Fifteen juvenile female Gyr-Saker (F. rusticolus × F. cherrug) hybrid falcons were obtained from 1 breeder. The birds received an intensive health evaluation, which included a general examination, radiographs, laparoscopy, blood cell count, blood chemistry analysis, and parasitologic examination; all results were within normal limits. The falcons were perched according to standard falconry techniques during the vaccination trial (22). For challenge infection, the animals were kept individually in stainless steel cages located in negatively pressurized isolation rooms within Biosafety Level 3 facilities. Seven 1-day-old chicks obtained from a disease-free stock, were provided to each bird each day as feed. Unconsumed chicks were removed to measure the daily feed intake of each bird.

Ten falcons were vaccinated (nos. 1–5 intramuscularly and nos. 6–10 subcutaneously) with 0.5 mL (hemagglutinating titer >16) of influenza (H5N2) inactivated vaccine (Intervet, Unterschleissheim, Germany) based on strain A/duck/Potsdam/1402/86; they were revaccinated with the same dose and by the same route 4 weeks later. As a negative control, 5 nonvaccinated falcons were kept with the vaccinated birds.

Before the first vaccination and in weekly intervals until 8 weeks after initial vaccination, individual blood samples were collected from the metatarsalis plantaris superficialis medialis vein directly into a serum tube (Sarstedt, Nümbrecht, Germany) by using a 0.7-mm × 30-mm needle (Sterican, Luer-Lock, Braun Melsungen, Germany). The serum was separated and tested for H5-specific antibodies by the HI test, with low pathogenic avian influenza subtype H5N2 (A/duck/Potsdam/619/85) as antigen according to standardized methods (23). A cloacal swab was also examined for influenza A virus RNA by using real-time reverse transcription–PCR (RT-PCR) targeting an M gene fragment to exclude a concurrent infection (24).

Five months after the initial vaccination, 5 falcons randomly selected from the 10 vaccinated birds (nos. 1, 2, 5, 8, 9) and 5 nonvaccinated control birds (nos. 11–15) were challenged with 10^6.0 50% egg infectious dose (EID₅₀) of influenza strain A/Cygnus cygnus/Germany/R65/2006, a highly pathogenic H5N1 strain that was isolated from a dead whooper swan (Cygnus cygnus) during an outbreak of HPAI virus (H5N1) among wild birds in Germany (25). Each bird received 1 mL cell culture medium by the oculo-oronasal route. The falcons were observed daily for 11 days after challenge. At the end of the trial, surviving birds were humanely killed. A serum sample was obtained by using the above-described method just before challenge and, for surviving birds, on the last day of the trial. The serum was used for the detection of antibodies against H5 by the HI test (see above) by using 2 different antigens (challenge and vaccine strain). Before challenge and at days 1, 2, 4, 7, and 11 after challenge, an oropharyngeal and a cloacal swab were taken for a semiquantitative detection of avian influenza virus–specific RNA by using a real-time RT-PCR targeting an M gene fragment as recommended in the Diagnostic Manual for Avian Influenza issued by the European Commission (26) and described by Spackman et al. (24). The method has been improved by using an internal control in parallel in a duplex reaction (27). In addition, virus isolation in embryonated chicken eggs was attempted as described by Werner et al. (28). Isolated viruses were characterized as HPAI virus (H5N1) by subtype-specific real-time RT-PCRs (26) and by a pathotype-specific real-time RT-PCR (29).

Necropsies were performed immediately after death. Samples of nasal cavity, trachea, lung, heart, cerebellum, cerebrum, spinal cord, proventriculus, small and large intestine, liver, pancreas, spleen, skin, and kidney were collected and either snap frozen or formalin fixed (48 h) and processed for paraffin embedding according to standardized procedures. For histopathologic examination, paraffin wax sections (3 µm) were dewaxed and stained with hematoxylin and eosin. Immunohistochemical examination for influenza virus A nucleoprotein (NP) was performed according to Klopfleisch et al. (30). Briefly, dewaxed sections were incubated with a rabbit anti-NP serum (1:500). As secondary antibody, biotinylated goat anti-rabbit IgG1 (Vector, Burlingame, CA, USA) was applied. By means of the avidin-biotin-peroxidase complex method, a bright red signal was produced. Positive and negative control tissues of chickens that had been infected experimentally with HPAI virus (H5N1) were included. Tissues from the central nervous system (CNS), small intestine, pancreas, trachea, and lung were used for real-time RT-PCR and for virus isolation.

During the entire trial, control birds remained negative for avian influenza virus H5-specific antibodies. In addition, influenza A virus RNA was not detected in any of the cloacal swabs. No adverse clinical effects were detected as a result of application of the 2 vaccine doses.

Nine of the 10 vaccinated birds mounted homologous H5-specific antibodies 3 weeks after the first vaccination; titers increased significantly after the booster vaccination (Table 1). The remaining bird (no. 5) showed a detectable titer of 8 only 6 weeks after initial vaccination (2 weeks after booster vaccination); HI titer for this bird remained at 8. Differences in titer development according to route of vaccination were not detected (Table 1). Clinical signs (i.e., decreased food intake or worsening general condition) were not observed in any of the vaccinated birds. HI titers against the heterologous challenge strain A/Cygnus cygnus/Germany/R65/2006 at the time of challenge are shown in Figure 1. The nonvaccinated birds remained seronegative.

All nonvaccinated birds died after challenge with HPAI virus (H5N1). The first falcon died on day 3 postchallenge, 3 died on day 4, and the rest died at day 5. Of these, 4 had reduced food intake starting from the day of challenge, and 3 had a slightly bloody tracheal exudate detectable the day after exposure. One bird died with no clinical signs.

All vaccinated birds survived. For 2, food intake was slightly reduced 1 day after challenge. No other vaccinated bird exhibited clinical signs. By 11 days after challenge, the titers of the vaccinated birds increased to 2,048 against the antigen used for vaccination and 1,024 against the challenge strain (Figure 1).

Necropsy showed multifocal acute hemorrhagic necrosis in the pancreas of 3 birds that died spontaneously and moderate to severe splenic hyperplasia in 3 birds. Histopathologic examination of the cerebellum, cerebrum, spinal cord, pancreas, spleen, and kidney of the nonvaccinated birds showed multifocal acute cellular degeneration and necrosis associated with minimal to mild infiltration of few heterophils and detection of HPAI virus antigen (Figure 2). Furthermore, antigen was present in the nasal cavity, trachea, bronchial epithelium, and gastrointestinal tract but not in the liver and skin. None of the euthanized vaccinated birds exhibited any gross or histologic lesions or presence of antigen in any of the tissues.

In the nonvaccinated falcons, after challenge infection viral RNA was detectable in all oropharyngeal swabs. Virus was also isolated from the pooled oropharyngeal swabs of these birds taken on the same days (Figure 3; Table 2). At day 1 postchallenge, viral RNA was detected in the cloacal swabs of 3 birds, although virus isolation failed. From day 2 postchallenge, all falcons demonstrated the presence of viral RNA and infectious virus in cloacal swabs (Figure 3; Table 2).

In the vaccinated falcons, 1 day after challenge viral RNA was detectable in all oropharyngeal swabs. At day 2 postchallenge, 1 falcon became negative for viral RNA, and at days 7 and 11, only 1 bird remained positive for viral RNA (Figure 3; Table 2). Virus isolation from a pool of all oropharyngeal swabs of all 5 vaccinated birds taken from day 1 postchallenge demonstrated a virus titer of 4.4 log₁₀ EID₅₀/mL at day 1 postchallenge and 1.2 log₁₀ EID₅₀/mL at day 2 postchallenge. From day 4 on, virus could no longer be isolated from the pooled oropharyngeal swabs. Viral RNA was only occasionally detected in cloacal swabs and completely absent in 1 bird (Figure 3, Table 2). Virus could not be isolated from any of the pools of the cloacal swabs.

In the nonvaccinated falcons, high loads of viral RNA were detected in the CNS, duodenum, pancreas, trachea, and lung of all control birds that died after challenge (Table 3). Virus isolation from these samples was not attempted.

In the vaccinated falcons, low to moderate loads of viral RNA were demonstrated in the brain and trachea of 3 birds that were euthanized on day 11 postchallenge, in the lung of 2 birds, in the duodenum of 1 bird, and in the pancreas of 1 bird. However, virus was isolated from the trachea of only 2 birds and from the lung of 1 (Table 3). The viral RNA load was as much as 6 log₁₀ lower than that of nonvaccinated animals.