Occurrence of avian bornavirus infection in captive psittacines in various European countries and its association with proventricular dilatation disease

Abstract

A total of 1442 live birds and 73 dead birds out of 215 bird collections in Spain, Germany, Italy, the UK and Denmark were tested for avian bornavirus (ABV) infection by four different methods. The majority of the birds were psittacines belonging to 54 different genera of the order Psittaciformes. In total, 22.8% of the birds reacted positive for ABV in at least one of the tests. Combined testing of swabs from the crop and cloaca, and serum for the diagnosis of ABV infection in live birds revealed that virus shedding and antibody production coincided in only one-fifth of the positive birds so that the examination of these three samples is recommended for reliable ABV diagnosis. By statistical analysis of this large number of samples, the ABV infection proved to be highly significant (P <0.001) associated with histopathologically confirmed proventricular dilatation disease (PDD) in dead birds as well as with clinically assumed PDD in live birds. However, ABV infection was also detected in psittacines without pathological lesions or clinical signs of PDD. Twelve non-psittacine birds belonging to the genera Aburria, Ciconia, Geopelia, Leucopsar and Pavo were tested negative for ABV infection. Within the order of Psittaciformes, birds belonging to 33 different genera reacted positive for ABV. In 16 of these psittacine genera, the ABV infection was demonstrated for the first time. The present study emphasizes the widespread occurrence of clinically variable ABV infections in Europe by analysing a large number of specimens from a broad range of bird species in several assays.

Introduction

In 2008 two independent research groups in Israel and the USA identified a group of novel non-segmented, negative-sense, single-stranded RNA viruses by high-throughput viral screens, which were classified as four different genotypes of a new genus of the family Bornaviridae designated avian bornavirus (ABV) (Honkavuori et al., 2008; Kistler et al., 2008). The ABV RNA was detected in parrots that died from proventicular dilatation disease (PDD). After these first publications, further reports of ABV infections in PDD-affected psittacines originated from Germany (Enderlein et al., 2009; Lierz et al., 2009; Rinder et al., 2009), Austria, Switzerland, Hungary, Australia (Weissenböck et al., 2009a) and Canada (Raghav et al., 2010). Besides, a report on the detection of ABV genome in a canary (Serinus canaria) with enteric ganglioneuritis and encephalitis indicates that not only psittacine birds may be susceptible for ABV infection (Weissenböck et al., 2009b).

PDD is a progressive, often fatal, disease of captive psittacine birds worldwide. The disease is characterized by lymphohistioplasmocytic infiltration of the ganglia of the central and peripheral nerve system leading to gastrointestinal dysfunction and associated wasting as well as neurological signs. PDD is one of the most threatening diseases for parrots, including endangered psittacine species. It has been found in about 60 species belonging to 20 different psittacine genera (Lutz & Wilson, 1991; Cazayoux Vice, 1992; Gregory et al., 1994; Sullivan et al., 1997; Berhane et al., 2001; Lierz, 2005; Lublin et al., 2006, Lierz et al., 2010), but also in some non-psittacine bird species (Staeheli et al., 2010). Since its first description in the late 1970s several viruses (adenovirus, herpesvirus, polyomavirus, eastern and western equine encephalitis viruses, paramyxovirus types 1 and 3) (Mannl et al., 1987; Cazayoux Vice, 1992; Gregory, 1995; Sullivan et al., 1997; Gregory et al., 1998; Grund et al., 2002; Gough et al., 2006; Lublin et al., 2006; Orosz & Dahlhausen, 2007) have been proposed to be the aetiologic agent of the disease, but this was not proven for any of these viral agents. However, results of two first experimental infections of a small number of cockatiels (Nymphicus hollandicus) and Patagonian conures (Cyanoliseus patagonis) provided evidence for the aetiological role of ABV in the development of PDD (Gancz et al., 2009; Gray et al., 2010). Besides, ABV infections were also found in psittacines with no clinical signs of PDD (De Kloet & Dorrestein, 2009; Lierz et al., 2009; Villanueva et al., 2009).

Meanwhile, different diagnostic methods have been developed for the direct and indirect diagnosis of ABV. ABV RNA can be detected by reverse transcriptase-polymerase chain reaction (RT-PCR) using various primer sets. In PDD-affected dead psittacine birds ABV RNA was demonstrated not only in the nervous system and the gastrointestinal tract but in nearly all other tissues (Kistler et al., 2008; Honkavuori et al., 2008; Lierz et al., 2009; Rinder et al., 2009; Villanueva et al., 2009; Enderlein et al., 2009; Gray et al., 2010; Kistler et al., 2010; Raghav et al., 2010). Further methods for the direct detection of ABV in dead birds as well as in crop biopsies are immunohistochemistry (IHC) (Rinder et al., 2009; Ouyang et al., 2009; Weissenböck et al., 2009a; Herzog et al., 2010; Raghav et al., 2010), virus isolation in the quail cell lines CEC-32 and QM7 (Rinder et al., 2009, Herzog et al., 2010) or duck embryo fibroblasts (Gray et al., 2010), and in situ hybridization (Weissenböck et al., 2010). In live birds, ABV can be diagnosed directly by detection of ABV RNA using RT-PCR in faeces, swabs of crop and cloaca, and blood (Rinder et al., 2009; Lierz et al., 2009; Enderlein et al., 2009; Gray et al., 2010; Kistler et al., 2010). As serological methods for the demonstration of anti-ABV antibodies, a western blot assay (De Kloet & Dorrestein, 2009; Lierz et al., 2009; Villanueva et al., 2009), an enzyme-linked immunosorbent assay (De Kloet & Dorrestein, 2009) and an indirect immunofluorescence assay (Gray et al., 2010; Herzog et al., 2010) have been developed.

In the present study, analysis of the results of testing a high number of samples from psittacine and non-psittacine birds from Spain, Germany, Italy, the UK and Denmark for ABV should provide further information on the suitability of different kinds of samples and the existing diagnostic tools for ABV detection as well as on the occurrence of ABV in different species and countries and its association with PDD.

Materials and Methods

Birds and samples

In total, 1515 birds out of 215 different bird collections in Spain, Germany, Italy, the UK and Denmark were included in the study (Table 1). The birds or different samples of them had been submitted by veterinarians, bird owners and breeders as well as zoos for diagnostic purposes, in most cases without any or with a poor history concerning number and species of other birds in the flock. The examined birds belonged to 54 different genera of the order Psittaciformes and five genera of non-psittacine birds (Table 2). Seventy-three of these birds with or without clinical suspicion of PDD were dead birds (n =42) or the organs of dead birds that were submitted fresh (n =20) or fixed in formalin or embedded in paraffin (n =11). From 1442 live birds, sera (n =1349), crop (n =347) and cloacal (n =381) swabs were received as either single or combined samples per bird. Sixty-three of these live birds were suspected of PDD due to clinical signs such as undigested feed in the faeces, loss of weight, proventricular dilatation in radiography and/or central nervous signs. From the other live birds, no clinical signs were reported, and samples of them were taken to detect subclinical ABV infections.

Table 1.  Origin and number of tested birds.

Country	Number of bird collections	Number of birds	Number of ABV-positive birds^a
Spain	14	996	142
Germany	171	4 42	170
Italy	12	32	17
UK	16	16	5
Denmark	2	29	9
Total	215	1515	343
^aDetection of ABV (RNA, virus, antigen) and/or ABV-specific antibodies.

Table 2.  Bird genera examined for ABV infection.

		Present study			Previous reports on
Family^a	Genus^a	Birds (n)	ABV-positive birds^b (n)	% ABV-positive birds	PDD	ABV
Micropsittacidae	Agapornis	56	7	12.5	Gregory et al. (1994)
Psittacidae	Forpus	17	0	0.0
	Bolborhynchus	20	1	5.0
	Anodorhynchus	23	3	13.0	Woerpel & Rosskopf (1984)	Weissenböck et al. (2009a)
	Cyanopsitta	2	0	0.0	Helstab et al. (1985)
	Ara	222	53	23.9	Ridgway & Gallerstein (1983)	Kistler et al. (2008)
	Guaruba	7	1	14.3	Helstab et al. (1985)	Villanueva et al. (2009)
	Nandayus	6	3	50.0	Phalen (1986)	Villanueva et al. (2009)
	Rhynchopsitta	7	1	14.3	Graham (1991)
	Aratinga	134	6	4.5	Ridgway & Gallerstein (1983)	Kistler et al. (2008)
	Pyrrhura	167	15	9.0
	Enicognathus	5	0	0.0	Berhane et al. (2001)
	Cyanoliseus	10	3	30.0	Cazayoux Vice (1992)
	Brotogeris	52	10	19.2	Gregory et al. (1994)
	Pionopsitta	12	1	8.3
	Amazona	179	54	30.2	Graham (1984)	Honkavuori et al. (2008)
	Deroptyus	8	2	25.0	Gregory et al. (1994)	Weissenböck et al. (2009a)
	Pionus	25	6	24.0	Gregory et al. (1994)	Kistler et al. (2008)
	Triclaria	4	0	0.0
	Pionites	7	0	0.0	Reavill & Schmidt (2007)
	Poicephalus	33	18	54.5	Suedmeyer (1992)	Shivaprasad et al. (2009)
	Psittacus	101	47	46.5	Ridgway & Gallerstein (1983)	Kistler et al. (2008)
	Coracopsis	7	3	42.9	Gregory et al. (1994)	Weissenböck et al. (2009a)
Psittaculidae	Psittacula	30	3	10.0	Gregory et al. (1994)
	Psittrichas	2	1	50.0
	Eclectus	47	17	36.2	Graham (1984)	Weissenböck et al. (2009a)
	Psittinus	3	1	33.3
Polytelidae	Aprosmictus	3	0	0.0
	Alisterus	6	3	50.0
	Polytelis	3	1	33.3
Loriidae					Reavill & Schmidt (2007)
	Lorius	8	0	0.0
	Neopsittacus	1	0	0.0
	Oreopsittacus	2	0	0.0
	Trichoglossus	8	2	25.0
	Eos	3	2	66.7
	Chalcopsitta	2	1	50.0
	Psittaculirostris	3	0	0.0
	Lathamus	4	3	75.0
Platyceridae	Purpureicephalus	5	1	20.0
	Barnadius	5	0	0.0
	Psephotus	3	0	0.0
	Platycerus	10	0	0.0
	Eunymphicus	3	0	0.0
	Cyanoramphus	1	0	0.0		Weissenböck et al. (2009a)
	Neophema	14	0	0.0
	Neopsephotus	2	0	0.0
Strigopidae	Strigops	18	0	0.0
Cacatuidae	Probosciger	6	2	33.3		Kistler et al. (2008)
	Calyptorhynchus	2	0	0.0
	Callocephalon	4	0	0.0
	Eolophus	32	6	18.8	Lublin et al. (2006)	Kistler et al. (2008)
	Cacatua	140	56	40.0	Ridgway & Gallerstein (1983)	Kistler et al. (2008)
	Nymphicus	6	2	33.3	Graham (1984)	Ouyang et al. (2009)
Nestoridae	Nestor	4	0	0.0	Grund et al. (2002)
Unknown	Unknown	19	8	42.1
Psttaciformes	Sub-total	1503	343	22.8
Cracidae	Aburria	2	0	0.0
Ciconiidae	Ciconia	4	0	0.0
Columbidae	Geopelia	1	0	0.0
Sturnidae	Leucopsar	4	0	0.0
Phasianidae	Pavo	1	0	0.0
Non-Psittaciformes	Sub-total	12	0	0.0
Total		1515
^aAccording to Wolters (1982). ^bDetection of ABV (RNA, virus, antigen) and/or ABV-specific antibodies.

General examination

All dead birds were necropsied, and the brain, spinal cord, retina, Nervus ischiadicus, crop, proventriculus, gizzard, small and large intestine, heart, liver, kidney, pancreas, pectoral muscle, and skin were collected. These as well as the submitted fresh organ samples (same range of organs) were processed for histopathological evaluation (fixation in 10% buffered formalin, embedding in paraffin, preparation of 5 µm sections, staining with haematoxylin and eosin). Additionally, in dead birds parasitological, microbiological (blood, Gassner and Kimmig agar) and virological (three passages in primary chicken embryo fibroblast and liver cell cultures as well as in specific pathogen free egg cultures [Valo; Lohmann, Cuxhaven, Germany]) examinations were performed for the detection of pathogenic infectious agents by standard methods (Dufour-Zavalla et al., 2008). The presence of Chlamydia spp. was excluded by PCR (Sachse & Hotzel, 2003). All sera were tested for antibodies against paramyxovirus type 1 by the haemagglutination inhibition test (OIE, 2004).

Specific testing for avian bornavirus

Reverse transcriptase polymerase chain reaction. Organs and swabs were tested for the presence of ABV RNA by quantitative real-time RT-PCR using the TaqMan® Fast Universal PCR Master Mix (Applied Biosystems, Carlsbad, California, USA) with two different primer sets (Honkavuori et al., 2008). In the case of negative results in these PCRs but positive results in other tests, a further primer set for standard PCR was used to detect other strains of ABV (Enderlein et al., 2009).

Immunohistochemistry

The different organ samples of the dead psittacines were analysed immunohistochemically for the presence of ABV antigen by the avidin–biotin complex method using a polyclonal rabbit antibody directed against the phosphoprotein (p24) of the Borna disease virus (BDV) (Herden et al., 1999; Herzog et al., 2010).

Virus isolation

Organ samples were processed in infectivity assays according to the method described by Narayan et al. (1983). Briefly, 10-fold dilutions of the organ homogenates (10%, w/v) were prepared in GMEM medium (Gibco, Invitrogen, UK) plus 10% foetal bovine serum (FBS), mixed with equal volumes of freshly dispersed cells of the quail cell line CEC32 and incubated on chamber slides (Lab-Tek Products; Nunc, Roskilde, Denmark) for 6 days at 37°C. Virus replication was demonstrated by indirect immunofluorescence using polyclonal sera from experimentally BDV-infected rats cross-reacting with ABV (Herzog et al., 2010).

Immunofluorescence assay

The demonstration of anti-ABV serum antibodies (doubling dilutions of serum starting with a dilution of 1:10) were performed by an indirect immunofluorescence assay on persistently BDV-infected Madin-Darby canine kidney or ABV-infected CEC32 cells using a fluorescein isothiocyanate-conjugated goat anti-bird IgG (Bethyl Laboratories, Montgomery, Texas, USA) in a dilution of 1:50 according to Herzog et al. (2010)

Statistical analysis

The chi-square test was applied for analysis of differences between bird groups. With one degree of freedom, calculated values ≥10.83 were considered highly significant (P <0.001) (Dixon, 1993).

Results

Post-mortem diagnosis of PDD and ABV

In 37 out of 73 examined dead psittacines PDD was confirmed by histopathological detection of typical lesions of non-suppurative ganglioneuritis in various parts of the gastrointestinal tract, often in combination with non-suppurative encephalitis, neuritis, myelitis and myocarditis. The affected birds belonged to the bird genera Ara (n=10), Psittacus (n=7), Cacatua (n=6), Amazona (n=5), Eclectus (n=2), Poicephalus (n=2), Guaruba (n=1), Nandayus (n=1), Pionus (n=1), Barnardius (n=1), and Nymphicus (n=1). All of these 37 cases were positive for ABV in at least one test (Table 3). In 26 cases, ABV RNA and antigen were detected by PCR and IHC, respectively. The ABV antigen was detected in the brain, spinal cord, retina, proventiculus, gizzard, and intestine by the anti-phosphoprotein antibody. Additionally, ABV was isolated from the brain, spinal cord and retina of 16 of these positive birds. Because of too small a sample size or the condition of the tissues (start of autolysis or formalin fixation/paraffin embedding), a further 11 birds were only tested by RT-PCR or IHC, and were found to be positive for either ABV RNA (n=5) or antigen (n=6). Parasites, fungi, bacteria including chlamydia, and viruses other than ABV were not detected in any of the 37 birds.

Table 3.  Association of PDD status and ABV detection in dead birds.

	ABV
PDD	Positive^a	Negative	Total number of birds	% ABV-positive birds
Positive^b	37	0	37	100
Negative	7	29	36	19
Total	44	29	73
Chi-square value: 49.45^c
^aABV-positive detection of ABV (RNA, virus, antigen) and/or ABV-specific antibodies. ^bPDD-positive pathohistological lesions typical for PDD. ^cHighly significant (P <0.001).

The other 36 out of the 73 examined dead birds revealed no histopathological lesions typical for PDD (Table 3). In 29 of these birds neither ABV RNA nor antigen was detected by RT-PCR and immunohistological staining. Virus isolation attempts were performed in 15 of these cases with negative results. However, seven (19%) of the birds without PDD lesions reacted positive for ABV in RT-PCR (n=4), IHC (n=1) or both tests (n=2).

The calculated value in the chi-square test of 49.45 indicates highly significant differences between the groups and a highly significant association of PDD with the ABV infection (Table 3).

Intra-vitam diagnosis of ABV

From 1442 live birds, 2077 samples composed of crop and cloacal swabs and sera were received for direct and indirect detection of ABV. In RT-PCR, 86 of 347 crop swabs and 99 of 381 cloacal swabs were found to be positive for ABV RNA. Of the 1349 sera tested, 228 revealed anti-ABV antibodies in the indirect immunofluorescence assay, while all sera were negative for antibodies against paramyxovirus type 1 in the haemagglutination inhibition test. Table 4 demonstrates the results of testing for ABV RNA and/or antibodies with respect to the number of birds of which different kinds and combinations of samples were examined. In total, 299 (20.7%) of the live birds reacted positive for ABV. The percentage of positive birds was even higher in those cases where several kinds of samples were investigated. As presented in Table 4, the ABV infection rate proved to be markedly higher (121/276 birds, 43.8%), when crop and cloacal swab and serum from a single bird were tested in combination. While only about one-fifth (25/121) of these birds showed positive reactions in all three samples submitted, the samples of the other cases were only partly positive for ABV: about one-third (n=45) reacted only indirectly positive for ABV by exhibiting specific antibodies, another third (n=35) was only positive for ABV RNA in both swabs and 16/121 reactors proved to be positive in only one swab in combination with the serum or even alone.

Table 4.  Detection of ABV RNA and/or antibodies in swabs and sera of live birds.

		Number of positive birds with
Sample(s)	Number of birds (samples)	Sample(s) +	Swab(s)–/S +	Swab(s) + /S-	Only CrS +	CrS + /S +	Only ClS +	ClS + /S +	Total number of positive birds	% positive birds
CrS	1 (1)	0	n.a.	n.a.	n.a.	n.a.	n.a.	n.a.	0	0
ClS	24 (24)	7	n.a.	n.a.	n.a.	n.a.	n.a.	n.a.	7	29.2
CLS, CrS	68 (136)	16	n.a.	n.a.	n.a.	n.a.	n.a.	n.a.	17	23.9
ClS, CrS, S	276 (828)	25	45	35	5	3	3	5	121	43.8
ClS, S	13 (26)	4	1	3	n.a.	n.a.	n.a.	n.a.	8	61.5
CrS, S	2 (4)	1	0	1	n.a.	n.a.	n.a.	n.a.	2	100.0
S	1058 (1058)	144	n.a.	n.a.	n.a.	n.a.	n.a.	n.a.	144	13.6
Total	1442 (2077)	197	46	39	5	3	4	5	299	20.7
+, positive; –, negative; n.a., not applicable; ClS, cloacal swab; CrS, crop swab; S, serum.

Table 5 shows the relationship of PDD status with the detection of ABV in the live birds tested. Birds with clinical signs of PDD (n=63) were positive for ABV to 66.6%. From 1379 birds without clinical signs of PDD, 81.4% were found to be negative for ABV, but in 257 birds (18.6%) ABV was detected directly or indirectly. The chi-square value of 84.56 indicated that the differences between the groups were highly significant, and hence that ABV infection was associated with PDD at a high significance level.

Table 5.  Association of PDD status and ABV detection in live birds.

	ABV
PDD	Positive^a	Negative	Total number of birds	% ABV-positive birds
Positive?^b	42	21	63	67
Negative?^c	257	1122	1379	19
Total	299	1143	1442
Chi-square value: 84.56^d
^aABV-positive, detection of ABV (RNA, virus, antigen) and/or ABV-specific antibodies. ^bPositive?, tentative diagnosis of PDD: clinical signs like indigested feed, loss of weight, PDD in radiography, central nervous system disorder. ^cNegative?, no report on clinical signs of PDD. ^dHighly significant (P <0.001).

Occurrence of ABV in countries and bird genera

ABV was detected in birds from all five European countries from which birds or samples had been submitted for testing (Table 1). The positive birds belonged to 33 of 54 different genera of the order Psittaciformes of which birds were examined (Table 2), and in 16 of them ABV infection was found for the first time. In total, 22.8% of all the psittacines reacted positive for ABV. Considering those genera with more than 30 birds tested, substantial differences in the percentages of ABV-infected psittacines were seen. While the groups of Ara, Brotogeris and Eolophus contained about 20% positive reactors, the psittacines of the genera Aratinga (4.5%), Pyrrhura (9.0%), Psittacula (10%) and Agapornis (12.5%) revealed a notably lower infection rate. In contrast to this, higher percentages of ABV-infected psittacines were observed in the genera Poicephalus (54.5%), Psittacus (46.5%), Cacatua (40.0%), Eclectus (36.2%) and Amazona (30.2%). The 12 non-psittacine birds included in the present study were negative for ABV.

Discussion

In the present study, the results of testing a large number of different psittacine and some other bird species from various European countries on the occurrence of ABV infection were analysed and correlated with the available data on the PDD status in the tested birds. Additionally, the suitability of different specimens and test methods for ABV diagnostics were compared.

In other studies for post-mortem diagnosis of ABV infection (Rinder et al., 2009; Gray et al., 2010, Raghav et al., 2010), virus isolation, RT-PCR and/or IHC were performed only with freshly dead or euthanized birds or frozen organ samples. In this study, however, birds or samples of birds had been submitted for ABV diagnosis with different grades of freshness and different organ fixation. For this reason, organ samples of only 31 cases were suitable for virus isolation attempts without negative influence of toxic reactions and bacterial contamination in cell culture additional to the use of RT-PCR and IHC. Organs of a further 31 birds could be examined in RT-PCR and IHC, although autolysis in a few cases hampered the IHC analysis due to high unspecific background staining. Tissue samples of 11 birds were already fixed in formalin or were paraffin embedded and could be only processed immunohistochemically. These examinations indicated that RT-PCR and IHC are suitable test methods also for diagnostic material of different quality. To ensure optimal examination results of dead birds that have to be sent over long distances to the diagnostic laboratory it is proposed that organ samples should be transported in an RNA stabilization reagent or in a formalin-fixed status.

By intra-vitam diagnostics for ABV about 20% of the nearly 1500 birds were found to be positive for ABV RNA or displayed ABV-specific antibodies. The detection of more than 40% positive reactors in the group of birds with combined testing of three different samples, however, gave indication of a higher occurrence of ABV infection in the population of captive psittacines. As in this group only one of five birds proved to be positive in all three samples, testing of only one kind of sample may imply a high risk of false negative results. The observation that the presence of ABV RNA in cloacal excretions and anti-ABV antibodies in serum did not always coincide has already been made by Villanueva et al. (2009) in a smaller number of birds. Possible explanations for seropositivity without detection of ABV RNA might be that: the currently used RT-PCRs are not able to detect all ABV genotypes (Enderlein et al., 2009); the virus is only intermittently present in the crop and/or cloaca, which has been demonstrated for urofaeces of five ABV-infected birds by daily testing over a period of 5 days (Raghav et al., 2010); or the virus was eliminated by the infected bird. ABV RNA detection in swabs of a bird without any humoral immune reaction might be due to sampling at an early stage of infection or to an ABV infection that is able to hide from the immune system. Regarding these results, we strongly recommend the combined molecular biological and serological testing of cloacal and crop swabs (eventually pooled) as well as serum for the diagnosis of ABV infection.

One further purpose of the present study was to relate the results on the occurrence of ABV in dead and live birds to their known status concerning signs of PDD. In post-mortem examinations, other working groups (Honkavuori et al., 2008; Kistler et al., 2008; Lierz et al., 2009; Ouyang et al., 2009; Rinder et al. 2009, Weissenböck et al., 2009a; Raghav et al., 2010) detected ABV infection in a high percentage (up to 100%) of the PDD-affected birds by different test methods, but without including non-affected birds. Due to the large number of birds with and without PDD in our examinations it was now possible for the first time by statistical analysis to prove this association as highly significant in naturally infected birds. However, we also found ABV RNA or antigen in psittacines without typical lesions of PDD. Until now, this has only been observed in single birds by Lierz et al. (2009) and Raghav et al. (2010), and may be an indication for inapparent infection with a low-pathogenic ABV strain or an early stage of infection, when PDD lesions have not yet been induced. Nevertheless, these results strongly underline the causative role of ABV for the development of PDD, which has also been shown in first successful experimental ABV infections (Gancz et al., 2009; Gray et al., 2010).

By calculation in the chi-square test, the relationship of PDD to ABV infection was also clearly demonstrated as highly significant in live psittacines. The lower percentage (67%) of ABV-positive birds within the group of PDD-suspected birds in comparison with dead birds with confirmed PDD reflects the current difficulties in intra-vitam diagnosis of PDD. The clinical signs of PDD like loss of weight, undigested grains in faeces, and dilated proventriculus sometimes in combination with central nervous signs can also be seen in other diseases of the gastrointestinal tract and/or the nervous system (Villanueva et al., 2009; Lierz et al., 2010). Thus, it seems probable that not all of the birds in this study with clinically suspected PDD really suffered from this disease. However, because of the now proven association of ABV infection with PDD, in all likelihood a psittacine showing PDD-like signs and being ABV-positive is affected by PDD and not another disease.

Although most of the clinically healthy birds were negative for ABV, in nearly 20% of them ABV was detected directly and/or indirectly. These results further support the findings of De Kloet & Dorrestein (2009), Lierz et al. (2009) and Villanueva et al. (2009), who tested a limited number of apparently healthy birds in a few aviaries, and found some of them to be ABV-positive. Whether these birds were still in the incubation phase of the disease or remained only virus carriers without developing manifestation of clinical disease, as is known for mammalian Borna disease (Herden & Richt, 2009), can only be speculated on with the present state of knowledge on ABV pathogenesis. It might be that similar immunopathological mechanisms as those that induce outbreaks of Borna disease in mammals (Richt et al., 2007) may play an important role in the course of ABV infections. Rossi et al. (2008) suggest an autoimmune mechanism is involved in the development of clinical signs, as they found anti-ganglioside antibodies in PDD-affected birds and hypothesize that ABV might just serve as trigger for the development of the disease.

Concerning the occurrence of ABV in different countries, ABV detections in captive psittacines have been reported from Israel, the USA, Australia and Canada (Kistler et al., 2008; Honkavuori et al., 2008; Weissenböck et al., 2009a; Raghav et al., 2010) as well as from various European countries like Germany (Enderlein et al., 2009; Lierz et al., 2009; Rinder et al. 2009), Austria, Switzerland and Hungary (Weissenböck et al., 2009a). In this study, psittacines from different bird collections in Spain, Italy, the UK and Denmark were found to be positive for ABV, and the presence of ABV was further confirmed in German psittacines. Based on these findings it is likely that ABV is distributed worldwide in captive psittacine collections, which may be due to the intense exchange and trade with these birds.

In tota l, about 20% of the 1503 psittacines tested revealed signs of previous or ongoing ABV infection. These positive birds belong to 33 different psittacine genera (Table 2) and can be grouped as follows: 15 genera that are known to be affected by PDD and in which ABV infection has been described before, partly revealing remarkably high infection rates (e.g. Ara, Amazona, Poicephalus, Psittacus, Eclectus, Cacatua) in the present study; five genera that are known to be susceptible to PDD and in which ABV has now been detected for the first time, some of them with only low percentages of positive birds (e.g. Agapornis, Anodorhynchos, Psittacula); one genus in which ABV has been reported before, but PDD is so far unknown; and a further 12 genera in which neither PDD nor ABV infection has been reported previously (e.g. Bolborhynchus, Pyrrhura, Pionopsitta), but have now been shown to be ABV-positive. Further systematic epidemiological investigations and experimental infection trials are necessary to investigate whether the observed differences are only due to the non-representative mixture of psittacine birds examined or are caused by differences of psittacine genera in susceptibility to ABV infections.

Acknowledgements

The present research was supported by the Loro Parque Fundacion, Tenerife, Spain (Project No. PP-65-2009-1), and the Association for the Conservation of Threatened Parrots, Schöneiche, Germany. The authors are grateful to Jürgen Richt for kindly providing the anti-phosphoprotein antibody.