Stochasticity, conflicts, and instability: Biological strategies for optimal growth in a complex environment

Abstract

Life is complicated. Even in the bacterium Escherichia coli, cell proliferation is dependent on the maintenance of over 2000 small-molecule metabolites, as well as the synthesis of more than 600 essential proteins. In addition to the sheer scale of this regulatory challenge, the regulatory processes themselves are stochastic in nature. In recent decades, biologists have made great progress in developing a functional map of cellular metabolism, but the dynamics and regulatory behavior of cellular processes remain largely opaque. In this dissertation, I have aimed to develop a series of mathematical and experimental methods that form a foundational framework for investigating and characterizing cellular dynamics and robustness. I first use the behavior of exponential growth to establish a direct correspondence between stochastic and deterministic cell models, bridging the gap between experimental stochasticity and observed population demographics. Using this correspondence, I then introduce the method of lag-time analysis, which experimentally characterizes the in vivo dynamics of replication for an exponentially-growing bacterial population. We use the method to measure replication pauses down to the precision of seconds, replication fork velocity in units of base pairs per second, and temporal oscillations in fork velocity in three evolutionarily-divergent species. Next, I introduce the robustness-load trade-off model, which incorporates stochasticity and an asymmetric fitness landscape to predict a lower limit for transcription of essential genes, metabolic load balancing between transcription and translation, and a generic overabundance of essential proteins. In the final chapter, I describe some preliminary work on regulatory feedback dynamics. We predict that regulation is a strategy that the cell uses to maintain robustness, complementary to the overabundance strategy. We also find an oscillatory signature that agrees with the lag-time analysis results, and demonstrate a trade-off between feedback strength, speed of return to equilibrium, and network stability. Although this research is not yet complete, this chapter provides a road map for further analysis and experimental tests. My hope is that the emergent phenomena described in this dissertation provide a solid foundation for future work on cellular dynamics and regulation.