A closed-loop multi-level model of glucose homeostasis

• DOI: 10.1371/journal.pone.0190627
• Publication Date: 2018-02-08T18:31:13Z
• Authors: Cansu Uluseker; Giulia Simoni; Luca Marchetti; Marco Dauriz; Alice Matone; Corrado Priami
• Keywords: glucose homeostasis; hierarchical modeling; closed-loop; diabetes; 0301 basic medicine; Science; glucose homeostasis, hierarchical modeling, closed-loop, diabetes; Q; R; Models, Biological; Modeling and simulation, bioinformatics, computational systems biology; 03 medical and health sciences; Glucose; Diabetes Mellitus, Type 2; Adipocytes; Medicine; Homeostasis; Humans; Insulin; Computer Simulation; Research Article; Signal Transduction

The pathophysiologic processes underlying the regulation of glucose homeostasis are considerably complex at both cellular and systemic level. A comprehensive and structured specification for the several layers of abstraction of glucose metabolism is often elusive, an issue currently solvable with the hierarchical description provided by multi-level models. In this study we propose a multi-level closed-loop model of whole-body glucose homeostasis, coupled with the molecular specifications of the insulin signaling cascade in adipocytes, under the experimental conditions of normal glucose regulation and type 2 diabetes.The ordinary differential equations of the model, describing the dynamics of glucose and key regulatory hormones and their reciprocal interactions among gut, liver, muscle and adipose tissue, were designed for being embedded in a modular, hierarchical structure. The closed-loop model structure allowed self-sustained simulations to represent an ideal in silico subject that adjusts its own metabolism to the fasting and feeding states, depending on the hormonal context and invariant to circadian fluctuations. The cellular level of the model provided a seamless dynamic description of the molecular mechanisms downstream the insulin receptor in the adipocytes by accounting for variations in the surrounding metabolic context.The combination of a multi-level and closed-loop modeling approach provided a fair dynamic description of the core determinants of glucose homeostasis at both cellular and systemic scales. This model architecture is intrinsically open to incorporate supplementary layers of specifications describing further individual components influencing glucose metabolism.