Survival of the Cheaters

In evolution, it's often said that only the strong survive. But among viruses, biologists have found, natural selection sometimes favors the simply devious. This happens, they report in tomorrow's issue of Nature, if the rules of the game resemble a mathematical puzzle called the "prisoner's dilemma."

Evolutionary biologists Lin Chao of the University of California, San Diego, and Paul Turner, who is now at the Universidad de Valencia in Spain, wanted to know what happens when two or more viruses infect the same cell. They grew a virus called phage phi-6 on a "lawn" of bacteria, making sure there were many more viruses than bacteria. They believed the viruses would trade genetic information to produce new strains--the viral equivalent of sex. Over time, natural selection would favor the "fittest" strains, which had the highest rate of reproduction. Instead, the cultures were overrun by a mutant strain that reproduced less rapidly than the original virus. "That went against everything we expected," says Chao.

The explanation, Chao and Turner believe, is that the new strain cheats. Normally, viruses share enzymes and nutrients with each other inside a cell. The mutant viruses don't share--although they're perfectly willing to accept donations. For any individual virus, this is a good strategy. Thus in the long run, Turner says, "only the cheaters survive." But that's bad news for the virus population as a whole. With an escalating proportion of selfish viruses, the population reproduces less prolifically.

Mathematicians call this situation the "prisoner's dilemma," because it resembles the plight of guilty prisoners who know they will get reduced sentences if they inform on each other, but also risk being tattled on themselves. To check if the viruses play this game, Chao and Turner studied the evolution of the normal and mutant strains over time and computed the actual "payoffs" for each strain: how much advantage a cheater gains when it shares a cell with a noncheater, how much the noncheater loses in that situation, and how much both viruses lose if they are both cheaters.

Although the prisoner's dilemma explains why the "selfish" form of phage phi-6 takes over when several strains infect the bacteria at the same time, experts say it leaves one question unanswered: Why, then, do cooperating viruses exist in nature? Perhaps some other mechanism favors cooperation, or in biological parlance "group selection." According to evolutionary biologist Jim Bull of the University of Texas, Austin, this type of experiment could be a good way to learn more about the mechanism.