Virus Genes (2012) 44:5154 DOI 10.

1007/s11262-011-0665-x

Complete genome characterization of a East European Type 1 subtype 3 porcine reproductive and respiratory syndrome virus
J. Van Doorsselaere M. S. Brar M. Shi U. Karniychuk F. C.-C. Leung H. J. Nauwynck

Received: 8 June 2011 / Accepted: 22 August 2011 / Published online: 27 September 2011 Springer Science+Business Media, LLC 2011

Abstract Porcine reproductive and respiratory syndrome (PRRS) is a swine disease of major economic importance that causes reproductive and respiratory problems in pigs. PRRSV strains are divided into European (Type 1) and North-American (Type 2) genotypes. Within the European PRRSV genotype, three subtypes have been delineated. Full genome sequences for North American and European subtype 1 strains have been described. Here, the rst complete genomic characterization of a European subtype 3 strain (Lena) is described. Amplication of Orf1a and Orf1b fragments was achieved using a set of degenerate oligonucleotides. Using RT-PCR with Lena-specic primers, the full length sequence (15001 nt) was obtained. Alignment of Lena with European subtype 1 reference strain Lelystad showed variation over the entire length (84% identity/89% similarity at amino acid level) with the most variation in Orf1a (Nsp2/NSP2) with a deletion of 29 amino acids. Phylogenetic relationships using different

Orfs supported Lenas genetic distinction from European subtype 1 strains. The availability of the European subtype 3 PRRSV full genome may be important for the understanding of PRRSV evolution and the more pronounced pathogenic nature of Lena. Keywords PRRSV RT-PCR Full genome sequencing Phylogenetic analysis

Electronic supplementary material The online version of this article (doi:10.1007/s11262-011-0665-x) contains supplementary material, which is available to authorized users.
J. Van Doorsselaere Department of Health Care and Biotechnology, KATHO Catholic University College of South-West Flanders, Wilgenstraat 32, 8800 Roeselare, Belgium M. S. Brar M. Shi F. C.-C. Leung School of Biological Sciences, The University of Hong Kong, 5 N-12, Kardoorie Biological Science Building, Hong Kong, China U. Karniychuk (&) H. J. Nauwynck Department Virology, Parasitology and Immunology, Faculty of Veterinary Medicine, Ghent University, Salisburylaan 133, 9820 Merelbeke, Belgium e-mail: uladzimir.karniychuk@ugent.be

Porcine reproductive and respiratory syndrome (PRRS) is a global viral swine disease. The virus can cause reproductive disorders and can give rise to respiratory problems in pigs of all ages [1]. PRRSV is an enveloped virus containing a positive-strand RNA genome of approximately 15,000 nt. The genome contains a 50 untranslated region (50 -UTR), genes (Orf1a and Orf1ab) coding for nonstructural proteins, and genes (Orf2-Orf7) coding for structural proteins, a 30 -UTR tail and poly (A) tail. Orf1 encodes for a minimum of 13 proteins, mainly involved in replication and transcription. The structural proteins, GP2a, E, GP3, GP4, GP5, and M interact to form multimeric complexes, which are present in the envelope. The capsid of PRRSV consists of nucleoprotein (N). Originally, PRRSV strains were divided into a European (Type 1) and North American (Type 2) genotype with approximately 60% identity at the genome level. Recent ndings revealed a more complex situation with considerable genetic intra-genotype variability, especially within European genotype [27]. Based on Orf5 and Orf7 sequences, Stadejek et al. [4] proposed the classication of Type 1 strains into three subtypes, namely, a Pan-European subtype 1 (present in Western and Eastern Europe and Russia), East European subtype 2 (present in Russia, Lithuania and Belarus), and subtype 3 (mainly found in

123

52

Virus Genes (2012) 44:5154

Belarus). Recently, Shi et al. [7] used Orf5 sequences to study genetic relationships. New in this classication was the division of subtype 1 into 12 different clades. A number of full length PRRSV Type 1 sequences have been described [8, 9]. All of these strains were classied as Type 1, subtype 1 (based on Orf5 sequences). However, no full length genomes are available for Type 1, subtypes 2 and 3 strains. Recently, a new East European subtype 3 PRRSV strain (Lena) from Belarus has been isolated and characterized [10]. PRRSV Lena-inoculated pigs showed more severe clinical symptoms compared to pigs inoculated with Type 1, subtype 1 strain. To expand the genetic knowledge of this clinically important isolate, in this study, the full length genome of PRRSV strain Lena was sequenced and analysed. PRRSV Lena was isolated from a Belarusian farm [10]. RNA extraction from macrophage propagated virus and cDNA synthesis was performed as previously described [10]. A set of degenerate oligonucleotides (Supplementary Material Table S1) was designed based on the alignment of Orf1a and Orf1b genes from Type 1 (LV, EuroPRRS, HKEU16, 07V063) and Type 2 (VR-2332) strains. PCR experiments were performed using Taq DNA polymerase at 45C annealing temperature. Amplicons were treated with Antarctic Phosphatase and Exonuclease I followed by direct cycle sequencing [10]. Subsequently, Lena specic primers were designed based on the sequences from the degenerate oligonucleotide PCR approach (see supplementary Table S1). Overlapping amplicons (spanning the complete genome) were obtained using RT-PCR. Both strands of these fragments were directly sequenced and sequences were compiled. The sequence of Orf2a-Orf7 region has been previously described [10, 11]. For the amplication of the 30 end, oligo dT was used in combination with orf7fw [10]. A 50 end primer (50 endfw) was designed based on the alignment of Type 1 strains and this primer was used in combination with orf1seq1rev to amplify the 50 end [12]. Sequence alignments were performed using ClustalW (workbench.sdsc.edu). Phylogenetic trees were constructed using the software package MrBayes v3.2 as described [7]. A degenerate oligonucleotide PCR approach was used to amplify Orf1a and Orf1b fragments from Lena with several primer combinations resulting in amplicons (Supplementary Material Table S1). Sequences from these amplicons served as the basis for designing Lena-specic primers which (after PCR and sequencing) resulted in the Orf1a and Orf1b genes. After amplication and sequencing of 50 and 30 ends, a full genome sequence of 15001 nt was obtained (GenBank: JF802085). The size of the 50 end (221nt) and 30 end (118nt) of Lena was identical to the 50 end and 30 end of LV, respectively, with a 90% identity and 24 nt difference for the 50 end and

an 89% identity and 14 nt difference for the 30 end. Several motifs such as the transcription regulatory sequence (UUAACC) and CACCC stretches (involved in binding of host cell transcription factors; [9]) were conserved in the 50 end of Lena. Comparison of the proteins from Lena with LV showed 6594% identity and 73100% similarity (Table 1) and overall 82% identity and 89% similarity. The most variation compared to LV was noticed in NSP1-b (80% identity/86% similarity) with a deletion of 1 aa at position 274 (in LV) and NSP2 (65% identity/73% similarity). A deletion of 29 aa in a variable region of NSP2 (at positions 710-739) was observed. Similar deletions in this region are known and recently Darwich et al. [13] showed that some Spanish strains (e.g., Cresa 3262) showed deletions of up to 74 amino acids. In this region, Lena showed more aa variation with LV as compared to similarity of LV with other subtype 1 strains. Pronounced variability is also noticed at the protein level from aa positions 2000-2500 (NSP5-NSP8), 3100-3600 (NSP10-NSP11), and 4100-4900 (GP2, GP3, GP4, GP5, and M). In addition, NSP12 (Orf1b) from Lena had an extra four amino acids at the C-terminus compared to LV (and other subtype 1 strains). In order to determine the phylogenetic relationship of Lena within the context of Type 1 PRRSVs, we used previously described Orf5 and Orf7 sequence collections

Table 1 Comparison of proteins from Lena (subtype 3) and prototype LV (subtype 1) Orf 1a Protein NSP1 NSP2 NSP3 NSP4 NSP5 NSP6 NSP7 NSP8 1b NSP9 NSP10 NSP11 NSP12 2a 2b 3 4 5 6 7 GP2 E GP3 GP4 GP5 M N Lena (aa) 384 832 447 203 170 16 269 45 645 441 224 156 249 70 249 184 201 173 124 LV (aa) 385 861 447 203 170 16 269 45 645 442 224 152 249 70 265 183 201 173 128 % Identity 80 65 90 87 90 87 89 96 94 91 96 90 92 96 81 85 83 94 92 % Similarity 86 73 95 95 94 87 94 100 98 95 99 94 94 99 87 92 89 98 95

Overall percentage identity/similarity between LV and Lena is 84% and 89%, respectively

123

Virus Genes (2012) 44:5154

53

which comprehensively characterize the presently known Type 1 diversity [7]. For both Orf5 and Orf7 phylogenies (Fig. 1), Lena was grouped within a cluster consisting of only Belarusian sequences, dened as subtype 3 under the current subtyping system [3, 4]. The most closely related

sequences to Lena were those isolated from Eastern Belarus. The phylogenetic analyses were also extended to other Orfs, where Lena was compared to 11 Type 1 PRRSV strains whose full genome sequences were available in the database (Table 2). For most of the Orfs, Lena was well-

Fig. 1 Phylogenetic characterization of Lena. Phylogenies were constructed using MrBayes 3.2 based on 8 full length Orf alignments of PRRSV, subtype classication is indicated to the right. Asterisks

indicate nodes with high support (posterior probability value greater than 80). All phylogenies (except those for Orf5 and Orf7) were rooted with VR-2332, which is the prototype of Type 2 PRRSV

Table 2 Overview of strains used for phylogenetic analysis Strain VR-2332 EuroPRRS Cresa3266 Cresa3262 Lelystad BJEU06-1 NMEU09-1 Amervac KNU-07 SD01-08 01-CB1 HKEU16 Lena Type 2 1 1 (subtype 1) 1 (subtype 1) 1 (subtype 1) 1 (subtype 1) 1 (subtype 1) 1 (subtype 1) 1 (subtype 1) 1 (subtype 1) 1 (subtype 1) 1 (subtype 1) 1 (subtype 3) Accession U87392 AY366525 JF276434 JF276431 M96262 GU047344 GU047345 GU067771 FJ349261 DQ489311 DQ864705 EU076704 JF802085 Location USA USA Europe, Germany Europe, Spain Europe, Netherlands Asia, China Asia, China Europe, Spain Asia, Korea USA Asai, Thailand Asia, Hong-Kong Europe, Belarus Year 1990 1999 1996 1992 1991 2006 2009 1992 2007 2001 2006 2007 2007 Reference [15] [9] [13] [13] [8] [16] [16] [16] [17] [18] Amonsin et al., unpublished Li et al., unpublished This study

The genotype of the strains is based on the work of Stadejek et al. [4]. The year refers to the year of isolation

123

54

Virus Genes (2012) 44:5154 4. T. Stadejek, M.B. Oleksiewicz, A.V. Scherbakov, A.M. Timina, J.S. Krabbe, K. Chabros, D. Potapchuk, Arch. Virol. 153, 14791488 (2008) 5. G. Balka, A. Hornyak, A. Balint, I. Kiss, S. Kecskemeti, T. Bakonyi, M. Rusvai, Vet. Microbiol. 127, 128135 (2008) 6. M.P. Murtaugh, T. Stadejek, J.E. Abrahante, T.T.Y. Lam, F.C.-C. Leung, Virus Res. 154, 1830 (2010) 7. M. Shi, T.T.Y. Lam, C.C. Hon, R.H.K. Hui, K.S. Faaberg, T. Wennblom, M.P. Murtaugh, T. Stadejek, F.C.-C. Leung, Virus Res. 154, 717 (2010) 8. J.J. Meulenberg, M.M. Hulst, E.J. de Meijer, P.L. Moonen, A. Den Besten, E.P. de Kluyver, G. Wensvoort, R.J. Moormann, Virology 192, 6272 (1993) 9. S.L. Ropp, C.E. Mahlum Wees, Y. Fang, E.A. Nelson, K.D. Rossow, M. Bien, B. Arndt, S. Preszler, P. Steen, J. ChristopherHennings, J.E. Collins, D.A. Beneld, K.S. Faaberg, J. Virol. 78, 36843703 (2004) 10. U.U. Karniychuk, M. Geldhof, M. Vanhee, J. Van Doorsselaere, T.A. Saveleva, H.J. Nauwynck, BMC Vet. Res. 6, 30 (2010) 11. W. Van Breedam, S. Costers, M. Vanhee, C.A. Gagnon, I.M. Rodrguez-Gomez, M. Geldhof, M. Verbeeck, J. Van Doorsselaere, U. Karniychuk, H.J. Nauwynck, Vet. Immunol. Immunopathol. 141, 246257 (2011) 12. J. Van Doorsselaere, M. Geldhof, H.J. Nauwynck, P.L. Delputte, Virol. J. 8, 160 (2011) 13. L. Darwich, M. Gimeno, M. Sibila, I. Diaz, E. de la Torre, S. Dotti, L. Kuzemtseva, M. Martin, J. Pujols, E. Mateu, Vet. Microbiol. 150, 4962 (2010) 14. D. Tian, H. Zheng, R. Zhang, J. Zhuang, S. Yuan, Virology 412, 18 (2011) 15. M.P. Murtaugh, M.R. Elam, L.T. Kakach, Arch. Virol. 140, 14511460 (1995) 16. N. Chen, Z. Cao, X. Yu, X. Deng, T. Zhao, L. Wang, Q. Liu, X. Li, K. Tian, J. Gen. Virol. 92, 880892 (2011) 17. E. Nam, C.K. Park, S.H. Kim, Y.S. Joo, S.G. Yeo, C. Lee, Arch. Virol. 154, 629638 (2009) 18. Y. Fang, R.R. Rowland, M. Roof, J.K. Lunney, J. ChristopherHennings, E.A. Nelson, J. Virol. 80, 1144711455 (2006)

diverged from entire subtype 1 cluster (Fig. 1). Although Orf2 showed exception, the posterior probability value for Lena to group within subtype 1 sequences was quite low to support such a topology. Therefore, the fact that Lena was genetically distinct lineage for different Orfs validate the subtyping scheme proposed by Stadejek et al. [3, 4] at the genomic level. In this study, the full genome of the East European subtype 3 PRRSV strain Lena was sequenced and analysed. Shi et al. [7] suggested that (ancestors of) subtype 3 PRRSV may have been present in Eastern Europe before the emergence of subtype 1 PRRSV in Western Europe. The availability of the subtype 3 full genome may be important for understanding of PRRSV evolution and PRRSV geographical relocation. It also opens possibilities for the construction of (chimeric) infectious clones [14], tools which can be used for the identication of genetic factors which dene the more pronounced pathogenic nature of Lena.
Acknowledgments This study was supported by VEWA Academic Fund. The authors would like to thank Ine Vanherpe for technical assistance.

References
1. G. Wensvoort, E.P. de Kluyver, J.M. Pol, F. Wagenaar, R.J. Moormann, M.M. Hulst, R. Bloemraad, A. den Besten, T. Zetstra, C. Terpstra, Vet. Microbiol. 33, 185193 (1992) 2. C. Prieto, E. Alvarez, F.J. Martinez-Lobo, I. Simarro, J.M. Castro, Vet. J. 175, 356363 (2008) 3. T. Stadejek, M.B. Oleksiewicz, D. Potapchuk, K. Podgorska, J. Gen. Virol. 87, 18351841 (2006)

123