Cooperation is seen in every corner of life from microbes to humans, many times with no obvious advantages to those that provide it at high costs. Given the existence of “freeloading cheaters” ready to exploit the resources of those cooperating, why is it that cooperation persist? In an article now published in the journal Current Biology Nogueira and colleagues suggest that in bacteria this can result from highly mobile genes that “jump” from one cell to the next carrying the cooperative traits, effectively turning everyone into a cooperator. They also show that, at least in Escherichia coli (E. coli), this new population remains stable through “punisher” genes that impose a mafia-like strategy of “cooperation or death”, ensuring that the new cooperators do not revert to freeloading. The work – by shedding some light on the complex interactions of microbes that ultimately determine which bacteria thrive or disappear – can have important implications for both human health and economy.

Electron miscrope photo E. coliScientists now agree that cooperation thrives if it benefits the cooperator or/and its relatives. The logic behind this last behaviour resides in the fact that to benefit relatives - which are genetically very similar to the cooperator - is to increase the probability that shared genes, including the cooperative traits, will pass to the next generation. Natural selection that preserves cooperative behaviours benefiting relatives - sometimes even at a cost to the cooperator’s survival and/or reproduction – is called kin selection. Classic examples of kin selection in action are “watchers” - individuals that, at the expense of their own security, guard and raise the alarm in the community - or the sterile workers found in ants and honey bees colonies. So cooperation is affected not only by its costs and benefits, but also by the genetic similarity between cooperator and those benefiting from the behaviour, with higher relatedness increasing the probability of cooperation occur.

Following this idea Teresa Nogueira (Centre for Environmental Biology, Univ. Lisbon, and Superior School of Health Technology of Oporto, Portugal), Eduardo P.C. Rocha and colleagues, at the Institute Pasteur and the UPMC Univ. Paris 06 (Paris, France) while trying to understand cooperation among bacteria wondered if horizontal gene transfer - a non-sexual process, very common in bacteria, where genes “jump” from one individual to another – by increasing the genetic similarity between “infected” individuals, could lead to cooperation via kin selection. Using mathematical tools and adding the effect of horizontal gene transfer to costs, benefits and genetic similarity the researchers confirmed that – theoretically at least – highly mobile genes carrying cooperative traits should promote cooperation, which is then be maintained by kin selection.

To confirm this hypothesis Nogueira and colleagues looked at E.coli, which is an abundant microorganism in the human gut flora, where the bacteria typically live in a mutually beneficial relationship with humans. However, changes in its social interactions with other microbes and/or the human host can turn E. coli into an extremely virulent organism, making it a particular interesting subject for this study. Not only that, but many of E. coli vital functions depend on secreted proteins (the so called secretome) that are easily exploited by other microbes, turning E. coli into a potential “collaborator”, even if an unintentional one.

The work now published analyses 21 E. coli genomes and their secretome genes and starts by finding that only a very small percentage of them are part of the “core genome” – those genes shared by all the strains of the species so, supposedly, those linked to functions crucial for survival – what agrees with the idea that the secretome contains cooperative traits, since, by definition, cooperative genes can be easily lost by natural selection.

Next Nogueira and colleagues try to identify the proteins of the secretome by comparing them to a large known sample from the human gut. And in fact, several of them were potential cooperative traits but, and even more interesting, they also found many E. coli secretome genes in other bacteria. This agrees with Nogueira’s prediction that cooperative traits accumulate in highly mobile genes that, by “moving around”, increase the genetic similarity of previously unrelated bacteria (in this case, even across species). This conclusion was further supported by the fact that large part of the secretome was found to be coded by a piece of extra-chromosomal DNA called plasmid, which is, simply, the most mobile part of the whole bacterial genome

Finally, secreted proteins are costly as they are not recycled and - if Nogueira’s hypothesis was correct - as cooperative traits liable to be exploited by cheaters, they should be under intense selection pressure. And indeed the secretome was found to comprise the proteins least expensive to produce in the entire organism, consistent with cooperation costs.

A last question remained - if highly mobile cooperative genes, by jumping between individuals, turn everyone into a cooperator, in the same way their mobility should lead them to be easily lost, creating, yet again, a new freeloading population. So how is this avoided? One way would be by imposing punishments and this possibility led Nogueira and colleagues to search for proteins known to stabilize gene integration by “punishing” the organism if the gene is lost.

And indeed they found two such mechanisms - the “Restriction Modification” and the “Bacterial Toxin–Antitoxin” system. Both systems work as a complex of two genes where one provokes the death of the host if the other goes missing. In E. coli these systems were found next to the secretome genes suggesting that the stability of the new cooperator population is maintained by making the cost of losing the cooperative trait higher then the benefit of becoming a freeloader.

In conclusion, Nogueira and colleagues’ work showed that the secreted proteins of E. coli behave as cooperative traits - they are part of the “disposable” part of the genome, show signs of intense selective pressure and some were even identified as potential cooperative traits. In agreement with the researchers’ model most of the genes coding for the secretome are located in the highest mobile part of the bacterial genome and found to be shared by other bacteria. This last increased genetic similarity allows the cooperative traits to stay in the population through kin selection, while “punisher” genes further assure the stability of the new population via a “cooperate or die” policy.

These results strongly suggests that - like Nogueira proposed - cooperation among bacteria can be enforced by extremely mobile genes containing the cooperative traits that jumping between individuals turn them into cooperators. This new population is then maintained by kin selection and punishment.

Microbes are used in a variety of jobs important for humans, from cleaning petroleum oil of the oceans or wastage treatment to food growth, and, of course, they are crucial agents of disease and health. They are also extremely social organisms normally living in a mix of many different species - the flora of the human gut is a good example – with different populations growing more or less in result of the social interaction between them and/or the host. Nogueira, Rocha and colleagues’ work is an important step in the understanding of these complex relationships and might contribute one day for better ways of manipulating bacterial growth, whether with the intention of stopping pathogenic populations or to stimulate beneficial ones. In the specific case of E. coli this knowledge is particularly important as many of its cooperative traits are virulence factors and responsible for turning this normally innocuous bacteria into a life-threatening dangerous pathogen.