Bacterial Kayaking

Image courtesy of Professor Köser

Probably our most well-known bacterium is Esch­erichia coli. These rod-shaped microorganisms come in a variety of strains – some harmlessly inhabit our intestines, while others cause infections. Still, many aspects of E. coli remain poorly understood. One very important yet under-characterized feature of the bacterium is its motion in a flowing liquid, known as hydrodynamics.

Hür Köser, Associate Professor of Electrical Engi­neering, came across bacterial hydrodynamics when studying chemotaxis, or the way bacteria direct their motion in response to chemical signals. “It became clear that something other than chemotaxis was hap­pening,” said Professor Köser.

As an E. coli bacterium swims, its flagella form a helical bundle that rotates in one direction to propel the cell. The bacterium’s subsequent path depends on the flow of the fluid. In the absence of flow, a bacterium near a surface swims in circles. As the flow is increased, it swims in larger circles, until in sufficiently high flow it swims upstream.

In a previous study, Professor Köser’s laboratory found that E. coli in flow near a surface swim upstream in a trajectory that balances the forces from both the flow and from bacterial propulsion.

To understand the hydrodynamic interactions between the bacteria, the flow and the nearby surface, researchers in the Kös˛er lab studied mutant bacteria that lacked flagella and therefore could not swim. “We want to find a simple model that we can use to predict the motion of bacteria that have flagella,” said Tolga Kaya, a postdoctoral research associate in the Köser lab.

In this study, published in the September 2009 issue of Physical Review Letters, the researchers used an optical microscope to observe non-flagellated bacteria carried downstream by flow. They found that the bacteria followed periodic trajectories similar to that of a kayak paddle.

The paddle-like motion of the bacteria corresponded with the orbits G. B. Jeffery predicted for rod-shaped par­ticles in 1922. Past research on Jeffery orbits has been limited to studies of long fibers. The Kös˛er lab’s study marks the first time Jeffrey orbits have been observed in bacteria, and it constitutes the most reliable and systematic data on the Jeffrey orbits of any particle.

“What happens in physics is that everyone models a bacterium as a sphere with a tail,” said Professor Köser. “Actually, the bacterium has more of a cigar shape. It does not get dragged isotropically – it interacts with the flow in the presence of the surface.”

When the bacterium comes closer to the sur­face, the surface forces increase, slowing down the bacterial motion. “The shape of the motion stays almost the same, but the period – how fast the bacteria paddle – changes depending on the distance from the surface,” explained Dr. Kaya.

The bacteria, like any rod-shaped particles, adjust their distance from the surface to find a stable position between the surface forces and their own motion. Flagellated bacteria use their distance and angle from the surface to find an equilibrium to move upstream.

At 2 to 5 microns long and under 1 micron wide, E. coli are strongly influenced by Brownian motion, which is the seemingly random flow of particles in liquid or gas. To reduce this effect, engineers in the Köser lab increased the viscosity of the medium.

The small scale and the flow also made it difficult to track the bacteria. “We used a high-resolution camera taking up to a hundred images per second,” said Dr. Kaya. “But the hard part was creating and debugging the algorithm to track each bacterium. This took months. When dealing with such large amounts – terabytes – of data, you need to write a very sophisticated program.”

“We really pushed our resources to the limit in terms of computation, data memory, and the speed at which the data could be taken,” con­cluded Professor Köser.

Understanding bacterial hydrodynamics could allow engineers to construct flow chambers that prevent E. coli upstream motility. Such chambers could reduce the rate of bacterial infection and water contamination without using antibiotics.

“We’ve characterized one hydrodynamic phe­nomenon,” said Professor Köser. “Next, we want to combine all components of bacterial motil­ity in flow near a surface. Since the equations governing the hydrodynamics are linear, we can combine these components by superposition. To understand motile cells, we would add swimming motility to what we have found here and see how that affects the hydrodynamics.”

Further Reading

  • Hill, J.; Kalkanci, O.; McMurry, J.L.; Köser, H. Hydrodynamic Surface Interactions Enable Escherichia Coli to Seek Efficient Routes to Swim Upstream. Phys. Rev. Lett., 2007. 98(6).
  • Kaya, T. and Köser, H. Characterization of Hydrodynamic Surface Interactions of Escherichia coli Cell Bodies in Shear Flow. Phys. Rev. Lett., 2009. 103(13).