INVESTIGATING THE MECHANISM OF BACTERIAL CELL DIVISION WITH SUPERRESOLUTION MICROSCOPY

The molecular mechanisms that drive bacterial cytokinesis are attractive antibiotic targets that remain poorly understood. The machinery that performs cytokinesis in bacteria has been termed the 'divisome' (see Chapter 1 for description). The most widely-conserved divisome protein, FtsZ, is an essential tubulin homolog that polymerizes into protofilaments in a nucleotide-dependent manner. These protofilaments assemble at midcell to form the ‘Z-ring’, which has been the prevailing candidate for constrictive force generation during cell division. However, it has been difficult to experimentally test proposed Z-ring force generation models in vivo due to the small size of bacteria (< 1 μm diameter for E. coli) compared to the diffraction-limited resolution of light (~ 0.3 μm). In this work, quantitative superresolution and time-lapse microscopy were applied to examine whether Z-ring structure and function indeed play limiting roles in driving E. coli cell constriction (Chapter 2). Surprisingly, these studies revealed that the rate of septum closure during constriction is robust to substantial changes in many Z-ring properties, including the GTPase activity of FtsZ, molecular density of the Z-ring, the timing of Z-ring disassembly, and the absence of Z-ring assembly regulators. Further investigation revealed that septum closure rate is instead highly coupled to the rate of cell wall growth and elongation, and can be modulated by coordination with chromosome segregation. Taken together, these results challenge the Z-ring centric view of constriction force generation, and suggest that cell wall synthesis and chromosome segregation likely drive the rate and progress of cell constriction in bacteria. These investigations were made possible by advancements in quantitative superresolution microscopy techniques (see Chapter 3 for overview). One major obstacle encountered during the course of this work, and shared by those utilizing localization-based superresolution microscopy techniques, was the overestimation of molecule numbers caused by fluorophore photoblinking. Thus, Chapter 4 describes a systematic characterization of the effects of photoblinking on the accurate construction and analysis of superresolution images. These characterizations enabled the development of a simple method to identify the optimal clustering thresholds and an empirical criterion to evaluate whether an imaging condition is appropriate for accurate superresolution image reconstruction. Both the threshold selection method and imaging condition criterion are easy to implement within existing PALM clustering algorithms and experimental conditions