H5N1 infection in sparrows, starlings and pigeons

It's taken longer than many of us wanted, but some new data on host susceptibility is now coming in. The influenza research group at St. Jude's has just published a paper in CDC's journal, Emerging Infectious Diseases, verifying that common land based birds can be infected with highly pathogenic avian influenza virus H5N1. The St. Jude's group inoculated house sparrows, European starlings and pigeons with four strains of H5N1 that were isolated in 2004 - 2006.

Hi path H5N1 was first found in poultry in southern China in 1996. It is lethal to chickens and other poultry. Ducks are usually unaffected, although some strains are lethal to ducks as well. The combination of poultry movements and movement of infected aquatic wildfowl have spread hi path H5N1 widely to poultry and birds from Asia to Europe to the Middle East to Africa. Most affected birds are poultry species like chickens, turkey, ducks and quail, but on occasion other birds have been found infected, including budgerigars, finches, mapgpies, tree sparrows, pigeons and a crow. The St. Jude study looked at three small landbased birds to get a better handle on the biology of the virus in these common species.

Sparrows, starlings and pigeons were inoculated with one or more of four H5N1 isolates to test their susceptibility to infection and see how much and how long they shed virus if infected. They also co-housed uninfected birds of the same species with infected ones to see if there was transmissibility. Two of the viruses were isolated from a duck and a quail in Thailand in 2005. The other two were obtained from dead wild birds, a Common Magpie and a Japanese White Eye, whose corpses were collected in early 2006 when Hong Kong intensified surveillance.

The sparrows fared poorly. All or most died within a week, with high viral loads in brain and lung. None of the starlings or pigeons died, but all of the starlings were infected and some of the pigeons, depending upon the virus. The virus isolated from the Common Magpie had the broadest host range, infecting all the sparrows, starlings and pigeons, but the duck virus, so lethal to the sparrows, could not be recovered from inoculated pigeons. There was no evidence of any transmission between birds of the same species, the place were transmission would be most likely. One instance of transmission was seen in starlings but couldn't be repeated with additional trials.

The authors observe that these experiments show pigeons, starlings and sparrows more susceptible to these recent H5N1 isolates than they were previously reported to be to an earlier 1997 isolate. They suggest this is consistent with the wider host range of the recent viruses, including mice, ferrets and cats. The evidence that host range has increased is still scant, however. They conclude:

A critical question concerning these small avian species is whether they can serve as intermediate hosts or reservoirs for influenza (H5N1) viruses and transmit them to poultry and mammals. Sparrows were highly susceptible to influenza (H5N1) infection; however, they did not transmit to sentinel contact birds, despite a relatively low infectious dose . . . and the fact that virus was common in drinking water and fecal samples. Although it is possible that the high pathogenicity of these viruses prevented bird-to-bird transmission, the data suggest that this species can act as an intermediate host and potentially transmit to both poultry and mammals but not serve as a reservoir for prolonged shedding of highly pathogenic influenza (H5N1) viruses. In contrast, the characteristics of influenza (H5N1) infection in starlings, i.e., nonfatal with longer-term shedding, suggest that starlings could act as an intermediate host and a reservoir for influenza (H5N1) virus. However, evidence of transmission to contact starlings was limited, which implies that these strains are unsustainable in a starling population. Because pigeons shed only low amounts of virus upon infection and they did not transmit to contact birds, their role in the ecology of influenza (H5N1) virus may be minor.

Our results indicate that there are considerable differences in susceptibility to influenza (H5N1) virus among various small terrestrial wild bird species. The high virulence of several recent isolates in sparrows suggests that this and other populations of small terrestrial birds may have substantial losses during current and future outbreaks. Further mutation of circulating influenza (H5N1) viruses might enhance their adaptation to hosts such as starlings and sparrows, further increasing virulence or allowing these species to become efficient intermediate hosts in the ecology of influenza (H5N1) viruses. (Boon et al., Emerging Infectious Diseases, [Epub ahead of print])

We agree that whether these birds or other animal species can serve (or do serve?) as a reservoir for this virus is a critical question. Given the variability in response seen here and the small number of animals, experiments and isolates, we think it premature to dismiss the possibility of small land birds as a reservoir. All that can be said is that these experiments do not show evidence this is currently or potentially the case. But the authors acknowledge the experiments are extremely limited.

Still, this is the kind of work that urgently needs to be done, not just in birds but in other possible reservoirs and intermediate hosts (usually a transitory or secondary host as a vehicle for an agent moving between one host and another). What about other small mammals, reptiles, even arthropods?

So many questions. So little time.