The influenza viruses used in this study included A/chicken/Vietnam/14/2005 (H5N1) and A/mallard/British Columbia/373/2005 (H5N2). Vietnam/05 stocks were grown and titrated on Japanese quail fibrosarcoma (QT-35) cells. This isolate bears a PQRERRRKR/GLF HA₀ cleavage site (GenBank accession no. EF535027), has an intravenous pathogenicity index of 2.97, and produced a 100% mortality rate in oronasally inoculated leghorn chickens receiving 10⁵, 10⁴, and 10³ PFU by 3, 4, and 6 days postinfection (dpi), respectively. British Columbia/05 stocks were grown and titrated in 9-day-old chicken embryos. Prior characterization of this isolate demonstrated that it has a PQRETR/GLF HA₀ cleavage site (GenBank accession no. DQ826532) typical for LPAI viruses.

Twenty-two Canada geese were captured with the permission of Environment Canada (Canadian Wildlife Service permit no. CWS06-M009) and were handled and cared for in accordance with Canadian Council on Animal Care guidelines and the animal use protocol approved by the Institutional Animal Care Committee. The geese consisted of 11 adult (6 male + 5 female) and 11 young-of-year (6 male + 5 female) birds. The latter were estimated to be ≈40 days of age at capture. Adult and juvenile birds were randomly assembled into 3 experimental groups, and each group subsequently housed in separate Biosafety Level-3 biocontainment cubicles: 1) a control group comprising 1 juvenile + 1 adult bird, 2) a pre-exposure group comprising 5 juvenile + 5 adult birds, and 3) a naive group comprising 5 juvenile + 5 adult birds.

After a 3-week acclimation period, the pre-exposure group was inoculated with 10⁶ 50% egg infectious dose (EID₅₀) of British Columbia/05 applied to the nares, oral cavity, and cloaca. Twenty-eight days later, pre-exposure and naïve groups were challenged with 1.7 × 10⁵ PFU of Vietnam/05 applied to the nares, oral cavity, and eye. The control group received a sham inoculum of minimal essential medium. Timed necropsies involving 1 juvenile and 1 adult bird from pre-exposure and naïve groups were performed on days 3 and 6 postchallenge (dpc). All remaining birds were either humanely euthanized when moribund or allowed to survive until 20 or 21 days if they showed mild disease or remained clinically normal.

Group A specific nucleoprotein (NP) antibodies were detected with a competitive ELISA as described previously (9). H5-specific antibodies were detected by microtiter plate HI test that used 4 HA U of A/duck/British Columbia/26–6/2005 (H5N2) and chicken erythrocytes.

We incubated 200 EID₅₀ of Vietnam/05 with an equal volume of 2-fold serially diluted test serum (1:4 to 1:512), incubated for 60 min at 37°C, and then used it to inoculate 9-day-old chicken embryos through the allantoic cavity. Egg deaths and HA titers were monitored and virus neutralization titers determined.

Specimens were stored at –70°C before RNA was extracted. Total RNA was extracted from 0.5 mL of 10% (wt/vol) tissue emulsions or clarified swab specimens by using an RNeasy Mini Kit (QIAGEN, Mississauga, Ontario, Canada). A semiquantitative real-time RT-PCR (10) that targets the M1 gene of influenza A virus segment 7 was conducted. Full-length, in vitro transcribed segment 7 RNA, serially diluted in buffer, was run with each assay to give a semiquantitative estimate of the viral load in each tissue.

Formalin-fixed, deparaffinized, and rehydrated 5-μm tissue sections were quenched for 10 min in aqueous 3% H₂O₂, rinsed in MilliQ water, and placed into Tris-buffered saline plus Tween (TBST) buffer for 5 min. Sections were pretreated with proteolytic enzyme (DakoCytomation, Carpinteria, CA, USA) for 15 min, rinsed twice with TBST, and incubated for 1 h with a monoclonal antibody specific for influenza A nucleoprotein (Clone 1331, Biodesign, Sasco, ME, USA) at a dilution of 1:5,000. The sections were washed with TBST, then incubated for 30 min with the Envision + anti-mouse (horse radish peroxidase–labeled) polymer kit (DakoCytomation), followed by a TBST rinse. Diaminobenzidine was used as the substrate chromagen, and slides were counterstained with Gill’s hematoxylin.

Upon arrival, 12 of 12 juvenile geese tested negative and 10 of 12 adult geese tested positive for influenza A virus NP antibodies (Table 1). To determine the HA subtype specificity of the seropositive birds, HI assays were run with 4 HA U of the following antigens: H1N1 (A/Ck/BC/3/98); H2N9 (A/Pintail/AB/293/77); H4N6 (A/Dk/BC/14/99); H5N2 (A/mallard/BC/373/05); H6N1 (A/Tk/ON/844–2/04); and H7N3 (A/Ck/BC/514/04). All tests were negative, indicating that the birds did not appear to have pre-existing H5-specific antibodies. Real-time RT-PCR–negative cloacal swab specimens indicated that the birds were also not actively infected.

After inoculation with 10⁶ EID₅₀ of British Columbia/05, all birds remained clinically normal. The juvenile birds gained weight, but 3 of 5 adult birds had a 6%–10% loss of bodyweight after infection. Cloacal swabs from juvenile birds were real-time RT-PCR positive at 3 dpi; swabs from adult birds were negative (oropharynegeal swabs not tested). At 6 and 10 dpi, cloacal and oropharyngeal swabs from both juvenile and adult birds were real-time RT-PCR negative, indicating that viral shedding was brief. Although most of the British Columbia/05 infected birds developed H5-specific HI antibody titers (Table 1), these sera did not neutralize Vietnam/05 in a chicken embryo–based neutralization assay.

Twenty-eight days after pre-exposure to British Columbia/05, birds in the pre-exposure and naïve groups were challenged with Vietnam/05. Juvenile birds were estimated to be 13 weeks of age at this time. Adult birds in the British Columbia/05 pre-exposure group exhibited mild decreases in feed consumption and mild depression 5–7 dpc. Except for 1 bird with a positive oropharyngeal swab sample at 6 dpc, oropharyngeal and cloacal swab specimens for the adults tested real-time RT-PCR negative at 2, 3, and 6 dpc. Juvenile birds in the British Columbia/05 pre-exposure group exhibited clinical signs similar to those of the adults with the addition of transient nervous signs manifested as repetitive jerking head movements. Viral shedding, as determined by real-time RT-PCR and confirmed by isolation, was detected at 3 dpc in oropharyngeal swab samples in 3 of 5 birds and in a cloacal swab sample in 1 of 5 birds. Complete necropsies showed no gross lesions in juvenile or adult birds at 3, 6, 11, and 21 dpc. The cerebrum, brain stem, and spinal cord of juvenile birds exhibited low levels of viral nucleic acid at 11 and 21 dpc (Appendix Table). Other organs were weakly positive by real-time RT-PCR to varying degrees.

In contrast, juvenile birds in the naïve group showed 100% morbidity after Vietnam/05 challenge; clinical signs included severe depression, inappetence, bright yellow diarrhea, ruffled feathers, hunched posture, repetitive jerking head movements, weakness, staggering gait, distressed vocalization, wing droop, and terminal coma. All birds died or were humanely euthanized by 5 dpc. Viral nucleic acid was detected in the oropharyngeal swab specimens collected at all time points before euthanasia or death; cloacal swab specimens were not as consistently positive. Adult birds also showed 100% morbidity but with clinical signs and viral shedding less pronounced than that observed in juveniles. Necropsies were performed on 2 adults on days 3 and 5; the remaining 3 birds survived until 20 dpc.

Gross pathologic lesions included congestion of the mucosal surface of the trachea, edema and multifocal pinpoint hemorrhages on the serosal surface of the pancreas, splenomegaly, hemorrhage within the ceca, conjunctivitis, congestion of the meninges and cerebral blood vessels, and hemorrhages on the surface of the brain. Virtually all tissues collected from juvenile birds in the naïve group were real-time RT-PCR positive; heaviest viral loads were found in cerebrum, brain stem, and spinal cord. Adult bird 841S/41Y, which required euthanasia at 5 dpi, also had levels of viral nucleic acid in the central nervous system (CNS) comparable to those found in naïve juveniles. This was one of the adult birds with no pre-existing NP antibodies at the beginning of the acclimation period (Table 1). Viral nucleic acid was found in the CNS of a second adult (840S/40Y), euthanized at 20 dpc, but at levels that were 5–7 logs lower than those found in juveniles or the adult bird euthanized at 5 dpc.

Specific influenza A virus immunolabeling was found in all tissues collected from naïve juvenile birds (Table 2). The most consistently affected tissues were the brain, spinal cord, parasympathetic ganglia of the gastrointestinal tract, heart, and pancreas (Figures 1, 2). Within the small intestine and cecum, the strongest and most consistent immunolabeling involved the parasympathetic ganglia of the submucosal and myenteric plexi (Figure 1, panel D) with only the occasional scattered smooth muscle and vascular endothelial cell within the gut mucosa positive for viral antigen. In the 3 birds in which the proventriculus was affected, viral antigen was detected in numerous cell types, including both surface columnar and glandular epithelium, smooth muscle cells of the muscularis mucosa, vascular smooth muscle, and the parasympathetic ganglia (Figure 2, panel C). In the lungs, antigen could be identified in a few capillary endothelial cells. Positive immunolabeling within trachea, liver, kidney, and breast muscle was minimal and observed in only a few birds. Immunohistochemical analysis of tissues collected from naive adult birds detected specific immunolabeling in only 1 bird (841S/41Y) euthanized at 5 dpc; tissues and cells affected were similar to those observed in naive juveniles.