Identification of Antimicrobial Drug Targets from Robustness Properties of Metabolic Networks
MetadataVis full innførsel
A reaction universe containing all 13,849 metabolic reactions known to exist was constructed and found to share many topological properties with real-world metabolic networks. Integration of the reaction universe into 43 different microbial genome-scale metabolic reconstructions led to improved viability and robustness. Five metabolic reactions remained essential in more than 70 % of these reconstructions after integration of the reaction universe and these absolutely superessential reactions were identified as potential targets for broad-spectrum antimicrobial drugs. One of the five reactions was involved in peptidoglycan biosynthesis and the remaining four were part of riboflavin metabolism. No reactions were absolutely superessential in all 43 cellular contexts, meaning that no set of reactions that are always essential in any metabolic network is likely to exist. Ten of the reconstructions into which the reaction universe was integrated were used to generate large ensembles of random viable metabolic networks. The method used for metabolic network randomization was evaluated and it was found that it produced networks with large fractions of blocked reactions. Aside from this, the reaction contents of random viable metabolic networks correlated very strongly with network size. Most importantly, small networks were less randomized than large ones. Even so, the increased size of the reaction universe relative to past studies allowed greater network randomization than what has previously been achieved. Many reactions that were essential or part of synthetic lethal pairs in random viable metabolic networks were capable of being so in all investigated cellular contexts. Based on this, it was postulated that essentiality and synthetic lethality is often caused by factors that are shared between different organisms and environments. Superessentiality indices, which indicate how frequently reactions are expected to be essential in metabolic networks in general, were calculated and found to correlate positively between cellular contexts. However, these correlations were only strong between indices obtained from very similar models, indicating that superessentiality is sensitive to cellular context. Also, a great deal of deviation between indices calculated in this study and previously reported ones was observed, primarily due to the increased size of the reaction universe. An average superessentiality index revealed that some reactions were highly superessential in all investigated cellular contexts and the ten reactions with highest average superessentiality indices, all of them involved in purine or histidine metabolism, were identified as potential antimicrobial drug targets. Synthetic lethality data obtained from random viable metabolic networks was used to construct graph representations of pairwise synthetic lethal interactions between reactions. All of these synthetic lethality networks contained a giant component in which most nodes were found and in all cases this giant component was highly clustered and single-scale and exhibited small-world properties. Indications of assortative network organization were also found. Finally, an algorithm was developed for identifying alternative metabolic pathways of essential reactions in metabolic networks and applied to all essential reactions in two models of potentially pathogenic bacteria. It was found that more than 500 alternative metabolic pathways existed in the reaction universe for most essential reactions in these models. The remaining essential reactions generally had few alternative pathways, most of which consisted of few reactions. Comparison to superessentiality indices showed that the key determinant for reaction superessentiality was most likely a combination of the number of alternative pathways and the lengths of these pathways.