Benefit and cost of siderophores

We determined that siderophore production is an appropriate cooperative trait for our experiment.Wemeasured the growth rates of a wild-type strain that produces the primary siderophore, pyoverdin (cooperator), and a mutant strain that does not (cheater), either when alone or in mixed cultures where the other strainwas present, and at a variety of iron availabilities. Both strains were grown overnight in 30ml glass universals containing 6ml standard King’s medium B (KB) in an orbital shaker (200 r.p.m.), at 37 8C. Cell densities did not differ between strains when grown under these conditions (n 1⁄4 12, 2-sample t-test: t 1⁄4 0.37, P . 0.2). Sixteen universals containing Casamino acids medium (CAA; 5 g Casamino acids, 1.18 g K2HPO4.3H2O, 0.25 g MgSO4.7H2O, per litre) media were inoculated with 10 overnight culture cells of either PA01, PA6609 or a 50:50 mixture of both, resulting in a total of 48 cultures. Sodium bicarbonate was added to each tube to create a 20mM solution, necessary for effective chelator activity. Four tubes inoculated with the three different bacterial populations (PA01, PA6609 and both) were exposed to three different iron-limitation treatments: 50mgml, 100 mgml and 200 mgml of human apo-transferrin (Sigma), a natural iron-chelator, were added. Human iron chelators (such as transferrin) bind iron, preventing non-siderophore-mediated uptake of iron by bacteria. Siderophores are also strong iron chelators, and so compete with transferrin for iron. The tubes were incubated at 37 8C in a static incubator for 24 h. Cultures were then plated onto KB agar and relative densities were determined. Siderophore (pyoverdin) production in these strains has the characteristics that make an appropriate cooperative trait for our study. First, there is a cost to individuals that can produce pyoverdin, as shown by monocultures of mutants being able to outcompete wild-type strains in an iron-rich environment (F (1,6) 1⁄4 12.61, P 1⁄4 0.01). Second, when iron is limiting, siderophore production is beneficial, as shown by populations containing wild-type strains growing to a higher density than those where pyoverdin producers are absent (F (1,21) 1⁄4 16.13, P 1⁄4 0.0006). Chelator concentration was also significant (F (1,21) 1⁄4 12.45, P 1⁄4 0.002), but not the interaction (F (1,20) 1⁄4 0.69, P 1⁄4 0.42). Third, individuals who do not produce pyoverdin are able to exploit the pyoverdin produced by the others, as shown by mutants growing to higher densities in the presence of wild type in iron-poor environments (t 1⁄4 6.37, d.f. 1⁄4 10, P , 0.005). Consequently, pyoverdin production is a costly altruistic trait that can potentially be exploited by cheaters.