Glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) are two peptides that function to promote insulin secretion. Dipeptidyl peptidase-4 (DPP-4) inhibitors increase the bioavailability of both GLP-1 and GIP but the dogma continues to be that it is the increase in GLP-1 that contributes to the improved glucose homeostasis. We have previously demonstrated that pancreatic rather than intestinal GLP-1 is necessary for improvements in glucose homeostasis in mice. Therefore, we hypothesise that a combination of pancreatic GLP-1 and GIP is necessary for the full effect of DPP-4 inhibitors on glucose homeostasis.

We have genetically engineered mouse lines in which the preproglucagon gene (Gcg) is absent in the entire body (GcgRA^ΔNull) or is expressed exclusively in the intestine (GcgRA^ΔVilCre) or pancreas and duodenum (GcgRA^ΔPDX1Cre). These mice were used to examine oral glucose tolerance and GLP-1 and GIP responses to a DPP-4 inhibitor alone, or in combination with incretin receptor antagonists.

Administration of the DPP-4 inhibitor, linagliptin, improved glucose tolerance in GcgRA^ΔNull mice and control littermates and in GcgRA^ΔVilCre and GcgRA^ΔPDX1Cre mice. The potent GLP-1 receptor antagonist, exendin-[9–39] (Ex9), blunted improvements in glucose tolerance in linagliptin-treated control mice and in GcgRA^ΔPDX1Cre mice. Ex9 had no effect on glucose tolerance in linagliptin-treated GcgRA^ΔNull or in GcgRA^ΔVilCre mice. In addition to GLP-1, linagliptin also increased postprandial plasma levels of GIP to a similar degree in all genotypes. When linagliptin was co-administered with a GIP-antagonising antibody, the impact of linagliptin was partially blunted in wild-type mice and was fully blocked in GcgRA^ΔNull mice.

Taken together, these data suggest that increases in pancreatic GLP-1 and GIP are necessary for the full effect of DPP-4 inhibitors on glucose tolerance.