By David N. Leff

A young Eskimo woman who died suddenly 83 years ago went to her grave harboring the secret of the greatest serial killer in human history. Her death was one of 72 in the remote Alaskan village of Brevig Mission. She and her stricken neighbors were buried 6 feet deep in the Arctic permafrost.

Fast forward 79 years from November 1918 to August 1997. That was when a retired pathologist named Johan Hultin exhumed the woman's remains from its frozen sepulcher. He found her lungs well preserved, and shipped their tissues to Jeffery Taubenberger, chief of molecular pathology at the U.S. Armed Forces Institute of Pathology (AFIP) in Washington. (See BioWorld Today, Feb. 9, 1998, p. 1.)

Hultin's gift was a personal contribution to Taubenberger's long-term project of sequencing the genome of the influenza virus that perpetrated what he calls "the most extensive infectious disease in history." Those 72 Alaskan Eskimo victims were part of a larger statistic: The 1918-1919 flu pandemic took the lives of about 675,000 people in the U.S., plus anywhere from 20 million to 40 million dead the world over. Nowadays, some 20,000 Americans a year - most of them elderly - succumb to flu.

Taubenberger is senior author of a paper in the current Proceedings of the National Academy of Sciences (PNAS), titled: "Characterization of the 1918 'Spanish' influenza virus neuraminidase gene." Its lead author is AFIP molecular biologist Ann Reid. Their project is to map the total sequence of that 1918 pandemic strain of the influenza virus. Besides Hultin's tissue, their starting RNA material includes lung biopsies from a 21-year-old soldier who succumbed to influenza at Fort Jackson, S.C., on Sept. 26, 1918, and a 30-year-old serviceman at Camp Upton, N.Y., who died the same day.

The AFIP team's point of departure was the virus' two most prominent surface glycoproteins - hemagglutinin and neuraminadase. They project as spikes studding the viral envelope. There are approximately 1,000 hemagglutinin projections per virion. Their job is to grapple their viral particle to its target host cell. The AFIP workers sequenced the viral hemagglutinin gene in 1998. Now, as reported in PNAS , they have done the same for the neuraminidase enzyme. Its main role in the virus is to cleave certain hemagglutinin receptors, thus preventing the virions from clumping together.

Second Key Envelope-Spike Protein Sequenced

"We've completed the sequence of neuraminidase to the point where it has been confirmed and is publishable," Reid told BioWorld Today. "And we've completed the first pass to sequencing two matrix and non-structural gene segments, and about half of the nuclear protein gene. With luck we'll be able to get them out and published in another six to eight months.

"After that," Reid went on, "we'll finish the nuclear protein, and what's left of the flu virus' three large polymerase genes. They are the three largest segments, so they are going to take a while. We hope to be able to complete those in another year to a year and a half.

"Our hope," Reid went on, "is that somewhere in those sequences will be the answer to the question as to why that 1918 virus was so lethal. But the more we study this virus the more we suspect that the answer as to its lethality is not going to be a simple, single mutational change in one of its genes. Very likely what made this pathogen so deadly is that it was an extremely well adapted, efficient virus that was able to replicate to extremely high copy number in humans, and spread from human to human very efficiently.

"That kind of ability," Reid pointed out, "is polygenic in nature, which requires that all of the genes really work well together. So we're somewhat less hopeful than we were at the beginning that there would be some single change that we could point to and say, 'That's the reason this was such a lethal virus.'"

Reid observed, "In terms of relating the genetic information from an influenza virus to how it behaves in nature - that is, relating phenotype to genotype - there's really not all that much known. So presumably, with the total 1918 virus sequence in hand, people will be able to test various hypotheses about which gene sequences were involved in viral virulence, and try to narrow down what made it work so well - or so badly, depending on one's point of view."

Reid surmised that the results of their work could have utility in predicting a future pandemic. "I think it can in two ways," she said. "We really have been trying all along to answer two questions: Where do pandemic viruses come from? What makes one systemic virus more lethal than another? We have two pandemics that we've been able to study in detail," she continued, "because their strains were actually isolated. They are the 1957 and 1968 pandemics, neither of which was as serious as 1918. The '57 pandemic was thought to cause 60,000 excess deaths, and the '68 about 25,000."

For Sequencing, Three Pandemics Beat Two

"So having the sequence of the 1918 virus gives us a third pandemic to put into that pot, and try to figure out what all three have in common. How can we predict how they can arise, and how serious they're going to be. So I think our work contributes in that way."

By way of preparing for such a predicted pandemic, Reid suggested, "You could certainly use the hemagglutinin sequence and the neuraminidase sequence to make a vaccine against the 1918 flu. But," she added, "I don't think anyone realistically thinks that the 1918 flu would ever recur in exactly that form. So that wouldn't necessarily have a lot of utility. I guess that for specific preparation the only way the 1918 flu might contribute is if we do find a specific change that contributes to its lethality, you might be able to design drugs or vaccines to counteract that particular mutation. However," she concluded, "we're leaning toward it being more complicated than that."