At least two different variants of the novel H7N9 flu have caused disease in China, a sign the virus diverged quickly once it emerged, researchers reported.

And the genes in both variants probably came from four bird sources, with no evidence of an intermediate mammal source, according to George Gao, DPhil, of the Chinese Academy of Sciences, Institute of Microbiology in Beijing, and colleagues.

But a complete genetic pedigree of the novel virus is hampered by lack of surveillance data in wild and domestic bird populations, Gao and colleagues reported online in The Lancet.

The novel flu emerged in eastern China early this year and was identified on March 30 in Shanghai. According to the World Health Organization on April 29, there have been 126 laboratory-confirmed cases and 24 deaths attributed to the virus.

Gao and colleagues analyzed the genetic sequences of H7N9 isolates from patients in China and compared them with each other and with sequences isolated from wild and domestic birds.

On the basis of those comparisons, they reported, the hemagglutinin gene (the 'H' in H7N9) appears to have come from wild ducks in the east Asian flyway – a pathway for migration -- that covers eastern China, South Korea, and Japan.

Similarly, the neuraminidase gene (the 'N' in H7N9) also appears to have come from wild ducks; the most closely related sequences were isolated from birds in South Korea in early 2011.

The six internal genes of the virus – dubbed PB1, PB2, PA, NP, M, and NS – all cluster together with genes from H9N2 viruses isolated from chickens in China.

But analysis suggested that at least two different H9N2 viruses contributed genes – one donated the NS gene and the other all the remaining internal genes, Gao and colleagues reported.

Taken together, the findings suggest at least four sources and multiple re-assortments were needed to produce the current strain, they concluded.

Gao and colleagues also reported there are 11 amino acid changes and a 5-amino acid deletions in the neuraminidase protein that distinguish the novel human-adapted H7N9 viruses from those found in wild birds.

But not all those changes are found in every isolate from patients, suggesting that at least two "subclades" of the virus are currently circulating, they reported.

Some of the changes affect the viral receptor-binding site, used by the virus to attach to host cells, they reported.

In some human isolates, the hydrophilic amino acid glutamine is found at position 217 on the hemagglutinin protein, suggesting it might still prefer to bind to avian receptors.

But in other isolates, the glutamine is replaced by the hydrophobic leucine, which is assumed to increase the affinity for human receptors, or by isoleucine, which is even more hydrophobic and therefore likely increases the avidity for human receptors, Gao and colleagues argued.

Those changes might help account for the pathogenicity of the novel flu, since previous H7 viruses have only rarely infected humans -- only about 100 recorded cases -- and only caused a single death.

After all the re-assortments that led to the new virus, Gao and colleagues suggested, it probably began circulating in chickens in eastern China, but had low pathogenicity among the birds.

Transporting the fowl from place to place allowed it to diversify and also brought more people in contact with it, leading to the increase in "confirmed sporadic cases of human infection," they concluded.

But, they argued, gaps in the surveillance data mean it's still not clear what "unknown intermediate hosts" might be involved, although there is no evidence yet of mammals other than humans being infected.

Gao and colleagues called for greater vigilance, including wide global surveillance and a close watch on domestic-poultry-to-human transmission. (MedPage Today)