AimsIn vitro, beta cells immediately secrete stored but readily releasable insulin in response to a rise of glucose. During a prolonged insulin response, this is followed by newly synthesized insulin. Our aim was to develop an in vivo test to determine the ratio between readily available and newly synthesized insulin after a stimulus in humans by labelling newly synthesized insulin.MethodsA stable isotope tracer of 1.0 g 13C leucine with C-peptide as target peptide was administered 45 min prior to 75 g glucose load of a frequently blood sampled 210-min oral glucose tolerance test (OGTT). Our OGTT also encompassed collection of urine, which has a high content of C-peptide. Prior, the optimal conditions under which the tracer 13C leucine was administered for enrichment of (pre) proinsulin were established. Also, techniques to obtain urinary C-peptide under highly purified circumstances were set up. Our main outcome measure was the stable isotope enrichment of de novo C-peptide, which we related to early plasma insulin and glucose AUC. Twelve healthy Caucasian individuals (M4F8, age 41.8 ± 2.3, BMI 28.3 ± 1.7) with normal glucose tolerance underwent our OGTT.ResultsWe found that during a 75-g OGTT, newly synthesized insulin contributed approximately 20 % of total insulin secretion. The pattern of isotope enrichment obtained by collecting multiple urine voids was suggestive that the newly synthesized insulin contributes to the late phase of insulin secretion. De novo C-peptide correlated negatively with both early plasma insulin AUC (r = −0.629, P = 0.028) and early plasma glucose AUC (r = −0.605, P = 0.037).ConclusionsWith stable isotope technique added to OGTT, we were able to measure newly synthesized insulin in healthy individuals. This new technique holds the promise that it is feasible to develop a direct in vivo beta cell function test.Electronic supplementary materialThe online version of this article (doi:10.1007/s00592-016-0896-3) contains supplementary material, which is available to authorized users.