ebook img

Laird, J., McInally, C., Carr, C., Doddiah, S., Yates, G., Chrysanthou, E., Khattab, A., Love, AJ, Geri PDF

14 Pages·2013·1.67 MB·English
by  
Save to my drive
Quick download
Download
Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.

Preview Laird, J., McInally, C., Carr, C., Doddiah, S., Yates, G., Chrysanthou, E., Khattab, A., Love, AJ, Geri

Laird, J., McInally, C., Carr, C., Doddiah, S., Yates, G., Chrysanthou, E., Khattab, A., Love, A.J., Geri, C., Sadanandom, A., Smith, B.O., Kobayashi, K., and Milner, J.J. (2013) Identification of the domains of cauliflower mosaic virus protein P6 responsible for suppression of RNA-silencing and salicylic acid-signalling. Journal of General Virology, 49 (12). pp. 2777- 2789. ISSN 0022-1317 Copyright © 2013 SGM. http://eprints.gla.ac.uk/86415/ Deposited on: 19 Nov 2013 Enlighten – Research publications by members of the University of Glasgow http://eprints.gla.ac.uk JournalofGeneralVirology(2013),94,2777–2789 DOI10.1099/vir.0.057729-0 Identification of the domains of cauliflower mosaic virus protein P6 responsible for suppression of RNA silencing and salicylic acid signalling Janet Laird,1 Carol McInally,1 Craig Carr,1 Sowjanya Doddiah,1 Gary Yates,1 Elina Chrysanthou,1 Ahmed Khattab,1 Andrew J. Love,13 Chiara Geri,1,2 Ari Sadanandom,3 Brian O. Smith,4 Kappei Kobayashi5 and Joel J. Milner1 Correspondence 1PlantScience Research Theme,School of LifeSciences andInstitute ofMolecular Cellular and JoelJ.Milner Systems Biology, College ofMedical, Veterinary & Life Sciences,University of Glasgow, [email protected] Glasgow G128QQ, UK 2Istituto diBiologiaeBiotechnologia Agraria, Consiglio NazionaleDelle Richerche, Pisa,Italy 3School of Biological andBiomedical Sciences, DurhamUniversity, DurhamDH13LE, UK 4InstituteofMolecularCellularandSystemsBiology,CollegeofMedical,Veterinary&LifeSciences, University of Glasgow,Glasgow G128QQ, UK 5PlantMolecular Biology andVirology, Faculty ofAgriculture, Ehime University, Ehime 790-8566, Japan Cauliflowermosaicvirus(CaMV)encodesa520aapolypeptide,P6,whichparticipatesinseveral essentialactivities inthevirus life cycle includingsuppressing RNAsilencing andsalicylic acid- responsivedefencesignalling.WeinfectedArabidopsiswithCaMVmutantscontainingshortin- framedeletions within the P6ORF. A deletion inthe distalend ofdomainD-I (the N-terminal 112aa) of P6did not affect virusreplication butcompromised symptom development and curtailedtheabilitytorestoreGFPfluorescenceinaGFP-silencedtransgenicArabidopsisline.A deletion intheminimum transactivator domainwasdefective invirusreplicationbutretained the capacity tosuppress RNA silencinglocally. Symptom expression inCaMV-infected plants is apparently linked tothe ability tosuppress RNAsilencing. When transiently co-expressed with tomatobushystuntvirusP19,anelicitorofprogrammedcelldeathinNicotianatabacum,WTP6 suppressed the hypersensitive response,butthree mutants,two withdeletions withinthe distal endofdomainD-IandoneinvolvingtheN-terminalnuclearexportsignal(NES),wereunabletodo so.Deleting the N-terminal 20aaalsoabolished thesuppression ofpathogen-associated molecular pattern-dependent PR1aexpression following agroinfiltration. However,thetwo other deletions indomain D-Iretained this activity, evidence thatthemechanisms underlying these functionsarenotidentical.TheD-IdomainofP6whenexpressedalonefailedtosuppresseither celldeathorPR1aexpressionandistherefore necessarybutnotsufficient forallthreedefence Received 30July2013 suppressionactivities.Consequently,concernsaboutthebiosafetyofgeneticallymodifiedcrops Accepted 19September2013 carryingtruncated ORFVI sequences appearunfounded. INTRODUCTION contains six known genera of which the most extensively studied member is cauliflower mosaic virus (CaMV), the Members of the family Caulimoviridae of pararetroviruses type member of the genus Caulimovirus. CaMV has a infect plants and replicate by reverse transcription of a genome of ~8 kb comprising six major ORFs (I–VI). Five circular dsDNA genome (Haas et al., 2002). The family of the six major virus proteins are translated sequentially from a single polycistronic RNA, the 35S RNA (Ryabova 3Presentaddress:Cell&MolecularSciences,JamesHuttonResearch et al., 2002, 2004). This unusual translational strategy Institute,Invergowrie,DundeeDD25DA,UK. is found in members of only two genera of viruses, Caulimovirus and the closely related Soymovirus (Ryabova Twosupplementaryfiguresandonetableareavailablewiththeonline versionofthispaper. et al., 2002, 2006). 057729G2013SGM PrintedinGreatBritain 2777 J.Lairdandothers P6, a 62 kDa polypeptide encoded by CaMV ORFVI, was (a) initially identified as the major component of cytoplasmic a inclusion bodies, which constitute the sites of virus b assembly (Haas et al., 2002). P6, which is translated from RH bip its own monocistronic mRNA, plays an essential role in c cc Path miniTAV RB-a RB-b different aspects of virus replication. It functions in 30 112 242 310 400 520 infected cells to facilitate translation of the downstream ORFs on the 35S RNA (Bonneville et al., 1989; Zijlstra & 1a 1b Hohn, 1992) via interaction with components of the D-I D-II D-III D-IV translational machinery (Bureau et al., 2004; Leh et al., 2000;Parketal.,2001;Ryabovaetal.,2004),amechanism (b) known as translation transactivation (TAV). P6 prevents ribosome detachment at the stop codon, enabling poly- TAVD2(Δ40–76) peptide synthesis to reinitiate at the next start codon (Ryabova et al., 2004). TAVD3(Δ80–110) At least four more roles for P6 have been identified. P6 TAVD6(Δ166–201) interacts with at least two of the other CaMV proteins involved in aphid transmission, P2 and P3 (Lutz et al., P6: Δ3–20 2012). It forms cytoplasmic inclusion bodies of various P6: T1–112 sizes; the smaller ones associate with microtubules and the endoplasmic reticulum and move dynamically along actin P6: T1–200 filaments(Harriesetal.,2009).Thismovementisprobably essential for intracellular virus trafficking and involves the P6: T111–520 interaction of P6 with the CaMV movement protein P1 P6: T183–520 (Hapiak et al., 2008) and CHUP1, which mediates association between chloroplasts and the cytoskeleton. P6: T1–112: Δ3–20 These findings suggest that P6 subverts the mechanism responsible for chloroplast movement for intracellular Fig.1.MapoftheP6domainsandmutantsusedinthisstudy.(a) trafficking of CaMV (Angel et al., 2013). SchematicrepresentationofP6domains:aminoacidnumbersat P6 is the major genetic determinant of virus pathogenicity theboundariesofknowndomainsareindicated.Openboxesshow (Baughman et al., 1988; Kobayashi & Hohn, 2004; Schoelz the coiled-coil (cc) a-helix, pathogenicity/host-range/avirulence etal.,1986;Stratford&Covey,1989)andexpressionfroma (Path),minimumtransactivator(miniTAV),RNaseH(RH)andRNA transgeneresultsinasymptom-likephenotype(Baughman binding(RB-aandRB-b).Thebipartitenuclearlocalizationsignals et al., 1988; Cecchini et al., 1997; Zijlstra et al., 1996). P6 (NLS; bip) and three non-conventional NLS (a, b and c) are exhibitsvirus-encodedsuppressorof RNA silencing (VSR) indicated by diamonds above and cross-hatching. The self- association domains D-I to D-IV and subdomains 1a and 1b are activity (Haas et al., 2008; Love et al., 2007), probably indicatedbysolidlines.DatafromHaasetal.(2005),Haasetal. through its interaction with the dsRNA-binding protein (2008), Kobayashi & Hohn (2004) and Hapiak et al. (2008). (b) DRB4 (Haas et al., 2008), a component of the Dicer4 Deletions in P6 coding sequences in CaMV-TAV mutants and in complex. thecorrespondingP6expressionconstructs.Filledboxesindicate Finally, ectopic expression of P6 in Arabidopsis and Nico- sequencefromCaMVCM1841,shadedboxesindicatesequence tiana benthamiana profoundly affects signalling mediated derived from CaMV Cabb B-JI and open boxes indicate internal by salicylic acid (SA), jasmonic acid, ethylene and auxin deletions. (Geri et al., 2004; Love et al., 2012; Smith, 2007). The ability of P6 to manipulate multiple components of the host defence suggests its central role as a pathogenicity predicted nuclear localization signals (NLSs) (Cerritelli determinant during virus infection, and the pleiotropic et al., 1998; De Tapia et al., 1993; Haas et al., 2008; phenotypesthatresultfromtransgene-mediatedexpression Kobayashi & Hohn, 2003; Ryabova et al., 2004). De Tapia inplantaderivefromitsactivityasapathogenicityeffector. et al. (1993) identified aa 111–242 (domain D-II) as How does a single protein achieve such a diverse range of containing the minimum functional domain (miniTAV activities? Outwith closely related members of the family domain) able to facilitate TAV. The miniTAV domain Caulimoviridae, P6 has no obvious homologues and its overlaps an RNase H domain (Cerritelli et al., 1998) and three-dimensional structure is unknown. Li & Leisner contains the interaction motif for RL18; those for RL24, (2002)definedfourdomainsbasedonself-association,and eIF4G and eIF2B are located within domain D-III sequenceanalysishasrevealedseveralstructuralmotifsand (Ryabova et al., 2004). Domains D-II and D-IV are functional domains (Fig. 1a). These include RNA binding, involved in the interaction with CHUP1 (and presumably RNase H, a short N-terminal helical domain and several therefore intracellular trafficking (Angel et al., 2013). P6 is 2778 JournalofGeneralVirology94 DomainanalysisofCaMVproteinP6 a nucleocytoplasmic shuttle protein, and both nuclear 2a). Plants inoculated with CaMV-TAVD2 and CaMV- localization and export functions are essential for infectiv- TAVD6 did not develop any symptoms (Fig. 2a). With ity. Mutation of a nuclear export signal (NES) at the CaMV-TAVD3 inoculation, plants were usually asympto- N terminus abolishes infectivity and mutation of the matic,althoughby28daysp.i.theoccasionalleafexhibited predicted NLS within the C-terminal domains abolishes subtleveinclearing.Wemeasuredvirustitresat28daysp.i. both VSR activity and infectivity (Haas et al., 2008; usingELISA(Fig.2c).Col-0plantsinoculatedwithCaMV- Kobayashi & Hohn, 2004). CW containedhightitres ofvirus,but CaMV-TAVD2 and CaMV-TAVD6werenotdetectablebyELISA.Surprisingly, AlthoughTAVactivityand virus-trafficking functionshave despitethelackofsymptoms,titresofCaMV-TAVD3were been mapped, the domain(s) responsible for VSR activity consistently verysimilartotitresof CaMV-CW.Thus,this and SA signalling suppression remain(s) to be identified. deletion did not appear to significantly reduce virus Domain D-I plays a major role in pathogenicity and host accumulation, at least under the conditions of our ex- range and acts as an avirulence domain in Arabidopsis and periment,butprofoundlyaffectedsymptomdevelopment. Solanaceous hosts (Agama et al., 2002; Baughman et al., 1988; Palanichelvam et al., 2000; Palanichelvam & Schoelz, We next tested whether we could complement the 2002;Schoelzetal.,1986;Stratford&Covey,1989).D-Ihas mutations in ORFVI by providing P6 from the transgenic beendividedintosubdomain1a,comprisingtheN-terminal Arabidopsis line A7, which expresses P6 at levels similar to 30 aa containing the NES (Haas et al., 2008; Haas et al., those in infected plants (Cecchini et al., 1997) (Fig. 2b). 2005),andsubdomain1b(aa31–110)containingavirulence Plantsstartedtoexhibitsubtlesymptomsofleafdistortion andpathogenicityfunctions.Mutantswithdeletionswithin ataround14daysp.i.,andby28daysp.i.,stuntingandleaf 1b retain replication competence but exhibit delayed virus distortion were visible on all A7 plants inoculated with spreadinturnip(Kobayashi&Hohn,2004). CaMV-CW. Plants inoculated with CaMV-TAVD6 also all WecarriedoutinfectionstudiesusingCaMVmutantswith developed symptoms similar to CaMV-CW and at around deletionsinP6andfoundthatatleastonemutationwithin the same time, but those inoculated with CaMV-TAVD2 subdomain 1b abolished both VSR activity and symptom and CaMV-TAVD3 did not. Titres of CaMV-CW were development without significantly reducing systemic virus approximately 30% of those in Col-0 plants, consistent titre. We transiently expressed WT and mutant P6 in with our previous reports of reduced titres in P6 NicotianabenthamianaandNicotianatabacumandassayed transgenics (Love et al., 2007, 2012). All three mutants the ability to suppress expression of an SA-responsive also accumulated to significant titres (Fig. 2c). These marker gene, PR1a, and cell death in response to a gene- results suggested that functional P6 provided from a for-gene elicitor. Deletions in subdomains 1a and 1b transgenecanactintranstofacilitatethereplicationofthe abolished VSR activity and also abolished the suppression mutants. ORFVI sequences provided from a transgene can ofthecell-deathresponseseenwithWTP6.However,only undersomecircumstancesrecombinewithdefectiveCaMV the deletion in subdomain 1a eliminated suppression of genomeswheninfectionproceedsoveranextendedperiod PR1a expression. Domain D-I evidently plays an essential (Kira´ly et al., 1998). Although we cannot absolutely rule roleinseveralpathogenicityeffectoractivities.Suppression out the possibility of recombination in our complementa- ofRNAsilencingandcelldeathmaybefunctionallylinked, tionexperiments,webelievethatitisunlikelybecausevirus but suppression of SA-responsive gene expression must titres for the WT and mutants were similar at 28 days p.i. involve an at least partially independent mechanism. and, in the case of CaMV-TAVD6, symptoms started to appear at a similar relatively early stage in infection. The absence of symptoms in A7 plants infected with CaMV- RESULTS TAVD2andCaMV-TAVD3suggestedthat,evenwhenWT P6 is provided from atransgene, symptom development is InfectivityofWTandP6deletionmutantsofCaMV blocked in the presence of virus-encoded P6 containing in Arabidopsis deletions in subdomain 1b. We did not observe this with CaMV-TAVD6 in which the deletion affects the miniTAV CaMV mutants with in-frame deletions in subdomain 1b domain. of P6 are replication competent in turnip but show delay- ed long-distance spread (Kobayashi & Hohn, 2004). We inoculated WT Arabidopsis (ecotype Col-0) with WT virus VSR activity of CaMV deletion mutants (CaMV-CW) and three mutants (Fig. 1b). CaMV-TAVD2 The transgenic Arabidopsis line GxA contains a 35S–GFP and CaMV-TAVD3 carry deletions in subdomain 1b, transgene whose expression is silenced by a second whilst CaMV-TAVD6 has a deletion in the miniTAV transgene, a potato virus X amplicon containing part of domain and cannot replicate in turnip. the GFP-coding sequence (Dalmay et al., 2000; Schwach Agroinoculation with CaMV-CW was remarkably effi- et al., 2005). CaMV infection of GxA suppresses silencing cient,withsymptomsappearingat~13dayspost-infection of the GFP transgene, restoring strong fluorescence to (p.i.) and essentially 100% of plants developing obvious infected tissue (Love et al., 2007). We used this assay to stunting, leaf distortion and mosaics by 28 days p.i. (Fig. comparetheVSRactivitiesofWTandmutantvirus.Virus http://vir.sgmjournals.org 2779 J.Lairdandothers (a) (i) Uninoculated (ii) CaMV-CW (iii) CaMV-TAVD2 (iv) CaMV-TAVD3 (v) CaMV-TAVD6 (b) (i) Uninoculated (ii) CaMV-CW (iii) CaMV-TAVD2 (iv) CaMV-TAVD3 (v) CaMV-TAVD6 (c) 1.2 1 U 0.8 A Col 0 e s titr 0.6 A7 u GxA Vir 0.4 0.2 0 CW TAVD2 TAVD3 TAVD6 Virus genotype Fig.2.InfectivityofCaMV(WTandmutants)onWTandP6transgenicArabidopsis.(a)SymptomsonCol-0at28daysp.i.:(i) uninfected,(ii)CaMV-CW,(iii–v)CaMVTAVmutantsasindicated.Bar,2cm.(b)SymptomsonP6transgenicplants(lineA7) at 28 days p.i.: (i) uninfected, (ii) CaMV-CW, (iii–v) CaMV TAV mutants as indicated. Bar, 2cm (note difference in scale betweenaandb).(c)Virustitresat28daysp.i.inCol-0,A7andGxAplantsdeterminedbyELISA.Barsshowsmeantitres (±SD)ofthreetissuesampleseachcomprisingthreepooledplants.Titresinarbitraryunits(AU)arenormalizedtothemeanof Col-0plantsinfectedwithCaMV-CW. levels in GxA were similar to those in Col-0, with CaMV- contained no detectable titres of virus. With CaMV- CW and CaMV-TAVD3 accumulating to high titres but TAVD2,we did not observe any fluorescent cells, but with CaMV-TAVD2andCaMV-TAVD6undetectablebyELISA CaMV-TAVD6 we consistently observed fluorescence in (Fig. 2c). As with Col-0, only CaMV-CW induced groups of cells within every leaf we examined between 8 symptoms in GxA. and11daysp.i.(Fig.3b),albeitatlowerintensitythanwith CaMV-CW; by 14 days p.i. fluorescence had become We examined the upper leaves for GFP fluorescence using undetectable. Kobayashi & Hohn (2003) showed using confocal microscopy (Fig. 3a). Uninoculated controls sensitive PCR that CaMV-TAVD6 is unable to replicate in showed no detectable fluorescence, even at high gain, but single cells. However, agroinoculation could provide tissuefromCaMV-CW-infectedplantsconsistentlyshowed transient P6 expression through direct transcription of strong fluorescence. We did not observe any fluorescent ORFVI (from its own 19S promoter) of the replication- cells in systemic leaves of GxA inoculated with any of the incompetent CaMV-TAVD6 genome (Kobayashi & Hohn, threemutants,despitethehightitresofCaMV-TAVD3.To 2003,2004).Thisresultdemonstratesthatdeletingaa166– test for local silencing suppression (around the sites of 201 does not abolish VSR activity. inoculation), we examined inoculated leaves (Fig. 3b). With CaMV-CW, by 8–11 days p.i. we consistently Mutations in P6 affect the ability to suppress SA- observed groups of cells showing strong GFP fluorescence. responsive cell death Inoculatedleaves recoveredfrom themicroscope slide and assayed by ELISA all contained moderate to high titres of Expression of P6 from a transgene in Arabidopsis reduces virus (with some leaf-to-leaf variation). Titres of CaMV- and delays cell death following treatment with SA or TAVD3 in inoculated leaves were similar to those of inoculation withan avirulent pathogen (Loveet al., 2012). CaMV-CW but we did not observe any GFP fluorescence. To identify the domain(s) responsible for this activity, we LeavesinoculatedwithCaMV-TAVD2andCaMV-TAVD6 exploited the ability of tomato bushy stunt virus (TBSV) 2780 JournalofGeneralVirology94 DomainanalysisofCaMVproteinP6 (a) (i) Uninoculated (ii) CaMV-CW (iii) CaMV-TAVD2 (iv) CaMV-TAVD3 (v) CaMV-TAVD6 (b) (i) Uninoculated (ii) CaMV-CW (iii) CaMV-TAVD2 (iv) CaMV-TAVD3 (v) CaMV-TAVD6 Fig.3.GFPfluorescenceinleavesofGxAplantsinoculatedwithCaMVWTandmutants.(a)Confocalmicroscopeimagesof representative upperleaves ofplantsat28daysp.i.: (i)uninoculated, (ii)CaMV-CW, (iii–v)CaMV TAVmutants.Thepanels showlow-magnificationimagesofGFPfluorescence.Allpanelsweretakenatthesamemicroscopegainsettings.(b)Confocal microscopeimagesofrepresentativeinoculatedleavesofplantsat28daysp.i.:(i)uninfected,(ii)CaMV-CW,(iii–v)CaMVTAV mutants.Notethattheimagesin(b)aretakenatahighermagnificationthanthosein(a).Allpanelsweretakenwiththesame microscopegainsettings.Bars,100mm. P19 to elicit a gene-for-gene hypersensitive response (HR) suppresstheHRelicitedbyP19(Fig.4h,i).Neitherdidthe in N. tabacum in an SA-dependent manner (Angel & corresponding C-terminally truncated variants (P6:T111– Schoelz, 2013; Sansregret et al., 2013). Agrobacterium- 520 and P6:T183–520) (Fig. 4j, k). The ability to suppress mediated expression of P19 in N. tabacum gave a strong cell death thus broadly paralleled the ability to suppress HRthatwascompleteby36 hbutcouldbeextendedto3– RNA silencing, with deletions in both D-I subdomains, 5 days by reducing the Agrobacterium titre to one-quarter but not in the miniTAV domain, affecting this activity. the usual level. At the higher titre of P19, co-infiltration However,theD-Idomainexpressedalonewasnotsufficient with a hypervirulent strain of Agrobacterium containing to suppress cell death; therefore other regions of P6 must pGWB-P6BJIW or pGWB-P6CW delayed the onset of HR alsoberequiredforthisactivity. by approximately 24 h. At the reduced titre of P19, co- infiltrationwitheitherWTP6constructsubstantiallyhalted Mutations in P6 affect the ability to suppress theprogress of thisHR compared with co-infiltration with expression of an SA-responsive marker gene emptyvector(EV)(Fig.4a,b,f).NeitherWTnormutantP6 elicitedHRintheabsenceofP19. Agroinfiltration of N. benthamiana elicits pathogen- associated molecular pattern (PAMP)-responsive expres- We cloned the P6 coding sequences from the three CaMV sion of PR1a, a reliable marker of SA-responsive gene mutants into Ti expression vectors and tested their ability tosuppressHR(Fig.4).IncontrasttoP6:CW(wild-typeP6 expression (Volko et al., 1998); this response is strongly from CaMV CM1841), neither of the two mutants with suppressedbytransientexpressionofP6(Loveetal.,2012). deletions in subdomain 1b, P6:D2 and P6:D3, was able to P6:CW gave the expected reduction in PR1a transcripts to suppressthedevelopmentofHR,butP6:D6,withadeletion ~30% that with EV (Fig. 5a). We anticipated that P6 in the miniTAV domain, suppressed cell death with an mutants with deletions in subdomain 1b might also fail to efficiencysimilar toWT(Fig. 4c–e).TheN-terminalsubdo- suppress PR1a expression. However, P6:D2 and P6:D3 as main1acontainstheNES,andmutationsthatabolishnuclear well as P6:D6 all reduced PR1a transcripts to a broadly export also abolish VSR activity (Haas et al., 2008). We similar level to that of P6:CW (Fig. 5a). All three mutants thereforedeleted18ofthe20 aaattheNterminustoproduce evidently retained the ability to suppress SA-responsive what we predicted would be a functionally equivalent gene expression. In contrast, infiltration with P6:D3–20 construct,P6:D3–20(Fig.1b).Whentransientlyco-expressed resulted in levels of PR1a expression similar to EV (Fig. withP19,P6:D3–20wasunabletosuppressHR(Fig.4g). 5b). Therefore, sequences required for suppression of SA- responsive gene expression are present in subdomain 1a To testwhether the D-I domain was able to suppressP19- but apparently not 1b. induced HR in the absence of the C-terminal domains, we produced a further series of expression constructs (for We next investigated whether expressing the N-terminal details,seeFig.1b).Truncatedpolypeptidescomprisingthe domain alone was sufficient to suppress PR1a expression. N-terminal 112or200 aa(P6:T1–112,P6:T1–200)didnot P6:T1–112 and P6:T1–200 not only failed to reduce PR1a http://vir.sgmjournals.org 2781 J.Lairdandothers (a) EV (b) P6:CW (c) P6:D2 (d) P6:D3 (e) P6:D6 (f) P6:BJIW (g) P6:Δ3–20 (h) P6:T1–112 (i) P6:T1–200 (j) P6:T111–520 (k) P6:T183–520 Fig.4.SuppressionofTBSVP19-dependentcelldeathbyco-infiltrationwithWTandmutantvariantsofP6.Photographsof leaf patches 4days after co-agroinfiltration with a construct expressing P19 plus EV control (a) or WT or mutant P6 as indicated(b–k).Allimagesareshownatsimilarmagnification.Bar,1.0cm. transcript levels but also produced a consistent increase of distribution.With P6:D6-GFP,weobservedveryfewsmall more than twofold over and above EV controls (Fig. 5b). inclusion bodies, although the large ones were still The D-I domain is therefore not sufficient for suppression abundant (Fig. 6a, v). Domain D-II, which contains the of SA-dependent gene expression but may play some role deletion in P6:D6 (aa 166–201), has been identified as in this activity because expression on its own promoted interacting with CHUP1 in connection with intracellular elevated expression of PR1a. virus trafficking (Angel et al., 2013). The region may be required for cytoskeletal association and for the formation of small inclusion bodies. Intracellular localization of mutant P6 The deletion in P6:D3–20–GFP included three residues Totestwhetherlossofdefencesuppressionactivityinsome identifiedasessentialfornuclearexport(Haasetal.,2008), mutants might be attributable to mislocalization and to so we anticipated that it would show enhanced co- confirm appropriate expression of P6, we analysed its localization with DAPI. Unexpectedly, the nuclear local- intracellular localization after transiently expressing the ization was similar to that of WT (Fig. 6b, ii). Haas et al. mutant forms of P6 as C-terminal GFP fusions in N. (2005) expressed the N-terminal 110 aa alone and benthamiana (Fig. 6). GFP-tagged P6BJIW (wild-type P6 compared localization with the same polypeptide with from CaMV Cabb B-JI) suppressed PR1a expression with the N-terminal a-helix deleted. We produced equivalent an efficiency similar to a myc-tagged construct (see constructs (although with a C-terminal GFP tag), P6:T1– Methods), indicating that this activity was unaffected by 112–GFP and P6:T1–112:D3–20–GFP. Whereas P6:T1– the GFP tag (data not shown). Epidermal cells expressing 112–GFP showed no nuclear localization whatsoever, WT P6 from the two isolates CM1841 and Cabb B-JI P6:T1–112:D3–20–GFP co-localized strongly with DAPI showed identical patterns of intracellular fluorescence, (Fig. 6b, iii, iv). P6:T111–520–GFP, which lacks the entire reminiscentofthosereportedbyHarriesetal.(2009)(Fig. D-I domain (including the NES), also showed enhanced 6a, i and ii). Cells contained cytoplasmic inclusion bodies nuclear localization (Fig. 6, v). Our results are therefore that were highly variable in size. Large numbers of small consistent with deletion of the N-terminal 20 aa affecting inclusion bodies were present, some of which appeared to nuclear export. Haas et al. (2005) fused GFP to the N be associated with cytoplasmic strands. Large inclusion terminus;ouruseofaC-terminaltagmightaccountforthe bodies were often clustered around the nucleus, but we differences for full-length P6. observed only weak GFP fluorescence co-localizing with DAPI(Fig.6b,i),consistentwiththefindingsofHaasetal. (2005) who reported rapid nuclear export of P6. Sequence conservation of domain D-I across members of the Caulimoviridae Localization of P6:D2-GFP and P6:D3-GFP was indistin- guishable from that of the WT (Fig. 6a, ii–iv). These We used the programs Jalview (Waterhouse et al., 2009) deletions did not cause obvious changes in intracellular and JPred3 (Cole et al., 2008) to align the sequences of P6 2782 JournalofGeneralVirology94 DomainanalysisofCaMVproteinP6 (a) The N-terminal sequences of the soymoviruses are shorter 1.2 than those of the caulimoviruses and are rather diverse V (Fig.S1).Possibly,membersofthegenusSoymoviruslacka E 1.0 o functional D-I domain. e t ativ 0.8 el s r 0.6 DISCUSSION pt nscri 0.4 WehaveshownthattheN-terminaldomainofP6contains a sequencesessentialforitsactivitiesasasuppressorofRNA 1a tr 0.2 silencing and of SA-dependent defence responses. These R P appear to be distinct from its TAV and virus-trafficking 0.0 functions.Deletingthedistalendofsubdomain1b(aa 80– U EV CW D2 D3 D6 110) abolished VSR activity within the context of an infectiousvirusclone.Thesamedeletionalsoabolishedthe (b) 3.5 ability to suppress one aspect of SA-dependent signalling, cell death triggered by the elicitor TBSV P19, but not V 3.0 another, PAMP-driven PR1a expression. The mechanisms E o underlying these three activities may therefore overlap but ve t 2.5 are clearly not identical. However, the N-terminal sub- elati 2.0 domain1a,whichincludestheNES,isessentialforallthree. s r pt CaMV-TAVD6, with a deletion within the miniTAV cri 1.5 domain, consistently produced a transient silencing sup- s n a tra 1.0 prerteasisniosnVSiRnaicntoivciutyla.tAedsClaeaMveVs-,TeAvVidDen6ciesctohmatpltehtieslymunuatabnlet 1 PR 0.5 to replicate in protoplasts (Kobayashi & Hohn, 2003), we assume that P6 mRNA is transcribed directly from CaMV- 0.0 U EV BJIW Δ3–20 T1–112T1–200 TAVD6 genomes introduced by agroinoculation. Although we were unable to detect CaMV-TAVD2 accumulation in inoculated leaves using ELISA, we might have expected Fig. 5. Quantification of PR1a expression in N. benthamiana similar limited P6 expression by direct transcription of the leaves following transient expression of WT and mutant P6 by T-DNA following agroinoculation. If so, the failure of agroinfiltration. (a) PR-1a transcripts, determined by quantitative CaMV-TAVD2 to stimulate similar transient GFP expres- PCR, in N. benthamiana leaves harvested 48h after agroinfiltra- sionininoculatedleavessuggeststhatittoomaybedeficient tion. Samples were uninfiltrated leaves (U), and leaves infiltrated inVSRactivity. withAgrobacteriumcarryingthefollowingvectors:pGWB17(EV), P6:CW(CW),P6:D2(D2),P6:D3(D3)andP6:D6(D6).(b)PR- Suppression of RNA silencing by VSR is a major 1a transcripts, determined as above. Samples were uninfiltrated contributor to symptom induction (Burgya´n & Havelda, leaves (U), and leaves infiltrated with Agrobacterium carrying the 2011). Despite titres similar to WT virus, CaMV-TAVD3 followingvectorspGWB17(EV),P6BJIW(BJIW),P6:D3–20(D3– was essentially asymptomatic on Col-0 plants, suggesting 20),P6:T1–112(T1–112)andP6:T1–200(T1–200).Barsshow that the symptoms of CaMV infection (at least in means±SD (in arbitrary units) of three independent biological Arabidopsis) are probably linked to the VSR activity of samples each comprising three pooled infiltrated leaf sections. P6. Complementation by transgene-derived P6 allowed all ValueswerenormalizedtovaluesforEV. three mutants to replicate, but, whereas CaMV-CW and CaMV-TAVD6causedobviousstuntingandleafdistortion in A7, CaMV-TAVD3 and CaMV-TAVD2 were both from CaMV with ten other members of the genus Cauli- asymptomatic. As A7 produces high levels of WT P6 movirus and four members of the genus Soymoviruses. P6 (Cecchinietal.,1997),theabsenceofsymptomsinCaMV- from all ten caulimoviruses showed significant homology TAVD2- and CaMV-TAVD3-infected plants must be a with CaMV over almost the entire sequence, but none of dominant-negative effect, presumably linked to the loss of the soymoviruses showed significant similarity to the VSR activity, further evidence that CaMV-TAVD2 is also caulimovirus D-I domain. (Figs 7 and S1, available in defective in this respect. JGV Online). Within domain D-I, homology varied between different members of the genus Caulimovirus ArolefordomainD-Iindefence suppression isconsistent (Fig. 7), but there was notable sequence conservation with its identification as the major genetic determinant of between aa 67 and 88, in particular the GK(D/E)X(S/ virus pathogenicity, host range and avirulence (Kobayashi T)NPLXXXXLXK motif (aa 74–88) conserved in 10 of 11 & Hohn, 2003; Palanichelvam & Schoelz, 2002; Schoelz & sequences. Interestingly, this motif extends across the Shepherd, 1988; Stratford & Covey, 1989). Both subdo- junction between the TAVD2 and TAVD3 deletions. mainsareinvolved.Subdomain1aclearlyplaysanessential http://vir.sgmjournals.org 2783 J.Lairdandothers (a) (i) P6:BJIW (ii) P6:CW (iii) P6:TAVD2 (iv) P6:TAVD3 (v) P6:TAVD6 (b) (i) P6:BJIW DAPI GFP Merge (ii) P6:Δ3-20 DAPI GFP Merge (iii) P6:T1-112 DAPI GFP Merge (iv) P6:T1-112:Δ3-20 DAPI GFP Merge (v) P6:T111-520 DAPI GFP Merge Fig. 6. Intracellular localization of WT and mutant P6 tagged with GFP. Confocal microscope images of tissue from N. benthamiana leaves 3days after agroinfiltration with WT and mutant variants of P6 fused at the C terminus to GFP. GFP fluorescenceisgreenandDAPIfluorescence(nuclearstaining)isblue.(a)IntracellulardistributionofWTandTAVDmutantP6: (i)P6BJIWinasingleepidermalcell.Representativelargeinclusionbodiesareindicatedbyyellowarrowsandsmallinclusion bodiesbypinkarrows.(ii)P6CW.Yellowandpinkarrowsareasin(i),whilstbluearrowsindicatenuclei(DAPIstaining).(iii–v) P6:D2, P6:D3 and P6:D6. Bars, 100mm. (b) High-magnification images showing nuclear localization of WT and truncated formsofP6:(i)P6BJIW,(ii)P6:D3–20,(iii)P6:T1–112,(iv)P6:T1–112:D3–20and(v)P6:T111–520.Panelsfromlefttoright: DAPI,GFP,merge.Nuclearfluorescenceisindicatedbyarrows.Bars,20mm. 2784 JournalofGeneralVirology94 DomainanalysisofCaMVproteinP6 role in pathogenicity as deleting it eliminated both the Theeffectsofmutationsinsubdomain1bonVSRactivityand suppression of cell death and PAMP-responsive gene thesuppressionoftheHRelicitedbyP19implythatitmust expression. Correct localization of P6 may be required play a key role in these functions. The motif GK(D/E)X(S/ for these activities. Deletions within subdomain 1b T)NPLXXXXLXK, which spans the ends of the TAVD2 and abolished the ability to suppress cell death in our assay TAVD3 deletions, is very highly conserved across 10/11 but not the ability to suppress PAMP-triggered expression members of the genus Caulimovirus. Such a degree of of PR1a. The apparent discrepancy between the effects of sequence homology provides additional support for the deletions in subdomain 1b on these two different SA- importanceofthisregionofP6tomembersofthisgenus. dependentresponsesmaybeexplainedbytherecentreport The pleiotropic phenotype(s) of P6-transgenic Arabidopsis thatextremeresistancetoTBSVinN.tabacumiselicitedby (Geriet al., 2004; Love et al., 2012; Smith, 2007) would be a complex of P19 plus small interfering RNAs (siRNAs) most elegantly accounted for by a common underlying (Sansregret et al., 2013). Interaction with DRB4, a mechanism, perhaps all involving RNA silencing, rather component of the Dicer4 complex (Haas et al., 2008), than by diverse direct interactions with multiple signalling providesaprobablemechanism fortheVSRactivityofP6. intermediates. However, because deletion mutants of As Dicer4 is involved in generating siRNAs, both defence subdomain 1b suppressed PAMP-responsive PR1a expres- suppression activities of P6 (on RNA silencing and SA sion with a similar efficiencyto WT P6, VSR activitymust signalling) could potentially play a role in inhibiting the not be essential for this activity. HR elicited by P19. Transgene-mediated expression of VSRs elicits pleiotropic The truncated proteins P6:1–112 and P6:1–200 elicited effects on jasmonic acid and other phytohormone elevated levels of PR1a transcripts compared with EV controls. The C-terminal region of P6, which is absent responses (Endres et al., 2010; Lewsey et al., 2010; from these constructs, contains four predicted NLSs (Fig. Lozano-Dura´n et al., 2011; Yang et al., 2008), and CaMV 1a).NuclearlocalizationisrequiredforVSRactivity(Haas infection is accompanied by profound changes in et al., 2008), and our results are consistent with it also microRNA (miRNA) and trans-acting siRNA (tasiRNA) being essential for cell death and suppression of SA- populations(Blevinsetal.,2006;Moissiard&Voinnet,2006; dependent gene expression. We did not observe obvious Shivaprasad et al., 2008). The 59 leader sequence of the differences in nuclear localization between WT and CaMV 35S RNA is a target for all four Arabidopsis Dicer truncated proteins (Fig. 6). However, because the N- complexes, producing siRNAs that appear to target host terminalNESpromotesveryefficientre-exportofP6from transcripts(Blevinsetal.,2006;Moissiard&Voinnet,2006; the nucleus, even WT P6, which is actively imported into Shivaprasadetal.,2008),evidenceofacomplexinteraction the nucleus (Haas et al., 2005), gave only weak GFP mediated at least partially by RNA silencing. miRNAs and fluorescence within nuclei. tasiRNAs regulate signalling pathways involving auxin CaMV IQ3HM11 1 63 HLV IQ5J1S0 1 64 LLDAV IB2CXY6 1 88 CERV IP05401 1 47 DaMV-HollIA9QKR8 1 94 EVCV IB2D1N1 1 58 MMV IQ8JTA1 27 120 DaMV IA9UD04 1 96 FMV IQ67458 1 96 CmYLCV IQ7TBL3 1 116 SVBV IQ88443 1 87 Consensus MEE–+L+A–LRL+LQQ–E––EKIL+LELK++++KI–––––––––––SLYEES+SSQ+++VSP+REK–QTE+SPLQTADGK++–+NPLKPDAL+KSITE+T–+S+L+D+S–KLVG+TK–S CaMV IQ3HM1164 140 HLV IQ5J1S0 65 140 LLDAV IB2CXY6 89 169 CERV IP05401 48 134 DaMV-HollIA9QKR8 95 144 EVCV IB2D1N1 59 134 MMV IQ8JTA1121 176 DaMV IA9UD04 97 150 FMV IQ67458 97 154 CmYLCV IQ7TBL3117 184 SVBV IQ88443 88 146 Consensus SD–––++SSPVQT+SGKDSSNPLMADSLPKS++–VQT–SRLVK––––––––––IPEDFSLRPN––––––––––––––––D–––––QGIPIP–KSE–HSSVA–PSY+E++IQ P+SG+KY Fig. 7. Alignment of sequences of the N-terminal amino acids of P6 from caulimoviruses. The sequences from CaMV, horseradish latent virus (HLV), lamium leaf distortion associated virus (LLDAV), carnation etched ring virus (CERV), dahlia mosaic virus-Holland (DaMV-Holl), eupatorium vein clearing virus (EVCV), mirabilis mosaic virus (MMV), dahlia mosaic virus (DaMV),figwortmosaicvirus(FMV),cestrumyellowleafcurlingvirus(CmYLCV)andstrawberryveinbandingvirus(SVBV)that precedetheRNaseHdomain(aa140inCaMV)arealigned,withtheconsensussequenceshownbelowinlogoform.Residues arecolouredaccordingtotheCLUSTAL_XcolouringschemeandtheUniprotaccessionnumbersindicated. http://vir.sgmjournals.org 2785

Description:
mosaic virus protein P6 responsible for suppression of RNA-silencing and salicylic Gary Yates,1 Elina Chrysanthou,1 Ahmed Khattab,1 Andrew J. Love,13. Chiara Geri .. Methods), indicating that this activity was unaffected by .. plants were combined, ground in 10 vols of Extraction Buffer (Bioreba),
See more

The list of books you might like

Most books are stored in the elastic cloud where traffic is expensive. For this reason, we have a limit on daily download.