Newfound flu protein may kill immune cells. (Science News of the week).

That flu shot you might have gotten recently faces a more complex challenge than scientists had formerly realized.

A strain of the influenza A virus perpetrated the "Spanish Flu" pandemic in 1918 that killed more than 20 million people. Related strains caused two other deadly global outbreaks in the 20th century, and virologists consider a future epidemic to be a perennial threat. All of that from a germ with a molecular toolkit of only 10 known proteins.

Systematic lab work and a dash of serendipity helped Jonathan W. Yewdell of the National Institute of Allergy and Infectious Diseases in Bethesda, Md., and his colleagues discover a new influenza protein. When the immunologists were investigating how the immune system recognizes virus-infected cells, they encountered the suspicious viral protein fragment, or peptide.

They then deduced the genetic sequence needed to make this peptide and subsequently found it within a flu gene already known to produce a protein. When that gene is translated from a different starting point than the already recognized one, however, the virus could theoretically make a protein that included the newly discovered peptide. That's like dropping the first few words from a written story, moving all the periods a few words to the right, and discovering that the text contains a second tale with a different plot.

Yewdell's group determined that the gene is, in fact, read from both starting points. When the researchers infected cells with the flu strain they were studying and then looked for the candidate protein, they found that the cells produced it in large quantities.

The researchers identified the RNA code for the novel protein, dubbed PB1-F2, in 64 of 75 influenza A strains recorded in the standard database known as GenBank. Its presence in the majority of strains suggests that it contributes to the virus' evolutionary success. However, when the researchers disabled PB1-F2 in their experimental strain, the resulting virus was still able to replicate in cell cultures. That, and the fact that some strains get by without the protein, indicates that it isn't essential for survival.

The protein aids the virus in some other way, perhaps by attacking immune cells, suggest Yewdell and his colleagues in the December NATURE MEDICINE. The researchers produced a synthetic version of the novel protein and exposed several types of cells to it. In some of the cells, including human immune cells, the protein migrated into cell substructures known as mitochondria, deformed them, and ultimately triggered programmed cell death, a process in which mitochondria are known to be instrumental. Strains in which PB1-F2 production had been disabled killed fewer cells than intact strains did.

Finding the new viral protein doesn't have any immediate application to flu vaccines, but it does offer new insights into how the virus causes disease, Yewdell says.

The protein's discovery represents "a highly significant contribution to our understanding of the flu," says Robert G. Webster of St. Jude Children's Research Hospital in Memphis, Tenn. Although the precise function of the newfound protein is unclear, "it opens up a new area of research," he says.

In the same issue of NATURE MEDICINE, Robert A. Lamb and Makoto Takeda of Northwestern University in Evanston, Ill., comment that the finding illustrates one way that flu viruses can use a single stretch of genetic code to produce several proteins. Yoshihiro Kawaoka, a virologist at the University of Wisconsin-Madison, suggests that other undiscovered proteins may well lurk in the influenza A genome.

"We are especially interested in determining if the [newfound protein] is present in the 1918 pandemic strain," says Jeffery K. Taubenberger of the Armed Forces Institute of Pathology in Rockville, Md., who studies the strain that was behind that unusually lethal outbreak.