Genome-scale reconstruction of the metabolic network in Staphylococcus aureus N315: an initial draft to the two-dimensional annotation

Abstract

Background: Several strains of bacteria have sequenced and annotated genomes, which have been used in conjunction with biochemical and physiological data to reconstruct genome-scale metabolic networks. Such reconstruction amounts to a two-dimensional annotation of the genome. These networks have been analyzed with a constraint-based formalism and a variety of biologically meaningful results have emerged. Staphylococcus aureus is a pathogenic bacterium that has evolved resistance to many antibiotics, representing a significant health care concern. We present the first manually curated elementally and charge balanced genome-scale reconstruction and model of S. aureus' metabolic networks and compute some of its properties.

Results: We reconstructed a genome-scale metabolic network of S. aureus strain N315. This reconstruction, termed iSB619, consists of 619 genes that catalyze 640 metabolic reactions. For 91% of the reactions, open reading frames are explicitly linked to proteins and to the reaction. All but three of the metabolic reactions are both charge and elementally balanced. The reaction list is the most complete to date for this pathogen. When the capabilities of the reconstructed network were analyzed in the context of maximal growth, we formed hypotheses regarding growth requirements, the efficiency of growth on different carbon sources, and potential drug targets. These hypotheses can be tested experimentally and the data gathered can be used to improve subsequent versions of the reconstruction.

Conclusion: iSB619 represents comprehensive biochemically and genetically structured information about the metabolism of S. aureus to date. The reconstructed metabolic network can be used to predict cellular phenotypes and thus advance our understanding of a troublesome pathogen.

Figure 1

Amino acid contributions to growth. The results of adding equivalent quantities of each amino acid to the media are shown here. Red bars represent amino acids that are reported essential in the literature, and green bars are amino acids that are not. Arginine is present in all cases and is the baseline against which the rest of the values are normalized. On average, adding an essential amino acid to the media allows better growth than does adding a non-essential amino acid. E stands for average essential amino acid, and NE stands for average non-essential amino acid. Two amino acids do not have transporters in the genome annotation and are not included here for that reason.

Figure 2

Relative growth efficiency with different carbon sources. The in silico growth of iSB619 varies depending on which carbon source is provided and the amount of oxygen present. The predicted efficiency of carbon incorporation into biomass is shown here as a function on the oxygen consumption. Growth rate is normalized relative to the number of carbon atoms per molecule. Oxygen consumption is normalized relative to optimal oxygen consumption for each carbon source. Trehalose, lactose, and sucrose all overlap (the trehalose line indicates all three). The legend is presented in the same order as the carbon sources appear in the figure, top to bottom.