GLUCAGON-LIKE PEPTIDE-1 INCREASES MASS BUT NOT FREQUENCY OR ORDERLINESS OF PULSATILE INSULIN-SECRETION

Glucagon-like peptide 1 (GLP-1) is a peptide hormone that is released from the gut after luminal stimulation. The hormone is a potent insuli n secretagogue and is a potential novel pharmaceutical adjuvant in the treatment of NIDDM. Insulin is secreted as a series of punctuated sec retory bursts superimposed on a basal insulin release. Recently, the c ontribution of these secretory bursts to overall insulin secretion has been evaluated, and studies using catheterization across the pancreas in a canine model and studies using deconvolution in humans have reve aled that the majority of insulin is released during these secretory b ursts. Moreover, the main regulation of insulin secretion is through p erturbation of mass and frequency of these secretory bursts. The mode of delivery of insulin into the circulation seems important for insuli n action, and it is therefore important to know the impact of a potent ial therapeutic insulin secretagogue on the mode of insulin secretion. To assess the effects of GLP-1 on the mass, frequency, amplitude, and overall contribution of pulsatile insulin secretion, we used a recent ly validated deconvolution model to examine these variables before and during infusion of GLP-1 in eight healthy men (age 28 +/- 2 years; BM I 24 +/- 2 kg/m(2)). At a constant glucose infusion (2.5 mg . kg(-1) . min(-1)), near-steady state was reached at 75 min, and sampling was p erformed every minute at t = 75-115 and 145-185 min. At t = 115 min, a n infusion of saline or GLP-1 (50 pmol . kg(-1) . min(-1)) was given. The regularity of insulin secretion was measured by approximate entrop y, a recently developed mathematical statistic, applied herein to asse ss the regularity in a hormone concentration time series. After GLP-1 infusion, there was an abrupt increase in the peripheral concentration s of serum C-peptide (696 +/- 65 vs. 1,538 +/- 165 pmol/l) and insulin (49 +/- 8 vs. 138 +/- 21 pmol/l) concentrations. This increase was ma inly due to an increase in the pulsatile component of insulin secretio n that was achieved by a fourfold increase in secretory burst mass (28 .2 +/- 4.4 vs. 100.1 +/- 15.8 pmol . l(-1) . pulse(-1); P < 0.001), an d amplitude (12.7 +/- 2.2 vs. 4.3 +/- 7.7 pmol . l(-1) . min(-1); P < 0.002), whereas the secretory burst frequency was not affected by GLP- 1 (11.5 +/- 0.7 vs. 12.6 +/- 0.6 pulses/h; P = 0.4). As a consequence, the detected contribution of pulsatile to overall insulin secretion w as increased from 56 +/- 4 to 77 +/- 4% (P < 0.005). The orderliness o f the insulin release process was not deteriorated by short-term GLP-1 infusion as assessed by approximate entropy (1.19 +/- 0.04 vs. 1.18 /- 0.04; P = 0.7).