I’m an epidemiologist, not an immunologist or a virologist but I like reading immunology and virology. It’s interesting, in some ways for me it’s more interesting than reading epidemiology. In an epidemiological paper I can see pretty quickly where things are going (or going wrong) and there isn’t much mystery. But the sheer number of moving parts in a cellular system is amazing and confounding. Navigating through the myriad bits and pieces that appear every week in the scientific literature is tough for experts and even tougher for the rest of us who aren’t experts. Vincent Racaniello over at Virology Blog is a great source of information and I read him in an effort not to fall too far behind and help me understand new papers as they come out. One appeared the other day in the Proceedings of the National Academy of Sciences (PNAS; hat tip reader hjmler) that was billed in the LA Times as showing that a bit of duck DNA protects that animal from the lethal effects of the flu virus. Well, maybe. There was the ever-present qualifier “might” that seems to appear in many “news” articles based on press releases. To me it’s a signal that a reporter or university media flack is trying to get a scientist to say more than they would to other scientists, and that’s the case here — “Bit of duck DNA might protect poultry from flu, scientists say”:

In a study published online March 22 in Proceedings of the National Academy of Sciences, researchers said they’ve found that a key influenza-fighting gene in wild ducks is absent in chickens. Genetically modifying chickens with a copy of that gene might render them resistant to influenza A, the most common form of flu infecting humans.

“If we could shut down influenza (in chickens), it would be of great commercial interest,” said lead author Katharine E. Magor, a comparative immunologist at the University of Alberta in Edmonton. (Amina Khan, LA Times via PhysOrg.com)

Sounds interesting, so I went to the paper and it is interesting. But I’m not sure what its public health significance is, nor am I sure I want chickens to stop being sentinels for lethal infections with H5N1. First, the science.

Wild birds are the natural reservoir of flu viruses and most of the many varieties of flu don’t cause problems for the birds. They live in symbiotic partnership. But domesticated land birds like chickens don’t fare so well. The same viruses that are harmless to the ducks are lethal to chickens. As Dr. Magor says, preventing that from happening would be of “great commercial interest.” So why don’t ducks get sick and chickens do?

The battle between host and virus is a complicated dance involving a move and counter move. Sometimes one of the partners steps on the feet of the other and the dance is over with one incapacitated and the other seeking a new partner. Sometimes they’ve worked things out so that the moves can be coordinated and both can keep dancing. Ducks and flu viruses are in the latter category. Chickens and flu viruses are in the former, with the chickens the incapacitated partner (incapacitated as in “dead”). What move is the chicken not making? This paper identifies at least one such response. Is it the one that stops the dance or are there more things involved? We don’t know yet. We do know that the dance is very, very complicated with lots of steps and lots of possible answering steps. Whether the difference between ducks and chickens shown in this paper is the big difference will require more work but the evidence is interesting.

What is the evidence? When a virus infects a host cell it sets off a whole raft of countermeasures. Among the first are a series of events that go under heading of “innate immunity” (for more on this see our post here and Vincent Racaniello’s post here and many more on his site and on ours). The innate immune system is itself very complicated and has lots of different parts to it. One of them is a sensing system inside the host that starts defensive measures when it “sees” viral genetic material being replicated. There are at least two such sensors in the duck cell (and in mammalian cells) called RIG-I and MDA5. You can read more about them and see some cartoon representations of the pathways at the Virology Blog here. When the buttons of either are pushed they start a series of events which converge on a final common pathway through another protein, IPS-1. IPS-1 then causes the cell to make another set of proteins (interferons) that send a message to neighboring cells to put up antiviral defenses. Here’s an analogy. IPS-1 is like the fire station dispatcher. It gets calls via 911 or from someone running into the fire station saying they see smoke or responding to a fire alarm somewhere connected to the station. Even though all these signals converge on the dispatcher, depending on where the signal comes from or how it might (or might not) result in a different kind of response.

Thus while MDA5 and RIG-I both sense replicating viral genetic material, they don’t necessarily produce the same responses. And it turns out that the duck (and the pigeon and mammals) have both of these sensors but chickens have only one: MDA5. This paper suggests chickens lost RIG-I early in their evolutionary separation, even before domestication. The loss of that pathway seems to make them much more vulnerable to the flu virus than the duck. It’s a dance step they don’t have in their repertoire and it seems it results in their feet getting stomped on. Is that alone enough to get them thrown of Dancing with the Stars? It “might” be.

Unfortunately there are a lot of untidy loose ends in this story. Some bird flu viruses can kill some ducks, but ducks also respond very vigorously with production of antiviral interferons when infected with these Highly Pathogenic viruses, even more so than with the low path variety that doesn’t make the ducks sick (although in the experiments in this paper those high path viruses didn’t kill the ducks, although they will do so sometimes in the wild). Chickens also have an intact MDA5 pathway, although it seems to produce one kind of interferon (IFNalpha) but not another (IFNbeta). Mammals also have both pathways (like ducks) but are killed by the virus, but the ducks don’t seem to react to pushing the RIG-I button the way mammals do (they don’t go through another protein, TRIM25). So it’s all very complicated and very interesting. It might wind up telling us a lot about the innate immune system, the first line of defense against infection. But it is just a sentence fragment of what is likely a pretty long story.

What if it were the case, though, that we could make the chickens as immune to bird flu as ducks, say by engineering a transgenic chicken with RIG-I? It would indeed be of great commercial value, because bird flu kills poultry by the tens or hundreds of millions. But many of those chicken deaths are the result of culling. Once it is evident there is bird flu in the flock, they become a danger to other flocks and, with H5N1 bird flu, a danger to humans. In much of Asia, southeast asia and Africa, poultry deaths are an early warning signal telling local and national authorities to be vigilant for unexplained pneumonia deaths that might be from bird flu.

So, yes, this is literally true: “Bit of duck DNA might protect poultry from flu, scientists say.” It might. Or it might not. And in either case, then what?

Citation: Megan R. W. Barber, Jerry R. Aldridge, Jr., Robert G. Webster, and Katharine E. Magor , Association of RIG-I with innate immunity of ducks to influenza, PNAS 2010 : 1001755107v1-201001755