点击显示 收起
Department of Veterinary Microbiology and Pathology, Washington State University, Pullman, Washington 99164-7040
The Zoological Society of San Diego, Center for Reproduction of Endangered Species, P.O. Box 120551, San Diego, California 92112-0551
The Peregrine Fund, 5668 West Flying Hawk Lane, Boise, Idaho 83709
ABSTRACT
An adenovirus was detected by electron microscopy in tissues from falcons that died during an outbreak of inclusion body hepatitis and enteritis that affected neonatal Northern aplomado (Falco femoralis septentrionalis) and peregrine (Falco peregrinus anatum) falcons. Molecular characterization has identified the falcon virus as a new member of the aviadenovirus group (M. Schrenzel, J. L. Oaks, D. Rotstein, G. Maalouf, E. Snook, C. Sandfort, and B. Rideout, J. Clin. Microbiol. 43:3402-3413, 2005). In this study, the virus was successfully isolated and propagated in peregrine falcon embryo fibroblasts, in which it caused visible and reproducible cytopathology. Testing for serum neutralizing antibodies found that infection with this virus was limited almost exclusively to falcons. Serology also found that wild and captive peregrine falcons had high seropositivity rates of 80% and 100%, respectively, although clinical disease was rarely reported in this species. These data implicate peregrine falcons as the natural host and primary reservoir for the virus. Other species of North American falcons, including aplomado falcons, had lower seropositivity rates of 43 to 57%. Falcon species of tropical and/or island origin were uniformly seronegative, although deaths among adults of these species have been described, suggesting they are highly susceptible. Chickens and quail were uniformly seronegative and not susceptible to infection, indicating that fowl were not the source of infection. Based on the information from this study, the primary control of falcon adenovirus infections should be based on segregation of carrier and susceptible falcon species.
INTRODUCTION
An adenovirus was detected by electron microscopy in tissues from falcons that died during an outbreak of inclusion body hepatitis and enteritis that affected neonatal Northern aplomado falcons (Falco femoralis septentrionalis) (12). The outbreak occurred in 1996 at a raptor propagation facility that, in addition to neonatal falcons, housed 72 adult aplomado falcons, 94 American peregrine falcons (Falco peregrinus anatum), three Vanuatu peregrine falcons (Falco peregrinus nesiotes), three gyrfalcons (Falco rusticolus), 11 harpy eagles (Harpia harpyja), 20 California condors (Gymnogyps californianus), and two Andean condors (Vultur gryphus). The affected neonatal aplomado falcons were all housed together in a single brooder room. The outbreak had an overall attack rate of 72/110 (65%) and a case fatality rate of 62/72 (86%). Strikingly, 102 American peregrine falcon neonates that were being raised simultaneously in the same brooder room had an attack rate of only 6/102 (6%). None of the adult falcons, eagles, or condors had any illness or death compatible with viral infection during the disease outbreak. The exclusive food source for the neonatal aplomado and peregrine falcons was Japanese quail (Coturnix japonica) raised at the facility. The quail were raised in a building along with domestic chickens (Gallus gallus domesticus) that were used to feed the other raptors. In addition, large numbers of pigeons (Columba livia) and chukar (Alectoris chukar) were free ranging on the grounds of the facility. No unusual disease was noted in any of these other bird species during the outbreak.
Epidemiologic investigation (12) indicated that the outbreak began with infection of a single neonatal aplomado falcon, with subsequent horizontal spread among the rest of the aplomado falcon chicks in the brooder room. However, the source of the initial infection was not known. The eggs from which these falcons hatched were naturally incubated for about 1 week by either peregrine or aplomado falcons, followed by artificial incubation. After hatching, the neonates were all hand raised and thus had no direct contact with any adult falcons. In the few previous reports of adenovirus infections of falcons (4, 10, 14), the source of infection was not determined. Forbes et al. (4) hypothesized that the agent was a poultry virus and that exposure of falcons occurred through food sources. This was based on isolation of an adenovirus from the chickens and turkeys used to feed the falcons. However, the ubiquitous nature of adenoviruses and the lack of any direct comparison of the falcon and fowl isolates made the association circumstantial (4).
Molecular characterization has identified the falcon virus as a new member of the aviadenovirus group (12); this new member is most closely related to the group I avian adenoviruses. The original working hypothesis concerning the outbreak was that the virus was introduced to the brooder room via the food quail. However, evidence that this adenovirus was distinct from the known poultry adenoviruses challenged the hypothesis that it originated from the food source. The purpose of this study was to isolate and propagate the falcon adenovirus and to use the virus for challenge studies and as an antigen in serologic assays to define the host range and natural history of the falcon adenovirus. This information would also be useful in determining the source of the virus in the original outbreak and in designing strategies to prevent future outbreaks.
MATERIALS AND METHODS
Virus isolation. Primary and permanent cell cultures were used to attempt in vitro propagation of the adenovirus. Primary fibroblast cell cultures were prepared from 11- to 13-day-old domestic chicken embryos, 12-day-old Japanese quail embryos, 12- to 21-day-old domestic duck embryos, and 17- to 25-day-old peregrine falcon embryos by methods described by Docherty and Slota (3). Primary duck and quail embryo fibroblast-like cultures were also prepared from the livers of 20- to 25-day-old duck embryos and 12-day-old quail embryos. The procedure for preparing liver fibroblasts was adapted from the method of Docherty and Slota (3) to obtain fibroblasts from connective tissues. Fibroblasts from the liver were selected for, and hepatocytes were removed by, the level of protease digestion (0.05% trypsin and 0.02% EDTA at 37°C for 15 min), and cells were subcultured to select for dividing cells. Although these cells have not been characterized, they have the typical spindle-shaped morphology of fibroblasts. The chicken, quail, and duck cells were maintained in Eagle's minimal essential medium (MEM), and the falcon cells were maintained in M199 medium. All media were supplemented with 10% (vol/vol) fetal bovine serum, 2 mM L-glutamine, 100 U ml–1 penicillin, 0.1 mg ml–1 streptomycin, and 2.5 μg ml–1 amphotericin B. Permanent cell lines of avian origin included a chicken liver cell line (LMH) and a Japanese quail fibroblast cell line (QT35) maintained as previously described (11).
Tissue pools of approximately 1 g of liver and 0.5 g of spleen from affected aplomado falcons were first homogenized with Dounce tissue grinders in 10 ml of MEM supplemented with 200 U ml–1 penicillin and 0.2 mg ml–1 streptomycin and then briefly centrifuged at 750 x g to remove tissue and cellular debris; then 500 μl of the sample was inoculated onto confluent monolayers of cells in 25-cm2 flasks. The sample was incubated with the cells for 1 h at 37°C, after which time an additional 5 ml of supplemented medium was added to the flasks. The cell cultures were incubated under 5% CO2 at 37°C for 7 to 14 days and observed for cytopathology. Cell cultures were passaged by freezing and thawing the flasks, collecting the cells and supernatants, and adding 1 ml of this lysate to new cells in 25-cm2 flasks.
Electron microscopy. Portions (1.5) ml of cell supernatants were collected from cell cultures with visible cytopathology, centrifuged at 1,000 x g for 10 min to pellet large cellular debris, and centrifuged at 8,000 x g for 10 min to further remove debris; then 1 ml of the clarified supernatant was centrifuged at 13,000 x g for 30 min to pellet virions. The virus pellet was resuspended in 50 μl of distilled water, negatively stained with 1% phosphotungstic acid for 30 s, mounted onto 200-mesh Formvar-coated copper grids, and viewed by transmission electron microscopy.
PCR assay and sequencing. Falcon adenovirus was detected in cell cultures or tissues by using a PCR assay specific for the falcon adenovirus hexon gene as described by Schrenzel et al. (12). DNA was extracted and purified from infected cell monolayers or tissue samples by standard phenol-chloroform and ethanol precipitation methods, and 1 μg of total DNA was used in each PCR. The identity of the amplified DNA as falcon adenovirus was confirmed by cloning the amplicon into the pCR2.1 sequencing vector (TOPO TA cloning kit; Invitrogen, Carlsbad, CA), and sequencing by automated dideoxy DNA methods.
Serology. The serum neutralization (SN) assay was modified from a previously published microneutralization procedure (5). Serum or plasma was collected from various species of raptors and other birds, stored at –20°C, and heat inactivated at 56°C for 30 min prior to use. Stocks of the falcon adenovirus were prepared by propagating the virus in peregrine embryo fibroblasts in 25-cm2 flasks until maximal cytopathology was evident. At this time, the cell monolayers were harvested by mechanically scraping free the cell monolayers, releasing intracellular virions by two freeze-thaw cycles, centrifuging to pellet cellular debris, and making 1-ml aliquots of the virus, which were stored at –80°C until use. The titer of the virus stock was determined by performing eight replicate 10-fold serial dilutions on peregrine embryo fibroblasts and scoring the well as positive or negative for cytopathology. The titer as the median tissue culture infective dose (TCID50) was then calculated by the method of Reed and Muench (9). Stocks of fowl adenovirus type 1 (chicken embryo lethal orphan strain) were similarly prepared and titered.
To perform the SN assay, 100 TCID50s of virus stock were mixed with test sera/plasma and cell culture media to give final serum dilutions of 1:5, 1:10, or 1:40 in a total volume of 100 μl. For selected cases, twofold serial dilutions of the sera/plasma were made (range, 1:10 to 1:5,120) to detect the end point titers. The serum-virus mixture was incubated at 37°C for 1 h, and then the entire 100 μl was added to newly confluent peregrine embryo fibroblasts in 96-well plates. The plates were then incubated at 37°C under a 5% CO2 atmosphere for 10 to 14 days, after which time the wells were scored as positive or negative for cytopathology. The titer was reported as the reciprocal of the highest dilution that completely inhibited cytopathology.
Experimental infections of live quail. All experiments using live birds were approved by the Washington State University Institutional Animal Care and Use Committee. To attempt to reproduce disease and determine the host range for the virus, Japanese quail were infected with tissue pools from infected aplomado falcons. These tissue pools were shown to have infectious virus by isolation in cell culture. The source flock for the quail was negative for adenovirus titers by agar gel precipitation assays against group I and group II aviadenoviruses (performed commercially by the Poultry Diagnostic and Research Center at the University of Georgia). Two groups of three 2-day-old quail each were inoculated either orally with 10 μl or by intramuscular injection with 10 μl of liver/spleen tissue homogenates from aplomado falcons RP5958 and 96JT1,6. One quail in each group was also left uninoculated as a contact control. At 19 days postinoculation, all of the quail in both groups, including the contact controls, were reinoculated by intramuscular injection with 100 μl of tissue homogenates from the respective falcons. Another two groups of two 8-week-old quail each were inoculated using 100 μl for the initial oral and intramuscular administrations and 200 μl for the subsequent intramuscular reinoculation. Groups of four age-matched control quail were housed in the same room but in separate cages. The birds were observed daily for any clinical evidence of disease. At day 29 postinoculation, all the birds were euthanized for collection of serum and tissues. SN assays for antibodies to fowl adenovirus type 1 in experimentally infected quail were performed commercially by the Poultry Diagnostic and Research Center, University of Georgia.
RESULTS
Virus isolation. Tissue pools were prepared from three aplomado falcons (RP5958, RP5963, and 96JT1,6) that were shown to be infected by histopathology and PCR (12). These three tissue pools were initially inoculated onto quail embryo fibroblasts and quail liver fibroblasts, because Japanese quail were hypothesized to be the primary host for the virus. Very slight cytopathology characterized by rounded, refractile cells was noted by day 3 postinfection in both cell types for all three cases. Electron microscopy on the supernatants from cases RP5958 and RP5963 collected at day 14 postinfection identified an adenovirus in the cultures from these two cases (Fig. 1). However, the cytopathology did not progress after day 3 postinfection and was lost upon subsequent passage. LMH and QT35 cells inoculated with samples from cases RP5958, RP5963, and 96JT1,6 did not have detectable cytopathology after three passages.
Duck embryo fibroblasts and duck liver fibroblasts were also inoculated with the tissue pool from 96JT1,6. Both cell types showed moderate cytopathology characterized by round, refractile cells up to the third passage. In the embryo fibroblasts, visible cytopathology was lost at the fourth passage and not detected with seven additional passages. However, in the liver fibroblasts, weak and sporadic cytopathology was detected between passages 4 and 11. Attempts to rescue cytopathology by passaging material from infected duck liver fibroblasts back to primary duck or chicken fibroblasts were unsuccessful. Subsequent electron microscopy of the third-passage duck liver fibroblasts detected adenovirus particles, and PCR testing revealed that all of the passages of both the duck embryo fibroblasts and duck liver fibroblasts were infected with the falcon adenovirus, although visible cytopathology was not apparent.
The cell culture supernatant from the third passage of duck liver fibroblasts for case 96JT1,6, which was positive by PCR and electron microscopy, was used to inoculate peregrine falcon embryo fibroblasts. In the peregrine cells, cytopathology characterized by rounded, refractile cells was detectable by day 4 to 5 postinfection and then progressed over the next 5 days to extensive cytopathology with many pyknotic cells (Fig. 2). Five subsequent passages onto fresh peregrine falcon fibroblasts resulted in extensive and reproducible cytopathology in all passages by about day 10 postinfection and titers of approximately 2 x 103 to 3 x 103 TCID50s ml–1. These passages were also all PCR positive. Comparison of the sequences of the PCR amplicons from falcon tissues and cell cultures confirmed that the cultured adenovirus was the falcon adenovirus.
Serology. Identification of peregrine embryo fibroblasts as an in vitro cell culture system with a reproducible detection method (visible cytopathology) was an important prerequisite for developing the SN assay. The initial focus was to determine SN titers in birds at the breeding facility. Serum samples were collected in September 1996, 2 months after the end of the outbreak, from surviving neonatal aplomado falcons (n = 10), adult aplomado falcons (n = 6), and adult peregrine falcons (n = 6). All 22 falcons tested were seropositive, with SN titers of 5 to 40. End point titers were determined for five of the adult peregrines; they ranged from 640 to 1,280. Attempts to determine the end point titers for five of the surviving neonatal aplomado falcons showed that all birds had very high titers of 5,120. These results are summarized in Table 1.
Archived plasma samples from 1995 and January 1996 provided an opportunity to test for SN titers in the six adult aplomado falcons and six adult peregrine falcons prior to the outbreak. All of these birds were positive, with titers of 5 to 40, indicating that the virus was present at the facility at least 1.5 years prior to the outbreak. Seven adult aplomado falcons and five adult peregrine falcons were tested in 1999, 3 years after the outbreak. Four of the aplomado falcons were positive with titers of 10 to 40, and three were seronegative at a dilution of 1:10. All five of the peregrine falcons were strongly seropositive, with titers of 40. These data indicated either that the titers were persistent or that exposure was ongoing, especially in the peregrine falcons. These results are summarized in Table 1.
Subsequently, other raptors and species of birds at the breeding facility were also tested for SN antibody titers (Table 1). This included seven adult gyrfalcons tested in 2002, of which three (43%) had titers of 10 to 40 and four were seronegative at a dilution of 1:10. All other birds tested were seronegative at titers of 1:10, including 5 harpy eagles in 1999, 4 pigeons in 2001, 2 chukar partridges in 2001, 4 of the food quail in 2001, and 10 of the food chickens in 2002. These initial results indicated that infection with the falcon adenovirus was limited to birds in the genus Falco, and the high and persistent titers in the adult peregrine falcons suggested that infection was endemic in this species. An additional five chickens from mixed flocks not associated with the breeding facility were also tested and found seronegative at dilutions of 1:10 (Table 2).
Testing for SN titers was also performed on raptors not associated with the breeding facility (Table 2). All species of hawks and eagles tested were seronegative at a dilution of 1:10, including six red-tailed hawks (Buteo jamaicensis), two rough-legged hawks (Buteo lagopus), two bald eagles (Haliaeetus leucocephalus), and two golden eagles (Aquila chrysaetos). Among owls, three great-horned owls (Bubo virginianus) were seronegative at a dilution of 1:10, while the single barred owl (Strix varia) tested had a titer of 10. Among falcons, two American kestrels (Falcon sparverius) were seronegative at a dilution of 1:10 while two others each had a titer of 10. Eleven captive teita falcons (Falco fasciinucha) and seven orange-breasted falcons (Falco deiroleucus) tested were all seronegative at dilutions of 1:10. In contrast, among 30 wild-caught tundra peregrine falcons (Falco peregrinus tundrius), 16 (53%) had titers of 40, eight (27%) had titers of 10, and only six (20%) were seronegative. The end point titer of one of the seropositive wild peregrines was determined to be 640.
Cross-neutralization SN assays indicated serologic cross-reactivity between the falcon adenovirus and its closest genetic relative, fowl adenovirus type 1 (12). The five neonatal aplomado falcons and the five adult peregrine falcons with very high SN titers to the falcon adenovirus had SN activity against fowl adenovirus type 1. However, the titers to the falcon adenovirus ranged from 640 to 5,120 (median titer, >3,200), while the titers to fowl adenovirus type 1 ranged from 40 to 320 (median titer, 80). Thus, the type-specific SN titer for the falcon adenovirus was >40-fold higher than that for fowl adenovirus type 1. Quail with SN titers of 2 to 16 against fowl adenovirus type 1 did not have SN titers to the falcon adenovirus at dilutions of 1:5.
Experimental infections of live quail. None of the experimentally infected quail showed any evidence of clinical disease at any time during the experiment. At necropsy, no gross or microscopic lesions suggestive of adenovirus infection were identified. Seroconversion to the falcon adenovirus was not detected in any of the quail by SN with screening dilutions of 1:5 (Table 3), suggesting that the quail were resistant to infection as well as to clinical disease. Although the quail were seronegative for fowl adenovirus type 1 (group I) by agar gel precipitation assays, some quail had titers ranging from 4 to 32 by the more sensitive SN assay (Table 3). However, the lack of falcon adenovirus titers in the exposed quail, and the positive titers to fowl adenovirus type 1 in the control quail, indicated that these titers were most likely preexisting and not related to the falcon adenovirus.
DISCUSSION
Phylogenetic analysis of the DNA sequences of the polymerase and hexon genes indicates that the adenovirus causing disease in neonatal aplomado falcons is a new member of the genus Aviadenovirus most closely related to fowl adenovirus types 1 and 4 (12). Cross-neutralization data showing cross-reactivity in the SN assay between the falcon adenovirus and fowl adenovirus type 1 also suggest that these viruses are related (5, 8). However the >40-fold neutralization index between these two viruses clearly indicates that the falcon virus is serologically distinct from the serotype 1 viruses found in chickens and quail (5, 7). In addition, the evidence that SN antibodies to the falcon adenovirus were detected almost exclusively in members of the genus Falco, and that the virus replicated optimally in cell cultures of falcon origin, also indicates that the virus is a naturally occurring adenovirus of falcons. The only other avian species that was seropositive to the falcon adenovirus was a single barred owl. This owl may have been infected with a closely related virus that serologically cross-reacted with the falcon adenovirus or, alternatively, it may actually have been infected with the falcon adenovirus. Although there is insufficient information to draw any conclusions, it is interesting that owls and falcons, which are phylogenetically disparate (13), also share susceptibility to another typically host-specific virus, columbid herpesvirus 1 (1, 6). Hawks and eagles, in the family Accipitridae and phylogenetically distinct from the falcons, do not appear to be infected with the falcon adenovirus, a finding consistent with the typically narrow host range of well-adapted adenoviruses (2, 7). The negative SN titers in fowl and the inability to cause disease or seroconversion in experimentally infected quail also indicate that this virus does not occur in fowl and that the disease is not spread to falcons by food as was hypothesized in previous outbreaks in Mauritius kestrels (Falco punctatus) (4) and American kestrels (14). In the experimentally infected quail it was possible that the titers to the fowl adenovirus type 1 were protective against infection with the falcon adenovirus. However, the lack of disease or seroconversion in the 50% of the quail that were also seronegative for fowl adenovirus type 1 suggests that this was not a factor.
The data from this study also implicate the peregrine falcon as a primary reservoir host for the falcon adenovirus. The very high and persistent 100% seroprevalence in the captive peregrine falcons suggests a very high infection rate and endemicity, respectively. The high seroprevalence may in part be explained by the close physical proximity of the birds at a single facility. However, adult aplomado falcons and adult gyrfalcons kept on the same premises had much lower seroprevalence rates (57% and 43%, respectively), suggesting that they are not as susceptible to infection. The 80% SN seroprevalence in wild peregrine falcons, and the lower seroprevalence in other falcon species outside of the breeding facility, suggests that this virus is naturally endemic in peregrine falcons, although infection with this or a closely related virus does occur in other wild falcon species (e.g., American kestrels). The other finding suggesting that peregrine falcons are the natural host for the virus is their apparent resistance to clinical disease. In the aplomado falcon outbreak in which the attack rate for aplomado falcons was 65%, the attack rate for peregrine falcons living in close proximity was only 6%. One possible explanation for the resistance of the neonatal peregrines to disease is the SN titers in the parent birds and passive protection by antibody. However, the deaths of six neonatal peregrines and the deaths of aplomado falcons hatched from females shown to have preexisting SN titers (MEMI, TWA, and LO) indicate that passive antibody either is not protective or requires high titers for protection. These data also caution that intervention strategies such as vaccination of adults may not be adequate for prevention of outbreaks in susceptible populations.
There also appears to be considerable variation in susceptibility to clinical disease among the other species of falcons. Of particular interest are deaths of adult falcons, which suggest that these species are highly susceptible to infection and disease. Lethal adenovirus infections have been described among adult Mauritius kestrels (4), Fijian peregrine falcons (12), orange-breasted falcons (12), American kestrels (14), and a merlin (Falco columbarius) (10). Interestingly, adult mortality appears to be most common in falcon species of island/tropical origin and/or in the smaller falcon species such as kestrels and merlins. The reason for the increased susceptibility of these species in not known, but it can be speculated to be due either to genetic susceptibility or to a lack of natural exposure and immunity, or both. Although the sample sizes were small, the lack of SN titers in a group of captive adult teita falcons and a group of captive orange-breasted falcons suggests that the virus is not endemic in these species. Thus, captive populations of these species should be considered at especially high risk for falcon adenovirus disease. Deaths have been recorded for two neonatal teita falcons (A. Wunschmann, personal communication) produced from the above group of adults when the neonates were moved to a new facility with gyrfalcons and peregrine falcons. Although adult gyrfalcons may be seropositive, adenovirus deaths among neonates or adults have not been recorded, and thus, this species appears to be relatively resistant to clinical disease. Deaths were recorded for two neonatal gyrfalcon-peregrine hybrids that were housed with, and died subsequently to, the teita falcons (A. Wunschmann, personal communication). In this case, as for the peregrine falcons, inherent resistance to disease may be overcome in neonates with overwhelming challenge doses. Despite the high mortality rate among neonatal aplomado falcons during the 1996 outbreak, the ability of adult aplomado falcons to seroconvert without clinical disease and the apparent ability of neonates to survive certain levels of exposure (mortality decreased after implementation of biosecurity measures which presumably decreased the challenge doses, although serology showed that the falcons were still infected) suggest that aplomado falcons are also relatively resistant to disease.
Based on the information from this study, the primary control measure for falcon adenovirus disease should be based on segregation of and biosecurity between the carrier falcons and known or suspected susceptible species. In particular, until more information is available regarding the epidemiology and pathophysiology of this virus, great care should be taken if peregrine falcons are housed with other species of falcons.
ACKNOWLEDGMENTS
We thank Tori Byington, Shannon Donahoe, Shirley Elias, and Sue Pritchard for technical assistance and John El-Attreche at the University of Georgia for performing serology assays on experimentally infected quail. Nikol Finch, Martin Gilbert, Willard Heck, Tom Maechtle, Erik Stauber, Nadia Sureda, Terry Tinker, and Catherine Ulibarri generously collected and provided serum samples used in this study.
REFERENCES
Aini, I., L. M. Shih, A. E. Castro, and Y. C. Zee. 1993. Comparison of herpesvirus isolates from falcons, pigeons and psittacines by restriction endonuclease analysis. J. Wildl. Dis. 29:196-202.
Benko, M., and B. Harrach. 2003. Molecular evolution of adenoviruses. Curr. Top. Microbiol. Immunol. 272:3-35.
Docherty, D. E., and P. G. Slota. 1988. Use of Muscovy duck embryo fibroblasts for the isolation of viruses from wild birds. J. Tissue Culture Methods 11:165-170.
Forbes, N. A., G. N. Simpson, R. J. Higgins, and R. E. Gough. 1997. Adenovirus infection in Mauritius kestrels (Falco punctatus). J. Avian Med. Surg. 11:31-33.
Grimes, T. M., and D. J. King. 1977. Serotyping avian adenoviruses by a microneutralization procedure. Am. J. Vet. Res. 38:317-321.
Mare, C. J., and D. L. Graham. 1973. Falcon herpesvirus, the etiologic agent of inclusion body disease of falcons. Infect. Immun. 8:118-126.
McCracken, R. M., and B. M. Adair. 1993. Avian adenoviruses, p. 123-144. In J. B. McFerran and M. S. McNulty (ed.), Virus infections of birds. Elsevier Science Publishers, Amsterdam, The Netherlands.
Pallister, J. A., K. M. Erny, and K. J. Fahey. 1993. Serologic relationships with the group E fowl adenoviruses. Intervirology 36:84-90.
Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty per cent endpoints. Am. J. Hyg. 27:493-497.
Schelling, S. H., D. S. Garlick, and J. Alroy. 1989. Adenoviral hepatitis in a merlin (Falco columbarius). Vet. Pathol. 26:529-530.
Schnitzlein, W. M., J. Radzevicius, and D. N. Tripathy. 1994. Propagation of infectious laryngotracheitis virus in an avian liver cell line. Avian Dis. 38:211-217.
Schrenzel, M., J. L. Oaks, D. Rotstein, G. Maalouf, E. Snook, C. Sandfort, and B. Rideout. 2005. Characterization of a new species of adenovirus in falcons. J. Clin. Microbiol. 43:.
Sibley, C. G., and J. E. Ahlquist. 1990. Phylogeny and classification of birds. A study in molecular evolution. Yale University Press, New Haven, Conn.
Sileo, L., J. C. Franson, D. L. Graham, C. H. Domermuth, B. A. Rattner, and O. H. Pattee. 1983. Hemorrhagic enteritis in captive American kestrels (Falco sparverius). J. Wildl. Dis. 19:244-247.