Abbreviations: NIDDM, Non-insulin-dependent diabetes mellitus; GIP, glucose-dependent insulinotropic polypeptide; Glp-1, glucagon-like peptide-1. porcine natural or synthetic or more recently, human GIP [15, 16, 21]. Indeed, in parallel studies of GLP-1 and GIP, the full efficacy of GLP-1 and the lack of effect of GIP has been documented in the same patients [15, 16]. Thus, the lack of effect of GIP in NIDDM could very well explain the absent incretin effect (for small glucose loads) in these patients. Conceivably, an impaired GLP-1 response to glucose as observed in some patients with NIDDM [14] and an increased activity of the GLP-1 degrading enzyme, dipeptidyl-peptidase IV [22] may contribute to an impaired incretin response in patients with NIDDM, but could hardly explain its complete absence. Why, then, is GIP ineffective while GLP-1 retains sufficient activity to be of therapeutic value? Both hormones exert their effect on the beta cell by interaction with a specific receptor. The two receptors, including the human forms, have been cloned and have been demonstrated to belong 1 to the family of seven transmembrane spanning G-protein coupling receptors [23, 24]. Furthermore, both couple to adenylate cyclase [19, 24, 25]. Given the ineffectiveness of GIP, but not GLP-1, in diabetic beta cells, we speculated that, by identifying differences between the intracellular actions of the two hormones, one would also reveal the site of the defect in the diabetic beta cell. Detailed attempts were, therefore, made to find such a difference. In a series of studies in single beta cells, islet cell lines or cell lines transfected with receptors, and more recently, human beta cells isolated from fresh pancreases obtained from multiorgan donors (unpublished data), we were able to study, in parallel, the effects of the two hormones using sophisticated modern techniques, including patch-clamping, direct determination of intracellular ion concentrations, transmembrane ion currents, membrane potentials, and hormone secretion by capacitance measurements [26–28]. In these studies, all effects of the two hormones were identical: they generated identical changes of membrane potential, intracellular calcium responses, membrane currents, and cAMP responses [26–28]. Moreover, the identical insulin responses to either peptide were both abolished by inhibitors of the actions of protein kinase A, activated by cAMP; similarly, the responses could be entirely mimicked by adenylate cyclase activation using forskolin, which bypasses the membrane receptors and their coupling to the enzyme. In addition, the insulinotropic effects of both peptides are similarly potentiated by sulphonylurea treatment and similarly augmented in the genetically obese (fa/fa) Zucker rat [29–31]. We conclude, therefore, that the two receptors activate the same intracellular machinery. The present assessment is that the incretin effect (for small loads of glucose which mainly release GIP) is absent in NIDDM, but the secretion of GIP is normal (or increased); the intracellular machinery of the beta cells of the diabetic pancreas functions normally when tested with GLP-1, the mechanism of action of which is identical to that of GIP. Conclusion? The diabetic beta-cell either does not express or, if it does, expresses a defective receptor for GIP. Indeed, two missense mutations in the coding region of the GIP receptor gene were recently described, one of which exhibited markedly decreased responsiveness to GIP in functional studies [32]. However, none of the mutations were associated with NIDDM in a Japanese study group. On the other hand, the many studies in Caucasians showing that GIP does not stimulate insulin …