EtudiantArtcile2

8/14/2019 EtudiantArtcile2

1/13

Tobacco Mosaic Virus Movement Protein Interacts withGreen Fluorescent Protein-Tagged MicrotubuleEnd-Binding Protein 11[W]

Katrin Brandner2, Adrian Sambade2, Emmanuel Boutant, Pascal Didier, Yves Mely,Christophe Ritzenthaler, and Manfred Heinlein*

Institut de Biologie Moleculaire des Plantes, laboratoire propre du CNRS (UPR 2357) conventionne aveclUniversite Louis Pasteur, 67084 Strasbourg cedex, France (K.B., A.S., E.B., C.R. M.H.); and Institut GilbertLaustriat, UMR CNRS 7034, Faculte de Pharmacie, Universite Louis Pasteur, 67401 Illkirch, France (P.D., Y.M.)

The targeting of the movement protein (MP) ofTobacco mosaic virus to plasmodesmata involves the actin/endoplasmic reticulumnetwork anddoes not require an intact microtubulecytoskeleton.Nevertheless, the abilityof MP to facilitatethe cell-to-cell spreadof infection is tightly correlated with interactions of the protein with microtubules, indicating that the microtubule system isinvolved in the transport of viral RNA. While the MP acts like a microtubule-associated protein able to stabilize microtubulesduringlate infection stages, theprotein wasalsoshownto cause theinactivation of thecentrosome upon expression in mammaliancells, thus suggesting that MP may interact with factors involved in microtubule attachment, nucleation, or polymerization. Tofurther investigate the interactions of MP with the microtubule system in planta, we expressed the MP in the presence of greenfluorescent protein (GFP)-fused microtubule end-binding protein 1a (EB1a) of Arabidopsis (Arabidopsis thaliana; AtEB1a:GFP).The two proteins colocalize and interact in vivo as well as in vitro and exhibit mutual functional interference. These findingssuggest that MPinteracts with EB1 and that this interaction may play a role in theassociations of MP with the microtubule systemduring infection.

The tobacco mosaic virus RNA (TMV/vRNA) re-quires the virus-encoded 30-kD movement protein(MP; Deom et al., 1987) for its intercellular spread viaplasmodesmata (PD), cytoplasmic pores in the plant

cell wall that interconnect adjacent cells (Heinlein,2002; Heinlein and Epel, 2004). During infection, theprotein targets PD and transiently increases their sizeexclusion limit (Oparka et al., 1997). MP also has thecapacity to bind single-stranded nucleic acids (Citovskyet al., 1990; Boyko et al., 2002), and because complexes

of MP and vRNA were isolated from TMV-infectedplants (Dorokhov et al., 1983, 1984) and coat protein-deficient virus can still move between cells (Dawsonet al., 1988), vRNA is proposed to be transported in

the form of a nonencapsidated ribonucleoprotein com-plex. Like other RNA viruses, TMV replicates in asso-ciation with the endoplasmic reticulum (ER; Heinleinet al., 1995, 1998a; Reichel and Beachy, 1998). Duringinfection, ER membranes transiently condense to forminclusion bodies (Heinlein et al., 1998a; Reichel andBeachy, 1998) that harbor viral replication complexesand accumulate vRNA, replicase, MP, and coat protein(Mas and Beachy, 1999; Asurmendi et al., 2004). Re-cently, it was shown that the actin/ER network is in-volved in the targeting of MP to PD (Wright et al.,2007). However, while the MP may target PD via ER,several in vivo studies indicate that the cell-to-cell

movement of vRNA also involves interactions of MPwith the microtubule cytoskeleton (Heinlein et al.,1995; McLean et al., 1995; Boyko et al., 2000a, 2000b,2000c, 2002, 2007). Consistent with a role of the ER/actin network in the targeting of MP to PD, MP mu-tants specifically deficient in microtubule associationand vRNA transport are still capable to accumulate inPD (Kahn et al., 1998; Boyko et al., 2000a, 2000c, 2007),suggesting that interactions with the microtubule sys-tem specifically contribute to the transport of vRNA.On the other hand, an intact microtubule cytoskeletonis not required for the spread of infection (Gillespieet al., 2002; Ashby et al., 2006), indicating that micro-tubules are either not essential or that individual

1 This work was supported by the Deutscher AkademischerAustauschdienst, Germany (postdoctoral fellowship grant to K.B.),the Generalidad Valenciana, Spain (postdoctoral fellowship grantsCTBPDC/2204/015 and BPOSTDOC06/072 to A.S.), le ministeredelegue a la recherche, France(grant no.ACI BCMS187 to M.H.), andthe CNRS, France. The fluorescence lifetime imaging microscopy

setup was supported by the Association pour la Recherche contre leCancer, France, the Association Francxaise contre les Myopathies, theFondation pour la Recherche Medicale, France, Sidaction, the pro-gram Physique-Chimie du Vivant of the CNRS, and the ReseauTechnologique en microscopie photonique, France, through the Mis-sions, Ressources et Competences Technologiques of the CNRS.

2 These authors contributed equally to the article.* Corresponding author; e-mail manfred.heinlein@ibmp-ulp.

u-strasbg.fr.The author responsible for distribution of materials integral to the

findings presented in this article in accordance with the policydescribed in the Instructions for Authors (www.plantphysiol.org) is:Manfred Heinlein ([email protected]).

[W] The online version of this article contains Web-only data.www.plantphysiol.org/cgi/doi/10.1104/pp.108.117481

Plant Physiology, June 2008, Vol. 147, pp. 611623, www.plantphysiol.org 2008 American Society of Plant Biologists 611


2/13

microtubules or localized activities of the microtubulecytoskeleton are sufficient to support the transport ofvRNA (Seemanpillai et al., 2006). Recent in vivo obser-vations imply the occurrence of mobile MP-associatedparticles, which were visualized proximal to microtu-

bules, in vRNA movement (Boyko et al., 2007). How-

ever, the exact mechanism by which the microtubulesystem contributes to vRNA movement is still un-known. Interestingly, point mutations in MP that affectthe ability of the protein to associate with microtubulesand to function in vRNA transport in a temperature-sensitive manner cluster in a domain with structuralsimilarity to the M-loop of tubulin known to mediatetubulin-tubulin interactions between microtubule pro-tofilaments (Nogales et al., 1999; Boyko et al., 2000a,2007). Functional mimicry of the tubulin M-loop mayallow MP to undergo direct interactions with tubulinas well as with tubulin-binding factors. Indeed, MPcan bind to polymerized microtubules with characteris-tics of a genuine microtubule-associated protein (MAP)

which, dependent on the amount of bound MP, can leadto the stabilization of the filaments against disruption

by cold or by treatment with microtubule-disruptingagents (Boyko et al., 2000a; Ashby et al., 2006; Ferralliet al., 2006). The potential ability of MP to interact withtubulin-binding factors is supported by heterologousexpression experiments, which demonstrated that MPexpression interferes with centrosomal microtubuleanchorage, g-tubulin recruitment, and microtubulenucleation activity in mammalian cells (Ferralli et al.,2006). This phenotype of MP-expressing cells is inde-pendent of microtubule association of the protein andalso independent of microtubules themselves and thus

suggests that this protein may interact with factorsinvolved in microtubule anchorage or polymerization.

To further investigate the interactions of MP withthe microtubule system in planta, we analyzed in-fected or MP-transfected cells expressing the Arabidopsis(Arabidopsis thaliana) microtubule end-binding proteinAtEB1a fused to GFP (AtEB1a:GFP). EB1 is a microtu-

bule plus-end-tracking protein that regulates microtubule dynamics and promotes end-on attachment todifferent cellular sites (Korinek et al., 2000; Vaughan,2005; Akhmanova and Steinmetz, 2008). AtEB1a repre-sents one of three EB1 proteins in Arabidopsis (AtEB1a,AtEB1b, and AtEB1c; Chan et al., 2003). In transgenic

Arabidopsis suspension cell lines, Arabidopsis plants,or tobacco (Nicotiana tabacum) BY-2 cells, the expressionof AtEB1a:GFP from either a constitutive or induciblepromoter resulted in a typical gradient- or comet-likelabeling of the plus ends of growing microtubules(Chan et al., 2003; Mathur et al., 2003; Dixit et al., 2006).Under conditions of constitutive expression, the pro-tein was also observed to label the minus ends, thusrevealing the dynamic behavior and localization of cor-tical microtubule nucleation sites (Chan et al., 2003). Byanalogy to EB1 function in other eukaryotes (Askhamet al., 2002; Rehberg and Graf, 2002; Louie et al., 2004)and because knockdown experiments do not indicate arole of EB1 in microtubule nucleation per se (Askham

et al., 2002; Rehberg and Graf, 2002; Rogers et al., 2002),it has been proposed that AtEB1 is involved in an-choring microtubules to their nucleation sites that maythen function as a reservoir for EB1 distribution to thegrowing end (Chan et al., 2003).

Here, we show that coexpression of MP fused to red

fluorescent protein (MP:RFP) together with AtEB1a:GFPcauses mutual interference between the proteins withrespect to both subcellular localization and function.The two proteins colocalize on microtubules and inter-act in vivo as well as in vitro. Based on these observa-tions, we propose that EB1 and MP represent mutualinteraction targets that mediate, guide, or controlmicrotubule-associated functions during infection.

RESULTS

TMV Infection Interferes with AtEB1a:GFP Dynamics

To test the effect of TMV infection on microtubuledynamics and nucleation sites, we used Nicotianabenthamiana leaves expressing AtEB1a:GFP followingagroinfiltration. As previously reported for Arabidop-sis and also BY-2 suspension cells (Chan et al., 2003;Van Damme et al., 2004; Dixit et al., 2006), cells in theagroinfiltrated N. benthamiana tissue exhibited the pro-tein in the form of comet-like gradients at the tip ofgrowing microtubules, thus confirming the potency ofthis marker to label the dynamic microtubule cyto-skeleton in heterologous plant species (SupplementalMovie S1). The average rate of the observed microtu-

bule polymerization was 4.8 (61.1) mm min21 (n 5 40

microtubules), which is comparable to the rates mea-sured in Arabidopsis and tobacco BY-2 suspensioncells (Chan et al., 2003; Dixit et al., 2006). In cells highlyexpressing AtEB1a:GFP, the protein sometimes labeledmicrotubules along their length, as has also been ob-served for other systems involving AtEB1a:GFP over-expression (Dixit et al., 2006). However, the microtubulesin such cells were nevertheless dynamic (see, for exam-ple, Fig. 1D; Supplemental Movie S5), and the comet-like staining pattern was always most prominentlyand clearly seen.

To test the effect of TMV infection on microtubuledynamics and nucleation sites, we infected the leaves

with TMV-MP:RFP, a TMV derivative expressing theMP in fusion to RFP (Ashby et al., 2006). At 2 d post-infection (dpi), the leaves were agroinfiltrated for ex-pression of AtEB1a:GFP and analyzed by microscopy40 h later(Fig.1A).Time-lapse microscopyrevealedthatcells in front of spreading infection sites exhibit numer-ous growing microtubules with the typical comet-like gradient of AtEB1a:GFP fluorescence at their tips(Fig. 1B; see also Supplemental Movie S2). However,cells within the infection site were characterized bythe absence of any growing microtubules and comet-like structures usually formed by AtEB1a:GFP. Instead,AtEB1a:GFP now accumulated along the length of themicrotubules (Fig. 1C; see also Supplemental Movie

Brandner et al.

612 Plant Physiol. Vol. 147, 2008


3/13

Figure 1. Viral infection inhibits microtubule dynamics in AtEB1a:GFP-expressing N. benthamiana epidermal leaf cells. A,Expanding infection site (4 dpi) caused by TMV-MP:RFP. Zones for observation by high magnification video microscopy andconfocal microscopy areindicatedby white andyellow rectangles, respectively, referring to B to L. Scale bar5 200 mm.B, In cellsoutside the spreadinginfectionsite, microtubulesare highlydynamic, and AtEB1a:GFPlocalizesto the tipsof growing microtubuleplus ends.Thebottomsegmentshows dynamic GFPpixelswithina timeframe of30 s.See also SupplementalMovieS2.Scalebar510 mm. C, In infected cells, AtEB1a:GFP is associated with microtubules along their length, and no dynamic behavior of theAtEB1a:GFP-associatedmicrotubulescan be detected by video timelapsemicroscopy. The bottom segment shows theabsence ofdynamic GFP pixels within a time frame of 30 s. See also Supplemental Movie S3. Scale bar 5 10 mm. D to F, Dynamics ofAtEB1a:GFP in cells at thefrontof infection, whichexpresslow levelsof MP. The cell on theleft showsAtEB1:GFPcomets, whereasin the cell on the right, which is marked by a white rectangle, the dynamic behavior of microtubules and AtEB1a:GFP is inhibited(see Supplemental Movie S5). The area within the white rectangle is magnified in E and F showing that in the cell on the right,MP:RFP (E) colocalizes with AtEB1a:GFP (F) on microtubules and microtubule-associated spots (arrows). Scale bar 5 10 mm. G,Confocal image of the leading front of a TMV-MP:RFP infection site. In the absence of AtEB1a:GFP, MP:RFP is localized to PD(arrows) and replication bodiesand does not showany accumulation on microtubules. Differentialinterferencecontrastis appliedto show the localization of cell walls. Scale bar 5 20 mm. H, Dual-color confocal image of the leading front of infection in thepresence of AtEB1a:GFP. The leading front cell is marked by an asterisk. An adjacent noninfected cell is marked by a doubleasterisk. The highlighted area shows parts of adjacent infected and noninfected cells and is magnified in I. Scale bar 5 50 mm. I,Whereas in the lower, noninfected cell (double asterisk) microtubulesare dynamic and showthe typical comet-likestainingpatternof AtEB1a:GFP (arrow), thedynamic behaviorof microtubulesand AtEB1a:GFP is inhibited in theupper, infected cell (asterisk;seealso Supplemental Movie S6). In the infected cell, MP:RFP colocalizes with AtEB1a:GFP in microtubule-associated spots. Theyellow color is due to colocalization of green AtEB1a:GFP and red MP:RFP signal (for individual red and green channels, seeSupplemental Fig. S1). Scale bar5 10 mm. J to L, Split red (J), green (K), and merged (L) dual-color confocal image of a cell justbehind the infection front. MP:RFP (J) colocalizes with AtEB1a:GFP (K) on microtubules. Scale bar 5 10 mm.

Tobacco Mosaic Virus Movement Protein Interacts with GFP:EB1

Plant Physiol. Vol. 147, 2008 613


4/13

S3). Thus, infection appears to interfere with the abilityof transiently AtEB1a:GFP-expressing cells to nucleateand polymerize new microtubules. The inhibitory ef-fect of infection on AtEB1a:GFP dynamics occurred al-ready in cells at the leading front of infection (Fig. 1, DF;SupplementalMoviesS4 andS5). In these cells (e.g. in the

cellmarkedbythewhiterectangleinFig.1D),AtEB1a:GFPshowed excessive colocalization with MP:RFP (Fig. 1, Eand F) within spots along microtubules. Given that MPusually does not yet accumulate on microtubules inleading front cells, this localization of MP is likely in-duced by AtEB1a:GFP. The colocalization of bothproteinssuggests that the inhibition of AtEB1a:GFP dynamicsoccurs in consequence of an interaction between thetwo proteins. Inhibition may occur as AtEB1a:GFP se-questers MP, thereby causing premature microtubuleassociation and, therefore, microtubule stabilization

by MP (Ashby et al., 2006). In parallel, or alternatively,MP may sequester AtEB1a:GFP and, potentially, en-dogenous EB1, thus leading to the inhibition of poly-

merization at the microtubule end.The aggregation of MP:RFP and AtEB1a:GFP to

microtubule-associated spots and inhibition of micro-tubule dynamics in cells at the leading front of infectionwas also revealed by confocal microscopy (Fig. 1, Hand I; Supplemental Movie S6; Supplemental Fig. S1).Colocalization of MP:RFP with AtEB1a:GFP was evenmore evident in cells of the second or third cell layer

behind the infection front, where more MP:RFP accu-mulated. Here, both proteins could be found to local-ize along the length of the dynamically inactivatedmicrotubules (Fig. 1, JL). Collectively, these findingsindicate that TMV-MP:RFP infection interferes with

microtubule dynamics in AtEB1a:GFP-expressing cells.The observed colocalization of MP:RFP and AtEB1a:GFPto microtubule-associated spots in cells at the leadingfront of infection expressing only low amounts of theprotein suggests that these two proteins interact andmutually interfere with their normal localization andfunction. This effect may be enhanced by the MAP-likeproperties of MP (Ashby et al., 2006), especially in cellsin which MP:RFP is highly expressed.

We note that the effect of infection on microtubuleand AtEB1a:GFP dynamics is transient. For example,

when cells at the leading front of infection were ana-lyzed at 72 h postinfiltration, the comet-like appearanceof AtEB1a:GFP had partially resumed (data not shown).Thus, the effect of infection on microtubule and AtEB1a:GFPdynamics maydepend on specific transientAtEB1a:GFP and MP:RFP expression conditions or may be over-

come by cellular mechanisms at later time points.

EB1a:GFP Expression Interferes with Virus Movement

The above findings indicate that EB1a:GFP expres-sion leads to microtubule localization of MP:RFP incells at the leading front of infection, where its mi-crotubule localization is not usually observed. To in-vestigate whether this change in localization of thevirus-encoded MP:RFP is correlated with changes inthe efficiency of TMV cell-to-cell spread, we comparedthe MP:RFP-mediated virus movement between tissuesexpressing either AtEB1a:GFP or free, nonfused GFP asa control. Thus, leaves were inoculated with TMV-MP:RFP and, at 3 dpi, images of individual infectionsites were acquired and their sizes measured. Subse-quently, the leaves were agroinfiltrated for expressionof AtEB1a:GFP or GFP, respectively. After 48 h (5 dpi),the infection sites observed at 3 dpi were again ana-lyzed to reveal the increase in their size over time.Eleven TMV-MP:RFP infection sites each were ana-lyzed in AtEB1a:GFP- and GFP-expressing leaves. As isshown in Figure 2, although variable to some extent, theincrease in the size of TMV-MP:RFP infection sites issignificantly reduced in AtEB1a:GFP-expressing leavescompared to GFP-expressing leaves. Thus, AtEB1a:GFPexpression interferes with the efficient spread of TMV-

MP:RFP infection. Given that TMV-MP:GFP infectionin turn interferes with microtubule and AtEB1a:GFPdynamics, it appears that MP:RFP and AtEB1a:GFPinterfere with each other in a mutual manner. This mu-tual interaction is supported by the observed colocal-ization of both proteins. Moreover, the finding that thecolocalization of MP:RFP and AtEB1a:GFP on micro-tubules is correlated with AtEB1a:GFP-induced inhibi-tion of TMV spread may confirm the concept that thequantitative accumulation of MP on microtubules seenduring late stages of normal infection (Heinlein et al.,

Figure 2. Expression of AtEB1a:GFP, but not expres-

sion of GFP, significantly reduces the efficiency ofTMV-MP:RFP cell-to-cell movement. The statisticalsignificance of the effect was confirmed by a Studentst test (P,, 0.01).

Brandner et al.



5/13

1998a) may be related to the inactivation of its trans-port function and that microtubule binding and inac-tivation of MP may be enhanced by ectopic expressionof microtubule-interacting proteins, suchas AtEB1a:GFP,as described here, or MPB2C, as described previously(Curin et al., 2007).

Interference of TMV Infection with AtEB1a:GFPDynamics Is MP Mediated

To investigate if the colocalization of MP:RFP andAtEB1a:GFP and the loss of microtubule dynamicsinvolve a function of MP that is independent of virusinfection, we analyzed the localization of MP:RFP andAtEB1a:GFP in N. benthamiana epidermal cells upontransient coexpression of both proteins in agroinfiltratedleaves. As in the previous agroinfiltration experiment,the cells were observed at 40 h postinfiltration. Whenexpressed alone in wild-type or tua-GFP-expressingplants under these conditions, MP:RFP localized pre-

dominantly to bodies of various sizes (Fig. 3A), ratherweakly to filaments (microtubules, weakly seen in Fig.3A) and strongly to PD-like structures (Fig. 3B), as has

been previously reported for MP:GFP expressed upontransfection by microparticle bombardment (Kotlizkyet al., 2001). Time-lapse analysis oftua-GFP-expressingplants indicates that microtubules are dynamic in MP:RFP-expressing cells (Fig. 3, D, 13, and E, 14) unlessthey are covered with the protein, in which case theyare stabilized (Ashby et al., 2006). The analysis oftua-GFP-expressing plants also revealed that at least someof the MP:RFP bodies occurred in association withmicrotubule y-junctions or intersections, which may

suggest that MP:RFP targets cortical microtubule nu-cleation sites (Supplemental Fig. S2).

When we used agroinfiltration to coexpress AtEB1a:GFP together with tua-GFP, a dynamic microtubulecytoskeleton was observed, and the growing microtu-

bule plus ends were highlighted by EB1a:GFP (Fig. 3F;Supplemental Movie S7). We note that, consistent withthe reports byChanet al.(2003) and Murata etal. (2005),AtEB1a:GFP also labeled microtubule nucleation siteson existing microtubules, from which new tua-GFP-and AtEB1a:GFP-labeled microtubules emerged, thusforming y-junctions (Fig. 3G, 13). Importantly, we alsoobserved that following microtubule growth, microtu-

bule shortening coincided with the disappearance ofthe AtEB1a:GFP-labeled capfrom the plus end(Fig. 3H,17; Supplemental Movie S7), thus providing evidencefor the functional integration of AtEB1a:GFP in micro-tubule polymerization.

We then replaced tua-GFP with MP:RFP and ana-lyzed the agroinfiltrated cells again at 40 h postagroin-filtration. Here, unlike in cells expressing AtEB1a:GFPalone or in combination with tua:GFP, the dynamic be-havior of microtubules and AtEB1a:GFP was impairedand AtEB1a:GFP localized along the length of the mi-crotubules (Fig. 3, IL; Supplemental Movie S8). More-over, also the MP:RFP localized predominantly tomicrotubules (Fig. 3, IK), which is in contrast to cells

expressing MP:RFP in the absence of AtEB1:GFP (Fig.3A). Control agroinfiltration experiments demonstratedthat the strong inhibitory effect of MP:RFP on microtu-

bule dynamics in AtEB1a:GFP-expressing cells cannotbe mimicked by coexpression of the RFP-tagged mi-crotubule-binding domain of MAP4 (RFP:MAP4-MBD;

Van Damme et al., 2004) and also the coexpression ofthe microtubule-stabilizing, GFP-tagged, ArabidopsisMAP65-5 (AtMAP65-5:GFP; Van Damme et al., 2004)does not producesuch inhibition (Supplemental MoviesS9 and S10; Supplemental Fig. S3). Based on thesefindings, we conclude that the inhibitory effect of TMV-MP:RFP infection on microtubule dynamics observedin AtEB1a:GFP-expressing cells at 40 h postinfiltration(Fig. 1C) is mediated by an interaction of AtEB1a:GFPwith MP:RFP and does not require infection. This inter-action is not caused bythe RFP moiety, and the ability ofMP to bind and stabilize microtubules (Ashby et al.,2006) appears to be insufficient to fully account for thiseffect.

We note that in agreement with the transient natureof inhibition of microtubule and AtEB1a:GFP dynam-ics and the colocalization of MP:RFP and AtEB1a:GFPon microtubules during infection, AtEB1a:GFP cometswere occasionally seen also in transiently expressingcells. In these cases, the MP:RFP localized to PD, or toER or punctate foci in protoplasts.

Moreover, in cells expressing only moderate levelsof MP:RFP and AtEB1a:GFP, microtubule colocaliza-tion of the two proteins appeared to be concentrated inrather distinct spots that often localized at microtubuley-junctions (Supplemental Fig. S2), which is consistentwith the notion that MP may target microtubule nu-

cleation sites.To determine whether our observations could be

dependent on agroinfiltration and transient expres-sion conditions in leaves, we prepared protoplasts ofa transgenic BY-2 suspension cell line stably express-ing AtEB1a:GFP under the control of a 35S-promotor(Van Damme et al., 2004) and infected them with TMV-MP:RFP. Consistent with our observations in planta,the typical dynamic EB1 and microtubule pattern seenin this cell line was maintained in noninfected proto-plasts (Supplemental Fig. S4), whereas infection re-sulted in the colocalization of AtEB1a:GFP with MP:RFPalong the length of the microtubules (Fig. 3, MO) and

in the inhibition of microtubule polymerization dy-namics (Fig. 3, P and Q). This finding confirms that theinhibitory influence of MP:RFP on microtubule dy-namics in AtEB1a:GFP-expressing cells is independentof both the plant system and the method used forexpression. We also infected the protoplasts with TMV-MPP81S:GFP. The MP:GFP encoded by this virus carriesan inactivating P81S amino acid exchange mutationthat does not interfere with the ability of the protein to

bind single-stranded nucleic acids but causes mislocal-ization of the protein to the cytosol (Boyko et al., 2002;Vogler et al., 2008). Unlike MP:GFP or MP:RFP, thismutant MP:GFP neither colocalized with AtEB1a:GFPnor interfered with microtubule and AtEB1a:GFP dy-




6/13

Figure 3. Transient expression of MP:RFP,AtEB1a:GFP, and both MP:RFPand AtEB1a:GFP. A to E4, Transient expression ofMP:RFP. A, Upon expression by agroinfil-tration, MP:RFP localizes to small andlarger bodies as well as (weakly) to fila-ments. Scale bar 5 10 mm. B, Transiently

expressed MP:RFP also localizes to PD.Scale bar 5 10 mm. C to D3, Expression ofMP:RFP upon agroinfiltration in tua:GFPplants. The presence of MP:RFP (C) doesnot interfere with dynamic microtubulegrowth (yellow arrows, movie framesD13 and E14) or shrinkage (red arrows,movie frames E3 and 4). Scale bars 5 5mm. F to H7, Expression of AtEB1a:GFP. F,AtEB1a:GFP highlights growing plus endsof tua:GFP-labeled microtubules (see alsoSupplemental Movie S7). Scale bar 5 10mm. G1 to 3, Movie frames showing thatAtEB1a:GFP also labels foci (arrow in G2)from which microtubules originate (arrow

in G3) and, thus, marks the location ofmicrotubule nucleation sites (arrowheadin G1). As the microtubule polymerizes,AtEB1a:GFP associates with the grow-ing plus end (arrow in G3). Scale bar 52.5 mm. H1 to 7, Loss of AtEB1a:GFPlabeling is associated with microtubuleshrinkage. Movie frames showing a grow-ing microtubule with AtEB1a:GFP labelingat the tip (H13, arrow) that upon loss ofthe AtEB1a:GFP cap (H4, arrow) exhibitsrapid shrinkage (H57, arrow). Scale bar 52.5 mm. I to L, Transient coexpression ofMP:RFP and AtEB1a:GFP. I to K, Greenchannel (I), red channel (J), and mergedchannel (K) movie frames showing tran-siently expressed MP:RFP and AtEB1a:GFPcolocalizing to microtubules, which showno signs of dynamic activity. See also Sup-plemental Movie S8. Scale bar5 10 mm. L,Projection of green channel dynamic pixelswithin 30 s of the time-lapse movie. Theabsence of dynamic pixels indicates thatthe AtEB1a:GFP-labeled microtubules shownin I are dynamically inactive. M to Q,Protoplasts derived from an AtEB1a:GFP-transgenic BY2-cell suspension line infectedwith TMV-MP:RFP. M to O, Confocal imagesillustrating that AtEB1a:GFP and MP:RFP

colocalize along the length of dynamicallyinhibited microtubules. Green (M) and red(N) channel images, showing the distributionof AtEB1:GFP and MP:RFP, respectively, aswell as a merged image (O), are shown.Scale bar 5 5 mm. P and Q, Movie datashowing the lack of microtubule dynamics inTMV-MP:RFP-infected protoplasts.

Brandner et al.



7/13

namics (Supplemental Fig. S4; Supplemental MovieS11).

MP:RFP and AtEB1a:GFP Form a Complex in Vitro andin Vivo

To test whether the inhibitory effect of MP:RFP onAtEB1a:GFP and microtubule dynamics may be me-diated by the formation of a complex between bothproteins, a pulldown assay was performed by using re-combinant MP:His6bound to NiNTA sepharose (Ashbyet al., 2006) as affinity matrix for proteins present inthe soluble fraction of a lysate prepared from theAtEB1a:GFP-expressing BY-2 cells. The cell lysate wasprepared in the cold and in the presence of TritonX-100 to disrupt microtubules and membranes, respec-tively. The extract was cleared by subsequent centrifu-gations at 20,000g and 100,000g to obtain a solubleprotein fraction, which was used for incubation withMP:His6. To control for nonspecific binding, the proteins

were also incubated with the nonconjugated NiNTAmatrix. Moreover, to disrupt unspecific ionic interac-tions between MP and proteins in the extract, the assaywas performed in the presence of 1% bovine serumalbumin (BSA) and 250 mM NaCl.

Western-blot analysis (Fig. 4A) leads to the detectionof AtEB1a:GFP in the elution fraction, along witha-tubulin. Binding of these proteins is specific forfunctional MP, because the amount of these proteinsfound in the elution fraction was significantly reducedwhen the assay was performed with MPP81S:His6 (Fig.4B),despitethattheamountofMPP81S:His6 was equal tothe amount of MP:His6 applied in this assay (data not

shown). The interaction between AtEB1a:GFP andMP indicated by this result was further tested by far-western analysis (Fig. 4C). Here, proteins in cell lysatesderived from the AtEB1a:GFP-expressing BY-2 cellsand from nontransgenic control cells were blotted onpolyvinylidene fluoride membrane following electro-phoresis, then denatured and renatured, and finallyincubated with soluble recombinant MPP81S:His6 (lanes1 and 2) or MP:His6 (lanes 36). Detection of MP

P81S:His6 and MP:His6 with anti-MP antibody a band ofabout 60 kD, which is present on the blot incubatedwith MP:His6 (lane 4, asterisk) but is absent on the blotincubated with MPP81S:His6 (lane 2), and which is also

absent from lanes in which proteins derived fromAtEB1a:GFP-nonexpressing control cells were sepa-rated (lanes 1 and 3). Reprobing the membrane withanti-GFP antibody (lanes 5 and 6) revealed that the 60-kD band detected in the MP-overlay corresponds toAtEB1a:GFP (lane 6, asterisk). This 60-kD band is notdetected when the overlay experiment is performedwith denatured MPP81S:His6 (lanes 7 and 8), denaturedMP:His6 (lanes 9 and 10), or no protein (lanes 11 and 12),and probing the membrane with anti-MP antibody,indicating that the binding of MP to AtEB1a:GFP (lane4) is specific and requires the tertiary structure of MP.Antibody cross-reactivity leads to detection of a singlenonspecific band around 40 kD (double asterisk) in all

experiments involving anti-MP antibody. Collectively,these results indicate that AtEB1a:GFP is an interactionpartner for MP.

To confirm in vivo the interactions observed betweenAtEB1a:GFP and MP:RFP, we analyzed the value offluorescence resonance energy transfer (FRET)fromthe

excitedfluorescent donor GFPto the RFP acceptor. FRETis dependent on both protein tags being in close prox-imity, generally up to a maximum of 5 to 10 nm, a dis-tance corresponding to intermolecular protein-proteininteractions (Bastiaens and Pepperkok, 2000; Hinket al., 2002). The FRET efficiency was straightforwardlymeasured with the fluorescence lifetime imaging mi-croscopy (FLIM) technique through the decrease of thefluorescence lifetime of the donor at each spatiallyresolvable element of a microscope image. Indeed, incontrast to fluorescence intensities, the fluorescencelifetimes areabsolute parameters that do notdepend onthe instrumentation or the local concentration of thefluorescent molecules. Thus, changes of the fluores-

cence lifetimes of the donor will provide a direct evi-dence for a physical interaction between the labeledproteins with high spatial and temporal resolution(Bastiaens and Squire, 1999).

The mean fluorescence lifetime of the donor mole-cule was first determined in cells expressing AtEB1a:GFP alone. An average fluorescence lifetime of 2.37 60.06 ns was determined from measurements performedon 170 individual microtubules analyzed in three sep-arate experiments (Table I; Fig. 4D). This average life-time for the S65T GFP variant moiety of AtEB1a:GFP iswell in agreement with the 2.4-ns lifetime reported forEGFP (F64L, S65T)-labeled proteins (Jakobs et al., 2000;

Treanor et al., 2005). Subsequently, FLIM-based FRETanalysis was performed on microtubules in cells ex-pressing AtEB1a:GFP together with MP:RFP as accep-tor molecule. Here, the fluorescence lifetime of thedonor AtEB1a:GFP was reduced to 1.896 0.13 ns(t test.0.05), which equals a FRET efficiency of 21% (Table I;Fig. 4E). As a control, no photons at wavelengths cor-responding to S65T-GFP emission could be detectedwhen only MP:RFP or free RFP was present (data notshown), confirming that FLIM measurements are spe-cific for the S65T-GFP-tagged donor molecule. To deter-mine whether FRET between MP:RFP and AtEB1:GFPmight occur because of overexpression or because of

colocalization and vicinity of binding of the two pro-teins to the microtubule surface, we replaced AtEB1a:GFP with GFP-fused MAP4-MBD (GFP:MAP4-MBD).In this case, the average lifetime of the donor (Fig. 4F)remained unchanged in the presence of MP:RFP (Fig.4G). We therefore conclude that MP:RFP interacts withAtEB1a:GFP in vivo.

DISCUSSION

Research presented here was undertaken to furtheraddress the interaction of the TMV MP with themicrotubule cytoskeleton. Recent evidence has shown




8/13

Figure 4. MP interacts with EB1a:GFP in vitro and in vivo. A, Pulldown assay with MP:His6 as bait. Immobilized recombinantMP:His6 binds AtEB1a:GFP as well as tubulin. MP:His6 resin and control resin (control) were incubated with protein extractsderived from BY2-cells that were either wild type (WT) or transgenic for AtEB1a:GFP (EB1a). Following electrophoresis andblotting of the eluted proteins, the membrane was incubated with antibody against a-tubulin (left, lanes 14) or antibody againstGFP (right, lanes 58; detection of AtEB1a:GFP). A slight cross reactivity of anti-GFP antibody with MP:His6 is indicated by anasterisk. B, Compared to MP:His6, MP

P81S:His6 has reduced capacity to bind EB1 or tubulin in vitro. Pulldown assay withMP:His6, MP

P81S:His6, and control resin for binding of proteins from extracts derived from AtEB1a:GFP-transgenic BY2-cells.Eluted proteins were blotted and probed with antibody against a-tubulin (left, lanes 13) or antibody against GFP (right, lanes46; detection of AtEB1a:GFP). The column chart displays the amount of eluted proteins averaged from four experiments.Amount of MP:His6-bound a-tubulin and AtEB1a:GFP in the eluate was set to 100%. C, Far-western assay. Recombinant solubleMP:His6, but not recombinant MP

P81S:His6, binds to immobilized AtEB1a:GFP. Proteins derived from BY2-cells that were eitherwild type (WT) or transgenic for AtEB1a:GFP (EB1a) were separated by electrophoresis and blotted. Following denaturation and

Brandner et al.



9/13

that during the course of infection, the MP first formsmobile, microtubule-associated particles (Boyko et al.,

2007) and, during later stages, accumulates on micro-tubules, leading to their stabilization (Boyko et al.,2000a; Ashby et al., 2006). However, although theseassociations with microtubules are related to the func-tion of the protein in TMV RNA movement (Boykoet al., 2000a, 2000b, 2000c, 2002, 2007), the microtu-

bules themselves may not represent the initial target ofMP. Indeed, results of experiments using heterologousexpression of MP in mammalian cells suggested thatthe MP may interact with microtubule nucleationfactors before accumulating on the microtubules them-selves (Ferralli et al., 2006).

The findings described herein provide confirmation

to this notion by showing that during infection and intransfected plant cells, MP colocalizes and interactswith AtEB1a:GFP. The ability of MP to interact withAtEB1a:GFP and to undergo FRET in vivo is demon-strated by FLIM and supported by the ability of His-tagged MP to form a complex with plant-derivedAtEB1a:GFP in vitro. These findings indicate that MPmay target EB1, a factor of central importance in theregulation of microtubule dynamics (Chan et al., 2003;Bisgrove et al., 2004; Lansbergen and Akhmanova,2006; Akhmanova and Steinmetz, 2008). An interactionwith EB1 could be in the pathway that during lateinfection leads to microtubule accumulation of MP

because MP

P81S

, which fails to efficiently interact with

AtEB1a:GFP in vitro, also fails to colocalize with mi-crotubules and AtEB1a:GFP in vivo.

Our experiments involved the expression of highamounts of AtEB1a:GFP and, except for cells at theleading front of infection, also of MP:RFP. Thus, someof our in vivo observations may be related to the high

expression of these proteins. For example, at a highconcentration, like during late stages of infection, theMP exhibits properties of a MAP that binds andstabilizes microtubules (Ashby et al., 2006). Thus, theobserved inhibition of microtubule and AtEB1a:GFPdynamics may involve manifestations of this MAPactivity. Nevertheless, interactions are implied by thefact that, generally, the proteins behaved normallywhen ectopically expressed alone, whereas they colo-calized to the length of microtubules when expressedtogether. In fact, several observations argue for directin vivo interactions between MP:RFP and AtEB1a:GFP.

First, MP:RFP colocalized with AtEB1a:GFP to mi-crotubules also in cells at the leading front of infection,

which express only low amounts of MP:RFP and inwhich MP:RFP usually does not occur at detectablelevels on microtubules. Thus, in this case, the inhibitionof microtubule and AtEB1a:GFP dynamics is indepen-dent of high MP:RFP expression but rather a result ofsequestration of MP:RFP by AtEB1a:GFP.

Second, we show by in vitro affinity binding and far-western experiments, as well as by in vivo FLIMexperiments, that MP:RFP and AtEB1a:GFP have thecapacity to interact. Given that the FLIM results areindicativeof FRET, which occurs only if protein tags areless than 10 nm apart, the interactions involve direct

binding interactions in vivo.

Third, coexpression of AtEB1a:GFP with othermicrotubule-binding proteins, i.e. RFP:MAP4-MBD andMAP65-5:GFP, did not result in an inhibition of cellularmicrotubule and AtEB1a:GFP dynamics to the extentseen with MP:RFP under the same experimental con-ditions. Thus, binding and stabilization of microtu-

bules seem insufficient to explain the effects of MP:RFPon microtubule dynamics occurring in the presence ofAtEB1a:GFP.

In addition, when expressed alone or togetherwith AtEB1a:GFP, MP was observed to accumulate atpunctate, microtubule y-junctions and othermicrotubule-associated sites, which may bear resemblance to previ-

ously described cortical microtubule nucleation sites

Figure 4. (Continued.)subsequent renaturation the blotted proteins were incubated with either soluble, recombinant MPP81S:His6 (lanes 1 and 2),MP:His6 (lanes 3, 4, 5, 6, 13, and 14), heat-denatured MP

P81S:His6 (lanes 7 and 8), heat-denatured MP:His6 (lanes 9 and 10), ormock solution (lanes 11 and 12). Following extensive washing, the blot overlays were incubated with antibody against either MP(lanes 1 4 and 712) or GFP (lanes 5 and 6). Blot overlays were also incubated with secondaryantibody only (lanes 13 and 14).*, Lanes 4 and 6, AtEB1a:GFP; **, all lanes except 5, 6, 13, and 14, unspecific cross reactivity with anti-MP antibody. D to G,FLIM assay. MP interacts with AtEB1a:GFP in vivo. Fluorescence intensity images acquired by FLIM are shown as gray scalepictures (left). Lifetime images (central) are represented as pseudo-color according to the color code ranging from 1 ns (blue) to3 ns (orange). The respective lifetime values measured for AtEB1a:GFP (D) and GFP:MAP4 (F) alone or upon coexpression withMP:RFP (E) and (G), respectively, are indicated on the color scales. Coexpression with MP:RFP strongly reduces the fluorescencelifetime of AtEB1a:GFP but not of GFP:MAP4-MBD.

Table I. FRET-FLIM analysis of interactions between AtEB1a:GFP(S65T) or GFP:MAP4-MBD and MP:RFP (mRFP1) upon transientexpression in N. benthamiana leaf cells

The average fluorescence lifetime (t) values and their respective SDs,as determined for AtEB1a:GFP alone or in the presence of MP:RFP, areshown. From the fluorescence lifetimes, the percentage of FRET wascalculated by the equation given in Materials and Methods. The totalnumber of microtubules analyzed is indicated by n, and N is thenumber of independent experiments.

Proteins Lifetime t SDFRET

EfficiencyN n

ns %

AtEB1a:GFP 2.37 0.06 3 170AtEB1a:GFP 1 MP:RFP 1.89 0.13 21 3 166GFP:MAP4-MBD 2.20 0.07 2 123GFP:MAP4-MBD 1 MP:RFP 2.21 0.07 0 2 179




10/13

(Chan et al., 2003; Murata et al., 2005). This finding mayreflect interactions of MP with endogenous EB1 atmicrotubule nucleation sites. An interaction of MP withendogenous EB1 at such sites would be consistent withour previous finding that the expression of MP in mam-malian COS7 cells interferes with the ability of centro-

somes to anchor and polymerize microtubules (Boykoet al., 2000a; Ferralli et al., 2006). Mammalian EB1localizes to the centrosome and is required for centroso-mal microtubule minus-end anchorage and polymeri-zation (Louie et al., 2004). The MP-induced inhibitionof centrosomal activity is independent of microtubuleassociation (Ferralli et al., 2006) and indeed appears tomimic the cytoskeletal phenotype of EB1-depleted cells(Louie et al., 2004), thus suggesting that MP may bindand sequester EB1 in mammalian cells.

Because AtEB1a:GFP binds MP:RFP, the effects ofectopic expression of AtEB1a:GFP may resemble the ef-fects of ectopic expression previously reported forMPB2C (Kragler et al., 2003). Like AtEB1a:GFP, MPB2C

also binds both microtubules and MP. Moreover, sim-ilar to the case of AtEB1a:GFP, the expression of MPB2Calso resulted in the binding of MP to microtubules andinhibition of PD-mediated transport (Kragler et al.,2003; Curin et al., 2007). It may be possible that EB1 andMPB2C share a role in regulating or mediating micro-tubule interactions of MP and lead to the production ofan inhibitory complex when overexpressed.

Considering that AtEB1a:GFP may be expressed to ahigher level than endogenous EB1, the observed inhi-

bition of microtubule dynamics in cells expressing bothAtEB1a:GFP and MP does not necessarily imply thatinteractions of MP with endogenous EB1 would also

lead to the formation of an inhibitory complex. Indeed,we noted here that in the absence of AtEB1a:GFP,microtubules are dynamic in MP-expressing cells, un-less microtubules are coated with the protein (Ashbyet al., 2006). Unfortunately, the analysis of potentialinteractions of MP with endogenous N. benthamianaEB1 is hampered by the lack of EB1 protein sequenceinformation and of suitable antibodies. Thus, withoutsuch important tools, further studies may have to beperformed with an Arabidopsis-infecting tobamovirus,

because EB1 protein sequences are known for Arabi-dopsis and can be addressed by mass spectroscopy.

Nevertheless, given the evidence indicating that MP

interacts with EB1, several potential roles of such aninteraction can be envisioned. Because in our experi-ments the overexpression of AtEB1a:GFP promoted the

binding of MP to microtubules, EB1 could participate inthe formation of MP:microtubule complexes usuallyobserved during late stages of infection, when highamounts of MP have accumulated. On the other hand,MP may target EB1 to manipulate microtubule poly-merization. For example, EB1b, another member of theEB1 family in Arabidopsis, was shown to localize tomicrotubule tips that upon extension can exert pullingforces on ER membranes and thus contribute to dy-namic endomembrane reorganization (Mathur et al.,2003). Considering that in the plant cortical cytoplasm,

microtubules are in close proximity to the endomem-brane network (Lancelle et al., 1987; Hepler et al., 1990;Lichtscheidl and Hepler, 1996), an interaction betweenMP and EB1 could thus contribute to the formation,distribution, or movement of membrane-associated rep-lication complexes. The interaction of MP with EB1 may

also allow MP to capture microtubule plus ends andthereby cause the local reorganization of the microtu-bule array. A precedent for such a model is provided bythe EB1-interacting adenomatous polyposis coli pro-tein. This protein occurs in the centrosome (Louie et al.,2004) but also in peripheral clusters able to capturegrowing microtubule ends and thus to contribute to theorganization of microtubule networks in mammaliancells (Barth et al., 2002; Reilein and Nelson, 2005). Aninteraction of MP with similar complexes may beinvolved in the MP-induced formation of protrusionson the surface of infected protoplasts (Heinlein et al.,1998a), of fibrous structures in the cavities of PD in MP-transgenic plants (Ding et al., 1992; Moore et al., 1992;

Lapidot et al., 1993),or of cytoskeletalfibers through thesepta that connect adjacent cells in MP-transgenic cy-anobacteria (Heinlein et al., 1998b). Finally, in otherorganisms, EB1 proteins were shown to interact withmicrotubule motor proteins (Korinek et al., 2000;Browning et al., 2003). Thus, MP could interact withEB1 to gain access to microtubule motor-mediated traf-ficking during early stages of infection.

MATERIALS AND METHODS

Plant Material

The in planta observationswere made in Nicotiana benthamiana using eitherwild-type plants or plants that express the Arabidopsis ( Arabidopsis thaliana)

TUA6 gene fused to GFP (tua-GFP) and produce GFP-labeled microtubules

(Gillespie et al., 2002).Wild-type andtransgenic plantswere grown from seeds

and maintained in approximately 70% humidity at 23C with a 16-h photo-

period. Three- to 4-week-old plants were used for infiltration assays and

inoculation experiments.

Constructs

The construction of infectious clones encoding TMV-MPP81S:GFP andTMV-

MP:RFP and conditions for the inoculation of plants are described elsewhere

(Boyko et al., 2002; Ashby et al., 2006). Also, the expression of His 6-tagged MP

(MP:His6) from expression vector pQE60 (QIAGEN) is described elsewhere

(Boyko et al., 2002).

Binary vectors pB7-MP:GFP and pB7-MP:RFP expressing the MP of TMV

fused C terminally to fluorescent protein under control of the 35S promoter

were created by Gateway cloning. The full-length MP sequence was amplified

from TMV-U1-encoding plasmid pU3/12 (Holt and Beachy, 1991), using

primers containing attB1 and attB2 sites (forward primer, GGGGACAAG-

TTTGTACAAAAAAGCAGGCTATGGCTCTAGTTGTTAAAGGAAAAGTG;

reverse primer, GGGGACCACTTTGTACAAGAAAGCTGGGTAAAACGA-

ATCCGATTCGGCGACAGTAGCC) and recombined into pDonR/Zeo (Invi-

trogen). Following sequence confirmation, this entry clone was used for

recombination with destination vectors pB7FWG2 or pB7RWG2, respectively

(Karimi et al., 2002). A binary pBIN-GWC vector encoding AtEB1a:GFP (Chan

et al., 2003) was kindly provided by Dr. Jordi Chan and Professor Clive Lloyd

(John InnesCentre, Norwich,UK). Binary vectorsencoding G/RFP:MAP4-MBD

and AtMAP65-5:GFP, respectively, were previously described (Van Damme

et al., 2004) and supplied by Danny Geelen and Dirk Inze (Ghent University,

Belgium). Plasmid pBinGFP, a binary vector encoding GFP under control of the

35S promoter, was kindly provided by Olivier Voinnet (Institut de Biologie

Moleculaire des Plantes, Strasbourg, France).

Brandner et al.



11/13

Agroinfiltration

Transformed agrobacteria (GV3101) were grown at 28C in 5 mL Luria-

Bertani medium containingselective antibiotics. Uponharvest(OD6005 0.5) by

centrifugation, the volume of bacteria was resuspended in the same volume

of water andinfiltratedwiththe help of a syringe into leaves. Forcoinfiltration,

the suspensions were mixed equally (1:1) just before infiltration. At 36 to 46 h

postinfiltration, the infiltrated leaf regions were analyzed by fluorescence

microscopy.

Infection of BY-2 Cells

Protoplasts of tobacco (Nicotiana tabacum) BY-2 cells transgenic for AtEB1a:

GFP (Van Dammeet al.,2004) werepreparedand inoculatedby electroporation

with infectious transcripts according to procedures described previously

(Heinlein et al., 1998a).

Fluorescence Microscopy and Image Processing

Plant tissues as well as protoplasts were observed with a Nikon TE2000

inverted microscope equipped for real-time imaging with a Roper CoolSnap

digital CCD camera, a piezo-drivenZ-focus,and a 603 1.45 NATIRF objective.

Excitation/emission wavelengths were 460 to 500 nm/510 to 560 nm for GFP

and 550 to 600 nm/615 to 665 nm for RFP. For simultaneous dual color acqui-sitions, a Dual-View beam splitter was used. The beam splitter was equipped

with GFP-mRFP1 exciter and mirror as well as with emission filters HQ510/

30m for GFP and HQ650/75m for RFP. To reduce red background emission

fluorescence caused by chlorophyll, an e640sp short pass filter (Chroma) was

inserted between the beam splitter and the CCD. Metamorph (6.2r6) and

ImageJ (1.32j) software was used for image acquisition, analysis, and process-

ing. Images showing dynamic movie pixels were created by projecting pixel

differences between movie frames using an Image J macro available at http://

rsb.info.nih.gov/ij/macros/Slice-to-Slice%20Difference.txt.

BY-2 protoplasts settled on a poly-L-Lys-coated coverslip and mounted into

an Attofluorcell chamber (Invitrogen)were observed witha ZeissLSM510laser

scanning microscope using a C-Apo-chromat (633; v1.2 W Korr) water ob-

jective lensunder multitrackmode. Excitation/emissionwavelengthswere 488

nm/505 to 545 nm for GFP and 543 nm/long pass 560 nm for mRFP. Confocal

images were processed using LSM510 version 2.8 (Zeiss), ImageJ (1.32j), and

Adobe Photoshop v7.0.

FLIM

Time-correlated single-photon counting FLIM measurements were per-

formed on a home-built two-photon system based on an Olympus IX70

inverted microscope with an Olympus 603 1.2NA water immersion objective,

as previouslydescribed(Azoulayet al.,2003; Clamme et al.,2003). Two-photon

excitation was provided by a mode-locked titanium:sapphire laser (Tsunami;

Spectra Physics), which was tuned to an emission wavelength of 900 nm. For

FLIM, the laser power was adjusted to give counting rates with peaks up to a

few 106 photons s21 so that the pile-up effect can be neglected. Imaging was

realized by a laser scanning system using two fast galvo mirrors (model 6210;

Cambridge Technology) operating in the descanned fluorescence collection

mode.

Photons were collected using a two-photon short pass filter with a cut-off

wavelength of 680nm (F75-680;AHF), anda band-passfilter5206 17 nm(F37-520; AHF). Fluorescence was directed to a fiber-coupled avalanchephotodiode

(SPCM-AQR-14-FC; Perkin Elmer), which was connected to a time-correlated

single photon-counting module (SPC830; Beckerand Hickl), whichoperates in

the reversed start-stop mode.

Typically, the samples were scanned continuously for about 30 s to achieve

appropriate photon statistics for the fluorescence decays. Data were analyzed

using a commercial software package (SPCImage V2.8; Becker and Hickl),

which uses an iterative reconvolution method to recover the lifetimes from the

fluorescence decays.

In FRET experiments, when coexpressing donor and acceptor proteins, the

FRET efficiency reflecting the distance between the two chromophores was

calculated according to:

E5R60

R6

01R

!5 1

tfret

tfree

where R0 is the Forster radius, R the distance between donor and acceptor, tfretis the lifetime of the donor in the presence of the acceptor, and tfree the lifetime

of the donor in the absence of acceptor.

Homogenization of BY-2 Cells

BY-2 cells were harvested by filtration with a vacuum pump. The vacuum-

dry pellet was shock-frozen in liquid nitrogen. Frozen cells were powderedwith a micro-dismembrator (Satorius) for 2 min at 3,000 rpm. Then 1 g of the

powdered cells was resuspended in 2 mL of pulldown buffer I (PDB-I: 50 mM

HEPES,pH 7,25 mM imidazole,250 mM NaCl, 2 mM dithiothreitol[DTT], 2 mM

MgCl2, 10% glycerol, 0.5% Triton X-100, 2 mm phenylmethylsulfonyl fluoride,

13 protease inhibitor cocktail [Roche]) and thawed on ice. To obtain a ho-

mogenous suspension, the lysed cells were passed five times through a 26G

needle. The soluble protein fraction (total fraction) was obtained by centrifu-

gation of the lysate at 20,000g for 10 min at 4C and subsequent centrifugation

of the supernatant at 100,000g for 1 h at 4C.

Pulldown Assay

Recombinant MP:His6 and mutant MP (MPP81S:His6) were prepared from

Escherichia coli as described (Boyko et al., 2002), except that the renaturation of

MP:His6 was performed by resuspension of the MP:His6- or MPP81S

:His6-complexedNiNTA Sepharosebeads inPDB-Iby incubation for1 h at4C.Forin

vitro binding studies, each soluble protein sample was preincubated with

NiNTA Sepharose for 1 h at4C to reducethe amountof nonspecific binding to

thebeads. Subsequently, eachof thesampleswas divided intotwo aliquots,one

of which was rotated for 2 h at 4 C with equally MP:His6- or MPP81S:His6-

complexed NiNTA beads, whereas the other was incubated under the same

conditions with empty NiNTA beads as control. Subsequently, the samples

were centrifuged at 1000 rpm at 4C for 2 min and washed 10 times in PDB-II

(50 mM HEPES, pH 7, 25 mM imidazole, 250 mM NaCl, 2 mM DTT, 2 mM MgCl2,

10% glycerol, 0.25% Triton X-100, 2 mm phenylmethylsulfonyl fluoride,

13 protease inhibitor cocktail). Following elution by incubation in elution

buffer (500 mM NaCl, 50 mM HEPES, 500 mM imidazole, pH 7.5),proteins were

separated by 12.5% SDS-PAGE and blotted onto membrane for detection with

GFP-specific rabbit anti-GFP-IV antiserum (kindly provided by D. Gilmer,

IBMP, Strasbourg), monoclonal anti-acetylated a-tubulin antibody (Sigma), or

polyclonal anti-AtEB1b antibody (Sigma).

Far Western

Wild-type BY-2 cellsand cellsexpressing AtEB1a:GFPwere resuspended in

SDSsample buffer(1% SDS, 10%glycerol,25 mM Tris-HCl,pH 6.8, 1 mM EDTA,

0.7 M mercaptoethanol) to obtain a final protein concentration of 100 mg/mL.

The cells were lysed by vortexing and heating for 3 min at 95 C, and 300 mg of

total protein was subjected to SDS-PAGE. Upon separation by electrophoresis,

the proteins was blotted onto polyvinylidene fluoride membrane using wet

transferand a transferbuffer (25mM Tris,192 mM Glycin) withoutmethanoland

SDS to ease protein renaturation. The proteins were denatured in buffer A (20

mM Tris-HCl, pH 7.5, 1 mM EDTA, 100 mM NaCl, 13 Denhart solution [0.02%

Ficoll, 0.02% BSA, 0.02% PVP-K90]) containing 6 M guanidine-HCl for 15 min

and renatured by consecutively washing the membrane at 4C with a buffer A

series with decreasing concentrations of guanidine-HCl, i.e. buffer A contain-

ing 3 M guanidine-HCl for 10 min, bufferA containing 1.5 M guanidine-HCl for

10 min, buffer A containing 0.75 M guanidine-HCl for 10 min, buffer A

containing 0.375 M guanidine-HCl for 10 min, twice with buffer A without

guanidine-HClfor 30 min,and finally againwith buffer A for 2 h. Subsequently,

the membrane was incubated with blocking solution (20 mM Tris-HCl, pH 7.5,

0.5 mM EDTA, 10% glycerol, 100 mM NaCl, 0.1% Tween 20) containing 5%

skimmed milk powder for 1 h at 4C, and after washing twice with blocking

solution, the membrane was incubated overnight at 4C with 50 mg of purified

MP:His6 in blocking solution supplemented with1 mM DTT, 2% skimmed milk

powder, and 0.5% BSA. Subsequently, the membrane was washed for 1 h at

4C in blocking solution containing 5% skimmed milk powder and incubated

with anti-MP-C (reactive against MP residues 209222) antibody for 2 h

at 4C. Bound MP antibody was detected by incubation with anti-rabbit IgG

antibody-horseradish peroxidase conjugate (Sigma) followed by a chemilu-

minescence reaction using ECL Plus western blotting detection reagent (GE

Healthcare).




12/13

Supplemental Data

The following materials are available in the online version of this article.

Supplemental Figure S1. Colocalization of MP:RFP and AtEB1a:GFP to

microtubule-associated spots.

Supplemental Figure S2. Localization of MP:RFP to microtubule junc-

tions and intersections.

Supplemental Figure S3. Coexpression of AtEB1a:GFP with MAPs.

Supplemental Figure S4. Microtubule and AtEB1a:GFP dynamics in BY-2

cells.

Supplemental Movie S1. AtEB1a:GFP forms comets in N. benthamiana

epidermal cells.

Supplemental Movie S2. Microtubules in AtEB1a:GFP-expressing cells

are dynamic outside TMV-MP:RFP infection site.

Supplemental Movie S3. Microtubules in AtEB1a:GFP-expressing cells

are not dynamic inside TMV-MP:RFP infection site.

Supplemental Movie S4. Nondynamic microtubules in AtEB1a:GFP-

expressing cells at the infection front.

Supplemental Movie S5. Patterns of AtEB1a:GFP in cells near the

infection front.

Supplemental Movie S6. Confocal movie showing the loss of AtEB1a:

GFP dynamics in a cell at the infection front.

Supplemental Movie S7. Dynamic pattern of AtEB1a:GFP in a tua:GFP

transgenic plant.

Supplemental Movie S8. Nondynamic pattern of AtEB1a:GFP and

MP:RFP in cotransfected epidermal cells.

Supplemental Movie S9. Pattern of AtEB1a:GFP in a cell also expressing

RFP:MAP4-MBD.

Supplemental Movie S10. Pattern of AtEB1a:GFP in a cell also transiently

expressing microtubule-stabilizing AtMAP65-5:GFP.

Supplemental Movie S11. AtEB1a:GFP-expressing BY-2 cell protoplast

infected with TMV-MPP81S:GFP.

ACKNOWLEDGMENTS

We thank the Functional Genomics Division of the Department of Plant

Systems Biology at the VIB-Ghent University for providing plasmids

pB7FWG2 and pB7RWG2, Professors Clive Lloyd, Dr. Jordi Chan, the John

Innes Centre, and the Biotechnology and Biological Sciences Research Council

for providing binary vector encoding AtEB1a:GFP, Professor David Gilmer

(IBMP, Strasbourg) for providing antibody against GFP, and Dr. Danny Geelen

(Ghent University) for providing the AtEB1a:GFP-expressing BY-2 cell line as

well as binary plasmids for expression of AtMAP65-5:GFPand G/RFP:MAP4-

MBD. We also are grateful to Richard Wagner and Chantal Fitterer for raising

plants and maintaining BY-2 cultures, Jerome Mutterer for assistance in

fluorescence microscopy, and Antonio Serrato for general technical support.

Received February 8, 2008; accepted April 1, 2008; published April 11, 2008.

LITERATURE CITED

Akhmanova A, Steinmetz MO (2008) Tracking the ends: a dynamic protein

network controls the fate of microtubule tips. Nat Rev Mol Cell Biol 9:

309322Ashby J, Boutant E, Seemanpillai M, Groner A, Sambade A, Ritzenthaler

C, Heinlein M (2006) Tobacco mosaic virus movement protein functions

as a structural microtubule-associated protein. J Virol 80: 83298344Askham JM, Vaughan KT, Goodson HV, Morrison EE (2002) Evidence that

an interaction between EB1 and p150(Glued) is required for the forma-

tion and maintenance of a radial microtubule array anchored at the

centrosome. Mol Biol Cell 13: 36273645Asurmendi S, Berg RH, Koo JC, Beachy RN (2004) Coat protein regulates

formation of replication complexes during Tobacco mosaic virus infection.

Proc Natl Acad Sci USA 101: 14151420

Azoulay J, Clamme JP, Darlix JL, Roques BP, Mely Y (2003) Destabilization

of the HIV-1 complementary sequence of TAR by the nucleocapsid

protein through activation of conformational fluctuations. J Mol Biol

326: 691700

Barth AI, Siemers KA, Nelson WJ (2002) Dissecting interactions between

EB1, microtubules and APC in cortical clusters at the plasma membrane.

J Cell Sci 115: 15831590Bastiaens PI, Pepperkok R (2000) Observing proteins in their natural

habitat: the living cell. Trends Biochem Sci 25: 631637

Bastiaens PI, Squire A (1999) Fluorescence lifetime imaging microscopy:

spatial resolution of biochemical processes in the cell. Trends Cell Biol 9:

4852

Bisgrove SR, Hable WE, Kropf DL (2004) 1TIPs and microtubule regu-

lation. The beginning of the plus end in plants. Plant Physiol 136:

38553863Boyko V, Ashby JA, Suslova E, Ferralli J, Sterthaus O, Deom CM,

Heinlein M (2002) Intramolecular complementing mutations in Tobacco

mosaic virus movement protein confirm a role for microtubule associa-

tion in viral RNA transport. J Virol 76: 39743980Boyko V, Ferralli J, Ashby J, Schellenbaum P, Heinlein M (2000a) Function

of microtubules in intercellular transport of plant virus RNA. Nat Cell

Biol 2: 826832

Boyko V, Ferralli J, Heinlein M (2000b) Cell-to-cell movement of TMVRNA is temperature-dependent and corresponds to the association of

movement protein with microtubules. Plant J 22: 315325

Boyko V, Hu Q, Seemanpillai M, Ashby J, Heinlein M (2007) Validation of

microtubule-associated Tobacco mosaic virus RNA movement and in-

volvementof microtubule-aligned particle trafficking. Plant J 51: 589603Boyko V, van der Laak J, Ferralli J, Suslova E, Kwon MO, Heinlein M

(2000c) Cellular targets of functional and dysfunctional mutants of

Tobacco mosaic virus movement protein fused to GFP. J Virol 74: 11339

11346Browning H, Hackney DD, Nurse P (2003) Targeted movement of cell end

factors in fission yeast. Nat Cell Biol 5: 812818

Chan J, Calder GM, Doonan JH, Lloyd CW (2003) EB1 reveals mobile

microtubule nucleation sites in Arabidopsis. Nat Cell Biol 5: 967971

Citovsky V, Knorr D, Schuster G, Zambryski P (1990) The P30 movement

protein ofTobacco mosaic virus is a single-stranded nucleic acid binding

protein. Cell 60: 637647Clamme JP, Azoulay J, Mely Y (2003) Monitoring of the formation and

dissociation of polyethylenimine/DNA complexes by two photon fluo-

rescence correlation spectroscopy. Biophys J 84: 19601968Curin M, Ojangu EL, Trutnyeva K, Ilau B, Truve E, Waigmann E (2007)

MPB2C, a microtubule-associated plant factor, is required for microtu-

bular accumulation ofTobacco mosaic virus movement protein in plants.

Plant Physiol 143: 801811Dawson WO, Bubrick P, Grantham GL (1988) Modifications of the Tobacco

mosaic virus coat protein gene affecting replication, movement, and

symptomatology. Phytopathology 78: 783789

Deom CM, Oliver MJ, Beachy RN (1987) The 30-kilodalton gene product of

Tobacco mosaic virus potentiates virus movement. Science 237: 384389

Ding B, Haudenshield JS, Hull RJ, Wolf S, Beachy RN, Lucas WJ (1992)

Secondary plasmodesmata are specific sites of localization of the Tobacco

mosaic virus movement protein in transgenic tobacco plants. Plant Cell 4:

915928

Dixit R, Chang E, Cyr R (2006) Establishment of polarity during organi-

zation of the acentrosomal plant cortical microtubule array. Mol Biol

Cell 17: 12981305Dorokhov YL, Alexandrov NM, Miroshnichenko NA, Atabekov JG (1983)

Isolation and analysis of virus-specific ribonucleoprotein of Tobacco

mosaic virus-infected tobacco. Virology 127: 237252

Dorokhov YL, Alexandrova NM, Miroshnichenko NA, Atabekov JG (1984)

The informosome-like virus-specific ribonucleoprotein (vRNP) may be in-

volved in thetransportofTobacco mosaic virus infection. Virology 137: 127134

Ferralli J, Ashby J, Fasler M, Boyko V, Heinlein M (2006) Disruption of

microtubule organization and centrosome function by expression of

Tobacco mosaic virus movement protein. J Virol 80: 58075821Gillespie T, Boevink P, Haupt S, Roberts AG, Toth R, Vantine T, Chapman S,

Oparka KJ (2002) Functional analysis of a DNA shuffled movement

protein reveals that microtubules are dispensable for the cell-to-cell

movement ofTobacco mosaic virus. Plant Cell 14: 12071222

Brandner et al.



13/13

Heinlein M (2002) Plasmodesmata:dynamic regulation and role in macro-

molecular cell-to-cell signalling. Curr Opin Plant Biol 5: 543552Heinlein M, Epel BL (2004) Macromolecular transport and signaling

through plasmodesmata. Int Rev Cytol 235: 93164Heinlein M, Epel BL, Padgett HS, Beachy RN (1995) Interaction of

tobamovirus movement proteins with the plant cytoskeleton. Science270: 19831985

Heinlein M, Padgett HS, Gens JS, Pickard BG, Casper SJ, Epel BL, Beachy

RN (1998a) Changing patterns of localization of the Tobacco mosaic virusmovement protein and replicase to the endoplasmic reticulum and

microtubules during infection. Plant Cell 10: 11071120Heinlein M, Wood MR, Thiel T, Beachy RN (1998b) Targeting and

modification of prokaryotic cell-cell junctions by Tobacco mosaic virus

cell-to-cell movement protein. Plant J 14: 345351

Hepler PK, Palevitz BA, Lancelle SA, McCauley MM, Lichtscheidl I

(1990) Cortical endoplasmic reticulum in plants. J Cell Sci 96: 355373Hink MA, Bisselin T, Visser AJ (2002) Imaging protein-protein interac-

tions in living cells. Plant Mol Biol 50: 871883Holt CA, Beachy RN (1991) In vivo complementation of infectious tran-

scripts from mutant Tobacco mosaic virus cDNAs in transgenic plants.

Virology 181: 109117

Jakobs S, Subramaniam V, Schonle A, Jovin TM, Hell SW (2000) EFGP

and DsRedexpressing cultures of Escherichia coli imaged by confocal, two-

photon and fluorescence lifetime microscopy. FEBS Lett 479: 131135

Kahn TW, Lapidot M, Heinlein M, Reichel C, Cooper B, Gafny R, BeachyRN (1998) Domains of the TMV movement protein involved in subcel-

lular localization. Plant J 15: 1525

Karimi M, Inze D, DepickerA (2002) GATEWAY vectors for Agrobacterium-

mediated plant transformation. Trends Plant Sci 7: 193195

Korinek WS, Copeland MJ, Chaudhuri A, Chant J (2000) Molecular

linkage underlying microtubule orientation toward cortical sites in

yeast. Science 287: 22572259

Kotlizky G, Katz A, van der Laak J, Boyko V, Lapidot M, Beachy RN,

Heinlein M, Epel BL (2001) A dysfunctional movement protein of

Tobacco mosaic virus interferes with targeting of wild type movement

protein to microtubules. Mol Plant Microbe Interact 7: 895904

Kragler F, Curin M, Trutnyeva K, Gansch A, Waigmann E (2003) MPB2C, a

microtubule-associated plant protein binds to and interferes with cell-

to-cell transport ofTobacco mosaic virus movement protein. Plant Physiol

132: 18701883

Lancelle SA, Cresti M, Hepler PK (1987) Ultrastructure of the cytoskeleton

in freeze-substituted pollen tubes of Nicotiana alata. Protoplasma 140:

141150Lansbergen G, Akhmanova A (2006) Microtubule plus end: a hub of

cellular activities. Traffic 7: 499507

Lapidot M, Gafny R, Ding B, Wolf S, Lucas WJ, Beachy RN (1993) A

dysfunctional movement protein of Tobacco mosaic virus that partially

modifies the plasmodesmata and limits spread in transgenic plants.

Plant J 4: 959970Lichtscheidl I, Hepler PK (1996) Endoplasmic reticulum in the cortex o f plant

cells. In M Smallwood, JP Knox, DJ Bowles, eds, Membranes: Special

Functions in Plants. BIOS Scientific Publishers, Oxford, pp 383402Louie RK, Bahmanyar S, Siemers KA, Votin V, Chang P, Stearns T, Nelson

WJ, Barth AI (2004) Adenomatous polyposis coli and EB1 localize in

close proximity of the mother centriole and EB1 is a functional compo-

nent of centrosomes. J Cell Sci 117: 11171128Mas P, Beachy RN (1999) Replication of Tobacco mosaic virus on endoplas-

mic reticulum and role of the cytoskeleton and virus movement in

intracellular distribution of viral RNA. J Cell Biol 147: 945958Mathur J, Mathur N, Kernebeck B, Srinivas BP, Hulskamp M (2003) A

novel localization pattern for an EB1-like protein links microtubule

dynamics to endomembrane organization. Curr Biol 13: 19911997McLean BG, Zupan J, Zambryski PC (1995) Tobacco mosaic virus movement

protein associates with the cytoskeleton in tobacco plants. Plant Cell 7:

21012114Moore P, Fenczik CA, Deom CM, Beachy RN (1992) Developmental

changes in plasmodesmata in transgenic tobacco expressing the move-

ment protein of Tobacco mosaic virus. Protoplasma 170: 115127

Murata T, Sonobe S, Baskin TI, Hyodo S, Hasezawa S, Nagata T, Horio T,

Hasebe M (2005) Microtubule-dependent microtubule nucleation based on

recruitment of gamma-tubulin in higher plants. Nat Cell Biol 7: 961968Nogales E, Whittaker M, Milligan RA, Downing KH (1999) High-resolution

model of the microtubule. Cell 96: 7088Oparka KJ, Prior DAM, Santa Cruz S, Padgett HS, Beachy RN (1997)

Gating of epidermal plasmodesmata is restricted to the leading edge of

expanding infection sites of Tobacco mosaic virus. Plant J 12: 781789

Rehberg M, Graf R (2002) Dictyostelium EB1 is a genuine centrosomal

component required for proper spindle formation. Mol Biol Cell 13:23012310Reichel C, Beachy RN (1998) Tobacco mosaic virus infection induces severe

morphological changes of the endoplasmic reticulum. Proc Natl Acad

Sci USA 95: 1116911174

Reilein A, Nelson WJ (2005) APC is a component of an organizing template

for cortical microtubule networks. Nat Cell Biol 7: 463473

Rogers SL, Rogers GC, Sharp DJ, Vale RD (2002) Drosophila EB1 is

important for proper assembly, dynamics, and positioning of the mitotic

spindle. J Cell Biol 158: 873884Seemanpillai M, Elamawi R, Ritzenthaler C, Heinlein M (2006) Chal-

lenging the role of microtubules in Tobacco mosaic virus movement by

drug treatments is disputable. J Virol 80: 67126715

Treanor B, Lanigan PM, Suhling K, Schreiber T, Munro I, Neil MA,

Phillips D, Davis DM, French PM (2005) Imaging fluorescence lifetime

heterogeneity applied to GFP-tagged MHC protein at an immunological

synapse. J Microsc 217: 3643Van Damme D, Van Poucke K, Boutant E, Ritzenthaler C, Inze D, Geelen

D (2004) In vivo dynamics and differential microtubule-binding activ-

ities of MAP65 proteins. Plant Physiol 136: 39563967Vaughan KT (2005) TIP maker and TIP marker; EB1 as a master controller

of microtubule plus ends. J Cell Biol 171: 197200

Vogler H, Kwon MO, Dang V, Sambade A, Fasler M, Ashby J, Heinlein M

(2008) Tobacco mosaic virus movement protein enhances the spread of

RNA silencing. PLoS Pathog 4: e1000038Wright KM, Wood NT, Roberts AG, Chapman S, Boevink P, Mackenzie

KM, Oparka KJ (2007) Targeting of TMV movement protein to plasmo-

desmata requires the actin/ER network; evidence from FRAP. Traffic 8:

2131



EtudiantArtcile2

Documents