Extracellular polysaccharides are compounds secreted by microorganisms into the surrounding environment and which are important for surface attachment and maintaining structural integrity within biofilms. They have been suggested to be metabolically costly to produce, and because they are secreted, to act as co-operative shared resources within biofilm communities. These assumptions have, however, not been experimentally well-examined. Here we empirically test the cooperative nature of the PSL polysaccharide, which is crucial for the formation of biofilms in Pseudomonas aeruginosa. We show that: (1) PSL provides population level benefits in biofilms, for both growth and antibiotic tolerance; (2) the benefits of PSL production are social and are shared with other cells; (3) the benefits of PSL production appear to be preferentially directed towards cells which produce PSL; (4) cells which do not produce PSL are unable to successfully exploit cells which produce PSL. Taken together, this suggests that PSL is a social but relatively non-exploitable trait, and that growth within biofilms selects for PSL-producing strains, even when multiple strains can interact (low relatedness).

The growth and proliferative success of many bacteria, including human pathogens, depends upon their ability to form biofilms in their respective environmental niches. Biofilms are multicellular three dimensional biomass structures, held together by extracellular matrix molecules that encapsulate cells and cause them to aggregate. These extracellular polysaccharides (EPS) which are secreted by the bacteria, typically function as adhesins that are used to attach to a surface and to maintain the three-dimensional biofilm structure, and sometimes aid in protection against a variety of stresses, including dehydration, antibiotics, and predators. The production of EPS represents a problem from an evolutionary perspective 3, because it appears to be a type of co-operative behaviour that can potentially provide a benefit to all cells in the community, and not just to those that produce EPS. Consequently, the question arises: “what prevents the invasion of potential cheats that do not produce EPS?”.

A possible solution to this problem is that EPS production may not be an exploitable co32 operative trait. Xavier & Foster showed, in an individual based simulation, that if the production of EPS facilitated growth into areas where nutrient availability was greater, then EPS producing lineages could spatially smother and outcompete non-producers. In this case, EPS production was costly, but the benefits were preferentially provided to nearby cells, which were likely to be EPS-producing clone mates. Some empirical support for this particular mechanism has been demonstrated in Vibrio cholerae, where it has been shown that EPS producing lineages are able to displace non-producers. In addition, EPS producers in V. cholerae are also able to share other secreted public goods with each other, which provides another benefit that is preferentially directed towards other EPS producers. However, the generality of these explanations for the evolutionary stability of EPS production remains unclear, and the work on V. cholerae represents the only empirical study to measure the social costs and benefits of EPS production. In addition, EPS produced by other bacterial species can vary greatly in both their chemical structure and the biological roles they play within biofilms. Furthermore, many species produce more than one type of EPS that are sometimes but not necessarily co-regulated. This means that there may be differences in the social nature of different types of EPS, and for different bacterial species.

Pseudomonas aeruginosa is an opportunistic pathogen that causes various biofilm infections such as chronic respiratory infections of cystic fibrosis, keratitis, and chronic wound infections. P. aeruginosa is known to produce at least three different types of EPS as major components of its biofilm matrix: alginate, PEL, and PSL polysaccharides. Alginate production is inversely regulated with PSL and is not expressed to high levels in the majority of non-CF isolates. In contrast, PSL is expressed in most P. aeruginosa natural and clinical isolates. PSL is a crucial adhesive scaffolding component of the biofilm matrix, promoting both cell-to-cell interactions and surface attachment. Here we test the social nature of PSL and find that (1) PSL production is not metabolically costly to P. aeruginosa cells; (2) PSL+ strains are significantly fitter than PSL- strains in mixed culture biofilms and PSL- strains cannot act as social cheats; (3) the benefit of producing PSL is enhanced when there are many PSL- cells present; (4) biofilms containing a high proportion of PSL- cells are more susceptible to antibiotics; (5) relatedness in biofilms does not matter since PSL+ strains are favoured in conditions of high and low relatedness. More generally we highlight that not all components of the biofilm matrix should be considered as shared resources.