Fluctuation Timescales in Bacterial Gene Expression

Abstract

The stochastic nature of intracellular chemistry guarantees that even genetically identical cells sharing an environment will differ in composition. The question of whether this chemical diversity translates into significant phenotypic individuality is tied to the relative timescales of the processes involved. In order for cells in a population to have distinct functional identities, they must maintain their states for an appreciable period of time. Quantification of these timescales requires accurate time-lapse measurements covering tens or even hundreds of generations, a technical hurdle that has left these questions largely underexplored. In this thesis I present three pieces of work that aim to provide a foundation for the study of fluctuation timescales in bacteria. In the first part, I describe modifications to a recently developed microfluidic platform for continuous culture of cells under constant conditions. These revised devices enable the high-throughput, long-term measurement of gene expression dynamics while eliminating several confounding experimental factors that interfere with timescale measurements. In the second part, I employ one of these devices to survey fluctuation timescales in ~50 reporters for Eshcerichia coli gene expression. Under rich conditions, all reporters exhibited nearly identical, rapid fluctuation dynamics that were captured by a simple model of gene expression. In contrast, under poor nutritional conditions gene expression states became correlated over several cell divisions. However, accounting for instantaneous growth rate fluctuations eliminated these slow timescales, revealing an exceedingly simple behavior. In the third part, I describe our work to dissect the stochastic transition between the solitary motile state and sessile multicellular state in exponentially growing Bacillus subtilis</italic.. By enforcing static environmental conditions, we uncover the cell's internal strategies for state switching. The transition to the multicellular state occurs without regard to the cell's state history, whereas commitment to the multicellular state is tightly timed. By manipulating the genetic circuit responsible for the switch, we also expose surprising functional modularity in the commitment. I believe that the striking range of gene expression timescales we observe--from the fast fluctuations in E. coli gene expression to the feedback-amplified noise in B. subtilis--will serve as a useful starting point for future studies.