AIMS/HYPOTHESIS: In the isolated perfused pancreas, gastric inhibitory polypeptide (GIP) has been shown to enhance glucagon secretion at basal glucose concentrations, but in healthy humans no glucagonotropic effect of GIP has yet been reported. Therefore, we studied the effect of GIP on glucagon secretion under normoglycaemic conditions.

METHODS: Ten healthy subjects (9 men, 1 woman; age 33+/-11; BMI 26.8+/-2.2 kg/m(2)) received three different doses of intravenous GIP (7, 20, and 60 pmol/kg body weight) and placebo. Venous blood samples were drawn over 30 min for glucagon and GIP concentrations (specific radioimmunoassays). In addition, 31 healthy subjects (16 men, 15 women; 42+/-11 years; BMI 24.4+/-2.7 kg/m(2)) were studied with 20 pmol GIP/kg. Statistics were done with RM-ANOVA and Duncan's post hoc tests.

RESULTS: Gastric inhibitory polypeptide dose-dependently stimulated glucagon secretion ( p=0.019) with a maximal increment after 10 min. Incremental glucagon concentrations (Delta(10-0 min)) were 0.1+/-0.7, 1.4+/-0.5, 2.4+/-0.5, and 3.4+/-0.8 pmol/l (for placebo and for 7, 20, and 60 pmol GIP/kg, respectively; p=0.017). After the injection of 20 pmol GIP/kg b.w. in 31 healthy subjects, glucagon concentrations increased over the baseline from 7.5+/-0.5 to 9.3+/-0.7 pmol/l ( p=0.0082).

CONCLUSIONS/INTERPRETATION: Glucagon secretion is dose-dependently stimulated by GIP at basal glucose concentrations. The absence of a glucagonotropic GIP effect in previous studies could be due to the hyperglycaemic conditions used in these experiments. Our results underline differences between GIP and the glucagonostatic incretin GLP-1.