Simulation and validation of modelled sphingolipid metabolism in Saccharomyces cerevisiae

• Published in: Nature 2005-01, Vol.433 (7024), p.425-430
• Main Authors: Alvarez-Vasquez, F, Sims, K.J, Cowart, L.A, Okamoto, Y, Voit, E.O, Hannun, Y.A
• Format: Article
• Language: English
• Subjects: Biochemistry; Biological and medical sciences; Computer Simulation; Fatty Acids - metabolism; Fundamental and applied biological sciences. Psychology; General aspects; Humanities and Social Sciences; Inositol - metabolism; Inositol - pharmacology; Kinetics; letter; lipid metabolism; Mathematical models; Mathematics in biology. Statistical analysis. Models. Metrology. Data processing in biology (general aspects); Metabolism; microbial physiology; model validation; Models, Biological; multidisciplinary; Palmitic Acid - metabolism; Palmitoyl Coenzyme A - metabolism; Phospholipids - metabolism; Reproducibility of Results; Saccharomyces cerevisiae; Saccharomyces cerevisiae - drug effects; Saccharomyces cerevisiae - enzymology; Saccharomyces cerevisiae - genetics; Saccharomyces cerevisiae - metabolism; Science; Science (multidisciplinary); simulation models; sphingolipids; Sphingolipids - metabolism; Yeast; Yeasts
• ISSN: 0028-0836; 1476-4687
• DOI: 10.1038/nature03232

Mathematical models have become a necessary tool for organizing the rapidly increasing amounts of large-scale data on biochemical pathways and for advanced evaluation of their structure and regulation. Most of these models have addressed specific pathways using either stoichiometric or flux-balance analysis, or fully kinetic Michaelis-Menten representations, metabolic control analysis, or biochemical systems theory. So far, the predictions of kinetic models have rarely been tested using direct experimentation. Here, we validate experimentally a biochemical systems theoretical model of sphingolipid metabolism in yeast. Simulations of metabolic fluxes, enzyme deletion and the effects of inositol (a key regulator of phospholipid metabolism) led to predictions that show significant concordance with experimental results generated post hoc. The model also allowed the simulation of the effects of acute perturbations in fatty-acid precursors of sphingolipids, a situation that is not amenable to direct experimentation. The results demonstrate that modelling now allows testable predictions as well as the design and evaluation of hypothetical 'thought experiments' that may generate new metabolomic approaches.