A physiologically based in silico kinetic model predicting plasma cholesterol concentrations in humans

Abstract

This article is available online at http://www.jlr.org these cholesterol modeling studies present models that focus on LDL cholesterol (LDL-C) metabolism in plasma ( 4-6 ) or on cellular cholesterol metabolism ( 7 ). These models do not represent all relevant components of whole body cholesterol homeostasis because they lack reactions such as cholesterol absorption and biosynthesis in organs, which are the reactions targeted by important cholesterol-lowering drugs, such as statins. This implies that the models cannot fully explain how relevant plasma cholesterol-associated biomarkers are infl uenced by these drug interventions.We have, therefore, developed an in silico physiologically based kinetic (PBK) model for plasma cholesterol in the mouse that includes all relevant reactions and that correctly predicts the plasma cholesterol levels of a large variety of mouse strains with gene knockouts related to cholesterol metabolism ( 11 ). The model, of which the structure is given in Fig. 1 , is able to predict HDL cholesterol (HDL-C), non-HDL cholesterol (non-HDL-C), and total plasma cholesterol (TC) concentrations ( 12 ) as well as the intra-organ pools representing hepatic free cholesterol (Liv-FC), peripheral cholesterol (Per-C), intestinal cholesterol ester (Int-CE), hepatic cholesterol ester (Liv-CE), and intestinal free cholesterol (Int-FC) in the mouse. A similar model for humans will be of considerable value in predicting effects of drugs and genetic variations on plasma cholesterol concentrations. Therefore, the aim of this work is to adapt our model to a human version.Turnover of cholesterol in humans is quantitatively and qualitatively different from that in mice ( 13 ). Qualitative difference resides mainly in the protein cholesteryl ester In silico modeling has proven to be a useful tool in biology because it allows the study of interspecies variation and regulation of homeostasis and allows the integration of information from various sources ( 1-3 ). There are several modeling efforts on cholesterol ( 4-7 ), an important biomarker for the risk for cardiovascular events ( 8-10 ). Most of Manuscript received 4 September 2012 and in revised form 28 September 2012. Published, JLR Papers in Press, September 29, 2012 DOI 10.1194 A physiologically based in silico kinetic model predicting plasma cholesterol concentrations in humans Abbreviations: C, cholesterol (sum of FC and CE); CE, cholesterol ester; CETP, cholesterol ester transfer protein; FC, free cholesterol; FH, familial hypercholesterolemia; GWAS, genome-wide association study; PBK, physiologically based kinetic; SC, sensitivity coeffi cient; SLOS, Smith-Lemli-Opitz syndrome; TC, total plasma cholesterol (sum of HDL-C and non-HDL-C).

Evaluating computational models of cholesterol metabolism

“…This would also allow for evaluation of the RCT in its more common definition, namely as the cholesterol transport from periphery to liver. For the whole body cholesterol homeostasis models, the model of van de Pas et al passed [74,75]. The relative simplicity is perhaps the most intriguing aspect of this model.…”

“…If the change predicted by the model was consistent with experimental findings in the respective knockout mice or homozygous deletions in humans, the model was considered valid [73,74]. The effect of LDL-receptor disruption was also evaluated and model simulations correctly predicted an increase in total plasma cholesterol.…”

“…Model topology of van de Pas et al [73,74]: Cholesterol is produced in the intestine liver and periphery. Cholesterol further enters the system through dietary intake.…”

“…Parameters were calculated by solving the steady state solutions, using steady state fluxes obtained from the literature [73]. This model was later adapted to describe cholesterol metabolism in man [74]. This required adding a reaction that accounted for the action of CETP, an enzyme that mice lack.…”

Efficient Generation and Selection of Virtual Populations in Quantitative Systems Pharmacology Models

“…17 We chose this model because we could use publicly available data from the NHANES database 16 to establish the empirical multivariate distribution for LDL cholesterol, HDL cholesterol, and total cholesterol (LDL c , HDL c , and TC, respectively). Note that the distribution of these variables is well approximated by a multivariate log-normal distribution (Supplementary Figure 1).…”