Relative flux trade-offs and optimization of metabolic network functionalities

• Published in: Computational and structural biotechnology journal Vol. 20; pp. 3963 - 3971
• Main Authors: Hashemi, Seirana, Razaghi-Moghadam, Zahra, Laitinen, Roosa A.E., Nikoloski, Zoran
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.01.2022; Research Network of Computational and Structural Biotechnology; Elsevier
• Subjects: Fluxes; Growth; Metabolic networks; Overexpression targets; Trade-offs; Overexpression targets; Fluxes; Growth; Trade-offs; Metabolic networks
• ISSN: 2001-0370; 2001-0370
• DOI: 10.1016/j.csbj.2022.07.038

[Display omitted] Trade-offs between traits are present across different levels of biological systems and ultimately reflect constraints imposed by physicochemical laws and the structure of underlying biochemical networks. Yet, mechanistic explanation of how trade-offs between molecular traits arise and how they relate to optimization of fitness-related traits remains elusive. Here, we introduce the concept of relative flux trade-offs and propose a constraint-based approach, termed FluTOr, to identify metabolic reactions whose fluxes are in relative trade-off with respect to an optimized fitness-related cellular task, like growth. We then employed FluTOr to identify relative flux trade-offs in the genome-scale metabolic networks of Escherichia coli, Saccharomyces cerevisiae, and Arabidopsis thaliana. For the metabolic models of E. coli and S. cerevisiae we showed that: (i) the identified relative flux trade-offs depend on the carbon source used and that (ii) reactions that participated in relative trade-offs in both species were implicated in cofactor biosynthesis. In contrast to the two microorganisms, the relative flux trade-offs for the metabolic model of A. thaliana did not depend on the available nitrogen sources, reflecting the differences in the underlying metabolic network as well as the considered environments. Lastly, the established connection between relative flux trade-offs allowed us to identify overexpression targets that can be used to optimize fitness-related traits. Altogether, our computational approach and findings demonstrate how relative flux trade-offs can shape optimization of metabolic tasks, important in biotechnological applications.