Parrot Bornavirus (PaBV)-2 isolate causes different disease patterns in cockatiels than PaBV-4

Transcription

1 Avian Pathology ISSN: (Print) (Online) Journal homepage: Parrot Bornavirus (PaBV)-2 isolate causes different disease patterns in cockatiels than PaBV-4 Anne K Piepenbring, Dirk Enderlein, Sibylle Herzog, Basim Al-Ibadi, Ursula Heffels-Redmann, Julia Heckmann, Hildburg Lange-Herbst, Christiane Herden & Michael Lierz To cite this article: Anne K Piepenbring, Dirk Enderlein, Sibylle Herzog, Basim Al-Ibadi, Ursula Heffels-Redmann, Julia Heckmann, Hildburg Lange-Herbst, Christiane Herden & Michael Lierz (2016) Parrot Bornavirus (PaBV)-2 isolate causes different disease patterns in cockatiels than PaBV-4, Avian Pathology, 45:2, , DOI: / To link to this article: Published online: 21 Apr Submit your article to this journal Article views: 449 View Crossmark data Citing articles: 12 View citing articles Full Terms & Conditions of access and use can be found at

2 AVIAN PATHOLOGY, 2016 VOL. 45, NO. 2, ORIGINAL ARTICLE Parrot Bornavirus (PaBV)-2 isolate causes different disease patterns in cockatiels than PaBV-4 Anne K Piepenbring a,b, Dirk Enderlein a, Sibylle Herzog c, Basim Al-Ibadi d, Ursula Heffels-Redmann a, Julia Heckmann a, Hildburg Lange-Herbst c, Christiane Herden d and Michael Lierz a a Clinic for Birds, Reptiles, Amphibians and Fish, Justus-Liebig-Universität Giessen, Giessen, Germany; b Tierarztpraxis Dr. E. Kellerwessel, Cologne, Germany; c Institute of Virology, Justus-Liebig-Universität Giessen, Giessen, Germany; d Institute for Veterinary Pathology, Justus- Liebig-Universität Giessen, Giessen, Germany ABSTRACT Psittaciform 1 bornavirus (PaBV) has already been shown to be the aetiologic agent of proventricular dilatation disease, a significant disease of birds. However, the pathogenesis of PaBV infection has not yet been resolved and valid data regarding the pathogenicity of different PaBV species are lacking. Thus, the present study was aimed to characterize the influence of two different PaBV species on the course of disease. Eighteen cockatiels were inoculated intracerebrally (i.c.) or intravenously (i.v.) with a PaBV-2 isolate under the same conditions as in a previous study using PaBV-4. Birds were surveyed and sampled for 33 weeks to analyse the course of infection and disease in comparison to that of PaBV-4. Similar to PaBV-4, PaBV-2 induced a persistent infection with seroconversion (from day 6 p.i. onwards) and shedding of viral RNA (from day 27 p.i. onwards). However, in contrast to PaBV-4, more birds displayed clinical signs and disease progression was more severe. After PaBV-2 infection, 12 birds exhibited clinical signs and 10 birds revealed a dilated proventriculus in necropsy. After PaBV-4 infection only four birds revealed clinical signs and seven birds showed a dilatation of the proventriculus. Clinically, different courses of disease were observed after PaBV-2 infection, mainly affecting the gastrointestinal tract. This had not been detected after PaBV-4 infection where more neurological signs were noted. The results provide evidence for different disease patterns according to different PaBV species, allowing the comparison between the infection with two PaBV species, and thus underlining the role of viral and individual host factors for disease outcome. ARTICLE HISTORY Received 12 November 2015 Accepted 15 December 2015 KEYWORDS Psittaciform 1 bornavirus; viral species; proventricular dilatation disease; pathogenesis; cockatiels; experimental infection Introduction Psittaciform 1 bornavirus (PaBV) is a member of the family Bornaviridae and was discovered in 2008 by two independent research groups in psittacine birds suffering from proventricular dilatation disease (PDD) (Honkavuori et al., 2008; Kistler et al., 2008). Up to now, seven different species of bornaviruses have been detected in 33 different genera of the order Psittaciformes (PaBV, PaBV-5 and PaBV-6) (Heffels- Redmann et al., 2011; Kuhn et al. 2015). Further bornaviruses were detected in other bird orders like Passeriformes (Passerine bornavirus) (Weissenböck et al., 2009; Rubbenstroth et al., 2014), Anseriformes (Waterbird 1 bornavirus, ABV MALL) (Delnatte et al., 2011) and, most recently, Charadriiformes (Guo et al., 2015). Clinical disease associated with PaBV infection in psittacine birds is extremely variable. Although clinical signs are predominantly associated with the gastrointestinal and nervous system, birds may be subclinically affected or show non-specific signs. The rate of disease development is also extremely variable, with both slow clinical progression and sudden death in apparently normal birds reported (De Kloet & Dorrestein, 2009; Gancz et al., 2009; Lierz et al., 2009; Kistler et al., 2010; Villanueva et al., 2010; Heffels-Redmann et al., 2011; Heffels-Redmann et al., 2012; Piepenbring et al., 2012). Recent studies have confirmed the aetiologic role of PaBV in the pathogenesis of PDD which is a fatal disease of psittacines and other bird orders worldwide (Daoust et al., 1991; Perpinan et al., 2007; Weissenböck et al., 2009) affecting especially large parrots (Heffels-Redmann et al., 2011). These studies were able to fulfil Henle- Koch s postulates (Gancz et al., 2009; Gray et al., 2010; Mirhosseini et al., 2012; Piepenbring et al., 2012). Gancz et al. (2009) inoculated three cockatiels with a PaBV-4-containing brain homogenate of an African grey parrot (Psittacus erithacus erithacus) via multiple routes and induced PaBV infection in the birds that reproduced PDD signs. After death, all birds contained PaBV-4 RNA. However, the brain homogenate also contained sequences similar to astro- and retroviruses, whereas these sequences were not identified in the tissues from the inoculated birds that developed clinical signs. (Gancz et al., 2009). Gray et al. (2010) inoculated two Patagonian conures CONTACT Michael Lierz 2016 Houghton Trust Ltd michael.lierz@vetmed.uni-giessen.de

3 AVIAN PATHOLOGY 157 (Cyanoliseus patagonis) intramuscularly with a PaBV-4 isolate from a yellow-collared macaw (Primolius auricollis) which was grown in duck embryo fibroblasts. Anti-PaBV antibodies were first detected in the inoculated birds on day 33 p.i. and PaBV-4 RNA was first amplified from one bird on day 60 p.i. PDD signs were observed on day 66 p.i. On post-mortem examination (66 d.p.i.), the birds revealed gross and histologic lesions consistent with PDD. However, the conures were chronic carriers of psittacine herpes virus (Gray et al., 2010). Mirhosseini et al. (2011) inoculated two cockatiels simultaneously via the intramuscular and oral routes with a PaBV-2 strain from a cockatiel and were able to reproduce PDD signs as well as typical gross and histopathological lesions. The birds did not shed viral RNA at any time during the trial, but contained PaBV-2 RNA in organ samples post-mortem. The cockatiels were known to be simultaneous chronic carriers of PaBV-4. However, PaBV-4 was not shed through the experiment nor identified at necropsy (Mirhosseini et al., 2011). Piepenbring et al.(2012) conducted a study employing 18 cockatiels which were inoculated via the intracerebral (i.c.) (n = 9) or intravenous (i.v.) route (n = 9) using a PaBV-4 isolate from a scarlet macaw (Ara macao). A non-inoculate was placed as a sentinel in the i.c. group. All inoculated birds became persistently infected, regardless of the route of infection. PaBV-4 RNA was first detected in crop and cloacal swabs on day 19 p.i. and anti-pabv antibodies were first detected on day 7 p.i. The sentinel bird tested positive for PaBV RNA in crop and cloacal swabs from day 76 p.i. on but did not seroconvert during the study period. Five of the 18 inoculated birds displayed signs characteristic of PDD, and the first clinical signs were observed on day 33 p.i. The remaining birds remained clinically healthy for the entire examination period. At postmortem examination, a dilated proventriculus was found in seven of the inoculated birds. PDD was confirmed by histopathology in all inoculated birds and PaBV-4 RNA and antigen were detected in all organs tested. PaBV-4 RNA was not detected in the organs of the sentinel bird which did not show any macroscopic alterations (Piepenbring et al., 2012). However, in field studies as well as in experimental infections, healthy PaBV carriers frequently occur (De Kloet & Dorrestein, 2009; Gancz et al., 2009; Lierz et al., 2009; Kistler et al., 2010; Villanueva et al., 2010; Heffels-Redmann et al., 2011; Heffels-Redmann et al., 2012; Piepenbring et al., 2012). This may indicate that so far unknown host and viral factors play an important role in disease induction. Although we have gained knowledge about the aetiologic role of PaBV, the pathogenic process remains obscure and to be solved. In order to investigate the effect of PaBV species on the clinical outcome, an experiment was undertaken using the same experimental methods as those described by Piepenbring et al. (2012), except with the use of a cockatiel-derived PaBV-2, rather than a scarlet macaw-derived PaBV-4, isolate. Both studies were performed under exactly the same conditions, which enables a valid and reliable comparison and statistical analysis of the results. Materials and methods Virus and sequence analysis The virus isolate used for experimental infection originated from a cockatiel (Nymphicus hollandicus) that presented in late stage PDD and was euthanized. PaBV (Ps 39) was isolated from the brain and passaged six times in the quail cell line CEC-32 (Zoller et al., 2000). The persistently PaBV-infected cells were suspended in medium with 2% FBS, sonicated, and clarified by centrifugation at 3000 g for 10 min. Afterwards, the supernatant was diluted with 0.9% NaCl and used for inoculation. For this purpose, an infectivity titre of ID 50 /ml (tissue culture infectious dose) was utilized. For sequence analysis, 200 µl of the virus suspension was used to isolate viral RNA applying the RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer s instructions. Total RNA was reverse transcribed using random hexamer primers. PCR for parts of the L- and the N-gene of PaBV was performed according to Kistler et al. (2008). PCR products were analysed by gel electrophoresis and purified for sequencing using the GeneJET PCR Purification Kit (Thermo Scientific, Dreieich, Germany). Sequencing was carried out by a commercial sequencing facility (LGC Genomics, Berlin, Germany). All generated sequence information was aligned using BlastN ( nlm.nih.gov/blast.cgi). Development of a PaBV-2 TaqMan-PCR Samples of organs from the cockatiel from which the infective PaBV isolate was derived were negative for PaBV using real-time-rt-pcr using both the primer pair and the primer pair 1367 according to Honkavuori et al. (2008). However, they were positive after applying conventional PCR for the L- and the N-gene of PaBV according to Kistler et al. (2008). In order to establish a quantitative analysis of viral load during the experimental infection, a PaBV-2 realtime-rt-taqman-pcr assay was established. Primers and probes were designed based on the known PCR assay with the primer set (Honkavuori et al., 2008) using an alignment of already published sequences of P gene from a PaBV-2 strain (EU781967) and a PaBV-4 strain (FJ169441) (Software: MegAlign, LaserGene DNAStar, Madison, WI, USA).

4 158 A. K. PIEPENBRING ET AL. The sequences of the forward primer and probe were modified as follows: 5 -CAGACAGCACGTC- GAATGAGG-3 ;probe:6-fam-5 -AGGTCTCCAA- GAAGGAAGCGA-3 -TMR (reverse primer: no modifications). PCR conditions were applied according to Honkavuori et al. (2008). The new PCR assay was tested against PaBV-4 using the PaBV-4 inoculum from the former experiment (Piepenbring et al., 2012). Experimental infection of cockatiels The set-up of the trial was identical to a previous study using PaBV-4 (Piepenbring et al., 2012) using genetically related birds from the same breeding flock. Nineteen cockatiels (N. hollandicus) of days of age were obtained from a flock that is kept under controlled conditions, and is regularly tested negative by PCR for Chlamydia psittaci, psittacine herpesvirus, psittacine circovirus, avian polyomavirus and PaBV. Prior to the experiment, crop and cloacal swabs from the birds were individually tested and were repeatedly negative for PaBV RNA by PCR using the new PCR protocol as well as the known PCR protocols published by Honkavuori et al. (2008) and Kistler et al. (2008). Blood was tested repeatedly and was negative for the presence of anti-pabv antibodies by indirect immunofluorescence assay (IIFA) (Herzog et al., 2010) and paramyxovirus-1 and -3 by haemagglutination inhibition assay. Physical examination and radiography revealed that the birds were in good health. Group 1, consisting of nine birds (three males, six females), was inoculated intracerebrally (i.c.) and group 2 (nine birds; five males, four females) was infected intravenously (i.v.) with 0.1 ml of the inoculum described above ( ID 50 /ml). One bird (male) remained untreated and served as a sentinel bird (se) in the intracerebrally inoculated group. In order to exclude any negative effect of the cell suspension on the birds, a group of four control birds was mock infected intracerebrally (n = 2, group 3 (i.c.)) and intravenously (n = 2, group 4 (i.v.)) with an uninfected cell suspension. All inoculations were performed under isoflurane inhalation anaesthesia. The study was governmentally approved by the Regierungspräsidium Giessen according to the guide for the care and use of birds (Reg.no GI 18/9 NR.02/210). Study design (sampling, clinical investigations and necropsies) Over a study period of 231 days, the birds were surveyed daily for their health status. Crop and cloacal swabs were taken three times per week. Crop and cloacal swabs from each time point were pooled per bird to investigate the presence of PaBV RNA using the newly established PaBV-2 real-time RT PCR assay. Ct values >36.0 were regarded as negative (Honkavuori et al., 2008). Thrice weekly swabbing was continued until all inoculated birds in the respective group tested positive, after which swabs were taken weekly. Blood samples (0.3 ml) were taken weekly from each bird for the detection of anti-pabv antibodies using indirect immunofluorescence assay (Herzog et al., 2010). In the control groups (3 and 4), all samples were taken once a week. If a bird exhibited clinical signs typical of PDD (emaciation, undigested seed in the faeces, neurologic signs) and if its general condition was reduced, it was euthanized for humane reasons. All remaining birds including the sentinel bird were euthanized at the end of the trial on day 231 p.i. and the mock-infected birds on day 100 post inoculation. All dead or euthanized birds were necropsied and organ samples from each bird (brain, eye, spinal cord, ischiadic nerve, adrenal gland, heart, liver, kidney, spleen, pancreas, crop, proventriculus, gizzard, intestine, pectoral muscle and skin with feathers) were processed the same way for further investigations. For histopathology and immunohistochemistry (IHC), organ samples were fixed in 10% buffered formalin, embedded in paraffin wax and used for the preparation of 5 μm sections stained with haematoxylin and eosin and immunohistological procedures. IHC was carried out by applying the avidin biotin complex method using a polyclonal rabbit p24 antibody directed against the phosphoprotein of mammalian 1 bornavirus (BoDV) according to previous studies (Herden et al., 1999; Herzog et al., 2010). Samples from the above listed organs were also fresh frozen for subsequent detection of PaBV RNA. Additionally, virus isolation was carried out on samples from brain, retina and kidney as previously described (Herzog et al., 2010; Piepenbring et al., 2012). To exclude other infections, blood, liver and lung were cultured on blood agar and Gassner agar, and samples from the crop and proventriculus were taken for mycological staining to exclude infection with yeast. Samples from all intestinal segments were examined for parasites. Statistical analysis and graphical data processing To ascertain differences between the intracerebrally and intravenously inoculated group and between the PaBV-4 (Piepenbring et al., 2012) and PaBV-2 trial, the Wilcoxon Mann Whitney-test was applied. Correlations were shown by the correlation coefficient. RNA loads are presented as 1/ct (values of ) or negative (value 0). Results Sequence analysis Using the BlastN ( Blast.cgi) sequence comparison of newly generated

5 AVIAN PATHOLOGY 159 sequences of parts of the N- and L-gene derived from the PaBV strain used for inoculation with sequences available at GenBank demonstrated high accordance (99%) for both genes (KM and KM053283) with PaBV-2. PaBV-2 RT TaqMan-PCR assay The newly established PaBV-2 RT TaqMan-PCR enabled a quantitative analysis of the PaBV-2 RNA load. RNA from the above described PaBV-2 virus suspension served as the positive control. The complementary DNA (cdna) was diluted 10-fold using the following dilutions: 10 1 (ct 22.7), 10 2 (ct 26.9), 10 3 (ct 29.3), 10 4 (ct 33.8) and 10 5 (ct 35.3). The PCR was not able to detect PaBV-4. Clinical observations and pathological findings Over the entire investigation period, 12 of the 18 inoculated birds (six birds in each group) exhibited clinical signs (Table 1). The signs ranged from reduced general condition (six birds in the i.v. group, six birds in the i.c. group) and reduced body weight (six birds in the i.v. group, three birds in the i.c. group) to diarrhoea (one bird in the i.c. group) and ataxia (two birds in the i.c. group). Five birds (two birds in the i.v. group, three birds in the i.c. group) displayed a rapid progression of disease and died unexpectedly within 1 16 days after demonstrating the first clinical signs. Six birds (four birds in the i.v. group, two birds in the i.c. group) were euthanized during the trial for humane reasons. One bird in the i.c. group recovered within a few days and remained clinically healthy until the end of the investigation period. Six other birds (three birds in the i.v. group, three birds in the i.c. group) and the sentinel bird showed no evidence of clinical disease. Necropsy revealed a dilated proventriculus in 10 of the 18 inoculated birds (six birds in the i.v. group, four birds in the i.c. group; Table 2). The degree of dilatation was severe in five cases (three birds in the i.v. group, two birds in the i.c. group), moderate in four birds (two birds in each group) and mild in one bird (i.v. group). The sentinel bird did not display any gross lesions. The inoculated birds fell into three clusters based on the severity of clinical signs and the days after inoculation when clinical signs occurred. Consequently, the study can be divided into three equal periods as follows (Table 1): Clinical group A, severe course of disease (days 1 77 p.i.): Five birds (three birds in the i.v. group, two birds in the i.c. group) died or were euthanized during the first third of the investigation period. Four of them showed emaciation, one bird displayed slight ataxia (i.c. 11) and another suffered from diarrhoea (i.c. 14). All had severe dilatation of the proventriculus. All birds in group A were female. Clinical group B, mild course of disease (days p.i.): Five birds (three birds in the i.v. group, two birds in the i.c. group) died or were euthanized within the second third of the trial. The clinical course of disease in these birds differed. One bird (i.c. 17) died one day after showing a reduced general condition and another one (i.c. 12) died one day after developing moderate ataxia. Four birds (one bird in the i.c. group, three birds in the i.v. group) were moderately emaciated Table 1. Intra vitam findings in cockatiels experimentally infected with PaBV-2 a. Death (dpi) Euthanasia (dpi) Group Bird ID b Sex/Age (d) c Antibodies/Titre d RNA e reduced GC reduced BW Diarrhoea Ataxia First detection of (dpi) First clinical signs (dpi) f A ic14 F/168 20/ iv13 F/196 27/ iv11 F/212 27/ ic11 F/216 20/ iv15 F/202 27/ B ic17 M/164 27/ ic12 F/212 27/ iv17 M/175 20/ iv10 M/235 27/ iv18 F/133 34/ C ic13 F / 196 6/ ic10 M/233 20/ ic15 F/182 34/ ic18 M/106 20/ iv14 M/202 55/ iv16 M/161 20/ iv12 M/196 76/ ic16 F/175 34/ se2 M/196 < Note: Group A: severe course of disease, Group B: mild course of disease, Group C: moderate course of disease. a Investigation period of 231 days, birds sorted by time point of death/euthanasia. b ic = Intracerebrally inoculated, iv = Intravenously inoculated, se = Sentinel. c Age at time point of inoculation. d Anti-PaBV antibody titre (1:x) detection by indirect immunofluorescence assay, the sentinel bird did not reveal anti-pabv antibodies at any time. e PaBV-2 RNA detection by real-time RT-PCR, Ct value >36.0 = Negative. f GC = general condition, BW = body weight.

6 160 A. K. PIEPENBRING ET AL. Table 2. Post-mortem findings in cockatiels experimentally infected with PaBV-2 a. Group Bird ID b (dpi) Death/Euthanasia Antibody titre d.o.d. c RNA load in organs d Dilatation of proventriculus e Histopathology f Antigen detection g Infectious virus h A ic PDD + (+) iv PDD + (+) iv PDD + (+) ic PDD + (+) iv PDD + (+) B ic PDD + + ic PDD + + iv PDD + (+) iv PDD + + iv PDD + + C ic PDD + + ic PDD + + ic PDD + (+) ic PDD + + iv PDD + + iv PDD + + iv PDD + (+) ic PDD + + se2 231 <1: (PDD) - - a For data of time point of death post-infection see Table 1. b ic = intracerebrally inoculated, iv = intravenously inoculated, se = sentinel. c Anti-PaBV antibody titre (1:x) detection by indirect immunofluorescence assay, d.o.d. = day of death. d PaBV-2 RNA detection by real-time RT-PCR, RNA load of all organs per bird as 1/ct (median). e 0 = No dilatation, 1 = Mild dilatation, 2 = Moderate dilatation, 3 = Severe dilatation. f PDD = Proventricular dilatation disease, histopathologically confirmed by lymphoplasmocytic infiltrates of/near ganglia in the central nervous system and/or upper gastrointestinal tract. g Detection of the phosphoprotein (p24) of Borna disease virus by IHC. h Re-isolation of PaBV-2 in the CEC-32 quail cell line, + = reisolation in three passages or less, (+) = Reisolation after three passages. the day they died or were euthanized. The proventriculus showed moderate dilatation in two cases (i.v. 10 and i.v. 17). Three birds in group B were male, and two birds were female. Clinical group C, moderate course of disease/clinically healthy birds (days p.i.): One bird in the i.c. group (female) was euthanized within the last third of the trial on day 172 p.i. after a mild progression of disease for 33 days with a reduced general condition. The proventriculus of this bird was moderately dilated. Three birds in the i.v. group and three birds in the i.c. group (five males, one female bird) and the sentinel bird (male) remained healthy until the end of the trial. One of them (i.v. 16) displayed mild enlargement of the proventriculus. The eighth bird of group C (i.c. 16, female) showed reduced body weight and reduced general condition from day 85/86 p.i., but recovered within a few days and stayed healthy until the end of the investigation period. However, the proventriculus was moderately dilated. The mock-infected birds remained healthy during the entire experimental period and did not show any macroscopic alterations. Histopathology PDD was confirmed in all inoculated birds by histopathological examination (Table 2). Typical lymphoplasmocytic infiltrations near or in ganglia and nerve fibres in the upper gastrointestinal tract were found in all inoculated birds. Those infiltrations were also present in the heart, liver, kidney, spleen, skin and in the pancreatic tissue of several birds. In the brain, no inflammatory lesions were noted but gliosis was found in 15 of the inoculated birds. The sentinel bird exhibited lymphoplasmocytic infiltration in the proventriculus and intestine, in the liver, kidney and skin, and the mock-infected birds displayed similar mild infiltrations in the liver, kidney, gizzard and intestine and in individual cases in the brain, spinal cord, spleen, crop, proventriculus and skin. However, these findings differed from the typical PDD lesions as the infiltrations revealed a milder and more disseminated distribution within the tissue and they were not found in nerves and ganglia but in the mucosal layers and parenchyma. Kinetics of PaBV RNA and anti-pabv antibody detection during the trial PaBV-2 RNA was first detected in the swabs between days 27 and 85 p.i. (Figure 1). Birds in group 1 (i.c. group) tested positive for the first time between days 27 and 83 p.i. and birds in group 2 (i.v. group) between days 43 and 85 p.i. Statistical analysis did not reveal any noticeable difference between the i.c. and i.v. groups regarding the timepoint of first RNA detection (calculated value for test statistic U = 17 which is >13 (critical value with n 1 = 8 and n 2 = 7) and thus considered not to be significant with α = 0.1). Three birds (i.c. 14, i.v. 13 and i.v. 11) did not shed viral RNA before they died or were euthanized (32, 46 and 50 dpi). The sentinel bird tested positive from day

7 AVIAN PATHOLOGY 161 The birds of the mock-infected group did not display any PaBV RNA in the swabs or anti-pabv antibodies in the serum at any timepoint. PaBV-RNA detection in organs Figure 1. Date of first detection of PaBV RNA (days post-infection) in combined crop and cloacal swabs in cockatiels experimentally infected intracerebrally (i.c.) and intravenously (i.v.) with PaBV-2. Statistical analysis did not reveal any significant difference between the i.c. and i.v. groups regarding the time point of first RNA detection. Three birds (i.c.14, i.v.11, i.v.13) did not shed viral RNA before they died or were euthanized. 64 p.i. onwards. RNA loads in the swabs of all birds increased constantly within the investigation period (1/ct 0.03 at first RNA detection until 1/ct 0.05 at the end of the trial median). Specific anti-pabv antibodies in birds in group 1 (i.c.) were first detected between days 6 and 34 p.i. (Figure 2). Birds in group 2 (i.v.) displayed seroconversion between days 20 and 76 p.i. Statistical analysis did not reveal any noticeable difference between the i.c. and i.v. groups regarding the timepoint of seroconversion (calculated value for test statistic U = 25 which is >21 (critical value with n 1 = 9 and n 2 = 9) and thus considered not to be significant with α = 0.1). Antibody titres increased in all inoculated birds until they died or were euthanized. Anti-PaBV antibodies were not detected in the sentinel bird during the study. PaBV-2-RNA was detected in the organs and tissue of every inoculated bird (Figure 3). The highest RNA loads were found in the eye, the gizzard and the adrenal gland (1/ct 0.05 and 0.06 median). The remaining organs revealed RNA loads between 0.03 and 0.04 (median). The Wilcoxon Mann Whitney-test indicated significant differences in median viral load by organ for the i.c. and i.v. groups. For each organ, the viral load was higher in the i.c. group with the highest 1/ct values (about 0.06) in the brain, eye, adrenal gland, crop, gizzard and intestine (calculated value for test statistic U = 29 which is 57 (critical value with n 1 =9andn 2 =9)andthusconsidered to be highly significant with α = 0.005). Birds with early euthanasia/death and severe proventricular dilatation (clinical group A; three of which did not shed viral RNA before death/euthanasia) displayed lower organ viral loads than birds in clinical groups B and C. A correlation coefficient of r = 0.51 indicates a moderate correlation between the time point of death/ euthanasia after inoculation and the median RNA load (1/ct) of the entire organ samples of the respective bird. Birds in group A (early time point of death) contained lower RNA amounts in organs than most of the birds in group C. The sentinel bird tested positive for PaBV RNA in the skin with feathers (1/ct 0.03) but was negative in the remaining organs. The mock-infected birds did not contain any PaBV RNA in any of the organ samples. Immunohistochemistry Viral phosphoprotein was detected in gastrointestinal tissue in all inoculated birds (Table 2). In 14 of the 18 inoculated birds (six in the i.v. group and eight in the i.c. group) antigen detection was possible in the central nervous system (CNS) as well. In birds which remained alive until the end of the trial, PaBV antigen was distributed more widely and was found more frequently in other organs like the kidney, liver, pancreas, heart, muscle and skin). In the sentinel bird as well as in the mock-infected birds, no viral antigen was detected in any organ. Figure 2. Date of first detection of anti-pabv antibodies (days post-infection) in cockatiels experimentally infected intracerebrally (i.c.) and intravenously (i.v.) with PaBV-2. Statistical analysis did not reveal any significant difference between the i.c. and i.v. group regarding the time point of first detection. Virus isolation Re-isolation of PaBV-2 was successful from all inoculated birds (Table 2). There was a positive correlation between an individual bird s median RNA load and the amount of re-isolated virus. Thus, the timespan

8 162 A. K. PIEPENBRING ET AL. Figure 3. PaBV-2 RNA load in organ samples collected at necropsy from cockatiels experimentally infected intracerebrally (i.c.) and intravenously (i.v.) with PaBV-2. RNA loads are presented by organ using the median value for each organ per group. between inoculation and death/euthanasia influenced the re-isolation process. In all birds of clinical group A (early death/euthanasia), re-isolation succeeded from either the brain or retina or both after passaging infected CEC cells only three times. Infectious virus was not isolated from the kidneys of these birds. The longer the birds lived, the higher the amount of re-isolated infectious virus. In clinical groups B (except bird i.v. 17) and C (except birds i.v. 12 and i.c. 15), cells were 100% infected five days after infection with material from the brain and retina. Re-isolation lasted longer in birds i.v. 17, i.v. 12 and i.c. 15 due to low RNA loads in the respective organs. Using kidney for infectivity assays, re-isolation succeeded in three birds in clinical group B (two birds in the i.c. group, one bird in the i.v. group) and in five birds of clinical group C (three birds in the i.c. group, two birds in the i.v. group). In the remaining birds, re-isolation from kidney failed. These birds presented moderate to low RNA loads in the kidney. No infectious virus was isolated from the sentinel bird or the mock-infected birds. Bacteriological, mycological and parasitological examinations did not detect any infection in any of the birds. Comparison of PaBV-2 and PaBV-4 infection (Piepenbring et al., 2012) Clinical observations and macroscopic findings Five of the PaBV-4-infected birds exhibited signs consistent with PDD in contrast to 12 birds inoculated with PaBV-2. Clinical signs first occurred on day 22 p.i. (PaBV-2 infection) and on day 33 p.i. (PaBV-4 infection). Proventriculi were enlarged in seven cases (PaBV-4 infection) and in 10 cases (PaBV-2 infection). After PaBV-2 infection, five birds had a severely dilated proventriculus in contrast to only one bird in the PaBV-4 group. RNA detection and seroconversion PaBV RNA was first detected between days 19 and 72 (PaBV-4 infection) and between days 27 and 85 p.i. (PaBV-2 infection) (Figure 4). Anti-PaBV antibodies were first detected between days 7 and 63 p.i. (PaBV- 4 infection) and between days 6 and 76 p.i. (PaBV-2 infection) (Figure 5). PaBV RNA was detected significantly earlier in PaBV-4-infected birds, whereas seroconversion occurred earlier in PaBV-2-infected birds

9 AVIAN PATHOLOGY 163 Figure 4. Comparison of date of first PaBV-RNA detection in combined crop/cloacal swabs from cockatiels after inoculation with PaBV-4 (Piepenbring et al., 2012) and PaBV-2. PaBV-RNA was detected significantly earlier in PaBV-4 infected birds (α = 0.005; Wilcoxon Mann Whitney test). (calculated value for test statistic U = 51/71 which is 64/81 (critical value with n 1 = 15/18 and n 2 = 18/18) and thus considered to be highly significant with α = 0.005). RNA detection in organs Birds infected with PaBV-4 displayed high RNA loads in all organs, regardless of the time point they died or were euthanized after infection or the route of inoculation (Figure 6). In contrast, RNA load in the organs of PaBV-2-infected birds depended on the route of infection and how long they stayed alive after inoculation; for example, RNA loads were lowest in group A(Table 2). Histopathology and IHC The characteristic PDD lesions were found in all inoculated birds either after PaBV-2 or after PaBV-4 infection. Viral antigen likewise was detected in all inoculated birds. However, in the PaBV-4-infected birds, IHC showed that the amount of antigen in the CNS was higher than in the PaBV-2 group. Re-isolation of infectious virus PaBV-4 was isolated from many tissues and induced a persistent infection of 100% of the CEC cells 5 days after infection (Herzog, unpublished data), whereas success of PaBV-2 re-isolation depended on how long the respective bird was infected until it died or was euthanized. Discussion PaBV has been shown to be the aetiologic agent of PDD, but the variety of outcomes of PaBV infections and unknown pathogenicity of different PaBV species emphasize the need for reliable and comparable investigations to resolve these open questions. Since the discovery of PaBV in 2008, several infection trials have been carried out using PaBV-4 (Gancz et al., 2009; Gray et al., 2010; Piepenbring et al., 2012), which is the predominant PaBV species in Europe and North America (Hoppes et al., 2010; Staeheli et al., 2010). So far, only one group has used a PaBV-2 isolate to inoculate two cockatiels, which were also co-infected with PaBV-4 that was shed intermittently. These Figure 5. Comparison of date of first anti-pabv antibody detection in cockatiels after inoculation with PaBV-4 (Piepenbring et al., 2012) and PaBV-2. Seroconversion was detected significantly earlier in PaBV-2 infected birds (α = 0.005; Wilcoxon Mann Whitney test).

10 164 A. K. PIEPENBRING ET AL. Figure 6. PaBV-4 RNA load in organ samples of experimentally infected cockatiels after death/euthanasia (Piepenbring et al., 2012). RNA loads are presented by organ using the median value for each organ per group. The birds displayed high RNA loads in all organs, regardless of the time point at which they died or were euthanized after infection or the route of inoculation. birds developed signs consistent with PDD and displayed the typical histopathological lesions. PaBV-2 RNA was not detected in faecal samples from living birds, but was detected in organ samples (Mirhosseini et al., 2011). Therefore, it remains questionable if the observed lesions were caused by PaBV-2 or PaBV-4. The present study is the first one that addresses the pathogenicity of PaBV-2 by investigating the course of infection and outcome of disease employing a statistically adequate number of healthy cockatiels. Moreover, the use of exact same study design as in a previous investigation with PaBV-4 as the infectious agent (Piepenbring et al., 2012) allows very precise comparison of the results. This provides consideration that the respective PaBV species indeed influences the outcome and progression of disease. Besides statistical differences between the results of both studies, there were also many parallels, substantiating once more the causative role of PaBV in the development of PDD. In both studies, all birds were persistently infected with PaBV and revealed lymphoplasmocytic infiltrates in neuronal tissues typical of PDD. No significant differences were noted according to the route of infection after PaBV-2 infection, as was found for PaBV-4 infection. This underlines once more the discrepancy between PaBV and BoDV infection, since it is not possible to infect rats with BoDV by i.v. inoculation (Narayan et al., 1983; Richt et al., 1992). However, remarkable temporal spatial differences between the infection trials were noted; for example, regarding the course and severity of disease as well as viral load. After PaBV-4 infection, there was no significant difference between the diseased birds. All of them seroconverted and shed virus. There was also no significant difference regarding PaBV detection in organs. However, after PaBV-4 infection, the clinical signs significantly differed. Some birds showed gastrointestinal signs, some birds had neurologic signs and one bird displayed

11 AVIAN PATHOLOGY 165 neurologic and gastrointestinal signs. After PaBV-2 infection, three different disease patterns were observed, depending on the time point of disease onset. In contrast to the PaBV-4 study, many more birds developed clinical signs (12 birds), and 10 birds showed dilation of the proventriculus when infected with PaBV-2. Five of these birds had a severely dilated proventriculus in contrast to one bird after PaBV-4 infection. This might indicate a greater pathogenicity of the PaBV-2 isolate used in these cockatiels. Most of the birds exhibited non-specific signs, like fluffing and weight loss. After PaBV-4 infection, birds showed undigested seed in the faeces and/or neurologic signs. In the PaBV-2 experiment, all birds of clinical group A(first third of the study period) displayed severe enlargement of the proventriculus. However, this seems not to be the result of a rapid and overwhelming viral replication and distribution because the viral load was significantly lower compared to the birds of clinical groups B and C. Whether this could be related to an earlier and/or stronger activation of the immune response needs to be further investigated. Moreover, RNA shedding was observed later after PaBV-2 infection than after PaBV-4 infection, and three PaBV-2- infected birds did not even shed viral RNA before they died. This might point to a lower replication efficiency of PaBV-2 in contrast to PaBV-4, possibly again due to an earlier and/or stronger activation of the antiviral immune response. However, it cannot be excluded that the two isolates of the PaBV species differ in their viral fitness per se. In addition, RNA amounts in the organs were significantly higher in the i.c. than in the i.v. group after PaBV-2 infection, which might indicate better conditions in or better adaptation to the nervous tissue for PaBV-2. In the first study using PaBV-4, all birds displayed high RNA amounts in multiple organs, independent of the time point of death or inoculation route indicating a more stable viral replication and distribution. Antibody titres also differed considerably within the PaBV-2 group. Whereas most of the PaBV-4 birds developed high antibody titres at the end of the trial up to 1:20,480, antibody titres in the PaBV-2 group were lower ranging from 1:320 to 1:20,480 at the time point of death even in the birds which survived until the end of the study (Figure 7). In contrast to PaBV-4 infection, the amount of virus re-isolated from birds inoculated with PaBV-2 depended on the duration of infection. This also points to slower replication of PaBV-2 when compared to PaBV-4. Antigen detection also differed between the two PaBVs and was possible in the CNS only in 14 birds within the PaBV-2-infected cohort. Eight of these birds belonged to the i.c. group and six to the i.v. group, fitting well to the higher load of viral RNA in the i.c. group which could be due to the intracerebral route of inoculation. After PaBV-4 infection, viral antigen was detected in all inoculated birds in the CNS as Figure 7. Anti-PaBV antibody titres in cockatiels experimentally infected with PaBV-2 and PaBV-4 (Piepenbring et al., 2012) at day of death or euthanasia. Antibody titres differed considerably within the PaBV-2 group. Whereas most of the PaBV-4 birds developed high antibody titres at the end of the trial up to 1:20,480, antibody titres in the PaBV-2 group were lower ranging from 1:320 to 1:20,480 at the time point of death even in the birds which survived until the end of the study. well as in the gastrointestinal tract, regardless of the inoculation route. Besides the respective PaBV species, the differences in outcome after either PaBV-2 or PaBV-4 infection and the variability within each infectious trial underline the role of host factors in the pathogenesis of PaBV infection. Whether they represent one of the important missing links to fully understand the disease process warrants further investigation. One key point might be the activation and efficiency of mounting an effective antiviral response. For mammalian 1 bornavirus (BoDV) infections, the immunopathogenic principle of the disease is well known. For instance, BoDV infections in rats have demonstrated that a minimum threshold of viral replication is sufficient to trigger clinical signs. Moreover, the onset and progression of disease depend on the host immune response to the virus (Carbone et al., 1987; Oldach et al., 1995). However, the precise comparison of experimental infection with PaBV-2 and PaBV-4 in our study also documents the influence of the respective viral species. Whether the origin of the isolate might also play an additional role cannot be ruled out. The PaBV-4 isolate used was obtained from a scarlet macaw, whereas the PaBV-2 isolate originated from a cockatiel; greater species adaption could have contributed to the higher

12 166 A. K. PIEPENBRING ET AL. Figure 8. Ganglioneuritis and periganglioneuritis in the crop (a) and proventriculus (b) of a PaBV-2 inoculated bird, typical for PDD, and non-typical follicular shape aggregations of mononuclear cells in the kidney of the sentinel bird (c) and in the liver of a mock infected bird (d). number of diseased birds after PaBV-2 infection. Nevertheless, both isolates clearly differed in their behaviour in terms of re-isolation. Whether gender had an effect on disease outcome remains speculative, but all birds in clinical group A (the early time point of death) were female. Group C contained six birds which remained completely healthy until the end of the trial. Five of these birds were male (initial distribution: i.c. group three male, six female birds; group i.v. five male, four female birds). However, in the present study, female and male birds were randomly distributed within the experimental cohorts (i.v. and i.c. groups). Thus, a possible influence of gender on the course of infection should be further investigated. The results of the sentinel bird were largely similar to the results obtained in the PaBV-4 experiment. In the present study, the sentinel bird did not seroconvert but tested positive for PaBV-2 RNA from day 64 p.i. onwards. Interestingly, this time point is earlier than in the PaBV-4 experiment, where PaBV RNA was first detected in the sentinel bird on day 76 p.i. In the organ samples, no PaBV RNA was detected, except in the skin with feathers. This might be due to contamination from the inoculated birds which constantly shed viral RNA. Moreover, re-isolation did not succeed. However, like the sentinel in the PaBV-4 trial, histopathological examination revealed mononuclear infiltrates in some organs. Thus, it remains unclear if the sentinel was infected subclinically or cleared the infection. The mock-infected birds neither shed viral RNA nor displayed anti-pabv antibodies during the entire experimental period. Interestingly, these birds revealed mild lymphoplasmocytic infiltration in some organs which might be due to other previous activations of the immune system since the birds were not specific pathogen-free, but rather tested negative for a panel of infectious agents. There was neither PaBV RNA nor PaBV antigen present in any organ sample tested. The newly established PaBV-2 detecting real-time RT PCR assay allows quantification of the results and, based on the same conditions and enzyme as the PCR assay published by Honkavuori et al. (2008), allows comparison of the PaBV-2 study and the PaBV-4 study (Piepenbring et al., 2012). Basically, results of one PCR assay should not be directly transferred to another. But it is unlikely that obvious differences are due to the different PCR protocols. In summary, the present study provides the first evidence for a role of the respective PaBV species in outcome and progression of infection, at least in cockatiels, as shown by the significantly different disease patterns after each PaBV infection. Future studies using additional isolates of each PaBV species will further contribute to attributing the pathogenic capacity of PaBV-4 and PaBV-2 per se. Additionally,

13 AVIAN PATHOLOGY 167 further studies should also focus on the important role of the host constitution. Acknowledgements The authors thank gratefully Kristina Maier, Helena Schneider, Luisa Ziegler and Hannah Wünnemann for the support in sampling of the birds and Antoinette Huhn and Christoph Kleinehakenkamp for support in microbiological and PCR testing. Hildburg Lange-Herbst was supported by the Margarete Ammon Stiftung. Funding The authors thank the Loro Parque Fundacion (Project ID: PP ), and the Association for the Conservation of Threatened Parrots (ACTP) for the financial support of the study. This work was supported by the Loro Parque Fundacion, Tenerife, Spain (Project ID: PP ) and Hildburg Lange-Herbst was supported by the Margarete Ammon Stiftung. References Carbone, K. M., Duchala, C. S., Griffin, J. W., Kincaid, A. L., & Narayan, O. (1987). Pathogenesis of Borna disease in rats: evidence that intra-axonal spread is the major route for virus dissemination and the determinant for disease incubation. Journal of Virology, 61, Daoust, P. Y., Julian, R. J., Yason, C. V., & Artsob, H. (1991). Proventricular impaction associated with nonsuppurative encephalomyelitis and ganglioneuritis in two Canada geese. Journal of Wildlife Diseases, 27, De Kloet, S. R., & Dorrestein, G. M. (2009). Presence of avian bornavirus RNA and anti-avian bornavirus antibodies in apparently healthy macaws. Avian Diseases, 53, Delnatte, P., Berkvens, C., Kummrow, M., Smith, D. A., Campbell, D., Crawshaw, G., Ojkic, D., & DeLay, J. (2011). New genotype of avian bornavirus in wild geese and trumpeter swans in Canada. Veterinary Record,169,108. Gancz, A.Y., Kistler, A.L., Greninger, A.L., Farnoushi, Y., Mechani, S., Perl, S., Berkowitz, A., Perez, N., Clubb, S., DeRisi, J.L., Ganem, D., & Lublin, A. (2009). Experimental induction of proventricular dilatation disease in cockatiels (Nymphicus hollandicus) inoculated with brain homogenates containing avian bornavirus 4. Virology Journal, 6, 100. Gray, P., Hoppes, S., Suchodolski, P., Mirhosseini, N., Payne, S., Villanueva, I., Shivaprasad, H.L., Honkavuori, K., Briese, T., Reddy, S.M., & Tizard, I. (2010). Use of avian bornavirus isolates to induce proventricular dilatation disease in Conures. Emerging Infectious Diseases, 16, Guo, J., Tizard, I., Baroch, J., Shivaprasad, H. L., & Payne, S. L. (2015). Avian bornaviruses in North American gulls. Journal of Wildlife Diseases, 51, Heffels-Redmann, U., Enderlein, D., Herzog, S., Herden, C., Piepenbring, A., Neumann, D., Muller, H., Capelli, S., Oberhauser, K., Gerlach, H., Kaleta, E.F., & Lierz, M. (2011). Occurrence of avian bornavirus infection in captive psittacines in various European countries and its association with proventricular dilatation disease. Avian Pathology, 40, Heffels-Redmann, U., Enderlein, D., Herzog, S., Piepenbring, A., Burkle, M., Neumann, D., Herden, C., & Lierz, M. (2012). Follow-up investigations on different courses of natural avian bornavirus infections in psittacines. Avian Diseases, 56, Herden, C., Herzog, S., Wehner, T., Zink, C., Richt, J., & Frese, K. (1999). Comparison of different methods of diagnosing Borna disease in horses post mortem. In U. Wernery, J. Wade, J. Mumford, & O. Kaaden (Eds.), Eqiune infectious diseases VIII (pp ). Newmarket: R&W Publications. Herzog, S., Enderlein, D., Heffels-Redmann, U., Piepenbring, A., Neumann, D., Kaleta, E. F., Müller, H., Lierz, M., & Herden, C. (2010). Indirect immunofluorescence assay for intra vitam diagnosis of avian bornavirus infection in psittacine birds. Journal of Clinical Microbiology, 48, Honkavuori, K. S., Shivaprasad, H. L., Williams, B. L., Quan, P. L., Hornig, M., Street, C., Palacios, G., Hutchison, S. K., Franca, M., Egholm, M., Briese, T., & Lipkin, W. I. (2008). Novel borna virus in psittacine birds with proventricular dilatation disease. Emerging Infectious Diseases, 14, Hoppes, S., Gray, P.L., Payne, S., Shivaprasad, H.L., & Tizard, I. (2010). The isolation, pathogenesis, diagnosis, transmission, and control of avian bornavirus and proventricular dilatation disease. The Veterinary Clinics of North America. Exotic Animal Practice, 13, Kistler, A.L., Gancz, A., Clubb, S., Skewes-Cox, P., Fischer, K., Sorber, K., Chiu, C.Y., Lublin, A., Mechani, S., Farnoushi, Y., Greninger, A., Wen, C.C., Karlene, S.B., Ganem, D., & DeRisi, J.L. (2008). Recovery of divergent avian bornaviruses from cases of proventricular dilatation disease: identification of a candidate etiologic agent. Virology Journal, 5, 88. Kistler, A.L., Smith, J.M., Greninger, A.L., Derisi, J.L., & Ganem, D. (2010). Analysis of naturally occurring avian bornavirus infection and transmission during an outbreak of proventricular dilatation disease among captive psittacine birds. Journal of Virology, 84, Kuhn, J.H., Dürrwald, R., Bào, Y., Briese, T., Carbone, K., Clawson, A.N., derisi, J.L., Garten, W., Jahrling, P.B., Kolodziejek, J., Rubbenstroth, D., Schwemmle, M., Stenglein, M., Tomonaga, K., Weissenböck, H., &, Nowotny N. (2015) Taxonomic reorganization of the family Bornaviridae, Archives of Virology, 160, Lierz, M., Hafez, H.M., Honkavuori, K.S., Gruber, A.D., Olias, P., Abdelwhab, E.M., Kohls, A., Lipkin, W.I., Briese, T., & Hauck, R. (2009). Anatomical distribution of avian bornavirus in parrots, its occurrence in clinically healthy birds and ABV-antibody detection. Avian Pathology, 38, Mirhosseini, N., Gray, P.L., Hoppes, S., Tizard, I., Shivaprasad, H.L., & Payne, S. (2011). Proventricular dilatation disease in cockatiels (Nymphicus hollandicus) after infection with a genotype 2 avian bornavirus. Journal of Avian Medicine and Surgery, 25, Mirhosseini, N., Gray, P. L., Tizard, I., & Payne, S. (2012). Complete genome sequence of avian bornavirus genotype 1 from a Macaw with proventricular dilatation disease. Journal of Virology, 86, Narayan, O., Herzog, S., Frese, K., Scheefers, H., & Rott, R. (1983). Pathogenesis of Borna disease in rats: immunemediated viral ophthalmoencephalopathy causing blindness and behavioral abnormalities. Journal of Infectious Diseases, 148, Oldach, D., Zink, M. C., Pyper, J. M., Herzog, S., Rott, R., Narayan, O., & Clements, J. E. (1995). Induction of protection against Borna disease by inoculation with high-doseattenuated Borna disease virus. Virology,206,