UC San Diego

Bacteria in changing environments must constantly adapt to grow and survive. The adaptive response to change can be extremely complex, as it depends not only on the current conditions, but also on the history of the cell over many generations. For this reason, most of our understanding of bacterial physiology comes from so-called steady state growth, where cultures are grown in static environments for a long time. As a first step toward understanding the kinetics of adaptation, we build on the foundation of steady-state growth and study transitions between two well -defined steady states of E. coli. We first focus on carbon upshifts, in which cells growing in steady state are supplemented with a better carbon source and transition to a new steady state with faster growth. We observe that the response of growth rate has multiple timescales and occurs over the course of several generations. The rate of biomass accumulation (i.e. flux), on the other hand, reaches its final behavior much more quickly. We develop a model that quantitatively reproduces these kinetics using only empirical observations of ribosome and catabolic enzyme synthesis in steady state and the known topology of regulatory interactions. The model is solved analytically and has only a single free parameter that captures the initial influx of the added carbon. We predicted that if this initial flux is high enough the growth rate can transiently exceed its steady state value for several hours; this is verified by synthetically titrating carbon transport enzymes. We also studied carbon downshifts, in which cells growing on a combination of two carbon sources deplete one of them and transition to slower steady state growth on only a single carbon source. We observe that the growth rate recovers more quickly than for upshifts, but the growth kinetics are also quantitatively captured by our model. We are able to reproduce surprising and counterintuitive kinetics of growth transitions with a conceptually simple model. The success of our approach demonstrates the power of empirical characterizations to quantitatively capture biological phenomena