Modeling in Systems Biology: Progress, Problems and Applications to Biotechnology and Biomedicine Oleg Demin Moscow State University, Institute for Systems Biology SPb MCCMB, 2007
Goals Development of quantitative description of biological processes and their application to biotechnology and biomedicine Areas of Expertise • Cellular metabolism ( mitochondria, chloroplast, bacteria, hepatocyte, platelet… ) • Cell signaling ( NFkB, EGFR, Ca 2+ , cAMP, caspases, MAP … ) • Gene regulatory networks ( PurR, IclR, FruR … ) • RNA metabolism ( yeast ) • Pathway reconstruction ( PTH, osteoblasts, iflammation… ) • Methods and software for control of industrial biotechnology processes ( PACS, strain improvement …. ) • Methods and software for kinetic modeling ( DBSolve, Model Creator, … ) • Databases and Information Systems ( Stem Cell Encyclopedia, Inflammation … ) • Protein-protein interactions (electrostatic interactions, prediction of binding energy and 3D-structure of complexes)
Partners and Collaborations Biosystems Informatics Institute (UK, NewCastle) Edinburgh University (UK, Edinburgh) Vytautas Magnus University (Lithuania) Centre National de la Recherche Scientifique Universite Montpellier (France) Amsterdam Free University (The Netherlands) University of Barcelona (Spain) Moscow State University (Russia) Institute of Cancerogenesis, N.N. Blokhin Cancer Research Center (Russia) RiboSys Consortium: EC FP 6 EUROCOLI: European alliance
Challenges in Biomedicine and Biotechnology • How to discover new drugs in fastest, cheapest and most optimal way? • How to minimize drug side effect or/and to find new drugs without it? • How to optimize production of new drugs and their components? Drug Drug Drug production discovery Safety Biosimulations and Informatics AIM To present model based strategy how to cope with the problems
Model Based Strategy Molecular Physiology Biology Data High Data Throughput Data New knowledge about functioning Mathematical OMICs and regulatory mechanisms of Experimental Model Data biological systems Biochemical Clinical Structural Data Data Data
What organisms, organs, tissues and cells have been processed? liver: Bacteria: E.coli, yeast hepatocyte, M. Tuberculosis mitocondria Model Based Stem cells osteoblast Strategy Cardiomiocyte: Blood system: Plant: platelets, endothelium mitochondria chloroplasts cells
What diseases and drug side effects have been processed? hepatotoxicity, Tuberculosis Osteoporosis, polluants effect osteoarthrit Model Based Breast cancer Strategy Heard attack Side effects of inflammation NSAIDs
What modeling techniques do we use? Kinetic Pathway Flux balance reconstruction modeling analysis Model Based Strategy Analysis of Enzyme Structural protein docking kinetics modeling
Biosimulation: what we are talking about? Systems Biology Biosimulation/ Data measurement data integration and interpretation Kinetic modeling Proteomics Metabolomics Pathway reconstruction Genomics Statistical models … …
Topics discussed in this presentation • What kinetic model is? - Stages of kinetic model development and validation • What types of experimental data can be integrated by kinetic model? • How kinetic model can integrate different experimental data? • Applications to biotechnology and biomedicine Examples: - strain improvement - drug safety assessment of NSAIDs
What kinetic model is? Kinetic model (KM) is system of ordinary differential equations describing dynamics and regulations of the corresponding biological system Key KM requirements KM should: • Take into account key properties of the biosystem (stoichiometry, dynamics and regulation) • Clearly describe the key properties of the biosystem in terms of easily understandable and measurable parameters (Vmax, Km, Kd, Ki, etc) • Reproduce correctly all known responses of the biosystem to external and internal perturbations Kinetic Modeling Approach M. Noble, A. Kolupaev, O. Demin, F. Tobin and I. Goryanin , Biotechnology & Bioengineering, 2006, accepted Metelkin, Е , Goryanin, I. and Demin, О . , Biophys. J., 2006, 90, 423-432
Main stages of kinetic model development • Stoichiometry of metabolic pathway and elucidation of the key enzymatic and genetic regulations: Kinetic scheme and N - matrix of stoichiometric coefficients • System of differential equations describing dynamics of the pathway: dx/dt=N·v(x;e,K) Here, x=[x 1 ,…x m ] is vector of metabolite concentrations and v=[v 1 ,…v n ] is vector of rate laws • Description of individual enzymes: - catalytic cycle; - derivation of the rate laws for enzymatic reactions; - estimation of kinetic parameters of enzymatic reactions from in vitro data, available from literature • Validation of the whole model using in vivo data Goryanin I., Lebedeva G., Mogilevskaya E., Metelkin E. and Demin O. Methods Biochem Anal. 2006;49:437-88
What types of experimental data can be integrated by kinetic model? in vivo in vitro Properties mRNA Enzyme Pertrubation Properties Pertrubation Measurement Protein of Purified Experiments of Purified Time Expression Experiments Of steady structure enzymes series profiles on cell culture Operators on Crude State fluxes And extract Regulatory proteins Metabolic networks Genetic networks
How KM integrates different types of experimental data 1) Takes into account in vitro data measured for individual enzymes 2) Uses in vitro data measured on crude extract 3) Is validated against in vivo experimental data Examples: • Signaling networks in platelets and endothelium cells initiated by prostaglandins • Metabolic pathways of prostaglandins’ biosynthesis • E.coli central catabolic pathways • E.coli purine biosynthesis pathway • E.coli histidine biosynthesis pathway
In vitro experimental data • Properties of purified enzymes - kinetics (Histidinol dehydrogenase) - pH and temperature dependence (Histidinol dehydrogenase) • Perturbation experiments on crude extract - Adenylate degradation of burines biosynthesis pathway of E.coli
Minimal kinetic model construction. Enzyme catalytic cycle construction • 3D structure analysis • pH adjustment • temperature adjustment • rate equation derivation • identification of the unique set of parameters to fit all available kinetic experimental data • minimal model from all available selection EXAMPLE: Histidinol dehydrogenase catalyzes two consecutive reactions in histidine biosynthesis pathway: 1) Hol + NAD = Hal + NADH ( histidinol oxidation ) 2) Hal + NAD = His + NADH ( histidinal oxidation ) Hol – histidinol, Hal – histidinal, His - histidine
Minimal kinetic model construction. Pathway of histidine biosynthesis ADP Ammonia Respiratory AMP + + assimilation chain 12 6 ppGpp HisHF 15 KG Glu Gln Glu P i NAD NADH - PPi - PPi HisC 7 10 PRPP PRATP PRAMP ProFAR PRFAR IGP IAP HolP Hol Hal His 1 2 3 4 5 9 13 14 16 17 11 HisG HisI HisI HisA HisHF HisB HisB HisD HisD HisC NAD NADH 8 ATP His His1 AICAR Purine biosynthesis Reactions catalyzed with histidinol dehydrogenase O.V.Demin, I. I. Goryanin, S.Dronov, G.V.Lebedeva Kinetic model of imidazole glycerol phosphate synthase from Escherichia coli , Biochemistry (Russian), 2004, 69(12): 1625-1638
Minimal kinetic model construction. Data on histidinol dehydrogenase available from literature 1) Structural data on catalytic site organization Order of substrate’s binding and product’s release 2) 3) pH dependence of maximal activity of histidinol oxidation (Hol and NAD as substrate) 4) pH dependence of maximal activity of histidinal oxidation (Hal and NAD as substrate) 5) Dependencies of initial rate of histidinol and histidinal oxidation at pH=7.5, pH=7.7 and pH=9.3 6) Time dependence of histidinol oxidation at pH=8.9
Minimal kinetic model construction. CATALYTIC CYCLE OF HISTIDINOL DEHYDROGENASE His E ° NADH His ° E ° NADH E ° NAD Hal Hal k 2 Hol NADH NAD k 1 His ° E Hol ° E ° NAD Hal ° E ° NADH Hal ° E ° NAD E k -1 NADH Hol NAD NAD Hol ° E NADH His Hal ° E E Hal We constructed “minimal” catalytic cycle using (1) structural data on catalytic site organization and (2) experimentally proved order of substrate’s binding and product’s release
Minimal kinetic model construction. pH DEPENDENCE OF CATALYTIC CYCLE OF HISTIDINOL DEHYDROGENASE (1) E ° NADH His ° E ° NADH His E ° NAD H H EH ° NADH His ° EH ° NADH H Hal EH ° NAD Hal H H NADH H k 2 EH 2 ° NADH Hol His ° EH 2 ° NADH EH 2 ° NAD Hol ° E ° NAD His ° E NAD Hal ° E ° NADH Hal ° E ° NAD EH H k 1 H H H H His ° EH Hol ° EH ° NAD Hal ° EH ° NADH Hal ° EH ° NAD EH H H H k -1 H H Hal ° EH 2 ° NAD NADH His ° EH 2 EH 2 Hal ° EH 2 ° NADH Hol ° EH 2 ° NAD Hol Hol ° E NAD NAD E Hal ° EH H H Hol ° EH H NADH His Hal ° EH EH H H H Hol ° EH 2 Hal ° EH 2 Hal EH 2 “Minimal” assumptions on pH dependence enabling rate equation to fit experimental data: 1) only once protonated states of enzyme are active 2) There are 3 groups of enzyme states which differ each other in proton binding affinity
Recommend
More recommend
