The interplay between metabolic stochasticity and regulation in single E. coli cells

Metabolism is inherently stochastic at the cellular level. Whether cells actively regulate processes in response to these random internal variations is a fundamental problem that remains unaddressed, yet critical to understanding biological homeostasis. Here, we show that in E. coli cells, expression of the main catabolic enzymes is continuously adjusted in response to metabolic fluctuations under constant external conditions. This noise feedback is performed by the cAMP-CRP system, which controls transcription of the catabolic enzymes by modulating concentrations of the second messenger cAMP upon changes in metabolite abundance. Using time-lapse microscopy, genetic constructs that selectively disable cAMP-CRP noise feedback, and mathematical modelling, we show how fluctuations circulate through this hybrid metabolic-genetic network at sub cell-cycle timescales. This circulation of stochastic fluctuations is explained by four distinct noise propagation modes, one of which describes the continuous cAMP-CRP regulation. The model successfully predicts how noise circulation is impacted by C-sector under and over-expression. The results raise the question whether the cAMP-CRP system, as well as other metabolic regulation mechanisms, have evolved to manage internal metabolic fluctuations in addition to external growth conditions. We conjecture that second messengers may broadly function to control metabolic stochasticity and achieve cellular homeostasis.