Bacteria normally rely on diffusion for material transport, such as nutrient uptake or signal exchange. Diffusion is slow if the distance is long. Many bacterial species use rotating flagella for locomotion. While they move, flagellated bacteria stir their fluid environment and generate micro-scale flows. An interesting question is, can bacteria make use of the flagellum-driven flows for long-distance transport? Here we show that flagellated bacteria in colonies can achieve this goal via self-organization.

Back in 2011, when I was working as a postdoc with Howard Berg at Harvard University, we discovered that expanding swarms of flagellated bacteria can generate directed flows for cargo transport at the swarm edge. Due to swarm expansion, however, those flows do not stay on track, and cargoes carried by the flows would get lost after traveling for less than 1 mm. What if we stop the swarm expansion? One would expect that the directed flows could then remain stable at the colony edge. Unfortunately, we did not succeed in observing any directed flows at the edge of non-expanding colonies. There cells appeared to be motionless.    

A few years later, when I was examining an E. coli colony routinely grown on streaked plates, I noticed that cells at the very edge were moving actively but not stuck as usual. A closer look revealed something exciting: Cells at the out-most edge were lining up orderly and together they circled clockwise around the entire colony; meanwhile, there appeared to be a rapid stream just adjacent to the edge and circulating counterclockwise around the colony. Immediately I knew this was what I had been looking for. After checking the conditions, I realized that humidity was the key -- The colonies had to be grown in a relatively humid environment to develop the organized motion and flows. Then my graduate student, Haoran Xu took the project. He elucidated the self-organization mechanism, and demonstrated that the rapid stream indeed provides a stable and high-speed avenue for directed material transport over centimeters. With strains kindly provided by Karine Gibbs at Harvard, we switched to Proteus mirabilis at some point, as its phenomena were more robust during observations.  Eric Lauga, an expert in fluid mechanics at University of Cambridge and his students helped us to explain the directionality of collective cellular motion.  It is essentially a result of hydrodynamic interaction between flagella and the substrate.

This new mechanism of long-range active transport may bring profound effect on the physiology of bacterial communities. Bacteria communities in natural and clinical settings normally grow in structured environments with heterogeneous nutrient distribution.  In such environments, the self-organization of motile sub-population may occur in various locations not limited to the colony edge; thus the long-range directed flows enabled by motile-cell self-organization may efficiently redistribute slowly-diffusing substances within bacterial communities.  We also wonder whether the long-range directed flows could be exploited to perturb the behavior of bacterial communities; for example, drugs and chemical effectors loaded in vesicles could be supplied locally and transported by the flows to manipulate distant parts of a bacterial colony.