Watching evolution on fast-forward

How do new species evolve and adapt to changes in their environment? The human lifespan is too brief to observe evolution occurring in vivo, while the fossil record is too fragmentary to track species in transit between ancestral and extant life forms.

But Professor Allen Rodrigo's team has spent the past half decade observing evolution in action, in real time.

The University of Auckland bioinformatician described his research team's work to the 2004 BioInfoSummer symposium at the Australian National University in December, hosted by the university's Centre for Bioinformatics Sciences.

Their subject isn't alive, in the strict sense, but it inhabits a hostile environment in which extreme selection pressures drive it to evolve more rapidly than any life form on the planet.

Their model is the Human Immunodeficiency Virus, the agent of AIDS, a pathogen notorious for its ability to mutate rapidly in its victims, under selection pressure from the immune system and potent anti-retroviral drugs.

Rodrigo's research treats the human body as a microcosm of global biosphere. He uses powerful computer algorithms to track the evolution of new strains arising from the virus' error-ridden replication process.

Computers facilitate the task of tracing evolutionary relationships as new strains rise, flourish and decline in response to intense selection pressure. The software tools are developed by PhD students who combine computational and biological skills.

"We can study the changing history of the virus in a person just as we can study the history of organisms around us," Rodrigo said.

"We're developing [bioinformatics] tools to study the changing histories of the virus in individuals. We sample the time dimension in a way that is otherwise extremely difficult with organisms that accumulate mutations at a lower rate.

"Certain changes in the amino acid sequences of viral proteins tell us whether the virus has been invading the immune system, because the changes occur in response to immune pressure."

Such studies have practical outcomes in terms of tailoring drug therapy to patients' needs - "We've done a study to determine whether patients who exert strong immune pressure on the virus survive longer than those with a weaker response.

"We found a correlation between those with strong immune response and long survival. In long-surviving patients, the viral genes showed greater evidence of being under positive selection than in those who died during the course of the studies."

Rodrigo said his team's tools can also detect changes in the virus's rate of mutation. Some researchers believe the virus stops replicating during very aggressive anti-retroviral therapy, so its rate of evolution drops to zero.

"How do you detect that? You sample the virus before and after the anti-retroviral therapy to see if it has mutated.

He said that when the patient's viral load drops after drug therapy; it disappears from the bloodstream and may take refuge in tissues or cells where it is relatively invisible to the immune system, and inaccessible to drug therapy.

Rodrigo likens the tissues that provide a safe haven for the dormant virus to the environmental 'refugia' where plant and animal species retreat to ride out extreme climate changes, such as glacial periods or epochal droughts. During these times, there is intense selection pressure for small, remnant populations of plants and animals in refugia to maintain genetic homogeneity, so they can survive in their restricted environment. The same appears to be true of HIV.

They have detected what seem to be bursts of rapid speciation in the virus, reminiscent of the controversial 'punctuated equilibrium' pattern predicted by evolutionary theorists Stephen Jay Gould and Niles Eldridge in the 1970s.

Rodrigo says it is unclear whether the pattern results from rapid speciation after a period of evolutionary stasis, or whether the different virus strains emerge from multiple compartments like the testes, lungs, central nervous system, liver and vaginal mucosa.

Their findings have implications for vaccine development - "If the virus is being transmitted by intercourse, you want to design a vaccine targeted to the strains found in the vaginal mucosa," he said.

"But we haven't found evidence for strong degrees of genetic differentiation, which indicates that the viruses found in these compartments come from transient populations that contract and expand again.

"It makes sense in terms of what the immune system is doing. Because immune system cells become infected with the virus, they carry it wherever they go, releasing pools of new virus particles that may establish in different compartments.

"If you think about the human host as an ecosystem, there's a great deal of niche expansion by new virus strains. The viruses may lie quiescent in dormant cells that may have a half-life as long as 20 years.

"The notion of the virus forming reservoirs becomes very important - if a person is infected today, the infection never goes away."

Rodrigo said that, during the course an average patient's infection, the virus evolves as much as vertebrates have evolved in the past 500 million years.

It's as if zoologists could place evolution on fast-forward, and watch the world's vertebrates diverging from the archetypal bony beast of Canada's Burgess Shales, Marrella splendens in only 10 years.