Integration of hormonal effects (insulin, glucagon and leptin) on key metabolic pathways through a computational perfused fed and fasted rat liver model

Abstract

A mechanistic model of the perfused rat liver was developed, integrating the hormonal regulation by insulin, glucagon, and leptin; under various nutritional states. This work is based on the recent model of Chalhoub E. et al. and Fayad R. et al., which included the detailed metabolic pathway, compartmentation between cytosol and mitochndria, shuttling mechanisms, and perfusion of different gluconeogenic substrates. The model was enhanced by including the glycogen metabolism, through the addition of GS and GP kinetic rates. It also included the activation and inhibitory effects of insulin, glucagon, and leptin, or a combination of them; on several key metabolic enzymes. The kinetic rate expressions are developed based on Michaelis-Menten, accounting for Haldane relation for reversible reactions, and modulation by ADP/ATP and/or NADH/NAD+. The regulatory effects of hormones were modeled based on a combination of Michaelis-Menten and Hill equations. Under fasted state, the model was fitted to several experiments of rat livers perfused with alanine in combination with insulin and leptin. In addition, during the fed state, the model was fitted to perfused rat liver experimental data of glucagon or glucagaon in combination with leptin or insulin; with no gluconeogenic precursors. The simulated model results are found in good agreement with experimental ones. The fasted liver model, under the perfusion of alanine with insulin, portrays a decrease in glucose production, lactate formation and urea synthesis by 32.3%, 49.2% and 78% respectively. Under perfusion of alanine with leptin, a decrease of 37.5, 23.571%, 40.09% was achieved for the rates of glucose, urea and lactate respectively. The simulation of glucagon in the fed rat liver model; in the absence of gluconeogenic precursors, promoted a rapid increase in hepatic glycogenolysis and glucose production by a rate of 55-52.35% and 90.6% respectively. Computational in silico models complement in vitro and in vivo tests to minimize the need for animal testing and decrease the cost and time for experimental procedures, in addition to the unique advantage of testing drugs and chemicals before they are synthesized. The model paves a way to study and analyse the distribution of key-regulatory enzymes to the alteration of physiological hormonal stimuli (such as insulin, leptin and glucagon) which allows for quantifying the contribution of these hormones in the liver. The future of computational hepatic metabolism modeling holds exciting opportunities as one of the ambitious aims of researchers is to have 90% of the human body computationally modeled by 2038.