Published in the Proceedings of the National Academy of Sciences, the study also suggests that a thorough characterization of circulating flu viruses may provide clues to an emerging pandemic before it hits.
According to the study's results, two genes from the 1918 influenza virus, which killed an estimated 50 million people, would have been present in human and swine flu viruses at least 6 years earlier.
Its authors say that during the intervening years, swine and human flu viruses would have swapped genes with avian viruses, ultimately giving rise to the dangerous assortment of genes carried by the 1918 virus.
"This work suggests that the generation of pandemic strains and the adaptation to humans could be much more involved than was previously thought," Nature magazine quoted Raul Rabadan, a biomedical informatician at Columbia University College of Physicians and Surgeons in New York, who was not involved in the study, as saying.
"It reinforces the idea that systematic surveillance, not only in humans but in other mammalian and avian hosts, is key to identifying possible pandemic strains and their future evolution," Rabadan added.
Yi Guan of the University of Hong Kong and Robert Webster of St. Jude Children's Research Hospital in Memphis, Tennessee have revealed that they compiled the available data on known bird, swine and human flu viruses, and created family trees based on DNA sequence information.
The researchers say that they estimated the amount of time it would take to accumulate the differences in DNA sequences found in human and swine viruses, which enabled them to determine that a precursor to at least one 1918 flu gene was present in mammals before 1911, and anther had been circulating in humans since the 19th century.
Their findings contrast previous hypotheses that the human 1918 flu strain had evolved directly from a bird flu virus, and instead suggest that an avian strain entered pig and human populations and, then, swapped genes with mammalian flu viruses before becoming a pandemic.
The researchers reckon that elements of the 1957 pandemic flu virus, which are considered to be a mosaic of human and avian flu genes, were probably introduced into human populations two to six years before the pandemic.
Even though the researchers had completed these analyses before the current pandemic swine flu strain made its mark, they insist that their findings might have implications for future pandemics.
They argue that the results from 1918 and 1957 pandemic flu suggest that public-health authorities should track the sequences of all influenza virus genes in emerging strains, rather than focusing largely on the gene that encodes the haemagglutinin' protein, which is critical for vaccine production, as is the current practice.
Meanwhile, University of Arizona researcher Michael Worobey has revealed that his own analyses have also suggested that human and swine forms of H1N1 shared a common ancestor years before 1918.
He, however, still remains unconvinced by the series of genetic swaps proposed by the paper.
"Using different assumptions, you can get very different results. We can't say much for certain about these events given the current data," he says.