In regard to the potential pathophysiological contribution of defective incretin effect, only a few studies have explored this important question, and the results have been controversial. For example, on the one hand, it has been shown that the insulin stimulatory effect of GIP is impaired in first-degree relatives of type 2 diabetic patients (9), which would suggest a pathophysiological involvement. On the other hand, it has also been shown that the insulinotropic effect of GIP is normal in women with a history of gestational diabetes (10), which instead suggests that the defect is secondary to developed diabetes. In regard to the question of a specific defect in incretin hormone effect in type 2 diabetes vs. a generalized β-cell defect, no studies have directly focused on this issue.

In the present issue of the JCEM, a novel study by Hansen et al. (11) has addressed these two fundamental questions by a novel and interesting approach. The authors examined the insulin secretory responses to iv GIP and GLP-1 (given in physiological concentrations) during a hyperglycemic clamp (10 mmol/liter) in healthy subjects before and after experimental induction of insulin resistance. In insulin resistance, insulin secretion is up-regulated, and in healthy subjects, the disposition index (i.e. insulin sensitivity × insulin secretion) remains normal (12). The investigators compared the up-regulated insulin secretion after glucose vs. incretin hormones. If there is a defective up-regulation of insulin secretion to insulin resistance in response to the incretin hormones, but not to glucose, in these healthy subjects, a conclusion can be drawn that defective insulinotropic actions of GIP and GLP-1 are secondary to insulin resistance. This in turn would suggest that the defective incretin effect in type 2 diabetes is not a primary condition for diabetes development but is secondary to the metabolic changes in insulin resistance. To make that case, great care was taken to ensure that the subjects were indeed completely healthy: they had no disease, no treatment, and no family history of diabetes, and they had normal glucose tolerance on an oral glucose tolerance test and a normal C-peptide response to iv glucagon. Insulin resistance was induced by a combination of sedentary lifestyle (rest for at least 8 h/d and no strenuous exercise), high-calorie diet (130% of daily recommended food intake), and daily administration of prednisone (37.5 mg/d for 12 d).

Previously, it has been demonstrated that a short-term high-calorie diet (13) or steroid administration (14) results in insulin resistance in humans with an adaptively increased insulin secretion; incretin hormone action after these challenges has, however, not been studied before. The novel approach taken in the study by Hansen et al. (11) used the combination of these challenges together with forced sedentary lifestyle with restriction of physical activity. A strength of this approach was that insulin resistance was induced already within the 12-d study period, which was verified by fasting hyperinsulinemia and the Matsuda index.

Before and after the induction of insulin resistance, the insulin secretory responses to glucose alone, glucose with GIP, and glucose with GLP-1 were studied. It was found that the insulin response to glucose alone was augmented by insulin resistance, which is a well-known adaptive response that is of importance for avoiding hyperglycemia in the setting of acute insulin resistance (12). The up-regulation of glucose-stimulated insulin secretion perfectly matched the degree of insulin resistance, as indicated by a normal disposition index. The main and important finding was, however, that the up-regulation of insulin secretion in response to the incretin hormones was impaired when compared with the up-regulation by glucose and glucagon. Hence, after GIP and GLP-1, an insufficient up-regulation of insulin secretion was evident in insulin resistance.

These findings have two important implications: 1) the impaired incretin effect does not seem to be a primary phenomenon in diabetes development but is rather a consequence of the dysglycemia in insulin resistance; and 2) the impaired incretin effect is not a generalized β-cell impairment but rather a specific dysfunction related to the effects of the incretin hormones. The results cannot, however, rule of that the incretin hormones, although not having a primary role for type 2 diabetes, still may contribute to the development of islet dysfunction that precedes the onset of the disease. The impaired β-cell action of the incretins induced by insulin resistance may thus still be a factor contributing to the progressive decline in β-cell function. The study also has an additional important finding—that it was mainly the late (10–120 min) phase of incretin hormone-induced insulin secretion that was impaired by insulin resistance.

Thus, several important findings evolve from this interventional study. Because this study group was small and the challenge could not distinguish between different aspects, the study has some weaknesses that require more study. Also, mechanistic explanations of the findings need to be explored, which might more efficiently be performed using animal models. These issues include GIP and GLP-1 receptor expression in insulin resistance, potential perturbations in post-receptor signaling during insulin resistance, and whether the defect observed is specific for the late phase of glucose-stimulated insulin secretion or whether it also exists for other qualities of β-cell function like the maximal insulin response.

The provocative study by Hansen et al. (11) therefore presents interesting data on the contribution of the incretin system to the pathophysiology of type 2 diabetes, presents evidence that defective incretin hormone action of β-cell function is not a primary pathophysiological trait, and opens the opportunity for further studies that may be of relevance for incretin-based therapy for the disease. The study also illustrates how powerful small interventional and mechanistic studies may be.

 Next SectionAcknowledgments
The work was supported by Faculty of Medicine, Lund University, Lund, Sweden.

Disclosure Summary: The author has nothing to declare.