Modelling of an allosterically-controlled branched metabolic system

Gilles Curien (Laboratoire de Physiologie Cellulaire Végétale, CEA Grenoble)

Quantitative modelling of complex biochemical networks is required to understand the logic of their organization. A major limitation for the understanding of metabolic networks is the lack of physiologically relevant kinetic data and quantitative data on the metabolic environment in vivo. The Aspartate-derived amino acid pathway is a model system to analyse the integrated properties of an allosterically-controlled branched metabolic system. This system ensures the synthesis of the essential amino acids Lysine, Threonine, Methionine and Isoleucine from the common precursor Aspartate. The kinetic properties of 15 enzymes from Arabidopsis, among which 12 are allosteric, were modelled according to data collected in vitro using purified recombinant proteins. To build a mathematical model of the system, kinetic data were compiled with data obtained from the plant material (enzyme concentrations, metabolic environment). Steady-state characteristics of the system were analysed by simulation. Metabolite pools and fluxes in the different branches were consistent with in vivo data. Thus, when care is taken to collect data in vitro under conditions mimicking the in vivo context, enzyme rate equations can be used to build accurate in silico replicates of biochemical pathways. Our model is the first one which takes into account the major features of real metabolic networks (branch-points, allosteric controls and isoforms). The model was primarily intended to the understanding of intricate allosteric interactions at the system level. The model allowed us to unravel the mechanisms ensuring a relative independence between competing pathways. The removal of some controls in the model (virtual mutants) permitted to identify a hierarchy in the controls. The probable function of a dual control by S-adenosylmethionine, a specificity of the plant pathway, was also elucidated. Model predictions were confronted to results obtained in planta providing explanations for observations that could not be interpreted in the absence of a model. Finally we observed that the parameters required for the modelling are much less abundant than expected provided that the metabolic context is taken into account. This work paves the way for the modelling of similar systems and for the integration of genetic controls.

Identification de motifs dans les réseaux métaboliques

Vincent Lacroix (Genome Bioinformatics Research Group - CRG PRBB, Barcelona - Spain)

Le travail présenté s’inscrit dans le cadre de l’analyse structurelle des réseaux biologiques. Nous proposons une nouvelle définition de motif dans le contexte des réseaux métaboliques. Un réseau métabolique est modélisé par un graphe coloré et un motif est défini comme un multiensemble de couleurs (une couleur correspond ici à un mécanisme réactionnel). Une occurrence d’un motif est définie comme un ensemble de noeuds connectés et colorés par les couleurs du motif. Nous proposons des algorithmes pour rechercher et inférer de tels motifs, ainsi qu’un critère statistique permettant de décider si un motif est sur-représenté. L’application de nos méthodes au métabolisme d’Escherichia coli révèle des structures locales répétées. Nous argumentons que ces structures peuvent être interprétées comme des blocs fontionnels et/ou évolutifs du métabolisme.

Systematic refinement of a global metabolic model of Acinetobacter baylyi using gene essentialities

Vincent Schachter (CNRS-Genoscope-Université d'Evry)

Gene essentiality screens can be used to significantly upgrade our knowledge on the metabolism of a given species. Genome-scale metabolic models can predict reactions essentiality by analyzing the capabilities of the underlying reaction network in a simulated environment : the relationship between these two types of essentialities is encoded in a genereaction correspondence. Previous work has shown that the identification of inconsistencies between experimental and predicted phenotypes can guide the expert search for model corrections. We introduce here a method, AutoGPR, that automatically corrects the genereaction correspondence in global metabolic models, by reasoning on a suitable model representation together with the set of experimental facts. This refinement strategy was applied to an initial metabolic model reconstruction of Acinetobacter baylyi ADP1, a recently sequenced soil bacterium for which a genome-wide single-gene knock-out library was phenotyped on several growth media. Two rounds of refinement yield a significant set of model corrections and new annotations, showing that large-scale genetics data can be used to rapidly and systematically obtain an accurate metabolic model of a recently sequencedbacterium.