Abstract
While bacterial motility has been well characterized in uniform liquids, only little is known about how bacteria propagate through complex environments, such as gel-like materials or porous media that are typically encountered in tissue or soil. Here, we study bacterial swimming in polysaccharide matrices formed by different concentrations of agar. We focus on the soil bacterium Pseudomonas putida (P. putida) that is known for its multimode swimming pattern, where a polar bundle of flagella may push, pull, or wrap around the cell body. In the gel matrix, P. putida cells display run-and-turn motility with exponentially distributed run times and intermittent turning phases that follow a dwell time distribution with power-law decay. An analysis of the turn angle distribution suggests that both, flagella mediated turning as well as mechanical trapping in the agar matrix are part of the overall swimming pattern. We compare these results to knockout mutants which differ from the wild-type in their swimming speed and show altered probabilities for the occurrence of the three swimming modes. Their run length distributions in the agar matrix are, however, identical demonstrating that run episodes of bacterial swimmers in a gel matrix are primarily determined by the surrounding geometry. We propose a minimal active particle model providing analytical solutions that quantitatively explain the observed time dependence of the mean squared displacement in the gel based on the experimentally observed motility pattern and the measured waiting-time distributions.