Perturbation-response analysis ofin silicometabolic dynamics: Hard-coded responsiveness in the cofactors and network sparsity

Abstract

Homeostasis is a fundamental characteristic of living systems. Unlike rigidity, homeostasis necessitates that systems respond flexibly to diverse environments. Understanding the dynamics of biochemical systems when subjected to perturbations is essential for the development of a quantitative theory of homeostasis. In this study, we analyze the response of bacterial metabolism to externally imposed perturbations using kinetic models ofEscherichia coli’s central carbon metabolism. We found that three distinct kinetic models consistently display strong responses to perturbations; In the strong responses, minor initial discrepancies in metabolite concentrations from steadystate values amplify over time, resulting in significant deviations. This pronounced responsiveness is a characteristic feature of metabolic dynamics, especially since such strong responses are seldom seen in toy models of the metabolic network. Subsequent numerical studies show that adenyl cofactors consistently influence the responsiveness of the metabolic systems across models. Additionally, we examine the impact of network structure on metabolic dynamics, demonstrating that as the metabolic network becomes denser, the perturbation response diminishes—a trend observed commonly in the models. To confirm the significance of cofactors and network structure, we constructed a simplified metabolic network model, underscoring their importance. By identifying the structural determinants of responsiveness, our findings offer implications for bacterial physiology, the evolution of metabolic networks, and the design principles for robust artificial metabolism in synthetic biology and bioengineering.Significance StatementUnderstanding the dynamics of metabolic systems is vital for deciphering cellular functions and their responses to environmental shifts. However, the advancement of dynamic theories for cellular metabolism lags behind its static counterparts. In this study, we delve into the dynamic responses ofEscherichia coli’s central carbon metabolism to perturbations. We found that the response of the cellular metabolism is stronger than that of the random network-based metabolic models. Also, our results highlight the critical role of adenyl cofactor dynamics, such as ATP, and sparse network structure in dictating response strength. These findings carry profound implications across bacterial physiology, evolutionary biology, and synthetic biology.