280803 Comparison of Network Structures That Confer Resilience Against Genetic Perturbations in Microbial Metabolism

Wednesday, October 31, 2012: 9:10 AM
Crawford East (Westin )
Chintan Joshi and Ashok Prasad, Chemical and Biological Engineering, Colorado State University, Fort Collins, CO

Due to their relatively high rates of division, most microorganisms need to be resilient against random genetic perturbations. In the metabolic network, microorganisms can survive a large number of gene deletions by altering the flux distribution to compensate for missing reactions. Understanding the design principles of the metabolic network that confer this resilience to the organism is of great interest for metabolic engineering strategies that aim to re-direct fluxes towards commercially important products like biofuels, nutraceuticals and pharmaceuticals. We used flux balance analysis (FBA) along with minimization of metabolic adjustment (MOMA) in order to predict the effects of gene knockouts on two genome-scale models of microbial metabolism under different conditions -- that of a bacterial network (E. coli iAF1260 – aerobic and anaerobic growth) and a cyanobacterial network (Synechocystis sp. PCC6803 iJN678 – autotrophic, mixotrophic, and heterotrophic growth). In autotrophic metabolism more than 52% of the mutants were lethal, while the remainder grew at nearly optimal growth rates, which was achieved by taking up higher amounts of light. We mapped genes leading to lethal knock-outs in one network on to the other and screened for non-lethality. We found that several lethal knockouts in one organism were non-lethal in the other, in other words the two organisms had different resilience properties. We identified some key subsystems of the E. coli metabolic network and the electron flow machinery of Synechocystis as being primarily responsible for conferring this resilience. We found that different sets of metabolites were very tightly regulated in these organisms, and genetic mutations that led to drastic perturbations in reactions involving these metabolites proved lethal for the organism. Analysis of epistasis by carrying out double gene deletions revealed that the autotrophic environment led to a large number of aggravating gene pair interactions. However, the heterotrophic environment and aerobic growth of E. coli led to relatively higher buffering interactions. Further, epistasis analysis was used to reveal more differences in the association between genes and reactions of metabolic network structures of the two organisms. Our analysis allows us to make predictions about; (i) why bacterial and cyanobacterial network exist the way they do, and (ii) how these organisms might benefit out of these unique gene-reaction associations. Thus photosynthetic and non-photosynthetic organisms have different resilience properties, arising in part from the differing structures of their metabolic networks, metabolite essentiality, and gene-reaction associations.

Extended Abstract: File Not Uploaded
See more of this Session: In Silico Systems Biology: Cellular and Organismal Models II
See more of this Group/Topical: Topical A: Systems Biology