Quantifying Bacterial Motion in Twitching Colonies and the Effect of the Agar-Glass Interface on Bacterial Twitching Motility
We have used bright-field optical microscopy, together with a custom-built, temperature- and humidity-controlled environmental chamber, to study the growth of colonies of Pseudomonas aeruginosa PAO1 due to twitching motility driven by type IV pili. The advancing front of colonies consisted of finger-like protrusions (fingers) containing many bacterial cells, with the cells within the expanding colony moving within a lattice-like pattern. We studied the expansion of twitching colonies at the interface between agar and glass for a range of agar concentrations 1.0 % w/v < C < 1.9% w/v, and we interpreted the microscopy results by characterizing the adhesion and local agar concentration at the interface using micropipette deflection and Attenuated Total Reflectance–Fourier Transform Infrared (ATR–FTIR) spectroscopy. To analyze the collective morphology and dynamics of the fingers, we used a combination of custom particle image velocimetry and Fourier analysis techniques. For agar concentrations below C ≤ 1.5% w/v, the average finger width and the density of the lattice region increased with increasing C, whereas the average edge speed remained constant. We observed a transition at C ~ 1.6% w/v in which the average edge speed dropped significantly while the average finger width remained constant. For C> 1.7% w/v, all measured quantities remained constant and the colonies were visually indistinguishable. We attribute this transition to a corresponding increase in the agar-glass adhesion that occurred because of an enhanced local concentration of agarose helices at the agar-glass interface at large agar concentrations. During the outward expansion of fingers along the interface, cells can vertically displace the agar to form multilayered regions. We observed a transition from monolayer to stable multilayer coverage within fingers at C = 1.5% w/v. We studied this transition by characterizing multilayer formation and dissolution, and transient and stable multilayer regions within fingers. We observed that a minimum finger width was required for multilayer stability, and we described the dependence of multilayer lifetime on finger width using a simple nucleation model.