Quantifying Antibiotic Response: Understanding Variability from Individual Cells to Community Dynamics

Abstract

Bacterial responses to antibiotic treatments have been precisely quantified. This has greatly enhanced our understanding of bacterial responses at 3 levels: cellular behaviors, clonal population characteristics, and community structure dynamics under antibiotic exposure. The significance of this dissertation is in establishing connections between these quantified dynamics, which facilitates robust predictive modeling and a deeper comprehension of bacterial systemic responses. We apply these approaches to study bacterial responses to beta-lactam antibiotic treatment. Beta-lactam antibiotics, widely prescribed for both prophylactic and therapeutic purposes, often lead to a transient increase in total population biomass followed by a decline. Despite individual cells often undergoing varying degrees of filamentation before lysis, the emergence of these clonal population-level responses from these individual behaviors during beta-lactam treatment remained unclear. Variations in the linear correlation within subpopulations and the interactions within microbial communities further complicate the quantitative mappings. Here, I demonstrate the quantitative mappings between those three levels of dynamics using precise measurements and mathematical modeling. My research explains how single-cell lysis probability can be quantified and establishes a connection between bacterial filamentation and time-delayed population reductions. It supports the well-established notion that growth and lysis rates are linearly correlated. My work further extends our understanding of how this linear correlation affects the community dynamics and restructures composition under antibiotic treatment, considering the variability and interactions between populations. Collectively, my work provides fundamental links between each level of bacterial responses through quantitative mapping, utilizing high-resolution measurements.