Psittacine beak and feather disease (PBFD) is a specific disease in psittacine birds caused by beak and feather disease virus (BFDV), a member of genus Circovirus [1]. Clinical form of PBFD included chronic, progressive loss of feathers and, in some species, deformities of the beak and claws [2]. Acute-form, which shows signs of septicemia accompanied by pneumonia, enteritis, rapid weight loss, and death, has also been described [2]. BFDV carries a single-stranded circular DNA with a complete genome size of approximately 2.0 kb. The genome contains two major open reading frames (ORFs), encoding the replication-associated protein (V1), and the capsid protein (C1). Another ORF (ORF V2) has been also described, but it is unclear what role the transcriptional product of this ORF may play in the cycle of the virus [1].

PBFD has since been reported to affect more than 60 psittacine species; it is highly probable that all psittacine bird species are susceptible [3]. So far, phylogenetic analysis of BFDV has revealed an apparent genotypic association with specific psittacine species [4, 5]. The level of genetic diversity has been reported to be similar among several countries such as Australia, New Zealand, and South Africa [5–7].

In the present study, 19 BFDV isolates derived from budgerigars (Melopsittacus undulatus) showing PBFD were detected by polymerase chain reactions (PCR), and of which six isolates were determined whole genome sequence. It was shown that 19 BFDV isolates in this study separated into two lineages, in addition, one was unique lineage to be reported previously.

Twenty samples including 13 feathers, 4 livers, and 3 bloods from 19 budgerigars in Japan suspected to be PBFD were examined in this study (Table 1). DNA was extracted from the blood or feathers with a SepaGene nucleic acid extraction kit (Sanko Junyaku Co., Japan) as described previously [8] and stored at −30°C until use. PCR was carried out using two sets of primer, Primer2/Primer4 (Ypelaar’s PCR) and PBFDdupF/PBFDdupR (Ogawa’s PCR) targeting BFDV genomic DNA previously described [8, 9]. As result of PCR, BFDV genomic DNA was detected in all 19 samples using primer set of PBFDdupF and PBFDdupR, whereas BFDV genomic DNA was amplified in only two samples using Primer2 and Primer4 (Table 1). These isolates were tentatively classified as budgerigar BFDV genotype 1, which was detectable by Ypelaar’s PCR, and genotype 2, which was not detectable.

Table 1 Birds used in this study and genotypes of the budgerigar BFDV
Full size table

Next, we performed whole genome sequence analysis of these BFDV isolates, 4 and 2, respectively, were from acute-form and chronic-form, to determine their genetic characteristics. Whole genome sequences of six budgerigar BFDVs, of which 2 (MU-JP1P and MU-JP2P) and 4 (MU-JP3P to MU-JP6P) were genotype 1 and 2, were determined. Four fragments covering the whole genome were amplified by PCR using four primers sets for each genotype (Supplementary Table S1). PCR products were cloned and sequenced as described previously [8]. The obtained sequences were edited using Genetyx-Mac version 13 (Genetyx Co., Japan) and Genetyx-Mac/ATSQ version 4.2.4 (Genetyx Co.). The sequences of six budgerigar BFDV isolates have been submitted to GenBank and have been given accession numbers AB277746–AB277751. The genome sizes ranged from 1996 to 2004 nt (MU-JP1P, 1996 nt; MU-JP2P, 1996 nt; MU-JP3P, 2001 nt; MU-JP4P, 2004 nt; MU-JP5P, 2004 nt, and MU-JP6P, 2002 nt) (Supplementary Table S2). The identities of whole genome sequences of budgerigar BFDV genotype 1 and genotype 2 compared with the 23 published BFDV isolates shown in supplementary Table S3 from 87 to 92% and from 83 to 92%, respectively. Budgerigar BFDV genotype 1 whole genome sequences were from 85 to 88% identical to genotype 2 sequences. The identities of deduced amino acid sequences of budgerigar BFDV genotype 1 and genotype 2 ORF V1 compared to 23 published BFDVs varied from 90 to 99% and from 87 to 93%, respectively. Those of genotype 1 and genotype 2 ORF C1 varied from 75 to 91% and from 74 to 85%, respectively. The deduced amino acid sequences of ORF V1 and C1 of budgerigar BFDV genotype 1 were from 91 to 94% and from 84 to 88% identical to those of genotype 2, respectively.

The deduced amino acid sequences of partial ORF V1 of budgerigar BFDV genotype 1 showed 96.7–99.4% homologies to the sequences of the published sequences derived from budgerigars (BG3-NZ, AY148301; UFS 3 to UFS 5; DQ384623 to DQ384625), whereas genotype 2 showed 91.8–94.6%. A phylogenic tree based on partial ORF VP1 using the UPGMA method revealed that genotype 1 isolates clustered together with BG3-NZ and UFS 3 to UFS 5, which classified into the BG lineage [7, 10], whereas genotype 2 isolates clustered a unique group (data not shown). These results suggested that budgerigar BFDV genotype 2 might be phylogenically distinct from previously described BFDV.

There were two clinical forms of the affected birds with budgerigar BFDV genotype 1, one was acute-form and another was chronic-form. Of the 17 budgerigar BFDV genotype 2 isolates, 3 forms were acute and the other 14 were chronic. No relationship between clinical forms and both genotypes was observed in the present study.

To confirm the further taxonomic position of these six budgerigar BFDV isolates, phylogenic trees were constructed based on the whole genome sequence analysis. The budgerigar BFDV genotype 1 and genotype 2 were located in independently different clusters (Fig. 1a). The genotype 2 isolates were particularly distinct from other BFDV isolates and clustered a unique lineage. This relationship was supported by analysis of the both ORF V1 and C1, respectively (Fig. 1b, c). The phylogenic analysis of whole genome sequence and both ORF V1 and C1 shows identical placement of the previously described BFDVs and the budgerigar BFDV genotype 2. Hereby, it is proposed that budgerigar BFDV genotype 2 belongs to a novel lineage (BG2). During review of this article, three budgerigar BFDV sequences were reported as a unique isolate in South Africa [11]. However, these isolates were also classified into the BG lineage, which was related to genotype 1, and not clustered a novel unique BG2 lineage.

Fig. 1
figure 1

Neighbor-joining tree of a complete genome sequence, amino acid sequences of b ORF V1 and c C1 of six budgerigar BFDVs and 23 BFDVs [1, 4–6]. The bootstrap values of each node were calculated using 1,000 replications. Bold fonts indicate budgerigar BFDV from the present study. Genbank accession numbers we used are AB277746–AB277751 in six budgerigar BFDV isolates, and AF311295–AF311302, AF080560, AF071878, AY521234–AY521238, AY450434, AY450436–AY450439 and AY450441–AY450443 in 23 non-budgerigar BFDV isolates, respectively. a Six budgerigar BFDV genome sequences formed different branches from 23 BFDVs, and budgerigar BFDV genotype 1 and genotype 2 were also in independent clusters. b and c Budgerigar BFDV genotype 1 was closely related to 23 BFDVs in the both trees, whereas genotype 2 was distantly related to them

Full size image

Ypelaar’s PCR targeting ORF V1 has been used as a universal method by other researchers [3, 6, 7]. However, it was reported that Ypelaar’s PCR might not detect all isolates because of the diversity of BFDV genotypes [3]. In fact, our PCR results showed the existence of budgerigar BFDV, which could not be detected by Ypelaar’s PCR. To investigate why the budgerigar BFDV genotype 2 was not detected by Ypelaar’s PCR, the sequences targeted by Primer2 and Primer4 were compared on the genomic sequences of both genotype 1 and 2. The sequences of genotype 1 were identical to both primer sequences and those of genotype 2 were identical to Primer4, whereas 6–7 of the 18 nucleotides were different in the sequences of genotype 2, which correspond to the central region of Primer2 (Fig. 2). In the case of Ogawa’s PCR, several nucleotide substitutions were scattered in budgerigar BFDV genotype 1 and genotype 2 sequences. However, consecutive nucleotide substitutions such as the genotype 2 in Primer2 were not found. The phylogenic classification of the six budgerigar BFDV genomes was the same as the tentative classification found in both PCR amplification results. It is conceivable that the combination of the results of both PCRs would be useful to identify the budgerigar BFDV genotypes. Therefore, it is possible that additional uncovered BFDV exist in other psittacine species.

Fig. 2
figure 2

Sequence comparison of the primer target sites of Ypelaar’s PCR [9] and Ogawa’s PCR [8] in budgerigar BFDV genotype 1 and genotype 2 sequences. Dots indicate positions in agreement with the primer sequences. The numbers to the left and right of the row following MU-JP1P indicate nucleotide positions of the MU-JP1P sequence. Asterisks indicate that corresponding positions of alignment are identical. 6–7 of 18 nucleotides were different in budgerigar BFDV genotype 2 sequences which correspond to the above Primer2 sequences. In PBFDdupF and PBFDdupR, several nucleotide substitutions were scattered in budgerigar BFDV genotype 1 and genotype 2 sequences

Full size image

In conclusion, this study revealed the presence of a novel lineage of BFDV in budgerigars, with further investigation involving the identification of any antigenic differences present.