Flu research.

How to make bird flu fly, part one.

The first of two controversial research papers is published.

ON APRIL 27th, after much toing and froing, the Dutch government gave Ron Fouchier of the Erasmus Medical Centre in Rotterdam permission to submit his paper on bird flu to Science. Dr Fouchier is the head of one of two groups studying how bird flu might become transmissible between people. In December the authorities in America and the Netherlands prevented both his group and the other, led by Yoshihiro Kawaoka of the University of Wisconsin-Madison, from publishing their findings, lest they get into the wrong hands. This official fear stemmed from the deadly nature of bird flu. Of the 602 human cases reported since 2003, 355 have been fatal. The factor that has stopped the death toll being worse is that people have to catch the virus directly from a bird (usually a chicken). It rarely, if ever, passes from one person to another.

Science has yet to publish Dr Fouchier's manuscript, but its rival Nature has gone ahead and published Dr Kawaoka's. This paper got clearance from the American authorities on April 20th. So it is now possible to see what all the fuss was about.

Dr Kawaoka was interested in the role a protein called haemagglutinin (HA) plays in the transmission of avian influenza. HA is the viral equivalent of a grappling hook: it lets the virus latch onto a cell by binding to substances called sialic acids that are found in receptors on the cell's surface. This done, the virus infects the cell with its DNA. But because the sialic acid found in birds is chemically different from that in mammals, and because bird flu has evolved to recognise only the avian variety, it cannot stick easily to mammalian cells, limiting its ability to infect people.

The researchers wanted to know what it would take to enhance that ability. They took the HA gene from avian influenza viruses found in Vietnam and made millions of mutant versions, each of which was spliced back into a copy of the original virus. They then screened 2.1m of these mutant viruses to see which bind to sialic acid of the mammalian variety. Just eight can do so, and just one of those has become, in effect, a mammalian specialist, because it can no longer bind to avian cells.

Dr Kawaoka decided to concentrate his efforts on this specialist. Further tests revealed that two mutations in its HA gene, called N224K and Q226L, were responsible for its characteristics.

It is here that things get complicated. Instead of looking at the doubly mutated gene in bird flu, the team transplanted it into the virus that caused the human-flu pandemic of 2009. By doing so, they virtually guaranteed they would make a virus that could pass between mammals.

And so it proved. When they tested it on ferrets (often used as proxies for people in these sorts of studies), it passed through the air from one animal to another. Moreover, closer inspection showed that two other mutations, N158D and T318I, had spontaneously got tangled up in the mix.

What all this means, then, is slightly obscure. Dr Kawaoka's purpose was to study how the mammalian-transmission mutations worked. In this, he succeeded. He found that N224K, Q226L and N158D all control areas in the head of the HA protein-ie, the hook-and probably help it attach to the mammalian flavour of sialic acid. T318I controls part of HA's stem and may help to stabilise the protein as it injects viral DNA into cells.

By transferring the genes into a virus that was known to be good at moving between mammals, Dr Kawaoka has shown that the HA gene is not itself an obstacle to bird flu gaining that ability, though its other genes may be. Perhaps Dr Fouchier's paper will shed more light on the matter.