Sept. 3, 2010
By Kim Luke and Sean Bettam

A team of scientists and economists from the University of Toronto (U of T), Harvard University and the University of East Anglia (UEA) have tackled one of the biggest questions facing biological scientists: how does cooperative behaviour evolve and what prevents “cheating” among organisms living in interdependent, symbiotic relationships.  Their study of mutualisms – cooperative interactions between species in which both partners benefit – looks at what prevents cheaters from taking the benefits of mutualism and giving nothing or very little in return.

“The prevailing consensus, known as Host Sanction theory, has been that host species have evolved to punish cheaters and to reward cooperators,” says Megan Frederickson of U of T’s Department of Ecology and Evolutionary Biology.  “However, an alternative explanation known as Partner Fidelity Feedback holds that instead of hosts evolving to punish and reward, symbionts have evolved to help their hosts because a healthy host automatically feeds back benefits to the symbionts. A cheating symbiont would seem to be treated like any other environmental setback, such as infertile soil, and a mutualistic symbiont elicits the same sort of investments that are triggered by the availability of new resources, like a patch of sunlight.”

The paper, published online this week in Proceedings of the National Academy of Sciences and as the cover story in the September 7 print edition, grew out of collaboration between economists – Harvard’s Glen Weyl – and evolutionary biologists – Frederickson, Harvard’s Naomi Pierce and the University of East Anglia’s Douglas Yu.  The team used the economic theory of employment contracts, which describes how employers should reward and/or punish their employees, and applied it to the world of biology to model how hosts should reward and/or punish their symbionts and to explore ideas about how cooperation evolves.

“Cooperation between species, known as mutualism, is remarkably commonplace in nature, but the evolutionary origins of such partnerships remain murky,” said Pierce, the Sidney A. and John H. Hessel Professor of Biology at Harvard. “Since symbiotic host species essentially ‘employ’ their symbionts, it occurred to us that economic theories explaining how and why employers reward or punish their workers might give us deeper insights into symbiosis. This approach reveals that punishment is far from ubiquitous in nature.”

Weyl used employment contract theory to design two tests to probe the evolution of mutualism, and the team found that published data from two well-known pairs of symbiotic species better supported the view that symbionts have evolved to help their hosts rather than the hosts evolving to monitor and manage their symbionts.

“Every time I created a model, it was inconsistent with the natural biological predictions and with the available evidence,” said Weyl, a junior fellow in Harvard’s Society of Fellows and lead author of the study.  “The great virtue of economic theory is not exactly that it is so often right, but that its precision allows it to be easily falsified when it is not.”

The analysis by Weyl and his colleagues showed that in two classic mutualisms where host sanctions were thought to prevent cheating – one between a legume and a soil bacterium, and the other between a yucca and a moth – the evidence is actually more consistent with the competing theory, Partner Fidelity Feedback.

“The mathematical tools in economic contract theory and mutualisms make quite a natural pairing, but in a way we hadn’t anticipated,” said Yu.  “Rather than just confirming expectations, the theory let us design powerful tests to choose between hypotheses. It’s always tempting to start with the assumption that an interesting behavior, such as punishment, is the product of specific adaptations, but Glen’s model forcefully reminds us that there is already a lot of biology out there, which natural selection makes use of.”

Most mutualism involves species, such as plants, that host smaller symbionts, such as bacteria or insects. Multiple studies have observed that hosts can selectively reward cooperative symbionts and punish cheaters, such as when soybean plants direct food and oxygen only to root nodules colonized by nitrogen-fixing bacteria, and kill off nodules with unhelpful bacteria. But it has remained unclear how such symbioses evolved.

“Most work on the evolution of cooperation has considered cooperation in humans or other animals with well-developed cognitive abilities such as memory of past interactions,” said Frederickson. “This paper furthers our understanding of how cooperation evolves and persists even in plants, insects, and bacteria.”

Photo courtesy of Megan Frederickson