Recombinant expression of a truncated capsid protein of beak and feather disease virus and its application in serological tests

Abstract

Beak and feather disease virus (BFDV) causes severe disease characterized by irreversible feather disorders and severe immunosuppression in many psittacine species. BFDV cannot be propagated in tissue or cell cultures, rendering virus propagation and thus diagnosis rather difficult. To develop reliable diagnostic methods, the region encoding the BFDV capsid protein C1 was cloned from an infected sulphur-crested cockatoo (Cacatua galerita). Phylogenetic analysis showed this gene had 76.3 to 83.2% amino acid identity to published sequences. No protein was detected after induction of full-length C1 expression in Escherichia coli. However, deletion of an amino-terminal arginine-rich sequence facilitated expression. C1_39–244-His, a polyhistidine-tailed variant of this protein, was purified and used for immunization of chickens. The immune sera detected C1 with an apparent molecular weight of 27 kDa in western blots of organ homogenates of BFDV-infected birds. Using C1_39–244-His as antigen, 11 psittacine sera were tested for the presence of BFDV-specific antibodies by enzyme-linked immunosorbent assay and immunoblotting. The results obtained correlated well with the BFDV-specific haemagglutination inhibition activity of the sera, suggesting C1_39–244-His has value as a recombinant antigen for BFDV-specific serological tests.

Introduction

Beak and feather disease virus (BFDV) causes severe disease in psittacine birds. Most species in the order Psittaciformes and some other bird species are susceptible. Usually nestlings and young birds are most frequently affected, but sometimes clinical signs of beak and feather disease are observed in adult birds as well (Ritchie, 1995). The main clinical features in the acute course of the disease are the irreversible loss of feathers all over the body and severe immunosuppression, which may result in lethal secondary infections. In chronic cases, the loss of feathers slowly progresses and, particularly in cockatoo species, beak elongation is observed.

BFDV belongs to the family Circoviridae. With a diameter of 17 to 25 nm, circoviruses are among the smallest animal viruses. The non-enveloped viral capsid has an icosahedral symmetry and surrounds a single-stranded circular DNA genome of 1700 to 2300 nucleotides (Todd et al., 2000). Recently, the family Circoviridae has been divided into two genera, Gyrovirus and Circovirus (Pringle, 1999). The sole member of the genus Gyrovirus is chicken anaemia virus (CAV), which can be morphologically and biologically differentiated from the other circoviruses. The genus Circovirus consists of the porcine circovirus and BFDV. Based on homologies in the viral genomes, pigeon circovirus, goose circovirus and canary circovirus should also be grouped in this genus (Phenix et al., 2001; Todd et al., 2001a,b). Members of the genus Circovirus possess similar genome organization, with two major open reading frames (ORF) in opposing orientations (Todd et al., 2001b). The ORF for the replication-associated (Rep) protein (ORF V1) is located in a virus-sense orientation, whereas the capsid protein (ORF C1) is encoded in a complementary-sense orientation. There are additional smaller ORFs in both orientations, mainly of unknown significance (Bassami et al., 1998, 2001).

The composition of the BFDV capsid is not definitively understood. In purified BFDV preparations, either three proteins with apparent molecular weights of 26.3, 23.7 and 15.9 kDa (Ritchie et al., 1989), or two proteins with apparent molecular weights of 26 and 23 kDa (Ritchie et al., 1990) have been observed. This is in contrast to CAV and the porcine circoviruses, which have only one capsid protein (Todd et al., 2000).

Routine diagnosis of BFDV infection is difficult because the virus cannot be propagated in tissue culture systems or in embryonated chicken eggs. Polymerase chain reaction (PCR) is routinely used to confirm clinical and histological observations (Ypelaar et al., 1999; Raue et al., 2004). However, extensive genetic diversity has been detected in BFDV isolates from a number of psittacine birds (Bassami et al., 2001; Ritchie et al., 2003; Raue et al., 2004), which may limit the use of this technique.

The haemagglutination inhibition (HI) assay is the only serological test available for detection of BFDV-specific antibodies (Raidal et al., 1993). However, there seems to be some disagreement about the source of suitable erythrocytes (Ritchie et al., 1991; Sexton et al., 1994; Sanada & Sanada, 2000). Furthermore, differences in the agglutinating ability of erythrocytes obtained from different individuals of the same species have also been reported (Sanada & Sanada, 2000). In general, erythrocytes collected from cockatoos are considered to be more sensitive for detection of BFDV than those of other avian species. The genetic and antigenic diversity of BFDV further limits the applicability of this test. There is still a need for standardized antigens to enable the comparison of tests.

In the study described here, the possibility of a recombinant expression of ORF C1 of BFDV in Escherichia coli was investigated. The immunogenicity of the recombinant protein and its suitability to serve as an antigen in serological tests was tested.

Materials and Methods

PCR amplification and cloning of the region encoding C1

The primers C1-s (5′-TGGGACATGTCTAACTACGCATG-3′ and C1-as (5′-GAGTAGATCTTAAGTACTGGGATT-3′), which were used in a PCR amplifying the region encoding the C1 protein of BFDV (Figure 1), were selected on the basis of a published BFDV genome sequence (Niagro et al., 1998). As the initiation codon for the gene does not appear to be ATG, an ATG sequence was introduced into primer C1-s upstream of the TCT codon at nucleotide position 1974 to 1972, which is thought to be the initiation codon for C1 translation in BFDV (Niagro et al., 1998). DNA isolated from the feather shaft of a sulphur-crested cockatoo (Cacatua galerita), which had been found to contain BFDV by PCR (sample M99-98 in Raue et al., 2004), was used as template for PCR amplification using the Expand High Fidelity PCR Kit (Boehringer, Mannheim, Germany). A PCR product of 755 base pairs was digested with AflIII and BglII, and ligated to the NcoI-digested and BglII-digested pQE-60 vector (Qiagen, Hilden, Germany). The resulting plasmid pC1 was used to transform E. coli strain XL1-blue MRF (Stratagene, Amsterdam, The Netherlands). Plasmid DNA was purified using the Qiagen Plasmid Mini Kit (Qiagen) and the insert was sequenced using dye terminators in an ABI PRISM device (Applied Biosystems, Foster City, CA, USA).

Figure 1. Genome organization of BFDV focusing on the two major ORFs encoding the Rep protein (ORF V1) and the capsid protein (ORF C1). ORFs are labelled according to their localization on the virus (V) or complementary (C) strand. The positions of the primers are indicated (small arrows), with nucleotide (nt) numbering according to Niagro et al. (1998).

Sequence analysis of the cloned C1 gene

Sequence data were compiled and analysed with the computer software package Lasergene (DNASTAR Inc., Madison, WI, USA). The complete nucleotide sequence of ORF C1 has been submitted to the GenBank database (accession number AY345131). The deduced amino acid sequence of the C1 protein was aligned with those of published BFDV strains (AF311295-AF31302, Bassami et al., 2001; AF080560, Bassami et al., 1998; AF071878, Niagro et al., 1998) using Clustal W (Thompson et al., 1994).

Expression of C1 variants in E. coli

An overview on the four C1 variants expressed in this study is presented in Figure 2a. For the construction of a plasmid expressing C1 with a deletion of amino acids 1 to 38, designated pC1_39–244, a PCR using primers C1₃₉-s (5′-AGACACATGTTCACAACCAATAG-3′) and C1-as was performed using pC1 as template. For the expression of His-tailed C1, plasmid pC1-His was constructed using a PCR product obtained with the primers C1-s and C1-His-as (5′-GAGTCTAGATCTAGTACTGGGATT-3′) and pC1 as template. The plasmid for expression of His-tailed C1 with the deletion of amino acids 1 to 39 (pC1_39–244-His) was constructed using a PCR with the primers C1₃₉-s and C1-His-as and pC1 as template. The respective PCR products were cleaved with AflIII and BglII, and ligated to NcoI-digested and BglII-digested pQE-60 (Qiagen). The sequences of the inserts were confirmed by DNA sequencing as already described.

Figure 2. Expression of C1 protein variants in E. coli. 2a: Schematic presentation of the amino acid sequences deduced from the nucleotide sequences of the C1 expression plasmids. 2b: Analysis of protein expression by SDS-PAGE followed by Coomassie brilliant blue staining. Bacterial lysates (lysates) were obtained at 5 h after induction with IPTG using a buffer containing 8 M urea. His-tailed proteins were purified (purification) from the lysates using Ni-NTA-agarose affinity chomatography. neg., negative control derived from bacteria transformed with the vector alone; M, molecular weight markers: 94.0 kDa, 67.0 kDa, 43.0 kDa, 31.0 kDa, 20.1 kDa, 14.4 kDa. 2c: Purification of C1₃₉₋₂₄₄-His using Ni-NTA-agarose affinity chomatography, analysed by SDS-PAGE followed by Coomassie brilliant blue staining. The bacterial lysate (L) was obtained by sonication at 5 h after induction of the culture with IPTG. F, flow-through fraction; W, flow-through fraction after washing; 50 to 500, fractions after elution with 50 to 500 mM imidazole solution; M, molecular weight markers (kDa). The fractions after elution with 120, 200 and 300 mM imidazole were pooled and used in further experiments (used fractions). The arrows indicate the positions of C1_39–244 or C1_39–244-His.

For the analysis of protein expression, 500 μl overnight cultures of E. coli carrying the appropriate expression plasmid, or pQE-60 as a negative control, was inoculated into 1.5 ml fresh medium and the culture was incubated for 30 min at 37°C. Thereafter, 40 μl of 100 mM isopropyl-beta-d-thiogalactopyranoside (IPTG) solution was added and the cultures were incubated for 5 h at 37°C. Cells were harvested by centrifugation and lysed by incubation in 200 μl buffer B (8 M urea, 0.1 M NaH₂PO₄, 0.01 M Tris, pH 8.0) for 5 min. Cellular debris was removed by centrifugation and the supernatants were stored at −20°C for sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis, or used for the purification of His-tailed proteins. In the latter case, 30 μl Ni-NTA-agarose (Qiagen) was added to the lysates, the mixture shaken for 30 min and the beads pelleted by centrifugation for 1 min at 17,000×g. After three washings each with 200 μl buffer C (8 M urea, 0.1 M NaH₂PO₄, 0.01 M Tris, pH 6.3), the agarose beads were dissolved in 20 μl buffer C containing 100 mM ethylenediamine tetraacetic acid and the solution incubated for 5 min at 37°C. The supernatants were then analysed by SDS-PAGE.

Expression of C1_39–244-His in E. coli and purification

Ten millilitres of overnight cultures of E. coli transformed with pC1_39–244-His was inoculated into 1 l fresh medium (Luria-Bertani medium supplemented with 100 μg ampicillin/ml). Cultures were incubated until the optical density at A ₆₀₀ reached 0.8 and expression was then induced with 1 mM IPTG. After 5 h, bacteria were harvested by centrifugation, resuspended in sonication buffer (10 mM Tris–HCl, pH 7.8, 50 mM KH₂PO₄, 300 mM NaCl, 10 mM beta-mercaptoethanol) and subsequently lysed by sonication on ice with six 30-sec bursts. The lysate was centrifuged at 10,000×g for 20 min, and the supernatant was used for the purification of the His-tailed proteins with Ni-NTA-agarose.

All purification steps were performed at room temperature. The Ni-NTA-agarose (4 ml) was added to a 20 ml column and equilibrated with 20 ml sonication buffer. The lysates were loaded onto the column, which was then washed with 20 ml washing buffer (40 mM Tris-HCl, pH 7.5, 20% glycerol, 100 mM KCl, 1 mM beta-mercaptoethanol) supplemented with 20 mM imidazole. The proteins were eluted in fractions of 4 ml containing 50 mM, 80 mM, 120 mM, 200 mM, 300 mM and 500 mM imidazole in washing buffer, respectively. Aliquots of each fraction were analysed by SDS-PAGE. Fractions containing high concentrations of the recombinant protein were pooled and the concentration of total protein was measured using a BCA protein assay (Pierce, Rockford, IL, USA). The purity and the apparent molecular weight of the recombinant proteins were determined by densitometric analysis of Coomassie brilliant blue-stained gels using Gel-Pro Analyzer software (Media Cybernetics, Silver Spring, MD, USA).

SDS-PAGE and immunoblotting

Proteins were analysed by SDS-PAGE and immunoblotting essentially as described previously (Stoll et al., 1993). Briefly, the samples were diluted in sample buffer (Laemmli, 1970), boiled for 5 min, and subjected to electrophoresis on 15% denaturating SDS-polyacrylamide gels. The low molecular weight markers calibration kit (Pharmacia, Erlangen, Germany) and biotinylated low molecular weight markers (Bio-Rad, München, Germany) were used as standards. Proteins were stained with Coomassie brilliant blue. For immunoblotting, proteins were transferred to nitrocellulose membranes. All of the dilutions were made in phosphate-buffered saline (PBS) supplemented with 0.1% Tween 20 (PBS-T). The membrane was incubated in a 10% solution of skim milk powder in PBS-T for 1 h, followed by two washings with PBS-T. For the analysis of multiple sera, the membrane was cut into strips, which were incubated in a 1:10 dilution of the sera for 1 h, followed by three washings in PBS-T. Chicken antibodies were detected by incubation with a 1:500 dilution of biotinylated anti-chicken IgG (Dianova, Hamburg, Germany) for 1 h. Psittacine antibodies were detected by incubation with a 1:100 dilution of a rabbit serum specific for IgG of an African gray parrot (Psittacus erithacus) (Grund et al., 2001) for 1 h, washing thrice and then incubation in a 1:1000 dilution of biotinylated anti-rabbit IgG (Sigma, München, Germany) for 1 h, followed by three washings in PBS-T. In both cases, the procedure was continued by incubation in a 1:2000 dilution of a streptavidin–horseradish peroxidase conjugate (Roche, Basel, Switzerland) for 30 min, two washings with PBS-T and one washing with PBS. Horseradish peroxidase activity was detected with 4-chloro-1-naphthol.

Immunization of chickens

Three 12-week-old White Leghorn chickens were inoculated intramuscularly with 500 μl volumes containing 50 μg purified C1_39–244-His emulsified in Specol (ID-DLO, Lelystad, The Netherlands), 50 μg purified C1_39–244-His without an adjuvant, or with washing buffer as negative control, respectively. One hundred days after the inoculation, the birds were booster immunized with the same amount of the inoculum used for the first immunization. Sera were collected from the birds at days 0, 7, 14, 28, 35, 100, 114 and 135 after the first inoculation.

Psittacine sera

A total of 11 psittacine sera were randomly selected from a collection of sera obtained from birds with feather disorders and found not to contain antibody against avian polyomavirus, in an antibody blocking enzyme-linked immunosorbent assay (ELISA) (Khan et al., 2000). The sera were derived from the following psittacine species: Poicephalus rueppelli (bird 1), P. erithacus (birds 2 and 9), Calyptorhynchus banksii (bird 3), Andorhynchus leari (birds 4 and 8), Amazona leucocephala (bird 5), Ara ararauna (birds 6 and 7), and C. galerita (birds 10 and 11). No information was available about the BFDV status of these birds. All of the sera were inactivated by incubation for 30 min at 56°C prior to serological analysis.

Enzyme-linked immunosorbent assay

An indirect ELISA was established for the detection of BFDV-specific antibodies in parrot sera. The wells of Nunc MaxiSorb ELISA plates were coated with 50 μl volumes containing purified C1_39–244-His (10 μg/ml) in coating buffer (0.1 M sodium carbonate/bicarbonate, pH 9.6) for 1 h at 37°C. Plates serving as negative controls were processed in parallel, with coating buffer without any protein used for this step. The wells were then washed five times with PBS-T and blocked with 200 μl volumes of 10% skim milk powder in PBS-T for 1 h at 37°C. The blocking solution was removed and the wells were washed twice with PBS-T. The parrot sera were diluted serially in two-fold steps in PBS-T, starting with 1:10 dilutions, and 100 μl volumes were added to wells and the plates incubated for 1 h at 37°C. After three washings with PBS-T, 100 μl volumes of a rabbit serum specific for IgG of an African gray parrot (P. erithacus) (Grund et al., 2001), diluted 1:100 in PBS-T, added and plates incubated for 1 h at 37°C. The plates were washed thrice and 100 μl of a 1:1000 dilution of biotinylated anti-rabbit IgG (Sigma) in PBS-T was added to each well. The plates were incubated for 1 h at 37°C. After three further washings in PBS-T, 100 μl of a 1:2000 dilution of a streptavidin–horseradish peroxidase conjugate (Roche) in PBS-T was added to each well and plates were incubated for 30 min, followed by two washings with PBS-T and one washing with PBS. Finally, a substrate solution containing o-phenylenediamine hydrochloride was added, with colour development stopped after 5 min using 2 M H₂SO₄, and the optical densities were measured at 490 nm (OD₄₉₀), with a reference wavelength of 650 nm. The titre was defined as the reciprocal of the highest dilution of the serum that showed a two-fold higher OD₄₉₀ on the plate coated with C1_39–244-His compared with the OD₄₉₀ on the plate coated without antigen.

Haemagglutination inhibition assay

Red blood cells (RBC) were derived from a sulphur-crested cockatoo (C. galerita) previously found to be BFDV-negative by PCR (Raue et al., 2004). The RBC were sedimented by low-speed centrifugation of a blood sample containing ethylenediamine tetraacetic acid and were washed twice with PBS. For stabilization of the RBC solution, the cells were treated with 0.2% glutaraldehyde for 30 min at room temperature, followed by five washings with PBS. The RBC were stored as a 10% suspension in PBS at 4°C for several weeks without showing haemolysis.

A crude virus preparation was extracted from the liver and spleen of a budgerigar, in which BFDV had previously been detected by PCR (Raue et al., 2004), by sonication of these organs in PBS and removal of cellular debris by centrifugation. The supernatant was tested in a haemagglutination assay. Wells of a U-shaped microtitre plate (Greiner, Frickenhausen, Germany) were inoculated with 200 μl of 10% skim milk powder solution in PBS and the plate incubated for 30 min to prevent non-specific binding of RBC to the plate. The plate was then washed twice with PBS and 25 μl volumes of two-fold dilutions of the virus preparation were added. Thereafter, 25 μl volumes of 1% RBC solution in PBS were added. After 20 min of incubation at room temperature, the titre was determined as the reciprocal of highest dilution of the virus suspension showing clear haemagglutination.

For the HI assay, the parrot sera were incubated with a 1% RBC solution for 30 min at room temperature. The RBC were removed by centrifugation and 25 μl volumes of two-fold dilutions of the sera were added to the microtitre plate, which had been pretreated as already described. A solution containing 4 haemagglutinating units of the virus preparation in 25 μl PBS was added to each well and plates incubated for 1 h at 37°C. Thereafter, 25 μl of 1% RBC suspension in PBS was added to each well and plates incubated for 20 min at room temperature. The HI titre was defined as the reciprocal of the highest serum dilution that completely inhibited haemagglutination.

Results

Cloning of the C1 gene and sequence analysis

The sequence encoding C1 of BFDV was amplified by PCR using the primers C1-s and C1-as, and DNA extracted from the feather of a BFDV-infected sulphur-crested cockatoo (C. galerita) as a template. The amplified PCR product was cloned into the procaryotic expression vector pQE-60. Two clones were selected and the plasmid insert sequenced. The nucleotide sequences of the insert in both clones were identical. The resulting plasmid was designated pC1.

The amino acid sequence deduced from the cloned sequence in pC1 was aligned with the C1 sequences of 10 published BFDV strains. In the phylogenetic tree, two distinct clusters were observed: two BFDV strains (AF311300 and AF311301) form a separate cluster, while the remaining isolates, including pC1, group into a second cluster. However, a high level of sequence diversity was observed in this second cluster, indicating variability of the capsid protein between the different BFDV isolates. The amino acid identity between the pC1 sequence and the published sequences ranged between 76.3 and 83.2%. An alignment based on the sequence of the Rep protein of the 10 published sequences revealed amino acid identities ranging between 86.9 and 98.3% (data not shown).

Expression of C1 protein variants in E. coli

Proteins from cultures of E. coli transformed with the plasmid pC1 were analysed at 5 h after induction using a denaturing buffer for cell lysis and SDS-PAGE. However, Coomassie brilliant blue staining revealed no additional protein bands compared with lysates of E. coli transformed with the expression vector pQE-60 alone (Figure 2b, left).

The analysis of the C1 nucleotide sequence revealed an accumulation of arginine codons at the 5′ end of the gene, including one AGG and five AGA codons, known to be the least used codons in E. coli. As an accumulation of these codons is known to inhibit efficient mRNA translation in E. coli (Varenne & Lazdunsky, 1986; Alexandrova et al., 1995), a deletion of the codons for the amino-terminal residues from positions 1 to 38 was introduced in plasmid C1_39–244 (Figure 2a). When cultures transformed with this plasmid were induced and analysed by SDS-PAGE and Coomassie brilliant blue staining, an additional protein band with an apparent molecular mass of 24 kDa was seen (figure 2b, left).

To facilitate concentration and efficient purification of C1 and C1_39–244, codons for carboxy-terminal polyhistidine tails were added in the plasmids pC1-His and pC1_39–244-His (Figure 2a). After induction of the cultures transformed with these plasmids, cells were lysed under denaturing conditions and His-tailed proteins were purified using Ni-NTA-agarose affinity chromatography. When the proteins were analysed by SDS-PAGE and Coomassie brilliant blue staining, an additional protein band was only detected in the case of C1_39–244-His. This protein had an apparent molecular weight of 25 kDa (Figure 2b, right).

Non-denaturing purification of C1_39–244-His

A protocol was developed for the expression and purification of C1_39–244-His under non-denaturing conditions. As demonstrated by SDS-PAGE and Coomassie brilliant blue staining (Figure 2c), the highest amounts of the C1_39–244-His were detected in fractions eluted with 120, 200 and 300 mM imidazole. Densitometric analysis of Coomassie brilliant blue-stained gels of the pooled fractions suggested that the preparation was 95% pure and the protein concentration was 0.1 mg/ml.

Immunogenicity of C1_39–244-His

The ability of C1_39–244-His to induce specific antibodies was analysed by immunization of chickens. By immunoblotting using purified C1_39–244-His as antigen and sera derived from different time-points after inoculation, seroconversion was detected in the birds immunized with C1_39–244-His, but not in the negative control (not shown). The chicken immunized with adjuvanted C1_39–244-His showed the greatest response (Figure 3a). In this chicken, antibodies specific for C1_39–244-His were first detected 7 days after inoculation.

Figure 3. Immunoblot analysis of sera collected from chickens immunized with C1_39–244-His. 3a: Analysis of reactivity with C1_39–244-His using sera obtained at 0, 7, 14, 28, 35, 100, 114 and 135 days after the first inoculation of the protein. An additional booster immunization was performed at day 100. 3b: Analysis of reactivity with supernatant of an organ suspension prepared from a BFDV-infected Rueppels parrot (P. rueppelli) and concentrated by ultracentrifugation (lane 1), crude organ suspensions of two BFDV-infected budgerigars (lanes 2 and 3), or C1_39–244-His (lane 4). Immunoblot analysis was performed using the chicken sera obtained before immunization (left) or 114 days after immunization with C1_39–244-His (right). Lane M, molecular weight markers: 97.4 kDa, 67.2 kDa, 45.0 kDa, 31.0 kDa, 21.5 kDa, 14.4 kDa. Arrows indicate the positions of C1 and C1_39–244-His. Additional faint bands in lanes 2 and 3, observed after using the chicken sera obtained 114 days after immunization, are indicated by white arrows.

The reactivity of the sera with BFDV was tested by immunoblotting antigen preparations derived from BFDV-infected birds (Figure 3b). One preparation from the skin of a BFDV-infected Rueppel's parrot (P. rueppelli) and preparations from the liver, spleen and kidney of two BFDV-infected budgerigars (Melopsittacus undulatus) were analysed. The preparation from the skin of the BFDV-infected Rueppel's parrot was further concentrated by ultracentrifugation of the virus in the supernatant of the organ homogenate. Using the serum collected at day 114 post-inoculation (Figure 3b, right), a prominent band with an apparent molecular weight of 27 kDa detected in the preparation from the BFDV-infected Ruppell's parrot, probably representing the C1 protein (lane 1). This serum also detected faint bands with an apparent molecular weight of 27 kDa and bands in closed position to that in the crude organ preparations from the BFDV-infected budgerigars (lanes 2 and 3). These were not detected by the serum collected from the chicken on day 0 (Figure 3b, left). However, many additional bands, considered non-specific, were detected by both sera.

Testing psittacine sera using purified C1_39–244-His

Eleven psittacine sera were tested for their reactivity with BFDV and the purified C1_39–244-His. A HI assay was established to demonstrate the presence of BFDV-specific antibodies. As shown in Figure 4a, four sera had HI titres of 8 (birds 2, 6, 7 and 11), four sera had HI titres of 4 (birds 4, 5, 8 and 9), one serum had an HI titre of 2 (bird 3) and two sera did not inhibit haemagglutination (birds 1 and 10).

Figure 4. Analysis of psittacine sera. 4a: Results of a HI assay using 4 haemagglutinating units of an organ homogenate prepared from a BFDV-infected budgerigar. 4b: ELISA using purified C1_39–244-His as antigen. A serum dilution of 1:160 is shown. 4c: Immunoblot analysis using purified C1_39–244-His as antigen. The arrow indicates the position of C1_39–244-His.

In an ELISA with purified C1_39–244-His as antigen (Figure 4b and Table 1), titres were 640 for one serum (bird 11), 160 for two sera (birds 2 and 7), 80 for four sera (birds 3, 4, 8 and 9) and 40 for four sera (birds 1, 5, 6 and 10). By immunoblotting using C1_39–244-His as antigen (Figure 4c and Table 1), a clearly visible band was detected with four sera (birds 2, 6, 7 and 11), a faint band with two sera (birds 1 and 8), and no distinct band with the remaining five sera (birds 3, 4, 5, 9 and 10). It was evident that a correlation existed between the HI titre, the ELISA titre and the intensity of the band in immunoblots. The correlation coefficient for the relationship between the HI and ELISA titres was 0.52, and that for the relationship between HI titres and the immunoblot band intensities (where a clearly visible band was scored as 2, a faint band as 1 and no band as 0) was 0.86. However, it has to be considered that only 11 samples were analysed.

Table 1. Comparison of HI titres, ELISA titres and immunoblot reactivity obtained for 11 randomly selected psittacine sera

Serum (bird number)	HI titre (scoring^a)	ELISA titre (scoring^b)	Immunoblot reactivity (scoring^c)
1	<1:2 (−)	1:40 (−)	+/− (?)
2	1:8 (+)	1:160 (+)	+ (+)
3	1:2 (−)	1:80 (−)	− (−)
4	1:4 (−)	1:80 (−)	− (−)
5	1:4 (−)	1:40 (−)	− (−)
6	1:8 (+)	1:40 (−)	+ (+)
7	1:8 (+)	1:160 (+)	+ (+)
8	1:4 (−)	1:80 (−)	+/− (?)
9	1:4 (−)	1:80 (−)	− (−)
10	<1:2 (−)	1:40 (−)	− (−)
11	1:8 (+)	1:640 (+)	+ (+)
^a HI titres ≥1:8 are scored positive (+), lower titres are scored negative (−).
^b ELISA titres ≥1:160 are scored positive (+), lower titres are scored negative (−).
^c A clearly visible band (+) is scored positive (+), a faint band (+/−) is scored questionable (?), no band is scored negative (−).

When criteria for scoring a serum positive were defined for each test (Table 1), a titre ≥1:8 for the HI assay, a titre ≥1:160 for the ELISA and a clearly visible band in an immunoblot, three of the sera were regarded positive in all tests (birds 2, 7 and 11) and five sera were scored negative in all tests (birds 3, 4, 5, 9 and 10). Sera of birds 1 and 8 were regarded negative in the HI assay and the ELISA, but a faint band was detected by immunoblotting in both cases. The serum of bird 6 was regarded positive in the HI assay and by immunoblotting, but it was scored negative in the ELISA.

Discussion

Beak and feather disease is one of the most important viral diseases of psittacine birds and has spread throughout the world due to the international trade in these birds. As no natural reservoir of BFDV is known in Europe, successful elimination should be possible in this region, provided that sensitive and reliable diagnostic tests are available. However, the development of diagnostic tests, as well as vaccines, is restrained by the inability to propagate the virus in tissue or cell cultures or embryonated chicken eggs. Therefore, recombinant antigens, as described here, are needed.

As C1 is considered to be the major capsid protein of BFDV, this protein is likely to induce an antibody response in BFDV-infected birds with titres sufficient for reliable serological tests. Serological tests based on recombinant C1 should be a useful tool for assessing the BFDV status of individual birds, as well as for epidemiological surveys. It is well known, however, that individual BFDV isolates differ significantly within the C1 gene (Bassami et al., 2001; Raue et al., 2004), which could limit the broad applicability of a serological test. The C1 gene used here had 76.3 to 83.2% amino acid identity with 10 other C1 sequences. It grouped with eight other C1 sequences amplified from different psittacine species including cockatoos, a lovebird, a lorikeet and a galah, and there was no indication of species specificity of BFDV strains. The variability of C1 seems to be higher than that of the Rep protein. However, the cross-reactivity of antibodies induced by the different BFDV strains, as well as the immune response to the Rep protein, need to be investigated further.

After induction of expression in E. coli carrying the C1 gene, no recombinant protein could be detected. An accumulation of arginine residues in the amino-terminal region of C1 could have been responsible for the low expression rate. After deletion of the first 38 amino acids, the protein was readily expressed with relatively high efficiency. The function of the amino-terminal region of C1 is not known. It could be speculated, however, that the accumulation of positively charged amino acids could be involved in packaging of viral DNA into the capsid, as has been proposed for corresponding regions within the capsid proteins of CAV and porcine circoviruses (Todd et al., 2000). If so, the amino-terminus of C1 would be located within the capsid, and thus may not contribute to the protective antibody response.

The truncated protein C1_39–244-His was easily purified using nickel affinity chromatography. After immunization of chickens, the protein induced specific antibodies that detected a protein with an apparent molecular weight of 27 kDa in organ homogenates of BFDV-infected birds. As C1 has a predicted molecular weight of 28.8 kDa, and a protein with an apparent molecular weight of approximately 26 kDa has been repeatedly detected in purified BFDV preparations (Ritchie et al., 1989, 1990), this band was considered to represent C1 of BFDV.

An ELISA and an immunoblotting assay using C1_39–244-His as antigen were developed for the detection of BFDV-specific antibodies in psittacine sera. The results of both assays were compared with those from an HI assay by simultaneously testing 11 randomly selected psittacine sera. The results of the tests correlated relatively well, indicating specific detection of antibodies against BFDV by ELISA and immunoblot. It has to be taken into account, however, that no ‘gold standard’ exists for the serological detection of BFDV infection. The only test in current use is the HI assay, with serious problems in standardization and interpretation of the results. Different results have been obtained using erythrocytes collected from different species and even with erythrocytes from different individuals of the same species (Ritchie et al., 1991; Sexton et al., 1994; Sanada & Sanada, 2000). Controlled infection experiments, as well as testing of a larger number of field sera, are required to assess the specificity and sensitivity of the tests. The applicability of recombinant C1 proteins as antigens in vaccines against the severe disease caused by BFDV will also be investigated in further studies.

Acknowledgements

The authors thank Marcellus Bürkle (Loro Parque Tenerife) and Birgit Ackermann (Labordiagnostik Leipzig) for providing the organ and blood samples from healthy, diseased and psittacine birds.