journa l homepage: www.e lsev ier .com/ locate /v i rusres
ecombination analyses between two strains of porcine reproductive andespiratory syndrome virus in vivo
an Liu, Rong Zhou, Jialong Zhang, Lei Zhou, Qiuyue Jiang, Xin Guo, Xinna Ge, Hanchun Yang ∗
ey Laboratory of Zoonosis of Ministry of Agriculture, College of Veterinary Medicine and State Key Laboratory of Agrobiotechnology, China Agricultural University, Beijing 100193,eople’s Republic of China
r t i c l e i n f o
rticle history:eceived 24 August 2010eceived in revised form0 November 2010ccepted 7 December 2010vailable online 15 December 2010
eywords:orcine reproductive and respiratoryyndrome virus (PRRSV)
a b s t r a c t
Porcine reproductive and respiratory syndrome virus (PRRSV) is characteristic of genetically extensivevariation. In this study, five SPF pigs were co-infected with two strains of PRRSV (JXwn06-81c and HB-1/3.9c), and 352 viruses were cloned by plaque assay from the sera of the infected pigs on days 3, 5, 7,10, 14, 21 postinfection (pi), and the recombinant events between the two viruses were systematicallyinvestigated by sequencing the ORF5, ORF3 and Nsp2 genes of each cloned virus and using SimPlot andGenetic Algorithm for Recombination Detection (GARD) analysis. Totally, 133 recombinant viruses outof the plaque viruses were acquired from four of five infected pigs during days 7–21 pi upon co-infectionwith JXwn06-81c and HB-1/3.9c. The intragenic recombination and intergenic fragment exchange of theORF5, ORF3 and Nsp2 genes between the two viruses exhibited different patterns, and the recombination
o-infectionecombination
n vivo
for ORF5 gene and Nsp2 occurred as early as on day 7 pi. The recombination between the ORF5, ORF3or Nsp2 gene resulted in the generation of chimeric GP5, GP3 or Nsp2. Of the three genes, Nsp2 geneexhibited more complicated recombination situation. Meanwhile, the putative recombination break-points and hotspots for the three genes were analyzed. Our findings not only provide valuable evidencesfor understanding that recombination is an important genetic mechanism contributing to the variationand evolution of PRRSV, but also suggest that extensive use of attenuated vaccine of PRRSV undoubtedly
ed di
contributes to the increas
. Introduction
Porcine reproductive and respiratory syndrome (PRRS) is onef the most economically important diseases for the pig industryorldwide (Garner et al., 2001; Neumann et al., 2005). The etio-
ogical agent of the disease, porcine reproductive and respiratoryyndrome virus (PRRSV), classified in the order Nidovirales, fam-ly Arteriviridae, genus Arterivirus, is an enveloped, single-strandedositive-sense RNA virus with approximately 50–65 nm in diam-ter (Benfield et al., 1992; Cavanagh, 1997). The viral genome ispproximately 15 kb in length and contains nine overlapping openeading frames (ORFs). ORF1a and ORF1b comprising 80% of theenome encode viral nonstructural proteins involved in genomeranscription and replication (Snijder et al., 2001), while ORF2a,RF2b, and ORFs 3–7 encode the viral structural proteins GP2,
, GP3, GP4, GP5, M, and N, respectively (Bautista et al., 2002;tadejek et al., 2002; Wootton et al., 2000). There are two dis-inct genotypes of PRRSV, the European isolate (Lelystad virus, LV)nd the North American isolate (VR-2332), representing two geno-
types with antigenic differences—type 1 and type 2, respectively(Larochelle and Magar, 1997). The two genotypes share only about60% nucleotide similarities at the genome level, but exhibit iden-tical pathobiological phenotypes (Meng et al., 1995a; Nelson et al.,1993).
PRRSV is characteristic of genetically extensive variation withthe genetic/antigenic diverse strains in both North American andEuropean isolates (Meng, 2000). The ORF5 is most variable amongthe different strains within the genome of PRRSV (Kapur et al.,1996; Meng et al., 1995b; Zhou et al., 2009a,c). The GP5 encodedby ORF5 of PRRSV is recognized to be essential for virus infectivityand virus neutralization (Pirzadeh et al., 1998), as well as for viralentry by interacting with the host cell receptor (Xia et al., 2009).The variations of ORF5 gene reflect the exceptional genetic diver-sity of PRRSV in field (An et al., 2007; Balka et al., 2008; Fang et al.,2007; Li et al., 2009a,b; Mateu et al., 2003; Stadejek et al., 2006).Meanwhile, the nonstructural protein 2 (Nsp2)-coding region isalso a remarkable variable gene which can tolerate natural point
mutations, nucleotide deletions and insertions (Gao et al., 2004;Ran et al., 2008). In addition, the ORF3 encoding GP3 of PRRSV isconsidered to be an variable gene as well (Mardassi et al., 1995).Thus, all these evidences point to the fact that PRRSV is a virus withnotorious genetic variability and rapid evolution ability.
474 D. Liu et al. / Virus Research 155 (2011) 473–486
Table 1Primers used in RT-PCR amplification and sequencing.
Primer Sequence (5′–3′) Use
ORF5F ATGACACCTGAGACCATGAGGT Amplification and sequencing of ORF5 geneORF5R TGTGCTATCATTGCAGAAGTCGORF3F CAGGGTCAAATGTAACCATAGTG Amplification and sequencing of ORF3 geneORF3R GGCAAGAAGAAAGCATGAGGAGNsp2F TTGAGGATCTCCTCAGAATCAG Amplification and sequencing of Nsp2gene
Previous studies have indicated that PRRSV exists during nat-ral infection as a quasispecies distribution of related genotypesRowland et al., 1999; Goldberg et al., 2003) and PRRSV can evolveontinuously by random mutation and intragenic recombinationChang et al., 2002; Dee et al., 2001; Kapur et al., 1996; Mengt al., 1995a). Thus, much attention to the phenomenon of recom-ination among different strains of PRRSV has been paid in theast few years. Co-infection of MA-104 cells with two strains ofRRSV could result in emergence of recombinant viral particlesYuan et al., 1999). Experimental data under cell culture revealedhat high frequency RNA recombination in porcine reproductivend respiratory syndrome virus occurred preferentially betweenarental sequences with high similarity (van Vugt et al., 2001).ubsequently, the observation of recombination was reported inild-type isolates of European genotype of PRRSV and appearance
f novel PRRSV isolates by recombination in the natural environ-ent (Forsberg et al., 2002; Murtaugh et al., 2001). Recent survey
eported a natural recombination among Chinese strains of PRRSVased on the genetic variation analysis of ORF5 (Li et al., 2009a,b).owever, little is known regarding the exact emergence of recom-ination events in pigs infected with different strains of PRRSV. Inhe present study, we carried out experimental co-infection of twotrains of PRRSV in pigs and analyzed the recombination eventsetween two viruses during the co-infection, in order to providealuable evidences for elucidating the variation and evolutionaryechanisms of PRRSV.
. Materials and methods
.1. Viruses and cells
MARC-145 cells were maintained in GIBCOTM Dulbecco’s modi-ed eagle medium (DMEM) (Invitrogen Corporation, Auckland, NY,SA) supplemented with 10% fetal bovine serum (FBS, Hycloneaboratories Inc., South Logan, UT, USA) at 37 ◦C, with 5% CO2.wo PRRSV strains, JXwn06 and HB-1/3.9, were used in thistudy. JXwn06, a highly pathogenic virus (Zhou et al., 2009b),as passaged on MARC-145 cells, and its 80th-passage virus wassed. HB-1/3.9, adapted in MARC-145 cells, was derived from HB-(sh)/2002, a low-virulence strain isolated in 2002 (Gao et al.,004).
.2. Purification of the viruses by plaque cloning and genomicequencing
The two viruses were purified by plaque cloning and the virusitres were assayed. The purified virus was individually designatedXwn06-81c and HB-1/3.9c. The full-length genomes of the puri-
ed viruses were sequenced according to the method describedreviously (Zhou et al., 2009b). Briefly, 14 overlapped fragmentsovering whole viral genome were amplified by RT-PCR using theorresponding primers. The 5′-region was amplified using a 5′ fullACE kit (TaKaRa, Dalian, China). The amplicons were cloned into
Nsp2 gene sequencing
pEASY-Blunt vectors (Transgen, Beijing, China) and then submit-ted to the company (Invitrogen, Beijing, China) for sequencing.Genomic analyses were conducted using the software DNAMAN(University of California) and DNAStar (Lasergene).
2.3. Animal infection and serum collection
Five 6-week-old SPF landrace piglets were obtained from theBeijing Center for SPF Swine Breeding and Management, and raisedin the animal facilities at China Agricultural University (CAU). Eachpiglet was simultaneously inoculated intranasally with 1 ml of viruscontaining 104 TCID50 JXwn06-81c and 1 ml of virus containing104 TCID50 HB-1/3.9c. Animals received food and water ad libitum.Serum samples were collected from each piglet on days 3, 5, 7, 10,14, 21 postinfection (pi), respectively. Simultaneously, serum sam-ples were collected from three 6-week-old SPF pigs infected withJXwn06-81c or HB-1/3.9c individually.
2.4. Virus isolation, gene amplification and sequencing
Viruses in serum samples were isolated by plaque assay, 10–12plaque viruses were randomly selected from each samples at eachtime point. MARC-145 cells were cultured in a 6 well tissue cul-ture dish and were infected by the serum sample for 1 h when theywere 100% confluent, then added agarose/growth media to eachwell. Plaques would be visible by day 2–4 after infection; the sin-gle plaque was selected by day 5–6 after infection. The isolatedplaque viruses were named as X-Y-Z (X = day pi; Y = serial num-ber of pig; Z = serial number of plaque virus). Each plaque viruswas propagated on MARC-145 cell. Viral RNA was extracted fromthe cell culture of each plaque virus and RT-PCR was conducted toamplify the ORF5, ORF3 and Nsp2 gene using three pairs of specificprimers (Table 1). PCR products were purified using a QIAquick GelExtraction Kit (Qiagen, Valencia, CA, USA) and sequenced in bothdirection using the amplification primers and compared with theparental viruses. Serum samples collected from three 6-week-oldSPF pigs infected with JXwn06-81c or HB-1/3.9c individually werealso used for amplification and sequencing of each gene to monitortheir variation.
2.5. Recombination analysis
Multiple sequence alignment between the parent viruses andcloned viruses from serum samples was performed with ClustalX1.83 program to determine the recombination occurred within theORF5, ORF3 and Nsp2 gene. The gene sequence was scanned forpossible recombination events using the soft package of SimPlot (v3.5.1) according to the methods described previously (Lole et al.,
1999). A window of 200 bp and a step size of 20 bp were appliedin the analysis using JXwn06 and HB-1/3.9 as the parent viruses,and the amplification sequences as the query. GARD (Genetic Algo-rithm for Recombination Detection) was used to search for putativebreakpoints delimiting sequence regions having distinct phyloge-
D. Liu et al. / Virus Research 155 (2011) 473–486 475
Table 2Summary of the recombinant events and patterns upon co-infection of JXwn06-81c and HB-1/3.9c.
Day postinfection # of the infected pig and the number of JXwn06-81c, HB-1/3.9c or recombinant virus Total JX:HB:R
Total JX:HB:R 32:0:3819a+5b+14c 71:0:0 46:0:254c+17d+4g 35:0:331c+16d+2e+1f+12g+1h 35:0:373a+2b+7c+6e+2g+2h 219:0:13322a+7b+26c+42d+8e+7f+18g+3h
JX, the ORF5, ORF3 or Nsp2 genes of the plaque-purified virus were derived from JXwn06-81c; HB, the ORF5, ORF3 or Nsp2 genes of the plaque-purified virus were derivedfrom HB-1/3.9c; R, the recombinant virus.a, with the recombination of both ORF5 and Nsp2 genes; b, with the recombination of ORF5, ORF3 and Nsp2 genes together; c, with the recombination of both ORF3 andN n of OO Nsp2a
nbrC
3
3H
1DT1(bgOgsa9v
3a
sJatoipTin
pv
gvt7m
sp2 genes; d, with the recombination of Nsp2 gene only; e, with the recombinatioRF3 genes from HB-1/3.9c and Nsp2 gene from JXwn06-81c; h, with the ORF5 and–f, intragenic recombination; g and h, intergenic fragment exchange.
ies (Stewart et al., 2009). Support for recombination is reflectedy changes in the goodness of fit between non-recombinant andecombinant models, as assessed by the AIC (Akaike Informationriterion).
. Results
.1. Full-length genomic nucleotide sequences of JXwn06-81c andB-1/3.9c
The full-length genomic sequences of JXwn06-81c and HB-/3.9c were determined and have been submitted to GenBankatabase (GenBanK accession numbers: HQ233604, HQ233605).he genomes of the two viruses were 15,320 nucleotides (nt) and5,407 nucleotides (nt) in length, respectively, excluding the polyA) tails. The nucleotide sequence identity of the complete genomeetween the two viruses was 96.76%. For the two viruses, the ORF5ene was 603 nt in length with 96.85% of nucleotide identity, theRF3 gene was 765 nt with 96.99% of nucleotide identity. The Nsp2ene of JXwn06-81c was 3498 nt in length with the existence of theame 90-nt deletion as its original virus JXwn06 (Zhou et al., 2009b),nd the Nsp2 gene of HB-1/3.9c was 3588 nt in length. There was3.87% of nucleotide identity between the Nsp2 genes of the twoiruses.
.2. Intragenic recombination of ORF5 gene between JXwn06-81cnd HB-1/3.9c
Totally, 352 plaque viruses were obtained from the serumamples of five pigs during days 3–21 pi upon co-infection withXwn06-81c and HB-1/3.9c. The ORF5, ORF3 and Nsp2 genes ofll the viruses were amplified and sequenced. The results showedhat out of them, 133 recombinant viruses were acquired from fourf five infected pigs, and 219 JXwn06-81c were gained from fivenfected pigs, and no HB-1/3.9c was obtained from five infectedigs. The recombinant events and patterns are summarized inable 2. In addition, the corresponding gene amplified from pigsnfected with JXwn06-81c or HB-1/3.9c alone had no variation (dataot shown).
The nucleotide sequences of amplified ORF5 genes for all thelaque viruses were determined and aligned with those of parentaliruses (Fig. 1A)
. The results showed that the intragenic recombination in ORF5
ene could be examined on day 7–21 pi; of the recombinantiruses, 37 viruses had recombinant ORF5 genes with seven pat-erns (Table 3); the recombinant sequence as a representative of-1-5 was the predominant, occupying 19/37 (Table 3). The align-ent analyses of amino acid deduced from the recombinant ORF5
RF5 gene only; f, with the recombination of ORF3 gene only; g, with the ORF5 andgenes from JXwn06-81c and ORF3 gene from HB-1/3.9c.
genes indicated that the recombination resulted in the generationof chimeric GP5 (Fig. 1B).
The recombination event between the ORF5 genes of the twoviruses was further confirmed by the SimPlot analysis, showingsimilar results to the sequence alignments. The analysis results ofrepresentative recombinant viruses 7-1-5 and 14-4-7 are shown inFig. 2A. The putative recombinant breakpoints were analyzed byGARD method using the maximized value of �2 combined witha genetic algorithm. The data indicated that the nucleotides atthe position 177 and 349 of ORF5 gene were the hotspots for therecombination of ORF5 gene for PRRSV (Table 4). We found thatthe recombination site of the ORF5 gene for the recombinant virus10-5-4 was similar to the recombinant of ORF5 gene describedpreviously (Yuan et al., 1999), and different from the natural recom-binant strain ZJJ07 described recently (Li et al., 2009a,b).
3.3. Intragenic recombination of ORF3 gene between JXwn06-81cand HB-1/3.9c
The intragenic recombination in ORF3 gene was found until 14day pi. Of the plaque viruses, 41 viruses had recombinant ORF3genes with three patterns (Table 3). The recombinant sequence asa representative of 14-1-1 was the predominant, occupying 38/41(Table 3) and sharing 100% identity to the parental strain HB-1/3.9cfor the first 429 nt and 100% identity to the parental JXwn06-81cfor the following 336 nt (Fig. 1C). As shown in Fig. 1D, the recom-bination resulted in the generation of chimeric GP3 as well. Therecombination analyses in ORF3 gene were performed by the pro-gram SimPlot 3.5, showing that ORF3 gene of the recombinant virus14-3-4 was highly close to JXwn06-81c before the breakpoint andclose to HB-1/3.9 behind the breakpoint (Fig. 2B). Three poten-tial recombination breakpoints (at the positions 402 nt, 409 nt and555 nt) with maximal �2 were found by GARD analysis (Table 4).
3.4. Intragenic recombination of Nsp2 gene between JXwn06-81cand HB-1/3.9c
The sequence analyses of the amplified Nsp2 genes from 352plaque viruses indicated that there were more complicated recom-bination events between the Nsp2 genes of the two viruses. Theintragenic recombination in Nsp2 gene could be found on day7–21 pi. Of the plaque viruses, 96 viruses contained recombinant
Nsp2 genes with fourteen patterns. Majority of the recombinantsequences was 3498 nt in length with the same 90 nt deletionlike the parental virus JXwn06-81c, and the Nsp2 gene of therecombinant virus 21-1-8 was the most complicated pattern thatappeared to be a quadruple cross-over recombinant. The recom-
476 D. Liu et al. / Virus Research 155 (2011) 473–486
Fig. 1. Alignment of gene sequences and deduced amino acids encoded by ORF5 (A and B) and ORF3 (C and D) genes of the recombinant viruses with representative patternswith the parental JXwn06-81c and HB-1/3.9c. Dashes indicate identical residues with JXwn06-81c.
D. Liu et al. / Virus Research 155 (2011) 473–486 477
(Cont
b7Otsf
Fig. 1.
inant patterns as representatives of the recombinant viruses
-1-5, 10-3-2 and 10-4-1 were the predominant (Table 3). LikeRF5 and ORF3 genes, the recombination of Nsp2 gene led to
he generation of chimeric Nsp2 as well (data not shown). Ashown in Fig. 2C, SimPlot nucleotide diversity profiles calculatedor Nsp2 genes confirmed that the recombinant viruses were gen-
inued )
erated through recombination, and revealed a reversal in sequence
similarity between recombinant and parental viruses in regionsupstream and downstream of a predicted breakpoint. GARD analy-ses indicated that the positions 264 nt, 259 nt, 2300 nt and 1818 ntwere the hotspots for intragenic recombination of Nsp2 gene(Table 4).
478 D. Liu et al. / Virus Research 155 (2011) 473–486
Fig. 1. (Continued )
D. Liu et al. / Virus Research 155 (2011) 473–486 479
Fig. 1. (Continued )
480 D. Liu et al. / Virus Research 155 (2011) 473–486
Fig. 1. (Continued )
D. Liu et al. / Virus Research 155 (2011) 473–486 481
Fig. 1. (Continued )
482 D. Liu et al. / Virus Research 155 (2011) 473–486
(Cont
3H
OtndbJiiet
4
itfawwa
Fig. 1.
.5. Intergenic fragment exchange between JXwn06-81c andB-1/3.9c
In addition to the intragenic recombination phenomenon ofRF5, ORF3 and Nsp2, the intergenic fragment exchanges among
hese genes were found (Table 5). Of the obtained 133 recombi-ant viruses, 18 recombinant viruses had ORF5 and ORF3 geneserived from HB-1/3.9c, and Nsp2 gene from JXwn06-81c; 3 recom-inant viruses possessed ORF5 and Nsp2 genes derived fromXwn06-81c, and ORF3 gene from HB-1/3.9c. In addition, otherntergenic fragment exchange patterns were determined as shownn Table 5. These results indicate that the intergenic fragmentxchanges among these genes occur during the co-infection of thewo parental viruses in vivo.
. Discussion
The recombination among different viral strains is an importantssue increasing much attention in the field of PRRSV. The objec-ive of our present study is to provide the first experimental data
or the fact that the recombination occurs during the co-infectionmong different strains of PRRSV in animal. Following co-infectedith two strains of PRRSV—JXwn0681c and HB-1/3.9c in SPF pigs,e purified the viruses from the blood sample of the co-infected
nimals by plaque assay, and sequenced the ORF5, ORF3 and Nsp2
inued )
gene of the purified viruses and compared their sequences togetherwith the parental viruses. Our results showed that 4 of 5 pigs co-infected with the two strains had the recombinant viral particles,and the recombination occurred as early as on day 7 post inocula-tions. Perhaps the real timing of the initial recombination event wasmuch earlier than 7 day post inoculation, because it was only whenthe level of the recombinant viruses reaches a relatively high levelwithin the population that they would be detected. Among the 352plaque-cloned viruses we obtained, the frequency of ORF5, ORF3and Nsp2 recombination were 10.51%, 11.64% and 27.27%, respec-tively, there were 133 recombinant viruses, occupying 37.78%,showing the high frequency of recombination between the twostrains of PRRSV in vivo. To exclude the possibility of recombinationoccurred during in vitro plaque isolation, we did direct RT-PCR andsequencing from some serum samples, and obtained correspond-ing recombinant genes which were similar to the viruses by plaqueisolation (data not shown).
Interestingly, in addition to the recombinant viruses, weobtained 219 plaque viruses with identical ORF5, ORF3 and Nsp2genes as JXwn06-81c, but no HB-1/3.9c from the recovered serum
samples of the infected pigs, implying that JXwn06-81c was thepredominating virus with stronger replication ability, during theirreplications upon the co-infection of the two viruses in vivo.
Homologous genetic recombination has been reported to occurexperimentally in PRRSV, in experimental co-infections in vitro
D. Liu et al. / Virus Research 155 (2011) 473–486 483
Fig. 2. Similarity plot and bootscan analyses of ORF5, ORF3 and Nsp2 genes of representative recombinant viruses by SimPlot (Version 3.5.1). (A) ORF5 gene. Pair-wisedivergence between the recombinant and each parental virus over the 603 bp ORF5 region employed a 200 nt sliding window at 20 nt intervals and the default Kimura (2-parameter) distance model. Percentage identities at each analysis point were plotted on a line chart. (B) ORF3 gene. (C) Nsp2 gene. The representative recombinant sequenceserved as query. For similarity plot analysis, the y-axis shows the percentage similarity between the parental sequences and the query sequence. For bootscan analysis, they-axis shows the percentage of permutated trees using a sliding window of 200 bases and a step size of 20 bases.
484 D. Liu et al. / Virus Research 155 (2011) 473–486
(Cont
(g(p
TR
Fig. 2.
Yuan et al., 1999). Natural recombination appears to result fromene rearrangements and causes the generation of chimeric virusesLi et al., 2009a; Yuan et al., 1999). In our study, by Pairwise com-arisons and SimPlot analysis of the amplified ORF5, ORF3 and
able 3ecombination pattern of PRRSV upon co-infection of two strains in vivo.
a Numbers refer to nucleotide positions within corresponding gene of JXwn06-81c andb 90-nt deletion in Nsp2 gene.c No deletion in Nsp2 gene.
inued )
Nsp2 genes, it was indicated that there were diverse and irregularrecombinant patterns in those three genes and sequence exchangesexisted within them, resulting in the generation of chimeric GP5,GP3 or Nsp2, and finally yielding new variants. Undoubtedly, the
umber of the recombinant virus Representative recombinant virus
ecombinations among different strains of PRRSV contribute to thencreased genetic diversity of the virus.
Recombination occurs when two parental viruses strains co-nfecting one cell come in physical contact with each other andoth generate subgenomic RNAs during viral replication (Chhabrat al., 2010), and any viable recombinant virus must be suitably fito survive competition from both parental and other recombinantiruses. In this study, irregular and complicated recombination pat-erns indicate that the recombination events are generally random,lthough the detailed understanding of mechanism involved inuch recombination phenomenon is to be clarified. Interestingly,hen compared with the natural ORF5 recombinant isolated byuan et al. (1999), we found that the recombination site of the ORF5ene of the plaque-cloned virus 10-5-4 was similar.
Current studies have indicated that the emergence and epidemicf highly pathogenic virus not only further aggravate the com-licated situation of PRRS, but also enhance the genetic diversity
f PRRSV in China (Zhou and Yang, 2010). Moreover, attenuatedaccines are extensively being used in preventing and controllingRRSV infection in Chinese swine industry. Although it has beenocumented that an attenuated vaccine derived from the highly
1 14-5-12
pathogenic virus by passage on MARC-145 cell could provide effec-tive protection for pigs against re-infection by the homologousvirus (Martelli et al., 2009; Tian et al., 2009), we need to pay muchattention to the recombination between the vaccine virus and thecirculating virus in field besides to its reversion to virulent virus.Our present data provide evidences for the finding that the recom-bination occurred in genetically similar attenuated and field strainsof PRRSV (Li et al., 2009a,b). Thus, the concerns associated withthe potential risk caused by extensive use of attenuated vaccineof PRRSV should be emphasized. However, further analyzing thebiological characteristics and pathogenicities of the recombinantviruses obtained in our study is essential.
Taken together, we systemically analyzed the recombinationevents between two strains of PRRSV following the co-infection ofpigs. Our data not only provide valuable evidences for the conclu-sion that recombination is an important genetic mechanism con-tributing to the variation and evolution of PRRSV, but also suggest
that extensive use of attenuated vaccine of PRRSV undoubtedly con-tributes to the increased diversity of PRRSV in field. Meanwhile, ourfindings point out valuable clues for further analyzing molecularmechanism associated with the variation and evolution of PRRSV.
4 search
A
fuPPiSI
R
A
B
B
B
C
C
C
D
F
F
G
G
G
K
L
L
L
L
M
M
86 D. Liu et al. / Virus Re
cknowledgments
This work was supported by National Natural Science Fundsor Distinguished Young Scholar (30825031) from National Nat-ral Science Foundation of China, and National Key Basic Researchlan Grant (#2005CB523204) and National Key Technology R&Drogram of China (grant no. 2006BAD06A03) from the Chinese Min-stry of Science and Technology, and the Program for Cheung Kongcholars and Innovative Research Team in University of China (no.RT0866).
eferences
n, T.Q., Zhou, Y.J., Liu, G.Q., Tian, Z.J., Li, J., Qiu, H.J., Tong, G.Z., 2007. Genetic diversityand phylogenetic analysis of glycoprotein 5 of PRRSV isolates in mainland Chinafrom 1996 to 2006: coexistence of two NA-subgenotypes with great diversity.Vet. Microbiol. 123, 43–52.
alka, G., Hornyák, A., Bálint, A., Kiss, I., Kecskeméti, S., Bakonyi, T., Rusvai,M., 2008. Genetic diversity of porcine reproductive and respiratory syn-drome virus strains circulating in Hungarian swine herds. Vet. Microbiol. 127,128–135.
autista, E.M., Faaberg, K.S., Mickelson, D., McGruder, E.D., 2002. Functional proper-ties of the predicted helicase of porcine reproductive and respiratory syndromevirus. Virology 298 (2), 258–270.
ee, S.A., Torremorell, M., Rossow, K., Mahlum, C., Otake, S., Faaberg, K., 2001. Iden-tification of genetically diverse sequences (ORF 5) of porcine reproductive andrespiratory syndrome virus in a swine herd. Can. J. Vet. Res. 65, 254–260.
ang, Y., Schneider, P., Zhang, W.P., Faaberg, K.S., Nelson, E.A., Rowland, R.R., 2007.Diversity and evolution of a newly emerged North American Type 1 porcinearterivirus: analysis of isolates collected between 1999 and 2004. Arch. Virol.152 (5), 1009–1017.
orsberg, R., Storgaard, T., Nielsen, H.S., Oleksiewicz, M.B., Cordioli, P., Sala, G., Hein,J., Bøtner, A., 2002. The genetic diversity of European type PRRSV is similar tothat of the North American type but is geographically skewed within Europe.Virology 299, 38–47.
ao, Z.Q., Guo, X., Yang, H.C., 2004. Genomic characterization of two Chinese isolatesof porcine respiratory and reproductive syndrome virus. Arch. Virol. 149 (7),1341–1351.
arner, M.G., Whan, I.F., Gard, G.P., Phillips, D., 2001. The expected economic impactof selected exotic diseases on the pig industry of Australia. Rev. Sci. Tech. 20 (3),671–685.
apur, V., Elam, M.R., Pawlovich, T.M., Murtaugh, M.P., 1996. Genetic variation inporcine reproductive and respiratory syndrome virus isolates in the midwesternUnited States. J. Gen. Virol. 77 (Pt 6), 1271–1276.
arochelle, R., Magar, R., 1997. Differentiation of North American and Europeanporcine reproductive and respiratory syndrome virus genotypes by in situhybridization. J. Virol. Methods 68, 161–168.
i, B., Fang, L., Xu, Z., Liu, S., Gao, J., Jiang, Y., Chen, H., Xiao, S., 2009a. Recombina-tion in vaccine and circulating strains of porcine reproductive and respiratorysyndrome viruses. Emerg. Infect. Dis. 15 (12), 2032–2035.
i, Y., Wang, X., Jiang, P., Chen, W., Wang, K., 2009b. Genetic variation analysis ofporcine reproductive and respiratory syndrome virus isolated in China from2002 to 2007 based on ORF5. Vet. Microbiol. 138 (1–2), 150–155.
ole, K.S., Bollinger, R.C., Paranjape, R.S., Gadkari, D., Kulkarni, S.S., Novak, N.G., Inger-soll, R., Sheppard, H.W., Ray, S.C., 1999. Full-length human immunodeficiencyvirus type 1 genomes from subtype C-infected seroconverters in India, withevidence of intersubtype recombination. J. Virol. 73 (1), 152–160.
ardassi, H., Mounir, S., Dea, S., 1995. Structural gene analysis of a Quebec referencestrain or porcine reproductive and respiratory syndrome virus (PRRSV). Adv.Exp. Med. Biol. 380, 277–281.
artelli, P., Gozio, S., Ferrari, L., Rosina, S., De Angelis, E., Quintavalla, C., Bottarelli, E.,Borghetti, P., 2009. Efficacy of a modified live porcine reproductive and respira-tory syndrome virus (PRRSV) vaccine in pigs naturally exposed to a heterologous
155 (2011) 473–486
European (Italian cluster) field strain: clinical protection and cell-mediatedimmunity. Vaccine 27 (28), 3788–3799.
Mateu, E., Martín, M., Vidal, D., 2003. Genetic diversity and phylogenetic analysisof glycoprotein 5 of European-type porcine reproductive and respiratory virusstrains in Spain. J. Gen. Virol. 84 (Pt 3), 529–534.
Meng, X.J., 2000. Heterogeneity of porcine reproductive and respiratory syndromevirus: implications for current vaccine efficacy and future vaccine development.Vet. Microbiol. 74 (4), 309–329.
Meng, X.J., Paul, P.S., Halbur, P.G., Lum, M.A., 1995a. Phylogenetic analyses of theputative M (ORF 6) and N (ORF 7) genes of porcine reproductive and respira-tory syndrome virus (PRRSV): implication for the existence of two genotypes ofPRRSV in the USA and Europe. Arch. Virol. 140 (4), 745–755.
Meng, X.J., Paul, P.S., Halbur, P.G., Morozov, I., 1995b. Sequence comparison ofopen reading frames 2–5 of low and high virulence United States isolates ofporcine reproductive and respiratory syndrome virus. J. Gen. Virol. 76 (Pt 12),3181–3188.
Murtaugh, M.P., Yuan, S., Faaberg, K.S., 2001. Appearance of novel PRRSV isolates byrecombination in the natural environment. Adv. Exp. Med. Biol. 494, 31–36.
Nelson, E.A., Christopher-Hennings, J., Drew, T., Wensvoort, G., Collins, J.E., Benfield,D.A., 1993. Differentiation of US and European isolates of porcine reproductiveand respiratory syndrome virus by monoclonal antibodies. J. Clin. Microbiol. 31(12), 3184–3189.
Neumann, E.J., Kliebenstein, J.B., Johnson, C.D., Mabry, J.W., Bush, E.J., Seitzinger,A.H., Green, A.L., Zimmerman, J.J., 2005. Assessment of the economic impactof porcine reproductive and respiratory syndrome on swine production in theUnited States. J. Am. Vet. Med. Assoc. 227 (3), 385–392.
Pirzadeh, B., Gagnon, C.A., Dea, S., 1998. Genomic and antigenic variations of porcinereproductive and respiratory syndrome virus major envelope GP5 glycoprotein.Can. J. Vet. Res. 62, 170–177.
Ran, Z.G., Chen, X.Y., Guo, X., Ge, X.N., Yoon, K.J., Yang, H.C., 2008. Recovery of viableporcine reproductive and respiratory syndrome virus from an infectious clonecontaining a partial deletion within the Nsp2-encoding region. Arch. Virol. 153(5), 899–907.
Rowland, R.R., Steffen, M., Ackerman, T., Benfield, D.A., 1999. The evolution of porcinereproductive and respiratory syndrome virus: quasispecies and emergence ofa virus subpopulation during infection of pigs with VR-2332. Virology 259,262–266.
Snijder, E.J., van Tol, H., Roos, N., Pedersen, K.W., 2001. Non-structural proteins 2 and3 interact to modify host cell membranes during the formation of the arterivirusreplication complex. J. Gen. Virol. 82 (Pt 5), 985–994.
Stadejek, T., Stankevicius, A., Storgaard, T., Oleksiewicz, M.B., Belak, S., Drew, T.W.,Pejsak, Z., 2002. Identification of radically different variants of porcine repro-ductive and respiratory syndrome virus in eastern Europe: towards a commonancestor for European and American viruses. J. Gen. Virol. 83 (Pt 8), 1861–1873.
Stadejek, T., Oleksiewicz, M.B., Potapchuk, D., Podgórska, K., 2006. Porcine reproduc-tive and respiratory syndrome virus strains of exceptional diversity in easternEurope support the definition of new genetic subtypes. J. Gen. Virol. 87 (Pt 7),1835–1841.
Tian, Z.J., An, T.Q., Zhou, Y.J., Peng, J.M., Hu, S.P., Wei, T.C., Jiang, Y.F., Xiao, Y., Tong,G.Z., 2009. An attenuated live vaccine based on highly pathogenic porcine repro-ductive and respiratory syndrome virus (HP-PRRSV) protects piglets againstHP-PRRS. Vet. Microbiol. 138 (1–2), 34–40.
van Vugt, J.J., Storgaard, T., Oleksiewicz, M.B., Botner, A., 2001. High frequency RNArecombination in porcine reproductive and respiratory syndrome virus occurspreferentially between parental sequences with high similarity. J. Gen. Virol. 82(Pt 11), 2615–2620.
Wootton, S., Yoo, D., Rogan, D., 2000. Full-length sequence of a Canadian porcinereproductive and respiratory syndrome virus (PRRSV) isolate. Arch. Virol. 145(11), 2297–2323.
Xia, P., Dang, Z., Qiu, H., Fan, X., Zhou, B., Cui, B., Chen, P., 2009. Function of PRRSVGP5 envelope protein by using pseudotyped virus. Vet. Microbiol. 138 (3–4),297–303.
Yuan, S., Nelsen, C.J., Murtaugh, M.P., Schmitt, B.J., Faaberg, K.S., 1999. Recombina-tion between North American strains of porcine reproductive and respiratorysyndrome virus. Virus Res. 61 (1), 87–98.
Zhou, L., Zhang, J., Zeng, J., Yin, S., Li, Y., Zheng, L., Guo, X., Ge, X., Yang, H., 2009b. The30-amino-acid deletion in the Nsp2 of highly pathogenic porcine reproductiveand respiratory syndrome virus emerging in China is not related to its virulence.J. Virol. 83 (10), 5156–5167.
