Surveillance for Highly Pathogenic Avian Influenza in Wild Turkeys (Meleagris gallopavo) of Minnesota, USA during 2015 Outbreaks in Domestic Poultry Open Access

Type A influenza viruses (AIVs) occur regularly in reservoir hosts, including waterfowl, gulls, and shorebirds (Webster et al. 1992), and most can be considered as low pathogenicity subtypes because of benign clinical effects of infection in chickens. There is great diversity in AIVs with 144 known subtypes possible–all combinations of viral glycoproteins hemagglutinin and neuraminidase with numerous strains (Fouchier et al. 2005; Alexander 2007). Typically, only H5 and H7 subtypes have the capability to mutate into highly pathogenic forms, which quickly kill domestic poultry and, in some cases, wild birds (Horimoto et al. 1995; Duan et al. 2007).

The novel strain of Eurasian/North American highly pathogenic avian influenza A (H5N2) virus (HPAIV) detected in North America in 2014 (World Health Organization 2015) was responsible for 50 million domestic poultry deaths in the US in 2015 (US Department of Agriculture 2016). It is hypothesized to be dispersed in the environment by movements of wild birds, although recent evidence suggests that HPAIV is not generally maintained in wild bird populations (Krauss et al. 2016). In Minnesota, US, from 3 March through 4 June 2015, the virus was detected in 23 counties with 104 poultry farms confirmed infected, along with six dangerous contacts (i.e., suspicion of exposure, facility depopulated; US Department of Agriculture 2016). Most of the nine million domestic birds killed or euthanized in Minnesota during the outbreaks were turkeys, and the peak of infection was coincidentally near the opening of the spring Wild Turkey (Meleagris gallopavo) hunting season, which begins the Saturday closest to 15 April. We initiated a pilot surveillance effort in central Minnesota to determine if hunter-harvested Wild Turkeys were infected and shedding HPAIV during the outbreak period, and whether hunter-harvested Wild Turkeys could constitute a viable AIV surveillance sampling stream.

The study area comprised all of central and southern Minnesota, with efforts focused in four counties (Kandiyohi, Meeker, Pope, and Stearns) where the majority of infected poultry facilities occurred during the 2015 outbreak (Fig. 1). Counties were categorized as either containing facilities with or without HPAIV infection, and we had an overall sample goal of 300 male Wild Turkeys. The Minnesota Department of Natural Resources (MNDNR) published a press release, conducted radio, television, and print interviews, and contacted the Minnesota chapter of the National Wild Turkey Federation to solicit samples from successful turkey hunters.

We used polyester-tipped swabs to separately sample the trachea and cloaca of male Wild Turkeys. We pooled both swab samples from each bird in a single vial of Brain Heart Infusion media (Docherty and Slota 1988) and kept samples refrigerated until testing, which generally occurred within 1 wk after collection. Hunter provision of harvested turkeys was voluntary, and we recorded harvest location, date, and age.

Through outreach on the MNDNR and Minnesota Board of Animal Health websites and official press releases, we solicited the public and agency personnel (state and federal employees) to report to the MNDNR and the US Geological Survey National Wildlife Health Center (NWHC) any live Wild Turkeys exhibiting neurological symptoms consistent with AIV infection or dead Wild Turkeys that were not associated with roadways. We emphasized the need to report Wild Turkeys as soon as possible to ensure collection of viable tissue samples; generally, we only collected samples from birds that were dead for <24 h. Birds were sampled by MNDNR staff in the same manner as hunter-harvested birds, with separate swabs used for sampling the tracheal and cloacal cavities. These swabs were placed in the same viral transport media vial, and kept cool in a portable cooler with ice packs or a refrigerator until testing. When viral transport media was not available, whole carcasses were collected, double bagged, and refrigerated until samples could be collected.

As per NWHC instructions, swabs from hunter-harvested Wild Turkeys were shipped overnight in coolers with ice packs to Madison, Wisconsin, US for diagnostic testing. Tissue samples for AIV testing were homogenized in viral transport media and centrifuged at 1,000 × G for 30 min at 4 C. The RNA from 50 μL of the supernatant of the tissue homogenate or swab material was recovered and tested for AIV by the current National Animal Health Laboratories Network reverse transcriptase PCR (rRT-PCR) protocols based on Spackman et al. (2002) and Pedersen et al. (2010).

For morbidity and mortality samples, whole carcasses were shipped overnight to the NWHC for necropsy and AIV screening, as mentioned earlier. Swab samples were submitted to the US Department of Agriculture National Wildlife Research Center (Fort Collins, Colorado, USA) or the University of Minnesota Veterinary Diagnostic Lab (St. Paul, Minnesota, USA) for AIV screening using the National Animal Health Laboratories Network protocol as mentioned earlier.

From 15 April through 28 May 2015, we collected tracheal and cloacal swab samples from 84 hunter-harvested male Wild Turkeys in 11 counties of central Minnesota (Fig. 1). From 7 April 2015 through 11 April 2016, we collected tracheal and cloacal swab samples from 23 Wild Turkeys in 17 Minnesota counties as part of our morbidity and mortality sampling efforts (Fig. 1). All 107 turkeys were negative for AIV using rRT-PCR, and we calculated the 95% detection threshold of AIV using methods of Cameron and Baldock (1998). For our calculations, we used a sensitivity of 0.95 and specificity of 1.0. The data suggested that the apparent shedding prevalence in male Wild Turkey was approximately between zero and 2.9% at a 95% confidence level.

Given the large number of domestic turkeys in Minnesota affected by the H5N2 HPAIV outbreak in 2015, it was important to investigate whether Wild Turkeys were infected and shedding this or other AIV. The outbreak coincided with the spring hunting season, which afforded a unique opportunity to collect samples from hunter-harvested Wild Turkeys in counties with high numbers of infected poultry facilities. Although our pilot effort did not achieve the sampling goal, we did not detect AIV from any samples and could conclude that AIV shedding in male turkeys during the study period was no more than 3%. We cannot make strong conclusions about Wild Turkey susceptibility to AIV infection, but previous serosurveys of Wild Turkeys in the US suggest exposure to AIV is uncommon (Davidson et al. 1988; Hopkins et al. 1990; Charlton 2000; Peterson et al. 2002). Sampling was limited by the number of successful hunters who chose to participate in the surveillance. For future Wild Turkey AIV sampling efforts during harvest seasons, we suggest that agencies make early contact with turkey hunters through press releases and local chapters of organizations such as the National Wild Turkey Federation in order to increase awareness and participation. We also suggest agencies collect data from multiple independent data sources including hunter-harvested (apparently healthy), sick, and found-dead Wild Turkeys to increase the likelihood of detecting AIV in this potential wild gallinaceous host, if it is present. Based on our results, we could not conclude that hunter-harvested Wild Turkeys constitute a valuable surveillance stream for detecting HPAIV in wild birds. However, if agencies can encourage hunter participation, spring and fall turkey harvest seasons lend themselves to a readily available source of this wild gallinaceous bird for AIV testing. Given the ability of AIVs to reassort into novel subtypes (Webster et al. 1992), we recommend consideration of hunter-harvested turkeys for future surveillance. The major challenges to address are how best to encourage hunter participation and provide a mechanism(s) for accessibility to successful hunters in the field (e.g., targeting public access parking areas on public lands).

We thank the wildlife staff from MNDNR who assisted with the sampling of hunter-harvested turkeys as well as the hunters that participated in this project. We also thank the members of the public and MNDNR that reported and collected sick or dead Wild Turkeys. We thank L. Minicucci for useful comments on an earlier draft. Funding was provided by the Minnesota Department of Natural Resources—Division of Fish and Wildlife, US Geological Survey, Ecosystems Mission Area, and US Department of Agriculture—Animal and Plant Health Inspection Service. The use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the US government.