在涉及混杂细菌菌落的实验演化的实验中，病原体Pseudomonas aeruginosa的个体被发现彼此之间进行合作，以便能够以铁为食。这种能力是细菌感染中生长和毒性的一个关键决定因素，因为铁对细菌生长来说经常是限制其生长速度的成分。寄主积极持留来自寄生虫的铁，而这些寄生虫可以通过合作的方式来克服寄主的防卫。这种现象取决于相关细菌间的合作，是一种亲情选择。病原体细菌间的合作很可能是决定细菌毒性的一个因素，但也可能成为新的干预方法的一个目标。
Cooperation and competition in pathogenic bacteria
Explaining altruistic cooperation is one of the greatest challenges for evolutionary biology1-3. One solution to this problem is if costly cooperative behaviours are directed towards relatives4, 5. This idea of kin selection has been hugely influential and applied widely from microorganisms to vertebrates2-10. However, a problem arises if there is local competition for resources, because this leads to competition between relatives, reducing selection for cooperation3, 11-14. Here we use an experimental evolution approach to test the effect of the scale of competition, and how it interacts with relatedness. The cooperative trait that we examine is the production of siderophores, iron-scavenging agents, in the pathogenic bacterium Pseudomonas aeruginosa15-17. As expected, our results show that higher levels of cooperative siderophore production evolve in the higher relatedness treatments. However, our results also show that more local competition selects for lower levels of siderophore production and that there is a significant interaction between relatedness and the scale of competition, with relatedness having less effect when the scale of competition is more local. More generally, the scale of competition is likely to be of particular importance for the evolution of cooperation in microorganisms, and also the virulence of pathogenic microorganisms, because cooperative traits such as siderophore production have an important role in determining virulence6, 9, 17-19.
Figure 1 Scale of competition and kin selection theory. We have plotted the effect of the scale of competition on selection for an altruistic trait, from the incorporation of Frank's3 scale of competition parameter, a, into a classic tragedy of the commons formulation3,17. This allows a simple and general graphical representation of the theoretical predictions. The y axis gives the evolutionary stable allocation of resources to a cooperative trait that is costly, but provides a benefit locally3,25. The scale of competition varies from global (a = 0) to local (a = 1). The different lines represent relatively high (r = 0.75) and relatively low (r = 0.25) relatedness. Higher levels of cooperation are favoured when relatedness is higher (higher r), and competition is more global (lower a). Furthermore, there is an interaction between scale of competition and relatedness: as competition becomes more local, the influence of relatedness on selection for cooperation is reduced. In the extreme, if a = 1, then competition is completely local and so kin selection cannot favour altruism3,22. The same conclusions can be reached using Queller's11 approach of allowing for local competition, or with a model specifically developed for siderophore production17.
Figure 2 Experimental design. We varied relatedness between interacting individuals by initiating each subpopulation with either a single bacterial clone (relatively high r) or two bacterial clones (relatively low r). We varied the scale of competition by either mixing the cultures from all of the subpopulations in a treatment before plating, and then transferring random colonies from this single plate to initiate new subpopulations (relatively global competition, lower a), or by allowing every subpopulation in a treatment to provide equal numbers of colonies to the next generation (relatively local competition, higher a). We use dark green to symbolize the siderophore-producing cooperator, white to symbolize a cheater that does not produce siderophores, and light green for a mixture of cooperators and cheaters.
Figure 3 The evolution of cooperation in response to relatedness and the scale of competition. The proportion of cooperating individuals who produce pyoverdin siderophores is plotted against time. The different lines represent relatively high (solid line) and low (dashed line) relatedness. The different symbols represent relatively global (circle) and local (triangle) competition. Each of the four treatments was replicated four times, and standard errors are shown for the final time point. Time is measured as transfers, between which cultures were allowed to grow for 24 h. Cooperation is favoured by higher relatedness and more global competition.
