A

oha(Esido©

K

o(agancetEbndaot

0d

Journal of Virological Methods 139 (2007) 111–115

Short communication

Isothermal amplification coupled with rapid flow-throughhybridisation for sensitive diagnosis of Plum pox virus

Antonio Olmos ∗, Edson Bertolini, Mariano CambraLaboratorio de Virologıa e Inmunologıa, Departamento de Proteccion Vegetal y Biotecnologıa, Instituto Valenciano de Investigaciones Agrarias (IVIA),

Carretera Moncada-Naquera Km 5, 46113 Moncada, Valencia, Spain

Received 26 May 2006; received in revised form 13 September 2006; accepted 19 September 2006Available online 7 November 2006

bstract

A nucleic acid sequence-based amplification method coupled with rapid flow-through hybridisation (NASBA-FH) was developed for diagnosisf Plum pox virus (PPV). The sensitivity level achieved by NASBA-FH was 10 times higher than that obtained by Co-PCR and 1000 timesigher than the sensitivity afforded by RT-PCR. In addition, samples from 262 stone-fruit trees collected during winter and spring seasons werenalysed. These samples were tested using methods recommended by the European and Mediterranean Plant Protection Organization to detect PPVDASI-ELISA, RT-PCR and Co-PCR) and by NASBA-FH. Winter PPV diagnostic results by ELISA and NASBA-FH coincided in 90.8%, whileLISA and PCR-based methods coincided in 91.6% and PCR-based methods with NASBA-FH agreed in 95.4%. In spring, diagnostic results were

imilar with all the molecular techniques, which agreed with ELISA results for 98.8% of the trees. NASBA-FH was able to detect more positivenfections in winter, which were later confirmed in spring. These results indicate that NASBA-FH is a suitable molecular method for routine PPVetection in the winter and spring. This user-friendly isothermal RNA amplification coupled with a very fast flow-through hybridisation (15 min)pens up new possibilities for rapid and reliable diagnosis of a variety of pathogens.

2006 Elsevier B.V. All rights reserved.

tabeshoApsolee

eywords: PPV; DASI-ELISA; RT-PCR; Co-PCR; NASBA

Plum pox virus (PPV) produces sharka disease, which isne of the most serious viral diseases affecting stone-fruit treesLopez-Moya et al., 2000), causing important economic andgronomical losses worldwide (Cambra et al., 2006). All strate-ies used to control PPV infection are based on rapid andccurate detection of the virus. Indeed, new methods to diag-ose sharka disease in the early stages are being developedontinuously in an attempt to reduce the impact which this dis-ase has in orchards and stone-fruit nurseries. To date, screeningechniques for PPV detection are based on biological indexing,LISA tests and PCR-based methods (EPPO, 2004), the lattereing the most sensitive. However, there are other alternatives toucleic acid amplification that are potentially applicable to theiagnosis of PPV. Among these, nucleic acid sequence-based

mplification (NASBA) method (Compton, 1991) is a techniquef choice. NASBA is an isothermal RNA amplification processhat involves three enzymes: avian myeloblastosis virus reverse

∗ Corresponding author. Tel.: +34 96 3424000; fax: +34 96 3424001.E-mail address: aolmos@ivia.es (A. Olmos).

tsrhttm

166-0934/$ – see front matter © 2006 Elsevier B.V. All rights reserved.oi:10.1016/j.jviromet.2006.09.012

ranscriptase (AMV-RT), RNase-H and T7 RNA polymerase,s well as two target sequence-specific primers (one of whichears a bacteriophage T7 promoter sequence appended to its 5′nd). The reaction begins when the first primer, bearing the T7equence tail, is elongated by the AMV-RT to yield a RNA–DNAybrid, which is hydrolysed by RNase-H. Subsequently, the sec-nd primer anneals to its target sequence and is elongated byMV-RT to yield a double-stranded DNA molecule. T7 RNAolymerase recognises its promoter site and begins RNA tran-cription. Hybridisation methods with specific probes in dot-blotr ELISA formats (Jean et al., 2001, 2002), or the use of molecu-ar beacon probes in real-time assays (Leone et al., 1998; Klerkst al., 2001) are required to detect the amplified RNA. How-ver, whilst on the one hand traditional passive hybridisation isime-consuming, the other molecular beacons require expen-ive probes and equipment. In this study, a very simple andapid hybridisation method is proposed, based on flow-through

ybridisation technology (US Patent 6,638,760), which directshe flow of the reacting molecules towards the targets and probeshat are immobilised on nylon membranes. The flow-through

ethod increases the molecular binding rate, speeding up the

1 logica

rt

3lftcfsftpm

nrpd

RiRwarbAsGuTe

(iBopLewia

n

taS1mhwsS1bNdsbp(TtTw

dthhbtact

aa

mtedsbR

12 A. Olmos et al. / Journal of Viro

eaction between the complementary molecules and reducinghe hybridisation time (Ou et al., 2005).

Two PPV isolates were used as positive controls, namely.30 RB/GF (PPV-D type) and Ms89/GF (PPV-M type) inocu-ated to GF305 peach seedlings and maintained under quarantineacilities. Healthy peach seedlings were used as negative con-rols. Control transcripts were prepared from the E. coli JM-109lone obtained previously by Olmos et al. (2005). Dilutionsrom 2 × 106 to 2 × 100 copies were used for sensitivity analy-is. In addition, to assess the suitability of the method, samplesrom 262 adult fruit trees of different Prunus species and cul-ivars (184 from Japanese plum, 69 from apricot and 9 fromeach) were collected in the winter (February) during the dor-ant period and also in the spring (April).Plant extracts were prepared in individual plastic heavy-

et bags (PlantPrint Diagnostics) by grinding plant mate-ial 1/20 (w/v) in PBS buffer, supplemented with 2% (w/v)olyvinylpyrrolidone (PVP-10) and 0.2% (w/v) sodium diethylithiocarbamate (Cambra et al., 1994).

Viral RNA was isolated from plant tissue samples with theNeasy Plant Mini Kit (Qiagen) according to the manufacturer’s

nstructions, using 200 �l of plant sap as sample and 350 �l ofLT solution as lysis buffer. Primers for the NASBA reactionere based on P1/P2 PPV universal PCR primers (Wetzel et

l., 1991). P1 primer sequence was extended in its 5′ terminalegion to bear the bacteriophage T7 RNA polymerase promoterinding region (T7P1 5′ AAT TCT AAT ACG ACT CAC TATGG GAC CGA GAC CAC TAC ACT CCC 3′). Nucleotide

equence of P2 primer was not modified (5′ CAG ACT ACACC TCG CCA GA 3′). Colorimetric detection was performedsing a PPV general probe labelled with digoxigenin (5′ TCGTT ATT TGG CTT GGA TGG AA-Digoxigenin 3′) (Olmost al., 2002).

NASBA reactions were undertaken according to Kievits et al.1991). The Nuclisens basic kit (BioMerieux) was used accord-ng to the manufacturer’s instructions with some modifications.riefly, 80 �l of diluent, 14 �l of molecular grade water, 16 �lf stock KCl (final concentration 80 mM), 4 �l of each requiredrimers (25 �M stock of each), were added to the reagent sphere.yophilised enzyme spheres were re-constituted with 55 �l ofnzyme diluent. Subsequently, reagent and enzymes solutionsere directly mixed. This amplification solution was aliquoted

nto 7.5 �l per reaction and 2.5 �l of purified RNA template wasdded. The reaction was carried out at 41 ◦C for 120 min.

One microliter of NASBA product was dispensed onto aylon membrane, positively charged (Roche), dried at room

F2(

Fig. 1. Detection of PPV transcripts by NASBA-FH, Co-PCR and RT-PCR. Ten

l Methods 139 (2007) 111–115

emperature and cross-linked by UV for 4 min. Prehybridis-tion was performed at 50 ◦C using a buffer containing 5×SC, 0.1% (w/v) N-lauroyl-sarcosine, 0.02% (w/v) SDS and% blocking reagent (Roche). Hybridisation was performedixing 10 pmol/ml of 3-DIG labelled specific probe with pre-

ybridisation buffer at 50 ◦C. Membranes were washed twiceith 2× SSC supplemented with 0.1% SDS at 25 ◦C. This

tep was repeated using 0.5× SSC supplemented with 0.1%DS. Membranes were equilibrated with 100 mM maleic acid,50 mM NaCl, 0.3% (v/v) Tween 20, pH 7.5 (Roche) andlocked with blocking buffer [100 mM maleic acid, 150 mMaCl pH 7.5 and 1% (w/v) blocking reagent (Roche)]. Anti-igoxigenin-alkaline phosphatase antibodies (Roche) were sub-equently added at a concentration of 150 mU/ml diluted inlocking buffer, and membranes were washed twice as in therevious steps with 100 mM maleic acid, 150 mM NaCl, 0.3%v/v) Tween 20, pH 7.5 and equilibrated for 2 min with 100 mMris/HCl, 100 mM NaCl, pH 9.5. The substrate used for detec-

ion contained 315 mg/ml NBT and 175 mg/ml BCIP in 100 mMris/HCl, 100 mM NaCl, pH 9.5 (Sigma). Thorough washingith water stopped the reaction.Previous hybridisation was carried out using a Hybrimax

evice (Hybribio Limited) that is based on the principle of flow-hrough hybridisation. A negative pressure under the airproofybridisation membrane was applied by vacuum pump. The pre-ybridisation solution, hybridisation solution, washing solution,locking solution, anti-digoxigenin-alkaline phosphatase solu-ion and colouring solution all flowed through the membraneutomatically. All steps required small volumes, just enough toover the membrane. Vacuum pressure reduced each hybridisa-ion step to 30–60 s, providing results in just 15 min.

Thus, target sequences were detected effectively, both quicklynd specifically, reducing the hybridisation step from a laboriousnd time-consuming task.

The theoretical detection limit of NASBA-FH was deter-ined by three repetitions of the assay using ten-fold serial dilu-

ions of control transcripts, and compared with RT-PCR (Wetzelt al., 1992) and Co-PCR (Olmos et al., 2002). The sensitivityetection limit was established at two copies of control tran-cripts, and was found to be 10 times higher than that obtainedy Co-PCR and 1000 times higher the sensitivity afforded byT-PCR (Fig. 1).

Plant samples collected in winter were tested by the NASBA-H and by the recommended screening tests for PPV (EPPO,004). The same plant extracts were employed for DASI-ELISACambra et al., 1994) based on the specific monoclonal antibody

-fold serial dilutions from 2 × 106 to 2 target copies. NC: negative control.

A. Olmos et al. / Journal of Virological Methods 139 (2007) 111–115 113

Table 1Stone-fruit tree species, number of trees and test results by DASI-ELISA, RT-PCR, Co-PCR and NASBA-FH Plum pox virus detection assays in winter and springseasons

Stone fruit treesanalysed

Winter Spring

No. of treesclustered byresults

ELISA RT-PCR Co-PCR NASBA-FH No. of Treesclustered byresults

ELISA RT-PCR Co-PCR NASBA-FH

Plums (184) 154 + + + + 154 + + + +

18 − − − − 11 + + + +7 − − − −

7 − + + + 7 + + + +

3 − − − + 2 + + + +1 − − − −

1 − + + − 1 + + + +1 + − − − 1 + + + +

Total positive plums 155 162 162 164 176 176 176 176

Apricots (69) 21 + + + + 21 + + + +

31 − − − − 2 + + + +29 − − − −

8 − + + + 3 − + + +5 − − − −

4 − − − + 1 + + + +3 − − − −

3 − + + − 1 + + + +2 − − − −

1 + − − + 1 + + + +1 + − − − 1 − − − −

Total positive apricots 23 32 32 34 26 29 29 29

Peaches (9) 9 – – – – 9 – – – –

T 0

T 198

5CsdPdt3ocbscfacsbo1C

cao(Pinpo

2ttawwb

otal positive peaches 0 0 0

otal 262 178 194 194

B-IVIA (Real-Durviz kit), RT-PCR (Wetzel et al., 1992),o-PCR (Olmos et al., 2002) and NASBA-FH. Results from

amples collected in winter are shown in Table 1. DASI-ELISAetected PPV in 178 trees (155 plum and 23 apricot trees). RT-CR and Co-PCR coincided in PPV diagnosis giving positiveetection for 194 stone-fruit trees (162 plum and 32 apricotrees). NASBA-FH detected PPV in 198 trees (164 plum and4 apricot trees). PPV was not detected in peach trees by anyf the techniques assayed. The results obtained were analysedonsidering the premise that none of methods employed coulde a “gold standard” technique with 100% specificity and 100%ensitivity. Cohen’s kappa coefficient (Staquet et al., 1981) washosen to compare methods, because compared with an imper-ect “gold standard” new tests will have bias in the error rates asresult of the imperfection of the gold standard. This is espe-

ially true for tests with greater detection limits than the “goldtandard” (Valenstein, 1990). Kappa is a measure of agreement

eyond chance that can be used to compare diagnostic tests with-ut designating one test as a gold standard (Bloch and Kraemer,989). Diagnostic Utility Statistics Software (Watkins, 2001;omputer Software, State College) was used to calculate kappa

aPEN

0 0 0 0

262 202 205 205 205

oefficient. DASI-ELISA and NASBA-FH coincided in thenalysis of 238 plants (176 positive and 62 negative trees) outf 262 stone-fruit trees, which represents 90.8% of coincidencek = 0.77 ± 0.06). RT-PCR and Co-PCR coincided totally inPV diagnosis. PCR-based methods and DASI-ELISA were

n agreement in 91.6% (k = 0.79 ± 0.06) (175 positives and 65egatives) whilst PCR-based methods and NASBA-FH dis-layed 95.4% agreement giving the same results in the analysisf 250 trees (190 positives and 60 negatives) (k = 0.88 ± 0.06).

All four techniques were in agreement in the diagnosis of33 out of 262 trees (90%). The rest of the trees were clus-ered on the basis of their results, grouping PCR-based methodsogether as one technique because they share a molecular basis ofmplification and primers. Thus, the rest of the trees, 29 in total,ere clustered in two groups. Cluster A that grouped 16 treeshich tested positive by two techniques (15 positive by PCR-ased methods and NASBA-FH and 1 positive by DASI-ELISA

nd NASBA-FH), and cluster B that grouped 13 trees in whichPV was detected by only one method (2 positive by DASI-LISA, 4 positive by PCR-based methods and 7 positive byASBA-FH).

1 logica

aaa(FsdD2f

fwdpmpNstd

ptoamtneptg(dtwStboticeaNc

b(oEpFd

tsimfiw

sttbefTPfPsMtlbmp

A

CeAgcr

R

B

C

C

C

E

J

J

14 A. Olmos et al. / Journal of Viro

Plant samples were also collected from the same 262 treesfter flowering in spring when full-expanded leaves appearednd then retested (8 leaves/tree) using all the techniques. Resultsre indicated in Table 1. DASI-ELISA detected PPV in 202 trees176 plum and 26 apricot trees). RT-PCR, Co-PCR and NASBA-H agreed with PPV diagnosis giving positive detection for 205tone-fruit trees (176 plum and 29 apricot trees). PPV was notetected in peach trees by the methods assayed. The results ofASI-ELISA and molecular methods agreed in the analysis of59 plants (202 positive and 57 negative trees) out of 262 stone-ruit trees, representing 98.8% agreement (k = 0.96 ± 0.06).

In order to validate whether NASBA-FH is a suitable methodor PPV diagnosis in plant material, agreement of techniquesas based on interpreting values of the kappa statistics asescribed previously by Landis and Koch (1977): k < 0.00 isoor, 0 < k < 0.20 is slight, 0.21 < k < 0.40 is fair, 0.41 < k < 0.60 isoderate, 0.61 < k < 0.80 is good and 0.81 < k < 1.00 is in almost

erfect agreement. The high concordance obtained amongASBA-FH and the recommended methods for PPV diagno-

is (ranging from good to almost perfect agreement) validateshe use of NABA-FH as a suitable routine technique for PPVetection in plant material in both winter and spring periods.

This study appraises critically PPV diagnosis in field sam-les in two aspects. First, it confirms that the best season toest plant material is spring, not only due to the higher numberf positive results but also because of the close concordancemong techniques. In fact, in spring all methods were in agree-ent completely with plum and peach trees. In the case of apricot

rees, molecular techniques detected PPV in three trees that wereegative by DASI-ELISA. This could be due to the higher inher-nt sensitivity of molecular methods, which can detect virionsresent in minute quantities. The second assessment focuses onhe discordance between results in winter and spring for the treesrouped in cluster A and B in winter. In the case of cluster Atrees giving positive results for PPV using two techniques) theiscordance was observed in 5 out of 16 trees and only in apricotrees. These five trees tested positive by molecular methods ininter; however, these results could not be confirmed in spring.uch results could be associated to the uneven distribution of

he virus in the apricot trees and consequently to sampling draw-acks. In the case of cluster B (trees giving positive results forne technique) discordant results were observed in 7 out of 13rees and could be due to false positive results or because PPVnfection was present in very low titer. These winter subclini-al infections very probably occur much more frequently thanxpected and constitute a subject of current research. Reliablend sensitive methods are extremely useful for this purpose andASBA-FH could improve detection potential, leading to betterontrol of this disease.

NASBA-FH was the method able to detect the largest num-er of PPV infections in winter that were confirmed in springTable 1). Specifically, 189 out of 198 cases by NASBA-FH, 187ut of 194 by PCR-based methods and 177 out of 178 by DASI-

LISA, tested positive in winter were confirmed. In the case oflums 163 out of 164 PPV infected trees diagnosed by NASBA-H in winter, were confirmed in spring. PCR-based methodsetected 162 infected plums and DASI-ELISA tested 155 posi-

J

l Methods 139 (2007) 111–115

ive trees in winter, respectively, all of which were confirmed inpring. In apricot trees, NASBA-FH diagnosed 34 infected treesn winter of which 26 were confirmed in spring; PCR-based

ethods detected 32 infections, 25 of which were later con-rmed, while ELISA-DASI tested 23 positive trees in winter ofhich 22 were confirmed in spring.This thermalcycler-free method is particularly suitable for

ingle-stranded RNA viruses such as PPV. It has been reportedhat NASBA generates the same number of copies as a conven-ional RT-PCR but in a shorter period of time (Jean et al., 2004)ecause PCR requires a previous retrotranscription step. How-ver, a more interesting aspect of this method, with implicationsor routine testing is that contaminating DNA is not amplified.he excellent detection limit obtained using NASBA-FH forPV detection was around 1000 times higher than that reportedor IC-PCR (Wetzel et al., 1992) and 10 times higher than Co-CR (Olmos et al., 2002). Sensitivity level by NASBA-FH wasimilar to that described for PPV by real time RT-PCR using Taq-

an probes (Olmos et al., 2005). This sensitivity level makeshis molecular technique suitable for routine control and surveil-ance programmes, mainly useful for clarifying the differencesetween results obtained with DASI-ELISA and PCR-basedethods. This method could easily be adapted to other RNA

athogens.

cknowledgements

This work has been supported by Spanish grants from theonsellerıa de Empresa, Universidad y Ciencia of the Gen-ralidad Valenciana, project GVA05/198 and CICYT projectGL2005-01546. Dr. E. Bertolini was recipient of a fellowshiprant (CTBPDC/2004/034) from Consellerıa de Cultura, Edu-acion y Deporte of the Generalidad Valenciana. English textevised by F. Barraclough and Carl Spetz.

eferences

loch, D.A., Kraemer, H.C., 1989. 2 × 2 Kappa coefficients: measures of agree-ment or association. Biometrics 45, 269–287.

ambra, M., Asensio, M., Gorris, M.T., Perez, E., Camarasa, E., Garcıa, J.A.,Lopez-Moya, J.A., Lopez Abella, J.J., Vela, D., Sanz, A., 1994. Detectionof Plum pox potyvirus using monoclonal antibodies to structural and nonstructural proteins. Bull. OEPP/EPPO Bull. 24, 569–577.

ambra, M., Capote, N., Myrta, A., Llacer, G., 2006. Plum pox virus andestimated cost associated to sharka disease. Bull. OEPP/EPPO Bull. 36,202–204.

ompton, J., 1991. Nucleic acid sequence-based amplification. Nature 350,91–92.

PPO., 2004. Diagnostic protocol for regulated pests. Plum pox potyvirus. Bull.OEPP/EPPO Bull. 34, 247–256.

ean, J., Blais, B., Darveau, A., Fliss, I., 2001. Detection of hepatitis A virusby the nucleic acid sequence-based amplification technique and compari-son with reverse transcription-PCR. Appl. Environ. Microbiol. 67, 5593–5600.

ean, J., Blais, B., Darveau, A., Fliss, I., 2002. Rapid detection of human rotavirususing colorimetric nucleic acid sequence-based amplification (NASBA)-

enzyme-linked immunosorbent assay in sewage treatment effluent. FEMSMicrobiol. Lett. 210, 143–147.

ean, J., D’Souza, D.H., Jaykus, L.A., 2004. Multiplex nucleic acid sequence-based amplification for simultaneous detection of several enteric viruses inmodel ready-to-eat foods. Appl. Environ. Microbiol. 70, 6603–6610.

logica

K

K

L

L

L

O

O

O

S

V

Wchain reaction assay adapted to plum pox potyvirus detection. J. Virol. Meth-

A. Olmos et al. / Journal of Viro

ievits, T., van Gemen, B., van Strijp, D., Schukkink, R., Dircks, M., Adriaanse,H., Malek, L., Sooknanan, R., Lens, P., 1991. NASBA isothermal enzymaticin vitro nucleic acid amplification optimized for the diagnosis of HIV-1infection. J. Virol. Methods 35, 273–286.

lerks, M.M., Leone, G.O., Verbeek, M., van den Heuvel, J.F., Schoen, C.D.,2001. Development of a multiplex AmpliDet RNA for the simultaneousdetection of Potato leafroll virus and Potato virus Y in potato tubers. J.Virol. Methods 93, 115–125.

andis, J.R., Koch, G.G., 1977. The measurement of observer agreement forcategorical data. Biometrics 33, 159–174.

eone, G., van Schijndel, H., van Gemen, B., Kramer, F.R., Schoen, C.D.,1998. Molecular beacon probes combined with amplification by NASBAenable homogeneous, real-time detection of RNA. Nucleic Acids Res. 26,2150–2155.

opez-Moya, J.J., Fernandez-Fernandez, M.R., Cambra, M., Garcıa, J.A., 2000.Biotechnological aspects of Plum pox virus. J. Biotechnol. 76, 121–136.

lmos, A., Bertolini, E., Cambra, M., 2002. Simultaneous and co-operationalamplification (Co-PCR): a new concept for detection of plant viruses. J.Virol. Methods 106, 51–59.

W

l Methods 139 (2007) 111–115 115

lmos, A., Bertolini, E., Gil, M., Cambra, M., 2005. Real-time assay for quanti-tative detection of non-persistently transmitted Plum pox virus RNA targetsin single aphids. J. Virol. Methods 128, 151–155.

u, Z.Y., Liu, N., Chen, C.J., Cheng, G., He, Y.S., 2005. Rapid and accu-rate genotyping of YMDD motif variants in the hepatitis B virus genomeby an improved reverse dot blot method. J. Clin. Microbiol. 43, 5685–5689.

taquet, M., Rozencweig, M., Lee, Y.J., Muggia, F.M., 1981. Methodology forthe assessment of new dichotomous diagnostic tests. J. Chronic. Dis. 34,599–610.

alenstein, P.N., 1990. Evaluating diagnostic tests with imperfect standards.Am. J. Clin. Pathol. 93, 252–258.

etzel, T., Candresse, T., Ravelonandro, M., Dunez, J., 1991. A polymerase

ods 33, 355–365.etzel, T., Candresse, T., Macquaire, G., Ravelonandro, M., Dunez, J., 1992. A

highly sensitive immunocapture polymerase chain reaction method for plumpox potyvirus detection. J. Virol. Methods 39, 27–37.