GLP-1 acts on the pancreatic islets, where the beta and delta cells express the specific GLP-1 receptor. Again it is a G-protein coupled receptor which resembles the GIP receptor very much, also with respect to intracellular signaling, although there are differences as we shall see later. In addition, GLP-1 has important actions on gastrointestinal secretion and motility (for instance by delaying gastric emptying) and also acts to inhibit appetite and food intake. The GLPs are released after nutrient intake and both carbohydrates, lipids and proteins are effective. Most small peptides are eliminated rapidly after secretion, if not for other reasons then because they are filtered by the kidneys and destroyed, but GLP-1 is eliminated extremely rapidly, and this is because of the actions of an enzyme, dipeptidyl peptidase-4, which cleaves off the two N-terminal amino acids of GLP-1 and thereby inactivates the molecule (at least with respect to insulin secretion). Because of this, GLP-1 has an apparent half-life of less than two minutes in the body, but this figure is actually misleading because a steady state is never reached. In fact, the peptide is degraded even before it is released from the endocrine organ secreting it, namely the gut. The peptide is stored in and released from the L-cells in the intact form and diffuses from the epithelium to the capillaries of the villi, but as soon as it enters the blood vessels it is degraded by DPP-4 expressed by the endothelial cells. In this way only a third or a quarter of what was released leaves the gut in the intact form. The GLP-1 molecules then reaches the liver, but here another DPP-4 system is ready to degrade about half of what reaches the liver so that only may be 12 % is left to reach the systemic circulation; and here circulating DPP-4 may destroy the rest. It has been demonstrated in a pig experiment that only about 8 % of the newly released GLP-1 actually reaches the peripheral targets with the arterial circulation. It may seem paradoxical that a hormone is inactivated so rapidly and extensively after its release but the explanation seems to be that it is acting not only as a regular hormone via the circulation, but also interacts with afferent sensory nerve fibers in the gut These nerves signal to the brain stem and the hypothalamus, where a reflex mechanism involving the efferent vagus regulates gastrointestinal motility and pancreatic secretion. Like its effect on insulin secretion, the appetite regulating action of endogenous GLP-1 also seems to involve the sensory vagal afferents, but high concentrations of active GLP-1 in plasma are also effective and may target GLP-1 receptors behind leaks in the blood brain barrier, for instance in the area postrema. Let us have a look at the secretion of the two hormones in healthy individuals. Actually, GIP is also degraded by DPP-4 but not nearly as rapidly and extensively as GLP-1. But how can we measure the secretion of the hormones if they are degraded so rapidly?. The trick is to employ assays that will react with both the intact hormone and the truncated metabolite, so-called “total” GLP-1 and GIP assays. Results of such assays will nicely reflect the actual secretion rate. Here we follow the concentrations of “total GLP-1 and GIP” in healthy volunteers throughout a day when they are receiving 3 large meals. Note the increase in the concentrations in response to the meals and also the similarities between the insulin concentrations and at least the GLP-1 concentrations. In a subsequent experiment the postprandial plasma responses of the two hormones were mimicked by intravenous infusions while glucose concentrations were clamped at fasting levels and at 1 and 2 mmol/l above this, similar to what is seen after meals. Here we see the results of the experiment. We can see that both hormones, when present in the typical postprandial concentrations will effectively stimulate insulin secretion even at the fasting glucose level, but more so as the glucose concentration rises in response for instance to the meal. This is what we mean when we say that their actions are “glucose dependent”. Looking at glucagon secretion, the predominant pattern is the inhibitory effect of the glucose clamps, but clearly GLP-1 further inhibits secretion, while GIP actually weakly enhances glucagon secretion. So here we see the differential effect of the two hormones: whereas both hormones glucose-dependently stimulate insulin secretion, GLP-1 inhibits glucagon secretion while GIP enhances it. Now, let us look at the incretin effect in type 2 diabetes. In this classical study by Michael Nauck, at that time working with Werner Creutzfeldt in Göttingen, the incretin effect was examined in type 2 diabetic patients and in healthy controls. Similar glucose excursions were obtained in the patients and the controls but while a nice amplification of insulin secretion was observed in the controls, there was not much of a difference between the insulin responses to oral and intravenous glucose in the patients; in other words, they appeared to have lost a considerable part of the incretin effect, and this of course ruins their ability to keep plasma glucose concentrations low after meal intake. This, in turn, has very serious consequences regarding development of diabetic complications. The question then arises: why did they lose the incretin effect? We can analyse that today because we know that the effect is mainly due to the actions of GIP and GLP-1. So what does the secretion of the two hormones look like in people with type 2 diabetes? Here is one of the largest studies of meal-stimulated GIP and GLP-1 secretion in people with type 2 diabetes and individually matched controls. While GIP secretion is nearly normal, GLP-1 secretion is clearly decreased in the patients, particularly in the late phase of the meal. Therefore, a decreased secretion of GLP-1 undoubtedly contributes to a decreased incretin effect. Similar findings were made recently in a very large study comprising almost 1500 individuals. In the next study we look at the effects of the two hormones, first in physiological concentrations. These were brought about by intravenous infusion during the conditions of a hyperglycemic clamp in both patients and controls. In the controls, both hormones very strongly stimulated insulin secretion, in addition to the effect of glucose alone, nicely illustrating the remarkable efficiency of the hormones. However, in the patients there was hardly any response to the glucose clamp or to the two hormones. In other words, the main factor responsible for the loss of incretin effect in type 2 diabetes is a loss of the effectiveness of both hormones. But could this be due to an impaired sensitivity of the beta cells to the incretin hormones? This was addressed in further experiments. Here larger amounts of the two hormones were infused during hyperglycemic clamps in patients and controls. Again the patients responded very poorly to the hyperglycemic clamp, compared to matched controls and again GIP was virtually inactive but GLP-1 stimulated insulin secretion up to levels that actually exceeded those observed in controls in response to the glucose infusion alone. In other words, the GLP-1 infusion actually restored to beta cell sensitivity to glucose to normal levels. Looking at glucagon secretion, the hyperglycemioc clamp strongly suppressed glucagon secretion in the controls, while in the patients the suppression was delayed and impaired. GIP infusion actually increased glucagon secretion, whereas GLP-1 elicited an inhibition similar to that observed in the controls. In other words, the GLP-1 infusion was also able to restore the alpha cell sensitivity to glucose. These findings clearly illustrated the potential of GLP-1 for diabetes therapy.
