Background: Characterization of beta-cell function in humans is essential for identifying genetic defects involved in abnormal insulin secretion and the pathogenesis of type 2 diabetes.
Materials and methods: We designed a novel test assessing plasma insulin and C-peptide in response to 3 different secretagogues. Seven lean, healthy volunteers twice underwent a 200 min hyperglycaemic clamp (10 mmol L-1) with administration of GLP-1 (1.5 pmol. kg-1. min-1) starting at 120 min and an arginine bolus at 180 min. We determined glucose-induced first and second-phase insulin secretion, GLP-1-stimulated insulin secretion, arginine-stimulated insulin response (increase above prestimulus, DeltaIarg) and the maximal, i. e. highest absolute, insulin concentration (Imax). Insulin sensitivity was assessed during second-phase hyperglycaemia. On a third occasion 6 subjects additionally received an arginine bolus at > 25 mM blood glucose, a test hitherto claimed to provoke maximal insulin secretion.
Results: Insulin levels increased from 46 +/- 11 pM to 566 +/- 202 pM at 120 min, to 5104 +/- 1179 pM at 180 min and to maximally 8361 +/- 1368 pM after arginine (all P < 0.001). The within subject coefficients of variation of the different secretion parameters ranged from 10 +/- 3% to 16 +/- 6%. Except for second-phase which failed to correlate significantly with DeltaIarg (r = 0.52, P = 0.23) and Imax (r = 0.75, P = 0.053) all phases of insulin secretion correlated with one another. The insulin concentration after the arginine bolus at > 25 mM glucose (n = 6) was 2773 +/- 855 pM vs. 7562 +/- 1168 pM for Imax (P = 0.003).
Conclusion: This novel insulin secretion test elicits a distinct pattern of plasma insulin concentrations in response to the secretagogues glucose, GLP-1 and arginine and is highly reproducible and can be used for differential characterization of islet function.