Alaska Science Center
Justification: In response to a threat of highly pathogenic (HP) H5N1 Asian avian influenza (AI), being carried into the United States by migratory birds, the U.S. government has developed an interagency strategic plan to detect the virus in migratory birds in Alaska and the rest of North America. To most effectively monitor AI in Alaska we need to know which avian species are most likely to carry the virus, and what the movement patterns are of suspect species. Information on the population structure of sampled populations will also be essential to predict movement patterns of infected birds and interpret the success of the on-going sampling effort.
Background: The avian influenza virus is commonly found in migratory waterfowl and shorebirds, and all 16 hemagglutinin (HA) and 9 neuraminidase (NA) subtypes have been characterized (Stallknecht and Shane 1988, Hanson et al. 2003, Fouchier et al. 2005) in these groups of birds. One subtype of AI, H5N1, has been found to be highly pathogenic to humans, with outbreaks in Asia, Europe, and Africa. The recent epizootic has yet to reach the western hemisphere, but there is concern that it could be carried here by migratory birds (Chen et al. 2006). It is presumed that the most likely incursion of the virus into North America by migratory birds will be from birds moving from East Asia into Alaska. The likelihood of a given species or population harboring the virus in Alaska is thus directly related to the probability of the bird having recently been in Asia. Although many Asian-wintering populations of birds are known to breed in Alaska, only a few do so exclusively, so movement history and hence probability of encountering Asian HP H5N1 will be unknown for most birds sampled in Alaska. To understand the potential threat of AI once it is detected in North America, we will also need to be able to predict the migratory pathways of birds once they leave Alaska. As most migratory birds are site-faithful across seasons, our best understanding of future movements will be based on knowledge of movement history.
It is possible to determine the migratory history of a bird by examining the ratio of stable isotopes in the birds feathers (Hobson 2005a). Knowledge of the isotopic ratio in the feathers of a bird, when used in conjunction with an understanding of the environmental distribution of different stable isotopes, can allow researchers to determine the location where feather growth takes place (Rubenstein et al. 2002). This is possible because bird feathers are unique as they grow in a very short time and then become metabolically inert. The chemical composition of a feather thus reflects the isotopic ratios of molecules contributing to feather growth at the time of development. Naturally occurring light stable isotopes (D/H, 13C/12C, 15N/14N, and 34S/32S) are known to vary geospatially and once incorporated into an animals tissue, can be used as an indicator of where the animal was when the sampled tissue was formed. The geospatial distribution of underlying environmental isotopes is best known for hydrogen and oxygen, for which the strongest spatial isotope gradients are known to occur at mid- to high-latitude continental regions (Araguas-Araguas et al. 1998, Bowen et al. 2005).
Developing a profile of stable isotopes in different species at various locations in Alaska will not only help us determine wintering areas, but may also reveal other aspects of a species life history, including evidence of population structuring. Many, if not most species of migratory birds exhibit some form of population structuring, in that they are composed of populations or components with distinctly different migration patterns that vary temporally and geographically (Ely and Takekawa 1996). Within-species variation in the use of time and space can be detected with stable isotopes; such differences may eventually lead to genetic variation and possibly speciation (Bearhop et al. 2005).
Another means of detecting population structuring that can also be used as a proxy for movement history and an indication of past associations association is through the construction of a disease history. The potential for an individual to encounter and harbor a virus (or parasite) will be largely dictated by the component of the population and social unit of which it is a member (Biek et al. 2006). Hence an individuals viral signature is a record of past exposure and possibly indicative of the use of common geographic areas. The incidence of viral exposure has been used to determine population structuring in wildlife (Poss et al. 2001, Biek et al. 2006) and humans (Ding et al. 2000). The occurrence of low path AI viruses in different species and populations of birds should provide insight into degree of contact and associations, especially when put in a community context with and compared across habitats and guilds.
1) Determine the origin of birds sampled for AI in Alaska by obtaining feather samples for stable isotope (SI) analysis, and quantify within-site variation as an index of homogeneity.
2) Determine the prior association of AI-sampled birds by identifying and comparing the HA and NA subtypes for each species.
3) Determine the relationship between isotopic and viral profiles within and across taxa and sampling sites and develop a predictive model to be tested with subsequent field sampling.
Species.-We will select 6-10 of the 26 species identified in the national strategic plan. Selection of target species will be based on probability of capture and likelihood of Asian origin. Other factors to be considered include: 1) probability of population structuring, 2) likelihood of obtaining samples at different locations and time to test for spatial and temporal variation. 3) knowledge base (prior research results to compare to current findings). Likely candidate species to include: Waterfowl: Tundra Swans. Emperor Geese, Wrangel Island Snow Geese, Black Brant, and Northern Pintail; Shorebirds: Dunlin and Bar-tailed Godwits; Landbirds: Arctic Warbler, Eastern Yellow Wagtail, and possibly Lesser Sandhill Cranes; Seabirds: Aleutian Terns.
Field sampling.-We will capture Emperor Geese in brood flocks on the Yukon Delta in late July. Tundra Swans will be captured in family groups and small molting aggregations on the Yukon Delta. We will work with staff from Izembek NWR to capture molting groups of migratory and non-migratory Tundra Swans on the Alaska Peninsula. We may assist the North Slope Bureau in capturing Lesser Snow Geese, and the USFWS in obtaining samples from hunter-killed birds on St. Lawrence Island. Samples from other species, at other localities will be obtained from birds captured as part of the overall assessment in Alaska of prevalence of avian influenza in Alaska.
Determination of wintering origin from analysis of stable isotopes.-Feathers will be collected from captured birds and stored in paper envelopes. A sample will be taken from the tail, wing, breast, and head of >50 birds from each locality (Smith et al. 2003). We intend to quantify variation in the proportions of several stable light isotopes, as geographic differences in isotope ratios in birds have been reported for hydrogen (Chamberlain et al. 1997, Hobson and Wassenaar 1997), as well as nitrogen, carbon, and sulfur (Kelly 2000, Wassenaar and Hobson 2000, Hobson 2005a, Hobson 2005b). The initial focus will be to characterize the isotopic ratios of hydrogen and oxygen, as environmental maps are most readily available for these isotopes. We will compare the isotopic ratio in the feathers of sampled birds with the ratios found in birds with known wintering locations. These samples will be obtained from cooperators in eastern Asia and North America, or from museum specimens.
Determination of associations and population structuring using viral markers.-Protocols outlined in the National Strategic plan will be followed for obtaining samples for viral testing. Following initial screening for H5N1, the other AI subtypes will be characterized using molecular testing at the National Wildlife Health Center. A comprehensive virological profile will be obtained from each of the cloacal samples, with all H and N subtypes characterized. Additional viruses will also be identified including Newcastle Disease and other avian paramyxoviruses.