HAL Id: dumas-02997823 https://dumas.ccsd.cnrs.fr/dumas-02997823 Submitted on 10 Nov 2020 HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés. Characterization of different pepper resistances to Potato virus Y through quantitative approaches to measure viral load Pierre Mustin To cite this version: Pierre Mustin. Characterization of different pepper resistances to Potato virus Y through quantitative approaches to measure viral load. Life Sciences [q-bio]. 2020. dumas-02997823
92
Embed
Characterization of different pepper resistances to Potato ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
HAL Id: dumas-02997823https://dumas.ccsd.cnrs.fr/dumas-02997823
Submitted on 10 Nov 2020
HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.
Characterization of different pepper resistances toPotato virus Y through quantitative approaches to
measure viral loadPierre Mustin
To cite this version:Pierre Mustin. Characterization of different pepper resistances to Potato virus Y through quantitativeapproaches to measure viral load. Life Sciences [q-bio]. 2020. �dumas-02997823�
díIngÈnieur de líInstitut SupÈrieur des Sciences agronomiques, agroalimentaires, horticoles et du paysage
de Master de líInstitut SupÈrieur des Sciences agronomiques, agroalimentaires, horticoles et du paysaged'un autre Ètablissement (Ètudiant arrivÈ en M2)
IntroductionL'améliorationdespratiquesdeprotectiondesculturescontrelesorganismesnuisiblesestl'undesprincipauxenjeuxdelarechercheagronomiqueactuelle.Parmiceux-ci,lesagentspathogènesdesplantessontcaractérisésparunediversitéd'organismesbiologiquescommeles nématodes, les champignons, les bactéries et les virus (Oerke, 2005). Cette diversitéd'agentspathogènesestliéeàdiversesfaçonsderéduirelaproductivitédescultures,depuisla consommation de tissus, l'assimilation de la sève jusqu'aux troubles dumétabolismecompromettantaumoins10%delaproductionalimentairemondiale(Booteetal,1983;StrangeetScott,2005).Depuis les années 1950, le développement industriel de l'agriculture a conduit à uneutilisationmassivedeproduitschimiquesdeprotectiondesculturesafindeminimiserlespertescauséesparlesagentspathogènesetlesparasites(Savaryetal,2019).Cependant,lesmaladiescauséesparlesvirusdiffèrentconsidérablementdespathologiesbactériennesou fongiques. Les infections virales sont persistantes, généralisées et généralementincurables (Devergne et Albouy, 1991). Par conséquent, les pratiques de protection desculturessontlimitéespuisqu'iln'existepasdetraitementchimiquepourguérirlesplantesinfectéespardesvirus(Mouryetal,2010).Certainesméthodesdeprophylaxietellesquelathermothérapieetlaprotectioncroiséedonnentdesrésultatsprometteurspourréduirelesinfectionsvirales,maissontassociéesàdesprocédurescoûteusesetexigeanteenmain-d'œuvre(ThompsonetTepfer,2010;Varverietal,2015;Wangetal,2018).Parailleurs,l'efficacitédelaluttecontrelesvecteursdépendfortementdel'espècedevirusetdesonmodedetransmission(Perringetal,1999).Comptetenudeceslimites,despratiquesdeprotection des cultures efficaces nécessitent la gestion des maladies virales avantl'apparitiondesfoyersdemaladie.Parconséquent,l'utilisationdesrésistancesdesplantespourraitêtreuncatalyseurmajeurpourproposerdessolutionsefficacesetàfaibleapportd'intrants(Mouryetal,2010;Boualemetal,2011).Lasélectionderésistancesvégétalesestdéjàuneméthodecourantepourlimiter les infectionsvirales.Cependant, l'évaluationdel'expression des résistances est une condition préalable à une utilisation durable desrésistancesdesplantesquipeutfacilementêtrecompromisepar lacontre-adaptationduvirus ciblé conduisant à la dégradation des résistances, un gaspillage de ressourcesgénétiquesassociéàdelourdesperteséconomiques.Le sujet de mon stage porte sur la gestion des maladies virales par l'utilisation desrésistancesdesplanteset l'évaluationdeleurexpressionphénotypiquesuruneétapeducycle viral. Afin de fournir un contexte permettant de mieux comprendre les résultatsprésentés dans ce rapport, l'introduction est divisée en quatre parties principales. Toutd'abord, lecycleviralseraprésenté,ainsiquelesprincipauxmécanismesdelarésistancedes plantes aux virus. Ensuite, lesméthodes actuelles d'évaluation de la résistance desplantes seront présentées, avant d'exposer les objectifs du stage et les connaissancespréalablessurl'étudedecasdecetravail:l'interactionentrelepiment(Capsicumannuum)etlePVY(PotatovirusY).
Fondamentalement,uncycleviraldanslesplantesestconstituédecinqétapesprincipales(figure1) : (i) inoculationduvirus, (ii)multiplicationduvirusdans les celluleshôtes, (iii)migration du virus de cellule à cellule, (iv) colonisation systémique du virus et (v)disséminationduvirusversd'autresplantes.Lesespècesviralespeuventavoirplusieursvoiesdetransmission(Matthews,1981;Astieret al, 2001). Par exemple, les tobamovirus comme le Tobaccomosaic virus (TMV) et leTomato brown rugose fruit virus (ToBRFV) sont transmis par les semences, les eaux decirculationet lescontactsmécaniquesentre lesplantes (Salemetal,2016).Cescontactsmécaniquesprovoquentdesblessures"fraîches"quipermettentàcesvirusdesurmonterles premières barrières de l'hôte à l'infection (cuticule et paroi cellulaire) (Devergne etAlbouy,1998).Cependant, latransmissiondesvirussefait leplussouventpardes insectesvecteursquipeuvent être telluriques ou aériens (Astier et al, 2001). Les vecteurs telluriquescomprennentdesgroupesrestreintsdenématodesetdechampignons(Walkey,1991).Lesvecteurs aériens comprennent des insectes comme les aleurodes, les cochenilles, lescicadelleset lespucerons(Jones,1987;Astieretal,2001).Lespuceronssontde loin lesvecteurslesplusimportants,carilstransmettentenviron30%detouslesvirusvégétauxconnusàce jour(Braultetal,2010). Enoutre, lesstratégiesdetransmissiondépendentd'interactionstrèsspécifiquesentrelesprotéinesviralesetlesrécepteursdesinsectes,cequimetenévidenceunecoévolutionintenseentrel'agentpathogèneetsonvecteur(Braultetal,2010).Ilexistedeuxprincipauxmodesdetransmissiondesvirusdesplantesparlesinsectesvecteurs:noncirculatoireetcirculatoire.-Latransmissionnoncirculatoireestlemodedetransmissionlepluscourantdanslesvirusdesplantes.Lesparticulesviralesrestentliéesauxstyletsdepuceronssurdesrécepteursspécifiques (Khelifa,2019).Cetteétapeestappelée rétentionet saduréepeutvarierdequelquesminutes(nonpersistante)àquelquesheuresoujours(semi-persistante)(Braultetal,2010).Lesparticulesviralesetlesstylesd'insectespeuventseliersoitdirectementgrâceàlaprotéinedecapside(Cucumbermosaicvirus),soitindirectementparl'intermédiaired'uncomposéprotéiquesupplémentaire.-Latransmissioncirculatoirenécessitelarétentiondelaparticuleviraleàl'intérieurdesonvecteuret le franchissementdedifférentesbarrièrespouratteindre l'hémolympheet lesautresorganes(glandessalivaires,organesreproducteurs...)(NgetPerry,2004;Braultetal,2010).Danslaplupartdescas,lapériodederétentionpeutdurerpendanttouteladuréedevieduvecteur (Braultetal,2010).Certainsviruspeuventmêmesemultiplierdans leurscellulesvectricescommelesRhabdoviridae(Sylvester,1980).Unefoisquelevirusapénétrédanssonhôte,ilprofitedelacellulehôteenutilisantsonénergieet samachineriepour la synthèsede sespropresprotéinesetacidesnucléiques
(figure2).PourlesvirusvégétauxàARN(+)(environ75%detouslesvirusvégétaux),lesacidesnucléiquessontséparésdelaprotéined'enveloppe(Astieretal,2001).Ensuite,l'ARNvirals'associeauxribosomesdel'hôtequiassurentlatraductiondel'ARNpolyméraseviraleetdecertainesautresprotéinesspécifiquesauviruscommelesprotéinesde l'enveloppe(Matthews,1981).L'ARN(+)estutilisécommemodèlepourtranscriredescopiesd'ARN(-)quiassurerontensuitelasynthèsedemoléculesd'ARN(+)complémentaires(DevergneetAlbouy,1998). Lesnouvellesparticulesvirales sontcomposéesd'ARNviralnouvellementrépliquéetencapsidéparlaprotéinedecapside(Astieretal,2001).Lorsquelesmultiplescyclesderéplicationetdetraductionseproduisent,ilyauneaccumulationd'ARNviral,departiculesviralesetdeprotéinesviralesdanslacellulehôteprimaireinfectée(DevergneetAlbouy,1998).Leviruspeutalorscoloniserlesplantesparl'intermédiairedesplasmodesmessousformedeparticulesviralesoudecomplexesribonucléoprotéiques(RNP)etinitierdenouveauxcyclesd'infectionvirale (Matthews,1981 ;DawsonetHilf,1992, Lucas,2006).Lorsquelevirusatteintlesystèmevasculaire,ilcolonisel'ensembledel'organisme,cequientraîneunemaladie généralisée reconnaissable à ses symptômes (DevergneetAlbouy,1998).Cessymptômescomprennentdesanomaliesdelapigmentation(mosaïques,aspecttacheté, taches annulaires, jaunisse et panachure des fleurs et/ou du feuillage), unralentissement de la croissance (rabougrissement), des déformations et des nécroses(Walkey, 1991 ; Astier et al, 2001 ; Valverde et al, 2012). De plus, les infections viralesprovoquent des changements physiologiques et biochimiques tels qu'une diminution del'efficacité de la photosynthèse, une augmentation de la fréquence respiratoire, uneaugmentationdecertainesactivitésenzymatiques(polyphénoloxydases),uneaccumulationdepolyphénolsoxydésetenfinuneaugmentationouunediminutiondes régulateursdecroissancedesplantes(Matthews,1981).Touscessymptômesentraînentuneréductiondurendementetdelaqualitédescultures,cequisetraduitpardesperteséconomiquespourlesproducteurs.Enfin,ladernièreétapeducycleviralestlatransmissionàuneautreplante(selonl'unedesvoies décrites précédemment) qui est essentielle pour la survie et la disséminationgéographiquedelapopulationvirale.
L'interférence de l'ARN constitue une défense contre les virus par un mécanisme derégulationde l'expressiongénétiqueoù l'ARNviral subitunedégradation très spécifique(Astieretal,2001).UnefoisquelacellulevégétaleestinfectéeparunvirusàARN,l'ARN
viral se réplique et forme des molécules d'ARN double brin (ARNdb) dont la présenceinhabituelleestdétectéedanslacellulehôte.CesARNdoublebrinsontensuitetransforméspardesenzymesdetypeDicerenpetitsARNinterférents(siRNA)dérivésduvirus(Sharmaetal,2013).CessiRNAsontprisdanslecomplexedesilençageinduitparl'ARN(RISC)quiinitieleclivagedesacidesnucléiquesvirauxciblés(Sharmaetal,2013,Nicaise,2014).Enoutre, l'intégrationdessiiRNAdans leRISCgénèreunsignaldesilencieuxmobilequiesttransféréauxcellulesadjacentesetàlaplanteentièreparlebiaisdesplasmodesmesetduphloèmeetamplifiéparunprocessusd'amplificationrelais(Nicaise,2014).Parconséquent,ce signal active l'extinction de l'ARN dans les cellules non infectées. Ce processus estimpliquédanslesstratégiesdeprotectioncroisée,quiempêchentl'infectionparuneespècede virus à la suite d'une infection antérieure par une souche apparentée (Sharma et al,2013).
1.2.2. RésistancesqualitativesetquantitativesLesrésistancesqualitativessontconnuespourconférerunerésistanceforteoucomplètecontre un parasite spécifique. Elles sont principalement basées sur la reconnaissancespécifiqued'unfacteurd'avirulencecodéparunvirus(commelaprotéinedel'enveloppe)paruneprotéineR(résistance)végétale,appartenantàlaclassedesrépétitionsrichesenleucine(NLR)(Nicaise,2014;Boualemetal,2016).Cetteinteractiondéclencheunecascadede signalisation par la protéine kinase activée par unmitogène (MAPK). L'activation desgènesRdéclencheuneréponsehypersensible(HR),quiestuneinductionrapidedelamortcellulaire programmée pour les cellules infectées et leurs voisins. Il en résulte unconfinementduvirusetempêchetoutenouvellecolonisationviraledanslaplante(Nicaise,2014). Elle peut également entraîner une accumulation rapide d'espèces réactives del'oxygène(ROS)etd'hormonesdedéfense(acidessalicyliqueetjasmonique),cequiconduitàunerésistancesystémiqueacquise(SAR;Nicaise,2014;Boualemetal,2016).D'unpointde vue sémiologique, cela conduit à l'apparition de taches nécrotiques sur les organesinfectésaprèsquelquesheures(Astieretal,2001).Enoutre,danscertainscas,l'infectionviraleesttotalementéliminéesansaucuneréactiondenécrose;c'estcequ'onappelleunerésistanceextrême(Astieretal,2001).Malgrélararetédesgènesderésistancequalitative,cetypederésistanceestfréquemmentutiliséparleséleveurscarcestraitsgénétiquessontfaciles à sélectionner (Fraser, 1992). Cependant, comme ils exercent une forte pressionsélective sur les virus qui sont connus pour présenter des taux demutation élevés parrapport à d'autres organismes biologiques, ces résistances peuvent être facilementsurmontéesparlesvirus(Gagoetal,2009)vialamutationdeleursfacteursd'avirulence.En revanche, les résistancespartiellesouquantitativesne fontqu'adoucir l'infectiondesagentspathogènes(réduisantainsil'impactsurlerendementoulaqualitédesplantes)sansempêcher l'infection. Elles sont plus abondantes dans les collections génétiques descultures.Cesrésistancesnesontpas liéesàunmécanismemoléculairespécifiqueet leurdéterminismegénétiquereposesurdenombreuxQTL(QuantitativeTraitLoci)(Lindhout,
2002). La résistance quantitative est souvent supposée plus durable que la résistancequalitative,cependant,commel'utilisationdelarésistancequantitativedanslesculturesestmoins fréquente que la résistance qualitative, peu de données sont disponibles pourdocumentercettehypothèse(Lecoqetal,2004).
1.2.3. LesrésistancesrécessivesCetypederésistanceestpluscourantpourlesvirusquepourlesautresagentspathogènesdes plantes et représente environ la moitié des 200 gènes de résistance connus (Diaz-Pendon,2004;Nicaise,2014).Ellereposesurdesgènesrécessifscodantpourdesfacteursd'initiation eucaryotes (eIF) principalement recrutés par les virus (dont notamment lespotyvirus) pour leur réplication et leur traduction (Nicaise, 2014). En effet, le cycle viralnécessiteuneséried'interactionscompatiblesentrel'hôteetlesfacteursvirauxàtouteslesétapes de l'infection virale, de la traduction et de la réplication de l'ARN viral à latranslocationà longuedistancedans le systèmevasculaire (Diaz-Pendonet al, 2004). Enconséquence,siunfacteurhôterequisfaitdéfautouprésenteuneversionmutée,levirusnepeutpasacheversoncycle,cequientraîneunerésistancedel'hôte(Astieretal,2001;Diaz-Pendonetal,2004).
Comme l'illustrent les paragraphes précédents, il existe de multiples mécanismes derésistancedesplantesauxvirusquisontcomplémentairesentermesdemomentdedéfense(auxpremiers ouderniers stadesde l'infection), de localisation (dans la première feuilleinfectée ou dans les tissus systémiques) et de molécules virales ciblées (génome ouprotéines virales) (Lecoq et al, 2004 ; Nicaise, 2014). En conséquence, on peut émettrel'hypothèsequedifférentsgènesderésistancepeuventciblerdifférentesétapesducycleinfectieuxviral.Cettehypothèseestétayéeparlesexemplessuivants.
Larésistanceàlatransmissionduvirusestassociéeàunerésistancedominantedelaplanteauvecteurlui-même,cequientraînel'échecdel'inoculationoudel'acquisition(Pochard,1977).Cesrésistancesontétédécritessurdenombreuxvecteurstelsquelesnématodes,les champignons et les insectes (Jones, 1987). Par exemple, le gène Vat (Virus aphidtransmission)dumeloninterfèreaveclatransmissionduvirusparl'espècedepuceronAphisgossypi.LesplantesporteusesdugèneVatnesontpas infectéesaprèsuneinoculationàmédiationparlespuceronsavecdifférentsvirusmaissontsensiblesàcesviruslorsqu'ellessontinoculéesmécaniquementoulorsqued'autresespècesdepuceronssontutiliséespourlatransmission(Martinetal,2005;Boissotetal,2010).
Larésistanceàlamultiplicationduvirusdanslacelluleinoculéepeutentraîneruneabsencetotaleouuneréductiondelaréplicationviraledanslacelluleetpeutreposersurdesgènesrécessifs ou dominants. Par exemple, dans le cas de l'infection du poivron par le PVY,lorsqu'uneplanteestporteusededeuxallèlesderésistancerécessifscommepvr21etpvr22,leurprésencecombinéeentraîneuneinteractionincompatibleentrelevirusetunfacteurdetraductiondelaplante(eIF4E)conduisantàl'absencedemultiplicationduvirus(Mouryetal,2004).
1.3.3. Résistance aumouvementdes cellulesdu virus et à la colonisationsystémique
Danscertainscas,leviruspeutsemultiplierdanslescellulesinoculéesmaisnepeutpassedéplacer vers les cellules voisines ou ne pas atteindre le système vasculaire. Ici desrésistancesdominantesontétéidentifiées.Parexemple,dansl'écotypeCol-0d'Arabidopsisthaliana,leTEV(Tobaccoetchvirus)semultiplieetsedéplacedecelluleencellulemaisnepeut pas atteindre le système vasculaire. Trois gènes dominants, RTM1 (Restricted TEVMovement1),RTM2etRTM3sontnécessairespourcetterésistance(Chisholmetal,2001).Malgrélesexemplescitésci-dessus,lalittératurescientifiquesurl'effetdelarésistancedesplantesauxdifférentesétapesducycleinfectieuxdesvirusestencorerare.Enparticulier,iln'a jamais été démontré que différents gènes de résistance peuvent cibler différentesétapesducyclepourunecombinaisonplante/virusdonnée.Pourtant,lamiseenplacederésistancesadaptéesetefficacesdanslesplantescultivéesestguidéeparlesconnaissancesdisponiblessurlesmécanismesmoléculairesetleseffetsphénotypiquesdesrésistancesdesvirus des plantes (Nicaise, 2014). La mesure de ces effets phénotypiques à des fins desélectionnécessitedesprotocolesappropriés,etnotammentdesoutilsdediagnosticetdequantification pour évaluer la présence du virus et sa limitation par lesmécanismes derésistance.
2.1. Lestestsd'infectivitéDe toutes les méthodes d'évaluation, les tests d'infectivité sont les plus élémentairespuisqu'ils ne reposent que sur l'observation des symptômes après l'inoculation du virus(Matthews,1981).Historiquement, ces testsétaientconsidéréscommedes facteurscléspourl'identificationdesviruscarilspouvaientaideràdéterminerlagammed'hôtesoulasémiologied'uneespècedevirusinconnue(Walkey,1991).Danslapratique,leschercheurscommencentpar inoculer biologiquement (via un vecteur) oumécaniquement le virus àl'unedesesplanteshôtesetsurveillentensuitel'apparitiondessymptômes(Walkey,1991).Cependant,l'expressiondessymptômesdépendfortementdelaplantehôte(génotypeetétat physiologique), de la souche virale, de la date d'inoculation et des conditionsenvironnementales (température, lumière) ; cela indique qu'une extrême prudence estnécessairepoureffectueretnormalisercestests(Walkey,1991;Astieretal,2001).Bienque ces tests soient toujours utilisés en laboratoire, l'introduction de l'essai immuno-enzymatique(ELISA)danslesannées1970-1980aconsidérablementraccourcietsimplifiélaréalisationdesdiagnosticsvirauxderoutine(Boonhametal,2014).
2.2. Lestechniquessérologiques
Lestechniquessérologiquessontfréquemmentutiliséespourleurspécificité,leurrapiditéetleurfacilitédestandardisation(ClarketAdams,1977).Cestechniquessontbaséessurdesprincipesderéactivitéantigéniquereposantsur la liaisonspécifiqueentredeuxtypesdemolécules:l'antigèneviraletsonanticorpsprovenantdemammifères(Walkey,1991;Astieretal,2001).Enrésumé,lesystèmeimmunitairedesmammifèresalacapacitédereconnaîtrespécifiquementlescaractéristiquesdesurfacedesmacromoléculesexogènesoudesmicro-organismes appelés antigènes (Crowther, 1995). Les composants des mammifères quieffectuentcettereconnaissancesontappelésanticorpsquipeuventêtremonoclonauxoupolyclonauxselonqu'ilsreconnaissentrespectivementunouplusieursépitopesdumêmeantigène. En virologie végétale, ces principes d'immunologie sont utilisés dans diversestechniquestellesquel'immunoprécipitationouletestELISA(EnzymelinkedImmunosorbentAssay).L'ELISAreposesurladétectionsensibledesréactionsnonprécipitantesgrâceàdesanticorpsmarquésparuneenzyme(ClarketAdams,1977).Enpratique,cetteméthodeestréalisée dans des plaques de microtitration sur lesquelles les antigènes sont d'abordadsorbés passivement sur la surface interne avant d'être incubés avec des anticorpsmarqués par des enzymes et suivis par le changement de couleur substrat approprié(Crowther,1995;Sakamotoetal,2018).Cetteadsorptionpassivedesanticorpsfacilitelaséparationdesréactifslibresetliés,cequipermetégalementunegrandesouplessedanslaconceptiondestests(Crowther,1995).Engénéral,laphosphatasealcaline(ALP)estutiliséecommeenzymechargéedelatransformation(encasdereconnaissanceantigène-anticorps)d'unsubstratincoloresoluble,lep-nitrophénylphosphate(PNPP),enp-nitrophénolquipeutêtredétectéà405nm(couleurjaune,Astieretal,2001;Sakamotoetal,2018).Laraisondelapopularitédel'ELISArésidedanssonrapportcoût-efficacité,sarobustesse,sacapacitéàtesterungrandnombred'échantillonsetsasimplicité.Cependant,danslesannées1990,de
nouvelles méthodes ont été développées pour détecter les acides nucléiques viraux(Boonhametal,2014).
2.3. LestechniquesmoléculairesLestechniquesmoléculairestellesquel'hybridationmoléculaireetl'amplificationenchaînepar polymérase (PCR) sont basées sur la complémentarité des séquences permettant laliaison des acides nucléiques viraux avec des sondes d'ADN ou d'ARN spécifiques auxséquences(Rubioetal,2020).LaPCRreposesurunesuccessiond'étapesdedénaturation,d'hybridationdesamorcesetd'extensiondelapolymérasethermostablequireproduit laséquenceviralecibledélimitéepar lesamorces(Astieretal,2001). Cependant,pour lesvirusàARN,uneétapedetranscriptioninverseestnécessaireavantl'amplificationparPCR.Elleconsisteàsynthétiserunbrind'ADNcomplémentaire(ADNc)basésurlebrind'ARNduvirusgrâceàunetranscriptaseinverse(Astieretal,2001;Websteretal,2004).Àlafin,unemigrationdesproduitsd'amplificationestgénéralementeffectuéeparélectrophorèsedansungeld'agarose,quirévèlelaprésenceoul'absenced'unvirus(Astieretal,2001).LaPCRetsesvariantessonttrèspopulairesenraisondeleurgrandesensibilitéetdeleurspécificitémodulableenfonctiondesobjectifs(Websteretal,2004).Eneffet,pluslesamorcessontconçuesdansunerégionvariabledelaséquenceduvirus,plusletestestspécifiquepourune espèce ou une souche virale (Astier et al, 2001). Toutefois, comme le test est trèssensible, ilaugmente lerisquederésultats faussementpositifsdusàdescontaminations(Websteretal,2004;Bustinetal,2009;Varverietal,2015).
2.4. PotentieldecestechniquespourlaquantificationdesvirusEnplusde leurpotentieldedétectionqualitativedesvirus, lestechniquesde laboratoiredécritesci-dessusontunpotentieldequantification.Pourlestestsd'infectivité,unerelationlinéaireaétédémontréeentrelenombredelésionslocalesetlaconcentrationdeparticulesviralesinfectieusesdansl'extraitd'inoculation(Matthews,1981;Astieretal,2001).Celasuggèreque lanotationde laprogressiondessymptômespeutêtreunmoyend'évaluerquantitativementl'accumulationvirale.LetestELISApeutêtreutilisépourlaquantificationdes protéines virales en fonction de l'intensité du signal (Rubio et al, 2020). Comme ladensité optique résultante à une longueur d'onde spécifique et après une périoded'incubationdonnéeestproportionnelleàlaquantitéinitialed'antigène,laconcentrationviraledansl'extraitdeplanteestmesurableparspectrophotométrie(ClarketAdams,1977;Astieretal,2001,Rubioetal,2020).Pourlesacidesnucléiquesviraux,laPCRquantitative(ouentempsréel)(qPCR)reposesurlaquantificationd'unsignalfluorescentgénérélorsdel'amplificationdel'ADN,soitviaunagentd'intercalation(parexempleSYBRGreen),soitviaunesonded'hydrolyse(parexempleTaqman).Elleconstitueuneméthodetrèsprécisepourestimer la concentrationd'acidesnucléiquesavecunegrande sensibilité (TseetCapeau,2010).
Comme décrit ci-dessus, l'absence de virucides et la difficulté de contrôler les insectesvecteurs tout en atténuant les problèmes environnementaux et humains limitentconsidérablement lepotentieldesproduits chimiquespourgérer lesépidémiesdevirus.Danscecontexte, l'utilisationde la résistancedesplantesestunestratégieprometteuse(Mouryetal,2010;Boualemetal,2011).L'utilisationdeméthodesdequantificationfiablesdel'infectionviraleestessentiellepourcaractériserleniveauderésistanced'uneplanteàl'accumulation de virus. Traditionnellement, l'état sanitaire d'une plante est révélé parl'observation des symptômes ou par la détection des composants du virus tels que lesprotéines(révéléesparELISA)oulesacidesnucléiques(révéléesparRT-PCR).Cesméthodessont donc utilisées pour distinguer les génotypes de plantes sensibles des génotypes deplantesrésistantes(Lecoqetal,2004).Cependant,l'estimationquantitativedelaprésenced'unviruspermetd'obteniruneinformationplusprécisesurleniveauderésistanced'uneplante.Parexemple,ilaétédémontréquel'accumulationdevirusestunestimateurprécisdelarésistancequantitativepourlepiment(Tamisieretal,2020).Eneffet,l'accumulationviraleestuneétapecrucialecarellecontrôlenonseulementlaquantitédevirusauseindesonhôtemaisestégalementessentiellepourtoutes lesétapessuivantesde lamigrationcellulaire, de la colonisation systémique et de la transmission du vecteur. Le principalobjectifdemonstageétaitl'évaluationdediversesapprochesquantitativespourmesurerl'accumulation du virus Y de la pommede terre (PVY, Potyvirus) dans le piment afin decaractériserlesdifférentesrésistancesdupoivronauPVY.Cevirusestassociéàdelourdesperteséconomiquesetbénéficied'unegrandeexpertiseauseindulaboratoire"PathologieVégétale"duCentreINRAEd'Avignonoùilestétudiédepuisprèsde50ans.Ilconstituedoncuncandidatparfaitpouratteindremesobjectifs.
4.1. LepimentLepiment(CapsicumL.)appartientàlafamilledesSolanacéesetaétéimportéd'Amériquelatine. Le genre Capsicum est composé de plus de 35 espèces dont 5 espèces sontdomestiquées : Capsicum annuum, Capsicum baccatum, Capsicum chinense, Capsicumfrutescens et Capsicum pubescens (Moscone et al, 2006). Actuellement, C. annuum estl'espècelapluscultivéeàl'échellemondialeetestreprésentéepardesmilliersdecultivars(Penella et Calatayud, 2018). En raison de leur distributionmondiale, les poivriers sontexposés à de nombreux agents pathogènes, notamment des champignons (Phytophtoracapsici,Rhizoctoniasolani,Verticilliumdahlia,Fusariumspp.),desbactéries(Xanthomonascampestris), des insectes (acariens, termites, pucerons et thrips), des nématodes(Meloidogyne incognita) et des virus (Penella et Calatayud, 2018). En particulier, vingtespèces de virus sont connues pour infecter le poivron, parmi lesquelles dix espècesappartiennentaugenrePotyvirus,notammentlevirusYdelapommedeterre(Mouryetal,2005).
4.2. LePotatovirusYLevirusYdelapommedeterre(PVY)appartientàlafamilledesPotyviridaeetaugenrePotyvirus. Son génome est un ARN simple brin (+), encapsidé dans une particulefilamenteuseetflexueuse.Ilpossèdeunelargegammed'hôtesparmilesquelsdesplanteséconomiquementimportantesdelafamilledesSolanacées(tomate,tabac,piment,pommedeterre).Pourcetteraisonetlafacilitéaveclaquelleceviruspeutêtremanipulédansdesconditions contrôlées, le PVY est signalé comme l'un des dix virus les plus étudiés enpathologievégétalemoléculaire (Scholthofetal,2010).La transmissionnaturelleduPVYimpliquedespuceronsvecteurs,maisdesinoculationsmécaniquespeuventêtreeffectuéesdansdesconditionsdelaboratoire(Astieretal,2001).Plusde40espècesdepuceronssontconnuespourtransmettrelePVYdemanièrenonpersistanteetnoncirculatoire(Scholthofetal,2010).
4.3. RésistancesdupimentauPVYLesinfectionsàPVYsontfréquentesdanslesculturesdepimentetpeuvententraînerdesperteséconomiquesconsidérables(Fereresetal,1996).Surlepiment,lessymptômesduPVYsontcaractérisésdesmosaïquesoudesnécrosesselonlegénotypedupimentetl'isolatduPVY(Dogimontetal,1996).Selonlesestimations,environ40%detouteslesaccessionsde piments sont résistantes aux isolats actuels de PVY, et plusieurs mécanismes derésistance sont impliqués (Charron et al, 2008). Parmi les sources de résistance bienidentifiées:
-pvr2,localisésurlechromosome4,confèreunerésistancerécessiveauPVY.Àceniveau,34allèlesderésistancesontclassésdepvr21àpvr234etunallèledesusceptibilitéest noté pvr2+. Ces allèles codent pour une protéine appartenant à la famille eIF4E(eukaryotictranslationinitiationfactor4E).Surungénotypesensible,eIF4Eestréquisitionnépar la protéine VPg du virus pour initier la traduction ou la réplication de l'ARN viral.Cependant, lorsqu'uneplanteportedeuxallèlesde résistancede cegène, l'infectionestinhibée(Ruffeletal,2004).
- Les QTLs pour les résistances quantitatives au PVY ont été identifiées sur desaccessionsdoublesdepimentshaploïdesobtenuesàpartird'hybridesF1entreuncultivardepimentrésistant,"Perennial"(portantl'allèlederésistancerécessifpvr23)etuncultivarsensible"Yolowonder"(portantunallèlesensiblepvr2+).Enraisonde laségrégationdesallèles parents au niveau du locuspvr2, ces accessions présentent différents niveaux derésistancepartielleauPVY.
5. ObjectifdustageetstratégieexpérimentaleEnrésumé,lestageaunobjectifmajeur:évaluerlesrelationsentrelesrésultatsfournispardifférentesméthodes de quantification afin de proposer un outil fiable pour le suivi del'accumulationde virus dans des plants de piments infectés. Pour atteindre cet objectif,l'approcheexpérimentaleconsisteàmesurerlachargeviraleassociéeàl'infectionparlePVYdanslesplantsdepimentparELISA,RT-qPCRetimageriedefluorescence,puisàévaluerlescorrélations et les relationsmathématiques entre les résultats. Historiquement dans lelaboratoirede"PathologieVégétale",laquantificationdesprotéinesviralesestassuréepardestestsELISA.Récemment,lestestsd'infectivitésontremplacésuneapprochequantitativenon destructive basée sur la mesure de fluorescence d'un isolat de PVY modifié maisnécessitant quelques ajustements. Enfin, la quantification des acides nucléiques du PVYnécessiteledéveloppementd'unprotocoledeRT-qPCR.
Acknowledgements:Firstofall,Iwouldliketothankmytrainingsupervisor,LoupRimbaud.Thankyouforalltheknowledges about plant virology, statistic and modelling you shared with me in a verypedagogicalway.Ialsothankyouforyourprecioustipsonthemasterthesisredaction,yourquick mindedness and of course your full support during the lockdown period. I reallyenjoyedworkingwithyouandIwillrememberallyouradvicesforalongtimeIthankalsomyco-supervisors,MarionSzadkowskiandJudithHirsch.Thankyoubothforyour pedagogy, enthusiasm and your unfailing support. I really appreciated all ourvideoconference during the lockdown and all our regular meetings during the wholeinternship.Thankyoualso,BenoîtMouryforallyoursoundadvicesduringthemeetings.Ialsothankmyreferentteacher,AlexandreDegraveandtheplantprotectionheadteachersforyourpedagogicalsupervisionandyourpositivewordsduringtheconfinement.I thank all the members of the Plant Pathology laboratory for their kindness andcommunicativejoy,Catherine,Gregory,Karine,Magali,Alexandra,Jonathan,Clara,Corinne,Claudine,PascaleandofcourseMarcBardinforwelcomingmeduringthesesixmonths.Andofcourse,thedoctoralstudentsandemployees,Paul,Estelle,ThomasandRoxanneforallourinterestingdailyconversationsandlaughter.Thankyouagainforwelcomingmewithsuchopenarms,duringmyunfortunatelyshortenedvisittothePlantPathologylaboratory.
Figure 9: RT-PCR protocol for detection and quantification of PVYson41p-115k-GFP inextractedRNAfromleafsamples.
Figure 10: Estimation of the viral load of PVYson41p-115k-GFP infected plants byfluorescence imaging followed by semi-quantitative ELISA assay (B, Blank; C, Negativecontrol)
Figure14:HeatmapofthenormalizedODvalues,averagedfrom9plates(loadedwithC.annuum cv. ‘Yolo Wonder’ leaf extracts) after 4h of incubation with p-nitrophenyl-phosphateatambienttemperature
Table 7: Nucleotide sequences of primers designed and used for RT-PCR to detectPVYson41p-115k-GFPand22otherPVYisolates
Figure 22: The 1.0% agarose gel electrophoresis analysis showing the detection ofPVYson41p-115K-GFP from infected leaf samples of C. annuum cv. ’Yolo Wonder’ bydifferent primers couples using a RT-PCR assay targeting the coat protein gene and theuntranslated3’region.C=control(RNAextractofanhealthyC.annuumcv.’YoloWonder’).
Figure 25: Regression curves between coat protein concentration and the proportion offluorescent surface in PVYson41p-115K-GFP infected leaves of (A) C. annuum cv. ‘YoloWonder’,(B)N.benthamiana,(C)N.tabacumcv.’Xanthi’.
Table8:Meanand95%confidence interval for theparametersof thenon-linear logisticequationsestimatedfortwocultivarsofC.annuum,‘YoloWonder’and‘Perennial’
Figure 26: Kinetics of the proportion of fluorescent leaf surface in PVYson41p-115k-GFPinfectedleavesofC.annuumcv.‘YoloWonder’and‘Perennial’.
3. Assessing the correlation of the variables provided by different quantificationmethods............................................................................................................................21
Theenhancementofcropprotectionpracticesagainstharmfulorganismsisoneofthemainissuesofcurrentagronomicresearch.Amongthese,plantpathogensarecharacterizedbyadiversityofbiologicalorganismslikenematodes,fungi,bacteriaandviruses(Oerke,2005).This diversity of pathogens triggers various sources of crop productivity reduction, fromtissue consumption, sap assimilation tometabolism disorders, globally compromising atleast10%oftheglobalfoodproduction(Booteandal,1983;StrangeandScott,2005).Sincethe1950’stheindustrialdevelopmentofagriculturehasledtoamassiveuseofcropprotectionchemicalproductsinordertominimisethelossescausedbypathogensandpests(Savary and al, 2019). However, the diseases caused by viruses differ significantly frombacterial or fungal pathologies. Viral infections are persistent, generalized and generallyincurable(DevergneandAlbouy,1991).Consequently,cropprotectionpracticesarelimitedsincenochemicaltreatmentisavailabletocureplantsinfectedbyviruses(Mouryandal,2010). Some prophylaxis methods such as thermotherapy and cross-protection showpromisingresultsonreducingviral infectionsbutareassociatedwith labor intensiveandexpensive procedures (Thompson and Tepfer, 2010; Varveri and al, 2015;Wang and al,2018).Inaddition,theefficiencyofvectorcontrolhighlydependsonthevirusspeciesanditstransmissionmode(Perringandal,1999).Takingintoaccounttheselimits,efficientcropprotectionpractices require themanagementofviraldiseasesbeforediseaseoutbreaks.Hence,theuseofplantresistancecouldbeamajorcatalystinproposingefficientandlow-inputsolutions(Mouryandal,2010;Boualemandal,2011).Theselectionofplantresistanceis already a current method to limit virus infections. However, assessing resistanceexpression is aprerequisite fora sustainableuseofplant resistancewhichcaneasilybecompromised by the counter-adaptation of the targeted virus. Such adaptation leads toresistancebreakdown,awasteofraregeneticresourcesassociatedwithheavyeconomiclosses.Myinternshipsubjectpertainstoviraldiseasemanagementbytheuseofplantresistancesand assessment of their phenotypic expression on a step of the viral cycle. In order toprovide background to better understand the results presented in this report, theintroductionisdividedinfourmainparts.First,theviralcyclewillbepresented,alongwiththemainmechanismsofplantresistancetoviruses.Next, thecurrentmethods forplantresistanceassessmentwillbepresented,beforestatingtheobjectivesoftheinternshipandpriorknowledgeonthecasestudyofthiswork:theinteractionbetweenpepper(Capsicumannuum)andPVY(PotatovirusY).
Figure1:Globalinfectiouscycleofanaphidtransmittednon-circulativeplantvirus(adaptedfromHipper,2013).Thisviralcycleisasimplifiedrepresentationofatype“cycle”inthecaseofacompatibleinfectionofahostbyaplantvirus.(1)Aviruliferousaphidinoculatesthevirustoahealthyplant.(2)Intheinoculatedcells,thevirusreplicatesanditsgenomeistranslatedinto viral proteins. (3) The newly formed virus particles and ribonucleoproteins migratethroughplasmodesmata.(4)Thevirusparticlesand/orRNPreachthesieveelementswheretheyaretransportedtothedistaltipsoftheplant.Thevirusentersanewsiteofinfectionandcontinuesitsreplication.(5)Thepropagationofthevirusinsidetheplantallowsitsacquisitionbyanaphidwhichmaytransmitittoanotherplant.
7
Virusspeciesmayhaveseveralpathsfortransmission(Matthews,1981;Astierandal,2001).Forinstance,tobamovirusesliketheTobaccomosaicvirus(TMV)andtheemergingTomatobrownrugose fruitvirus (ToBRFV),are transmitted throughseeds,circulatingwatersandmechanicalcontactsbetweenplants(Salemandal,2016).Thesemechanicalcontactscause“fresh”woundswhichallowsthesevirusestoovercomethefirsthostbarrierstoinfection(cuticleandcell-wall)(DevergneandAlbouy,1998).However,themostcommonwayforvirusestobetransmittedisviainsectvectorswhichmaybetelluricoraerial(Astierandal,2001).Telluricvectorsincluderestrictedgroupsofnematodes and fungi (Walkey, 1991). Aerial vectors include insects like whiteflies,mealybugs,leafhoppers,planthoppersandaphids(Jones,1987;Astierandal,2001).Aphidsarebyfarthemostimportantvectorsastheytransmitabout30%ofallplantvirusesknownto date (Brault and al, 2010). Besides, transmission strategies depend on highly specificinteractions between viral proteins and insect receptors which highlights an intensecoevolutionbetweenthepathogenanditsvector(Braultandal,2010).Therearetwomainmodesofplantvirustransmissionbyinsectvectors:non-circulativeandcirculative.
- Non-circulative transmission is themost commonmodeof transmission in plantviruses.Theviralparticlesremainboundtotheaphidstyletsonspecificreceptors(Khelifa, 2019). This step is called retentionand itsduration canvary froma fewminutes (non-persistent) to a few hours or days (semi-persistent) (Brault and al,2010).Theviralparticleandinsectstyletscaneitherbinddirectlythankstothecapsidprotein(Cucumbermosaicvirus)orindirectlyviaanadditionalproteincomponent,the helper component protein (HC) (Brault and al, 2010). These viruses are nottransmittedtotheprogenyoftheirinsectvector(Astierandal,2001).
- Circulativetransmissionrequirestheretentionoftheviralparticleinsideitsvectorandthecrossingofdifferentbarrierstoreachthehemolymphandtheotherorgans(salivaryglands,reproductiveorgans…)(NgandPerry,2004;Braultandal,2010).Inmostcases,theretentionperiodcanlastforthevectorentirelifespan(Braultandal,2010).Somevirusescanevenmultiply in theirvectorcells like theRhabdoviridae(Sylvester,1980).
Oncethevirushasentereditshost,ittakesadvantageofthehostcellbyutilizingitsenergyandmachineryforthesynthesisofitsownproteinsandnucleicacids(figure2).For(+)RNAplantviruses(about75%ofallplantviruses)thenucleicacidsareseparatedfromthecoatprotein(Astierandal,2001).ThentheviralRNAassociateswiththehostribosomeswhichensurethetranslationoftheviralRNApolymeraseandsomeothervirus-specificproteinslikecoatproteins(Matthews,1981).The(+)RNAisusedasatemplatetotranscribecopiesof (-)RNA which will then ensure the synthesis of complementary (+) RNA molecules(DevergneandAlbouy,1998).Newviralparticlesarecomposedofnewly replicatedviralRNA encapsidated by the coat protein (Astier and al, 2001). As the multiple cycles ofreplicationandtranslationoccur,thereisanaccumulationofviralRNA,viralparticlesandviralproteinsintheprimaryinfectedhostcell(DevergneandAlbouy,1998).Thentheviruscancolonizetheplantsthroughplasmodesmataasvirusparticlesorasribonucleoprotein(RNP)complexesandinitiatenewvirusinfectioncycles(Matthews,1981;DawsonandHilf,
Figure2:Mainstepsoftheviralcycleatthecellularscale(adaptedfromDevergneandAlbouy,1998;Astierandal,2001).(1)Theviralparticleisintroducedtoahostcellandits(2)RNAisseparatedfromthecoatprotein.(3)TheRNApolymeraseistranslatedbyahostribosome.(4)The host ribosomes translate viral proteins and the RNA polymerases ensure virus RNAsreplication.(5)ViralRNAs,particlesandproteinsareaccumulatinginthehostcell.
8
1992, Lucas, 2006).When the virus reaches the vascular system, it colonizes thewholeorganism leading to a generalized disease recognizable by its symptoms (Devergne andAlbouy, 1998). These symptoms include pigmentation anomalies (mosaics, mottledappearance,ringspots,jaundicesandflowerand/orfoliagevariegations),reducedgrowth(stunting),deformationsandnecrosis(Walkey,1991;Astierandal,2001;Valverdeandal,2012).Furthermore,virusinfectionscausephysiologicalandbiochemicalchangessuchasadecreaseinphotosynthesisefficiency,anincreaseinrespiratoryrate,anincreaseofcertainenzymeactivities(polyphenoloxidases),anaccumulationofoxidizedpolyphenolsandfinallyanincreaseoradecreaseofplantgrowthregulators(Matthews,1981).Allthesesymptomsleadtoareductionincropyieldandquality,resultingineconomiclossesforproducers.Finally,thelaststepoftheviralcycleisthetransmissiontoanotherplant(accordingtooneofthepathwaysdescribedpreviously)whichisessentialforthesurvivalandgeographicaldisseminationoftheviralpopulation.
RNAinterferenceconstitutesadefenseagainstvirusesbyamechanismofgeneexpressionregulationwheretheviralRNAundergoesahighlyspecificdegradation(Astierandal,2001).Oncetheplantcell is infectedbyaRNAvirus, theviralRNAreplicatesandformsdoublestrandedRNAmolecules(dsRNA)whoseunusualpresenceisdetectedinthehostcell.ThesedsRNAsarethenprocessedbyDicer-likeenzymesintovirus-derivedsmallinterferingRNAs(siRNAs), (Sharma and al, 2013). These siRNAs are taken into theRNA induced silencingcomplex(RISC)whichinitiatesthecleavageofthetargetviralnucleicacids(Sharmaandal,2013,Nicaise,2014).Inaddition,theintegrationofsiRNAsintheRISCgeneratesamobile-silencing signal which is transferred to the adjoining cells and the whole plant throughplasmodesmata and the phloemand amplified by a relay-amplification process (Nicaise,2014).Consequently,thissignalactivatesRNAsilencinginnon-infectedcells.Thisprocessisinvolvedincross-protectionstrategies,whichpreventinfectionbyavirusspeciesfollowingpreviousinfectionbyarelatedstrain(Sharmaandal,2013).
1.2.2. Qualitativeandquantitativeresistances
Qualitativeresistancesareknowntoconferstrongorcompleteresistanceagainstaspecificparasite.Theyaremostlybasedonthespecificrecognitionofavirus-encodedavirulencefactor(likethecoatprotein)byaplantR(resistance)protein,belongingtothenucleotidebinding leucine-rich repeat (NLR) class (Nicaise, 2014; Boualem and al, 2016). Thisinteraction initiates a mitogen-activated protein kinase (MAPK) signaling cascade. TheactivationofRgenestriggersahypersensitiveresponse(HR),whichisarapidinductionofprogrammedcelldeathfortheinfectedcellsandtheirsurroundingneighbors.Itresultsina
9
confinementofthevirusandpreventsanyfurtherviralcolonizationintheplant(Nicaise,2014).Itcanalsoresultinafastaccumulationofreactiveoxygenspecies(ROS)anddefensehormones (salicylic and jasmonic acids) leading to a systemic acquired resistance (SAR;Nicaise,2014;Boualemandal,2016).Fromasemiologicalpointofview,thisleadstotheappearanceofnecroticspotsontheinfectedorgansafterafewhours(Astierandal,2001).Besides,insomeinstances,viralinfectionistotallyclearedwithoutanynecrosisreaction;this is calledextremeresistance (Astierandal,2001). In spiteof the rarityofqualitativeresistancegenes,thiskindofresistanceisfrequentlyusedbybreedersasthesegenetictraitsareeasily selected (Fraser,1992).However,as theyapplya strong selectivepressureonviruses which are known to present high mutation rates compared to other biologicalorganisms,theseresistancesmaybeeasilyovercomebyviruses(Gagoandal,2009)viathemutationoftheiravirulencefactors.In contrast, partial or quantitative resistances only soften pathogen infection (hencereducingtheimpactonplantyieldorquality)withoutpreventinginfection.Theyaremoreabundantincropgeneticcollections.TheseresistancesarenotlinkedtoaspecificmolecularmechanismandtheirgeneticdeterminismrelyonnumerousQTLs(QuantitativeTraitLoci)(Lindhout,2002).Quantitativeresistanceisoftensupposedmoredurablethanqualitativeresistance,however,astheuseofquantitativeresistanceincultivatedcropsislessfrequentthanqualitativeresistance,fewdataareavailabletodocumentthishypothesis(Lecoqandal,2004).
1.2.3. Recessiveresistances
This kind of resistance ismore common for viruses than for other plant pathogens andrepresentsaboutonehalfofthe200knownresistancegenes(Diaz-Pendon,2004;Nicaise,2014). It relieson recessivegenesencoding foreukaryotic initiation factors (eIFs)mainlyrecruitedbyviruses(includingespeciallypotyviruses) fortheirreplicationandtranslation(Nicaise,2014).Indeed,theviralcyclerequiresaseriesofcompatibleinteractionsbetweenhostandviral factorsalongallstepsoftheviral infection,fromviralRNAtranslationandreplicationtolong-distancetranslocationthroughthevascularsystem(Diaz-Pendonandal,2004).Asaconsequence,ifarequiredhostfactorislackingorpresentsamutatedversion,theviruscannotcompleteitscycle,resultinginhostresistance(Astierandal,2001;Diaz-Pendonandal,2004).
1.3. Plantresistancesandviralcycle
Asillustratedinthepreviousparagraphs,therearemultiplemechanismsofplantresistanceagainstviruseswhicharecomplementaryintermsofdefensetiming(atearlyorlatestepsofinfection),location(inthefirstinfectedleaforinsystemictissues)andtheviralmoleculesthat are targeted (viral genome or proteins) (Lecoq and al, 2004; Nicaise, 2014). As aconsequence, itcanbehypothesizedthatdifferentresistancegenesmaytargetdifferentstepsoftheviralinfectiouscycle.Thishypothesisissupportedbythefollowingexamples.
Resistancetovirustransmissionisassociatedwithaplantdominantresistancetothevectoritself leading to the failure of the inoculation or the acquisition (Pochard, 1977). Theseresistancesweredescribedonmanyvectorssuchasnematodes,fungiandinsects(Jones,2006). For instance, themelonVat (Virus aphid transmission) gene interfereswith virustransmission by the aphid species Aphis gossypi. Plants carrying the Vat gene are notinfectedafteranaphid-mediatedinoculationwithdifferentvirusesbutaresusceptibletotheseviruseswhentheyaremechanicallyinoculatedorwhenotheraphidspeciesareusedforthetransmission(Martinandal,2005;Boissotandal,2010).
Resistancetovirusmultiplicationintheinoculatedcellmayresult inatotalabsenceorareductionofviralreplicationinthecellandmayrelyonrecessiveordominantgenes.Forexample, in the case of pepper infection by PVY, when a plant carries two recessiveresistancealleleslikepvr21andpvr22,theircombinedpresenceresultsinanincompatibleinteractionbetweenthevirusandaplanttranslationfactor(eIF4E)leadingtotheabsenceofvirusmultiplication(Mouryandal,2004).
In some cases, the virus may multiply in the inoculated cells but cannot move to theadjoiningcellsorcannotreachthevascularsystem. Inthesecases,dominantresistanceswereidentified.Forinstance,intheArabidopsisthalianaecotypeCol-0,TEV(Tobaccoetchvirus)multipliesandmovesfromcelltocellbutcannotmovethroughthevascularsystem.Threedominantgenes,RTM1(RestrictedTEVMovement1),RTM2andRTM3arerequiredforthisresistance(Chisholmandal,2001)Inspiteoftheexampleslistedabove,thescientificliteratureontheeffectofplantresistanceonthedifferentstepsoftheviral infectiouscycleisstillscarce. Inparticular, ithasneverbeendemonstratedthatdifferentresistancegenesmaytargetdifferentstepsofthecyclefor a given plant/virus combination. Yet, the implementation of suitable and efficientresistancesincropplantsisguidedbyavailableknowledgeonthemolecularmechanismsand phenotypic effects of plant virus resistances (Nicaise, 2014). Measuring thesephenotypic effects for breeding purposes requires appropriate protocols, and especiallydiagnostic and quantification tools to assess the virus presence and its limitation byresistancemechanisms.
Ofallassessmentmethods,infectivitytestsarethemostbasicsincetheyrelyonlyontheobservationofsymptomsaftertheinoculationofthevirus(Matthews,1981).Historically,thesetestswereconsideredaskeyfactorsforvirusidentificationbecausetheymighthelptodeterminethehostrangeorthesemiologyofanunknownvirusspecies(Walkey,1991).Inpractice, investigators startbybiologically (via vector)ormechanically inoculating thevirustooneofitsplanthostandthenmonitortheoccurrenceofsymptoms(Walkey,1991).However, symptom expression strongly depends on the host plant (genotype andphysiological state), the viral strain, the date of inoculation and the environmentalconditions(temperature,light);thisindicatesthatextremecautionisrequiredtoperformandstandardizethesetests(Walkey,1991;Astierandal,2001).Althoughthesetestsarestillusedinlaboratories,theintroductionofenzyme-linkedimmunosorbentassay(ELISA)inthe1970-80s deeply shortened and simplified the achievement of routine viral diagnostics(Boonhamandal,2014).
2.2. Serologicaltechniques
Serological techniques are frequently used for their specificity, speed and ease tostandardize(ClarkandAdams,1977).Thesetechniquesarebaseduponantigenicreactivityprinciplesrelyingonthespecificbindingbetweentwotypesofmolecules:theviralantigenanditsrelatedantibodyoriginatingfrommammals(Walkey,1991;Astierandal,2001).Inbrief,themammalianimmunesystemhasanabilitytospecificallyrecognizesurfacefeaturesof exogenousmacromolecules ormicroorganisms called antigens (Crowther, 1995). Themammal components performing this recognition are called antibodies which may bemonoclonalorpolyclonaldependingonwhethertheyrecognizerespectivelyoneorseveralepitopesof thesameantigen. Inplantvirology, these immunologyprinciplesareused invarioustechniquessuchas immunoprecipitationorEnzyme linked ImmunosorbentAssay(ELISA). ELISA relies on the sensitive detection of non-precipitating reactions thanks toenzymelabelledantibodies(ClarkandAdams,1977).Inpractice,thismethodisachievedinmicrotiterplatesonwhichantigensarefirstpassivelyadsorbedontheinnersurfacebeforebeingincubatedwithenzymelabelledantibodiesandfollowedbycolordevelopmentwiththeappropriatesubstrate(Crowther,1995;Sakamotoandal,2018).Thispassiveadsorptionofantibodiesfacilitatestheseparationoffreeandboundreagentswhichalsoallowsagreatflexibilityinassaydesign(Crowther,1995).Generally,alkalinephosphatase(ALP)isusedastheenzymeinchargeofthetransformation(incaseofantigen-antibodyrecognition)ofasolublecolorlesssubstrate,p-nitrophenylphosphate(PNPP)intop-nitrophenolwhichcanbedetectedat405nm(yellowcolor,Astierandal,2001;Sakamotoandal,2018).Thereason
12
forELISApopularity resides in itscosteffectiveness, robustness,capacityof testing largenumberofsamplesandsimplicity.However,inthe1990s,newmethodsweredevelopedtodetectviralnucleicacids(Boonhamandal,2014).
2.3. Moleculartechniques
Moleculartechniquessuchasmolecularhybridizationandpolymerasechainreaction(PCR)are basedon sequence complementarity allowing the binding of viral nucleic acidswithsequence-specificDNAorRNAprobes(Rubioandal,2020).ThePCRreliesonasuccessionofdenaturation,primershybridizationandthermostablepolymeraseextensionstepswhichreplicatesthetargetviralsequencedelimitedbyprimers(Astierandal,2001).However,forRNAviruses,areversetranscriptionstepisrequiredbeforethePCRamplification.ItconsistsinsynthetizingacomplementaryDNA(cDNA)strandbasedontheRNAstrandofthevirusthankstoareversetranscriptase(Astierandal,2001;Websterandal,2004).ThisaddedstepisperformedpriortothePCRamplification(Astierandal,2001).Attheend,amigrationoftheamplificationproductsisgenerallyperformedwithanelectrophoresisinanagarosegel,which reveals the presence or absence of a virus (Astier and al, 2001). The PCR and itsvariants are very popular because of their high sensitivity and modulable specificitydependingonthedesiredobjectives(Websterandal,2004).Indeed,themoretheprimersaredesignedinavariableregionofthevirussequence,themorespecificthetestisforaviral species or strain (Astier and al, 2001). However, as the test is highly sensitive, itincreases the riskof falsepositive results due to contaminations (Webster andal, 2004;Bustinandal,2009;Varveriandal,2015).
In addition to their potential for virus detection in a qualitative way, the laboratorytechniquesdescribedabovehaveapotentialforvirusquantification.Forinfectivitytests,alinear relationshiphasbeendemonstratedbetween thenumberof local lesionsand theconcentrationofinfectiveviralparticlesintheinoculationextract(Matthews,1981;Astierand al, 2001). This suggest that scoring the progression of symptomsmay be a way toevaluatequantitativelyviralaccumulation.ELISAcanbeusedforvirusproteinquantificationbasedonthesignal intensity(Rubioandal,2020).Sincetheresultingopticaldensityataspecificwavelengthandafteragivenincubationperiodisproportionaltotheinitialquantityofantigen,theviralconcentrationintheplantextractismeasurablebyspectrophotometry(ClarkandAdams,1977;Astierandal,2001,Rubioandal,2020). Forviralnucleicacids,quantitative (or real-time)PCR (qPCR) relieson thequantificationofa fluorescent signalgeneratedduringDNAamplification,eitherviaanintercalatingagent(e.g.SYBRGreen)orahydrolysisprobe(e.g.Taqman).Itconstitutesofaveryaccuratemethodtoestimatenucleicacidtiterwithawidedynamicrangeandhighsensitivity(TseandCapeau,2010).TherearealsoothertechniqueslikeHighThroughputSequencing(HTS),DNAarrays,lateralflow and molecular hybridization, which allow a quantitative assessment of viral
Asdescribedabove,theabsenceofvirucidesandthedifficultytocontrolinsectvectorswhilemitigatingenvironmentalandhealthissuesconsiderablylimitsthepotentialofchemicalstomanagevirusepidemics.Inthiscontext,theuseofplantresistanceisapromisingstrategy(Mouryandal,2010;Boualemandal,2011).Theuseofreliablequantificationmethodsofvirus infection is essential to characterize the resistance level of a plant to virusaccumulation.Traditionally,thesanitarystateofaplantisdeterminedbytheobservationofsymptomsorbythedetectionofviruscomponentssuchasproteins(revealedbyELISA)ornucleicacids(revealedbyRT-PCR).Thesemethodsweretraditionallyusedtodistinguishsusceptible fromresistantplantgenotypes (Lecoqandal,2004).However,quantitativelyestimatingtheamountofavirusallowsamorepreciseinformationonthelevelofagivenplant resistance. For example, it has been shown that virus accumulation is an accurateestimator of quantitative resistance for pepper (Tamisier and al, 2020). Indeed, viralaccumulationisacrucialstepasitcontrolsnotonlythevirusquantitywithinitshostbutisalsoessentialforallthefollowingstepsofcellmigration,systemiccolonizationandvectortransmission.Themaingoalofmyinternshipwastheassessmentofvariousquantitativeapproachestomeasure the accumulation ofPotato virus Y (PVY, Potyvirus) in pepper and destined tocharacterize different pepper resistances to PVY. This virus is associated with heavyeconomiclossesandbenefitfromagreatexpertiseatthe“PathologieVégétale”laboratoryof the INRAE Center of Avignon where it has been studied for nearly 50 years. It thusprovidesaperfectcandidatetofulfillmyobjectives.
4. Caseofstudy:infectionofpepperbythePotatovirusY
4.1. Pepper
Pepper (Capsicum L.) belongs to the Solanaceae family and was imported from latinAmerica.TheCapsicumgenusiscomposedofmorethan35speciesamongwhich5speciesaredomesticated:Capsicumannuum,Capsicumbaccatum,Capsicumchinense,Capsicumfrutescens andCapsicumpubescens (Mosconeandal,2006).Currently,C.annuum is themore cultivated species at a global scale and is represented by thousands or cultivars(Penella and Calatayud, 2018). Due to their worldwide distribution, pepper plants areexposedtonumerouspathogens including fungi (Phytophtoracapsici,Rhizoctoniasolani,Verticillium dahlia, Fusarium spp.), bacteria (Xanthomonas campestris), insects (mites,termites,aphidsandthrips),nematodes(Meloidogyneincognita)andviruses(PenellaandCalatayud, 2018). In particular, twenty virus species are known to infect pepper, among
PotatovirusY(PVY)belongstothePotyviridaefamilyandthegenusPotyvirus.Itsgenomeissinglestranded(+)RNA,encapsidatedinafilamentousandflexuousparticle(figure3).IthasawidehostrangeamongwhicheconomicallyimportantplantsfromtheSolanaceafamily(tomato,tobacco,pepper,potato).Forthisreasonandtheeasewithwhichthisviruscanbemanipulatedincontrolledconditions,PVYisreportedasoneofthetenmoststudiedvirusinmolecularplantpathology,accordingtoScholthofandal(2010).NaturalPVYtransmissioninvolves aphid vectors, but mechanical inoculations can be performed in laboratoryconditions(Astierandal,2001).Morethan40aphidspeciesareknowntotransmitPVYinanon-persistentandnon-circulativeway(Scholthofandal,2010).
4.3. PepperresistancestoPVY
PVY infection are frequent in pepper and may result in considerable economic losses(Fereres and al, 1996). Symptoms are characterized by systemic mosaic or necroticsymptomsdependingonthepeppergenotypeandthePVYisolate(Dogimontandal,1996).Accordingtoestimates,about40%ofallpepperaccessionsareresistanttocurrentisolatesofPVY,andseveralresistancemechanismsareinvolved(Charronandal,2008).Amongthewellidentifiedresistancesources:
Tosumup,theinternshipheldonemajorpurpose:assessingtherelationshipsbetweentheresultsprovidedbydifferentquantificationmethodsinordertoproposeareliabletoolformonitoring virus accumulation in pepper. To address this objective, the experimental
approach consisted inmeasuring the viral load associatedwith PVY infection in pepperplantsbyELISA,RT-qPCRandfluorescenceimagingandthenassessingthecorrelationsandmathematical relationships between the results. Historically in the “Plant Pathologylaboratory", the quantification of virus proteins was ensured by ELISA assays. Recently,instead of doing infectivity tests, a non-destructive quantitative approach based on themeasurementofafluorescentlytaggedPVYisolate,hasbeendevelopedbutrequiredsomeadjustments.Finally,thequantificationofPVYnucleicacidsneededthedevelopmentofacompleteRT-qPCRprotocol.However,developingafullRT-qPCRprotocoltowardthisobjectivewasdifficultbecauseofthecurrentCovid-19pandemicandthelockdownperiod,soitremainedincomplete.
Two contrasted pepper genotypes were used for the experiments. The cultivar ‘YoloWonder’hasasusceptibleallele(pvr2+)andsowasusedasasusceptiblegenotypewhereas‘Perennial’wasusedasaresistantgenotypeduetoitsrecessiveresistanceallele(pvr23)andatleastitsthreeQTLsconferringaquantitativeresistancetoPVY(Montarryandal,2012;Quenouille and al, 2013). Moreover, Nicotiana tabacum cv. ‘Xanthi’ and Nicotianabenthamiana also from the Solanaceae familywere used for viral multiplication beforemechanicalinoculationofpepperplants,aswellasforsomeviralquantificationtests.For all plant species and cultivars, the seeds were first manually sown in batches andapproximately two weeks after germination, the plantlets were manually disposed inindividualpots.Alltheplantsweregrowningreenhouseswithnaturallightconditionsandnothermiccontrol(exceptviafanstoavoidextremesummertemperatures).
1.2. Virusmaterial
ThePVYSON41p-115K-GFPisolateisnamedaftertheweedSolanumnigrum(‘SON’),onwhichitwas originally collected in Montfavet in 1982 (Moury and al, 2003). For years, it wasconservedbyserialinoculationsofC.annuumplantsinthe“PathologieVégétale”laboratory(hencetheletter‘p’).Asingleaminoacidsubstitutionoccurred,andthethreonine(115thaminoacidintheVPgaminoacidsequence)wasreplacedbyalysine(K)allowingthevirusto overcome the major resistance allele pvr23 of the pepper cultivar ‘Perennial’. Morerecently,afluorescentmarker,thegreenfluorescentprotein(‘GFP’)sequencewasaddedtothegenomeoftheisolateinordertobeabletovisuallymonitorthecolonizationofthevirus.Consequently,thisPVYisolateisconsideredasageneticallymodifiedorganism(GMO)anditsmaintenanceandstudyfollowdrasticrules.Accordingly,theplantsinoculatedwithPVY‘son41p-115K-GFP’weredisposedina“S3type”greenhousewhichconsistsofahighlevelofconfinementtopreventtheundesirabledisseminationofsuchmaterial.ThecollectionofPVYisolateswasmaintainedinthelaboratoryintheformofdehydratedinfected plant material (Bos, 1969). This process relies on thinly cutting one gram ofsystemicallyinfectedleavesandstockingitat4°Cinairtightboxeswithcalciumchloride.
minbeforebeingwashedwithtapwater.Approximatelysevendayspostinoculation(dpi),the occurrence ofmosaic symptoms (figure 5) on the leaves ofN. benthamiana andN.tabacum cv. ‘Xanthi’ revealed that thevirushadefficientlymultiplied.Thecotyledonsofhealthypepperplantscouldthenbeinoculatedfollowingthesameprotocolbyusingcrudeextract(1g)ofN.benthamianaandN.tabacumcv.‘Xanthi’leavesassourcesofvirus.Intheabsenceofsymptoms(i.e.atearlystagesofinfection)thesanitarystatusoftheplants(i.e.PVYinfectedornot)couldbecheckedbyfluorescenceimagingorDAS-ELISA.Twotypesofcontrolsweresystematicallyadded:mock-inoculatedplantswereinoculatedonlywiththeinoculationmix(buffer,coalandCarborundum®)andwithoutanyviralsource,andhealthyplantcontrolswerenotinoculatedatall.
2. Quantitativeestimationsoftheviralload
Inthe“PlantPathologylaboratory",theassessmentofvirusaccumulationwastraditionallyperformedby semiquantitativeELISA, a reliablemethod for coatproteinquantification.However,asthismethodisusedregularly,manipulatorswantedtooptimizetheirprotocolby assessing the influences of different factors on ELISA results. Moreover, in order tosimplify and increase the efficacy of the quantification, two candidate quantificationmethods based on fluorescence imaging and RT-qPCR required respectively a furtheranalysisandafulldevelopment.Consequently,thethreemethodsusedforestimatingtheviralloadandtheirrespectiveoptimizationordevelopmentweretobedescribedseparately.
2.1. ELISA
2.1.1. GeneralDAS-ELISAprocedure
Foralltheexperiments,onlyF96Maxisorpnun-immunoplate®(Thermoscientific,Denmark)wereused.First,the96wellsofeachplatewerecoatedwith150µLofantibodies(titteredat1:2000)beforebeingplacedinanincubatorfor3hoursat37°C(figure6).Thentheplateswerewashed three timeswithwater and oncewith phosphate-buffered saline solution(PBS)andpatteddry.Straightafter,thewellswerefilledwith150µLofleafextract(1gofleafgroundin4mLbuffer,seetableS1)andplacedinarefrigeratorat4°Covernight(about15hours).Thentheplateswerewashedagainasdescribedabovebeforebeingfilledwith150µLofthesecondlayerofantibodies(titteredat1:2000)labelledwithanenzyme,thealkalinephosphatase. Theplateswere incubatedagain at 37°C for 3hoursbeforebeingwashedandfilledwith150µLofthesubstratesolutionofp-nitrophenylphosphatedilutedin thesubstratebuffer (1%).During the incubation,acoloredreactionoccurredwhenp-nitrophenyl phosphate was dephosphorylated by phosphatase activity leading to theformationoftheyellowcoloredp-nitrophenol.Theopticaldensitieswerethenmeasuredafter four hours of substrate incubation thanks to a Multiskan FC (Thermo scientific,Denmark)microplatereadercalibratedat405nm.
Figure8:Relativequantification ina leafsample (greencurve) incomparisontoapositivestandard (redcurve). In thegreyarea, it ispossible todeterminetherelativecoatproteinconcentrationinthesamplethankstothedifferenceintheabscissasassociatedwithagivenopticaldensitybetweenthetwocurves.
Severalauthorsreportedtheexistenceofopticaldensity(OD)gradientswithinmicrotiterplates and reported especially an “edge” effect (Burt and al, 1979;Oliver and al, 1981).Knowing these possible variations, manipulators at the research unit “Plant Pathologylaboratory”oftenchoosetoavoidtheperipheralwellsforquantitativeELISAtests.However,this represents a loss of 36 wells by plate. Moreover, protocol differences betweenmanipulatorsandresearchunitssuggestedthattwootherfactorscouldinfluencetheODvalues:lightconditionduringthesubstrateincubation(whichmaybeacauseoftheedgeeffect) and the plates washing method (which could trigger contaminations betweenneighboringwells).Consequently, we ran some preliminary experiments to choose the best procedure forquantitativeELISA tests. ThePVYSON41p-115K-GFPwasmechanically inoculatedonpepper (C.annuumcv.‘YoloWonder’)andtobacco(N.tabacumcv.‘Xanthi’)plants.Foragivenplantspecies,leaveswerebulkedtoserveaspositivesamplesinELISA.Totesteachfactor,differentdesignswereelaboratedonELISAplates:
TherelativeconcentrationofPVYSON41p-115K-GFPin infectedplantswasdeterminedbysemiquantitativeELISA,inordertoestimatethevirusloadintheleavesrelativetoareferencesample.Tocomparevirusconcentrationbetweensamples,eachleafsamplewasweightedandanadjustedvolumeofgrindingbufferwasaddedtomakea1:5dilution.Then,foreachsample,1/5thdilutions ingrindingbufferweredeposited(dilutionfactors1:25to1:3125,seefigure10,step3).Auniquereferencesample(andassociateddilutions)wasalsoaddedto all the plates. ODs of all wells were obtained using the same protocol as decribedpreviously(seesection2.2).Then,foragivensample,theODsassociatedwiththedifferentdilutions give a curve (ideally linear, figure 8). The comparison of this curve (slope and
interceptofoneofitssegments)withthecurveassociatedwiththereferencesampleallowstheestimationof a relative viral concentration.AnR script developedby Lucie Tamisier(formerPhDstudent)andenhancedbyLoupRimbaud(Researcher)wasusedtocomputetherelativevirusconcentrations.An absolute quantification could be obtained if the viral concentration in the referencesamplewasalreadyknown.Thiscouldbeeasilydonewithapurifiedvirussolution.However,asitwouldhavebeenlaborintensive,nopurificationoftheisolatePVY-son41p-115K-GFPwas performed. Consequently, all the obtained concentrations were considered as“relative”concentrations.
Here the purposewas to quantify the PVYnucleic acids in the samples by developing aReverseTranscriptionquantitativePolymeraseChainReaction (RT-qPCR).As themethodwasnotyetusedonpepperinfectedbyPVYinthe“PathologieVégétale”laboratory,thewholeprocedurehadtobedeveloped.
2.3.1. Primerdesign
Primersforvirusdetectionandquantificationweredesignedinthecoatproteingeneand3’untranslated region of the genome (figure 3),which is a highly conserved region of the
In order tominimize RNA degradation, the leaves were separated from the plants andinstantlyfrozeninliquidnitrogen(-196°C)inamortar.Thealiquotsofgroundpowderwerestoredat-80°C.TwoRNAextractionmethods,TRI-Reageant®-chloroform(MRC,USA)andRNeasy®PlantMiniKit(Qiagen,Germany)weretestedinordertofindthebestcompromisebetween RNA yield (measured via Nanodrop®ND-1000 Spectrophotometer, Thermo-Scientific,USA),purity(assessedthroughmigrationon1%agaroseelectrophoresisgelwith0.005%BET,andNanodrop®ND-1000Spectrophotometer)andthemethodcost.Extractionswereperformedaccordingtomanufacturers’instructions.
2.3.3. Primersassessment
AlltheprimersweretestedfollowingaclassicaltwostepRT-PCRprocedure(figure9).Thereversetranscriptionwasensuredbyfillinga0.2mLmicrotubewith2µLofextractedRNA,7µLofRNasefreewaterand1µLofreversePrimer(1:10dilution).AfirstRNAdenaturationwasachievedbyincubatingthetubesat80°Cfor3min(PTC-100™ProgrammableThermalController,MJResearch,USA)beforeadding4.8µLofRNasefreewater,0.8µLofdNTP,4µLof5XbufferRT(Promega®,USA)and0.4µLofreversetranscriptaseAMVRT(Promega®,USA).Thetubeswerethenincubatedat42°Cfor1hfortheannealingstep.ThenthePCRwasachievedbyadding3µLofcDNAina0.2mLmicrotubewith21.8µLofRNasefreewater,8µLof5Xbuffer(Promega®,USA),2.4µLofMgCl2,2µLofforwardprimer(1:10dilution),2µLofreverseprimer(1:10dilution),0.2µLofdNTPand0.6µLofTaqpolymeraseenzyme(Promega®, USA). The PCR program consisted in a prior denaturation at 95°C for 5minfollowedby35cyclesofdenaturation,annealingandextensionforrespectively30sat95°C,30sat62°Cand1minat72°C.Thefinalstepwasanextensionduring10minat72°C.The
In this part, three plant species: C. annuum cv. ‘YoloWonder’,N. benthamiana andN.tabacumcv.‘Xanthi’infectedwithPVYSON41p-115K-GFPwereusedtoperformthetests.Alltheplant leaves were first analyzed by fluorescence imaging to measure GFP quantitativevariablesbeforebeingprocessedbysemi-quantitativeELISA(figure10).
4. Assessingvirusaccumulationinpepper
AsmentionedbyTamisierandal (2020),quantifyingthepresenceofPVYinpepper isanaccuratemethod to characterize plant resistances against virus accumulation. Here twopeppergenotypespresentingcontrastedresistanceprofilestoPVYwereused.
4.1. Fluorescencemonitoring
Beforemyinternship,asetof11peppers‘YoloWonder’plants(including1mock-inoculatedplantand1healthycontrol)and9peppers‘Perennial’plants(including1mock-inoculatedplantand1healthycontrol)hadbeeninoculatedwithPVYSON41p-115K-GFPviaaphids.From5to22 days post inoculation (dpi), the leaves of the inoculated plants were monitored byfluorescenceimagingat5,6,8,9,12,15and22dpi.However,atthistimetheimagescouldnot be analyzed without knowing which was the most relevant variable to assess viralaccumulation(averageleveloffluorescenceorproportionoffluorescentleafsurface,seeabove). Given the results ofmy experiments (see Results section), I took in charge theanalysisoftheseimagesandusedtheproportionoffluorescentleafsurfacetomodelthekineticsofviralaccumulation.
the probability for a pixel to be fluorescent (i.e. of successful viral invasion of thecorresponding part of the leaf), which depended on the number of days after aphid-mediatedinoculation(x).ThislinkfunctionwasalogisticequationpreviouslydesignedbyLoupRimbaud(Rimbaudandal,2015).Themodelparametershadthefollowingbiologicalinterpretations:
AllstatisticalanalysesandgraphicrepresentationsweremadewiththeRsoftwareversion3.5.1(RcoreTeam2012).Allthetestswereconsideredassignificant,whentheobtainedp-valueswerebelowthestatisticalthresholdof5%.FortheoptimizationoftheELISAprotocol,allthetestedfactorswereanalyzedseparately(nointeractions).Meanscomparisonandvarianceshomogeneitytestswereperformedonthe variable to explain (OD values) between themodalities of the explanatory variable(sample position, light condition during incubation, washing procedure). When it waspossible,inordertosimplifytheanalysisofthetests,ODvalueswherenormalizedbytheaverageOD value of the plate,which allowedus to compare different plates filledwithdifferent leafextract concentrations. Finally, if thevalidationconditionsof the statisticaltestswerenotensured(normalityofresiduals,varianceshomogeneityandindependenceof residuals with respectively Shapiro-Wilk, Bartlett and Durbin-Watson tests), moreadequateornon-parametrictestswereperformedinstead(Welch’sttest,Kruskal-WallisHtest).The relationship between the viral load estimated via quantitative ELISA and viafluorescenceimagingwasanalyzedusingPearson’scorrelationcoefficientandageneralizedlinearmodel(GLM)withamodifiedlogarithmlinkfunction.
Figure12:ObservedfluorescenceonC.annuumcv.‘Perennial’mechanicallyinoculatedwithPVY-son41p-115K-GFP(A:stems,B:leaves,C:roots)andmockinoculated(D:stems,E:roots)at42dpi.Fluorescenceaveragelevelisshownbyvirtualcolorsfromdeep-blue(null)tored(highintensity).Table 2: Detection of PVY-son41p-115K-GFP by ELISA in different plant organs for twoC.annuumcultivars(‘YoloWonder’and‘Perennial’).
Plants from two different cultivars of pepper (‘Yolo Wonder’ and ‘Perennial’) weremechanically inoculated with PVYSON41p-115K-GFP. The proportion of infected plants wasassessedbyfluorescenceimagingandserologicaldiagnostic42dayspostinoculation(dpi).All‘YoloWonder’leavespresentedthetypicalfluorescenceinfectionpattern(figure11A)whereasno‘Perennial’leafwasfluorescent(figure12B).Toinvestigateiftheviruscouldbestoredinotherorgansoftheplant,especiallyin‘Perennial’,fluorescenceimagingwastakenon stems, roots and flowers. However, for all the plants genotypes and inoculationprocedures(inoculatedandmockinoculated),therootsandthestemsfluoresced(figures11B, 11D, 11E, 11F, 12A, 12C, 12D, 12E). Flowerswere only present in an infected YoloWonderandwerefoundfluorescent(figure11C).In distinct plants, serological detectionsof PVYSON41p-115K-GFPwereperformedondifferentplantorgans.Here,onlyoneplantof‘YoloWonder’wasusedasapositivecontrol(table2).Asexpected,theleavesofthepositivecontrol‘YoloWonder’showedaΔODvalueabovethepositivitythresholdwhichconfirmeditsinfectedstate(1.87>0.63).Moreover,onlytherootsandtheleavesof‘YoloWonder’wereinfectedaccordingtothepositivitythreshold(1.86 > 0.49). As the OD values for ‘Perennial’ tissues were always under the positivitythresholds(0.06<0.63),no‘Perennial’leavesorothertissueswereconsideredinfected.Inconclusion, the fluorescence of the roots and the stems cannot be considered as goodindicatorsofinfection.Consideringthefailuretomechanicallyinfectthe‘Perennial’cultivar,thisgenotypewasnotincludedinfurtherexperiments.
2. OptimizingthesemiquantitativeELISAprotocol
Todeterminetheinfluencesofthreefactors(wellposition,lightconditionduringsubstrateincubation and washing procedure) on OD values, leaf samples of C. annuum cv. ’YoloWonder’andN.tabacumcv.’Xanthi’infectedwithPVYSON41p-115K-GFPwereusedtoperformDAS-ELISAtests.
wellasthefirstandthetwelfthcolumn.However,wenotedthat6wellspresentverylowaveragenormalizedODvalues(0.90to1).AWelch’st-testshowedthatthemeansoftheaveragenormalizedODvaluesfromtheperipheryandthecenterweresignificantlydifferent(p-value:9.13x10-7).Nevertheless,whentheplatesareanalyzedseparately,1ofthe3and6ofthe9platesloadedwithrespectivelyN.tabacumcv.‘Xanthi’andC.annuumcv.‘YoloWonder’showedsignificantdifferencesbetweentheaveragenormalizedODvaluesofthecenterandtheperipheralwells(table3;p-valuerangingfrom3.77x10-9to0.03).Thefigures15to18showedthegeneraldistributionsofthemeanODvaluesobtainedforeachlineandeachcolumnofanELISAmicrotiterplate.FortheplatesloadedwithN.tabacumcv.‘Xanthi’leafextracts(figure15),thecurvescorrespondingtothelowestconcentrations(1:625and1:15625) showed flat distributions of the mean OD values whereas the highestconcentration(1:25)showedaslightdecreaseofthemeanODvaluesthroughthelinesandcolumns of the plate. Among the plates loaded with C. annuum cv. ‘YoloWonder’ leafextracts(Figures16to18),sixplatesshowedU-orn-shapecurvesrepresentingrespectivelyhigherorlowermeanODvaluesfortheperipheralwells(table3).
2.2. LightinfluenceonthenormalizedODvalues
TheODvalueswerenormalizedbydividingtheODvalueofeachwellbythemeanODvalueofthewholeplatewhichallowedthecomparisonofallplates.Theinfluenceoflightduringthe incubation period of thep-nitrophenyl phosphate showed no significant differences(figure19;p-value=1)onthenormalizedODvalueswhethertheplateswereplacedinthelightduringincubationorinthedark.
2.3. InfluenceofthewashingmethodonODvalues
ThecomparisonbetweenthetwowashingmethodswasbasedonthenormalizedODvaluesassociatedwiththreesampletypes:infectedsample,blankandnegativecontrol.Asthep-values ranged from0.45 to1 (table4), the influenceof thewashingmethodshowednosignificantdifferencesbetweenthenormalizedODvaluesforallsampletypes(figure20).
Figure19:BoxplotsofnormalizedODvaluesmeasuredafter4hofsubstrateincubationinthelight or in the dark. According to aWelch’s t test, no significant differencewas observedbetweenthemeanofthenormalizedODvaluesobtainedwhentheplateswereincubatedinthelightorinthedark(p-value=1)
aWelch’sttestwaspreferredasStudent’sttestbecausethedatadidn’trespectthevarianceshomogeneityhypothesisbKruskal-WallisHtestwasperformedbecausethedatadidn’thaveanormaldistributionTable5:Pearson’scorrelationcoefficient(r)betweencoatproteinconcentrationsandGFPfluorescence (proportion of fluorescent surface and average level of fluorescence) for C.annuumcv.‘YoloWonder’,N.benthamiana,N.tabacumcv.’Xanthi’.
Plantspecies
Relationshipbetweencoatproteinconcentrationsand
proportionoffluorescentleafsurface
Relationshipbetweencoatproteinconcentrationsand
averagelevelofleaffluorescence
r p-value r p-valueC.annuumcv.’YoloWonder’(N=74) 0.63 3.128x10-9 0.39 5.8x10-4
N.benthamiana(N=19) 0.32 0.19 0.70 0.001
N.tabacumcv.‘Xanthi’(N=19) 0.56 0.013 0.55 0.014
Table 6: Summary of RNA isolation from leaf samples using the TRI-Reageant® or theRNeasy®PlantMiniKitprocedure
fluorescence(0.39and0.55forC.annuumcv.’YoloWonder’andN.tabacumcv.’Xanthi’,respectively).Moreover, the general distributions of the data (figure 25) highlighted that for all plantspeciestherelativeviruscoatproteinconcentrationcouldbedescribedbyamonotonicandincreasingfunctionoftheproportionoffluorescentleafsurface.Incontrast,therelationshipbetweenviruscoatproteinconcentrationandtheaverageleveloffluorescencewasnon-monotonic (seefigureS4 inappendices),sotherewasno logical“mathematicalpattern”between the two variables. As a result, in the following we decided to use only theproportion of fluorescent leaf surface as an estimator of the viral load in fluorescenceimaging.
As a first step,we extracted total RNA fromeach sample using twodifferent extractionprotocols:a“home-made”protocolusingTRI-Reageant®(MRC,USA)andathree-yearoldRNeasy®PlantMiniKit(Qiagen,Germany).TheseprotocolswerecomparedbasedoncriteriasuchasfinalconcentrationoftotalRNAandpurityratios.ThecomparisonofaverageRNAconcentrations(table6)revealedsignificantlyhigheryieldsinsamplesextractedwithTRI-Reageant®thantheseobtainedwiththeRNeasy®PlantMiniKit(p-value=0.031).However,RNApurity(table6)wasbetterwithRNeasy®PlantMiniKit,asdemonstratedbytheaverageA260/230andA260/280ofrespectively1.57and2.13whichweretogethermuchclosertothe optimum value of 2.00 for RNA samples. This observation was completed by anelectrophoresismigrationshowingthatmoreRNAsamples(77%ofthe30testedsamples)werevisible(orlessdegraded)withtheRNeasy®PlantMiniKit(figure21)thanwiththeotherkit. It is noteworthy that no DNA contamination was observed for either kits. As aconsequence of all these results, we used the RNeasy®Plant Mini Kit for further RNAextractions.
Five pairs of primerswere designed to amplify a conserved region of the PVY genome,located at the junctionbetween the coat protein sequence and theuntranslated region(3’UTR) (table 7). The size of the respective PCR amplification products were 146 bp(PM2014),150bp(PM202andPM2011),176bp(PM2012)and184bp(PM2013).WhentestedonRNAextractfrompeppers(’YoloWonder’)infectedbyPVYSON41p-115K-GFP,ampliconsoftheexpectedsizeswereobtainedforalltheprimers(figure22).The range of specificity of primers was evaluated using 14 PVY isolates in addition toPVYSON41p-115K-GFP. These tests were performed using only the PM202F-PM202R and
Figure 21: The 1.0% agarose gel electrophoresis analysis showing the quality of the RNAsamplesextractedbytheTRI-Reageant®-chloroformmethod(A)orbytheRNeasy®PlantMinikit (B). Lanes 1 to 30 are different RNA samples extracted from30different PVY infectedC.annuumcv.’YoloWonder’leaves.L=1kbDNAladder(Promega®,USA)Table7:NucleotidesequencesofprimersdesignedandusedforRT-PCRtodetectPVYson41p-115k-GFPand22otherPVYisolates
Name Amplifiedregion Primersequences Length
(pb)Tm(°C)
Expectedsize
PM202F CP 5’-TACACAAGAGGAGAACACAGAGAGG-3’ 25 61.2 150PM202R 3’UTR 5’-CAGGAAAAGCCAAAATACTTACTGC-3’ 25 61.6 150PM2011Fa CP 5’-GGGCTTATGGTTTGGTGCAT-3’ 20 62 150PM2011R CP 5’-GAAATGTGCCATGATTTGCCTA-3’ 22 62.1 150PM2012R CP 5’-TCTATATACGCTTCTGCAACATCTGAG-3’ 27 62.1 176PM2013R CP 5’-TGCGCATTTCTATATACGCTTCTG-3’ 24 62.6 184PM2014R CP 5’-TGTGCCATGATTTGCCTAAG-3’ 20 59.7 146a PM2011F is the adequate forward primer for four different reverse primers (PM2011R,PM2012R,PM2013R,PM2014R)
Figure22:The1.0%agarosegelelectrophoresisanalysisshowingthedetectionofPVYson41p-115K-GFP from infected leaf samples ofC. annuum cv. ’YoloWonder’ by different primercouplesusingaRT-PCRassaytargetingthecoatproteingeneandtheuntranslated3’region.C=control(RNAextractofahealthyC.annuumcv.’YoloWonder’).P=positivesamples(RNAextractsoftwoPVY infectedC.annuumcv. ’YoloWonder’).B=Blank.L=1kbDNAladder(Promega®,USA)
Figure 25: Regression curves between coat protein concentration and the proportion offluorescent surface in PVYson41p-115K-GFP infected leaves of (A) C. annuum cv. ‘YoloWonder’, (B)N.benthamiana, (C)N.tabacumcv. ’Xanthi’.Circles:experimentaldata;solidline:best-fittingnonlinearmodel;dashed line:best fitting linearmodel forN.tabacum cv.’Xanthi’.R2valuesare0.85,0.89and0.86forA,BandC,respectively.
B
A
C
' = +STU?V.X
' = 1.7[
' = +S\.]U?S\.V
' = +V.XU?V − 0.001
26
PM2011F-PM2011Rprimer couples as theyboth have 150 bp ampliconsand the lowestprobabilities to form primer dimers and to hybridize with C.annuum sequences. Theelectrophoresis(figure23)revealedthatbothprimercoupleswereabletodetectalltestedisolatesalthoughthistimethePM2011F-PM2011RcouplefailedtodetectPVYSON41p-115K-GFP.
As the RT-qPCRwas still in development and also as the proportion of fluorescent leafsurface was determined as the best variable for fluorescence imaging, the relationshipbetweentheviruscoatproteinconcentrationandtheproportionoffluorescentleafsurfacecould be assessed. The relationships were established on the same data as before tocomputePearson’scorrelationcoefficient(table5).ConsideringthesetwovariablesforC.annuum cv. ’Yolo Wonder’, the boxplots (figure 24) helped to identify the outlier leafsamplesamongtheleafranks.Indeed,forbothvariables,thehighestvalues,proportionoffluorescent leafsurfacebetween0.84and0.98,andrelativecoatproteinconcentrationsbetween0and5.34,wereachievedbyleavesfromthefirstleafrankfollowedbytheleavesofthesecondleafrank.Onthecontrary,thelowestvaluesofbothvariables,proportionoffluorescent leaf surface between 0 and 0.85, and relative coat protein concentrationsbetween0and4.23,wereachievedbythemoreapicalleafranks(thethirdandthefourth).Surprisingly, two leaf samples (cotyledons) showed high proportions of fluorescent leafsurfacebutverylowrelativecoatproteinconcentrations.Becausethiswasprobablyduetothe fact that cotyledons presentedwounds induced by themechanical inoculation (andwounds emitting a spontaneous fluorescence), cotyledons were removed from furtheranalyses.Then to linearize the relationship between relative coat protein concentration and theproportionoffluorescentleafsurfaceforeachplantspecies,weusedthelogarithmofthecoat protein concentrations (as frequently done for concentration values), including aconstant of 0.001 for C. annuum cv. ’Yolo Wonder’ to avoid infinite values when theconcentration is zero (figure25).ThecalculatedR2valuesare0.85,0.86and0.89 forC.annuumcv. ’YoloWonder’,N. tabacum cv. ’Xanthi’andN.benthamiana, respectively. Inaddition, forN. tabacum cv. ’Xanthi’ a linearmodel fitted on fluorescent surface valuesbetween0and0.92providedaR2valueequalto0.98.Thegeneraldistributionofthedatashowedthatforallplantspecies,thecoatproteinconcentrationvariancesbecomeslargeras the proportion of fluorescent leaf surface increases. Indeed for C. annuum cv. ‘YoloWonder’,thecoatproteinconcentrationsmeasuredwhentheleavesshowedbetween97and 100% of fluorescent surface range from 0.17 to 4.41. The same pattern could beobserved for N. benthamiana and N .tabacum cv. ‘Xanthi’ where the coat proteinconcentrationsrangedrespectivelyfrom0.18to1.52and5.8to77.8andforproportionsoffluorescentleafsurfacebetween97and100%.
Table 8: Mean and 95% confidence interval for the parameters of the non-linear logisticequationsestimatedfortwocultivarsofC.annuum,‘YoloWonder’and‘Perennial’
Figure 26: Kinetics of the proportion of fluorescent leaf surface in PVYson41p-115k-GFPinfectedleavesofC.annuumcv. ‘YoloWonder’and‘Perennial’.Circles:experimentaldata;solid lines: prediction of the dynamics following a single and successful aphid mediatedinoculation using the best-fitting logisticmodel; dashed vertical lines: time to 50% of themaximumvalueoffluorescentleafsurface.
These resultswere obtainedbymonitoring theGFP fluorescence on pepperplants afteraphid-mediated inoculation of PVYSON41p-115K-GFP (figure 26). Fluorescence imaging wasperformedon5,6,8,9,12,15and22dayspostinoculation(dpi).Accordingtotheresults(table8),‘Perennial’showedalesser,sloweranddelayedviralaccumulationincomparisonto‘YoloWonder’.Indeed,themaximumproportionoffluorescentleafsurfacewas77%for‘YoloWonder’andonly9.8%for‘Perennial’.Theinfectedleavesof‘YoloWonder’reached50%of theirmaximum fluorescent surfaceproportion about11.5 days post inoculation,whileitwasabout16dayspostinoculationfor‘Perennial’.Finally,theslopeattheinflectionpointwasabout19%perdayfor‘YoloWonder’and13%perdayfor‘Perennial’.
28
Discussion
1. Arobustquantitativeapproachtoestimateviralload
Inacontextofsustainableuseofplantresistancesagainstviruses,severalmethodologiesforquantifyingtheviral loadinplants inordertoassesstheirresistancelevelcouldhavebeenused.Indeed,astrategycouldhavebeentheestablishmentofacorrelationbetweenasymptomexpressionquantitativevariable(non-destructivemethod)andviruscoatproteinconcentration(destructivemethod).However,someauthorsreporteddifficultiestocloselycorrelate virus titer to symptom expression (Zhu and al, 2010) and distinct resistancemechanismsmaygovernsymptomexpressionandvirusaccumulationleadingtoabreakinthecorrelation(Kaweesiandal,2014).Heretwomethodsforquantifyingthevirusinplantsinfectedbyamodifiedvirusweretestedandcompared.Thefirstonebasedonfluorescenceimaging focusesonGFPemissionmeasurementswhile thesecondbasedonquantitativeELISAfocusesonviruscoatproteins.Amajorresultwasthemonotonicrelationshipbetweenrelativecoatproteinconcentrationandtheproportionoffluorescentleafsurfaceonpepperandtobaccoleafsamples.Biologicallyitseemedintuitivethatasthemodifiedvirusisolatemultipliedwithinitshost,itincreasedthecoatproteinconcentrationintheinfectedtissuesand simultaneously the concentrationofGFP. It shouldbenoted thatnoGFP titerwasactually performed, but itwas estimated by fluorescence imaging. Indeed,whenGFP ispurifiedfromasample,itsfluorescenceintensityisdirectlylinkedtoitsconcentration,whichallowsthedesignofnon-destructivequantitativemethods(Remansandal,1999;Richardsandal,2003).Howeverherewedecidedtousetheproportionoffluorescentleafsurfaceinstead of the average level of fluorescence because the link between the coat proteinconcentrationandtheproportionoffluorescentsurfacewasunequivocallyexponentialforallstudiedplantspecies.ThePearson’scorrelationcoefficient(r)ofthelinearizedmodelswere all above 0.92which indicates a very strong correlation between the coat proteinconcentrationsandtheproportionsoffluorescentleafsurface.Inaddition,forN.tabacumcv. ’Xanthi’,thefitofa linearmodelwasgoodenoughtoprovidearvalueequalto0.99whentheproportionoffluorescentsurfacevaluesarebetween0%and92%.Ithighlightedthat when the proportion of fluorescent leaf surface became close to 100%, there is asaturation effect of this variable while the other (coat protein concentration) keptincreasing.Thus,thelinearitywaslostandthevalueoffluorescencehadtobeconsideredwithcare.This work is destined to be complemented by other experiments, as the relationshipbetween the twopresentedquantificationmethods concernsonly viralproteins. In fact,whenavirusreplicates inthehostcell, theviralproteinsareseparatedfromthenucleicacids(Walkey,1991;Astierandal,2001).AsacompletevirusparticleisdefinedbytheviralRNAsurroundedby its relatedproteins,quantifyingbothviralproteinsandnucleicacidswouldnicelycompletethiswork.
ThedevelopmentofaRT-qPCRmethodwouldbeusefultolinkthesevirusrelatedproteinquantitativevariablestotheRNAquantity.SuchcorrelationhasalreadybeenmadeonN.tabacum cv. ’Xanthi’ infectedwith a GFP labelled PVY isolatewhere themeasured GFPfluorescenceintensityshowedalinearcorrelation(R2=0.76)withvirusRNAconcentration(Rupart and al, 2015). Such result is essential to get a fine understanding of viralquantificationbyRT-qPCRwhichhasahigherspecificitythanELISA(Websterandal,2004;Bustinandal,2009;Varveriandal,2015).However,theentireRT-qPCRprotocolhadtobedeveloped. To date the target region chosen for quantification was the contiguoussequencesofthecoatproteingeneandthe3’untranslatedregionwhichishighlyconservedamong PVY isolate strain groups (Kim and al, 2016). Here all the designed primerssuccessfullyamplifiedthetargetedsequenceinthecoatproteinregion(PM2011,PM2012,PM2013 and PM2014) as well as the 3’ untranslated region (PM202) of the isolatePVYson41p-115K-GFP.Moreover,twooftheprimerpairs(PM202F-PM202RandPM2011F-PM2011R)weretestedon14otherPVYisolatesandwereabletodetectallofthem.TheseprimersarethusgoodcandidatesforthelaboratorytodetectandquantifyanyPVYisolate.Morespecificallytheprimerpair,PM202F-PM202Rmustapparentlybepreferredasitalsopresentedthelowestprobabilitytoformprimerdimerbandsandtohybridizewithpeppersequences.Inaddition,thenextexperimentswillalsousetheRNeasy®PlantMiniKitforRNAextractiongiventhequalityofextractedRNAeventhoughtheyieldissignificantlylowerandthe price is higher thanwith the TRI-Reageant®-chloroformmethod (1.48€ vs 1.00€ perextracted sample respectively). Finally, the assessment of the relationship between viralRNA concentration and the proportion of fluorescent leaf surface will require furtheradjustmentslikechoosingthemostsuitableprotocolforreversetranscription(oneortwosteps)andthebestcandidatereferencegenefornormalizationamongthethreecandidates(actin, ubiquitin and tubulin identifiedbasedon literature). In addition, as a primerpairwhichamplifiesaPCRproductinstandardRT-PCRmayhaveapoorefficiencyinRT-qPCR,thequantificationefficiencyortheprimerswillalsorequirefurtheranalysis.ThemethodforgeneexpressionanalysiswillbebasedonSYBRGreenasitisrelativelycosteffective(Tajadiniandal,2014).
Theproportionoffluorescentleafsurfaceappearstobeagoodestimatoroftheviralload(at least the amount of coat proteins), which can therefore be monitored in a non-destructive way by fluorescence imaging. A non-destructive method presents the mainadvantage toallowadirectquantificationof theviruswithoutany sampleprocessing incontrasttoserologicalormolecularmethods.Moreover,comparedtoELISAandRT-qPCR,fluorescence imaging is well less labor intensive (at least for capturing images) and ischeaper(noneedforexpensiveextractionkitsorantibodies)(Bustin,2000;Sakamotoandal,2018).However,asshownbythefigures,theautofluorescenceofplanttissueswasaseriousdrawbacktohavereliableresults.Indeed,autofluorescencemaybelinkedtothe
30
planttissues(stems,roots),tothepresenceofnecroticspotsortodamagescausedbytheinoculation procedure (Baulcombe and al, 1995). Here, the achievement of acomplementaryELISA test forvirusdetectionshowedthatonly the leavesand the rootswereeffectivelyinfectedin‘YoloWonder’.However,therootsofhealthyplantspresentedhighautofluorescenceresponse.Thishighlightedthatnotallplantorgansweregoodtargetsfor virus detection or quantification by fluorescence imaging; leaveswere definitely theorgantofavor.Indeed,theautofluorescenceisaknownasarecurrentlimitinfluorescenceimagingandgenerallyencouragesscientiststodesignspecificstrategiestoovercomeautofluorescenceoflivingtissues(Kodama,2016).Likewise,severeinterferenceofchlorophyllwith GFP are known to disrupt the proportional relationship between GFP titer andfluorescenceleadingtosubstantialerrorsinestimatingthetargetconcentration(Zhouandal, 2005). In our experiments, these issueswere limited in estimating the proportion offluorescent leaf surface by adding an auto fluorescence threshold. Nevertheless, someuncertainty may lie in the fluorescence measurements because the calculation of theproportion of fluorescent leaf surface is provided by manually cutting-our surfaces onimages (which is, in addition, labor intensive). In order to evaluate the reliability of thismeasurement,wetested thecorrelationsbetweentheestimatedtotal leafareaandtheactual weight of each leaf. The obtained Pearson’s correlation coefficients were alwaysabove 0.76, which suggested that ourmeasurementswere accurate. Further works isneededtoautomateimageanalysisusinganimageprocessingsoftwarelikeImageJtospeeduptheanalysisandimprovetheprecisionoftheresults.Finally,astheproportionoffluorescentleafproportionwasnotimpactedbyleafsizeandallowed comparisons of phenotypically different plant cultivars, this method may beextendedtothephenotypingofvirusaccumulationindifferentgenotypes.
Asdemonstratedabove,theproportionoffluorescent leafsurfacecouldthusbeusedtoassessresistancestargetingviralaccumulation.Inthiscontext,C.annuumcv.‘Yolowonder’and ‘Perennial’ were chosen for their contrasted resistance profile to PVY and wereinoculated via aphids with the isolate PVYSON41p-115K-GFP. Unsurprisingly, ‘Yolo Wonder’presentsahigherlevelofvirusaccumulation(k=77%)andafastervirusaccumulation(µ=11.5days)than‘Perennial’(k=9.8%andµ=16.31days).Consequently,allthenon-linearlogisticmodelestimatorshighlight thehigher susceptibilityof thecultivar ‘YoloWonder’than‘Perennial’.Biologically,itseemedthatPVYaccumulatedmorein‘YoloWonder’thanin‘Perennial’whichareknowntopresentrespectivelyasusceptibilityallele(pvr2+)andaresistance allele (pvr23) combined with three QTLs. These observations confirmed thecharacteristicsofthesecultivars,andinthecaseof‘Perennial’theyallowedtoassessthelevelofresistanceonPVYaccumulation(Montarryandal,2012;Quenouilleandal,2013).Here the low accumulation PVYSON41p-115K-GFP in ‘Perennial’ shows that the quantitativeresistancetoPVYoccursatanearlystepoftheviralinfectionprocess(Mouryandal,2004).
Comparingtheresultsoftwodifferentvirusinoculationprocedures(mechanicalandaphidmediated) obtained with two pepper cultivars, it was very surprising to measure viralaccumulation in ‘Perennial’ following an aphid-mediated inoculation even though amechanical inoculation never succeeded in efficiently infecting this cultivar. To ourknowledge, ‘Perennial’ is described to be completely resistant to the isolate PVYSON41p.Nevertheless,amutation in theVPg (at115thaminoacid) restores infectivityandbreaksdownthemajorresistancegene(pvr23).Thus,thefailureofvirusinfectionaftermechanicalinoculation wasn’t expected, especially because potyviruses are notorious to be easilytransmittedviamechanicalinoculations(Kerlan,2006,Quenouilleandal,2013).Ourcurrenthypothesisreliesontheanatomicdifferencesbetweenthetwopeppercultivars. Indeed,leavesofC.annuumcv.’Perennial’areharderandtwicesmallerthantheseofC.annuumcv.’Yolo Wonder’. Moreover, on other pathosystems, authors reported several factorsaffectingmechanicaltransmissionofplantvirusessuchasthepresenceofahighwaxlayerontheleaveswhichcouldbeovercomebytheuseofNa2SO3andmercaptoethanolintheinoculationbuffer(Mandalandal,2001).Theyalsorecommendedtheuseofmoreabrasivepowders likediatomaceousearth toensuremore injurieson the leaves (Mandal andal,2001).TheseideascouldbeconfirmedbyfurtheradjustmentsoftheinoculationprocedureofPVYon‘Perennial”orbyscanningelectronmicroscope.Inanycase,theevidentsuccessofaphidmediated inoculationconfirmedthataphidmouthpartsandfeedingprocessareremarkablyadaptedtovirusinoculation(Braultandal,2010).
ThelackofreproducibilityoftheELISAmethodmaybeexplainedbynumerousfactorsthatcouldinterfereatdifferentstepsofthetestandhighlydependsonthestudiedpathogen(Cardinandal,1984).However,asatthe“PathologieVégétale”researchunit,manipulatorsperformELISAonadailybasis,theyreporteddivergencesinachievinganentireELISA,likethe lightconditionsforsubstrate incubationandthewashingmethodforremovingplantextracts. Here, the assessment of possible factors influencing OD values in ELISA wasinvestigatedonPVY.First,theeffectofthemethodtoremoveplantextractsfromtheplateswasnotsignificantandnocontaminationleadingtofalsepositiveresultswasobserved.ThismeantthatthesensibilityoftheELISAtestsforPVYdiagnosticandquantificationonpepperwas not compromised by the washing method. In addition, the light effect showed nosignificanteffectonthenormalizedODvaluesmeasuredonplatesincubatedinthelightorin thedarkeven though it is incontradictionwith thesupplier recommendations (Esser,2014).Withregardtothewashingmethodandlightconditionforsubstrateincubation,the
32
methods requiring the least effort (sinkwashing and substrate incubation in light)wereretainedforallsubsequentsELISAmeasurements.According toCardinandal (1984), theODvariationsamong thewellsofaplatearenotrandomlyassigned.ThisobservationwassupportedbythedeterminationofasignificantdifferencebetweenthemeanofthenormalizedODvaluesfromtheperipheralwellsandthe center wells of the plates loaded with C. annuum cv. ‘Yolo Wonder’ leaf extracts.However,nosignificantdifferencewasdetectedbetweenthemeanofthenormalizedODvaluesfromtheperipheralandthecenterwellsoftheplatesloadedwithN.tabacumcv.‘Xanthi’. As these tests usingN. tabacum cv. ‘Xanthi’ leaf extractswere performed firstduringmyinternship,itledtotheuseofthe96wellsforallsemi-quantitativeELISAtests.Thisavoidedthelossofthe36peripheralwellsandlimitedthereforethenumberofrequiredplates which matched the SME (Environment Management System) certification of theresearchunit and its globalprocedureofwaste reduction. Indeed, theuseof thewholemicrotiterplatesreducedbyhalf thenumberofplatesrequiredforthewhole internship(56%ofreduction).Inaddition,asshownbythefiguresthegeneraldistributionsoftheODvaluesvariedsharplybetweentheplateswhichmademoredifficulttheestablishmentofafinalconclusion.Indeed,thedeterminationofasignificantdifferenceinthenormalizedODmeansbetweenthecenterandtheperipheralwellswasn’talwayslinkedtoaU-orn-shapetendencyofthemeanODvaluesalonglinesandcolumns.Nevertheless,wekeptanalyzingthe influenceof thewellpositionuntil theendofmy internshipand finally identifiedan“Edge” effect within microtiter plates. This justifies the traditional avoidance of theperipheralwells.According to severalauthors, thevariationsbetween theODmeasuredfromtheperipheraltothecenterwellscanbeattributedtodifferencesinthermicisolationandplasticcharacteristicswithintheplates,leadingtodifferencesinenzymeactivityandinantigenattachment,respectively(Burtandal,1979;Oliverandal,1981;Shekarchiandal,1984).
33
Conclusion
Breedingforresistancewithconventionalapproachesremainsalongandexpensiveprocess.Therefore, in order tobe cost-efficient and tomaximize the successprobability of plantresistancesuse,thepriorassessmentofplantresistancesisessential.Andyet,despitetherecentdevelopmentandspreadofeasyandreliableserologicalandmoleculartechniquesfor virusdetectionandquantification, thephenotypingofplant resistancesamongplantaccessionsbankusinglaboratorymethodsisstilllaborintensive.Inordertofacethesemainissues,thisreportpresentedthefirstevaluationofanon-destructivemethodforquantifyingvirus accumulation, based on fluorescence imaging. Measures of viral accumulation inpepper (C.annuumcv. ‘Yolowonder’)and tobacco (N.benthamianaandN. tabacum cv.’Xanthi’)plantsinoculatedbyamodifiedPVYisolate(PVYSON41p-115K-GFP)weresignificantlyandpositivelycorrelatedwhenobtainedbyELISA(toquantifycoatproteins)andfluorescenceimaging(toestimatetheproportionoffluorescentleafsurface).Hence,theestimationofthe viral load by fluorescence imaging allowed to observe the phenotypic expression ofresistancetoPVYaccumulationinaresistantcultivar(‘Perennial’)comparedtosusceptibleone (‘YoloWonder). Thus, if the relationshipbetweenvirusRNAconcentrationsand theproportionoffluorescentleafsurfaceleadtothesameobservations,itwillcorroboratethatthequantificationofGFPappearsasapromisingmacroscopicandnon-destructivemethodwhichmaybeusedforfurtherphenotypingpepperresistancestoPVYaccumulation.Indeed,thirteenpepperdoublehaploidlinesissuedfromtheF1hybrid(PerennialxYoloWonder)wentthroughthesameprotocolas‘Perennial’and‘YoloWonder’plantsinordertoassessPVYson41p-115K-GFPvirusaccumulation.Accordingtotheirgenotypescharacteristics,theirresponse to virus accumulation may provide different resistance profiles. Finally, thequantificationofvirusremainsareliablemethodforphenotypingplantresistancesagainstvirusaccumulation.However,thecharacterizationofplantresistancesonotherviralcyclesteps (inoculation, systemic colonization, acquisition) would require the design of otherspecificmethodologies.
Boissot,N., S. Thomas,N. Sauvion,C.Marchal,C.Pavis, etC.Dogimont.«MappingandvalidationofQTLsforresistancetoaphidsandwhitefliesinmelon».TheoriticalandAppliedGenetics121(2010):9-20.
Boonham, N., J. Kreuze, S. Winter, R. Van der Vlugt, J. Bergervoet, J. Tomlinson, et R.Mumford.«MethodsinvirusdiagnosticsfromELISAtonextgenerationsequencing».VirusResearch,no65(2014):473-503.
Boote, K.J., J.W. Jones, J.W. Mishoe, et R.D. Berger. « Coupling pests to crop growthsimulatorstopredictyieldreductions».Phytopathology,no73(1983):1581-87.
Charron,C.,M.Nicolaï,J.L.Gallois,C.Robaglia,B.Moury,etC.Caranta.«Naturalvariationand functionnal analyses provide evidence for co-evolution between plant eIF4E andpotyviralVPg:Co-evolutionbetweeneIF4EandVPg».ThePlant Journal54,no1 (2008):56-68.
Díaz-Pendón, J.A., R. Fernández-Muñoz, M.L. Gómez-Guillamón, et E. Moriones.«InheritanceofresistancetoWatermelonmosaicvirusincucumismelothatimpairsvirusaccumulation, symptom expression and aphid transmission ». Phytopathology 95, no 7(2005):840-46.
Dogimont, C., A. Palloix, A.M. Daubèze, G. Marchoux, K. Gebre Selassie, et E. Pochard.«Geneticanalysisofbroadspectrumresistancetopotyvirusesusingdiubledhaploidlinesofpepper(CapsicumannuumL.)».Euphytica88(1996):231-39.
Jones, R.A.C. « Control of Plant VirusDiseases ». Advances in Virus Research 67 (2006):205-44.
Kaweesi, T., R. Kawuki, V. Kyaligonza, Y. Baguma, G. Tusiime, etM.E. Ferguson. « Fieldevaluationofselectedcassavagenotypesforcassabrownstreakdiseasebasedonsymptomexpressionandvirusload».Virologyjournal11,no216(2014):1-14.
Kodama, Y. « Time gating of chloroplast autofluorescence allows clearer fluorescenceimaginginplanta».PlosOne11,no3(2016):1-8.
Kim, S.,W. Kang, H.N. Huy, S. Yeom, J. An, S. Kim,M. Kang, H.J. Kim, Y.D. Jo, et Y. Ha.« Divergent evolution of multiple virus-resistance genes from a progenitor in Capsicumspp.»NewPhytologists213(2017):886-99.
Oliver, D.G., A.H. Sanders, D. Hogg, et J. Woods Hellman. « Thermal gradients inmicrotitrationplates.Effectsonenzyme-linkedimmunoassay».Journalof ImmunologicalMethods,no42(1981):195-201.
Palloix,A.,V.Ayme,etB.Moury.«Durabilityofplantmajorresistancegenestopathogensdependson theplant genetic background, experimental evidence and consequences forbreedingstrategies».NewPhytologists183(2009):190-99.
Penella, C., et A. Calatayud. « Pepper crop under climate change: grafting as anenvironmental friendly strategy ». In Climate Resilient Agriculture - Strategies andPerspectives,2018.
Perring, T.M., N.M. Gruenhagen, et C.A. Farrar. « Management of plant viral diseasesthrough chemical control of insect vectors ». Annual Review of Entomology 44 (1999):457-81.
Quenouille, J., L. Saint-Felix, B.Moury, et A. Palloix. « Diversity of genetic backgroundsmodulatingthedurabilityofamajorresistancegene;AnalysisofacorecollectionofpepperlandracesresistanttoPotatovirusY».MolecularPlantPathology2,no17(2016):296-302.
Remans, T., P.M. Schenk, J.M. Manners, C.P.L. Grof, et A.R. Elliott. « A protocol forfluorometric quantification of mGFP5-ER and sGFP (S65T) in transgenic plants ». PlantMolecularBiologyReporter17(1999):385-95.
Rimbaud, L., S.Dallot,A.Delaunay,S.Borron,S. Soubeyrand,G.Thébaud,etE. Jacquot.« Assessing the Mismatch Between Incubation and Latent Periods for Vector-BorneDiseases :ThecaseofSharka».Phytopathology105,no11(2015):1408-16.
Ruffel, S., M. Dussault, M. Lesage, A. Moretti, A. Palloix, M. Daunay, B. Moury, A.Bendahmane, C. Robaglia, et C. Caranta. « Involvement of the eukaryotic translationinitiation factor eIF4E in Solanaceae-Potyvirus interactions. » European Association forResearchonPlantBreeding,2004,167-70.
Tajadini,M.,M.Panjehpour,etS.H.Javanmard.«ComparisonofSYBRGreenandTaqManmethods in quantitative real-time polymerase cain reaction analysis of four adenosinereceptorsubtypes».AdvancedBiomedicalResearch85,no3(2014).
Figure S1:OD values in function of the dilution factor for the 18 samples of a plate (16thsample:referencesample).Dilutionfactors,1:1:5,2:1:25,3:1:125,4:1:625,5:1:3125)
Figure S5: Absorption in function of wavelength of 30 RNA samples extracted with twoextractionmethods:(A)RNeasy®PlantMiniKit(Qiagen,Germany)and(B)TRI-Reageant®-chloroform (MRC, USA). All the absorptions were measured via Nanodrop®ND-1000Spectrophotometer(Thermo-Scientific,USA)RscriptexampleforassessingthewashingeffectonnormalizedODvalues