A Unifying Organ Model of Pancreatic Insulin Secretion.

Andrea De Gaetano,Simona Panunzi,Pasquale Palumbo,Claudio Gaz

doi:10.1371/journal.pone.0142344

Abstract

The secretion of insulin by the pancreas has been the object of much attention over the past several decades. Insulin is known to be secreted by pancreatic β-cells in response to hyperglycemia: its blood concentrations however exhibit both high-frequency (period approx. 10 minutes) and low-frequency oscillations (period approx. 1.5 hours). Furthermore, characteristic insulin secretory response to challenge maneuvers have been described, such as frequency entrainment upon sinusoidal glycemic stimulation; substantial insulin peaks following minimal glucose administration; progressively strengthened insulin secretion response after repeated administration of the same amount of glucose; insulin and glucose characteristic curves after Intra-Venous administration of glucose boli in healthy and pre-diabetic subjects as well as in Type 2 Diabetes Mellitus. Previous modeling of β-cell physiology has been mainly directed to the intracellular chain of events giving rise to single-cell or cell-cluster hormone release oscillations, but the large size, long period and complex morphology of the diverse responses to whole-body glucose stimuli has not yet been coherently explained. Starting with the seminal work of Grodsky it was hypothesized that the population of pancreatic β-cells, possibly functionally aggregated in islets of Langerhans, could be viewed as a set of independent, similar, but not identical controllers (firing units) with distributed functional parameters. The present work shows how a single model based on a population of independent islet controllers can reproduce very closely a diverse array of actually observed experimental results, with the same set of working parameters. The model’s success in reproducing a diverse array of experiments implies that, in order to understand the macroscopic behaviour of the endocrine pancreas in regulating glycemia, there is no need to hypothesize intrapancreatic pacemakers, influences between different islets of Langerhans, glycolitic-induced oscillations or β-cell sensitivity to the rate of change of glycemia.

Highlights

Over the past 40 years or so, several experimenters have focused their attention on the mechanisms and modalities with which pancreatic β-cells secrete insulin in response to glycemic stimuli
The model presented here is an extension of [16] where, inspired by the seminal work of Grodsky [1], control of glucose-stimulated pancreatic insulin secretion is effected by a discrete set of independent controllers, that is, no direct control is exerted on a secretory unit either by other units or by neural or endocrine mechanisms, the only connection among the units being the common input signal represented by blood glucose concentration in in vivo situations or environment glucose concentration in in vitro experiments, as sensed by each secretory unit
In order to test the robustness of the model in reproducing this type of experiment, responses to the test of hypothetical subjects with different glucose tolerance states were simulated: a Normal Glucose Tolerance (NGT) subject, an Impaired Fasting Glucose (IFG) subject, an Impaired Glucose Tolerance (IGT) subject, an IFG+IGT subject and a Type 2 Diabetes Mellitus (T2DM) subject

Summary

Introduction

Over the past 40 years or so, several experimenters have focused their attention on the mechanisms and modalities with which pancreatic β-cells secrete insulin in response to glycemic stimuli. Other experiments have been carried out on animal models (monkeys [4, 5], dogs [6, 7], minipigs [8], rats [9]) as well as human subjects [10,11,12,13,14,15], administering variable amounts of glucose in different ways, and observing the corresponding insulin serum concentrations The results of these many experimental procedures have shown that insulin secretion, in response to glucose stimuli, exhibits a number of diverse and interesting properties, ranging from pulsatility, oscillations, entrainment to exogenous stimuli, first and second phases of release, potentiation, etc. Both in-vivo and in-vitro experimental results will be reproduced: in particular, the in-vitro experimental framework under investigation is the one detailed in the pioneering work of Grodsky [1], still considered a standard benchmark to test mathematical models aimed at accounting for the biphasic pattern of insulin release (see, e.g. the works by Bertuzzi, Salinari and Mingrone [17] and by Pedersen et al [18])

Methods

Results

Conclusion