By David N. Leff

When the great ¿Spanish influenza¿ pandemic of 1918-19 destroyed 20 million lives around the world, there was no bin Laden to blame for that holocaust. That scourge¿s prime perpetrator was a shifty pathogen since named ¿influenzavirus A.¿ Its co-conspirators were probably infected pigs and birds in China.

Since that record viral massacre, two far lesser influenza epidemics have marked the 20th century ¿ the Asian flu in 1972, and the Hong Kong flu in 1968. In 1997 in Hong Kong, a lethal avian influenza virus was transmitted directly to humans from chickens, and of the 18 known cases, six were fatal. But influenzavirus A is biding its time ¿ big time.

¿What everyone is worried about,¿ observed viral immunologist Jonathan Yewdell at the National Institute of Allergy and Infectious Diseases (NIAID) in Bethesda, Md., ¿is that something akin to the 1918 strain, which wiped out 20 million people ¿ most of them in the prime of life, between 18 and 30 ¿ is going to happen again.

¿The pandemic¿s comeback,¿ he went on, ¿is just a matter of chance. Either the same pathogen or something like it will come around once more. The virus showed back then that it was capable of generating a strain that was highly and rapidly pathogenic for people. In 1918 they couldn¿t even blame it on human terrorists. This time they will. The good news is that no one knows enough to make a really lethal influenza strain.¿

Yewdell is senior author of a paper in the December 2001 issue of Nature Medicine titled: ¿A novel influenza A virus mitochondrial protein that induces cell death.¿

¿Thanks to serendipity,¿ Yewdell told BioWorld Today, ¿we discovered a protein produced by the viral genome that people didn¿t know before. For nearly 20 years,¿ he pointed out, ¿scientists have labored under the assumption that the influenza virus comprises only 10 protein molecules that form its structure and carry out its activities. We report finding a new hidden¿ influenza virus protein that may kill immune system cells that fight it, thereby enhancing the virus¿s virulence. So now there are 11.

¿We weren¿t looking for new proteins at all,¿ he recounted. ¿We assumed the 10 known flu proteins were all there were. Our lab is primarily interested in how the immune system can recognize virus-infected cells. We¿re focused on a particular part of the immune machinery, namely CD8⁺ T cells, also known as cytotoxic (cell-killing) T lymphocytes.

¿What these T cells recognize are short bits of viral proteins, peptides and amino acids, sometimes as long as 11 or 12 residues.¿

A Suspiciously Long Gene Sequence

¿A major question in the field, which still hasn¿t been answered,¿ Yewdell continued, ¿is, Where do these hidden peptides come from? What are their sources? One hypothesis that has been around in the field for more than 15 years is that all parts of a gene ¿ viruses, tumor cells, lots of things ¿ encoded by all possible reading frames are translated at some low level. It¿s not just normal proteins, which are processed into peptides, but at some low level there¿s translation of all potential reading frames.

¿We wanted to do a systematic survey of immunogenic peptides and their ability to elicit an immune response in mouse strains,¿ he recalled. ¿And lo and behold, one of those peptides turned out to elicit one of the strongest responses in our mouse strain. This guy was in the middle of a gene sequence potentially 87 amino acids long.

¿Then I started thinking, Could this be a protein instead of a peptide?¿ So I went to the Internet, and looked at the sequence; sure enough, it was conserved. Of the 75 strains that I identified, it was present in 64. And that¿s really suspicious. The fact that it had such a long reading frame that¿s so conserved really suggested that it was a protein. We injected synthetic peptides corresponding to the ends of the protein into rabbits, and saw from their antibodies that in fact it was produced by virus-infected cells, and turned out to be a very unusual protein indeed.

¿First of all,¿ Yewdell explained, ¿it¿s located in the mitochondria, where no other flu protein has ever been found before. And second, it¿s got a very short half-life. It¿s degraded extremely rapidly, compared to other viral proteins, which basically are completely stable. This guy is degraded with a half-life of maybe 30 minutes. We then found evidence that what this protein does is enhance apoptosis in a subset of infected cells. So our working hypothesis is that what the protein evolved to do was promote apoptosis in cells that are responding to the virus infection, but actually get infected with the virus.

¿The influenza virus is not interested in killing people,¿ Yewdell pointed out. ¿That pathogen has only one interest ¿ according to Darwin ¿ which is simply to replicate. That is, not just replicating in a person, but spreading from individual to individual. Often, in fact, that¿s helped by making the virus less pathogenic. Because if it is too lethal, it¿s got no place to go. So a gene encoding the novel protein we report possibly has as its function to lessen the mortality of the virus, its pathology, so it can spread more easily. But we really don¿t know for sure.¿

A New Vaccine Potential? No And Yes

Yewdell made the point, ¿Our discovery holds no implications for a new vaccine. There are good flu vaccines already, based on neutralizing antibodies. The trouble with flu is that it¿s a very easy virus to vaccinate against. The problem is it¿s so variable. The virus stays ahead of the immune system, and nobody has solved the riddle of how to make an effective flu vaccine.¿

But on another level he suggested: ¿The high conservation of most of the determinant recognized by the cytotoxic T lymphocytes could form the basis of a new kind of vaccination, where instead of a neutralizing antibody response, you elicit a CD8⁺ T-cell response. That sort of vaccine might save your life in a flu epidemic,¿ Yewdell concluded, ¿because it gives you a partial immunity, but it wouldn¿t completely prevent infection.¿