TORONTO - Dutch scientists have reported they have found what was thought to be a key mutation in some swine flu viruses from the Netherlands, a change many virologists feared would give the viruses the ability to cause more severe disease.

But so far the evidence seems to suggest this mutation does not make the new H1N1 virus more virulent, the researchers said Tuesday.

The change, at position 627 on the PB2 gene of the virus, is known to increase the ability of flu viruses to replicate; prolonged viral replication can lead to more serious illness. The mutation has been found in all known human flu viruses, including the three that caused the pandemics of the last century.

"Everybody predicted that this mutation is going to have a big impact on virus replication of the new H1N1," said Dr. Ron Fouchier, one of the authors of the report and a molecular virologist at Erasmus Medical Centre in Rotterdam.

"If you would have asked me three months ago, I would have bet my car on it. But nobody placed that bet because everybody was sure that it would increase (virulence)."

If they had, it seems they might have been able to claim the keys to Fouchier's SUV.

Three people either known or suspected of having been infected with the mutated H1N1 viruses suffered only mild disease. And ferrets infected with a laboratory synthesized H1N1 virus with this change also did not suffer more severe disease. Ferrets are the standard animal model for human flu.

"Given the information that we have at present, we have no indication for increased virulence," said Dr. Marion Koopmans, chief of virology in the infectious diseases laboratory of the National Institute of Public Health, The Netherlands.

"This is a mutation that is in the textbooks as something to look out for, but whether it really confers something to these (H1N1) viruses remains to be seen."

Koopmans, Fouchier and a number of colleagues disclosed the surprising findings through ProMed, a website and mailing list that serves as an early warning system for infectious diseases developments. It is closely scrutinized by scientists and public health officials in the infectious diseases sphere.

The Dutch scientists reported finding two viruses with this change that appear to have been transmitted between mid-July and mid-August in the West Frisian Islands in the north of The Netherlands. The area is a popular destination for Dutch and German campers.

One of the mutated viruses was recovered from a male who had been there and who started developing symptoms on Aug. 9. The second was found in a girl who hadn't been to the area, but whose sister had been camping there at the time. The sister was also sick, but there was no specimen from her to test. Koopmans said the working assumption is that the sister who went camping was also infected with this virus.

The first virus was only discovered in mid-September, when it made its way to Koopmans' lab. An investigation at that point showed the male and the sister had been part of a group of 24 who shared two tents on the island. Most of the members of the camping party reported having been ill.

Koopmans said officials have looked at specimens from the area and from the regions from whence the campers came, but haven't found more viruses with this change.

"There's no evidence yet that this virus has spread any further in Holland," Fouchier said. "Of course we're currently still looking for it. Every virus we get our hands on we check (position) 627. But we haven't found any more."

Labs have been looking for this mutation from the moment the new H1N1 virus was fully analyzed and it was seen it didn't have the same amino acid at position 627 as other human viruses have.

Some scientists even suggested the virus might not be fully adapted to spread among humans because it didn't have this change, but instead had an amino acid at position 627 that is normally seen in avian flu viruses. The pandemic virus, which is a never-before-seen hybrid of swine, avian and human genes, has an avian PB2 gene.

Dr. Richard Webby, head of the World Health Organization's influenza collaborating centre at St. Jude Children's Research Hospital in Memphis, Tenn., said there is good evidence this mutation is associated with adaptation of avian influenza viruses to humans, but the proof that it is linked to severity of disease is less clear.

Viruses with this change show increased virulence in mice and sometimes in ferrets, but not always, he said, suggesting the ferret data probably are more reliable. "I think this is one instance where mice are probably lying a little bit," Webby said.

In some flu viruses this change is known to allow the virus to replicate at cooler temperatures, meaning they can infect the upper airways, rather than the warmer deep lung area preferred by avian flu viruses.

That might actually be a good thing with this H1N1, Webby said, noting autopsies have shown that in severe cases the pandemic virus wreaks havoc deep in the lungs.

Dr. Nancy Cox, who heads the influenza division at the U.S. Centers for Disease Control, warns people should not take too much comfort from the fact this change doesn't seem to make the virus more virulent at this point.

"We just know that influenza can change in unpredictable ways through mutation and reassortment," she said, referring to the process by which flu viruses swap genes with each other.

"The unexpected can arise, and arise very quickly. So we shouldn't write this off. It is causing hospitalization. It is causing fatalities. And in every single case that you hear about, it's a tragedy."