History dependence of chemosensing strategy in Escherichia coli

Princeton University, Princeton NJ

Investigators

Linked publications, trials & patents

Abstract

DESCRIPTION (provided by applicant): To survive, organisms must continually adapt to changing conditions. While some of these responses represent relatively simple mechanisms of homeostasis, others are more complex, reflecting associative learning and even predictive ability. Such sophisticated responses are not solely the domain of multicellular organisms - bacteria have also evolved refined strategies to deal with their complex, changing environments, e.g. circadian rhythms and temperature/oxygen association. While such examples primarily reflect metabolic adaptation, there is recent evidence that bacterial sensory systems are also reshaped by the cell's growth environment. In particular, the chemotaxis network of Escherichia coli undergoes ~10-fold changes in protein levels and ratios in response to nutrient abundance, temperature, and cell density. The large number of assays available for this system and the existence of well-tested quantitative models of its operation make it an ideal target to explore the principles of history-dependent sensing. To carry out this exploration, we will grow E. coli cells under a wide range of physiologically relevant conditions, including nutrient type and abundance, temperature, pH, O2 levels, osmolarity, cell density, and the presence of multiple chemical signals. We will then characterize the chemotactic network at three levels: protein abundances, signaling response to stimulation (via fluorescence resonance energy transfer), and chemotactic behavior (via tracking of single cells swimming in microfluidic gradients). We will exploit the well-established model for chemotactic signaling to interpret our experimental results, and to develop a working model for how growth conditions reshape the chemosensory apparatus. The molecular mechanisms underlying history-dependent regulation, both known and newly discovered, will be characterized by assaying mRNA and protein levels/stability and by exploiting a variety of fluorescent reporters. Finally, we will extend the existing model for chemotactic signaling to determine how chemotactic performance depends on network composition, predict optimal scaling relations between protein levels and receptor cooperativity, and test these predictions with microevolution experiments.