Constrains-Based De-Composition of the Metabolic Robustness in Bacteria

With the exception of the Enterobacteria group, the current bacterial genome-scale reconstructions (GENREs) represent, roughly, models of primary metabolism rather than true organism-specific genome-scale models. Therefore, efforts focused on the high-quality reconstruction of bacterial groups other than Enterobacteria are required in order to expand the current biological reactome suitable for computation. Pseudomonads comprises a heterogeneous and large group of species (> 100), which share a remarkable metabolic and physiological versatility, and high adaptability enabling colonization of diverse ecological niches. Pseudomonas are of great interest because of their importance in plant and human diseases, their potential as plant growth promoting, and as biocatalysts. Among this group, P. putida has been widely used as model environmental bacterium lacking undesirable virulence factors. P. putida is a paradigm of metabolic versatility and exhibits a notable resistance to environmental stresses. These traits, together with its amenability for genome editing, make P. putida as an ideal workhorse for synthetic biology-based cell factories.

Aimed by the lack of coverage of strain-specific metabolism in published bacterial reconstructions, we performed an exhaustive study of P. putida KT2440 metabolism.  First, a deep coverage of P. putida bibliome and its manual curation, followed by an extensive experimental validation process, led to a strain-specific metabolic network reconstruction comprised of 1411 genes, 2826 metabolic and transport reactions, and 2083 metabolites. The final reconstruction, iJN1411, represents a significant expansion over previous GENREs from P. putida, even over current reconstructed bacteria. We first used iJN1411 for the assessment of the metabolic capabilities of P. putida as a species by means of the multi-strains modeling of ten P. putida strains. iJN1411 was then used to gain insights on P. putida metabolism including: i) the quantification and expansion of its well-known metabolic versatility, ii) the identification of hidden phenotypes and metabolic mechanisms such as an alternate ubiquinone-independent oxidative metabolism, iii) the assignment of the molecular mechanisms providing metabolic robustness under both environmental and genetic perturbations, and iv) the identification of metabolic capacitors based on ATP-fueled metabolic cycles.

In summary, we show that iJN1411 can be used not only for increase our knowledge about KT2440 and other P. putida strains, but also as a platform for modeling the whole Pseudomonas group while providing a new scaffold for deciphering new and unexpected bacterial systems properties. In this sense, our analysis provides solid computational and experimental evidences about how bacteria increase metabolic robustness, paving the way for future engineering of more robust and stable biocatalysts based on minimal cells.