Four influenza A (H5N1) virus strains were studied, 2 from previously known susceptible hosts (duck and quail) and 2 from previously unknown hosts (common magpie and Japanese white-eye). The A/duck/Thailand/144/2005 (A/DK/TH/144/05) and A/quail/Thailand/551/2005 (A/Q/TH/551/05) viruses were isolated from western Thailand and tested for their pathogenicity in ducks (15). The other 2 viruses, A/common magpie/Hong Kong/645/2006 (A/CM/HK/645/06) and A/Japanese white-eye/Hong Kong/1038/2006 (A/JW/HK/1038/06) were provided to us by K.C. Dyrting and C.W.W. Wong (Agriculture, Fisheries and Conservation Department in Hong Kong). They were isolated from dead wild birds collected in January and February 2006 during the heightened Hong Kong territory-wide avian influenza surveillance of dead wild birds that started in October 2005.Upon arrival at St Jude Children’s Research Hospital, the viruses were propagated in 10-day-old embryonated chicken eggs.

Wild house sparrows (Passer domesticus) and European starlings (Sturnus vulgaris), both members of the order Passeriformes, were captured. Six-week-old white Carneux pigeons (Colomba spp.), members of the order Columbiformes, were purchased from Palmetto Pigeon Plant (Sumter, SC, USA) and Double T farms (Glenwood, IA, USA). Birds were housed in cages in the St Jude Children’s Research Hospital Animal Biosafety Level 3+ containment facility, food and water were provided ad libitum, and general care was provided as required by the Institutional Animal Care and Use Committee. Before inoculation with virus, oropharyngeal and cloacal swabs were collected to exclude preexisting influenza A virus infection.

Three sparrows and pigeons were inoculated intranasally with 1 million 50% egg infectious doses (EID₅₀) in 50 μL or 500 μL phosphate-buffered saline, respectively, for each virus. Because of their limited availability, starlings were inoculated with 3 viruses (1 million EID₅₀ in 150 μL), and group sizes were reduced (1 bird for A/DK/TH/144/05, 3 birds for A/CM/HK/645/06, and 2 birds for A/JW/HK/1038/06). One day after inoculation, uninfected contact birds, at a ratio of 1:1 for sparrows and starlings or 2:3 for pigeons, were housed together with inoculated animals to study intraspecies transmission. Birds were monitored daily for death and illness for a 14-day period. After inoculation, oropharyngeal and cloacal swabs were collected on days 2, 4, 6, 8, and 11 for sparrows and starlings and days 3, 5, and 7 for pigeons. Influenza virus was detected by using 10-day-old embryonated chicken eggs as previously described (7). EID₅₀ virus titers were determined in positive swabs by using the method of Reed and Muench (15). The lower limit of quantitation of the assay is 10^0.75 EID₅₀/mL, and average virus titers in organs and swabs were calculated by using the log₁₀ value of each sample.

Fourteen days after inoculation with virus, serum specimens were collected from inoculated and contact birds, and hemagglutination-inhibition (HI) titers were determined according to standard methods (16,17) by using chicken erythrocytes and 4 hemagglutinating units of virus. An HI titer >10 suggested a recent influenza virus infection; an HI titer <10 was considered negative.

The ability of 4 different influenza A (H5N1) viruses to infect and cause disease in house sparrows, European starlings, and white Carneux pigeons was determined. Infection of sparrows caused death in 66%–100% of the infected animals, depending on the inoculated virus (Table 1). The average time to death varied from 4.2 days for A/DK/TH/144/05 to 6.3 days for A/Q/TH/551/05 virus (data not shown). High viral loads were detected in brain and lung tissues of deceased sparrows (Figure, panel C). In contrast, none of the starlings or pigeons died after inoculation with these viruses.

Re-isolation of virus from oropharyngeal and cloacal swabs obtained at various time points after inoculation indicated that all the sparrows and starlings were infected by all viruses tested. In contrast, the frequency of virus re-isolation from inoculated pigeons varied widely among viruses. Of the 4 different H5N1 subytpes, A/CM/HK/645/06 demonstrated the broadest host range, infecting not only sparrows and starlings but also all of the inoculated pigeons. The A/DK/TH/144/05 virus, which caused 100% mortality in sparrows within 4.2 days after inoculation, was not re-isolated from inoculated pigeons.

Quantification of the virus titer in the swabs demonstrated that sparrows and starlings shed similar amounts of virus in oropharyngeal swabs. However, virus titers in the cloacal swabs of sparrows were higher than in those obtained from infected starlings (Table 1). Comparison of peak virus titers in oropharyngeal swabs confirmed the similarity in oral shedding between sparrows and starlings. In contrast, peak virus titers in the cloacal swabs of starlings were lower (Figure, panels A and B). The 2005–2006 influenza (H5N1) viruses replicated relatively poorly in pigeons, as shown by average oropharyngeal and cloacal shedding on days 3 and 5 (Table 1) and by peak virus titers in oropharyngeal and cloacal swabs (Figure, panels A and B).

The capacity of current influenza (H5N1) viruses to transmit from infected birds to same-species uninfected birds was assessed for these 4 viruses. No evidence of transmission in sparrows and pigeons was found, as attempts to isolate the virus from contact birds failed (Table 2). Also, no virus-specific antibodies were detected by HI in the contact birds (data not shown). In starlings, transmission of virus to contact birds was observed once for A/CM/HK/645/06 virus, but this was not seen in 2 further experiments.