Metabolic comparison of the in-silico phenotype-genotype relationship of Pseudomonas putida and Pseudomonas aeruginosa

Vitor A.P. Martins dos Santos 1, Miguel Godinho de Almeida 2, Jeremy S. Edwards2, Kenneth N. Timmis1, National Centre for Biotechnology Research; 2University of Delaware

The relationship between the genotype and the phenotype is complex, highly non-linear and cannot be predicted from simply cataloguing and assigning gene functions to genes found in a genome. Comprehensive understanding of cellular metabolism requires placing the function of every gene in the context of its role in attaining the set goals of a cellular function. This demands the integrated consideration of many interacting components. Mathematical modelling provides us a powerful way of handling such information and allows us to effectively develop appropriate frameworks that account for these complexities. We report on an in silico representation of Pseudomonas putida and Pseudomonas aeruginosa that describes their metabolic capacities within the scope of their environmental constraints. Using annotated genome sequence data, biochemical information and strain-specific knowledge, we analysed (and we expect to be able to interpret extensively and ultimately predict) the cellular behaviour of these micro-organisms under a wide range of conditions relevant for both human health and environmental applications. This has been done using flux balance analysis for the entire metabolic networks of these bacteria, as well as by determining the elementary modes for relevant subsets of the networks. Preliminary results show that the number of elementary pathways that represent the metabolic potential of P. aeruginosa is 2 to 6 times higher than those for P. putida, although the former has only two more reactions than the latter. This reflects a higher flexibility of the central metabolism of Ps. aeruginosa as compared to Ps. putida. This is clearly a emergent property of the system that could not be predicted solely on basis of the linear comparison of gene lists. The construction of comprehensive metabolic maps provides a framework to study the consequences of alterations in the genotype and to gain insight into the phenotype-genotype relation. Ultimately, this analysis defines the entire metabolic space of the possible flux distributions and metabolic interactions within the network. A direct comparison of this „phenotypic space" for both bacteria will possibly help in identifying „orphan genes", evolutionary features and genetic plasticity. Such in silico models can be as well used to choose the most informative knockouts and ultimately, to rationally design experiments relevant for the elucidation of the behavior of these bacteria in polluted environments or within the scope of their relationships (as pathogens) with an infected host.