A kinetic model of the lipid network in yeast

Lipids play a very important role in cell structure and function, as well as in the physiopathology of many diseases. Maintenance of the lipid profiles should be tightly regulated as it is very important for preserving membrane permeability, cell integrity and several other functions. Large-scale kinetic models of metabolic networks are essential in order to accurately capture and predict such behaviors of cellular systems when subject to perturbations. We have thus developed a detailed model of the lipid metabolism for the yeast S. cerevisiae, in order to identify how the stoichiometric and kinetic coupling determines lipid homeostasis and its regulation. The model encompasses 308 reactions and 212 unique metabolites, and includes the following subsystems: glycolysis, fatty acid biosynthesis and elongation, biosynthesis of phospholipids, sphingolipids, cardiolipin and sterols, triacylglycerides decomposition and the mevalonate pathway. We curated this model using thermodynamic data as well as lipidomic measurements and we used the Optimization and Risk Analysis of Complex Living Entities (ORACLE) framework to generate populations of parametrized kinetic models that are consistent with the given physiology, while satisfying the stoichiometric and thermodynamic constraints. We used these models to identify parameters (i.e. enzyme activities) that determine lipid distribution and homeostasis.

Identification of the enzymes that determine the robustness of lipid profiles in yeast

Vassily Hatzimanikatis, Howard Riezman, Georgios Savoglidis, Sofia Tsouka

Lipids play a very important role in cell structure and function, as well as in the physiopathology of many diseases. Maintenance of the lipid profiles should be tightly regulated as it is very important for preserving membrane permeability, cell integrity and several other functions. Large-scale kinetic models of metabolic networks are essential in order to accurately capture and predict such behaviors of cellular systems when subject to perturbations. We have thus developed a detailed model of the lipid metabolism for the yeast S. cerevisiae, in order to identify how the stoichiometric and kinetic coupling determines lipid homeostasis and its regulation. The model encompasses 308 reactions (of which 230 are enzymatic reactions, 32 are transport reactions and 29 are elementary reactions that contribute to biomass formation) and 212 unique metabolites, and includes the following subsystems: glycolysis, fatty acid biosynthesis and elongation, biosynthesis of phospholipids, sphingolipids, cardiolipin and sterols, triacylglycerides decomposition and the mevalonate pathway. We curated this model using thermodynamic data as well as lipidomic measurements and we used the Optimization and Risk Analysis of Complex Living Entities (ORACLE) framework to generate populations of parametrized kinetic models that are consistent with the given physiology, while satisfying the stoichiometric and thermodynamic constraints. We computed and analyzed the distributions of these models’ flux and concentration control coefficients (FCCs and CCCs, respectively), which quantify the magnitude to which a change in a system parameter (i.e. enzyme activities) will affect and control fluxes through reactions and metabolic concentrations at a representative steady state. We used these coefficients to reverse engineer changes in enzyme activities that will lead to desired phenotypes as well as to identify mutations based on lipidomic measurements.

2017

Kinetic reconstruction and analysis of sphingolipid metabolism

Vassily Hatzimanikatis, Howard Riezman, Georgios Savoglidis

Lipids are major constituents of the cell. They are responsible for major properties of the cellular membranes: hydrophobicity, selective permeability and being the scaffold of signaling proteins. Many diseases are associated with alterations in the lipid distribution in the cell and the composition of membrane domains. Metabolic syndrome, obesity, atherosclerosis, as well as Alzheimer’s, Huntington’s diseases and cancer, have an impact in the levels of lipids, with observed alterations in their concentrations compared to the healthy state. Applying computational techniques along with systematic modeling of lipid metabolism can provide insights that can guide biomedical research and develop potential strategies for prevention and cure. In the present study we developed a comprehensive model of the sphingolipid biosynthesis in the yeast Saccharomyces cerevisiae. Sphingolipids are one of the four major lipid categories, along with (glycero)phospholipids, sterols and fatty acids synthesized in the yeast S. cerevisiae. The importance of sphingolipids present in any higher eukaryote has been demonstrated in many recent studies. For this study, S. cerevisiae has been chosen as a model organism due to its high homology of cellular processes with mammalian cells. The developed model will be an essential part towards the construction of a detailed kinetic model of the whole lipidome of the cell. We first constructed a stoichiometric model that contains all the currently known reactions for the biosynthesis of ceramides and complex sphingolipids in the yeast. Additionally, we have accounted for all five reported hydroxylation states, along with the reactions synthesizing the necessary precursor metabolites from other lipid pathways (i.e. palmitate-CoA from fatty acid synthesis and phosphatidylinositol from the phospholipids metabolism). We next developed a kinetic model and we used a large number of lipidomic measurements of wild type yeast to consistently calibrate our model. Curation of the kinetic information of the model came from the comprehensive mining of references for operation of the enzymes as well as ranges of kinetic parameters from online databases. These datasets created the pool for a sampling technique that accounts for the uncertainty in the parameters of the model. This lead to a robust dynamic model, containing mass balances for all the components of the biochemical network as well as terms that accounted for the dillution of these molecules in the cell due to growth. We performed a thorough kinetic analysis of the system by examining the impact of different assumptions in enzyme operation on the levels of ceramides and complex sphingolipids (e.g. substrate competition, (un)competitive inhibition). By applying the principles of Metabolic Control Analysis (MCA) we were able to quantify the effect of enzyme activities on the lipid profiles and we identified enzymes of the biochemical reaction network as efficient targets for metabolic engineering towards a desired state. We also applied a reverse engineering method and we were able to identify enzyme perturbations responsible for an observed altered state compared to the wild type cellular lipidomic profile. Such an approach could lead to the identification of genetic mutations by imploring the information containd within metabolomic measurements. Although we demonstrate the application of the method in models of lipid metabolism it is possible to be applied to biochemical systems of various parts of metabolism and sizes creating a modular platform for kinetic analysis of cellular operations.

2014

Integrating omics data into metabolic models of lipids: a platform for the identification of gene mutations from lipidomics data

Paolo Angelino, Vassily Hatzimanikatis, Howard Riezman, Georgios Savoglidis

Kinetic models of metabolic networks are mapping the biochemical reactions taking part in a cell to a mathematical representation. They are the blueprints of the biochemical network, connecting enzymatic activities and metabolic compounds. Kinetic models have been used in systems biology to produce estimates of response of an organism to stress and to reveal potentially efficient targets for cellular engineering. Lipids are major constituents of the cell membrane. They are responsible for major properties of the cellular membranes: hydrophobicity, selective permeability and being the scaffold of signaling proteins. Many diseases are associated with alterations in the lipid distribution in the cell and the composition of membrane domains. Metabolic syndrome, obesity, atherosclerosis, as well as Alzheimer’s, Huntington’s diseases and cancer, all originate from alterations in some stage of lipid biosynthesis. Therefore, advancement of knowledge in the field of lipid metabolism will provide novel insights for further biomedical research and potential strategies for drug development. In the current study we combine lipidomics data and metabolic models of lipid metabolism in yeast Saccharomyces cerevisiae. We use the models as platform for the application of sensitivity analysis and metabolic control analysis, which involves direct and inverse sensitivity analysis. In direct sensitivity analysis we quantify how the changes in enzymatic activities, either by gene deletion or RNAi, are mapped through the metabolic pathway model into variations in lipid profiles. This analysis allows us to identify the effects of the perturbations in enzyme activities (inhibition or overexpression) on the distribution of the lipids synthesized as the end products of the network. We further extend the analysis by casting the inverse problem. With the inverse sensitivity analysis framework, we consider as input an altered lipidomic profile and we are able to identify, the responsible perturbations in enzyme activities that account for such a change. The results of the analysis have been confirmed based on lipidomics analysis of mutant yeast cells. Although we demonstrate the application of the method in models of lipid metabolism it is possible to be applied to biochemical systems of various parts of metabolism and sizes creating a modular platform for identification of mutations.

2013

Kinetic reconstruction and analysis of sphingolipid metabolism

Vassily Hatzimanikatis, Howard Riezman, Georgios Savoglidis

2014

Identification of the enzymes that determine the robustness of lipid profiles in yeast

Vassily Hatzimanikatis, Howard Riezman, Georgios Savoglidis, Sofia Tsouka

Lipids play a very important role in cell structure and function, as well as in the physiopathology of many diseases. Maintenance of the lipid profiles should be tightly regulated as it is very important for preserving membrane permeability, cell integrity and several other functions. Large-scale kinetic models of metabolic networks are essential in order to accurately capture and predict such behaviors of cellular systems when subject to perturbations. We have thus developed a detailed model of the lipid metabolism for the yeast S. cerevisiae, in order to identify how the stoichiometric and kinetic coupling determines lipid homeostasis and its regulation. The model encompasses 308 reactions (of which 230 are enzymatic reactions, 32 are transport reactions and 29 are elementary reactions that contribute to biomass formation) and 212 unique metabolites, and includes the following subsystems: glycolysis, fatty acid biosynthesis and elongation, biosynthesis of phospholipids, sphingolipids, cardiolipin and sterols, triacylglycerides decomposition and the mevalonate pathway. We curated this model using thermodynamic data as well as lipidomic measurements and we used the Optimization and Risk Analysis of Complex Living Entities (ORACLE) framework to generate populations of parametrized kinetic models that are consistent with the given physiology, while satisfying the stoichiometric and thermodynamic constraints. We computed and analyzed the distributions of these models’ flux and concentration control coefficients (FCCs and CCCs, respectively), which quantify the magnitude to which a change in a system parameter (i.e. enzyme activities) will affect and control fluxes through reactions and metabolic concentrations at a representative steady state. We used these coefficients to reverse engineer changes in enzyme activities that will lead to desired phenotypes as well as to identify mutations based on lipidomic measurements.

2017

Integrating omics data into metabolic models of lipids: a platform for the identification of gene mutations from lipidomics data

Paolo Angelino, Vassily Hatzimanikatis, Howard Riezman, Georgios Savoglidis

2013