CMPG-SPI logo

Swarming behavior of Rhizobium etli and Pseudomonas aeruginosa

Research carried out by Maarten Fauvart, Natalie Verstraeten

Swarming is the fastest known mode of surface translocation, allowing rapid colonization of nutrient-rich environments and host tissues. This complex multicellular behavior requires integration of chemical and physical signals, leading to physiological and morphological differentiation of the bacteria into swarmer cells (Verstraeten et al., 2008). Two model systems for swarming are studied in our group: the opportunistic pathogen Pseudomonas aeruginosa and the nitrogen-fixing bean symbiont Rhizobium etli (Figure 1).

Swarming patterns of P. aeruginosa (left) and R. etli (right)
Figure 1. Swarming patterns of P. aeruginosa and R. etli.
P. aeruginosa produces a distinctive 'fingering' pattern (left), while R. etli swarming colonies are typified by flowering-like structures (right).

Previously, we identified genetic determinants affecting swarming in R. etli by screening a transposon mutant library (Braeken et al., 2008). In addition, we showed that specific R. etli quorum sensing molecules serve a dual role during swarming by acting as signaling molecules as well as biosurfactants (Daniels et al., 2006). These results inspired us to initiate a research program focused on the interaction between biological and physico-chemical processes leading to swarming migration and pattern formation in P. aeruginosa and R. etli. This multidisciplinary effort is carried out in close collaboration with the group of Prof. Jan Vermant, a leading expert on complex fluid interfaces.

Physicochemical aspects of bacterial swarming
Despite the often intricate genetic mechanisms that regulate swarming, there are also several ways in which physico-chemical phenomena could play a role in the dynamics of swarming and biofilm formation. Possible parameters intervening in these are the heterogeneity of substrates, the surface-active nature of signaling molecules, and the dependence of viscosity on the concentration of bacteria and its effect on thin film hydrodynamics. Likewise, the extracellular slime is a non-Newtonian fluid whose viscosity strongly depends on local deformation rates that will affect the spreading dynamics of the bacterial film. Finally, at high concentrations of bacteria, large-scale coherent movements of bacteria with vortex-like motions might appear due to hydrodynamic coupling, in which the collective motion of bacteria through the viscous slime drives the fluid flow. In particular, the striking ‘fingering’ patterns, formed by some swarmer colonies on relatively soft subphases, have attracted attention as they could be the signatures of instability (see Movie below).

Movie. Fingering pattern formation during swarming of Pseudomonas aeruginosa. Left: wildtype strain. Right: rhlA mutant strain unable to swarm due to a lack of rhamnolipid (biosurfactant) production. For details, see Fauvart et al., 2012.


Marangoni flow-driven pattern formation
Two different approaches of treating the bacterial film as a continuum have been proposed to explain the pattern formation. A first approach starts from the experimentally observed sensitivity of bacterial swarming to the condition of the agar substrate. A non-linear reaction-diffusion model has been proposed, in which the branching is due to the sensitivity of the system to local irregularities in the substrate. This approach successfully reproduces the patterns, but a detailed comparison of growth kinetics has not been presented. Alternatively, a parallel has been drawn with the spreading of viscous drops under the influence of a surfactant, which leads to similar patterns. Starting from the observation that several of the molecules essential in swarming systems are strong biosurfactants, the possibility of flows driven by gradients in surface tension has been proposed. Marangoni flows also lead to the observed fingering patterns. For R. etli, both the pattern formation and the spreading speed are consistent with those expected for Marangoni flows for surface pressures, thicknesses, and viscosities found experimentally (Daniels et al., 2006). However, complications due to transport of oxygen or signaling molecules could arise. Hence, further work describing the pattern formation kinetics in quantitative terms is clearly warranted. Recently, we succeeded in recording signature height profiles indicative of Marangoni flow-driven pattern formation in swarming P. aeruginosa colonies, an important step towards a better understanding of the physical processes underlying swarming motility (Fauvart et al., 2012).

Height profiling of a P. aeruginosa swarming colony
Figure 2. Time-lapse recording and height profiling of a P. aeruginosa swarming colony. For details, see Fauvart et al., 2012.

Related publications:

Fauvart M., Phillips P., Bachaspatimayum D., Verstraeten N., Fransaer J., Michiels J., Vermant J. (2012). Surface tension gradient control of bacterial swarming in colonies of Pseudomonas aeruginosa. Soft Matter 8:70-76 -- Publisher -- PDF -- Supplementary Movie 1 -- Supplementary Movie 2

Verstraeten, N., Braeken, K., Debkumari, B., Fauvart, M., Fransaer, J., Vermant, J., and J. Michiels (2008) Living on a surface: swarming and biofilm formation. Trends in Microbiology. 16, 496-506. -- PubMed -- PDF

Braeken, K., Daniels, R., Vos, K., Fauvart, M., Bachaspatimayum, D., Vanderleyden, J. and J. Michiels (2008) Genetic determinants of swarming in Rhizobium etli. Microb. Ecol. 55, 54-64. -- PubMed -- PDF

Daniels, R., Reynaert, S., Hoekstra, H., Verreth, C., Janssens, J., Braeken, K., Fauvart, M., Beullens, S., Heusdens, C., Lambrichts, I., E. De Vos, D., Vanderleyden, J., Vermant, J., and J. Michiels (2006) Quorum signal molecules as biosurfactants affecting swarming in Rhizobium etli. Proc. Natl. Acad. Sci. USA. 103, 14965-14970. -- PubMed -- PDF