Samples of the original flu were kept from some of the victims.© SPL

Disease experts have come a step closer to understanding the deadly secrets of the Spanish flu virus, which killed around 20 million people in 1918–19. By reconstructing genes from the killer and introducing them into a modern virus, researchers have recreated some of its disease-causing power.

The resulting virus makes mice very ill, even though they don't normally get flu. Learning how exactly they get sick might help health experts to contain future outbreaks of deadly flu in humans.

The researchers, led by Yoshihiro Kawaoka, who works at the universities of Wisconsin and Tokyo, used an approach called reverse genetics to reconstruct the genes. This involves piecing together a desired DNA sequence using small molecules called oligonucleotides.

For this, they needed to know the sequences of genes from the original Spanish flu virus. Several of these were deciphered in 1999 and 2000, using samples from preserved victims. Kawaoka's group was able to reconstruct two genes from the virus, responsible for making the proteins haemagglutinin (HA) and neuraminidase (NA).

The researchers inserted these genes into a modern influenza virus and used it to infect lab mice. The mice developed a flu-like lung disease characterized by severe bleeding. "What we saw was similar to what we saw in 1918 in humans," Kawaoka says.

Catching the culprit

Of the two genes, HA is the culprit, Kawaoka adds. As his team reports in this week's Nature¹, when they repeated the experiment without the NA gene, the mice still got sick. The NA gene alone had no such effect.

This gives a valuable clue why Spanish flu was such a potent killer, Kawaoka says. Haemagglutinin is responsible for binding the virus to host cells and injecting the viral contents. Perhaps Spanish flu's HA was particularly good at this.

The reverse genetics approach is a very useful one, comments Robert Lamb, a virologist at Northwestern University in Evanston, Illinois. But he adds that to unravel completely the mystery of Spanish flu, we should sequence its entire genome, recreate it in its entirety, and study the disease in a non-human primate.

"The case is not yet solved — this is a clue, but I don't think it's the whole story," he says. Perhaps, he adds, the 1918–19 pandemic was so bad because the virus caught people's immune systems by surprise. "It might have been a lack of pre-existing immunity in that population," he suggests. "They'd met flu before, but maybe not this immunological type."

Running riot

The latest study might nevertheless help us to treat new outbreaks of deadly flu, Kawaoka says. His mice showed increased levels of proteins called cytokines and chemokines, suggesting that their immune systems were running riot and this was what killed the human victims. Perhaps the disease could be treated using drugs that quell this effect, he suggests.

People infected with H5N1 — a related virus that causes the deadly "bird flu" — also show high cytokine levels. A spate of infections in Thailand has led health experts to fear an impending global epidemic on a similar scale to Spanish flu, although so far most cases have been caught directly from birds rather than from people.

But despite our increased knowledge, prevention will be much better than treatment, says Kawaoka. "The only way to stop the spread is international collaboration," he says. "We need to control outbreaks in chickens."

If outbreaks are not contained, says Lamb, the results could be disastrous. "Spanish flu killed one in 100 people; H5N1 is killing 90-something per cent of those infected," he says. "We have no vaccine and limited antiviral drugs. The 1918 virus doesn't worry me particularly — we have a vaccine. But H5N1 scares the hell out of me." 

Universities of Wisconsin and Tokyo

Northwestern University

• References

  1. Kobasa D., et al. Nature, 431. 703 - 707 (2004). | Article | PubMed | ISI | ChemPort |