Jeff Gore - It pays to cooperate
Yeast cells that share food have a survival
edge over their freeloading neighbors— particularly when there is bacterial
Anne Trafton, MIT News Office
November 13, 2012
Many species exhibit cooperative survival
strategies — for example, sharing food or alerting
other individuals when a predator is nearby.
However, there are almost always freeloaders in the
population who will take advantage of cooperators.
This can be seen even among microbes such as
yeast, where “cheaters” consume food produced by
their neighbors without contributing any of their own.
In light of this, evolutionary biologists have long
wondered why cooperation remains a viable survival
strategy, since there will always be others who
cheat. Now, MIT physicists have found a possible
answer to this question: Among yeast, cooperative
members of the population actually have a better
chance of survival than cheaters when a competing
species is introduced into an environment.
This experimental setup, in which yeast must coexist with a bacterial competitor, more closely mimics natural
environments, where species often have to compete with one another for scarce food and other resources.
“It’s very difficult to study these things in the truly natural context,” says Jeff Gore, an assistant professor of
physics at MIT and senior author of the new study, which appears in the journal Molecular Systems Biology on
Nov. 13. “These experiments can act as a bridge between single-species experiments and very complicated
ecosystem dynamics that are out there.”
Hasan Celiker, an MIT graduate student in electrical engineering and computer science, is the paper’s lead
Cheaters do prosper, sometimes
Gore and Celiker studied a strain of yeast that relies on small sugars such as glucose and fructose for
nourishment. When yeast cells are grown in a test tube containing sucrose, some secrete an enzyme that breaks
the sucrose down into smaller sugars, most of which diffuse away and are available to any nearby yeast cell.
In 2009, Gore and colleagues published a study showing that in a stable population of these yeast, freeloaders
dominate. Only 14 percent of the yeast cells cooperate by secreting metabolic enzymes, while the rest enjoy the
bounty of others’ work.
In the new study, the researchers wanted to bring their experimental system closer to the complexity of natural
environments. “Our approach is to start with the most simple system you can,” Gore says. “Then we’d like to try
to understand these more complicated interactions between different species, so as a baby step in that direction,
we added a bacterial competitor.”
This bacterial competitor consumes much of the sugar produced by the cooperative yeast, and also competes for
other resources, such as nutrients. After these bacteria were added, the percentage of cooperators in the yeast
population increased to about 45 percent.
This isn’t because the yeast “decide” to become more cooperative, as humans might when faced with an external
threat, Gore says. It’s determined purely by genetics. “That’s the point of doing experiments with microbial
populations: There should be a mechanism that we can understand,” he says.
In this case, the researchers found two mechanisms at work. First, in the face of extra competition for sugars, the
cooperators have a slight advantage over cheaters because they have slightly easier access to the sugar they
A related mechanism involves population density. When bacteria are present, the overall yeast population stays
smaller and is forced to spread out more, making it harder for the freeloaders to find food.
“The cheater cells can really spread when there are a lot of other yeast, because there’s a lot of sugar out there
for them. If there aren’t very many yeast, then it’s challenging for cheaters to spread,” Gore says.
The study represents an important step toward integrating interactions between species into scientists’
understanding of the evolution of cooperation, says Michael Brockhurst, a professor of biology at the University
“This suggests that the effects of species interactions on social traits in natural communities could be highly
dynamic and complex, and thus provide an area ripe for future research,” says Brockhurst, who was not part of
the research team.
Though the experiment does not precisely replicate natural conditions, it does show that it is important to try to
get as close to those conditions as possible, Gore says.
“If you just study something in isolation, such as yeast in a test tube, you might conclude that cooperation is very
unstable,” he says. “But if you look at these populations in the wild, where their densities are limited and they
have to interact with all these other species, you might come to very different conclusions for how difficult it is for
the cooperative behavior to survive.”
In future studies, Gore hopes to add further complexity to the experimental microbial ecosystems, including
varying the size of the yeast habitat.
The research was funded by the National Institutes of Health, the National Science Foundation, a Pew
Fellowship, a Foundational Questions in Evolutionary Biology grant, a Sloan Foundation Fellowship, and a Siebel