Of the various factors that can stimulate insulin secretion, glucose is physiologically the most important. This is reflected by the moment-to-moment fluctuations in plasma concentration that accompany fluctuations in plasma glucose concentration. The data shows that glucose metabolism within the cell, rather than a signal from a membranal “glucose receptor” produces the stimulus for insulin release. Supporting this contention is the observation that compounds that inhibit glucose metabolism, for example mannoheptulose, interfere with insulin secretion.
It would appear that the products, or intermediates of glycolysis are responsible for insulin secretion. Glucose increases the concentration of glycolytic intermediates within islet cells and so promotes insulin secretion. Mannoheptulose is a sugar that inhibits glycolysis and its presence reduces the amount of insulin secreted. As with many intracellular processes, cAMP participates in the insulin secretory process. cAMP is believed to act as a positive synergistic modulator of a glucose-sensitive secretory step.
An increase in cAMP concentration, without glucose, is not sufficient to stimulate insulin secretion. Glucose therefore leads to an increased intracellular concentration of cAMP that is in turn thought to promote insulin secretion by depolarising the cell, that is, by making the resting potential become more positive. IF 15mM of glucose is added to a beta-cell within an isolated Islet of Langerhans, within about a minute its membrane potential is found to change from its resting potential of about -60mV to -30mV.
This depolarization is a consequence of the decrease in the membrane’s permeability to potassium ions that is observed in the presence of glucose. It has recently been possible to obtain records of a single ion channel in a patch of membrane on an intact beta-cell, thanks to a technique known as “cell-attached patch-clamp recording”. These records show the current level when the ion channel is in the open and closed state, allowing one to calculate the proportion of time that the channel is open, or the “open-state probability”. In the complete absence of glucose the open state probability is about 0.4, but in the presence of 20mM of glucose this probability becomes almost zero, ie. the channel is practically always closed. Subsequent experiments showed that this kind of channel is selective for potassium ions.
The main products of glycolysis are ATP and NADH, which is further broken down, via oxidative phosphorylation, into more ATP. It is thought that it is this ATP that leads to the blocking of the potassium channels. Evidence for this has been obtained from “inside-out patches” of the potassium selective channel. This is a patch of membrane that has been pulled off the cell so that the solution in the recording chamber is on the side of the patch of membrane that usually faces the cytosol.
If 1mM of ATP is added to this solution the open-state probability of the potassium channel decreases to zero. Further evidence is obtained from the cell-attached patch-clamp recording. In the presence of 20mM glucose and 20mM mannoheptulose the channel has an open-state probability of approximately 0.2. This is because mannoheptulose inhibits glycolysis and the production of ATP, thus preventing the blocking of potassium channels.
It is possible that rather than there being two separate mechanisms leading to insulin secretion, the process of glycolysis is linked to the observed increase in intracellular cAMP levels. The main products of glycolysis are ATP and NADH, the latter being quickly broken down to yield the former. The ATP can then be converted by the membrane-bound enzyme adenylate cyclase into cAMP, which in turn could block the potassium channel producing depolarisation.
It is thought that this depolarisation leads to an increase in intracellular calcium concentration, which acts as the final triggering mechanism whereby glucose, or other stimuli, result in the release of insulin from beta cells. If the beta cells are exposed to 13mM of glucose in the extracellular fluid bath, intracellular concentrations are found to rapidly rise from 100nM to 300nM. If the experiment is repeated in the absence of calcium in the superfusate this change is not observed. This tells us that intracellular stores of calcium, for example in the endoplasmic reticulum, have little role to play in the release of insulin. Instead, the transport of calcium from the extracellular fluid is the most important influence on intracellular calcium levels. This transport must be active since the calcium must be moved against its concentration gradient.
Further evidence for the importance of changes in intracellular calcium concentration comes from studies using ionophores, molecules that act as membrane carriers for ion transport. In the presence of A23187, a specific divalent cation ionophore that transports calcium across biological membranes, addition of calcium to beta cells results in a burst of insulin secretion in the absence of rises in glucose availability or intracellular cAMP concentration. This secretion quickly tails off in a couple of minutes, which probably reflects the fact that all the preformed insulin has been released.
It should also be noted that glucose, in addition to stimulating calcium entry, may, under certain conditions, lead to intracellular sequestration and extrusion of calcium. Therefore, high levels of glucose may paradoxically cause an inhibition of insulin release, which may be a factor in the low beta cell function, noted in persons with poorly controlled type II diabetes. The overall sequence of events leading to glucose-stimulated insulin secretion is summarized below: However, one must remember that this is only a small part of a very complex physiological story. The hormone gastric inhibitory peptide (GIP), amino acids and stimulation by the parasympathetic nervous system are all found to increase insulin secretion, whereas adrenalin, noradrenaline and somatostatin all inhibit it.