Section: New Results

Models of carbon metabolism in bacteria

All free-living bacteria have to adapt to a changing environment. Specific regulatory systems respond to particular stresses, but the most common decision bacteria have to make is the choice between alternative carbon sources, each sustaining a specific, maximal growth rate. Many bacteria have evolved a strategy that consists in utilizing carbon sources sequentially, in general favouring carbon sources that sustain a higher growth rate. As long as a preferred carbon source is present in sufficient amounts, the synthesis of enzymes necessary for the uptake and metabolism of less favourable carbon sources is repressed. This phenomenon is called Carbon Catabolite Repression (CCR) and the most salient manifestation of this regulatory choice is diauxic growth, a phenomenon discovered by Jacques Monod more than 70 years ago. Although this system is one of the paradigms of the regulation of gene expression in bacteria, the underlying mechanisms remain controversial. Carbon catabolite repression involves the coordination of different subsystems of the cell - responsible for the uptake of carbon sources, their breakdown for the production of energy and precursors, and the conversion of the latter to biomass.

The complexity of this integrated system, with regulatory mechanisms cutting across metabolism, gene expression, signaling and subject to global physical and physiological constraints, has motivated important modeling efforts over the past four decades, especially in the enterobacterium Escherichia coli. Different hypotheses concerning the dynamic functioning of the system have been explored by a variety of modeling approaches. In an article in Trends in Microbiology [3] , which was initiated during the sabbatical of Andreas Kremling in Grenoble in 2013, we have reviewed these studies and summarized their contributions to the quantitative understanding of carbon catabolite repression, focusing on diauxic growth in E. coli. Moreover, we have proposed a highly simplified representation of diauxic growth that makes it possible to bring out the salient features of the models proposed in the literature and confront and compare the explanations they provide.

A bottleneck in the development of dynamic and quantitatively predictive models of bacterial metabolism, explicitly accounting for the different regulatory mechanisms on the molecular level, is information on the kinetic parameters describing the enzymatic reactions and other molecular interactions. One particularly important piece of information is knowledge of enzyme concentrations. Recent technological advances in quantitative proteomics have made mass spectrometry-based quantitative assays an interesting alternative to more traditional immuno- affinity based approaches for quantifying enzyme concentrations. In particular, these advances have improved specificity and multiplexing capabilities. In a study carried out at CEA Grenoble, a quantification workflow to analyze enzymes involved in central metabolism in E. coli was developed. This workflow combined full-length isotopically labeled standards with selected reaction monitoring analysis. The workflow was used to accurately quantify 22 enzymes involved in E. coli central metabolism in a wild-type reference strain and two derived strains, optimized for higher NADPH production. Delphine Ropers and Hidde de Jong participated in the analysis of these data. In combination with measurements of metabolic fluxes, we showed that proteomics data can be used to assess different levels of regulation, in particular enzyme abundance and catalytic rate. This is key to the development of predictive kinetic models, but also provides information that can be used for strain design in biotechnology. An article based on this work was published in Molecular and Cellular Proteomics [8] .

Other ongoing work on the analysis of bacterial metabolism is carried out by Delphine Ropers in collaboration with Inra/INSA in Toulouse, in the framework of the PhD thesis of Manon Morin, supported by a Contrat Jeune Scientifique Inra-Inria. In their respective PhD theses, Stéphane Pinhal and Valentin Zulkower also study specific aspects of carbon metabolism, using both models and experimental data.