'Rock-paper-scissors' preserves biodiversity in bacterial neighborhoods

The hand game of "rock-paper-scissors" is perpetual one-upmanship: rock crushes scissors, scissors cut paper and paper covers rock. In the July 11 issue of the journal Nature, researchers at Stanford and Yale report that certain bacteria can play their own version of that game. Populations chase each other around the petri dish in a set of incessant skirmishes that lack a clear victor but conserve biodiversity of the overall ecosystem. But biodiversity breaks down in the uniform environment of a well-mixed flask - demonstrating that spatial separation may be necessary for different populations to coexist.
"The scale at which organisms interact and disperse can have profound effects on the maintenance of biodiversity," says co-author Brendan Bohannan, an assistant professor of biological sciences at Stanford. "Organisms exist in neighborhoods in nature, and historically that's been overlooked by many ecologists. And the fact that they exist in neighborhoods has profound implications for the maintenance of biodiversity."
Other authors of the Nature paper are biological sciences graduate student Benjamin Kerr and Professor Marcus Feldman, both of Stanford, and Associate Professor Margaret Riley of Yale, who spent her sabbatical at Stanford from 1999 to 2000. Feldman, the Burnet C. and Mildred Finley Wohlford Professor in the School of Humanities and Sciences, is a theoretical biologist. Bohannan is interested in both theoretical and experimental aspects of microbial ecology, and Riley is a microbiologist interested in ecology and evolution. Kerr brought this group together, creating computer simulations with Feldman and Bohannan to explore these ideas theoretically and developing microbial systems with Bohannan and Riley to test these ideas experimentally.
The researchers worked with three populations of the world's best-studied bacterium, Escherichia coli. One population carries a natural genetic element (called a plasmid) that serves as the blueprint to make an antibacterial toxin, or antibiotic, called colicin. The plasmid also encodes a protein that renders the bacterium immune to its own colicin and a protein that causes colicin to be released when the cell is under stress.
Colicin-producing bacteria can kill bacteria that are sensitive to colicin. But occasionally, sensitive bacteria mutate to become resistant. To understand why these three populations - colicin-producing, colicin-sensitive and colicin-resistant - are entwined in a rock-paper-scissors relationship requires understanding two facts about their metabolism. First, the colicin-sensitive bacteria have a molecule on their surfaces that lets in colicin - but it also lets in certain nutrients. That molecule has been mutated, or changed, in the colicin-resistant bacteria, so they no longer let in colicin. This results in a reduction in the molecule's ability to let in certain nutrients as well. That means sensitive bacteria grow faster than do resistant bacteria. Second, the resistant bacteria grow faster than do the colicin producers because they avoid the metabolic cost of carrying the genetic machinery to encode the toxin. The result? Colicin-sensitive bacteria displace colicin-resistant ones; colicin-resistant bacteria displace colicin producers; and colicin producers displace colicin-sensitive bacteria. "[The community] satisfies a rock-paper-scissors relationship," the authors write.
But that triangular relationship alone is not enough to preserve biodiversity, they found. Biodiversity disappeared when the strains were grown in environments that allowed them to interact more thoroughly. Instead of gangs of E. coli fighting for local turf on a petri dish, mixing populations on the petri dish or in a flask produced the bacterial equivalent of world war. Greater dispersal and interaction of populations turned competition for an ecological niche into a winner-take-all battle. That quickly led to demise of all but the resistant strain, as colicin producers killed sensitive strains but colicin-resistant strains outgrew colicin producers.
All three of the populations with which Bohannan and colleagues work are found in nature. "One of the mysteries that microbiologists were interested in was why you could find all three populations in nature - because if you took all three into the laboratory and put them in a flask ... very soon you'd have just one type left," Bohannan says.
Localizing the interactions preserves genetic diversity, Bohannan says. "Say you have a genotype that's not a particularly strong competitor. Having interactions and dispersal localized might allow it to persist over time because it essentially has a refuge from its competitors."
The researchers are now exploring whether genetic diversity can persist throughout the long time scales required for evolution. One new idea they are testing is "survival of the weakest," a counterintuitive theory that suggests in a rock-paper-scissors relationship, it isn't necessarily advantageous to grow fast. "By growing faster, you end up putting more pressure on the next member of the triplet, which then frees the third member to grow faster, which then comes back negatively and affects you," Bohannan explains.
The Nature paper demonstrates the importance of the scale of interaction and dispersal on maintaining biodiversity, Bohannan says. "Understanding the processes that maintain and generate biodiversity is crucial to conserving biodiversity," he says. "Whether this particular process is important in a given system is something that ecologists working in the field have to establish."
Rock-paper-scissors may be common in many ecosystems, such as coral reefs, where scientists first observed the dynamic. The strategy of producing toxins to kill or slow the growth of a competitor is called allelopathy. It occurs in many plants, marine invertebrates, fungi and essentially every major bacterial group. The toxins that one population of bacteria uses to poison others are exploited in medicine as antibiotics.
"There are interests in developing colicins as commercial pharmaceuticals," Bohannan says. That in fact is a research interest of co-author Riley. "Our work potentially could have implications for the ecology and evolution of bacteria that produce antibiotics that are of potential medical importance." This work also has the potential to help answer one of the biggest questions in ecology and evolution: Why are there so many different types of organisms out there? The localization of dispersal and interaction, coupled with rock-paper-scissors, is an example of one process that might prove important in maintaining all that diversity.
For at least a decade, ecologists have made an assumption in their theories and models that may need revision in light of the findings put forth in the Nature paper. Their so-called "mean field assumption" posited that organisms interact with each other in proportion to their abundances across space. "It's essentially assuming that the organisms are mixed up together, like our bacteria in our flasks," Bohannan says. "Because of advances in theory contributed by physicists and increases in computational ability, we're now able to think about how populations interact in space and relax this assumption. We can assume now that organisms actually have a relationship with place, and they interact more with their neighbors than they do with organisms farther away. And that's a big revolution in ecology."
Bohannan and colleagues are now crafting experiments to explore the role of neighborhoods in ecology and evolution. "I'm excited about this project," he says. "It is an example of what I think Stanford is very good at, which is bringing together people with very different points of view to ask big questions. This project brought together theoretical biologists, ecologists, microbiologists from other institutions, graduate students and professors to ask important questions in novel ways."

Stanford funded the work presented in the Nature paper. The National Science Foundation is now funding an extension of this work to explore how evolution occurs in bacterial environments when dispersal is local versus global.

By Dawn Levy

The paper by Kerr et al. gives a new impetus to such
investigations, by stressing the importance of the geometry of
neighbourhoods. Many habitats resemble the surface of a pizza
more than a well-stirred bowl of soup”
M. A. Nowak and K. Sigmund, ‘news and views’ in Nature 418, 138 (2002)


Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer