A multiplex RT-PCR assay for the rapid and differential diagnosis of classical swine fever and othe

A multiplex RT-PCR assay for the rapid and differential diagnosis of classical swine fever and other pestivirus infectionsHeidy Dı´az de Arce a ,*,Lester J.Pe ´rez a ,Maria T.Frı´as a ,Rosa Rosell b ,c ,Joan Tarradas b ,Jose´I.Nu ´n ˜ez b ,Llilianne Ganges b aCentro Nacional de Sanidad Agropecuaria (CENSA),La Habana,CubabCentre de Recerca en Sanitat Animal (CReSA),UAB Bellaterra,08193Barcelona,Spain c Departament d´Agricultura,alimentacio ´n i Accio ´Rural de la Generalitat de Catalunya (DAR),Spain 1.IntroductionClassical swine fever (CSF),a highly contagious viral infection of domestic pig and wild boar,is one of the most devastating porcine diseases worldwide (Moennig et al.,2003).The disease is endemic in Asia and prevails in many Central and South American countries,as well as in Eastern Europe (Greiser-Wilke et al.,2007).The classical swine fever virus (CSFV)together with the viruses of bovine viral diarrhea types 1and 2(BVDV-1and BVDV-2)and Border disease (BDV)belongs to the genusPestivirus within the family Flaviviridae.The pestiviruses have a positive single-strand RNA genome of approxi-mately 12.3–16kb in length that encodes one large open reading frame (ORF),flanked by highly conserved 50-and 30-untranslated regions (UTR).The 50-UTR has secondary structure and highly conserved regions (Meyers and Thiel,1996).The CSFV non-structural protein 5B (NS5B)encodes an RNA dependent RNA polymerase (RdRp),pestivirus polymerase amino acid sequences are 70–75%identical and still appears to be a conservation of tertiary structure rather than primary sequence (Zhang et al.,2005)and is a valid candidate to CSFV and non-CSFV pestiviruses differentiation.CSFV is predominantly restricted to pigs;viruses of bovine viral diarrhea and Border disease infect many different ruminant species as well as swine (van Rijn,Veterinary Microbiology 139(2009)245–252A R T I C L E I N F O Article history:Received 26December 2008Received in revised form 25May 2009Accepted 3June 2009Keywords:Classical swine fever virus PestivirusMultiplex RT-PCR Differential diagnosisA B S T R A C TClassical swine fever is a highly contagious viral disease causing severe economic losses in pig production almost worldwide.All pestivirus species can infect pigs,therefore accurate and rapid pestivirus detection and differentiation is of great importance to assure control measures in swine farming.Here we describe the development and evaluation of a novel multiplex,highly sensitive and specific RT-PCR for the simultaneous detection and rapid differentiation between CSFV and other pestivirus infections in swine.The universal and differential detection was based on primers designed to amplify a fragment of the 50non-coding genome region for the detection of pestiviruses and a fragment of the NS5B gene for the detection of classical swine fever virus.The assay proved to be specific when different pestivirus strains from swine and ruminants were evaluated.The analytical sensitivity was estimated to be as little as 0.89TCID 50.The assay analysis of 30tissue homogenate samples from naturally infected and non-CSF infected animals and 40standard serum samples evaluated as part of two European Inter-laboratory Comparison Tests conducted by the European Community Reference Laboratory,Hanover,Germany proved that the multiplex RT-PCR method provides a rapid,highly sensitive,and cost-effective laboratory diagnosis for classical swine fever and other pestivirus infections in swine.ß2009Elsevier B.V.All rights reserved.*Corresponding author.Tel.:+5347863206;fax:+5347861104.E-mail addresses:heidy@.cu ,heidydal@yahoo.es (H.D.d.Arce).Contents lists available at ScienceDirectVeterinary Microbiologyj o u rn a l ho m e pa g e :w w w.e l s e v i e r.c o m /l o c a t e /v e t m i c0378-1135/$–see front matter ß2009Elsevier B.V.All rights reserved.doi:10.1016/j.vetmic.2009.06.0042007).Infections in swine with pestiviruses other than CSFV can give rise to a clinical disease that is indistinguishable from CSF(Terpstra and Wensvoort, 1988;Paton et al.,1992;Frey et al.,1995).The rapid and accurate differentiation of CSFV from non-CSFV pesti-viruses is decisive because whereas CSFV infection of domestic pigs demands strict control measures to prevent vast economic losses for other pestivirus infections in pigs,no immediate zoo-sanitary measures are prescribed.Since different pestiviruses are closely related,both immunologically and genetically,the ruminant’s pesti-virus infections in swine can result into a false CSF diagnosis(OIE,2008a).Virus neutralization tests(VNT), monoclonal antibody assays and nucleic acid based techniques are the methods described for pestivirus discrimination(Edwards et al.,1991;Wirz et al.,1993; Sullivan and Akkina,1995;McGoldrick et al.,1998;Barlic-Maganja and Grom,2001;OIE,2008a).While procedures based on monoclonal antibodies have limitations such as moderate sensitivity(identification on cryostat sections of organs and antigen capture ELISA)or long turn-around time(identification after virus isolation or VNT),standard (gel-based)and real time RT-PCR assays provide very useful tools for rapid and highly sensitive detection (Edwards et al.,1991;Katz et al.,1993;Wirz et al., 1993;Sullivan and Akkina,1995;Gilbert et al.,1999; Barlic-Maganja and Grom,2001;Blome et al.,2006;Bela´k, 2007).Several RT-PCR assays(Katz et al.,1993;Wirz et al., 1993;Canal et al.,1996;Sandvik et al.,1997;Dı´az de Arce et al.,1998),one RT-PCR–ELISA(Barlic-Maganja and Grom,2001)and a multiplex DNA microarray method(Deregt et al.,2006)have been reported for the detection and discrimination between classical swine fever virus and other pestivirus infections in swine.Albeit some of these assays have limitations for routine diagnostic purposes.In particular,the method described by Wirz et al.(1993)has poor sensitivity for CSFV specific detection in organ extracts and the nested RT-PCR assay is not able to discriminate between CSFV and BDV(Sandvik et al.,1997).Moreover some protocols have not proved to be suitable for routine diagnosis since none have been assessed in swine clinical samples.Real-time RT-PCR assays have been reported for rapid and sensitive detection of pestiviruses and CSFV differ-entiation(Cheng et al.,2008;Hoffmann et al.,2005; Jamnikar Ciglenecki et al.,2008;Liu et al.,2007; McGoldrick et al.,1999).Despite of the many advantages of the real-time RT-PCR techniques over conventional gel-based RT-PCR assays,including rapidity,a higher sensi-tivity and a lower contamination rate,the high costs of the instruments and probes required to perform the real-time assays restricted their use to laboratories with good financial resources.Therefore,the ultimate objective of this work was the development,optimization and evaluation of a novel and highly sensitive multiplex RT-PCR assay for the simulta-neous detection and rapid differentiation between CSFV and other pestivirus infections.2.Materials and methods2.1.VirusesCSFV and non-CSFV pestiviruses strains available in ours laboratories as well as other porcine viruses used throughout this study are listed in Table1.2.2.Cell cultures and virus propagationViruses were grown and titrated in PK15cell line(CSFV strains)and MDBK cell line(BVDV and BDV strains) following standard procedures as described at the OIE Manual(OIE,2008a,b,c).The Cuban EMCV744/03strain was propagated on BHK-21cells in MEM supplemented with5%FBS.After approximately16h incubation at378C, when more than80%of the cells showed cytopathology, the cells were exposed to three freeze–thaw cycles.PK15, MDBK,and BHK-21non-infected cell cultures were also used in the specificity assays.The cell cultures and media were previously tested to be free of contaminating pestivirus(Bolin et al.,1994;Edwards,1993).2.3.Experimental standard clinical samplesA panel of40pig serum,from two International Inter-laboratory Comparison Tests2006and2007,conducted by the CRL,Hanover,Germany,aimed to evaluate the quality and diagnostic performance of RT-PCR assays in different laboratories for the detection of pestivirus infections in pigs was used.The serum samples were obtained by experimental infection of pigs with pestivirus strains from different subtypes and genotypes and comprised samples with different titers of the virus.BVDV positive fetal calve serum and negative commercial pig serum were also included.The serum samples were tested at least three times with all routine procedures at the CRL.The methods used at CRL to obtain the RT-PCR results for CSFV and panpestivirus detection were the real time RT-PCR assay described by Hoffmann et al.(2005)and one SYBR Green based real-time RT-PCR using the primer pair described by Vilcek et al.(1994),respectively.2.4.Field samplesA collection of18tissue homogenate samples(lymph nodes,spleens or tonsils)from naturally infected animals from CSFV Cuban outbreaks and12healthy non-pestivirus infectedfield animals were used for the assessment of the assays performance in organs.Samples were confirmed to be CSFV positive or negative by virus isolation and identification,as it is described in the Manual of Diagnostic Tests and Vaccines for Terrestrial Animals2008(OIE, 2008a).2.5.Primers selectionCSFV specific primer pair CSFV1/CSFV2,previously described(Dı´az de Arce et al.,1998),was employed in the multiplex RT-PCR.This primer pair was identified within the sequence region corresponding to the NS5BH.D.Arce et al./Veterinary Microbiology139(2009)245–252 246protein with the aid of a computer program that allowed selection of sequences highly conserved among CSFV strains but that were also divergent enough when compared to other pestiviruses(Dı´az de Arce et al.,1998).Panpestivirus specific primers were designed in the conserved50-UTR of viruses from different pestiviral genotypes and subgenotypes,available in GenBank. Nucleotide sequences belonging to genotypes BVDV-1 (BVDV-1a and BVDV-1b),BVDV-2,CSFV and BDV as well as hepatitis C virus were aligned using Clustal W1.8software and manually examined for oligonucleotides correspond-ing to pestivirus.Primers were designed from highly conserved nucleotide region using the Oligo6.31program (Molecular Biology Insights,Inc.,USA).A BLAST search at National Center for Biotechnology Information(NCBI)site ()was performed using blastn algorithm for calculating sequence similarity with primers selected as query sequences against nucleotide databases of different pestivirus genotypes and random nucleotide sequences.The oligonucleotide sequence of the CSFV and panpes-tivirus primer sets and their main characteristics are summarized in Table2.2.6.RNA extraction and cDNA synthesisTotal RNA was extracted from250m L amounts of samples(cell cultures,serum,tissue homogenate),with aTable1Virus used in this study.Virus Genotype/subtype Reference strain/isolate SourceCSFV Genotype1.1Alfort187CReSA,SpainAmes CENSA,CubaClinical samples from experimentallyinfected pigs with CSF0382reference strainEU Reference Laboratory for Classical Swine Fever,GermanyGenotype1.2PAV-250Centro de Investigacio´n en Sanidad Animal,SpainMargarita a CENSA,Cuba15Cubanfield isolates a CENSA,CubaClinical samples from experimentallyinfected pigs with Kozlov reference strainEU Reference Laboratory for Classical Swine Fever,Germany Genotype2.2Clinical samples from experimentallyinfected pigs with CSF0018reference strainEU Reference Laboratory for Classical Swine Fever,GermanyGenotype2.3Clinical samples from experimentallyinfected pigs with CSF0864reference strainEU Reference Laboratory for Classical Swine Fever,GermanyClinical samples from experimentallyinfected pigs with CSF0634reference strainEU Reference Laboratory for Classical Swine Fever,Germany Genotype3.4Clinical samples from experimentallyinfected pigs with CSF0309reference strainEU Reference Laboratory for Classical Swine Fever,GermanyBVDV-1Subtype1a NADL EU Reference Laboratory for Classical Swine Fever,GermanyOregon Centro de Investigacio´n en Sanidad Animal,SpainSinger Centro de Investigacio´n en Sanidad Animal,Spain Subtype1b Osloss EU Reference Laboratory for Classical Swine Fever,Germany Non-determinated7field isolates from cattle,Cuba/2002-2003CENSA,Cuba8field isolates from cattle,Spain/1990-1994b CReSA/DAR,SpainBVDV-2New York Hipra laboratory,SpainBDV Genotype1Frijters id-dlo Laboratory,Netherland.Moredun EU Reference Laboratory for Classical Swine Fever,Germany137/4Central Veterinary Laboratory of Weybridge,United Kingdom Genotype3Clinical samples from experimentally infectedpigs with Gifhorn reference strainEU Reference Laboratory for Classical Swine Fever,Germany Genotype45BDV isolates from sheep c CReSA/DAR,Spain2BDV isolates from sheep Neiker,SpainBDV isolated from pig Spain/2006d CReSA,SpainTGEV Purdue115Pathobiology laboratory,Gelph University,CanadaEMCV Field isolate CENSA,CubaDı´az de Arce et al.(2005).b Vega et al.(2000).c Vega et al.(2002).d Rosell et al.(2008).Table2Selected CSFV and panpestivirus primer sets.Primer Nucleotide sequence(50–30)Genome position(50–30)Amplicon length(bp) CSFV1CCTGAGGACCAAACACATGTTG nt10.240–10.261aCSFV2TGGTGGAAGTTGGTTGTGTCTG nt10.413–10.392a174Panpest1CGAGATGCCACGTGGACGAG nt226–245bPanpest2CCTCTGCAGCACCCTATCAGG nt344–324b119c Genome position according to CSFV Alfort/187(Accession no.X87939.1).b Genome position according to BVDV strain NADL(Accession no.AF039181.1).c The length vary from114bp of some CSFV to119of some BVDV strains.H.D.Arce et al./Veterinary Microbiology139(2009)245–252247commercial reagent(TRI Reagent LS,SIGMA,San Louis, Missouri,USA),as recommended by the supplier.RNA was resuspended in10m L of nuclease free water(Promega, Madison,WI,USA)and used in the assay.First strand complementary DNA(cDNA)was synthe-sized using Moloney-Murine leukemia virus reverse tran-scriptase(M-MLV RT)(Promega,Madison,WI,USA)in20m L final reaction volume.Briefly,5m L of RNA was incubated with1m L of random primers(50ng/m L)(Promega,Madison, WI,USA)and4m L of nuclease free water(Promega, Madison,WI,USA)in10m Lfinal reaction volume at708C for10min and then cooled on ice to accomplish nucleic acid denaturing.After incubation on ice3.5m L of nuclease free water,4m L of5Âreaction buffer[250mM Tris–HCl(pH8.3 at258C),375mM KCl,15mM MgCl2,50mM DTT],1m L 10mM of each deoxynucleoside triphosphate,0.5m L of 40U/m L RNAs in ribonuclease inhibitor(Promega)and1m L of200U/m L M-MLV RT was added and the reaction mixture was further incubated at378C for60min.2.7.Optimization of the CSFV/panpestivirus multiplex RT-PCR assayA number of experiments were performed to optimize the CSFV/panpestivirus multiplex RT-PCR protocol,includ-ing reagent concentration and PCR cycling parameters.The primers,as well as magnesium chloride,were titrated in chequerboard assays to determine the reagents concentra-tions that yielded the best results.Different annealing/ elongation temperatures and times were evaluated.The assays werefinally optimized as follows:for PCR amplifica-tion,the reaction mixture was prepared in a volume of50m L comprised of2m L of cDNA,1ÂGoTaq Green Master Mix (Promega)[200m M of each dNTP,1.5mM MgCl2(pH8.5)] and1m M of each primer.The PCR reaction was done under the following conditions in a thermal cycler(Eppendorf Mastercycler):1cycle of2min at958C;35cycles of denaturation at948C for30s,annealing at508C for30s,and elongation at728C for30s;and1cycle of5min at728C.The amplified products were analyzed by electrophoresis in a2% agarose gel,stained with ethidium bromide0.5g/mL)in TBE buffer pH8.4(89mM Tris;89mM boric acid;2mM EDTA) and visualized under UV light.As control,at least one negative serum for each batch of RNA extracted samples tested and one water control were included in the PCR. 2.8.Sequences determination and analysisAmplicons of Margarita strain obtained from CSFV/ panpestivirus multiplex RT-PCR assays were purified from the agarose gel using Wizard1PCR Preps DNA Purification System(Promega),and sequenced by BigDye Terminator v3.1Cycle Sequencing Kit(Applied Biosystems)in an ABIPRISM3730.Quality of each sequence obtained was analyzed manually and the sequence similarity was checked against sequences deposited in the EMBL/Gen-Bank using a BLAST search at NCBI site(http:// ).Both sequences,CSFV and panpestivirus multiplex RT-PCR fragments,have been submitted to the EMBL/GenBank database with the accession numbers FM954979and FM954980,respectively.2.9.Analytical specificityThe analytical specificity of the multiplex RT-PCR assays was evaluated by analysis of viral RNA genomes from representatives of BVDV-1,BVDV-2,BDV,and the major subgroups(1.1,1.2,2.1,2.2,2.3and3.4)of CSFV strains(Table1).2.10.Analytical sensitivityRT-PCR assays were evaluated by testing viral RNAs purified from sequential tenfold dilutions in negative serum of a viral suspension of Alfort/187strain with a titre of107.86TCID50/mL.The analytical sensitivity of the assays was determined by considering the volumes of specimen used for each RNA extraction(250m L),RNA dissolution (10m L nuclease free water),the RNA added to each cDNA reaction(5m L)and the cDNA added(2m L)to each PCR reaction(represented by TCID50per reaction).2.11.CSFV and panpestivirus simultaneous detection by multiplex RT-PCRIn order to assess the possible preferential amplification of one target sequence over another,different known amounts of BVDV(NADL)and CSFV(Alfort/187)strains were diluted in serum from a non-infected donor pig and virus sample mixtures were evaluated.All possible sample mixtures of each virus included within sensitivity rate of the test were analyzed.2.12.Application of multiplex RT-PCR to standard clinical samplesTo determine the ability of the multiplex RT-PCR method to detect different pestivirus genotypes and virus concentrations as well as the competence of the RT-PCR test in clinical samples,40standard pig sera samples(20 each)from two European Inter-laboratory Comparison Tests2006and2007for Classical Swine Fever conducted by the European Community Reference Laboratory,Hann-over,Germany were evaluated.The serum samples were obtained from experimentally infected animals at different post-inoculation days with different CSFV strains that included a wide range of genotypes(Table3).Besides,BVDV positive fetal calve serum and negative pig serum were included.The samples contained a full representation of the virus concentration to test sensitivity and were sent in duplicates in order to test reproducibility.3.Results3.1.Primer selection and optimization of the RT-PCRTwo primer pairs,each one specific for CSFV and panpestivirus,were collectively selected in a multiplex RT-PCR.Well-conserved CSFV genome but divergent from other pestivirus and highly Pestivirus conserved regions were chosen as the best candidates for specific primers setting.Primer design parameters such as annealingH.D.Arce et al./Veterinary Microbiology139(2009)245–252 248temperatures,homologies either internally or to one another and target sizes were also considered.The optimization of the multiplex RT-PCR reaction components and cycling parameters was performed.Most suitable combination of the two primer sets concentration was meticulously selected.Finally,the optimized concen-trations of reaction components and the thermal profile were established and used in further experiments(as described in Section2).3.2.Specificity of the multiplex RT-PCR assayWith the panpestivirus specific primers,designed in a highly conserved region from the50-UTR RNA genome from different pestivirus strains,PCR products of the expected sizes(114or119bp)were amplified from all of the examined pestivirus strains and isolates in multiplex RT-PCR assay.A PCR product from the expected size (174bp)was obtained when RNA from all CSFV strains and isolates were used as templates for amplification reactions in the multiplex RT-PCR test.Moreover,all ruminant pestivirus(non-CSFV)strains and isolates tested in the multiplex RT-PCR were negative for CSFV primers ampli-fication.Some representative results are shown in Fig.1.No amplification signals were obtained from hetero-logous RNA porcine viruses(TGEV and EMCV)tested in both RT-PCR assays.Finally,nucleic acids from PK15, MDBK,and BHK-21non-infected cell cultures were also tested showing no positive products(data not shown).The BLAST analysis of the amplicon sequences corre-sponding to174bp,amplified from CSFV strains,and114 or119bp(amplicon length depending of the strain) amplified from pestivirus strains by multiplex RT-PCR assay,confirmed that the PCR products were,respectively, CSFV and panpestivirus specifics.3.3.Sensitivity of multiplex RT-PCR assayThe detection limit using negative serum spiked with serially diluted virus suspensions was consistently observed to be0.89TCID50per amplification reaction for the multi-plex RT-PCR as well as for the two single PCR assays(Fig.2).Table3Multiplex RT-PCR results obtained from the samples evaluated in the2006and2007Inter-laboratory Comparison Tests.Samples Code Virus strain/group Virus titer(TCID50/mL)Multiplex(panpest/CSFV)CRL(non-CSFV/CSFV)PCR-1/2006CSFV Kozlov/1.2101.6+/++/+PCR-2/2006CSFV Kozlov/1.2103.6+/++/+PCR-3/2006CSFV Kozlov/1.2103.6+/++/+PCR-4/2006CSFV Kozlov/1.2103.8+/++/+PCR-5/2006BDV Gifhorn/3n.d*+/À+/ÀPCR-6/2006BVDV Osloss/1b n.d*+/À+/ÀPCR-7/2006CSFV Kozlov/1.2101.8+/++/+PCR-8/2006CSFV Kozlov/1.2103.8+/++/+PCR-9/2006CSFV Kozlov/1.2103.6+/++/+PCR-10/2006CSFV Kozlov/1.2102.8+/++/+PCR-11/2006CSFV Kozlov/1.2102.6+/++/+PCR-12/2006CSFV Kozlov/1.2101.8+/++/+PCR-13/2006CSFV Kozlov/1.2104.8+/++/+PCR-14/2006CSFV Kozlov/1.2102.6+/++/+PCR-15/2006CSFV Kozlov/1.2101.8+/++/+PCR-16/2006CSFV Kozlov/1.2102.8+/++/+PCR-17/2006CSFV Kozlov/1.2102.6+/++/+PCR-18/2006CSFV Kozlov/1.2101.6+/++/+PCR-19/2006CSFV Kozlov/1.2101.6+/++/+PCR-20/2006CSFV Kozlov/1.2102.8+/++/+PCR-1/2007CSF0864/2.3103.75+/++/+PCR-2/2007Non-virus(pig serum)–À/ÀÀ/ÀPCR-3/2007CSF0864/2.3101.75+/++/+PCR-4/2007CSF0864/2.3101.75+/++/+PCR-5/2007CSF0018/2.2104.50+/++/+PCR-6/2007CSF0382/1.1103.75+/++/+PCR-7/2007Non-virus(pig serum)–À/ÀÀ/ÀPCR-8/2007Non-virus(pig serum)–À/ÀÀ/ÀPCR-9/2007Pestivirus non-CSFV(Fetalcalf serum diluted1:2)–+/À+/ÀPCR-10/2007Pestivirus non-CSFV(Fetalcalf serum diluted1:2)–+/À+/ÀPCR-11/2007CSF0864/2.3103.75+/++/+PCR-12/2007Non-virus(pig serum)–À/ÀÀ/ÀPCR-13/2007CSF0018/2.2104.5+/++/+PCR-14/2007CSF0018/2.2n.d*(sample code PCR-5/2007diluted1:103)+/++/+PCR-15/2007CSF0018/2.2n.d*(sample code PCR-5/2007diluted1:104)+/++/+PCR-16/2007CSF0018/2.2n.d*(sample code PCR-5/2007diluted1:105)À/À+/+PCR-17/2007CSF0018/2.2104.50+/++/+PCR-18/2007CSF0018/2.2n.d*(sample code PCR-5/2007diluted1:103)+/++/+PCR-19/2007CSF0018/2.2n.d*(sample code PCR-5/2007diluted1:104)+/++/+PCR-20/2007CSF0018/2.2n.d*(sample code PCR-5/2007diluted1:105)À/À+/+ non-determined.H.D.Arce et al./Veterinary Microbiology139(2009)245–2522493.4.CSFV and panpestivirus simultaneous detection by multiplex RT-PCRThe multiplex assay detected all variable numbers of viruses spiked onto pig serum according to their propor-tions.The lowest detection limit of the multiplex assay for CSFV and for other pestiviruses was the same when only CSFV or CSFV plus other pestivirus was evaluated in the same reaction (data not shown).Hence,each target inthe mixture reaction had the same amplification efficiency when both targets were evaluated simultaneously.3.5.Application of multiplex RT-PCR assay to standard samplesAmplification from two samples (code numbers PCR-16/2007and PCR-20/2007),failed for both CSFV and non-CSFV detection since they were duplicates from a CSFV low-titer serum (Table 3).Therefore,the sensitivity of the assay was 94%for both CSFV (30/32)and non-CSFV virus (34/36)detection (Table 3).The specificity of the assay was 100%with no false positives for both CSFV (8/8)and non-CSFV (4/4)detection (Table 3).3.6.Application of multiplex RT-PCR to field samples Multiplex RT-PCR assay was performed on a collection of 18CSFV tissue field samples (lymph nodes,spleen or tonsil)and 12tissue samples from healthy animals.Specific CSFV was detected in all clinical samples from natural infected pigs and no positive results were obtained from non-infected animals (data not shown).4.DiscussionThe multiplex PCR methods use multiple primers to allow amplification of multiple templates within a single reaction at the same time providing the possibility to detect more than one infectious agent(s)in a single assay.Multiplex PCR has the potential to produce considerable savings of time and effort within the laboratory without compromising the robustness and sensitivity of the virusdetection assays (Bela´k,2007).Additional advantage is a reduced sample requirement,which is particularly impor-tant when sample material is limited (Persson et al.,2005).Multiplex gel-based RT-PCR tests for simultaneous detec-tion of multiple porcine viruses have been documented(Agu¨ero et al.,2004;Cheng et al.,2008;Ferna ´ndez et al.,2008;Giammarioli et al.,2008;Zhao et al.,2008).The development of a multiplex PCR method is a complex task mainly due to the fact that the presence of more than one primer pair in the same reaction mix can limit the sensitivity or specificity and/or cause preferential amplification of specific targets (Elnifro et al.,2000;Markoulatos et al.,2002).In this study,a gel-based multiplex RT-PCR assay was developed for the simultaneous detection and rapid discrimination between CSFV and other pestiviruses infections in swine.Primer selection and protocol optimi-zation were considered to be the key factors and were carefully addressed.The CSFV specific primers used in thisassay,as described previously (Dı´az de Arce et al.,1998),were designed in the NS5B gene of CSFV genome from a region highly conserved between the genome of CSFV strains,but which was highly divergent from the corresponding sequences in the genome of other pestivirus strains.Panpestivirus specific primers were designed in the conserved 50-UTR of viruses from different pestiviral genotypes sequences and from hepatitis C virus,a related genome belonging to the Flaviviridae family,in ordertoFig.1.Specificity of multiplex RT-PCR for CSFV (1–4)and panpestivirus (5–8).(1)CSFV strain Alfort/187;(2)CSFV strain Ames;(3)CSFV strain Margarita;(4)CSFV strain PAV-250;(5)BVDV strain NADL;(6)BVDV strain Oregon;(7)BVDV strain Singer;(8)BDV strain Moredun;M:molecular weight marker 100bp (Promega,Madison,WI,USA).Fig. 2.Multiplex RT-PCR and single RT-PCR sensitivity assays.RNAs extracted from serial,tenfold dilutions of CSFV strain Alfort/187with an initial titer of 107.86TCID 50/mL in negative pig serum were employed under reaction conditions described in Section 2;M:molecular weight marker 100bp (Promega,Madison,WI,USA.(A)Multiplex RT-PCR sensitivity assay for CSFV and panpestivirus detection.(B)RT-PCR sensitivity assay for panpestivirus detection.(C)RT-PCR sensitivity assay for CSFV detection.H.D.Arce et al./Veterinary Microbiology 139(2009)245–252250obtain a pair of primers with broad pestivirus specificity and that does not cross-react with related viruses.To ensure specificity,genus-primers were designed to detect as many viral sequences as necessary to encompass genetic diversity. The pestivirus50-UTR contains stretches of highly conserved nucleotides and has frequently been targeted for pestivirus specific PCR primers(Hamel et al.,1995;Canal et al.,1996; Sandvik et al.,1997;Barlic-Maganja and Grom,2001).The RT-PCR test proved to detect the major genotypes of CSFV and demonstrated to be very sensitive because the experimental data showed a limit of detection of less than 1TCID50/PCR reaction.In addition,positive amplifications were obtained in all the tissue samples,from CSFV naturally infected pigs,evaluated.The detection from clinical samples from infected animals demonstrates the potential usefulness of the method for a rapid disease diagnosis fromfield cases.Nevertheless an adequate number of samples from the suspected herd should be tested to prevent a positive result in the pestivirus PCR reaction with a false negative result for CSFV detection which could cause a misleading diagnosis.The assessment of the multiplex RT-PCR assay in two proficiency testing,conducted by the European Commu-nity Reference Laboratory for CSF and which included differential diagnosis with other pestiviruses,demon-strated the ability of the assay to detect different CSF genotypes as well as a full representation of virus concentration from standard samples.Only a duplicated sample was not detected.This is explained because of the low-titer of these samples that was below the detection limit of the assay.However,the detection limit of the multiplex RT-PCR should be able to detect pigs infected not only with highly virulent but also with moderately and low virulent strains at an early stage(Weesendorp et al.,2009).This assay was designed to include two systems,a ‘‘general’’to detect genus Pestivirus and a specific to detect CSF infections in a single reaction.The parallel use of ‘‘general’’and specific PCR assays provides a rapid and effective diagnosis of pestivirus infections in pigs(Bela´k, 2007).The multiplex RT-PCR assay described here provides a sensitive tool for simultaneous detection and rapid differentiation between CSFV and other pestiviruses infections in swine.The reaction can be performed in a rapid and cost-effective manner.This method could be a good alternative for diagnostic laboratories with limited economic resources probably located in developing nations where CSF outbreaks occur frequently. AcknowledgementsWork at CENSA was supported by the project00300248 from the Cuban Ministry of Science,Technology and Environment.Work at CReSA was supported by the project CONSOLIDER-INGENIO2010CDS2006-00007from the Spanish Government.ReferencesAgu¨ero,M.,Ferna´ndez,J.,Romero,L.J.,Zamora,M.J.,Sanchez,C.,Bela´k,S., Arias,M.,Sanchez-Vizcaino,J.M.,2004.A highly sensitive and specific gel-based multiplex RT-PCR assay for the simultaneous and differ-ential diagnosis of African swine fever and classical swine fever in clinical samples.Vet.Res.35,551–563.Barlic-Maganja,D.,Grom,J.,2001.Highly sensitive one-tube RT-PCR and microplate hybridisation assay for the detection and for the discri-mination of classical swine fever virus from other pestiviruses.J.Virol.Methods95,101–110.Bela´k,S.,2007.Molecular diagnosis of viral diseases,present trends and future aspects a view from the OIE Collaborating Centre for the application of polymerase chain reaction methods for diagnosis of viral diseases in veterinary medicine.Vaccine25,5444–5452. Blome,S.,Meindl-Bohmer,A.,Loeffen,W.,Thuer,B.,Moennig,V.,2006.Assessment of classical swine fever diagnostics and vaccine perfor-mance.Rev.Sci.Technol.25,1025–1038.Bolin,S.R.,Ridpath,J.F.,Black,J.,Macy,M.,Roblin,R.,1994.Survey of cell lines in the American type culture collection for bovine viral diarrhea virus.J.Virol.Methods48,211–221.Canal,C.W.,Hotzel,I.,de Almeida,L.L.,Roehe,P.M.,Masuda,A.,1996.Differentiation of classical swine fever virus from ruminant pesti-viruses by reverse transcription and polymerase chain reaction(RT-PCR).Vet.Microbiol.48,373–379.Cheng,D.,Zhao,J.J.,Li,N.,Sun,Y.,Zhou,Y.J.,Zhu,Y.,Tian,Z.J.,Tu,C.,Tong,G.Z.,Qiu,H.J.,2008.Simultaneous detection of classical swine fevervirus and North American genotype porcine reproductive and respira-tory syndrome virus using a duplex real-time RT-PCR.J.Virol.Meth-ods151,194–199.Deregt,D.,Gilbert,S.A.,Dudas,S.,Pasick,J.,Baxi,S.,Burton,K.M.,Baxi, M.K.,2006.A multiplex DNA suspension microarray for simultaneous detection and differentiation of classical swine fever virus and other pestiviruses.J.Virol.Methods136,17–23.Dı´az de Arce,H.,Nunez,J.I.,Ganges,L.,Barreras,M.,Frias,M.T.,Sobrino,F., 1998.An RT-PCR assay for the specific detection of classical swine fever virus in clinical samples.Vet.Res.29,431–440.Dı´az de Arce,H.,Ganges,L.,Barrera,M.,Naranjo,D.,Sobrino,F.,Frı´as,M.T., Nu´n˜ez,J.I.,2005.Origin and evolution of viruses causing classical swine fever in Cuba.Virus Res.112,123–131.Edwards,S.,Moennig,V.,Wensvoort,G.,1991.The development of an international reference panel of monoclonal antibodies for the differ-entiation of hog cholera virus from other pestiviruses.Vet.Microbiol.29,101–108.Edwards,S.,1993.Bovine viral diarrhea virus.In:Doyle,A.,Griffiths,J.B., Newell,D.G.(Eds.),Cell&Tissue Culture:Laboratory Procedures.John Wiley&Sons,Chichester,UK,Module7B:5,pp.1–8.Elnifro,E.M.,Ashshi,A.M.,Cooper,R.J.,Klapper,P.E.,2000.Multiplex PCR: optimization and application in diagnostic virology.Clin.Microbiol.Rev.13,559–570.Ferna´ndez,J.,Agu¨ero,M.,Romero,L.,Sanchez,C.,Bela´k,S.,Arias,M., Sanchez-Vizcaino,J.M.,2008.Rapid and differential diagnosis of foot-and-mouth disease,swine vesicular disease,and vesicular stomatitis by a new multiplex RT-PCR assay.J.Virol.Methods 147,301–311.Frey,H.R.,Roder,B.,Depner,K.,Liess,B.,1995.Epidemiological charac-terization of a pestivirus isolate from a viremic pig out of a mixed pig-cattle herd.Dtsch.Tierarztl.Wochenschr.102,181–183. Giammarioli,M.,Pellegrini,C.,Casciari,C.,De Mia,G.M.,2008.Develop-ment of a novel hot-start multiplex PCR for simultaneous detection of classical swine fever virus,African swine fever virus,Porcine Circo-virus type2,porcine reproductive and respiratory syndrome virus and porcine mun.32,255–262.Gilbert,S.A.,Burton,K.M.,Prins,S.E.,Deregt,D.,1999.Typing of bovine viral diarrhea viruses directly from blood of persistently infected cattle by multiplex PCR.J.Clin.Microbiol.37,2020–2023.Greiser-Wilke,I.,Blome,S.,Moennig,V.,2007.Diagnostic methods for detection of classical swine fever virus—status quo and new devel-opments.Vaccine25,5524–5530.Hamel, A.L.,Wasylyshen,M.D.,Nayar,G.P.,1995.Rapid detection of bovine viral diarrhea virus by using RNA extracted directly from assorted specimens and a one-tube reverse transcription PCR assay.J.Clin.Microbiol.33,287–291.Hoffmann, B.,Beer,M.,Schelp, C.,Schirrmeier,H.,Depner,K.,2005.Validation of a real-time RT-PCR assay for sensitive and specific detection of classical swine fever.J.Virol.Methods130,36–44. Jamnikar Ciglenecki,U.,Grom,J.,Toplak,I.,Jemersic,L.,Barlic-Maganja,D., 2008.Real-time RT-PCR Assay for rapid and specific detection of classical swine fever virus:comparison of SYBR green and TaqMan MGB detection methods using novel MGB probes.J.Virol.Methods 147,257–264.Katz,J.B.,Ridpath,J.F.,Bolin,S.R.,1993.Presumptive diagnostic differ-entiation of hog cholera virus from bovine viral diarrhea and border disease viruses by using a cDNA nested-amplification approach.J.Clin.Microbiol.31,565–568.H.D.Arce et al./Veterinary Microbiology139(2009)245–252251。