Incretin secretion, action in the natural history of type 2 diabetes

Incretin hormones contribute a major portion to the insulin secretory responses after meals in healthy people. The incretin effect describes the phenomenon that oral glucose elicits approximately threefold greater insulin responses than the elevation in glucose (achieved with glucose administered intravenously) alone.

The incretin effect is the result of nutrient-stimulated secretion of the incretin hormones glucose-dependent insulinotropic hormone (GIP) and glucagon-like peptide-1 and their insulinotropic effect (ie, the augmentation of insulin secretion at elevated plasma glucose concentrations). In patients with type 2 diabetes, this incretin effect is severely impaired or even absent.

It is the purpose of this commentary to highlight current knowledge in incretin research and to answer the question of whether and to which degree abnormalities in incretin hormone secretion and action accompany the development of type 2 diabetes or even contribute to this process.

Michael A. Nauck

Irfan Vardarli

Juris J. Meier

History of incretin

The reduced incretin effect in patients with type 2 diabetes was first noticed in 1967 and was clearly established in 1986.

Three types of questions arose from this finding. What is the mechanism behind the reduced incretin effect? Is the secretion or insulinotropic action of GIP and GLP-1 at fault? Are defects in the enteroinsular axis (the signaling system between the gut, from where incretin hormones are secreted, and the endocrine pancreas, the main target tissue that incretin hormones act on) important for the development and/or progression of type 2 diabetes? Can the pathophysiological characterization of the incretin system in type 2 diabetes provide clues for the development of new approaches for the treatment of this metabolic disease?

A severe impairment in the insulinotropic activity of GIP in type 2 diabetes explains the reduced incretin effect. A large cross-sectional study by Toft-Nielsen and colleagues comparing GLP-1 responses after meal stimulation suggested a reduced release of GLP-1 in patients with type 2 diabetes and, to a lesser extent, impaired glucose tolerance (“prediabetes”). This widely quoted study was sometimes interpreted to indicate a progressive loss in the capacity of GLP-1 secretion in the natural history of type 2 diabetes, starting from normal secretion as long as glucose tolerance was normal with slight impairments when IGT develops, with a further deterioration after the diagnosis of type 2 diabetes and little residual capacity for GLP-1 secretion when the condition has progressed.

The logical consequence was to replace a missing hormone by advocating incretin-based antidiabetic agents (GLP-1 receptor agonists or dipeptidyl peptidase-4 inhibitors). However, not all studies that have compared the secretion of GLP-1 in patients with type 2 diabetes and in matched healthy people come to the same conclusions. A recent meta-analysis suggested no uniform reduction in L-cell secretion between healthy and type 2 diabetic patients, but a large interindividual variation, in part determined by age, obesity and plasma levels of glucagon and free fatty acids. In nondiabetics, the amount of GIP and GLP-1 secreted is significantly correlated to the incretin effect in quantitative terms. Thus, a low secretion of GLP-1 may determine a reduced incretin effect on an individual level, but does not explain the reduced incretin effect in patients with type 2 diabetes by and large.

If secretion is not the culprit, is there any peculiarity regarding the action of incretin hormones in type 2 diabetes? As originally described using GIP of the porcine amino acid sequence, and later confirmed using synthetic human GIP, the endocrine pancreas shows very little secretory response, even if exposed to supraphysiological concentrations of GIP. This inability to respond to GIP appears to be acquired, since populations at high risk for developing type 2 diabetes do not display a similar defect. Basically, the response to GIP seems to be normal in any form of prediabetes (first-degree relatives, patients recovering from gestational diabetes, etc.), but after diagnosis (ie, with a fasting glucose ≥126 mg/dL), the incretin effect is reduced or lost, as is the ability to respond to exogenous GIP.

Most likely, the inability to elicit insulin secretory responses with GIP, even at hyperglycemia, is explained by a generalized impairment in beta-cell secretory capacity, as is typical for type 2 diabetes, no matter which stimulus is looked at (hyperglycemia, amino acids, sulfonylureas, etc).

Source: Michael A. Nauck, MD, PhD

Furthermore, rodent studies have suggested a down-regulation of the GIP receptor by chronic hyperglycemia. The fact that this defect becomes apparent when glucose concentrations rise above the normal level has raised the question of whether this phenomenon is reversible by glucose normalization. A recent study by Højberg and colleagues suggested that this may be the case. However, although the insulin secretory responses to GIP and GLP-1 were significantly improved, normalization was not achieved after improved glucose control.

Abnormalities in the incretin system accompany the development of type 2 diabetes and may contribute to the velocity of progression. Figure 1 depicts the natural history of developing type 2 diabetes and also the progression of the disease after the diagnosis has been made. Changes in insulin secretory capacity, based on homeostasis model assessment (HOMA) estimation of beta-cell function, and insulin sensitivity preceding the diagnosis were taken from a recent analysis by Tabak and colleagues. The development after the diagnosis of diabetes was based on analyses from the UKPDS and ADOPT study.

Regarding the secretion of GIP and GLP-1, we refer to our recent review indicating no general abnormalities in K-cell (GIP) and L-cell (GLP-1) secretion associated with a diagnosis of type 2 diabetes. The fact that in none of the studies examining prediabetic populations, insulinotropic GIP effectiveness was impaired, but after the diagnosis, uniformly, a severe inability to respond to GIP with secreting insulin was documented, is the basis for assuming a substantial drop in beta-cell responsiveness to GIP around the time of diagnosis, with no further changes afterward (Figure 1). In a recent review, we have explained reasons to assume that this inability to secrete insulin in response to GIP stimulation goes along with a general impairment of beta-cell function, which is demonstrable with most other secretagogues as well.

Whether this inability of the endocrine pancreas to respond to GIP contributes to the natural history of type 2 diabetes can only be evaluated by quantitative considerations. If a mechanism to stimulate insulin secretion after meals that normally contributes two-thirds of the overall secretory responses is at fault, this almost certainly has the effect to accelerate the progression of type 2 diabetes because without the additional incretin stimulus, overall insulin secretion should be further impaired.

In the case of GLP-1, the insulinotropic activity is somewhat reduced after the diagnosis of type 2 diabetes, and even worse under the condition of uncontrolled hyperglycemia compared with healthy controls. High pharmacological doses of GLP-1, nevertheless, have the potential to raise insulin concentrations and to suppress glucagon secretion, with the overall result of normalizing glucose concentrations in the fasting state and after meals over a wide range of patients with type 2 diabetes, ranging from those treatable with lifestyle modification (“diet and exercise”) to those requiring insulin treatments.

Thus, the “resistance” to GIP of the type 2 diabetic beta-cell can be overcome by a compensatory exposure to high concentrations of the incretin hormone, GLP-1. GLP-1 itself appears to be less important than GIP for postprandial glucose control in healthy people and does not seem to be involved in the pathogenesis of type 2 diabetes. However, because of its preserved efficacy in type 2 diabetes, GLP-1 is an effective agent to treat hyperglycemia in type 2 diabetic patients, with the added benefits of inducing weight loss and avoiding hypoglycemia.

Michael A. Nauck, MD, PhD, and Irfan Vardarli, MD, are from Diabeteszenstrum Bad Lauterberg. Juris J. Meier, MD, is from St. Josef-Hospital, Ruhr-University of Bochum, Germany.

Disclosures: Drs. Meier, Nauck and Vardali report no relevant financial disclosures.

For more information:

• Creutzfeldt W. Diabetologia. 1979;16:75-85.
• Creutzfeldt W. Diabetes/Metab Rev. 1992;8:149-177.
• Drucker DJ. Lancet. 2006;368:1696-1705.
• Fasching P. Metabolism. 1994;43:385-389.
• Højberg PV. Diabetologia. 2008;51:632-640.
• Kahn SE. N Engl J Med. 2006;355:2427-2443.
• Krarup T. Metabolism. 1988;36:677-682.
• Kjems LL. Diabetes. 2003;52:380-386.
• Knop FK. Diabetes. 2007;56:1951-1959.
• Meier JJ. Diabetes. 2010;59:1117-1125.
• Nauck MA. Diabetes. 2004;53(Suppl. 3):S190-S196.
• Nauck MA. Diabetologia. 1986;29:46-54.
• Nauck MA. Diabetologia. 1993;36:741-744.
• Nauck MA. Diabetologia. 1999;42:373-379.
• Nauck MA. Diabetologia. 2011;54:10-18.
• Nauck MA. J Clin Invest. 1993;91:301-307.
• Nauck MA. Regul Pept. 2004;122:209-217.
• Nauck MA. Regul Pept. 2005;124(Suppl.):135-148.
• Perley MJ. J Clin Invest. 1967;46:1954-1962.
• Rachman J. Diabetologia. 1997;40:205-211.
• Tabak AG. Lancet. 2009;373:2215-2221.
• Toft-Nielsen MB. J Clin Endocrinol Metab. 2001;86:3717-3723.
• Turner R. Ann Int Med.1996;124:136-145.
• Vilsbøll T. Diabetologia. 2002;45:1111-1119.
• Willms B. J Clin Endocrinol Metab. 1996;81:327-332.