A computational study of liver X receptor induced hepatic steatosis

Background The development of hepatic steatosis is related to increased de novo lipogenesis, which is believed to be highly regulated by the liver X receptor (LXR). Administration of synthetic LXR agonist T0901317 in C57BL/6J mice resulted in development of hepatic steatosis and secretion of enlarged, atherogenic very low density lipoproteins. LXR activates a wide range of signaling pathways involved in sterol and fatty acid metabolism, from which the exact kinetics and molecular mechanisms remain to be determined. Aim/Objective Our aim was to identify kinetic changes in the complex network of hepatic metabolic pathways upon LXR activation. These collectively result in a change of metabolic state of the liver, hereby indicating possible mechanisms responsible for development of hepatic steatosis and production of enlarged lipoproteins. Methods To address the kinetics of different hepatic responses, a computational model was constructed including reactions representing lipogenesis, cholesterol synthesis, as well as lipoprotein assembly, secretion and remodeling. Different datasets from wild-type C57BL/6J mice were used for model parameterization. To explore the complete set of potential system behaviors, a large-scale search in parameter space was carried out to locate regions of high likelihood. Parameter sets were subsequently optimized by applying a nonlinear least squares parameter estimation method. Data from LXR activated C57BL/6J mice were used to re-optimize resulting parameter sets (presenting wild-type C57BL/6J mice) in a step-wise manner to describe the diseased mouse phenotype. Results We were able to quantitatively integrate data of different experiments into a consistent model and to identify a collection of parameter sets describing the wild-type C57BL/6J mouse. Furthermore, necessary parameter modifications were identified to describe the LXR activated mouse. It appeared that several parameters changed significantly and consistently from the healthy to the diseased phenotype.