Elucidating Synergistic Interactions in Microbial Communities Consisting of Complimentary E. Coli Auxotrophs
Date
2018-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
In nature, microorganisms can exist in diverse, integrated communities wherein metabolic intermediates, in addition to substrates and products, are shared synergistically to enable biochemical phenotypes unobserved in clonal monocultures. In this work, co-cultures of complementary Escherichia coli auxotrophs are studied through 13C metabolic flux analysis (13C-MFA) to elucidate the cooperative interactions between strains. Co-culture models outlining the central carbon metabolism reaction network of each strain as well as the exchange reactions between species within a co-culture were developed and fit to isotopomer distributions obtained through 13C labeled glucose tracer experiments to identify cross-fed metabolites and estimate metabolic fluxes. In a preliminary analysis, isoleucine and �-ketoglutarate were identified as two metabolites likely to be exchanged within a co-culture consisting of two auxotrophic E. coli strains with icd and ilvC knockouts, respectively, while alanine and �-ketoglutarate were identified for a co-culture consisting of ∆icd and ∆argE mutants. In addition, amino acid supplemented agar plating assays were utilized to reveal feedback controlled growth dynamics that follow mutualism by invested benefits. Furthermore, experiments with Transwell® permeable supports reveal the employment of an efficient exchange mechanism in which metabolites are shared by secretion into the cell medium. Adaptive laboratory evolution experiments demonstrated the ability for this mechanism to be improved with a 56% increase in the growth rate of the ∆icd/∆ilvC co-culture (from 0.170 h-1 to 0.265 h-1) over 38 subcultures. Lastly, more complex metabolic models for the co-culture of ∆icd and ∆ilvC were developed to test current hypotheses regarding metabolite exchange. Through 13C-MFA, a model predicting the exchange of amino acids between strains was developed which obtained flux estimates with SSRs that fell near the 95% acceptable confidence level. With further tuning, we believe this model can accurately estimate central carbon metabolism and exchange fluxes for co-cultures of conditionally lethal E. coli mutants