Tuesday, November 08, 2005

Ecology of viruses

From a fascinating article on the evolution of the 1918 flu, a hypothesis that sounds familiar. They sequenced avian flu from some waterfowl at the Smithsonian, collected in 1918. Turns out:
the viruses had not even evolved. Human influenza viruses change every year, mutating slightly so they can reinfect people who had just had the flu and developed antibodies against it. But birds, Dr. Slemons said, do not have much of an immune response to influenza, and so there is no particular pressure for the virus to mutate.

Another reason the viruses stay the same, he said, is that some birds live for only a couple of years and so, every year, the viruses have a new bird population to infect. Finally, he said, birds are chronically infected with lots of flu viruses at once, and all the viruses coexist peacefully.

"There are so many that there is no selective pressure on any virus," he said.
I don't know if Dr. Slemons is familiar with The Unified Neutral Theory of Biodiversity and Biogeography by Stephen P. Hubbell, but if not, he's reinventing it. Hubbell works in tropical forests, looking at the spatial distribution of tree species.

A persistent question that bugs ecologists is "Why are there so many species in the tropics?" There are about as many theories on that question as there are ecologists, and it's almost certain that there are multiple answers to the question. I won't go into all of them, but there are area effects (the tropics have the largest land area), productivity effects (more sun, more precipitation, more available energy, more biomass=more species, though that's complicated), and so on and so forth.

All of these ideas assume that species are adapting and evolving to produce new diversity, that there are new niches opening and that species are evolving to occupy them. The extent to which competition and other interactions are producing diversity is subject to regular debate.

Hubbell argues from a different angle. Most theories of directional selection involve one species interacting with every other species in the community. But trees are different. Each tree is only surrounded by about 6 trees, but there are thousands of species in an area. Each individual tree will experience selection pressures, but they will be different for each tree in a species. On average, there's no selection, or what there is is small and in a random direction. Species evolve in a "neutral" environment (hence the unified neutral theory). Populations randomly increase and decrease, one new tree replacing each tree that falls, but which species wins the lottery is random.

Mutations accumulate, and trees at each end of the species range get so different that they can't interbreed, and then the two species drift back into coexistence. New species fade into existence and fade out, completely at random.

When there are fewer species, each individual faces basically the same pressures. Species evolve in consistent ways, ways predictable by competition theory.

All this sounds very similar to the idea proposed for the relative stasis of the viral genome in birds compared to humans. In birds, you have lots of viruses, though exactly which ones occur in a given bird will vary, and any immune response varies randomly between birds and viral strains.

This is all obviously grossly simplified. I'd need an easy way to draw matrices to do it all justice, but I hope you get the idea. It's always fun when science in one field can inform another.

"Every Time I Go Around Here" by Frank Black from the album Frank Black (1993, 3:31).