Amid all the political jostling over evolution being "just a theory," it is sometimes easy to forget that in reality it is.

And as a scientific theory, it is open to being tested through science and amended as new experimental evidence is presented.

The Dec. 4, 2005, early online edition of Nature presented evidence on the evolution of RNA viruses, specifically on natural selection. Researchers from the University of California at San Francisco and Pennsylvania State University showed that at least in RNA viruses, selection can occur at the population level, rather than being directed at individual variants.

Asked to comment on the fact that population-level selection flies in the face of some basic ideas about how natural selection works, Raul Andino, senior author, replied: "Tell me about it. It took 10 months to convince the reviewers."

He went on to say that "the results are counterintuitive to how we all think about evolution, because people think simplistically about mutation." In an experimental context, in which the environment changes one variable at a time, viruses can overcome changes through single mutations; but things are not so simple in a living host.

"This paper explains the evolution of the virus in a complex environment, where there is no one mutation that can compensate for every insult," Andino told BioWorld Today.

Classic genetics holds that evolution of a population occurs through selection of individuals. Random mutations either increase or decrease an individual's genetic fitness; those with beneficial mutations will reproduce at a greater rate. But another theory, known as the quasi-species hypothesis, proposes that for RNA viruses, evolution occurs through selection of interdependent viral subpopulations.

Population selection may not just occur in viruses.

"In my opinion, it is quite possible that population-level selection also occurs in other organisms, including ants and humans," Andino said. But "the ideas of the quasi-species theory are more evident on RNA virus populations because of the high mutation rates and high numbers of individuals in those RNA virus populations."

To test the quasi-species theory, the scientists first isolated a poliovirus that had a point mutation in its polymerase, leading to a reduced error rate and copies that are more similar to the original than in the case for wild-type viruses. The viral population was less infectious than wild-type polio virus. The reduced infectivity was not due to the polymerase mutation itself. When the researchers treated the high-fidelity virus with a mutagen, increasing its error rate back to wild-type levels, the virus was once again as infectious as a wild-type virus.

Further evidence that infecting the brain takes the cooperation of several viral strains came when the scientists isolated virus directly from the brain. While wild-type polio isolated from the brain was able to cross the blood-brain barrier and re-infect the brain of another animal when injected into the bloodstream, the high-fidelity virus was unable to do so. Since the virus was able to replicate in the brain when it was injected directly into the brain of another animal, the investigators concluded that what the virus lacked was cooperation from other variants.

Andino said the paper has two main points. First, "it clearly demonstrates that you need diversity in the population" of RNA viruses. And second, "diversity is not needed to generate a swarm of beneficial mutants, but for what we call positive cooperation."

He added that the positive cooperation "is like an army - maybe one guy is good at running, another at planning and another at fighting. And together they are able to conquer."

Asim Dasgupta, professor in the department of microbiology and immunology at the University of California at Los Angeles, told BioWorld Today that the conclusions Andino and his colleagues drew from their data seem valid and added that "the idea that different viral variants within the population could provide different functions (such as colonization, immunemodulation, crossing the blood-brain barrier etc.) in the host ultimately leading to successful viral pathogenesis is certainly novel."

Apart from their theoretical significance, the results also suggested that the replication error rate for RNA viruses is fine-tuned. Increased error rate can lead to what is termed "error catastrophe" in the quasi-species hypothesis - too many errors will drive the virus into extinction. But too few errors also are a problem for the virus. They decrease its infectivity.

Andino said his findings suggested that determining the diversity requirement for viruses could be useful for vaccine development; it also meant that "we need to start thinking about how to measure diversity in viral populations" that are easier than cloning and sequencing, which is not a practical large-scale approach.

Andino also said that if a mutagenesis strategy could be found that specifically works on RNA viruses - no small task - that could possibly be used in antiviral development. His research also suggested that finding a way to restrict the population diversity might render polio virus infection less devastating, since only the worst infections cause paralysis.

"The polio virus does not care about entering the brain and causing paralysis," Andino said. "All it wants to do is to go from host to host. That's where the selection pressure is."