SUMMARY.
Gulls are widely recognized reservoirs for low pathogenic avian influenza (LPAI) viruses; however, the subtypes maintained in these populations and/or the transmission mechanisms involved are poorly understood. Although, a wide diversity of influenza viruses have been isolated from gulls, two hemagglutinin subtypes (H13 and H16) are rarely detected in other avian groups, and existing surveillance data suggests they are maintained almost exclusively within gull populations. In order to evaluate the host range of these gull-adapted influenza subtypes and to characterize viral infection in the gull host, we conducted a series of challenge experiments, with multiple North American strains of H13 LPAI virus in ring-billed gulls (Larus delawarensis), mallards (Anas platyrhynchos), chickens (Gallus domesticus), and turkeys (Meleagris gallopavo). The susceptibility to H13 LPAI viruses varied between species and viral strain. Gulls were highly susceptible to H13 LPAI virus infection and excreted virus via the oropharynx and cloaca for several days. The quantity and duration of shedding was similar between the two routes. Turkeys and ducks were resistant to infection with most strains of H13 LPAI virus, but low numbers of inoculated birds were infected after challenge with specific viral strains. Chickens were refractory to infection with all strains of H13 LPAI virus they were challenged with. The experimental results presented herein are consistent with existing surveillance data on H13 LPAI viruses in birds, and indicate that influenza viruses of the H13 subtype are strongly host-adapted to gulls, but rare spill-over into aberrant hosts (i.e., turkeys and ducks) can occur.
Keywords: avian influenza, chickens, gulls, H13, mallards, poultry, turkeys
RESUMEN.
Susceptibilidad en diferentes especies de aves a los virus de la influenza aviar de baja patogenicidad subtipo H13 aislados en América del Norte.
Las gaviotas son reservorios ampliamente reconocidos del virus de la influenza aviar de baja patogenicidad, sin embargo, los subtipos mantenidos en estas poblaciones y/o los mecanismos de transmisión involucrados no están bien esclarecidos. Aunque, una amplia diversidad de virus de influenza se han aislado de gaviotas, dos subtipos de hemaglutininas (H13 y H16) rara vez se detectan en otros grupos de aves, y los datos existentes de vigilancia sugieren que se mantienen casi exclusivamente dentro de las poblaciones de gaviotas. Con el fin de evaluar el rango de hospederos de estos subtipos del virus de influenza adaptados en gaviotas y para caracterizar la infección viral en las gaviotas como huéspedes, se llevó a cabo una serie de experimentos con varias cepas del virus de la influenza aviar de baja patogenicidad subtipo H13 aisladas en América del Norte en el desafío de gaviotas Delaware (Larus delawarensis), en patos de collar (Anas platyrhynchos), en pollos (Gallus domesticus) y en pavos (Meleagris gallopavo). La susceptibilidad a los virus de influenza aviar de baja patogenicidad H13 variaron entre las especies y las cepas virales. Las gaviotas son muy susceptibles a la infección por el virus de la influenza aviar de baja patogenicidad H13 y el virus se excreta a través de la orofaringe y la cloaca durante varios días. La cantidad y la duración de la eliminación fue similar entre las dos rutas. Los pavos y los patos fueron resistentes a la infección con la mayoría de las cepas de virus de la influenza aviar de baja patogenicidad H13, pero un bajo número de aves inoculadas fueron infectadas después del desafío con las cepas virales específicas. Los pollos fueron refractarios a la infección con las cepas de virus de la influenza de baja patogenicidad H13 con las que fueron inoculados. Los resultados experimentales presentados en este informe son consistentes con los datos de vigilancia existentes con los virus de la influenza aviar de baja patogenicidad H13 en las aves, e indican que los virus de influenza del subtipo H13 están fuertemente adaptados a las gaviotas como hospederos, aunque la diseminación a huéspedes aberrantes (pavos y patos) es rara pero puede ocurrir.
The first isolation of avian influenza (AI) virus from a wild bird was during an outbreak of H5N3 highly pathogenic (HP) AI in common terns (Sterna hirundo) in South Africa during 1961 (1). In the 50 yr since this event, AI viruses have been commonly isolated from wild birds in the family Laridae (gulls and terns), including HP and low pathogenic (LP) viruses representing numerous hemagglutinin (HA) and neuraminidase (NA) subtypes (9,14,19,23). Two HA subtypes (H13 and H16) have rarely been detected in mammalian or avian hosts outside of the family Laridae, and are thought to be maintained almost exclusively in gull and tern populations (6). Collectively, surveillance data indicate gulls are permissive hosts for a wide diversity of AI viruses and recognized reservoirs for the H13 and H16 subtypes; however, it is currently unclear how these viruses are transmitted or maintained in these avian populations.
The H13 subtype was first identified and characterized in the early 1980s (12). Although, it is the most common HA subtype detected in gull species around the globe, the epidemiologic patterns of H13 LPAI virus are not well defined. The limited existing experimental and surveillance data indicate the host range of H13 LPAI viruses is unique, but confusing. Other than gulls and terns, natural infections with H13 LPAI virus have sporadically been reported in only a few other species, including domestic turkeys (Meleagris gallopavo), shorebirds (family Scolopacidae), and pilot whales (Globicephala sp.) (12,13,16,17,24). Experimentally, the infectivity of H13 LPAI viruses for different hosts appears highly variable between species and viral strains. Turkeys and ferrets (Mustela putorius furo) were permissive hosts that excreted detectable levels of virus and seroconverted after challenge with H13 LPAI viruses (12,16). Studies in mallards (Anas platyrhynchos) and chickens (Gallus domesticus) have yielded variable results, ranging from complete resistance to sporadic infections (12,14,16). Interestingly, although gulls are the primary reservoir for H13 LPAI viruses, to date, infection has not been characterized in this host experimentally. This lack of data precludes understanding the mechanisms of viral transmission between gulls and from gulls to aberrant hosts (including domestic animals and other wild avian taxa), and limits any attempts to improve surveillance efficiency for H13 LPAI viruses.
The overall goal of the experimental trials presented herein was to complement existing surveillance data and provide insights into the epidemiology of H13 LPAI virus. The principal objectives were 1) to evaluate the susceptibility of ring-billed gulls (Larus delawarensis) to infection with a North American H13 LPAI virus and characterize patterns of viral shedding; and 2) to determine the potential risks for spill-over of North American strains of H13 LPAI virus into important poultry or wild bird species known to be highly susceptible to AI viruses.
MATERIALS AND METHODS
Animals.
Four avian species were used in this study: 4-wk-old mallards, 6-wk-old ring-billed gulls, 4-wk-old specific pathogen free (SPF) white-leghorn chickens, and 3-wk-old SPF small Beltsville white turkeys. The mallards, chickens, and turkeys were captive-bred (McMurray Hatchery, Webster City, IA [mallards]) or acquired from in-house flocks at the Southeast Poultry Research Laboratory [SEPRL], Athens, GA [chickens and turkeys]) and cared for under indoor confinement. At 3-wk (turkeys) or 4-wk (mallards and chickens), birds were transferred to a biosafety level (BSL)-2Ag+ facilities at the University of Georgia (UGA; mallards) or the SEPRL (chickens and turkeys), where the experimental challenge trials were conducted.
Ring-billed gulls used in this study were acquired through the Southeastern Cooperative Wildlife Disease Study (SCWDS), UGA, under appropriate federal and state permits. Young nestling ring-billed gulls were hand-caught in St. Lawrence County, New York, and maintained at the UGA under semi-indoor confinement until 6 wk of age, at which time they were transferred to a BSL-2Ag+ facility at the UGA for the experimental challenge trial. Using the methodology described below, all gulls tested negative for antibodies to the nucleoprotein (NP) of AI virus at 48 hr after capture.
The procedures and methods used during the bird capture, husbandry, and experimental trials were performed according to animal care and use proposals approved by the Institutional Animal Care and Use Committee at the UGA and/or SEPRL.
Viruses.
The following eight strains of H13 LPAI virus were used in the experimental challenge trials: A/laughing gull/NJ/AI08–1460/2008 (H13N9), A/ring-billed gull/DE/AI10–1708/2010 (H13N9), A/sand (environment)/NJ/1–24/2010 (H13N6), A/turkey/MN/1012/1991 (H13N2), A/ruddy turnstone/NJ/AI09–294/2009 (H13N6), A/laughing gull/NJ/AI08–1388/2008 (H13N9), A/gull/ND/44036/1992 (H13N6), A/laughing gull/NJ/AI08–1460/2008 (H13N9), and A/gull/MD/704/1977 (H13N6). Working stocks for each strain were propagated and titrated in 9-to-11-day-old SPF embryonating chicken eggs (25). The viral inocula used in each trial were prepared by diluting stocks in sterile brain-heart-infusion (BHI) media to the desired titer.
Virus detection.
All oropharyngeal and cloacal swabs collected during the trials were stored at −70 C until virus isolations (all trials), titrations (gulls), and/or quantitative real-time reverse transcription PCR (qRRT-PCR [chickens and turkeys]) testing were conducted. Virus isolations and titrations were performed in 9-to-11-day-old SPF embryonating chicken eggs (25). For viral titrations, infectious titers were calculated using the methodology described by Reed and Muench (18) and reported in median embryo infectious doses (EID50) per milliliter.
For qRRT-PCR testing on samples collected during the turkey and chicken trials, RNA was extracted using a previously described combination of Trizol LS reagent (Invitrogen Inc., Carlsbad, CA) and the MagMax AI/ND RNA isolation kit (Ambion, Inc., Austin, TX) (4). The extracted chicken and turkey samples were tested with qRRT-PCR targeting influenza matrix gene (21) using the Applied Biosystems 7500 Fast Real Time PCR instrument (Foster City, CA), and the Applied Biosystems/Ambion AgPath-ID One-Step RT-PCR kit. Samples with a cycle threshold (Ct) less than 38 were considered positive.
Antibody detection.
Serum samples were stored at −20 C until testing was performed. All samples collected during the trials were tested for antibodies against the influenza NP using a commercial blocking enzyme-linked immunosorbent assay (bELISA; FlockChek AI MultiS-Screen antibody test kit; IDEXX Laboratories, Westbrook, ME).
Histopathology and immunohistochemistry.
Tissues collected during the ring-billed gull trial for histopathologic examination were fixed in 10% neutral buffered formalin, routinely processed, and embedded in paraffin. Sections were cut at 5 μm and stained with hematoxylin and eosin. Duplicate sections were cut and immunohistochemically stained using a commercial mouse monoclonal antibody to the NP of influenza A virus at a 1:1000 dilution (Biodesign International, Sako, ME). The methodology used to perform the immunohistochemical staining followed those previously described (5).
Experimental design.
Gulls.
A single viral strain was used in the gull challenge trial. Viral stock for A/laughing gull/NJ/AI08–1460/2008 (H13N9) was diluted in sterile BHI media to a titer of 104 EID50/0.2 ml (single-bird inoculum). This lower viral dose (relative to most wild or domestic bird AI virus challenge studies) was chosen to more closely mimic natural levels of exposure.
Eight ring-billed gulls were inoculated with 0.2 ml of the H13N9 LPAI virus inoculum, split evenly between the intranasal and intratracheal routes. Two negative control ring-billed gulls were inoculated with 0.2 ml of sterile BHI media by the same combined routes. After inoculation, all gulls were monitored twice daily for any behavioral changes or overt signs of disease. Oropharyngeal and cloacal swabs were collected from all gulls on post-inoculation day (PID) 0 to 7, 10, 12, and 17, and placed in separate cryogenic vials (Corning Inc., Corning, NY) containing 2.0 ml of BHI media supplemented with antimicrobial drugs (250 μg/ml gentamicin, 500 μg/ml kanamycin, 1000 μg/ml streptomycin, 1000 U/ml penicillin G, and 25 μg/ml amphotericin B). All swab samples were tested with virus isolation and titration. Two LPAI virus–inoculated gulls were humanely euthanatized and necropsied on PID 1, 3, 7, and 17. The two negative control gulls were euthanatized and necropsied on PID 17. At necropsy, samples were collected from all sections of the respiratory and gastrointestinal tracts, liver, and kidneys, and placed in 10% neutral buffered formalin for routine histopathologic and immunohistochemical examination. A sample of blood was collected from the right jugular vein of each gull prior to inoculation and after they were euthanatized. Blood samples were placed into serum separator tubes (Becton, Dickinson and Company, Franklin Lakes, NJ), centrifuged at 1509 × g for 15 min, and the serum was harvested.
Mallards.
Three viral strains were used in the mallard challenge trials. Viral stocks for A/laughing gull/NJ/AI08–1460/2008 (H13N9), A/ring-billed gull/MN/AI10–1708/2010 (H13N6), and A/sand (environment)/NJ/1–24/2010 (H13N6) were diluted in sterile BHI media to a titer of 106 EID50/0.2 ml (single-bird inoculum). Existing field and experimental data suggest that mallards are relatively resistant to H13 LPAI viruses. Consequently, the higher dose used for the mallard trials was chosen to evaluate the potential for this species to become infected with H13 LPAI viruses, even after exposure to high infectious titers.
Twenty mallards were evenly divided into four groups (5 birds/group), and each group was challenged with one of the three LPAI viruses listed above or a sham inoculum. Individual ducks were inoculated with 0.2 ml of H13 LPAI virus or sterile BHI media split evenly between intranasal and intratracheal routes. At PID 1, two additional mallards were placed into each of the isolators containing LPAI virus–inoculated birds. All ducks were monitored twice daily for any behavioral changes or overt signs of disease. Oropharyngeal and cloacal swabs were collected from all ducks on PID 0, 2, and 4 (or post-contact day [PCD] 0, 2, and 4) and placed in separate cryogenic vials containing 2.0 ml of BHI media and antimicrobial drugs, as already described. Swab samples were tested by virus isolation in embryonating chicken eggs. Blood was collected from all ducks prior to inoculation and immediately after euthanasia for serologic testing using the bELISA. The trial was terminated and all ducks humanely euthanatized on PID 14 (PCD 13).
Chickens and turkeys.
Eight viral strains were used in the chicken and turkey challenge trials. Viral stocks for A/laughing gull/NJ/AI08–1460/2008 (H13N9), A/ring-billed gull/MN/AI10–1708/2010 (H13N6), A/sand (environment)/NJ/1–24/2010 (H13N6), A/turkey/MN/1012/1991 (H13N2), A/ruddy turnstone/NJ/AI09–294/2009 (H13N6), A/laughing gull/NJ/AI08–1388/2008 (H13N9), A/gull/ND/44036/1992 (H13N6), and A/gull/MD/704/1977 (H13N6) were diluted in sterile BHI media to a titer of 107 EID50/0.2 ml (single-bird inoculum). As with mallards, the higher dose used for these trials was selected to provide a sensitive measure of risk for H13 LPAI virus infection in these poultry species.
For both chickens and turkeys, 72 birds were separated into nine groups (8 birds/group), and each group was challenged with one of the viruses listed above or a sham-inoculum consisting of sterile BHI media. At PID 3, two naïve contact birds were placed into each isolator containing LPAI virus–inoculated birds. The methods for inoculation, monitoring, and swab sample collection used in the chicken and turkey trials were identical to those described for the mallard trials.
All swab samples collected during the chicken and turkey trials were initially screened with qRRT-PCR, and all positive samples (Ct value < 38) were further tested with virus isolation in embryonating chicken eggs. Blood was collected from all birds prior to inoculation and after euthanasia for serologic testing using a bELISA. The trial was terminated and all birds humanely euthanized on PID 14 (PCD 11).
Data analysis.
In each of these trials, individual birds were considered infected with a H13 LPAI virus based on any one of the following criteria: 1) evidence of seroconversion (i.e. pre-inoculation serum samples tested negative and post-inoculation serum samples tested positive by bELISA); 2) detection of AI viral antigen in immunohistochemically-stained tissues (gull trial only); or 3) virus isolated from cloacal swabs collected on any day or from oropharyngeal swabs collected on or after PID 2. Virus isolated from an oropharyngeal swab at PID 1 was not included as a stand-alone criterion due to the potential for detecting residual inoculum. Samples that were positive by virus isolation but did not have a quantifiable titer were assigned a titer of 101 EID50/ml.
RESULTS
No morbidity, mortality, or gross lesions were observed in any of the LPAI virus- or sham-inoculated birds during the four species trials. No virus was isolated from oropharyngeal or cloacal swabs collected from sham-inoculated gulls, mallards, chickens, and turkeys, and all pre- and post-inoculation serum samples from these birds were negative for antibodies to AI virus based on the bELISA.
Based on the infection criteria described above, the susceptibility to H13 LPAI viruses varied among the four species included in this study (Table 1). Ring-billed gulls were highly susceptible to infection, with seven of the eight inoculated birds fulfilling one or more infection criteria. The one uninfected gull was euthanatized on PID 3. This gull was virus isolation positive on the PID 1 oropharyngeal swab but was excreting low concentrations of virus (below the detectable limit of titration), was virus isolation negative on all other collected oropharyngeal and cloacal swabs, did not seroconvert, and had no detectable viral antigen in immunohistochemically stained tissues. Consequently, the PID 1 positive oropharyngeal swab sample presumably reflects the detection of residual inoculum.
Table 1.
Summary of productive infection results obtained from inoculation of mallards, turkeys, and chickens with various H13 LPAI virus strains. No contact birds in any of the challenge groups listed below had detectable evidence of infection based on viral shedding or serologic testing.
| Species | Virus | Viral shedding |
SerologyA |
|
|---|---|---|---|---|
| PID 2 | PID 4 | PID 14 | ||
|
| ||||
| Mallard | A/laughing gull/NJ/AI08–1460/2008 (H13N9) | 0/5 | 0/5 | 0/5 |
| Mallard | A/ring-billed gull/MN/AI10–1708/2010 (H13N6) | 3/5 | 0/5 | 3/5 |
| Mallard | A/sand (environment)/NJ/1–24/2010 (H13N6) | 0/5 | 0/5 | 0/5 |
| Turkey | A/ laughing gull /NJ/AI08–1460/2008 (H13N9) | 0/8 | 0/8 | 0/8 |
| Turkey | A/ring-billed gull/MN/AI10–1708/2010 (H13N6) | 0/8 | 0/8 | 0/8 |
| Turkey | A/sand (environment)/NJ/1–24/2010 (H13N6) | 0/8 | 0/8 | 0/8 |
| Turkey | A/turkey/MN/1012/1991 (H13N2) | 1/8 | 0/8 | 1/8 |
| Turkey | A/ruddy turnstone/NJ/AI09–294/2009 (H13N6) | 0/8 | 0/8 | 0/8 |
| Turkey | A/laughing gull/NJ/AI08–1388/2008 (H13N9) | 0/8 | 0/8 | 0/8 |
| Turkey | A/gull/ND/44036/1992 (H13N6) | 1/8 | 0/8 | 0/8 |
| Turkey | A/gull/MD/704/1977 (H13N6) | 0/8 | 0/8 | 0/8 |
| Chicken | A/laughing gull/NJ/AI08–1460/2008 (H13N9) | 0/8 | 0/8 | 0/8 |
| Chicken | A/ring-billed gull/MN/AI10–1708/2010 (H13N6) | 0/8 | 0/8 | 0/8 |
| Chicken | A/sand (environment)/NJ/1–24/NJ/2010 (H13N6) | 0/8 | 0/8 | 0/8 |
| Chicken | A/turkey/MN/1012/1991 (H13N2) | 0/8 | 0/8 | 0/8 |
| Chicken | A/ruddy turnstone/NJ/AI09–294/2009 (H13N6) | 0/8 | 0/8 | 0/8 |
| Chicken | A/laughing gull/NJ/AI08–1388/2008 (H13N9) | 0/8 | 0/8 | 0/8 |
| Chicken | A/gull/ND/44036/1992 (H13N6) | 0/8 | 0/8 | 0/8 |
| Chicken | A/gull/MD/704/1977 (H13N6) | 0/8 | 0/8 | 0/8 |
Serologic testing was performed in all groups using a commercial bELISA.
Turkeys and mallards were resistant to infection with most H13 LPAI viral strains, but low numbers of inoculated birds were infected after challenge with specific strains (Table 1). One of eight turkeys inoculated with A/turkey/MN/1012/1991 (H13N2) seroconverted and was virus isolation positive on the PID 2 oropharyngeal swab, and one of the eight turkeys inoculated with A/gull/ND/44036/1992 (H13N6) was virus isolation positive on the PID 2 cloacal swab; post-inoculation antibodies were not detected in the latter turkey. No virus was isolated or antibodies detected in the remaining inoculated or contact turkeys in any group. Three mallards inoculated with A/ring-billed gull/MN/AI10–1708/2010 (H13N6) were virus isolation positive on PID 2 swabs and were also seropositive on post-inoculation serum samples. Two of these mallards were virus isolation positive on both the PID 2 oropharyngeal and cloacal swabs, while the third was positive only on the oropharyngeal swab. All PID 4 swab samples from these three mallards were virus isolation negative. No infections were detected in any of the remaining inoculated or contact mallards in any of the challenge groups. No virus was isolated from PID 2 or 4 swab samples and antibodies were not detected in post-inoculation serum samples collected from any of the inoculated or contact chickens exposed to the eight H13 LPAI viruses (Table 1). Although only two turkeys swab samples were positive by virus isolation, an additional 17 oropharyngeal or cloacal swab samples from chickens and turkeys inoculated with virus tested positive with qRRT-PCR but were negative on virus isolation (Table 2). The following six contact-exposed chicken and turkey samples were qRRT-PCR positive/virus isolation negative on PCD 2: two turkeys (oropharyngeal swabs) in the A/laughing gull/NJ/AI08–1460/2008 (H13N9) group; one turkey (cloacal swab) in the A/laughing gull/NJ/AI08–1388/2008 (H13N9) group; one turkey (oropharyngeal swab) in the A/turkey/MN/1012/1991 (H13N2) group; and two chickens (oropharyngeal) in the A/turkey/MN/1012/1991 (H13N2) group.
Table 2.
Summary of qRRT-PCR test results on oropharyngeal (OP) and cloacal (CL) swabs collected from turkeys and chickens inoculated with various H13 LPAI virus strains.
| Species | Virus | Viral shedding |
|||
|---|---|---|---|---|---|
| PID 2 |
PID 4 |
||||
| OP | CL | OP | CL | ||
|
| |||||
| Turkey | A/ laughing gull /NJ/AI08–1460/2008 (H13N9) | 1/8 (37.2) | 1/8 (37.2) | 0/8 | 0/8 |
| Turkey | A/ring-billed gull/MN/AI10–1708/2010 (H13N6) | 0/8 | 0/8 | 0/8 | 0/8 |
| Turkey | A/sand (environment)/NJ/1–24/2010 (H13N6) | 0/8 | 0/8 | 0/8 | 0/8 |
| Turkey | A/turkey/MN/1012/1991 (H13N2) | 3/8 (33.4–37.7)A | 1/8 (37.0) | 1/8 (26.3) | 0/8 |
| Turkey | A/ruddy turnstone/NJ/AI09–294/2009 (H13N6) | 0/8 | 2/8 (37.8–37.9) | 0/8 | 0/8 |
| Turkey | A/laughing gull/NJ/AI08–1388/2008 (H13N9) | 2/8 (37.4–37.1) | 0/8 | 0/8 | 0/8 |
| Turkey | A/gull/ND/44036/1992 (H13N6) | 0/8 | 2/8 (36.5–37.2)A | 0/8 | 0/8 |
| Turkey | A/gull/MD/704/1977 (H13N6) | 2.8 (35.0–37.3) | 0/8 | 0/8 | 0/8 |
| Chicken | A/laughing gull/NJ/AI08–1460/2008 (H13N9) | 0/8 | 0/8 | 0/8 | 0/8 |
| Chicken | A/ring-billed gull/MN/AI10–1708/2010 (H13N6) | 0/8 | 0/8 | 0/8 | 0/8 |
| Chicken | A/sand (environment)/NJ/1–24/NJ/2010 (H13N6) | 1/8 (35.9) | 0/8 | 0/8 | 0/8 |
| Chicken | A/turkey/MN/1012/1991 (H13N2) | 2/8 (32.5–35.5) | 0/8 | 0/8 | 0/8 |
| Chicken | A/ruddy turnstone/NJ/AI09–294/2009 (H13N6) | 0/8 | 0/8 | 0/8 | 0/8 |
| Chicken | A/laughing gull/NJ/AI08–1388/2008 (H13N9) | 1/8 (37.9) | 0/8 | 0/8 | 0/8 |
| Chicken | A/gull/ND/44036/1992 (H13N6) | 0/8 | 0/8 | 0/8 | 0/8 |
| Chicken | A/gull/MD/704/1977 (H13N6) | 0/8 | 0/8 | 0/8 | 0/8 |
Virus isolated from the qRRT-PCR positive sample.
Virus was isolated and titrated from all seven ring-billed gulls infected with A/laughing gull/NJ/AI08–1460/2008 (H13N9). In all gulls, viral shedding started between PID 0 and 2 and continued, in surviving birds, up to PID 10, at which time low levels of virus were detected at sporadic time points until the end of the trial at PID 17. All seven infected gulls excreted the H13N9 LPAI virus via both the oropharynx and cloaca. Viral shedding was similar between the two routes, but the concentration and duration were slightly greater via the oropharynx (Fig. 1). Peak titers in oropharyngeal swabs occurred between PID 1 to 3 and ranged from 2.1 to 5.4 log 10 EID50/ml. Peak titers in cloacal swabs occurred later, from PID 3 to 6, and ranged from below a quantifiable titer to 4.7 log 10 EID50/ml. The two gulls euthanatized on PID 1 had mild tracheal lesions, consisting of patchy deciliation (n = 2) and mild lymphocytic tracheitis (n = 1). The gull with mild tracheitis had nuclear staining for AI viral antigen in rare tracheal epithelial cells (Fig. 2). No viral antigen or microscopic lesions associated with LPAI viral infection were detected in any of the remaining gull tissues.
Fig. 1.
Mean (6 standard error) oropharyngeal and cloacal infectious viral titers excreted by ring-billed gulls (n = 8) experimentally inoculated with A/laughing gull/NJ/AI08–1460/2008 (H13N9). Two gulls were euthanatized and necropsied on PID 1, 3, 7, and 17, resulting in a decreasing sample size over the course of the study: PID 0–1 (n = 8); PID 2–3 (n = 6); PID 4–7 (n = 4); and PID 8–17 (n = 2). Viral titers are reported in log 10 EID50/ml.
Fig. 2.
Trachea from a ring-billed gull infected with A/laughing gull/NJ/AI08–1460/2008 (H13N9) at PID 1. Rare tracheal epithelial cells contain nuclear staining for AI virus NP antigen (brown staining). Immunoperoxidase labeling, hematoxylin counterstain (bar = 50 μm).
DISCUSSION
The experimental results presented herein are consistent with existing field and experimental data and indicate H13 LPAI viruses are strongly adapted to gulls, but rare spill-over into aberrant hosts is possible. The strong adaptation to gulls was evidenced by a high proportion (7/8; 87.5%) of infection in gulls challenged with the reduced titer inoculum (104 EID50), and extensive viral shedding via the oropharynx and cloaca without overt clinical signs of disease. This is in contrast to the poor adaptation of all H13 LPAI virus strains used in this study to the remaining three species, which were either refractory or experienced rare infections after challenge with high titer inoculums (106–107 EID50), had very transient viral shedding, and no intraspecies transmission to contact birds. The mechanisms driving the adaptation of AI viruses to different host species are not completely understood but likely involve multiple interactions associated with viral infection and replication in host tissues, including HA binding to sialic acid (SA) receptors on the host cell, the ability to express viral polymerase genes and utilize the host cell machinery for viral replication, and efficient NA-dependent release of progeny virus from the host cell membrane (15).
In regards to the HA binding, the expression of different avian-type SA receptors in the tissues of laughing gulls, ring-billed gulls, and mallards generally corresponds with the observed host ranges of wild bird-origin LPAI viruses in these species (7). Fine receptor specificity of HA for SA receptors on the host cell is dependent on the linkage between the SA 2,3Gal-disaccharide and the penultimate N-acetylhexosamine residue of the carbohydrate chain of SA. Based on in vitro trials, duck-origin LPAI viruses preferentially bind to SAα2,3Galβ1,3GalNacβ1 (β1,3Gal), whereas gull-origin LPAI viruses prefer SAα2,3Galβ1,4GlcNacβ1 (β1,4Glc) (8). Using lectin histochemistry, Franca et al. (7) recently reported laughing gulls and ring-billed gulls strongly express both avian-type SA receptors in the respiratory and intestinal tracts, and consequently, should be susceptible to both duck- and gull-origin AI viruses. In contrast, mallards predominately expressed β1,3Gal in the respiratory and intestinal tracts, suggesting they would be more susceptible to duck-origin viruses. The results of these receptor studies are consistent with field and experimental data on these three species. Although multiple HA subtypes (H1–12) are routinely isolated from mallards, viruses of the H13 and H16 subtypes, to our knowledge, have not. Experimentally, mallards are highly susceptible to duck-origin LPAI viruses, but are comparably much more resistant to gull-adapted strains of the H13 subtype (3,12,26). In contrast, natural infection with numerous HA subtypes have been reported from laughing and ring-billed gulls (9,14,19), including viruses of the H13 and H16 subtypes, and experimentally, gulls are permissive hosts for both duck- and gull-origin LPAI viruses (3).
Based on virus isolation and serologic testing, none of the chickens became infected after challenge with any of the eight strains of H13 LPAI virus. Two inoculated turkeys had evidence of infection based on virus isolation, and one of these birds seroconverted (Table 1). None of the contact chickens or turkeys in any challenge group had serologic or virologic (isolation) evidence of infection. Twenty-three oropharyngeal or cloacal swab samples collected from inoculated (n = 17) or contact-exposed (n = 6) chickens or turkeys were positive on qRRT-PCR, but attempts to isolate virus from these samples were unsuccessful (Table 2). The majority of these qRRT-PCR positive/virus isolation negative turkey and chicken samples were oropharyngeal swabs (74%; 17/23), samples collected on PID 2 (96%; 22/23), and had a relatively high Ct value of >35 (78%; 18/23). Potential explanations for the discordant test results on these samples include detection of residual inoculum, low levels of replication below the detection threshold of isolation, and/or freeze-thaw artifact. These explanations are not mutually exclusive and all may have contributed to these results. Regardless of the mechanism involved, collectively our data suggest these discordant results are biologically insignificant.
One of the two H13 LPAI virus strains that infected turkeys in this study was originally isolated from a turkey (A/turkey/MN/1012/1991 [H13N2]) (20). Our experimental results with this strain are consistent with a previous study by Laudert et al. (16), in which A/turkey/MN/1012/1991 (H13N2) produced infections in turkeys after oculonasal inoculation, but not in chickens. The biologic properties of this H13N2 LPAI virus (i.e., infectivity for turkeys) may reflect some degree of viral adaptation to this species; however, our experimental results suggest this host-range is quite unique because turkeys are generally resistant to infection with most strains of H13 LPAI virus.
Ring-billed gulls excreted the H13N9 LPAI virus via the oropharynx and cloaca for several days; however, the duration and concentration of shedding were slightly greater via the oropharynx. Both gulls at PID 1 had mild microscopic changes in the trachea and minimal viral antigen was detected in one of these birds. Although the microscopic changes in the trachea may have been associated with viral replication, it is also possible that they were a result of trauma from intratracheal inoculation. The lack of viral antigen detection in the remaining immunohistochemically stained tissues of the infected gulls likely reflects the low level of viral replication and shedding at the time of examination, which ranged from below the detectable limit of titration to 3.1 log 10 EID50/ml in oropharyngeal swabs and 0 to 2.9 log 10 EID50/ml in cloacal swabs. This is supported by the one gull with positive staining for viral antigen excreting a higher viral titer in the PID 1 oropharyngeal swab (5.4 log 10 EID50/ml).
The excretion of virus via the oropharynx is consistent with the results reported for juvenile laughing gulls infected with multiple duck-origin LPAI viruses (3) and is in contrast to LPAI virus infection in mallards, in which fecal shedding is dominant (3,26). Such differences in viral shedding between gulls and mallards may have impacts on routes of transmission within these distinct reservoir groups, as well as the potential risk for virus spillover to aberrant hosts. Within duck populations, LPAI viruses are thought to be transmitted by the fecal-oral route through contaminated aquatic habitats (11). This is an extremely efficient transmission mechanism in ducks due to the high concentrations of virus excreted in the feces (2), the tenacity of LPAI viruses in water (22), and the high concentrations of susceptible ducks that can congregate on aquatic habitats. This is also an efficient means for LPAI virus spillover to domestic animals because both the infected feces and contaminated water can serve as a potential source of virus (10). The routes of LPAI virus transmission in gulls are currently not known; however, considering the extent of viral shedding via the respiratory and intestinal tracts, there may be multiple potential routes, particularly on breeding colonies where the density of naïve chicks is high. The pattern of LPAI virus shedding in gulls also has important implications for sampling these birds. Surveillance efforts focusing on understanding the natural history of AI (i.e., LPAI viruses) generally sample wild bird species by collecting cloacal swabs and/or feces, which is based largely on LPAI virus infection in ducks. However, existing experimental data on gulls suggests that oropharyngeal swabs or pooled oropharyngeal and cloacal swabs may be more efficient samples to collect than cloacal swabs or feces alone.
Collectively, these experimental trials provide valuable data that improve our understanding on the unique biology of H13 LPAI viruses and support ongoing field studies, both for interpreting surveillance results and improving future sampling efforts in gulls. Although not examined in this current study, terns are taxonomically related to gulls and are also frequently cited as reservoirs for AI virus; however, surveillance to date has been limited in this group of birds, relative to gulls. Consequently, although AI viruses have been reported from multiple tern species, there currently is not enough field or experimental data on influenza in terns to understand their role in the ecology of AI virus. Many unanswered questions regarding the biology of H13 LPAI viruses remain and further research is needed to identify mechanisms underlying the unique host-range of these viruses; particularly as such information may have value in understanding the inherent resistance of some poultry species to AI (i.e., chickens).
ACKNOWLEDGMENT
We thank Whitney Kistler, Shamus Keeler, Lee Harper, and the personnel of the New York Department of Environmental Conservation for assistance in acquiring the ring-billed gulls used in this study. Additionally, we thank the faculty and staff of the Southeastern Cooperative Wildlife Disease Study (UGA), Poultry Diagnostic and Research Center (UGA), and SEPRL for support with animal husbandry and laboratory testing. We also acknowledge the histotechnicians in the Department of Pathology at the UGA, particularly Abbie Butler and Patricia Rowe for performing the hematoxylin and eosin and immunohistochemical staining. Funding for this work was provided by the National Institute of Allergy and Infectious Diseases, National Institutes of Health (NIH), Department of Health and Human Services, under Contract No. HHSN266200700007C and the National Science Foundation under grant No. DEB-0917853. The opinions expressed herein are those of the authors and do not necessarily reflect the views of any of the funding agencies.
Abbreviations:
- AI
avian influenza
- bELISA
blocking enzyme-linked immunosorbent assay
- BHI
brain-heart-infusion
- BSL
biosafety level
- EID50
median embryo infectious doses
- HA
hemagglutinin
- HP
highly pathogenic
- LP
low pathogenic
- NA
neuraminidase
- NP
nucleoprotein
- PCD
post-contact day
- PID
post-inoculation day
- qRT-PCR
quantitative real-time reverse-transcription polymerase chain reaction
- SA
sialic acid
- SCWDS
Southeastern Cooperative Wildlife Disease Study
- SEPRL
Southeast Poultry Research Laboratory
- SPF
specific pathogen free
- UGA
University of Georgia
REFERENCES
- 1.Becker WB The isolation and classification of tern virus A/Tern/South Africa/61. J. Hyg. 64:309–320. 1966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Brown JD, Berghaus RD, Costa TP, Poulson R, Carter DL, Lebarbenchon C, and Stallknecht DE. Intestinal excretion of a wild bird-origin H3N8 low pathogenic avian influenza virus in mallards (Anas platyrhynchos). J. Wildl. Dis. in review. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Costa TP, Brown JD, Howerth EW, and Stallknecht DE. Variation in viral shedding patterns between different wild bird species infected experimentally with low-pathogenicity avian influenza viruses that originated from wild birds. Avian Pathol. 40:119–124. 2011. [DOI] [PubMed] [Google Scholar]
- 4.Das A, Spackman E, Pantin-Jackwood MJ, and Suarez DL. Removal of real-time reverse transcription polymerase chain reaction (RT-PCR) inhibitors associated with cloacal swab samples and tissues for improved diagnosis of avian influenza virus by RT-PCR. J. Vet. Diagn. Invest. 21:771–778. 2009. [DOI] [PubMed] [Google Scholar]
- 5.Driskell EA, Jones CA, Stallknecht DE, Howerth EW, and Tompkins SM. Avian influenza virus isolates from wild birds replicate and cause disease in a mouse model of infection. Virology 399:280–289. 2010. [DOI] [PubMed] [Google Scholar]
- 6.Fouchier RAM, Munster VJ, Keawcharoen J, E Osterhaus ADM, and Kuiken T. Virology of avian influenza in relation to wild birds. J. Wildl. Dis. 43(Suppl.):S7–S14. 2007. [Google Scholar]
- 7.Franca M, Stallknecht DE, Poulson R, Brown J, and Howerth EW. The pathogenesis of low pathogenic avian influenza in mallards. Avian Dis. 56(suppl. 1):676–980. 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Gambaryan A, Yamnikova S, Lvov D, Tuzikov A, Chinarev A, Pazynina G, Webster R, Matrosovich M, and Bovin N. Receptor specificity of influenza viruses from birds and mammals: new data on involvement of the inner fragments of the carbohydrate chain. Virology 334:276–283. 2005. [DOI] [PubMed] [Google Scholar]
- 9.Graves IL Influenza viruses in birds of the Atlantic flyway. Avian Dis. 36:1–10. 1992. [PubMed] [Google Scholar]
- 10.Halvorson DA, Kelleher CJ, and Senne DA. Epizootiology of avian influenza: effect of season on incidence in sentinel ducks and domestic turkeys in Minnesota. Appl. Environ. Microbiol. 49:914–919. 1985. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hinshaw VS, Webster RG, and Turner B. Water-borne transmission of influenza A viruses? Intervirology 11:66–68. 1979. [DOI] [PubMed] [Google Scholar]
- 12.Hinshaw VS, Air GM, Gibbs AJ, Graves L, Prescott B, and Karunakaran D. Antigenic and genetic characterization of a novel hemagglutinin subtype of influenza A viruses from gulls. J. Virol. 42:865–872. 1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hinshaw VS, Bean WJ, Geraci J, Fiorelli P, Early G, and Webster RG. 1986. Characterization of two influenza A viruses from a pilot whale. J. Virol. 58:655–656. 1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kawaoka Y, Chambers TM, Sladen WL, and Webster RG. Is the gene pool of influenza viruses in shorebirds and gulls different from that in wild ducks? Virology 163:247–250. 1988. [DOI] [PubMed] [Google Scholar]
- 15.Kuiken T, Holmes EC, McCauley J, Rimmelzwaan GF, Williams CS, and Grenfell BT. Host species barriers to influenza virus infections. Science 312:394–397. 2006. [DOI] [PubMed] [Google Scholar]
- 16.Laudert E, Sivanandan V, Halvorson D, Shaw D, and Webster RG. Biological and molecular characterization of H13N2 influenza type A viruses isolated from turkeys and surface water. Avian Dis. 37:793–799. 1993. [PubMed] [Google Scholar]
- 17.Olsen B, Munster VJ, Wallensten A, Waldenstrom J, Osterhaus AD, and Fouchier RA. Global patterns of influenza A virus in wild birds. Science 312:384–388. 2006. [DOI] [PubMed] [Google Scholar]
- 18.Reed LJ, and Muench H. A simple method of estimating fifty percent endpoints. Am. J. Hyg. 27:493–497. 1938. [Google Scholar]
- 19.Sinnecker R, Sinnecker H, Zilske E, and Kohler D. Surveillance of pelagic birds for influenza A viruses. Acta Virol. 27:75–79. 1983. [PubMed] [Google Scholar]
- 20.Sivanandan V, Halvorson DA, Laudert E, Senne DA, and Kumar MC. Isolation of H13N2 influenza A virus from turkeys and surface water. Avian Dis. 35:974–977. 1991. [PubMed] [Google Scholar]
- 21.Spackman E, Senne DA, Myers TJ, Bulaga LL, Garber LP, Perdue ML, Lohman K, Daum LT, and Suarez DL. Development of a real-time reverse transcriptase PCR assay for type A influenza virus and the avian H5 and H7 hemagglutinin subtypes. J. Clin. Microbiol. 40:3256–3260. 2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Stallknecht DE, Shane SM, Kearney MT, and Zwank PJ. Persistence of avian influenza virus in water. Avian Dis. 34:406–411. 1990. [PubMed] [Google Scholar]
- 23.Stallknecht DE, and Brown JD. Ecology of avian influenza in wild birds. In: Avian influenza, 1st ed. Blackwell Publishing, Ames, IA. pp. 43–58. 2008. [Google Scholar]
- 24.Stallknecht DE, Luttrell MP, Poulson R, Goekjian V, Niles L, Dey A, Krauss S, and Webster RG. Surveillance and testing approaches to detect and recover avian influenza viruses from shorebird populations. J. Wildl. Dis 48:382–393. 2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Swayne DE, Senne DA, and Suarez DL. Influenza. In: A laboratory manual for the isolation and identification of avian pathogens, 5th ed. American Association of Avian Pathologists, Kennett Square, PA. pp. 128–134. 2008. [Google Scholar]
- 26.Webster RG, Yakhno M, Hinshaw VS, Bean WJ, and Murti KG. Intestinal influenza: replication and characterization of influenza viruses in ducks. Virology 84:268–278. 1978. [DOI] [PMC free article] [PubMed] [Google Scholar]