus on the mechanism responsible for the oscillations of the insulin secretion stimulated by glucose. It was found in our laboratory that β-cells connexin36 hemichannels (Cx36Hs), non-coupled to neighbor’s cells to form gap junction channels, allowed the cytosol communication with the extracellular space under determined circumstances. They are opened by plasma membrane depolarization when expressed in Xenopus oocytes and exhibit a molecular exclusion limit of 1 KD for the diffusion of molecules and ions. There is strong experimental evidence that their opening by high KCl in β-cells leads to loss of adenine nucleotides, amino acids, and K+, provoking a suppression of insulin secretion stimulated by non-glucose secretagogues like branched-chain α-ketoacids. Glucose has been shown to inhibit the opening of Cx36 hemichannels expressed in Xenopus oocytes with a dose dependency like that established for the stimulation of insulin secretion. Islets from mice with a knockout of the connexin 36 were glucose intolerant and did not show any loss of cytosolic molecules by KCl depolarization. The objective of this review is to offer a hypothetical explanation of the possible existence of a functional interaction between K+ATP and Cx36 hemichannels. It is based on the supposition that closure of K+ATP channels might cause the opening of Cx36 hemichannels, contributing to the plasma membrane repolarizations of the silent phases. Nowadays, the alternation of active and silent phases in the oscillating behavior of plasma membrane voltage variations responsible for the oscillating nature of glucose-stimulated insulin secretion is attributed to either oscillation of β-cell glycolysis or mitochondrial metabolism. It is first discussed whether β-cell glycolysis might oscillate like in the skeletal muscle under the light of recent metabolic data. Secondly, the new suggested hypothesis is compared with previous proposals supporting that metabolic or [Ca2+]Cyt oscillations drive β-cell plasma membrane voltage and insulin secretion oscillations stimulated by glucose. The main conclusion is that this new hypothesis supports that mitochondrial oxidative phosphorylation oscillations drive membrane voltage and secretory oscillations.