| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Short Communication |
1 Department of Zoology, The University of Hong Kong, Hong Kong SAR
2 Department of Zoology, The University of Oxford, Oxford OX1 3PS, UK
Correspondence
Frederick Chi-Ching Leung
fcleung{at}hkucc.hku.hk
 |
ABSTRACT |
|---|
 TOP ABSTRACT MAIN TEXTREFERENCES |
|---|
The GenBank/EMBL/DDBJ accession numbers for the sequences determined in this study are DQ997815–DQ997817.
Supplementary material is available with the online version of this paper.
|
MAIN TEXT |
|---|
|
TOPABSTRACTMAIN TEXTREFERENCES |
|---|
). PCV was classified into PCV type 1 (PCV1) and PCV type 2 (PCV2) based on their genotypes. Serological tests suggest that PCV1 is widely found in pigs, but no evidence has associated PCV1 to animal diseases (Tischer et al., 1987). However, PCV2 is now considered as the causative agent of post-weaning multisystemic wasting syndrome (PMWS), which was first identified in Canada during the early 1990s (Harding, 2004). PMWS has become a major cause of wasting disease in swine-producing areas including Asia, North America and Europe (Chae, 2004; Liu et al., 2002; Mankertz et al., 2000; Wen et al., 2005).
Point mutation and recombination are major forces of viral evolution. Analysis of the genomes of worldwide PCV2 showed a very high nucleotide sequence identity (>90 %) among them (Fenaux et al., 2000; Hamel et al., 1998). Nevertheless, evidence of natural recombination has been reported in members of the family Circoviridae, including beak and feather disease virus (Heath et al., 2004) and Torque teno virus (Manni et al., 2002), implying that recombination may also contribute to the genetic variations within the family Circoviridae. Recently, natural recombination in PCV2 has been proposed in Hungarian wild boars (Csagola et al., 2006) and by analysis of worldwide PCV2 GenBank sequences (Olvera et al., 2007), although further phylogenetic evidence is needed to validate these hypotheses. Herein, we present a comprehensive analysis of complete PCV2 genomic sequences from Hong Kong, providing strong evidence of natural recombination among PCV2 of different lineages.
The complete genomic sequences of three PCV2 strains from Hong Kong, namely HKS02-04, HKS03-04 and HKS091-04 (collectively designated HKS04 viruses), with GenBank accession nos DQ997815, DQ997816 and DQ997817, were reported. These viruses were found in pigs showing PMWS symptoms from three different herds during 2004. Viral DNA was extracted from 500 µl serum using TRI reagent (Invitrogen) according to the manufacturer’s instructions. The DNA pellet was then dissolved in 300 µl 8 mM NaOH and subjected to PCR amplification with Platinum High Fidelity Taq DNA polymerase (Invitrogen) according to the manufacturer’s instructions. The complete genomes were amplified using primers F-PCVSAC2 and R-PCVSAC2 (Fenaux et al., 2002), and the amplicons were cloned into pCR2.1-TOPO (Invitrogen). To avoid misleading results caused by PCR artefacts, five random clones were sequenced for each of the viruses, using M13 universal forward and reverse primers, as well as two internal primers, CV1 and CV4 (Fenaux et al., 2000). DNA sequencing was performed using a BigDye Terminator cycle sequencing kit (Applied Biosystems) according to the manufacturer’s instructions. A limited number of non-recurrent mutations were found among different clones of the same virus, which are thought to be PCR-generated sequence mutations. Therefore, the genomic sequences of each virus were represented by the majority consensus of the five clones. The complete genomes of HKS04 viruses were found to be 1767 bp long and shared over 99 % in nucleotide sequence identity.
To investigate the phylogenetic origin of HKS04 viruses, all available complete PCV2 genomic sequences were downloaded from GenBank. Sequences were screened to exclude patent, artificial, defective and potential PCV2 recombinants as proposed in Hungary (Csagola et al., 2006). The complete genomic sequences (n=164), as shown in Supplementary Table S1 (available in JGV Online), were aligned using CLUSTAL W with one gap column removed (Thompson et al., 1994). A maximum-likelihood (ML) phylogeny of 500 bootstrap replicates was constructed based on the complete genomic sequences of worldwide PCV2 strains using PHYML (http://atgc.lirmm.fr/phyml/) under the general time-reversible nucleotide substitution model (GTR model) (Yang, 1994) with estimated gamma and invariable-site values, and this optimal model was selected using MODELTEST (Posada & Crandall, 1998). Seven well-supported major lineages (bootstrap values >80 %) were identified and designated lineages A, B, C, D, E, F and G, respectively (Fig. 1). The mean Kimura’s two-parameter (K2P) distance of all taxa is 0.03, while the intra-lineage K2P distance has a mean of 0.01, ranging from 0.008 to 0.015. Preliminary analyses of different genomic regions of HKS04 viruses showed discordant phylogenetic relationships with other PCV2 strains. For instance, HKS04 viruses clustered within lineage A based on complete genome and ORF2 sequences, but they clustered out of lineage A when ORF1 sequences were considered (data not shown). These conflicting phylogenies implied the possible recombinant origin of HKS04 viruses.
|
). Involvement of large number of sequences in the RDP analysis results in noisy recombination signals. As the intra-lineage genetic variations are relatively low, i.e. mean pairwise K2P distance ranging from 0.008 to 0.015, the RDP analysis was done using phylogenetically distant strains (n=33) selected from different lineages. Their details are shown in Supplementary Table S2, available in JGV Online. With the support of several recombination detection methods implemented in RDP, including Geneconv, RDP and MaxChi methods (cut-off P value of 0.001 with Bonferroni correction), parental strains of HKS04 viruses possibly come from lineage A and lineage E or F. It is noted that the mean genetic distances between lineage A/lineage E and lineage A/lineage F are 0.042 and 0.045, respectively. The specific settings used in each method in RDP are available upon request. To further investigate this potential recombination event, a recombination detection method called bootscanning analysis, implemented in SIMPLOT (Lole et al., 1999), was performed on the dataset. Briefly, a set of neighbour-joining (NJ) phylogenies were generated to show the clustering of the query sequences (potential recombinants) and reference sequences (parents) in a moving window along the alignment with user-defined number of bootstrap replications. In particular, we performed a bootscanning analysis (Fig. 2a) on four groups of complete PCV2 genomic sequences, including a group of potential recombinants, i.e. HKS04 viruses (n=3), two groups of reference sequences from lineage A (P1, n=3, GenBank accession nos AY682990
[GenBank]
, AY691679
[GenBank]
and AY732494
[GenBank]
) and lineage F (P2, n=3, accession nos AY424401
[GenBank]
, AY424402
[GenBank]
and AY424403
[GenBank]
), and an outgroup from lineage D (O2, n=2, accession nos AF166528
[GenBank]
and AY146991
[GenBank]
). The bootscanning analysis was performed under the K2P model for 500 replications, with a window size of 400 bp, step size of 10 bp and a transition/transversion ratio of 2. The result of bootscanning analysis suggested discordances in the phylogenetic origins of the HKS04 viruses, supporting the hypothesis that HKS04 viruses might be recombinants from different PCV2 lineages.
|
). In brief, an alignment of sequences of two potential parents and a potential recombinant was divided by a partition (breakpoint), and two separated ML phylogenies were constructed for each of the two regions. Likelihood score of this ‘recombination model’ with a particular breakpoint was combined from the likelihood scores of the two ML phylogenies. This procedure was repeated for all possible breakpoints, which is implemented in likelihood analysis of recombination in DNA (LARD) (Holmes et al., 1999). The breakpoint with the highest combined likelihood score was then considered to be the most likely breakpoint. To assess whether the recombination model gave a significantly better fit to the data than the null hypothesis of no recombination, the combined likelihood score of the best ‘recombination model’ was compared with the likelihood of ML phylogeny of the unbroken alignment, i.e. null hypothesis of no recombination, using the likelihood ratio test (LRT) (Holmes et al., 1999). Simultaneous estimation of multiple breakpoints, e.g. two breakpoints in this case, is computationally intensive. Therefore, the complete genome alignment was divided into two overlapping alignments, and only one breakpoint was estimated for each of the trimmed alignments. The trimmed alignments are available upon request. Due to the fact that HKS04 viruses share over 99 % sequence identity, only the LARD results of HKS02-04 were shown for simplicity. Breakpoints of HKS02-04 were estimated to be at nt 1737 and 391, designated breakpoint A and B, respectively (Fig. 2b
), the estimated breakpoints in other HKS04 viruses were similar (P<0.005 in LRT of all cases, data not shown). All nucleotide positions in this manuscript were assigned according to strain NL_PMWS_4 (GenBank accession no. AY484416
[GenBank]
).
To further assess the statistical significance of these putative recombination events, the likelihood ratios of our datasets were evaluated against the null distributions of likelihood ratios of the 1000 simulated datasets assuming no recombination (Holmes et al., 1999). The simulated datasets were generated using Seq-Gen (Rambaut & Grassly, 1997), with the ML model parameters and sequence lengths from the corresponding real dataset. The likelihood ratios of the real and simulated datasets were generated using the breakpoint analysis in LARD as described. The likelihood ratios for all of putative breakpoints in HKS04 viruses were greater than any of the likelihood ratios of the corresponding simulated datasets (Supplementary Fig. S1, available in JGV Online), implying that the discordant phylogenetic relationships for different genomic regions of HKS04 viruses were unlikely to be the result of chance.
To investigate the phylogenetic origins of the parents of HKS04 viruses, two ML phylogenies were constructed based on the major and minor parental regions of 33 representative PCV2 strains (Fig. 3). The phylogenies were constructed with PHYML under the GTR model with 500 bootstrap replications. In the phylogeny based on the major parental region, HKS04 viruses clustered within lineage A with high bootstrap support. On the other hand, in the phylogeny based on the minor parental region, HKS04 viruses clustered with PCV2 from lineage E and lineage F with high bootstrap support, despite the general low bootstrap support of the internal nodes in this phylogeny. In addition, three PCV2 strains (GenBank accession nos AY691169
[GenBank]
, DQ180393
[GenBank]
and DQ195679
[GenBank]
) were shown to have a recombination pattern and phylogenetic origins similar to those of the HKS04 viruses (Fig. 3). Taken together, the results above suggest that HKS04 viruses might have resulted from a recombination event between parental PCV2 that were phylogenetically close to lineage A and lineage E or F, respectively.
|
). Circoviruses were further proposed to have resulted from recombination between a plant nanovirus and a vertebrate-infecting virus, e.g. a single-strand RNA calicivirus, during a host-switch event (Gibbs & Weiller, 1999). Moreover, extensive natural recombinations between strains of another member of the family Circoviridae (beak and feather disease virus) have been documented (Heath et al., 2004). These results suggested the possibility of natural recombination in other circoviruses. In the present study, the recombination breakpoints were predicted to be located within the ori and rep gene (ORF1). Within the ori of PCV, there is a stem–loop structure flanked by a pair of palindromic sequences. It has been proposed that PCV1 replicated via a rolling-circle melting-pot replication model. This provided insights into the illegitimate recombination of any circular DNAs with an ori-flanking palindrome (Cheung, 2004). Whether the chance of recombination is enhanced with the ori-flanking palindrome remains to be determined.
Until now, there is still no conclusive correlation of PCV2 strains to particular PCV2-associated diseases such as PMWS and porcine dermatitis and nephropathy syndrome (Larochelle et al., 2002). It has been suggested that all PCV2 strains belong to a single pathogenic genotype (Chae, 2005), and that genetic variations of PCV2 might be associated with geographical origins (Fenaux et al., 2000; Meehan et al., 2001). Based on our analysis of the worldwide PCV2 complete genomes, correlation of the five major lineages with particular geographical origins was not observed. Interestingly, our results suggested that recombination event(s) happen in PCV2 natural populations. In particular, the common ancestor of HKS04 viruses in this study resulted from a recombination event between a major parent of lineage A and a minor parent of lineage E or F. Our result is unlikely to be a laboratory artefact because three HKS04 viruses were cloned and sequenced individually at different time periods. Furthermore, several PCV2 strains from different provinces of China (GenBank accession nos AY691169
[GenBank]
, DQ180393
[GenBank]
and DQ195679
[GenBank]
) were shown to have a similar recombination pattern. As shown in Fig. 3, the genomic regions of these viruses shared the same phylogenetic origins with the corresponding genomic regions of the HKS04 viruses, suggesting that HKS04 and its related viruses might have resulted from a single recombination event and this recombinant genotype might already be widespread within China.
Recombination requires the simultaneous infection of a single cell by two different viral strains. In pigs infected with PCV2, concurrent coinfections with other viral and bacterial agents, such as porcine reproductive and respiratory syndrome virus, swine influenza virus, porcine parvovirus, Haemophilus parasuis, Actinobacillus pleuropneumoniae, Streptococcus suis and Mycoplasma hyopneumoniae, have been frequently detected (Kim & Lyoo, 2002). However, the detection of coinfections with divergent PCV2 strains has not been reported. The fact that PCV1 and PCV2 share a highly similar genomic organization and that coinfections are reported (Calsamiglia et al., 2002) raises the possibility of recombination between PCV1 and PCV2. To conclude, the present study provides strong statistical evidence of the presence of natural recombination among different lineages of PCV2. Our results also suggest that caution has to be taken on the interpretation of PCV2 phylogenies based on a single partial genomic region.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTMAIN TEXTREFERENCES |
|---|
Chae, C. (2004). Postweaning multisystemic wasting syndrome: a review of aetiology, diagnosis and pathology. Vet J 168, 41–49.[Medline]
Chae, C. (2005). A review of porcine circovirus 2-associated syndromes and diseases. Vet J 169, 326–336.[CrossRef][Medline]
Cheung, A. K. (2004). Palindrome regeneration by template strand-switching mechanism at the origin of DNA replication of porcine circovirus via the rolling-circle melting-pot replication model. J Virol 78, 9016–9029.
Csagola, A., Kecskemeti, S., Kardos, G., Kiss, I. & Tuboly, T. (2006). Genetic characterization of type 2 porcine circoviruses detected in Hungarian wild boars. Arch Virol 151, 495–507.[CrossRef][Medline]
Fenaux, M., Halbur, P. G., Gill, M., Toth, T. E. & Meng, X. J. (2000). Genetic characterization of type 2 porcine circovirus (PCV-2) from pigs with postweaning multisystemic wasting syndrome in different geographic regions of North America and development of a differential PCR-restriction fragment length polymorphism assay to detect and differentiate between infections with PCV-1 and PCV-2. J Clin Microbiol 38, 2494–2503.
Fenaux, M., Halbur, P. G., Haqshenas, G., Royer, R., Thomas, P., Nawagitgul, P., Gill, M., Toth, T. E. & Meng, X. J. (2002). Cloned genomic DNA of type 2 porcine circovirus is infectious when injected directly into the liver and lymph nodes of pigs: characterization of clinical disease, virus distribution, and pathologic lesions. J Virol 76, 541–551.
Gibbs, M. J. & Weiller, G. F. (1999). Evidence that a plant virus switched hosts to infect a vertebrate and then recombined with a vertebrate-infecting virus. Proc Natl Acad Sci U S A 96, 8022–8027.
Hamel, A. L., Lin, L. L. & Nayar, G. P. (1998). Nucleotide sequence of porcine circovirus associated with postweaning multisystemic wasting syndrome in pigs. J Virol 72, 5262–5267.
Harding, J. C. (2004). The clinical expression and emergence of porcine circovirus 2. Vet Microbiol 98, 131–135.[CrossRef][Medline]
Heath, L., Martin, D. P., Warburton, L., Perrin, M., Horsfield, W., Kingsley, C., Rybicki, E. P. & Williamson, A. L. (2004). Evidence of unique genotypes of beak and feather disease virus in southern Africa. J Virol 78, 9277–9284.
Holmes, E. C., Worobey, M. & Rambaut, A. (1999). Phylogenetic evidence for recombination in dengue virus. Mol Biol Evol 16, 405–409.[Abstract]
Kim, J. H. & Lyoo, Y. S. (2002). Genetic characterization of porcine circovirus-2 field isolates from PMWS pigs. J Vet Sci 3, 31–39.[Medline]
Larochelle, R., Magar, R. & D'Allaire, S. (2002). Genetic characterization and phylogenetic analysis of porcine circovirus type 2 (PCV2) strains from cases presenting various clinical conditions. Virus Res 90, 101–112.[CrossRef][Medline]
Liu, Q., Wang, L., Willson, P., O'Connor, B., Keenliside, J., Chirino-Trejo, M., Melendez, R. & Babiuk, L. (2002). Seroprevalence of porcine circovirus type 2 in swine populations in Canada and Costa Rica. Can J Vet Res 66, 225–231.[Medline]
Liu, J., Chen, I. & Kwang, J. (2005). Characterization of a previously unidentified viral protein in porcine circovirus type 2-infected cells and its role in virus-induced apoptosis. J Virol 79, 8262–8274.
Lole, K. S., Bollinger, R. C., Paranjape, R. S., Gadkari, D., Kulkarni, S. S., Novak, N. G., Ingersoll, R., Sheppard, H. W. & Ray, S. C. (1999). Full-length human immunodeficiency virus type 1 genomes from subtype C-infected seroconverters in India, with evidence of intersubtype recombination. J Virol 73, 152–160.
Mankertz, A., Domingo, M., Folch, J. M., LeCann, P., Jestin, A., Segales, J., Chmielewicz, B., Plana-Duran, J. & Soike, D. (2000). Characterisation of PCV-2 isolates from Spain, Germany and France. Virus Res 66, 65–77.[CrossRef][Medline]
Manni, F., Rotola, A., Caselli, E., Bertorelle, G. & Di Luca, D. (2002). Detecting recombination in TT virus: a phylogenetic approach. J Mol Evol 55, 563–572.[CrossRef][Medline]
Martin, D. & Rybicki, E. (2000). RDP: detection of recombination amongst aligned sequences. Bioinformatics 16, 562–563.
Meehan, B. M., Creelan, J. L., McNulty, S. & Todd, D. (1997). Sequence of porcine circovirus DNA: affinities with plant circoviruses. J Gen Virol 78, 221–227.[Abstract]
Meehan, B. M., McNeilly, F., McNair, I., Walker, I., Ellis, J. A., Krakowka, S. & Allan, G. M. (2001). Isolation and characterization of porcine circovirus 2 from cases of sow abortion and porcine dermatitis and nephropathy syndrome. Arch Virol 146, 835–842.[CrossRef][Medline]
Olvera, A., Cortey, M. & Segales, J. (2007). Molecular evolution of porcine circovirus type 2 genomes: phylogeny and clonality. Virology 357, 175–185.[CrossRef][Medline]
Posada, D. & Crandall, K. A. (1998). MODELTEST: testing the model of DNA substitution. Bioinformatics 14, 817–818.
Rambaut, A. & Grassly, N. C. (1997). Seq-Gen: an application for the Monte Carlo simulation of DNA sequence evolution along phylogenetic trees. Comput Appl Biosci 13, 235–238.
Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res 22, 4673–4680.
Tischer, I., Peters, D., Rasch, R. & Pociuli, S. (1987). Replication of porcine circovirus: induction by glucosamine and cell cycle dependence. Arch Virol 96, 39–57.[CrossRef][Medline]
Wen, L., Guo, X. & Yang, H. (2005). Genotyping of porcine circovirus type 2 from a variety of clinical conditions in China. Vet Microbiol 110, 141–146.[CrossRef][Medline]
Yang, Z. (1994). Estimating the pattern of nucleotide substitution. J Mol Evol 39, 105–111.[Medline]
Received 9 October 2006;
accepted 31 January 2007.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
