By David Brown
Washington Post Staff Writer
Tuesday, May 5, 2009 6:23 PM
An analysis of small changes between samples of the swine influenza virus gathered so far suggests it probably came into being sometime before mid-September last year.
If it then started spreading person-to-person immediately, the new virus may have been present during the entire winter in Mexico. It may have become visible only at the end of the flu season -- a time when seasonal flu strains ebb -- as it continued to infect the huge number of people still susceptible to it.
If that theory holds up, it would explain away the apparent disparity in the virus's severity in Mexico and elsewhere in the world. The deaths in Mexico would simply be the expected mortality in a huge epidemic of more-or-less normal influenza.
"If this virus has been circulating in Mexico for several months, we can only presume that it wasn't killing everyone it infected or we might have been expected to pick it up straight away," said Oliver G. Pybus, an evolutionary biologist at the University of Oxford.
"It would be very surprising if this was a highly virulent virus that had been spreading since September but wasn't detected until April," he added.
This is of more than academic interest. It telegraphs to the places without swine flu that when the virus arrives it is likely to cause fairly mild illness -- unless, of course, it mutates in a way that makes it more dangerous, which is always possible.
Pybus is the leader of one of four groups of mathematical biologists estimating the age of the new swine flu strain, which is a product of a genetic "reassortment" between two, and possibly three, previous strains of flu virus.
This is done by looking at small variations in the RNA sequence that creep into influenza virus genes over time. The number (and to some extent type) of changes between samples can be used to infer how long ago all samples had a common ancestor -- which is to say, the time when they were all the same.
Other estimates put the "birth" of the new strain as August or November.
In an interview with Science magazine's online site, ScienceInsider, the head of Mexico's influenza laboratory, Celia Alpuche, said that flu peaked in late November and December, went down, and then peaked again in February. There were strains of both influenza A and influenza B circulating, as is often the case.
Overall, the season was not remarkable, she said.
"It was pretty much the same as we see every year except [there] was a prolongation of the flu season," she told Science correspondent Jon Cohen.
Although Mexico has by far the largest number of cases of illness, that country is not necessarily the place where the reassortment happened. It could have occurred in the United States or Canada -- two other places in North America where it is found -- or someplace else and been carried to Mexico.