The focus of this article is on the analysis of the release and postrelease fate of the incretin hormones, glucagon-like peptide-1 and glucose-dependent insulinotropic polypeptide. Their actions are dealt with to the extent that they are linked to their secretion. For both hormones, their posttranslational processing is analyzed in detail, because of its importance for the understanding of the molecular heterogeneity of the hormones. Methods of analysis, in particular regarding measurements in plasma from in vivoexperiments, are discussed in detail in relation to the molecular heterogeneity of the hormones, and the importance of the designations “total” versus “intact hormones” is explained. Both hormones are substrates for the ubiquitous enzyme, dipeptidyl peptidase-4, which inactivates the peptides with dramatic consequences for their physiological spectrum of activities. The role of endogenous and exogenous antagonists of the receptors is discussed in detail because of their importance for the elucidation of the physiology and pathophysiology of the hormones. Regarding the actual secretion, the most important factors are discussed, including gastric emptying rate and the influence of the different macronutrients. Additional factors discussed are the role of bile, paracrine regulation, the role of the microbiota, pharmaceuticals, and exercise. Finally, the secretion during pathological conditions is discussed. © 2019 American Physiological Society. Compr Physiol9:1339-1381, 2019. J. J. Holst, N. J. Wewer Albrechtsen, M. M. Rosenkilde, C. F. Deacon. Physiology of the Incretin Hormones, GIP and GLP-1—Regulation of Release and Posttranslational Modifications. Compr Physiol9: 2019, 1339–1381. Glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) are important incretin hormones, responsible for a major part of postprandial insulin secretion and, therefore, for glucose tolerance.The insulinotropic action of GLP-1 is preserved in type 2 diabetes (T2DM), while that of GIP is lost. Glucagon-like peptide-1 receptor agonists (GLP-1RAs) are, therefore, utilized for T2DM treatment.Because of their (patho)physiological role, knowledge of their secretion and metabolism is important.Their secretion is mainly prandial, and all macronutrients stimulate their secretion, often secondary to their absorption.Studies of their physiological role in metabolism are greatly facilitated by the use of receptor antagonists, exendin 9-39 for the GLP-1 receptor and, more recently, GIP 3-30 amide for the GIP receptor.Both hormones are rapidly metabolized by the ubiquitous enzyme, dipeptidyl peptidase-4 (DPP-4), which truncates and inactivates the peptides.The extremely rapid degradation of GLP-1 suggests that the endogenous peptide mainly acts via neuronal signaling.Inhibitors of DPP-4 significantly enhance the survival of GIP and GLP-1 and are used to treat T2DM.Incretin hormone posttranslational processing and their metabolism create considerable molecular heterogeneity, which not only impacts upon their spectrum of biological activities, but also complicates their measurement in plasma.In general, secretion is best studied using assays that codetermine the intact hormone as well as any metabolites (total hormone levels). The following legends to the figures that appear throughout the article are written to be useful for teaching. Figure 1.The incretin effect and its impairment in type 2 diabetes. Patients with type 2 diabetes and matched healthy controls received 50 g of glucose orally (black circles). On a subsequent day, they received an intravenous infusion of glucose (white circles), which was adjusted so that the plasma glucose concentrations were similar to those obtained on the day with the oral glucose administration. Subsequently, insulin and C-peptide concentrations were measured in peripheral blood. Asterisks indicate significant (p< 0.05) differences. Didactic explanation. The figure illustrates the classical method of investigating the incretin effect: isoglycemic oral and intravenous glucose administration. In the control subjects, much more insulin is secreted during the oral compared to the intravenous administration. The difference is the incretin effect. C-peptide measurements show about the same difference between oral and intravenous glucose, but the changes are more accurately related to the actual insulin secretion rate, since C-peptide is not extracted in the liver. In the people with type 2 diabetes, the differences between responses to oral and intravenous glucose are much smaller, and for C-peptide, the difference is not significant. This indicates the lack of incretin effect. From Nauck et al. (290). Figure 2.A schematic representation of the differential processing of proglucagon in the pancreas and in the intestine. The numbers refer to the numbering of the amino acid residues of proglucagon, counting from the N-terminus. GRPP, Glicentin-related pancreatic polypeptide; IP-1, intervening peptide-1; IP-2, intervening peptide-2. GLP-1 (glucagon-like peptide-1) thus corresponds to proglucagon 78-107NH2 (meaning that its carboxyterminus is amidated). Didatic explanation. It is seen that in the (human) pancreas, proglucagon is cleaved (by the prohormone convertase PC-2) to the pancreatic glicentin-related peptide (GRPP) corresponding to proglucagon (PG) 1-30; glucagon (PG 33-61); a small intervening peptide; and the major proglucagon fragment (MPF) consisting of amino acid residues PG 72-158. In the gut, the processing products (PC1/3) are glicentin (PG1-69), about 1/3 is processed further to GRPP and oxyntomodulin (PG 33-69); GLP-1 (PG 78-107amide) and GLP-2 (PG 126-158) (the missing amino acids are cleaved off by the processing enzymes). Figure 3.Healthy subjects were given glucose orally with or without a concomitant infusion of exendin 9-39, and had insulin and glucagon concentrations measured in their peripheral blood. Didactic explanation. During exendin infusions, insulin and C-peptide responses to glucose were clearly reduced, whereas glucagon responses increased. The interpretation is that GLP-1 secreted from the gut in response to glucose normally acts to enhance insulin secretion but also to inhibit glucagon secretion, and that both responses are prevented or even reversed by exendin 9-39. From Salehi et al. (367). Figure 4.Total GLP-1 responses to a mixed meal (538 kcal) in 54 patients with type 2 diabetes (HbA1c 8.4%—black circles), 33 matched controls with normal glucose tolerance (white circles), and 15 subjects with impaired glucose tolerance (white squares). Asterisks indicate significant differences between patients and controls. Didactic explanation. The study shows that in these cohorts of patients and controls (which were carefully matched for body weight, age, and gender), there was a significantly impaired GLP-1 response to the mixed meal in the patients with T2DM. Interestingly, the fasting concentrations in the patients were significantly higher than that of the other groups, and it is clear that the impaired response occurs rather late, with the curves separating around 1 h after meal intake. Note the tendency to a rise already after 10 min. A biphasic response has been mentioned in the literature, but clearly is not seen here. The figure shows “total GLP-1” responses (see text for explanation). In younger and leaner populations, greater meal responses may be seen (322). From Toft-Nielsen et al. (422). Figure 5.Glucose absorption and GLP-1 secretion from isolated perfused rat upper small intestine. (A) Effluent concentrations of glucose (blue) and GLP-1 (black) during brief luminal administration of a 20% glucose solution in saline. Bombesin (BBS), a Gq-activating ligand of the bombesin 2 receptor, was infused toward the end of the experiment as a positive control. (B) Correlation between GLP-1 and glucose concentrations in the venous effluent. (C and D) Responses to repeated stimulations with 20% glucose. (D and F) Effect of the SGLT-1 inhibitor, phloridzin, on GLP-1 responses to glucose. Didactic explanation. The experiments illustrate the effect of luminal administration of glucose on the secretion of GLP-1 from the gut. The gut preparation also absorbs the luminal glucose load, measured as increasing glucose concentrations in the venous effluent from the perfused segment. Not only are GLP-1 secretion and glucose absorption highly correlated (panel B), but the powerful sodium-glucose cotransporter (SGLT-1) inhibitor, phloridzin, nearly completely blocked the GLP-1 response. This indicates that the glucose absorption is required for GLP-1 secretion, and suggests that it is the glucose entry into the L-cell that is responsible for stimulation of secretion. Further details may be found in Kuhre et al. (230). Figure 6.Diagrammatic representation of the gastrointestinal anatomy after a Roux-en-Y gastric bypass operation. Didactic explanation. Importantly, the gastric pouch is very small (~30 mL), which means that there is no reservoir for retention of food, which therefore passes directly on to a more distal segment of the small intestine. The so-called alimentary limb draining the pouch does not receive any of the digestive secretions (bile, gastric, and pancreatic secretion) and, therefore, depends entirely on its brush-border enzymes for digestive capacity. Digestion requiring the digestive secretions can only begin distal to the entero-entero-anastomosis between the secretory limb and the alimentary limb, together forming the “common limb,” but this where the density of, for instance, the L-cells is high. Figure 7.A diagrammatic representation of the degradation of GLP-1 by dipeptidyl peptidase-4 (DPP-4). Didactic explanation. The figure shows a (very enlarged) intestinal villus with a single yellow open type L-cell. Newly secreted GLP-1 is indicated by black dots. Upon stimulation, GLP-1 is released by exocytosis and diffuses across the lamina propria of the mucosa until it finds and enters a capillary. Here, the GLP-1 molecules meet DPP-4 expressed on the luminal surface of the endothelial cells lining the capillary. As a result, only about 33%-25% of what leaves the gut remains in the intact form. In the liver, there is also the DPP-4 activity degrading about half of what the liver receives. This means that only about 10%-15% of what was released makes it to the systemic circulation in the intact form. In plasma, there is soluble DPP-4 causing further degradation, and it has been calculated that only about 8% of what was originally released reaches the target organs (e.g. the pancreas) in the intact form (170). Figure 8.Degradation of GLP-1 by DPP-4. (A) DPP-4 mediated degradation of the GLP-1 molecule. The diagram shows that the enzyme cleaves off the two N-terminal amino acids leaving the inactive metabolite, GLP-1 9-36/37, and that this occurs rapidly (resulting in a plasma half-life in humans of about 2 min) and with a high plasma clearance. If DPP-4 is blocked, it can be determined that the kidneys alone are responsible for a high plasma clearance, resulting in a plasma half-life of 4-5 min. (B) Concentrations in plasma of intact GLP-1 and total GLP-1 (intact GLP-1 7-36amide + the metabolite GLP-1 9-36amide) after i.v. injection of the maximally tolerated dose of GLP-1 (1.5 nmol/kg) in patients with type 2 diabetes. In spite of the high dose injected, very little GLP-1 survives in the intact form (from Deacon et al. (83)). Figure 9.An alternative signaling pathway for GLP-1 via sensory afferent of the vagus nerve. Didactic explanation. Confer with Figure 7. During the diffusion across the lamina propria, newly secreted GLP-1, which is still intact, has the chance of binding to GLP-1 receptors expressed on nerve fibers of sensory afferents of the parasympathetic nervous system (f). Indeed, such receptors are synthesized in the cell bodies of these neurons in the nodose ganglion (c). The neurons project to the nucleus of the solitary tract (a) where they may interact with other neurons, projecting to the hypothalamus or to the dorsal vagal motor nuclei (b), eventually leading to stimulation of efferent nerve fibers reaching the peripheral organs (g, h) via the vagus (from Holst and Deacon (173)). Figure 10.Processing of proglucagon and antigenic determinants utilized for antibody generation allowing specific measurement of the various circulating components. The processing scheme is identical to the one shown in Figure 2, but also shows the various molecular forms of GLP-1 resulting from the differential pancreatic and intestinal processing, the absence or presence of amidation, and DPP-4-mediated degradation products. Didactic explanation. Obviously, the black antibodies, directed against a mid-region of GLP-1 (the so-called side-viewing), will react with all of the molecular forms of GLP-1 regardless of the origin (including the major proglucagon fragment), and therefore completely lack specificity. The C-terminal antibodies (terminal wrapping) will react with GLP-1 1-36NH2, 7-36NH2, and 9-36NH2, and have therefore been used for measuring “total GLP-1” (although the cross-reaction with GLP-1 1-36NH2, which is derived from the pancreas, represents a problem). Similar results may be obtained with a sandwich ELISA combination of black and red antibodies, which currently is the combination used in commercial ELISAs for “total GLP-1.” For specific measurement of the Gly-extended molecular forms, a terminal wrapping antibody against the free acid in position 37 is required (yellow antibody). For specific measurement of the individual molecular forms, a combination of the C-terminally wrapping antibodies and N-terminally wrapping antibodies may be used. Thus, for intact GLP-1 sandwich ELISAs, blue and red antibodies or blue and yellow antibodies may be used. It is clear that specific measurement of GLP-1 in the circulation is challenging, and not many assays are sufficiently specific. Indeed, most antibodies claimed to be truly terminal are somewhat “side-viewing” and, therefore, suffer from lack of specificity. Figure 11.Secretion of GLP-1 and GIP in patients with type 2 diabetes (blue curves) and in healthy control subjects (green curves) in response to increasing doses of glucose (25, 75, and 125 g) and in response to intravenous infusions of glucose resulting in similar glucose excursions (isoglycemia). Color tones indicate each set of OGTT and IGII [light, 25-g OGTT (closed symbols) and corresponding IIGI (open symbols); medium, 75-g OGTT (closed symbols) and corresponding IIGI (open symbols); and dark, 125-g OGTT (closed symbols) and corresponding IIGI (open symbols)]. Didactic explanation. In these experiments, none of the intravenous infusions resulted in any changes in hormone secretion, as expected. The responses to oral administration were clearly dependent on the dose, with short-lived increases in GIP secretion after the low glucose dose and more protracted responses to the larger doses, although reaching almost the same peak or plateau levels. For GLP-1, a similar pattern was observed for the two larger doses, whereas the response to the low dose was clearly smaller. These responses are readily explained by the markedly different rates of gastric emptying associated with the three doses, as shown in the bottom panel. The nutrient regulation of gastric emptying ensures that a rather constant amount of nutrients (in kcal per min) is delivered to the small intestine as long as there is something left in the gastric reservoir. The hormone responses, which are generated by the presentation and absorption of the nutrients, therefore follow the gastric emptying rate. From Bagger et al. (19). Figure 12.The relationship between the intestinal somatostatin cells and the L-cells. Didactic explanation. The diagram shows three cells, an L-cell, a somatostatin producing D-cell, and an interspaced enterocyte. Both the D-cell and the L-cell are of the “open” type with a projection with microvilli reaching the gut lumen, but the D-cell has the characteristic basal cytoplasmic process of the paracrine cells, which, in this case, contacts the L-cells. The L-cell expresses somatostatin receptors, and the D-cell expresses GLP-1 receptors, so that GLP-1 reaching the D-cells stimulates somatostatin secretion, which in turn restrains L-cell secretion. If the feedback cycle is interrupted with receptor antagonists, GLP-1 secretion increases markedly (based on Hansen et al. (146)). Figure 13.ProGIP and its intestinal processing. Didactic explanation. Once the signal peptide of Prepro-GIP is cleaved off in the Golgi apparatus, the remaining proGIP is processed further in the granules by the prohormone convertase 1/3 to release full-length GIP 1-42 (proGIP 52-93). In the circulation, about half of circulating GIP is degraded by dipeptidyl peptidase-4 (DPP-4) to generate GIP 3-42 (whereas GIP, unlike GLP-1, does not seem to be degraded locally in the gut). Small amounts of a C-terminally truncated and amidated form, GIP 1-30NH2, which is a full agonist on the GIP receptor, may also be formed and, upon DPP-4 digestion, this form may be degraded to generate GIP 3-30NH2, which is a potent GIP receptor antagonist. The figure also indicates how antibodies against specific regions of the molecules may be used to measure the various molecular forms identified in the circulation.