Molecular Diagnosis of Beak and Feather Disease in Native Brazilian Psittacines

I Laboratory of Avian Diseases, Department of Preventive Veterinary Medicine, School of Veterinary Medicine, Federal University of Minas Gerais, Brazil II Forest Institute of the state of Minas Gerais (Instituto Estadual de Florestas), MG, Brazil. III Brazilian Institute for Environment and Renewable Natural Resources (Instituto Brasileiro de Meio Ambiente e Recursos Naturais Renováveis), IBAMA, Belo Horizonte, MG, Brazil.


INTRODUCTION
The highest global incidence of psittacine beak and feather disease (PBFD) is recorded in Australia (McOrist et al., 1984;Pass and Perry, 1984;Raidal et al., 1993), where at least one species is threatened of extinction.Infection by the PBFD virus (BFDV) is disseminated in more than 40 species of companion psittacines in Asia, Europe, Africa, and North and South America as a result of the international avian trade (Borthwick et al., 2005;Rahaus et al., 2003).However, species of the Cacatuidae family appear to be more susceptible (Bassami et al., 2001).BFDV is a DNA virus of Circoviridae family with tropism for epithelial cells of several tissues, including skin and gastrointestinal mucosa (Latimer et al., 1991).An acute form occurs in about 90% of wild cockatoos and results in hepatic necrosis, leucopenia, and greenish diarrhea in nestlings (Raidal and Cross, 1995).The chronic form results in progressive feather loss (alopecia), beak deformation, and predisposition to opportunistic infections (Pass and Perry, 1984).All psittacines may be susceptible, despite the morbidity differences possibly associated with host species and virus strains, and in Australia, PBFD is considered partially responsible for the successful predation of affected birds (Shearer et al., 2008).
Transmission occurs by inhalation or ingestion, including cloacal drinking, of the infective virus exported from sloughed feather follicles, feces with bile, and mucus secretion from crop, with high concentrations of virus detected in such tissues, and will appear in skin and feather dander and feces (Ritchie, 1995;Ritchie et al., 1990).Vertical transmission is suspected (Ritchie, 1995), similarly to chicken anemia virus and porcine circovirus 2. Diagnosis may be based on clinical signs (alopecia), history, histopathology (feather follicle biopsy) showing intranuclear inclusion bodies, and BFDV DNA detection in feathers, feces, or other tissues, such as blood, feather follicle, liver, and spleen, by PCR (Ypelaar et al., 1999).The liver is considered a site of virus replication (Raidal et al., 1993).Differential diagnosis should consider other causes of alopecia, such as polyomavirus, nutritional diseases, bacterial and fungal diseases, and psychogenic feather picking (Ritchie, 1995).In Brazil, the diagnosis of PBFD was first reported in 1997, in a Cacatua alba with alopecia (Werther et al., 1998).However, BFDV incidence and its clinical impact on infected Brazilian native avian species are yet unknown.This paper describes an investigation on the incidence of BFDV DNA in native Brazilian psittacines housed in triage centers and examines its possible association with clinical disease.

Sampling
Cloacal swabs, blood, and/or liver samples were collected between January 2009 and October 2010 at triage centers (Table 1).Cloacal swab (n=128) and blood (n=46) samples were collected upon bird arrival at the triage centers of Belo Horizonte and Uberaba, Minas Gerais, Brazil, avoiding contact with resident birds.Cloacal swabs were collected and immediately placed in microtubes with PBS (0.15M) and stored at -20ºC.Blood samples (0.5 mL) were collected by brachial vein puncture after physical restriction, using sterile disposable syringe and needle, and allowed to clot.The serum was separated and both serum and clot were frozen (-20ºC) until DNA extraction.Liver samples were collected from psittacines dead on arrival or which died at the triage center during the period of study (n=167).Both cloacal swab and liver samples were collected from 19 A. aestiva individuals.The clinical evaluation of 16 birds arriving at triage center was performed using a protocol based on literature (Rupley, 1999).

DNA extraction
Liver DNA extraction was performed by sodium iodide (NaI) protein denaturation and silica DNA adsorption (Boom et al., 1990;Vogelstein and Gillespie, 1979).Two hundred microliters of macerated liver tissue or blood clot were added to three volumes (600 µL) of saturated (6M) NaI (Synth, Diadema, SP, 09990-080, Brazil) and incubated at 55 °C for 15 min.After centrifugation (4,000 g/15 min), the supernatant was recovered and 50 µL of silicon dioxide (Sigma, Chemical Co., St, Louis, MO, USA) packed suspension were added, rapidly mixed in a vortex, and incubated in an end-over-end mixer for 10 min at room temperature.The mixture was centrifuged (14,000 g/30 s), the supernatant discarded by inversion, the silica pellet suspended in 1 mL NaI, vortexed, and centrifuged (14,000 g/30 s).The pellet was washed three times with ethanol wash buffer (ethanol 50%, 50mM Tris-HCl pH8,0, 10mM EDTA at pH 8.0), the ethanol supernatant was removed with pipette, 1 mL acetone was added to the pellet, and then vortexed and centrifuged (14,000 g/30 s).The pellet acetone residue was allowed to evaporate (55 ºC/10 min) and the DNA adsorbed into silica eluted with 50 µL of TE (5 mM Tris-HCl at pH 8.0, 0.5 mM EDTA at pH 8.0), incubated at 55 °C/10 min, and centrifuged (14,000 g/30 s).The supernatant DNA was quantified, its purity verified, and then it was stored at -20 ºC until use.Cloacal swab DNA was similarly extracted, except that 600µL NaI was added instead of 1 mL.Microtubes were vortexed, incubated at 55 ºC/15 min, centrifuged (14,000 g/4 min), the swabs were removed and the silicon dioxide was added to supernatants, which were then treated as described above.Blood DNA extraction procedure was similar to that applied to liver DNA, except that 600 μL NaI were added to 50 μL unclotted samples.DNA concentration and purity were estimated in 1 μL of each extraction sample by spectrophotometry (NanoDrop ND-1000, Thermo Fisher Scientific, Wilmington, DE, USA) and frozen at -20 ºC until tested.

Sequencing
Sequencing was performed by the dideoxynucleotide method (Sanger et al., 1997), in an automated capillary sequencer (ABI 310, Applied Biosystems, Bedford, MA, 01730, USA), after PCR reaction with Big Dye Terminator Mix and Save Money buffer (Applied Biosystems, USA), according to the manufacturer's instructions, using 1µL of the PCR product in a thermocycler (PTC-100, MJ Research Inc., USA).Labelled products were purified by precipitation with isopropanol and ethanol, homogenized in formamide, rapidly denatured at 95 o C/2 min, and the microtubes placed in flaked ice.The sequencing software employed was Sequencing Analyses version 5.2 (Applied Biosystems, USA).

Analyses of sequences
In order to perform electropherogram analysis of both sense and anti-sense sequences, the Bioedit software (Tamura et al., 2011) was employed to evaluate the quality of bases (Hall, 1999).Database sequences were obtained at the National Center for Biotechnology Information (NCBI -http://www.ncbi.nlm.nih.gov) and aligned with Genbank sequences using ClustalX (Thompson et al., 1997).The algorithms for nucleotide similarity or deduced amino acids were evaluated, respectively, using BLAST 2.0 (Basic Local Alignment Search Tool, BLASTn) or BLASTx (http:// www.ncbi.nlm.nih.gov/BLAST/)Altschul et al., 1997.The software ClustalW version 1.6 was utilized in the Molecular Evolutionary Genetics Analysis (MEGA 5.0/ www.megasoftware.net)version 5.0 for Windows for nucleotide alignment or deducing amino acid sequences.
The phylogenetic analyses of VP1 encoding nucleotide sequences or deduced amino acids were performed by the neighbor-joining method, using MEGA 5.0 software, with topology confidence (bootstrap) of 1,000 oversampling, with the nucleotide substitution by Kimura 2 method (Kimura, 1980) or the amino acid JTT substitution model (Jones et al., 1992).

Histopathology
A few dead animals (n=38) were fully necropsied and liver samples were collected for PCR and histopathology.At necropsy, heart, kidney, liver, and spleen fragments were collected in buffered formaline at 10% for histopathology (H&E).

RESULTS AND DISCUSSION
A 717 base pair product was obtained from liver or blood samples (Fig. 1), which was compatible with the fragment described in literature (Ypelaar et al., 1999).
Among birds with alopecia (n=16), only one was positive for BFDV in the blood and all tested cloacal swabs were negative (Table 2).The clinical evaluation of the 16 symptomatic psittacines with abnormal feathering is shown in Table 3.The clinical signs of feather abnormalities, despite being suggestive of PBFD (Rupley, 1999), were not accompanied by beak deformities.The successful confirmation of BFDV by PCR was obtained in only the blood sample of Ara ararauna individual, which cloacal swab, however, was tested negative.BFDV infection of New World psittacines may result in clinical disease (Phalen, 2006;Piçarra, 2009), reducing their chances of survival in the wild, as previously documented in New Zealand psittacines (New Zealand, 2012).However, efficient immune response may result in a latent subclinical infection, which may become clinically apparent only under stressful conditions (Dahlhausen & Radabaugh, 1997;Khalesi et al., 2005;Phalen, 2006;Piçarra, 2009).

Molecular Diagnosis of Beak and Feather Disease in Native Brazilian Psittacines
Out of the 46 blood samples tested, only one collected from an A. ararauna was positive.Nineteen birds that presented negative cloacal swab samples were retested, revealing 21% (4/19) positive liver samples, which suggests higher sensitivity of liver sampling.
Only Amazona aestiva and Ara ararauna individuals were positive for BFDV (Fig. 1).Previous descriptions in Amazona aestiva (Huff et al., 1988) and Ara ararauna (Piçarra, 2009) were documented.New World psittacines (Amazona sp., Ara sp.) present lower incidence of the disease compared with Old World species (Eclectus sp., Agapornis sp.) (Shearer et al., 2008;Piçarra, 2009).Although BFDV was detected in Amazona aestiva individuals in the present study, the birds were asymptomatic, in agreement with a previous report (Piçarra, 2009).BFDV DNA was detected in a symptomatic Blue-and-Yellow macaw (Table 2).Although there are few reports available on the incidence of BFDV in Brazil, the virus has been described in other continents (Pass & Perry, 1984;Shearer et al., 2008).In our study, swab and blood samples were collected at arrival of the birds at the triage centers, avoiding contact with resident birds.Out of the 15 birds symptomatic showing alopecia, only individual (Ara ararauna) was positive for BFDV by PCR.However, there may have been false negative BFDV results in truly positive, but nonviremic, birds.The negative results obtained in cloacal swabs may have been due to the presence of inhibitors in the DNA extract (Das et al., 2009).Testing multiple tissue samples in individual is a strategy for diagnosis (Khalesi et al., 2005;Piçarra, 2009).
The regular diagnosis and monitoring associated to biosecurity have enabled the reduction of prevalence the disease in the USA (Bert et al., 2005), and may possibly reduce individual losses by PBFD also in Brazilian psittacines in captivity, triage centers and or in the wild.Quarantine should be employed when purchasing birds, and all birds, including those clinically normal, should be tested (Olsen & Speer, 2009).
The sequence (strain BH-27) extracted from the blood of the rose-ringed parakeet (positive control) resulted in a complete match with BFDV VP1 (ORF-1) sequences stored in the GenBank.Sequencing of purified PCR products of BH-215 and BH-732 strains, both extracted from Amazona aestiva liver, allowed BLAST (Altschul et al., 1997) search for matches in the NCBI, which showed similarity only with BFDV published sequences.
The two local BFDV sequences of isolates BH-215 (Accession number JQ649409) and BH-732 (Accession number JQ649410), both from Amazona aestiva, and isolate BH-27 from Psittacula krameri (Accession number JQ649411) were evaluated.The analysis of the nucleotide sequences of BH-215 and BH-732 strains comparing the sequences available in the GenBank, showed identity only to BFDV.
The alignment (ClustalW in MEGA 5.0) of strain BH-215 with 11 published BFDV sequences revealed 19 nucleotide substitutions.However, a 98% homology (677/696 bp) was observed with the Cacatua galerita Australian strain AU-SCC1-WA-2000 (AF311302.1)(Massaro et al., 2012), with 1181 bits score and no base insertion nor deletion.The analysis the 232 deduced amino acids sequences of strain BH-215 revealed 99% similarity with the strain BKS3ZA-86, with one substitution of threonine for alanine.This substitution could result from genomic mutation or polymerization error, although high-grade polymerase was employed.

Molecular Diagnosis of Beak and Feather Disease in Native Brazilian Psittacines
acids sequence between residues 25 and 243.In this case, a similar property amino acid (polar, negative side chain) substitution, which apparently would not change protein properties, has enabled adaptation to mice lung for influenza virus hemagglutinin (Sakabe et al., 2011).Isolate BH-732 was also very similar (98%) to the Japanese strain MU-JP1P (AB277746.1).Considering that circoviruses affecting other animal species have very high mutation and recombination rates, it would be necessary to study the complete genomes to understand interspecies transmission (Massaro et al., 2012).
A phylogenetic tree based on the nucleotide sequences of strains BH-27, BH-215 and BH-732 was constructed (Fig. 2) in order to evaluate their potential geographical association, as compared to geographically distinct strain sequences available in the GenBank.No geographical marker was detected for local isolates; hence, none was considered unique, but similar to the strains described in Asia and Oceania, in contrast with the strains characterized in New Zealand (Massaro et al., 2012), which are unique.Isolate BH-215 was grouped with the Cacatua galerita strain AU-SCC1-WA-2000 and the Eolophus roseicapillus strain AU-Galah-WA-2000, both from Australia, whereas isolate BH-732 was grouped with the Japanese strain MU-JP1P and the New Zealand strain ZA-BKS3ZA_86-2008.The least related strains, as compared to BH-27, BH-215 and BH-732, were TW04-13 (Taiwan), BFDV-16 (Thailand), AFG4-ZA (South Africa), and Isolate 05-106 (Australia).The deduced amino acid tree topology (not shown) was in agreement to nucleotide sequences.
The Amazona aestiva individuals tested positive for BFDV DNA by PCR that died during this study were necropsied and examined by histopathology (n = 5).No histopathological findings could be directly associated to BFDV infection, including no intranuclear inclusion bodies.The liver of bird n. 15 presented multi-focal coalescent hepatocytes, lymphocytic proliferation, and blastic mytotic erythrocyte precursors.Bird n. 412 was a male, and presented cachexia, hepatomegaly, splenomegaly, and air sacculitis.Its liver histopathology (not shown) revealed plasmacytosis, macrophage hemosiderosis,.and a focal necrosis, which may also suggest Pacheco disease or Chlamydophila psittaci infection.However, no indication of intranuclear or intracytoplasmic inclusion bodies were found.The spleen presented eosinophilic areas with lymphoid depletion.The main gross lesions found in bird n. 732, a female, were cachexia, air sacculitis, hepatomegaly, splenomegaly, hemorrhagic gastroenteritis, and pale lungs.At histopathology, hemosiderosis and particularly periportal lymphocytosis and multifocal lymphoplasmacytic infiltrations associated with focal necrosis were found in the liver.The histology of bird n. 865, which presented hepatomegaly and air sacculitis, also showed hemosiderosis, with large hemosiderin deposits in the hepatocytes and splenocytes, liver multifocal lympho-plasmacytic infiltrations, proliferation of bile ducts, which indicates chronic hepatosis; however, no intranuclear or cytoplasmic inclusion bodies were found.Bird n. 885 had hemorrhagic gastroenteritis and splenomegaly, and presented moderate lymphocytic infiltration in the liver and a moderate population of spleen macrophages with hemosiderin, moderate lymphoid depletion and moderate lymphoid activation.In addition, megakaryocytes, which are platelet precursors, were visible, indicating extra-bone marrow erythropoiesis.The histopathological findings indicative of PBFD were previously described (Doneley, 2003), and included the demonstration of intranuclear inclusion bodies in bone marrow, hepatocytes, kidney and thymus (Pass & Perry, 1984).
No beak deformities were observed.Birds n. 412, 732, 865 and 885 did show hepatic lesions suggestive of virus infection, although not typical of BFDV and/ or other DNA virus, except for bird n. 412 (Pacheco disease).

CONCLUSIONS
The silica extraction was the most successful method for BFDV detection by PCR in blood and liver samples; however, this method may need to be modified when testing cloacal swabs.The DNA of BFDV was detected in psittacines arriving at Belo Horizonte triage center.However, BFDV was not detected in birds presenting alopecia, suggesting other causes of feather dystrophy.Among the clinically affected birds, BFDV DNA was detected in the blood of only one Ara ararauna .No beak deformities were observed in any of the birds examined.Out of the tested psittacines, BFDV was detected only in Amazona aestiva and Ara ararauna individuals, which, in general, presented normal feathers and had died due to other conditions.The higher detection rate of BFDV in A. aestiva could be associated with the higher number of individuals available for testing relative to the other species.However, this hypothesis cannot be applied to A. ararauna, which sample was very small, or to Amazona amazonica, which comprised the second largest sample and all individuals were tested negative for BFDV.Moreover, BFDV detection could be related to a higher susceptibility to infection.
Histopathology of livers did not allow the demonstration of BFDV intranuclear inclusions bodies.
The genomic similarity of the tested isolates with reference strains from Asia and Oceania indicates the exotic origin of the BFDV strains detected in the present study, suggesting their introduction in Brazil by the legal or illegal trade of psittacines from those continents.
The partial VP1 gene sequence of BH-215 and BH-732 strains are the first published for Brazilian isolates.No sequence was unique or exclusive of a geographical location.
Due to mostly asymptomatic occurrence of the BFDV in psittacines, the establishment of strict biosecurity measures, including quarantine, is strongly recommended in wild bird rescue facilities.

Table 1 -
Occurrence of psittacine beak and feather disease virus (BFDV) as detected by PCR a.

Table 2 -
Clinical evaluation, blood chemistry and molecular investigation of BFDV DNA by PCR in blood of Brazilian native psittacines with impaired feather coverage.
Right eye blepharitis; head, around the eyes, and dorsal alopecia; abnormal facial pigmentation, with red spots; exposed right wing feather shafts.aPCR polymerase chain reaction (blood) based on 717 bp ORF-1 sequence amplification. 47 BC: body condition; c Ht% hematocrit; d PP% plasmatic protein; e ND: not done.Araújo AV,