Physiologic effects of incretin hormones
The GLP-1 receptor has been detected on numerous organs and tissues. High levels of GLP-1 receptors have been found on pancreatic β cells, δ cells, endothelial cells, cardiomyocytes, the gastrointestinal tract, and the kidneys . The actions of GLP-1 have been summarized in Figure 11.3.
There is an ongoing debate as to the expression of GLP-1 receptors on pancreatic acinar cells and thyroid C-cells, with discrepant results being reported in different studies. Also, the expression of GLP-1 receptors of pancreatic α cells has been controversially discussed. Whereas most studies have failed to detect such receptors on α cells, others have reported small GLP-1 receptor numbers . The discrepant results of different studies may be partly related to methodological problems with unequal specificities and binding characteristics of different commercially available GLP-1 receptor antibodies.
Modulation of islet hormone secretion
Intravenous administration of GLP-1 under fasting conditions leads to an immediate rise in insulin secretion [8,41]. These effects can be detected both at pharmacologic as well as physiologic plasma concentrations. In line with this, GLP-1 receptor knockout mice display significant reductions in insulin secretion . Notably, GLP-1 stimulates insulin secretion only in the presence of hyperglycemia , and various experiments have demonstrated that the insulinotropic effect ceases once normoglycemia has been reached . Based on stepwise hypoglycemic clamp experiments, plasma glucose concentrations of more than ∼65 mg dL-1 will be necessary to allow for GLP-1 to augment insulin release .
The β-cell effects of GLP-1 have been shown to be mediated through binding to the GLP-1 receptor on the cell surface . Intracellular signaling involves generation of cAMP via protein kinase A . This favors closure of ATP-dependent potassium channels, opening of voltage-gated calcium channels and exocytosis of insulin granules at elevated glucose concentrations . Coupling of these processes to glucokinase action explains the glucose-dependency of the GLP-1 effects. However, GLP-1 alone cannot close ATP-dependent potassium channels and initiate insulin secretion at low glucose concentrations.
In addition to the direct insulinotropic effect, GLP-1 has also been demonstrated to enhance insulin biosynthesis . Accordingly, islet insulin content was found to be increased after incubation with GLP-1 under culture conditions . These properties may explain the durability of GLP-1 action on insulin secretion. Finally, experiments in isolated β cells have suggested that GLP-1 treatment confers glucose-competence to previously quiescent β cells . This means that GLP-1 may recruit additional, previously inactive, β cells to the insulin secretory machinery. As a clinical correlate to these cell culture experiments, improvement of glucose-responsiveness in patients with diabetes has been demonstrated in “glucose ramp” experiments .
Suppression of glucagon concentrations is the second major islet cell effect of GLP-1 [8,41]. Similar to the insulinotropic effect, the suppression of glucagon levels by GLP-1 is strictly glucose-dependent, and hypoglycemic clamp experiments did not reveal any alterations of hypoglycemia counterregulation during GLP-1 infusion . The effect of endogenous GLP-1 is also believed to be relevant for the physiologic control of glucagon secretion, since experiments with i.v. administration of the GLP-1 receptor antagonist exendin (9-39) have demonstrated a significant increase in glucagon concentrations after oral glucose ingestion . There is some controversy as to the mechanisms underlying the suppression of glucagon secretion by GLP-1, since most studies did not detect sufficient amounts of GLP-1 receptors on pancreatic α cells. Inhibition of glucagon secretion secondary to the increase in insulin secretion cannot plausibly explain the effects either, because a similar reduction in glucagon levels has also been reported in patients with type 1 diabetes . However, the finding that co-administration of a somatostatin receptor 5 antagonist abolishes the effects of GLP-1 on glucagon secretion in the isolated perfused pancreas suggests that the glucagonostatic effect of GLP-1 is secondarily mediated through an increase in somatostatin release from pancreatic δ cells .
Finally, reduction of PP secretion has been demonstrated in response to GLP-1 administration. This effect is held to be mediated via an inhibition of vagal nervous activation .
Inhibition of gastric emptying during intravenous infusion of GLP-1 has been demonstrated using various techniques both in healthy individuals and in patients with type 2 diabetes [54 – 56]. Unlike the effects on islet cell secretion, the deceleration of gastric emptying by GLP-1 is rather independent of plasma glucose concentrations, but exhibits a clear dose-response relationship . More specifically, a prolongation of the lag period along with a dose-dependent inhibition of antral propagations and antroduodenal contractions has been reported during GLP-1 administration.
The finding that administration of the GLP-1 receptor antagonist exendin (9-39) increases gastric antral motility and reduces pyloric tone suggests a role for endogenous GLP-1 in the physiologic control of gastric emptying. These actions are believed to contribute to the so-called “ileal brake”, that is, the inhibition of gastric emptying by humoral and neural signals from the lower parts of the small intestine .
A transient reduction of exocrine pancreatic secretion has also been reported during GLP-1 administration . However, it is difficult to judge whether these effects are due to a direct pancreatic effect of GLP-1 on the exocrine pancreas or rather secondary to the delay in gastric emptying leading to reduced secretion of gastrointestinal hormones that would in turn stimulate exocrine pancreatic secretion .
Furthermore, some earlier studies in humans have suggested a modest reduction in gastric acid secretion by intravenous GLP-1 infusion .
Mechanistically, the deceleration of gastric emptying by GLP-1 is believed to be mediated by an inhibition of vagal activation . In support of this, co-administration of atropine in humans as well as afferent vagal denervation in rats have abolished the gastric effects of GLP-1 .
Because retardation of gastric emptying inevitably results in delayed and protracted glucose appearance in the systemic circulation, postprandial insulin concentrations are typically reduced during GLP-1 administration [55,56]. This observation has led to some controversy as to whether GLP-1 really fulfills the classical criteria of an incretin hormone (i.e., an augmentation of postprandial insulin secretion at typical plasma concentrations) . The importance of the gastric emptying effect of GLP-1 also becomes apparent from experiments, in which the prokinetic drug erythromycin was co-administered along with GLP-1 in order to antagonize the deceleration of gastric emptying. Under those conditions, the GLP-1-induced reduction of postprandial glycemia was less pronounced than with GLP-1 alone, and the inhibition of insulin release was largely abolished .
Notably, the effect of GLP-1 on gastric motility appears to be subject to rapid tachyphylaxis and tends to wane with continued exposure to high GLP-1 plasma concentrations.