Plasmodesmata and the control of symplastic transport

Plant, Cell and Environment

(2003)

26,

103–124

© 2003 Blackwell Publishing Ltd

103

Blackwell Science, LtdOxford, UKPCEPlant, Cell and Environment0016-8025Blackwell Publishing Ltd 2002

26Original Article

Plasmodesmata and symplastic transportA. G. Roberts & K. J. Oparka

Correspondence: Alison G. Roberts. e–mail: [email protected]

Plasmodesmata and the control of symplastic transport

A. G. ROBERTS & K. J. OPARKA

Scottish Crop Research Institute, Invergowrie, Dundee, DD2 5DA, UK

‘There are holes in the skywhere the rain gets in,but they’re ever so smallthat’s why rain’s so thin’

Spike Milligan (1968)

In 1879 Eduard Tangle discovered cytoplasmic connectionsbetween cells in the cotyledons of

Strychnos nuxvomica

,which he interpreted to be protoplasmic contacts. This ledhim to hypothesize that ‘the protoplasmic bodies . . . areunited by thin strands passing through connecting ducts inthe walls, which put the cells into connection with eachother and so unite them to an entity of higher order’ (Carr1976). This challenged the then current view that cells func-tioned as autonomous units. It was after much research inmany other species and cell types that Strasburger, in 1901,named these structures plasmodesmata (Carr 1976).

During the division and differentiation of meristematiccells, plasmodesmata are formed across each developingcell plate, allowing cytoplasmic and endomembrane conti-nuity to occur between all daughter cells, and ultimately,between all cells in a developing tissue (Mezitt & Lucas1996). Those plasmodesmata that form during cell divisionare termed primary plasmodesmata (Jones 1976). Thosethat form

de novo

across existing cell walls are called sec-ondary plasmodesmata (Ehlers & Kollmann 2001). The for-mation of secondary plasmodesmata allows cells to increasetheir potential for molecular trafficking and allows connec-tions to be created between cells that are not relatedcytokinetically.

As cells expand and differentiate, their fate determinesthe extent to which their cytoplasmic connectivity to othercells is maintained (Mezitt & Lucas 1996). Some cell types,such as those of the leaf mesophyll, remain closely con-nected to their neighbours, and may even lay down addi-tional plasmodesmata to increase the continuity (Ding

et al

.1992a). In other areas of the plant, for instance in vasculartissue, certain cells greatly reduce the number of plas-modesmata in their adjoining walls (Gamalei 1989). In thisway, the cytoplasmic continuity can be altered dependingon the tissue type (Botha & Evert 1988; Brown

et al

. 1995).However, although reductions in the number of plasmodes-mata are common, only guard cells surrounding stomata(Erwee, Goodwin & van Bel 1985; Palevitz & Hepler 1985)and differentiating xylem elements (Lachaud & Maurous-set 1996) lose all symplastic connections at maturity. In all

other cells, some degree of intercellular connection is main-tained. This plasmodesmal continuum that potentiallyexists throughout the whole plant is termed the symplast(Münch 1930). However, the symplast is not the open con-tinuum that Münch originally hypothesized, but is dividedinto functional domains, each tightly regulated by differentforms of plasmodesmata (Erwee & Goodwin 1985; Ehlers& Kollmann 2001). Plasmodesmata are now thought of asfluid, dynamic structures that can be modified both struc-turally and functionally to cope with the requirements ofspecific cells and tissues.

THE STRUCTURE OF PLASMODESMATA

Based on structure, two basic types of plasmodesmata havebeen characterized; simple and branched. Simple plas-modesmata consist of a single pore traversing the cell wall,whereas branched plasmodesmata have two or more chan-nels on either side of the middle lamella, often joined by acentral cavity. Electron micrographs of simple plasmodes-mata are shown in Fig. 1A and B, and in diagrammatic formin Fig. 2. Simple plasmodesmata may be grouped togetherin primary pit fields and this form has been found to pre-dominate in immature plant tissues (Oparka

et al

. 1999).Simple plasmodesmata are common in the lower plantssuch as algae and mosses (Franceschi, Ding & Lucas 1994;Cook

et al

. 1997) whereas branched plasmodesmata arefound extensively in mature tissues and appear to representmore evolutionarily advanced structures (Lucas, Ding &van der Schoot 1993; Oparka

et al

. 1999). Both primary andsecondary plasmodesmata are initially simple in structurebut, during tissue development, may intrinsically formbranches or fuse with neighbouring plasmodesmata to pro-duce complex, branched structures. During the sink–sourcetransition that occurs in leaves, simple plasmodesmata aregradually converted to branched plasmodesmata by amechanism that involves the formation of bridges betweenadjoining simple pores (Fig. 2; see Oparka

et al

. 1999;Roberts

et al

. 2001).Due to the difficulty of isolating intact plasmodesmata

from cell-wall fractions, structural models have tradition-ally been based on data from transmission electron micro-graphs (Robards 1971; Robards & Lucas 1990; Beebe &Turgeon 1991; Ding, Turgeon & Parthasarathy 1992b;Botha, Hartley & Cross 1993; Overall & Blackman 1996;Waigmann

et al

. 1997). Most models depict simple plas-modesmata as linear pores in the cell wall (Figs 1C & D),each lined by the plasma membrane, which is continuous

104

A. G. Roberts & K. J. Oparka


,


,

26

, 103–124

between adjacent cells. In the centre of the pore lies astrand of modified cortical endoplasmic reticulum, the des-motubule (Robards & Lucas 1990; Ding

et al

. 1992b; Lucas& Wolf 1993; Epel 1994). Different structural models of thedesmotubule have been postulated. Gunning & Overall(1983) showed it to have a cylindrical, membranous struc-

ture with an internal lumen, whereas Tilney

et al

. (1991)speculated that it exists as a solid, proteinaceous rod. Thedesmotubule has been found in both an appressed anddilated state (Overall, Wolfe & Gunning 1982; Robinson-Beers & Evert 1991; Waigmann

et al

. 1997).It is thought that most endogenous cytoplasmic mole-

Figure 1.

(A) Longitudinal section through simple plasmodesmata between adjacent vascular parenchyma cells in a mature sugarcane leaf. These plasmodesmata show constricted desmotubules and neck constrictions lacking sphincters. ER, endoplasmic reticulum. (Reproduced with permission from R.F. Evert and

Planta

; Robinson-Beers & Evert, 1991) Bar

=

200 nm. (B) Transverse section through simple plas-modesmata between adjacent vascular parenchyma cells in a mature sugarcane leaf. The inner leaflet of the desmotubule appears as a central, electron-opaque dot. The cytoplasmic sleeve appears mottled due to the presence of electron-opaque, spoke-like extensions that extend from the outer leaflet of the desmotubule toward the inner leaflet of the plasma membrane. (Reproduced with permission from R.F.Evert and

Planta

; Robinson-Beers & Evert, 1991) Bar

=

100 nm. (C) and (D) Diagrammatic representation of the substructure of simple plasmodesmata in longitudinal (C) and transverse (D) sections. ER, endoplasmic reticulum; CW, cell wall; CS, cytoplasmic sleeve; D, desmotubule; CR, central rod; PM, plasma membrane; SP, spoke-like extensions; PMP, plasma membrane-embedded proteins; DP, desmo-tubule-embedded proteins. (Based on model in Ding

et al

. 1992b).

Plasmodesmata and symplastic transport

105


,


,

26

, 103–124

cules, and plant viruses, utilize the cytoplasm between thedesmotubule and plasma membrane, a region known as thecytoplasmic sleeve (Esau & Thorsch 1985), to move fromcell to cell (Lucas & Wolf 1993; Epel 1994; Kragler, Lucas& Monzer 1998a). However, trafficking between cells viathe desmotubule has also been reported (Cantrill, Overall& Goodwin 1999). At both ends of the plasmodesma, inthe neck region, the channel is frequently constricted(Fig. 1A), and the cytoplasmic sleeve may be partiallyoccluded with globular subunits. The subunits are thoughtto exist in a helical arrangement that may functionallydivide the space into a number of spiralling channels (Zee1969; Robards 1976; Olesen 1979; Overall

et al

. 1982; Wolf

et al

. 1989; Olesen & Robards 1990; Robards & Lucas 1990;Ding, Turgeon & Parthasarathy 1991; Robinson-Beers &Evert 1991; Lucas & Wolf 1993; Lucas

et al

. 1993; Overall& Blackman 1996; Waigmann

et al

. 1997). This arrange-ment appears to reduce the functional diameter of thecytoplasmic sleeve, producing microchannels that arebetween 3 and 4 nm in diameter (Lucas

et al

. 1993; Fisher1999). It is thought that these proteins may be part of animportant regulatory structure that can alter the size exclu-sion limit (SEL) of plasmodesmata, and which viruses mayexploit to modify plasmodesmata in order to move fromcell to cell (Evert, Eschrich & Heyser 1977; Carrington

et al

. 1996). Electron micrographs have also shown an elec-tron-opaque ring of material surrounding the neck of theplasmodesma, the wall ‘collar’ or ‘sphincter’, that mayfunction to alter the SEL of the pore (Olesen 1979; Olesen& Robards 1990; Beebe & Turgeon 1991; Badelt

et al

. 1994;Turner, Wells & Roberts 1994). While the structure of mostplasmodesmata conforms to this general model, a rangeof substructural variations have been found in differentplant species and within tissues of the same plant

(Robinson-Beers & Evert 1991; Waigmann

et al

. 1997).Very little is currently known about the proteins that con-tribute to the structure of plasmodesmata, and many, ifnot most, plasmodesmal proteins await isolation andfunctional characterization.

TRANSPORT VIA THE CYTOPLASMIC SLEEVE

Transport of many small molecules through plasmodesmataoccurs via the cytoplasmic sleeve. Passively loaded fluores-cent probes (Duckett

et al

. 1994; Roberts

et al

. 1997) andmicroinjected dyes (Goodwin 1983; Erwee

et al

. 1985;Madore & Lucas 1986; Oparka

et al

. 1991) utilize this path-way, and larger molecules, including green fluorescent pro-tein (GFP; see below) may also move via the cytoplasmicsleeve. Barclay, Peterson & Tyree (1982) showed that dyemovement was not disrupted when cytoplasmic streamingwas inhibited by treatment with cytochalasin B, althoughplasmodesmal disruption by plasmolysis did cause a much-reduced rate of fluorescein movement. Tucker (1987) alsonoted that cytoplasmic streaming has no effect on the inter-cellular movement of molecules. These data suggest thatthe movement of small molecules occurs by diffusion, andthat movement through plasmodesmata is the rate-limitingstep for intercellular transport (Barclay

et al

. 1982; Tucker,Mauzerall & Tucker 1989).

TRANSPORT VIA THE ENDOMEMBRANE SYSTEM

Many electron micrographs show that the cortical endo-plasmic reticulum (ER) is closely associated with plas-modesmata, and the desmotubule provides a potentialpathway for movement between cells. Different structures

Figure 2.

Diagrammatic representation of the structural change from simple to branched plasmodesmata. During development, the simple pore develops into a complex, branched structure with a central cavity aligned along the middle lamella of the cell wall. This transformation occurs via an H-shaped intermediate that appears to form by the introduction of a new protoplasmic bridge between neighbouring pairs of simple plasmodesmata. WC, wall collar; CW, cell wall; ML, middle lamella; D, desmotubule; PM, plasma membrane; ER, endoplasmic reticulum; CC, central cavity. (Re-drawn from Oparka

et al

. 1999)

106



,


,

26

, 103–124

of the desmotubule have been postulated. Although Tilney

et al

. (1991) speculated that the desmotubule might exist asa solid proteinaceous rod, most models consider it to be acylindrical, membranous structure with an internal lumen(Gunning & Overall 1983). Grabski, de Feijter & Schindler(1993) showed that fluorescent lipid and phospholipid ana-logues could diffuse between cells and redistribute follow-ing photobleaching if these lipid-based probes were locatedin the ER. However, they were unable to move betweencells when targeted to the plasma membrane. This suggeststhat at least some lipid-based molecules may diffusebetween cells along the membranes that comprise the des-motubule. Recently, Cantrill

et al

. (1999) showed that ifsmall dyes were microinjected directly into the cortical ERof a single cell they were capable of movement into adjoin-ing cells. Such movement can only occur through the des-motubule. Trafficking via the ER/desmotubule couldpotentially occur in three different ways: (1) by passive flowwithin the desmotubule lumen; (2) by diffusion along theinner desmotubule membranes; or (3) by specific attach-ment of molecules to the cytoplasmic face of the desmotu-bule followed by facilitated transport through thecytoplasmic sleeve. This ER-mediated mode of transportmay be particularly important at plasmodesmata betweensieve element and companion cells, as there is virtually nocytoplasm in sieve tubes, but the desmotubule provides acontinuum between the parietal ER of the sieve elementand the normal, cortical ER of the companion cell (Black-man

et al

. 1998; Oparka & Santa Cruz 2000; van Bel &Knoblauch 2000).

THE PLASMODESMAL SEL

If molecules are smaller than the basal SEL of the plas-modesma, intercellular transport appears to occur by diffu-sion. The basal SEL is defined as being the ‘natural’ SELof plasmodesmata in mature tissues that have been unmod-ified by, for example, movement proteins, chemicals orenvironmental stresses that increase the SEL. Sugars,metabolites, ions and amino acids are all thought to moveby diffusion through the cytoplasmic sleeve of plasmodes-mata (see Lucas

et al

. 1993). Molecules larger than the basalSEL require selective transport, in which case conforma-tional changes must occur in the plasmodesmal pore. Suchchanges are not necessarily permanent structural modifica-tions as plasmodesmata are capable of transient increasesin their SEL (Schulz 1999). It is now apparent that the SELvaries throughout plant development, within different tis-sues of an organ, and between different plant species. TheSEL can also be modified by interactions with a number ofendogenous proteins, viral movement proteins (MP) andother endogenous molecules. Thus, although many earlyreports suggested that the ‘basal’ SEL of plasmodesmatawas between 850 and 900 Da (reviewed in Lucas

et al

.1993), and some cell types (e.g. epidermal cells) can have aSEL less than 370

M

r

(Erwee & Goodwin 1985; Duckett

et al

. 1994), it now appears that for many cells the SEL isconsiderably higher than these values.

Ultimately, the factor that determines the diffusive per-meability of a molecule through plasmodesmata is theStokes radius (

R

S

). Terry & Robards (1987) showed thatthe rate of diffusion is directly correlated to the radius ofthe permeant molecule, and that small changes in theStokes radius could cause large differences in the mobilitiesof molecular probes. Based on these experiments, it wasproposed that the diameter of individual channels in thecytoplasmic sleeve of a plasmodesma is approximately3 nm. This is a figure that corresponds well to estimatesfrom high resolution electron micrographs (Ding

et al

.1992b) and dye-coupling studies (Tucker 1982; Erwee &Goodwin 1983, 1984). However, a more recent recalcula-tion of the Terry & Robards (1987) data has shown that theabove figure may be closer to 4 nm, a value that makes aconsiderable difference to the potential SEL of the pore(Fisher 1999). For instance, an increase in diameter from 3to 4 nm may alter the maximum

R

S

of dextrans that canpass through the plasmodesmal pore from approximately6 nm to 10 nm, and increase the maximum

R

S

of proteinsfrom approximately 4 nm to 14 nm (data derived from Jør-gensen & Møller 1979; le Maire

et al

. 1986).

REGULATION OF TRANSPORT

Trafficking of molecules through plasmodesmata may benon-specific, as described above, or may require a specificinteraction between a structural motif or sequence-specificelement of the transported protein with other proteinsclose to, or within, plasmodesmata (Itaya

et al

. 2000; Craw-ford & Zambryski 2000). Both endogenous plant proteinsand viral proteins can effect specific trafficking (Mezitt &Lucas 1996; Ghoshroy

et al

. 1997; McLean, Hempel &Zambryski 1997). Non-specific trafficking is influenced byplant development (Oparka

et al

. 1999) and also by a num-ber of cellular factors. For example, GFP can traffic non-specifically through plasmodesmata in sink-leaf tissues oftobacco, where plasmodesmata are predominantly simplein structure and have a relatively high SEL (Imlau, Truernit& Sauer 1999; Oparka

et al

. 1999). However, movement ofGFP is restricted in mature source-leaf tissues of tobacco,where plasmodesmata are branched and have a low SEL(Oparka

et al

. 1999; Roberts

et al

. 2001). A reduction in theSEL of plasmodesmata during maturation of a tissue seemsto be common. For example, symplastic communication isextensive in young embryos, but this changes with develop-ment, leading to increased symplastic isolation as cells andtissues mature (Rinne & van der Schoot 1998; Gisel

et al

.1999). In

Arabidopsis

roots, undifferentiated epidermalcells in the meristem and elongation zone are extensivelyconnected (as measured by dye coupling), but become lessso as they mature, until in the mature root the epidermisand root hairs are completely isolated from underlying celllayers (Duckett

et al

. 1994). Similarly, Cantrill, Overall, &Goodwin (2001) found that the symplastic permeability ofcells in tissue culture was high during early regenerationevents, but decreased as shoot formation was initiated.Studies of GFP trafficking have shown that plasmodesmata

Plasmodesmata and symplastic transport

107


,


,

26

, 103–124

are differentially permeable in different tissues or organsand in different plant species. GFP was shown to trafficbetween cells in

Arabidopsis

leaf and stem epidermis at alldevelopmental stages and yet was unable to move in theepidermis of either tomato or cucumber (Itaya

et al

. 2000).Movement between cell layers also appears to be regulateddifferently from movement within a cell layer. GFP couldnot pass from the epidermis into the mesophyll of cucum-ber cotyledons although it could move from the epidermisinto the cortex of hypocotyls (Itaya

et al

. 2000). Duringmorphogenesis of protoplast-derived calluses, some plas-modesmata were found to be blocked by a dense, osmio-philic material that produced numerous symplast domainsof varying size within the developing tissue (Ehlers,Binding & Kollmann 1999. The authors noted that theblockages were always produced by the surrounding cellsand not within the isolated domain, and suggested thatdemarcation of symplastic domains might be a general pre-requisite for differential morphogenesis. Plasmodesmalregulation may therefore use a number of distinct mecha-nisms, depending on the developmental stage, tissue andspecies of plant, the structure of plasmodesmata present,and the presence of modifying factors such as endogenousor viral MPs.

THE CYTOSKELETON AND PLASMODESMATA

Both actin and myosin have been localized to the plas-modesmal pore (White

et al

. 1994; Blackman & Overall1998; Radford & White 1998). Some authors have sug-gested that this distribution may allow constriction andrelaxation of the entire length of the pore (Zambryski &Crawford 2000), whereas others have suggested that thesecytoskeletal elements are involved in transport

per se

,rather than having a role in regulating the permeability ofthe plasmodesma (Blackman & Overall 2001). The latterpoint is supported by the fact that inhibitors of actin andmyosin do not appear to influence the movement of smallmolecules that pass through plasmodesmata by diffusion,indicating that actin and myosin are not involved in con-trolling the permeability of plasmodesmata, at least at lowSELs (Tucker 1987). In contrast, complete depolymeriza-tion of the actin cytoskeleton by cytochalasin resulted in awidening of the neck region of plasmodesmata in

Nephrol-epis exaltata

(White

et al

. 1994), and caused an increase inthe SEL of tobacco plasmodesmata from 1 kDa to over20 kDa (Ding, Kwon & Warnberg 1996). However, whenactin microfilaments were stabilized by treatment withphalloidin, cell–cell transport was inhibited. Additionalstudies, in which a GFP-talin fusion was biolistically bom-barded into a single cell, showed that the actin microfila-ments which became labelled by GFP-talin did notthemselves move between cells (Crawford & Zambryski2000). These results suggest that actin filaments may forma static scaffold within plasmodesmata, along which mole-cules move using a myosin-driven mechanism. The cytosk-eleton has also been hypothesized to be important inmaintaining an extensive communication pathway in the

long-lived ray and parenchyma cells of tree species(Chaffey & Barlow 2001).

The ER is known to be closely associated with actinfilaments (Quader, Hofmann & Schnepf 1987; Boevink

et al

. 1998), and the highly motile cortical ER has beenpostulated to move along actin cables (Quader

et al

. 1987),possibly using myosin or other motor proteins that link thecytoskeleton to the endomembrane system. Fluorescentlipid analogues have been shown to move between cells inthe ER membrane but not in the plasma membrane (Grab-ski

et al

. 1993). This suggests that the ER may ‘flow’through plasmodesmata whereas the plasma membrane isa more stationary structure. If intrinsic ER proteins canmove between cells in this way, it may be possible thatother molecules could ‘hitch a ride’ on the ER. Moleculesinserted into, or temporarily attached to, the ‘flowing’ ERmembrane could conceivably move through plasmodes-mata via the desmotubule. In addition, if myosin motorsare required for movement along actin cables, this couldgive rise to polarity of ER-based movement, as all myosinsmove uni-directionally towards either the pointed (–) orthe barbed (

+

) end of the actin filament (Wells

et al

. 1999).Such behaviour could help to explain the apparent direc-tionality of movement sometimes observed through plas-modesmata. In

Nicotiana clevelandii

trichome cells,movement of low molecular weight dyes and dextransoccurred preferentially towards the tip of the trichome(Waigmann & Zambryski 1995), although this could bedue to an increase in simple plasmodesmata towards thetrichome tip (Waigmann

et al

. 1997). In

Chara

cells, asimilar linear system, the intercellular transport ofphoto-assimilate also shows directionality (Ding

et al

.1991). Tightly controlled and directional movement oftranscription factors has also been discovered, even intissues with a high SEL (see later).

The presence of both actin and myosin along the lengthof the pore, possibly as helically arranged ‘spokes’ thatconnect the desmotubule to the plasma membrane (Over-all & Blackman 1996), provides a possible contractilemechanism for controlling the aperture of the cytoplasmicsleeve (Radford & White 1998; Reichelt

et al

. 1999).Myosin VIII, an unconventional myosin found only inplants, has also been localized to plasmodesmata (Reichelt

et al

. 1999) and its activity could be regulated by calcium(Knight & Kendrick-Jones 1993). At present, seven puta-tive myosin VIII proteins have been reported (Baluska

et al

. 2001). All of these have a characteristic C-terminalstructure that includes a probable phosphorylation site forprotein kinases, as well as four calmodulin-binding motifs(Reichelt & Kendrick-Jones 2000). Myosin VIII thereforeemerges as a good candidate for a plasmodesmal motorprotein whose activity may be regulated by calcium orcalmodulin.

Plasmodesmal regulation may occur along the full lengthof the plasmodesmata (as suggested by the presence ofboth actin and myosin along the length of the pore), butregulation could also occur only at the neck regions, pro-ducing the same effect. Centrin, a calcium-binding contrac-

108



,


,

26

, 103–124

tile protein, has been localized to the neck region ofplasmodesmata (Blackman, Harper & Overall 1999) andcould regulate the plasmodesmal aperture. An increase inthe concentration of cytoplasmic calcium causes a decreasein the phosphorylation of this protein (Martindale &Salisbury 1990). This dephosphorylation causes the centrinnanofilaments to contract rapidly, potentially closing theplasmodesma (Martindale & Salisbury 1990; Blackman

et al

. 1999). Additional data to support this hypothesisinclude evidence that increased levels of calcium lead toplasmodesmal closure (Erwee & Goodwin 1983; Tucker1990; Holdaway-Clarke

et al

. 2000), and the fact that twoprotein kinases, one calcium-dependent, have been local-ized to plant cell walls or plasmodesmata (Citovsky

et al

.1993; Yahalom

et al

. 1998). In addition, both ATPase activ-ity and calcium-binding sites have been localized to plas-modesmata in barley roots (Belitser, Zaalishvili &Sytnianskaja 1982) and the calcium-sequestering protein,calreticulin, has also been localized to plasmodesmata(Baluska

et al

. 1999, 2001). Calreticulin is normally foundin the lumen of the ER but has calcium-buffering activitythat could potentially regulate plasmodesmal transport.However, although alterations in cytoplasmic calcium havebeen reported to increase the electrical resistance of plas-modesmata, cytoplasmic acidification with butyric aciddoubled the calcium concentration and yet did not affectplasmodesmal resistance (Holdaway-Clarke

et al

. 2001).Citovsky

et al

. (1993) showed that a cell-wall-associatedkinase was developmentally regulated, and its activity wascorrelated with the developmental maturation of plas-modesmata. This kinase was also able to phosphorylate the30 kDa MP of

Tobacco mosaic virus

(TMV), a proteinknown to be able to interact with, and functionally alter,plasmodesmata (Citovsky

et al

. 1993; see below). Further-more, phosphorylation of this MP led to altered plas-modesmal permeability properties (Waigmann

et al

.2000), and may be a common mechanism to control viralmovement, as the movement proteins of

Potato leafrollvirus

(PLRV; Sokolova

et al

. 1997; T

omato mosaic virus

(Matsushita

et al

. 2000), and C

ucumber mosaicvirus

(CMV; Matsushita

et al

. 2002b) have all been foundto be phosphorylated. The presence of phosphorylationsites at, or near, plasmodesmata strongly suggests thatmodification of plasmodesmal-specific, contractile proteinsby either phosphatases or kinases may play an importantrole in plasmodesmal control.

TURGOR AND PRESSURE EFFECTS

Plasmodesmata are known to alter their permeability inresponse to turgor changes. In contrast to the fine regula-tion of the plasmodesmal SEL imposed by cytoskeletal ele-ments, it seems likely that responses to turgor changes aredesigned to cope with physiological traumas that arise as aresult of stress or wounding. In

Nicotiana clevelandii

tri-chome cells, pressure differentials of more than 200 kPabetween adjacent cells prevented the transport of microin-jected Lucifer yellow dye (Oparka & Prior 1992). How-

ever, at lower pressure differentials, dye continued to movebetween neighbouring cells. The closure of plasmodesmataoccurred relatively slowly, over a period of at least 10 min,and the effect lasted for some time, possibly permanently.In the alga,

Chara corallina

, pressure differences alsocaused plasmodesmata to seal off, preventing further inter-cellular exchange (Ding & Tazawa 1989; Reid & Overall1992). In contrast, an increase in the turgor of

Arabidopsis

root hairs by microinjection of oil droplets did not cause analteration in the electrical coupling of cells, apparentlyindicating that plasmodesmata did not close completely(Lew 1996). Increasing osmotic tissue stress has also beenshown to cause changes in plasmodesmatal permeability.Schulz (1995) showed that increasing mannitol concentra-tions caused the usually constricted neck regions of plas-modesmata to widen to approximately the same diameteras the central portion of the pore. In parallel to the plas-modesmal widening, these osmotically stressed root tipsshowed increased phloem unloading. In

Egeria densa

leaves, the SEL was also increased for a period of 20 hfollowing plasmolysis (Erwee & Goodwin 1984). Together,these results suggest that a sudden pressure differentialbetween adjacent cells, as might occur during wounding,will cause isolation of plant cells from their neighbours,whereas an overall drop in tissue turgor, characteristic ofwater stress (hyperosmotic stress), will lead to an increasedSEL. In the latter case, the increase in plasmodesmalpermeability may be associated with the plant’s need tosuddenly increase solute fluxes into organs undergoingosmotic stress.

Other physiological perturbations can also alter plas-modesmal SELs. In a study of roots under anaerobic stress,reduced levels of ATP caused the plasmodesmal SEL toincrease from less than 1 kDa in controls, to between 5 and10 kDa in stressed roots (Cleland, Fujiwara & Lucas 1994).These authors suggested that ATP was required to maintainplasmodesmata in a constricted state. However, althoughhypoxia has been shown to depolarize the membranepotential of wheat root cells, it had little effect on the mem-brane resistance or the electrical coupling ratio, which is ameasure of plasmodesmal resistance (Zhang & Tyerman1997). These authors suggested that their results pointed tothe possibility that transport of water and solutes couldoccur via both the cytoplasm and the ER, and that only theER pathway is interrupted by hypoxia. Furthermore, theypostulated that hypoxia may cause the desmotubule to con-dense, which would in turn enlarge the cytoplasmic annulus,and might explain the increased SEL found by Clelandet al. (1994) in anaerobically stressed wheat roots (Zhang& Tyerman 1997. A range of metabolic inhibitors have alsobeen shown to increase symplastic movement out of thetransport phloem in Arabidopsis roots (Wright & Oparka1997), whereas increased cytoplasmic concentrations ofgroup II ions decreases plasmodesmal permeability (Erwee& Goodwin 1983). Simply changing plant growth condi-tions has been reported to alter plasmodesmal permeabil-ity; plants grown in a glasshouse exhibited increasedmovement of GFP compared with plants grown in tissue-

Plasmodesmata and symplastic transport 109

© 2003 Blackwell Publishing Ltd, Plant, Cell and Environment, 26, 103–124

culture containers (Crawford & Zambryski 2000). In short,plasmodesmata appear to be extremely sensitive to a rangeof physiological perturbations, a fact that underlines theircentral importance in regulating exchange between cellsand tissues.

CALLOSE

Callose ((1Æ3)-b-glucan) deposition at plasmodesmata hasalso been postulated to control intercellular exchange,although the rates of deposition are generally slow, sug-gesting that this mechanism may be used more in responseto wounding or pathogenesis than as a rapid control mech-anism. However, callose plugs at plasmodesmata haverecently been implicated in the maintenance of dormancyby symplastically isolating the meristem from surroundingtissues (Rinne & van der Schoot 1998; Rinne, Kaikuranta& van der Schoot 2001). During wounding, callose isdeposited at plasmodesmata (Hughes & Gunning 1980),but callose deposition and degradation have been shown tobe variable both in time and in quantity. During theresponse of oat coleoptiles to plasmolysis, callose wasdeposited at plasmodesmata, but most had disappeared 4–6 h after full turgor was regained. Recovery of electricalcoupling in these cells followed the time course of calloseremoval (Drake, Carr & Anderson 1978), taking severalhours to resume. In contrast, in Egeria densa, electricalcoupling was re-established within 10 min of deplasmolysis(Erwee & Goodwin 1984). Callose deposition at the neckof plasmodesmata has also been linked to alterations in thehormonal balance or the osmotic potential of plant mate-rial, and plasmodesmata with callose collars were observedto also contain electron dense material in the cytoplasmicsleeve which apparently obscured the desmotubule (Botha& Cross 2000). Wolf et al. (1991) showed that a reductionin the SEL, and an inhibition of dye movement in trans-genic tobacco plants expressing the movement protein oftobacco mosaic virus, could be alleviated by treatment ofthe tissue with an inhibitor of callose synthesis, providingevidence that callose deposition is involved in regulatingthe plasmodesmal SEL. Dye coupling has also been shownto be reduced in cells adjacent to TMV infection sites, inareas known to have callose deposits at plasmodesmata(Susi 2000; see below). Enhanced callose deposition in a b-1,3-glucanase-deficient mutant also caused a reduction inthe SEL (Iglesias & Meins 2000) but not an absolute clo-sure of the plasmodesmata. It is interesting to note that,although the effects of callose are generally thought to becoarser than other control mechanisms, many recent pub-lications have shown that callose can regulate the SEL,rather than simply constricting or closing the plasmodes-mal pore. Additional experiments have shown that callosemay be deposited only in response to specific stresses. Cal-lose deposition is known to be induced by micromolarchanges in intercellular calcium concentrations (Kauss1987). Induction of stress by application of both aluminiumor hydrogen peroxide caused an increase in cytoplasmiccalcium concentration and yet, callose deposition occurred

only in response to aluminium, and not to peroxide appli-cation (Robards & Lucas 1990; Jones et al. 1998), suggest-ing that callose deposition at plasmodesmata may be asurvival mechanism for aluminium toxicity in plants(Sivaguru et al. 2000). Although some callose is alwaysfound in sieve pores as a remnant of sieve-pore genesis,phloem transport can also occur in the presence of largecallose deposits across sieve plates (Peterson & Rauser1979). However, phloem loading in the sucrose export defi-cient (sxd1) maize mutant is prevented by callose depositswhich specifically block plasmodesmata at the interfacebetween bundle sheath and vascular parenchyma cells(Botha et al. 2000).

During pathogenesis, callose has been found at plas-modesmata, suggesting that callose deposition is a ubiqui-tous plant response to pathogen spread. Most publishedreports have concentrated on viral infections, but callosehas also been found to block plasmodesmata during infec-tion by the oomycete Phytophthora sojae (Enkerli, Hahn& Mims 1997). Callose has been found in pit fields andalong the surrounding walls of plants infected with Potatovirus X (PVX; Allison & Shalla 1974). In these plants,callose deposition was extensive within necrotic virallesions, but could also be detected outside the visiblelesion, suggesting that formation of callose collars at plas-modesmata could be an early defence strategy against thisvirus. Callose deposition has also been reported at plas-modesmata in plants infected with many other viruses,including TMV (Wu & Dimitman 1970; Moore & Stone1972; Leisner & Turgeon 1993; Beffa et al. 1996), Tomatobushy stunt virus (Pennazio et al. 1978), and Maize dwarfmosaic virus (Choi 1999). In addition, plants deficient inb-1,3-glucanases (callose-degrading enzymes) showed adecreased susceptibility to infection with TMV (Beffa et al.1996; Iglesias & Meins 2000) and to PVX and CMV (Igle-sias & Meins 2000). These plants also showed increasedcallose deposition at plasmodesmata in response to viralinfection, and a reduction in the plasmodesmal SEL.Complementary experiments showed that TMV mutantsthat overexpressed the b-1,3-glucanase gene displayedincreased movement and spread on either wild-type or b-1,3-glucanase-deficient plants (Bucher et al. 2001). Moore& Stone (1972) also showed that levels of b-1,3-glucanhydrolase were increased in leaves infected with a numberof other plant viruses. Collectively, these authors havesuggested that induction of b-1,3-glucanases might be astrategy used by viruses to enhance their cell-to-cell move-ment and spread (Moore & Stone 1972; Beffa et al. 1996;Iglesias & Meins 2000).

LOSS OR REDUCTION OF PLASMODESMATA

In addition to changes in the structure of plasmodesmata,their frequency can also be developmentally altered in dif-ferent tissue types throughout the plant. Plants are capableof sealing off, or removing plasmodesmata both tempo-rarily and permanently. Erwee & Goodwin (1985) injecteda range of differently sized fluorescent dyes into Egeria

110 A. G. Roberts & K. J. Oparka


densa and showed that the SEL varied between tissues. Thisproved that the symplast is not a cellular continuum withunlimited communication, but is subdivided into functionaldomains. These domains, and the limits they place on themovement of signal molecules, are thought to allow co-ordinated mitosis, morphogenesis and development inplants (Kragler et al. 1998a; Ehlers & Kollmann 2000;Pfluger & Zambryski 2001).

As guard cells differentiate, cell-wall material is depos-ited across the pore on the side of the neighbouring epider-mal or subsidiary cell wall, effectively truncatingplasmodesmata and symplastically isolating the guard cellcomplex (Wille & Lucas 1984; Palevitz & Hepler 1985). Indifferentiating xylem tissue of Sorbus torminalis, manyplasmodesmata are found in the pits that connect imma-ture xylem elements to the surrounding mesophyll cells.However, during the final stages of programmed cell deaththe pits become sealed off by the deposition of new cell-wall material across both ends of the plasmodesmatalpores (Lachaud & Maurousset 1996). The sucrose exportdeficient (sxd1) mutant of maize is unable to load sucroseinto minor veins due to a blockage of plasmodesmata atthe bundle sheath–vascular parenchyma cell-wall interface(Russin et al. 1996). This blockage also occurs by deposi-tion of wall material and/or callose (Russin et al. 1996;Botha et al. 2000) across the plasmodesmatal pores. It hasbeen hypothesized that sxd1 mutants have a defect in thesignalling mechanism between the chloroplasts andnucleus in bundle sheath cells. This defect in the communi-cation pathway affects the differentiation of bundle sheathcells during the sink-source transition, leading to deposi-tion of callose across plasmodesmata at the bundlesheath–vascular parenchyma cell-wall interface (MezittProvencher et al. 2001). Since the permanent blockage ofplasmodesmata may share similar underlying mecha-nisms, the cloning and study of genes such as sxd1 willhopefully elucidate the molecular events that underlieplasmodesmatal modifications that lead to loss of sym-plastic continuity.

During the sink–source transition in tobacco leaves, thenumbers of simple plasmodesmata are dramaticallyreduced (Roberts et al. 2001). Some of this loss appears tooccur by the conversion of simple plasmodesmata tobranched plasmodesmata (Oparka et al. 1999). However, itappears that many simple plasmodesmata are ‘sacrificed’during leaf development. Most of these plasmodesmataare lost during the rapid phase of leaf cell expansion whenintercellular air spaces are forming in the mesophyll. Notrace of these plasmodesmata remains in the mature leaf,and they are literally ripped apart during rapid leaf expan-sion. It would appear that these simple plasmodesmata areutilized for a brief developmental period during whichthe leaf functions as a sink for assimilates (Roberts et al.2001).

In the Arabidopsis shoot, communication between cellsin different areas of the meristem changes during the tran-sition from vegetative growth to flowering (Gisel et al.1999), whereas plasmodesmata in the cambium of Lupinus

plants were also gradually shut off as cells differentiatedinto phloem mother cells (van Bel & van Rijen 1994). Sym-plastic communication is also reduced during Arabidopsisembryogenesis when a developmental transition reducesthe plasmodesmal size exclusion limit at the torpedo stageof development (Kim et al. 2002a). Such data suggest thatsymplastic isolation is a common mechanism to allow reg-ulated and co-ordinated development and differentiationin plant tissues. By preventing the movement of signal mol-ecules, ions and photo-assimilate, plasmodesmal isolationalso affects the turgor potential within a given symplasticdomain. This aspect of symplastic isolation has long beenunderstood in the function of stomata, but has also recentlybeen shown to control development of cotton fibres (Ruan,Llewellyn & Furbank 2001). In these cells, a developmentalswitch from simple to branched plasmodesmata, concomi-tant with a transient closure of plasmodesmata, drives therapid elongation required in these specialized cells (Pfluger& Zambryski 2001; Ruan et al. 2001).

At least some plasmodesmal proteins can be targeted fordegradation via the ubiquitin pathway, as occurs for cyto-solic proteins. In regenerating protoplasts, discontinuousplasmodesmata showed high levels of ubiquitin whereasthose able to establish secondary contacts between neigh-bouring cells did not (Ehlers, Schulz & Kollmann 1996).The accumulation of ubiquitin in these half-plasmodesmatamay point to a role for this enzyme in the degradation andremoval of plasmodesmata.

INTERCELLULAR PROTEIN ANDRNA TRAFFICKING

Functional parallels exist between plasmodesmata andnuclear pore complexes, and it seems that common mech-anisms may underlie the transport of mRNA and proteinsthrough both types of pore (Lee, Yoo & Lucas 2000). Oneof the earliest examples of cell–cell signalling was reportedby Sussex (1951) who found that the signal that determinesthe dorsiventrality of leaves was initiated in the shoot apicalmeristem (SAM). It is now known that signals that passbetween the abaxial and adaxial cell layers of leaves arealso required for organ shape (Bowman 2000). During the1990s, many proteins and RNAs were discovered to movein plants, often displaying non-cell-autonomous functionsthat had previously been assumed to be the function ofsmaller signal molecules (Hake 2001). The SAM gives riseto all aerial parts of the plant and in most species, consistsof three layers, L1 (the outermost), L2 and L3, which giverise to the epidermis, cortical tissues and inner tissues(including the vasculature), respectively. In dicotyledonousplants, cells in the L1 and L2 layers divide anticlinally,effectively producing clonal cell layers (Poethig & Sussex1985; Sussex 1989). By using periclinal chimeras, where onecell layer is not genetically related to the others, it wasshown that signals could be transmitted between layers(Szymkowiak & Sussex 1992, 1993). In tomato periclinalchimeras, signals produced within the L3 layer determinethe number of carpels and size of the floral meristem



Figure 3. Diagrammatic representation of the expression pattern of various transcription factors in apical and floral meristems. Relative quantities of RNA and protein are represented by gradations of colour. In contrast to the unrestrained movement of GFP in meristems (see Fig. 4), a number of RNAs and proteins have been found to move in a tightly regulated manner within the meristem region, or not at all. (a) KNOTTED1 mRNA is found only in the L2 and L3 layers of the meristem whereas KNOTTED1 protein is found in all three layers of the meristem (Jackson et al. 1994). (b) The DEFICIENS mRNA is also present only in the L2 and L3 layers, but the protein shows polar trafficking, being found predominantly in the L1 layer, with lesser quantities present in the L2 and L3 layers (Perbal et al. 1996). (c) Genetic chimeras expressing LEAFY mRNA only in the L1 layer, showed LEAFY protein expression in all layers, most strongly within the L1 layer, but with a gradient of expression extending into inner layers (Sessions et al. 2000). (d) In experiments using genetic chimeras, the APETALA3 protein has been found only in the cell layers in which the APETALA3 RNA is expressed, showing that this protein does not traffic between cells of the meristem (Jenik & Irish 2001). (e) CLAVATA3 mRNA accumulates only in a small zone of cells at the meristem apex in the L1 and L2 layers, plus a few underlying cells in the L3 layer. It is hypothesized that the CLAVATA3 protein is secreted to the extracellular space and then moves through the apoplast into the L3 layer of the meristem and, by acting on the CLAVATA receptor found there, allows co-ordinated development of the meristem (Fletcher et al. 1999).



although these tissues are formed from cells in the L1 andL2 layers (Szymkowiak & Sussex 1992).

The protein KNOTTED1 (KN1) was the first transcrip-tion factor reported to move between cells (Jackson, Veit& Hake 1994). The KN1 protein was found in the L1 layerof the SAM although its RNA was restricted to the L2 andL3 layers (see Fig. 3). The fact that KN1 could movebetween cells was initially demonstrated by microinjectionof fluorescently labelled KN1 into tobacco leaf cells (Lucaset al. 1995) and has since also been shown using GFP-fusions (Kim et al. 2002b). These experiments showed thatthe protein was not only capable of trafficking betweencells in the leaf and the SAM (Kim et al. 2002b), but alsoincreased the plasmodesmal SEL and could specificallytraffic its own RNA between cells in tobacco mesophyll(Lucas et al. 1995). However, it is perhaps puzzling thatalthough KN1 can move into the L1 layer, and specificallytraffic its own mRNA in mesophyll, that the mRNA isunable to move with the protein into the L1 layer of themeristem. This suggests that expression of the mRNA isunnecessary, or perhaps deleterious in the L1 layer and thatmovement of the RNA, even in the presence of the KN1protein is tightly controlled. Several other non-cell-autonomous factors may move through plasmodesmata(Mezitt & Lucas 1996) but this may not be the case for allsuch signals. KN1 has since been shown to require theactivity of a chaperone molecule to unfold the protein, anda receptor molecule to allow movement from cell to cell(Kragler et al. 1998b). The same receptor apparently playsa role in the movement of CMV (Kragler et al. 1998b),suggesting that a single protein is able to mediate the traf-ficking of both an endogenous protein and viral RNA. Kra-gler et al. (2000) subsequently showed that a peptideantagonist could interact with a motif involved in plas-modesmal dilation, preventing the movement of KN1RNA, but still allowing limited movement of the KN1 pro-tein. Murillo, Cavallarin & San Segundo (1997) found thatPRms, an 18 kDa maize pathogenesis-related protein thatis produced in response to fungal infection, could also moveindependently between cells. This protein was found inplasmodesmata, and was also detected in both the paren-chyma cells of the protoxylem and central pith of maizeradicles, whereas the PRms RNA was found only inparenchyma cells of the protoxylem.

A number of non-cell-autonomous transcription factorsthat control flower development have now been described(see Fig. 3 and Sharma & Fletcher 2002). Transcription fac-tors that have been found to move symplastically includeLEAFY (LFY), in which the RNA is located only in theL1 layer of the Arabidopsis floral meristem whereas theprotein can be found in a number of inner cell layers(Sessions, Yanofsky & Weigel 2000), and DEFICIENS(DEF) which can move within developing Antirrhinumflowers (Perbal et al. 1996). The DEF protein shows direc-tional movement; it can move from the L2 upwards into theL1 layer, but not in the opposite direction (Perbal et al.1996). GNARLEY1, another non-cell-autonomous factorfrom maize leaves also shows directional movement, having

an effect only on cells in underlying layers, and not on cellsin the same plane (Foster, Veit & Hake 1999). In contrast,other proteins, such as APETALA3 and PISTILATA, alsofound in the SAM, do not move at all (Jenik & Irish 2001).In Arabidopsis, the maintenance of a functional SAMrequires a gene family called CLAVATA (Fletcher et al.1999; Brand et al. 2000; Sharma & Fletcher 2002). It isthought that a small protein encoded by CLA3 binds toCLA1, a transmembrane serine/threonine kinase, in orderto modulate transcription factors that are required for mer-istem maintenance. CLA3 RNA is expressed only in a smallgroup of stem cells at the centre of the meristem apex butthe encoded protein is hypothesized to move into deeperlayers of the meristem where it activates the CLV stem cellsignalling pathway to allow co-ordinated growth (Fletcheret al. 1999; Rojo et al. 2002). In contrast to the symplasticpathway thought to be used by other transcription factors,recent and elegant work by Rojo et al. (2002) has shownthat the CLV3 protein is transported through the secretorypathway and localized to the apoplast. The protein is nowthought to move through the apoplast into the L3 layerwhere it binds with, and activates, the CLV1–CLV2 com-plex at the plasma membrane to co-ordinate stem cell activ-ity via the transcription factor WUSCHEL (Brand et al.2000; Rojo et al. 2002). Fluorescent tracer studies by Rinne& van der Schoot (1998) showed that the SAM of birchseedlings maintains symplastic domains to allow co-ordinated morphogenesis. It has been proposed that anapoplastic means of transport, as has been shown forCLA3, may allow signalling between areas of the SAM thatare not symplastically coupled (Rojo et al. 2002). Figure 3summarizes the reported movement of some known tran-scription factors in the SAM.

A recent publication has highlighted the biological sig-nificance of the movement of a transcription factor intocells beyond its site of synthesis. In Arabidopsis, signalsinitiated in older cells and propagated to immature cells aretransmitted throughout the root meristem in order to deter-mine the positional fate of cells (van den Berg et al. 1995,1997). One specific gene has recently been identified thatcontrols radial patterning of the Arabidopsis root. TheSHORT ROOT (SHR) gene encodes a transcription factorrequired for formation of the endodermal layer, andmutants that lack this protein do not form an endodermis(Helariutta et al. 2000). In situ hybridization has shown thatthe SHR RNA is expressed only in the vascular tissues, butthe protein product is found in the vascular tissue and inthe neighbouring endodermis (Nakajima et al. 2001). Inaddition, the localization of SHR in the vascular tissues wasboth nuclear and cytoplasmic, but in the endodermis it wasnuclear only. This suggests that the cytoplasmic SHR movesone cell away from the vascular origin, presumably throughplasmodesmata, into the endodermis where it becomesnuclear localized. These authors also showed that whenSHR was expressed from either the endodermis-specificSCARECROW (SCR) promoter, or the constitutive Cau-liflower mosaic virus (CaMV) 35s promoter, supernumer-ary endodermal cell layers were produced, leading to



malformed roots and showing that expression of a gene inthe same tissues as the required protein product can actu-ally be detrimental to plant development (Nakajima et al.2001). These results highlight the fact that the controlledmovement of signal molecules and gene products is essen-tial for accurate development, and may go some way toexplaining why, for instance, the KNOTTED mRNA is notfound in the L1 layer of the meristem.

When considering the movement of transcription factorsin meristems, an apparent paradox emerges. The SAM isknown to have a high SEL (Ormenese et al. 2000), a char-acteristic of immature plant tissues in which simple plas-modesmata predominate (Oparka et al. 1999). Similarly theroot meristem shows extensive cell-to-cell trafficking ofphloem-unloaded GFP (K.M. Wright et al. unpubl. results;Fig. 4). Given the apparent high basal SEL of root andshoot meristems, what is surprising is not that transcriptionfactors move from cell to cell, but that they do not movemore extensively away from their site of cellular synthesisand throughout tissues (Fig. 3). Clearly, a tightly regulatedmovement of transcription factors occurs, allowing thisclass of proteins to modify the plasmodesmata leading intoneighbouring cells, but at the same time preventing theirfurther spread into cells beyond. Such highly specific move-ment points to a complex and orchestrated interactionbetween the transcription factor and the plasmodesmatapresent in meristematic tissues. Identifying the plasmodes-

mal components that interact with transcription factorsremains a challenge for the future.

LONG-DISTANCE MOVEMENT OF PROTEINS AND RNA IN PLANTS

A number of proteins have been found to traffic withinphloem tissues but these are described elsewhere in thisissue (see review by van Bel, pages 125–149 in this issue).However, many of the proteins known to move betweencells in the phloem also traffic over long distances throughthe phloem to subsequently unload in sink tissues. In het-erografts between members of the Cucurbitceae, rootstock-specific structural phloem proteins and their precursorswere found to be present in scion tissues (Golecki et al.1998), having moved extensively through the phloem(Golecki, Schulz & Thompson 1999). It should be notedhowever, that although a number of proteins have nowbeen found to pass through the plasmodesmata that con-nect sieve elements and companion cells of the phloem, andto unload in sink tissues, that no common structural orsequence-specific motifs have been identified among thisgroup of proteins. This fact suggests that protein traffickingin the phloem must occur in one of two ways. Transport maybe very highly specialized, with specific interactionsbetween each protein and plasmodesmal receptors. How-ever, it seems unlikely that such a wide array of receptorscould exist at the plasmodesmal pore, and even moreunlikely that receptors would exist for non-plant proteinssuch as GFP. The alternative, and more likely scenario, isthat trafficking within and through the phloem isrelatively unrestricted and does not rely on specificinteractions with plasmodesmal proteins.

In addition to proteins, plant RNAs are also trafficked along distance through the phloem, in some cases in conjunc-tion with endogenous proteins, and the RNA can subse-quently exert effects in sink tissues. The CmPP16 phloemprotein from Cucurbita maxima allows the transport ofboth sense and antisense RNA in the phloem, and canmove from cell to cell, increasing the plasmodesmal SELduring the process (Xoconostle-Cázares et al. 1999). Thesecharacteristics may allow this protein, and other similarhost proteins, to facilitate the long-distance delivery ofRNAs through the phloem to allow the regulation of trans-lational events in developing tissues of sink organs (Jor-gensen et al. 1998; Xoconostle-Cázares et al. 1999). Anotherendogenous phloem protein, PP2, a dimeric lectin and themost abundant component of Cucurbita spp. phloem exu-date, has also been shown to interact with RNA (Owens,Blackburn & Ding 2001). This protein can interact withHop stunt viroid. Viroids are small, circular, unencapsi-dated RNA pathogens that lack mRNA activity. PP2 is ableto move from cell to cell through plasmodesmata (Bal-achandran et al. 1997) and can itself traffic through thephloem (Golecki et al. 1999), suggesting that, in addition toviroids, other RNAs may be moved through the phloem byinteractions with chaperone proteins (Owens et al. 2001).Viral proteins can also act in a non-specific way to improve

Figure 4. Emerging lateral root of a transgenic Arabidopsis plant expressing GFP from the AtSUC2 companion cell-specific pro-moter. GFP unloads from the phloem and moves freely within the developing tissues of the root tip, including the meristem. GFP (green) is present in the cytoplasm of all cell layers. The cell walls are stained red with propidium iodide. (Image courtesy of K. Wright, SCRI). Bar = 50 mm.



phloem transport. Groundnut rosette virus (GRV) isunusual in that it does not encode for a coat protein, nor-mally required to protect the viral RNA during systemicmovement. The protein encoded by GRV ORF3 has beenshown to act as a trans-acting long-distance movement fac-tor that can facilitate the movement of unrelated viralRNAs through the phloem (Ryabov, Robinson & Taliansky1999).

Endogenous Cucurbita maxima CmNACP mRNA hasbeen found to move through the phloem and unload inapical tissues of heterografts. CmNACP, part of the NACdomain gene family which may be involved in apical mer-istem development, was found to move over long distancesand accumulate in vegetative, root and floral meristems.These results proved the existence of a transport system forspecific mRNA transcripts into developing apices, provid-ing evidence of a novel mechanism that could integratedevelopmental and physiological processes throughout theplant (Ruiz-Medrano, Xoconostle-Cázares & Lucas 1999).A further example of specific mRNA trafficking has proventhat developmental regulation is possible using this sys-temic RNA signalling mechanism. Kim et al. (2001) showedthat RNA transcripts of the dominant leaf mutant fromtomato, Mouse ears (Me), could move a long distancethrough the phloem and cause a change in leaf morphology.The Me mutant transcripts were graft transmissible, local-izing to the same areas of the normal scion apex as in non-grafted Me plants, and proving that the pattern of transcriptaccumulation was due to controlled trafficking of the Metranscripts (Kim et al. 2001). This study does not confirmthat the altered morphology seen in scion tissue of thegrafted plants was due directly to the trafficked RNA; asecondary signal, such as a growth regulator or hormone,might cause the morphological change, as pointed out byJackson (2001). However, this elegant study does illustratethe involvement of RNA trafficking in plant development.It now seems highly likely that plants routinely use endog-enous RNAs as long-distance signals in the phloem to inte-grate and regulate plant development (Jorgensen et al.1998; Ruiz-Medrano, Xoconostle-Cázares & Lucas 2001;Wu, Weigel & Wigge 2002).

Post-transcriptional gene silencing (PTGS) in plants is aresponse to foreign nucleic acids; either inserted transgenesor viral nucleic acids (Vaucheret et al. 1998; Voinnet, Pinto& Baulcombe 1999). Plant gene silencing signals are alsothought to move through plasmodesmata; for example,they can enter guard cells of immature leaves, but areexcluded from these symplastically isolated cells in maturetissue (Voinnet et al. 1998). A silencing response that isgenerated by introduction of ectopic DNA into a few cellscan spread, via plasmodesmata, into neighbouring cells andsubsequently the phloem, allowing systemic spread of thesignal (Voinnet et al. 1998). Palauqui & Vaucheret 1998)showed that PTGS can spread to cells that do not expressthe source transgene, causing the silencing of genes thathave homology with the short RNAs which act as signals(Hamilton & Baulcombe 1999; Di Serio et al. 2001). Thesignals are thought to be short RNA species (25 nucle-

otides; Hamilton & Baulcombe 1999), which can enter thephloem and subsequently unload in sink tissues (Voinnetet al. 1998) in a manner that resembles the unloading pat-terns of photo-assimilate or viruses (Oparka & Santa Cruz2000). This PTGS mechanism is used as a defence againsta number of plant viruses. A recent report suggests that theshort RNAs are not required for methylation of the trans-gene, a state required to allow and maintain gene silencing,and are unnecessary for the production and transmission ofthe silencing signal (Mallory et al. 2001). These results sug-gest that alternative or additional signals are involved, andthat gene silencing, and symplastic RNA signalling in gen-eral, is likely to be far more complex than we currentlyunderstand.

VIRUS MOVEMENT

Until relatively recently, plant viruses were the only agentsknown in which a single protein was able to interact withplasmodesmata to allow the transfer of proteins and nucleicacids from one cell to another. As such, many studies onintercellular protein movement in plants were carried outusing recombinant viral MPs that were microinjected intoplant cells along with a fluorescent reporter molecule (Fuji-wara et al. 1993; Noueiry, Lucas & Gilbertson 1994; Waig-mann et al. 1994; Ding et al. 1995). The fact that facilitatedprotein trafficking can occur in the absence of a viral infec-tion (see above) showed that these mechanisms were likelyto be of general relevance to plant physiology. This led toa debate as to whether viral MPs may have originally beenplant proteins that were ‘trapped’ by viral genomes andused to facilitate viral movement (Lucas & Gilbertson1994; see also reviews by Maule 1994 and Mezitt & Lucas1996). Although essential for plant development and func-tion, plasmodesmata represent an ‘Achilles heel’ that canbe exploited and manipulated by viruses to allow them tospread throughout plant tissues (Oparka & Roberts 2001).

Local virus movement occurs via plasmodesmata,whereas long-distance virus spread uses the vascular tissues(Samuel 1934; Hull 1989; Maule 1991; Leisner & Turgeon1993; Lucas & Gilbertson 1994; Carrington et al. 1996).Most viruses utilize the phloem to spread systemically. Evi-dence for this has come from a variety of experimentsshowing that movement is influenced by the flow of metab-olites (Roberts 1952; Leisner & Turgeon 1993) and thatwhen stems are ‘ringed’ or steamed to kill the phloem, virusspread is prevented or delayed (Roberts 1952). Virus par-ticles have also been seen in electron micrographs ofphloem (Price 1966; Esau, Cronshaw & Hoefert 1967).However, very few studies have examined the entire path-way taken from the infection site to the phloem, or howviruses are transported into and out of the vascular tissue(Roberts et al. 1997; Nelson & van Bel 1998).

Some viruses move through plasmodesmata as intact vir-ions, causing permanent modification to plasmodesmalstructure. Viruses such as CaMV, Tomato spotted wilt virusand Cowpea mosaic virus use protein tubules, encoded byviral proteins, to line the plasmodesma; allowing the virions



to pass through the tubule (Hull 1992; Storms et al. 1995).Others viruses, such as PVX, Dahlia mosaic virus andTobacco etch virus (TEV) do not use tubules, but passthrough the pore as intact virions (Kitajima & Lauritis1969; Weintraub, Ragetli & Leung 1976; Santa Cruz et al.1998). Another group of plant viruses, typified by TMV,cause more subtle and often transient alterations to plas-modesmata, allowing the viral genome to move as a ribo-nucleoprotein complex, and encode for viral MPs whichmodify plasmodesmata (Deom, Lapidot & Beachy 1992;Citovsky & Zambryski 1993; McLean et al. 1993; Lucas &Gilbertson 1994). The plethora of viral movement mecha-nisms suggest that a unified ‘strategy’ for viral movement isunlikely to occur, given the variety of viral proteins andgenome organizations involved.

Viral genomes can consist of either single- or double-stranded DNA or RNA. For those viruses that move asribonucleoprotein complexes, it has been postulated thatmovement occurs using single-stranded genomes (Citovskyet al. 1992), which would be narrower than double-strandedcomplexes, therefore requiring less plasmodesmal modifi-cation. In support of this hypothesis, it was found that theMP of CaMV, a double-stranded DNA virus whichreplicates via an RNA intermediate, can selectively bindsingle-stranded nucleic acids, and binds RNA with greaterefficiency than single-stranded DNA (Citovsky, Knorr &Zambryski 1991; Thomas & Maule 1995). However, moreevidence will be required to validate this model and takeinto account larger, circular genomes such as those pos-sessed by Geminiviruses.

Some viral MPs have been shown to be antigenicallyrelated to endogenous plant proteins (Xoconostle-Cázareset al. 1999), and have provided essential tools for the studyof intracellular transport. The most studied viral MP is the30 kDa MP of TMV. Some time ago, this protein was shownto modify the basal SEL of plasmodesmata in infected cellsto permit virus movement from cell to cell (Deom, Oliver& Beachy 1987; Wolf et al. 1989; Derrick, Barker & Oparka1990; Waigmann et al. 1994). The TMV MP has been shownto localize to plasmodesmata (Fig. 5A; Tomenius, Clapham& Meshi 1987; Ding et al. 1992a) and, in common with theCMV 3a MP, and probably the PLRV (Prüfer et al. 1997)MP, accumulates only in branched plasmodesmata (Dinget al. 1992a; Itaya et al. 1998; Hofius et al. 2001; Robertset al. 2001). The TMV MP binds single-stranded nucleicacids (Citovsky et al. 1990, 1992), and interacts with the ER,actin and microtubules (Heinlein et al. 1995; McLean,Zupan & Zambryski 1995; Mas & Beachy 1999, 2000;Boyko et al. 2000a; Gillespie et al. 2002). The MP appearsto form a complex with the viral genomic RNA, and manymodels of TMV movement implicate microtubules as thecytoskeletal element responsible for transporting the MPto the plasmodesmal pore (Heinlein et al. 1995; Boyko et al.2000b; Mas & Beachy 2000). The TMV MP has a conservedmicrotubule-binding motif, and mutants with an alteredmotif showed no interaction with microtubules, which pre-vented both viral infectivity and movement, although themutant MP continued to target plasmodesmata. (Boyko

et al. 2000b). A second, contrasting study involving a differ-ent MP mutant showed that this mutant could localize tomicrotubules, but was unable to target plasmodesmata ormove between cells, and interfered with wild-type MPtargeting to microtubules (Kotlizky et al. 2001). However,a recent study has shown that TMV is able to replicateand move in the absence of microtubules, forcing a re-evaluation of current models for TMV movement(Gillespie et al. 2002). The cytoskeleton is involved in somestages of the viral infection process, as is the ER, but todate, the exact mechanism for TMV movement remains tobe finalized. In animal cells, actin and myosin have beenimplicated in the transport and anchoring of mRNAs tospecific sites within the cell, suggesting that the cytoskele-ton could also play a role in endogenous mRNA trafficking(Bassel & Singer 1997; Sundell & Singer 1991), and similardata has recently been published for the unicellular algaAcetabularia acetabulum (Vogel, Grieninger & Zetsche,2002).

Figure 5. (A) Confocal image showing epidermal cells on a source leaf of transgenic Nicotiana tabacum plants expressing the TMV movement protein-GFP fusion. The movement protein-GFP fusion is localized to branched plasmodesmata (green) in the walls between the epidermal cells (imaged in false transmission; red). Bar = 10 mm. (B) When the TMV 30K-GFP fusion protein is expressed in transgenic tobacco plants under the AtSUC2 compan-ion cell-specific promoter, the protein is only found in plasmodes-mata within phloem tissues. Confocal micrograph showing the movement protein localized to plasmodesmata in the longitudinal walls between phloem elements (running left to right). Bar = 10 mm.



CMV also requires the ER for infection. CMV moves asa ribonucleoprotein complex that includes a MP and viralcoat protein (Canto et al. 1997; Blackman et al. 1998). Oncethe complex enters sieve elements, the MP associates withthe parietal ER, and viral assembly occurs in membrane-bound vesicles derived from the ER (Blackman et al. 1998).However, there are viruses that move without any interac-tion between the ER or cytoskeletal elements (Satoh et al.2000) and viral MPs that, although essential for movementdo not localize to plasmodesmata (Solovyev et al. 2000).Yeast two-hybrid and far-western screens of cDNA librar-ies have recently shown that viral MPs can also interactwith a number of host proteins such as putative transcrip-tional co-activators that may modulate host gene expres-sion during pathogenesis (Matsushita et al. 2001, 2002a),and also proteins with homologies to myosin, kinesin andDnaJ-like chaperones, again suggesting an involvement ofthe cytoskeleton during viral infection (von Bargen et al.2001).

INTERACTION BETWEEN PLASMODESMAL PROTEINS AND VIRAL MPS

The continued study of viral MPs is beginning to offer cluesabout plasmodesmal proteins that may be involved in viralintercellular transport. The TMV MP interacts with pectinmethylesterase (Dorokhov et al. 1999; Chen et al. 2000), anenzyme that modifies the pectin in plant cell walls. Whenthis enzyme was screened for interactions with other viralMPs, both Turnip vein clearing virus and CaMV MPs boundto pectin methylesterase (Chen et al. 2000). A pectinmethylesterase-binding domain has been identified on theTMV MP, and deletion of this domain prevents cell–cellmovement of the virus (Chen et al. 2000). Pectin methyl-esterase has been localized to plasmodesmata (Morvanet al. 1998), and has also been shown to have RNA-bindingproperties (Dorokhov et al. 1999), suggesting that it is agood candidate for a MP receptor in plasmodesmata.

Additional genes involved in the viral movement processare being discovered in plants resistant to viral infection.For instance, the Arabidopsis mutation vsm1 restricts thesystemic movement of TMV and the related Tobamovirus,Turnip vein clearing virus (Lartney, Ghoshroy & Citovsky1998), and at least three genes, RTM1, RTM2 and RTM3in Arabidopsis restrict the movement of TEV, preventingsystemic spread (Mahajan et al. 1998; Chisholm et al. 2000,2001). RTM1 and RTM2 are located only in sieve elementsand their interaction is specific to TEV, and does notinvolve a hypersensitive response or induction of systemicacquired resistance (Mahajan et al. 1998; Whitham et al.2000). Further analysis of such mutants may reveal if thesegenes operate by preventing movement of the viral genomethrough plasmodesmata in the phloem.

A potential plasmodesmal protein was discoveredrecently using the yeast two-hybrid system to screen forArabidopsis proteins that interacted with the CaMV MP, aprotein known to be necessary for movement through plas-modesmata (Huang et al. 2001). One protein, MPI7, was

found to interact strongly with the CaMV MP, and dis-played fluorescence resonance energy transfer (FRET) insitu. MPI7 was widely expressed throughout different tis-sues of Arabidopsis and was localized to punctate spots atthe cytoplasmic periphery, which may be plasmodesmata.Sequence data suggests that the protein may be akin to amammalian rab acceptor protein (Huang et al. 2001), afamily of proteins involved in protein transport through theendomembrane system, and possibly associated with theER. More detailed studies are required to confirm whetherMPI7 is indeed associated with plasmodesmata.

Biochemical purification of plasmodesmal proteins fromwall preparations has proved difficult due to the smallamounts of protein present in plasmodesmata, and the dif-ficulty in obtaining pure plasmodesmal extracts that arefree of wall contaminants (Epel 1994; Epel et al. 1995).Antibodies raised to such extracts have been immunolocal-ized to plasmodesmata, but none of the genes encoding theproteins have yet been isolated or characterized (Epel1994; Waigmann et al. 1997). One of the problems in iden-tifying plasmodesmal genes with whole-plant geneticscreens is likely to be that any mutants arising will have ahigh chance of lethality, making screens at the embryonicstage necessary. Many plasmodesmal proteins and struc-tural components are probably essential for plant functionand development. A potential plasmodesmal regulatoryprotein, encoded by the gene increased size exclusion 1 (ise-1), increased the movement of molecules between cells butwas indeed lethal at the embryonic stage (Zambryski &Crawford 2000).

PLASMODESMATA IN AND AROUNDTHE PHLOEM

Plasmodesmata in phloem tissue are particularly difficult tostudy as they are deeply embedded within the plant. Theunique pore-plasmdesma units (PPUs) that connect sieveelements and companion cells are described elsewhere inthis issue (see review by Aart van Bel, pages 125–149 in thisissue) and will not be dealt with here. For many viruses,there is a fundamental difference between movementthrough plasmodesmata in mesophyll tissues and move-ment into and through the vascular tissue (Gilbertson &Lucas 1996). There may be a control point for virus entryinto the phloem at the junction between the vascular paren-chyma cells and the companion cell (Leisner & Turgeon1993; Oparka & Turgeon 1999; Santa Cruz 1999), but ourunderstanding of viral entry into the phloem is limited.Once viruses enter the phloem, they are carried with theflow of photo-assimilate and unload in sink tissues, appar-ently without the same level of control that accompaniedtheir entry (Roberts et al. 1997; Santa Cruz 1999). With onlya few exceptions (Petty & Jackson 1990; Ryabov et al.1999), additional viral proteins (most commonly the coatprotein) are required to facilitate a systemic infection(Scholthof et al. 1995). In one recent study, domains withinthe coat protein of African cassava mosaic geminivirus werefound to mediate nuclear import and export, as well as



targeting GFP-fusions to the cell periphery, presumably toplasmodesmata (Unseld et al. 2001). Viral MPs can alsodiffer in their interactions with mesophyll plasmodesmataand phloem plasmodesmata. Neither the TMV MP, norantibodies raised to the MP, localize to phloem plasmodes-mata in transgenic plants expressing the MP under a con-stitutive promoter (Ding et al. 1992a), although this MPdoes target phloem plasmodesmata when expressed from acompanion cell-specific promoter (K.M. Wright, personalcomm.; Fig. 5B). It has been suggested that the TMV MPis not required in the phloem (Blackman & Overall 2001)but we do know that, even though plasmodesmata are nottargeted by this MP in sink leaves, it is essential for viralmovement in sink leaves (Roberts et al. 2001). In contrast,the MP of CMV does bind to all plasmodesmata in thephloem, including those that connect sieve elements andcompanion cells (Blackman et al. 1998). Similarly, thePLRV MP has been found within phloem plasmodesmatain both minor and major veins (Hofius et al. 2001).

THE FUTURE

Previous views of plasmodesmata as static pores in thewalls between cells are no longer tenable. The current pic-ture that emerges is one in which plasmodesmata functionas highly flexible structures that can exert considerable con-trol of the molecules that pass through them. Althoughmany published models depict simple plasmodesmata suchas those shown in Fig. 1, it is clear that there are fundamen-tal differences in architecture between plasmodesmata indifferent tissues, and at different interfaces within a singletissue. In the same way that extensive families of solutetransporters have been discovered in recent years (forexample over 20 types of sucrose transporter have nowbeen described in different plant species; Lalonde et al.1999; Williams, Lemoine & Sauer 2000), in the future it willnot be surprising to find that plasmodesmata, and theircomponent proteins, will be grouped into families that per-form specific functions at discrete locations within theplant. This concept of a plasmodesmal ‘family tree’ is onlynow beginning to emerge with the demonstration that dif-ferent types of plasmodesmata have different basic perme-ability properties (Oparka et al. 1999).

Only a small number of plasmodesmal proteins havebeen discovered by design (Epel 1994; Epel et al. 1995). Themajority of plant proteins known to be located in plas-modesmata were discovered serendipitously. There isclearly a need for high-throughput screening proceduresthat will lead to the discovery of novel plasmodesmal genesand proteins. Although relatively unsuccessful to date, pro-teomics approaches involving the large-scale isolation ofplasmodesmal proteins from cell-wall fractions may lead tothe discovery of novel proteins that regulate plasmodesmalfunctions. Such methods will have to ensure that the plas-modesmal ‘fraction’ is entirely free of contaminants fromother cellular membranes.

Although it is likely that the discovery of novelplasmodesmal proteins will advance our understanding of

plasmodesmal composition, not every protein that regu-lates plasmodesmal function will be discovered through thisroute. Some genes/proteins may interact only transientlywith plasmodesmata to exert their influence, whereas oth-ers may exert effects on plasmodesmal function withoutever locating to plasmodesmata. A case in hand is the sxd1mutant of maize (Russin et al. 1996; Botha et al. 2000)whose mutation leads to a downstream effect on plas-modesmal development that is not intuitive, given its pri-mary role in chloroplast-to-nucleus signalling (MezittProvencher et al. 2001). Unravelling the pleiotropic effectsof different genes on plasmodesmal structure/functionremains a major challenge for the future.

With the onset of the functional genomics era, imagina-tive whole-plant or tissue screens, such as that developedfor the ise mutants (Zambryski & Crawford 2000; Kim et al.2002a), will be required to characterize the plethora ofgenes and proteins arising from genomics databases. Thechallenge for the future is not to provide further descriptiveanalyses of plasmodesmal structure and function, but toprovide future researchers in this rapidly moving field withthe tools necessary to initiate the molecular dissection ofplasmodesmata.

ACKNOWLEDGMENTS

The authors are supported by grants awarded by the Scot-tish Executive Environment and Rural Affairs Department(SEERAD). The authors would like to thank Kath Wrightfor providing Fig. 4, and Don Fisher for discussions regard-ing the calculation of Stokes radii.

REFERENCES

Allison A.V. & Shalla T.A. (1974) The ultrastructure of locallesions induced by potato virus X: a sequence of cytologicalevents in the course of infection. Phytopathology 64, 784–793.

Badelt K., White R.G., Overall R.L. & Vesk M. (1994) Ultrastruc-tural specialisations of the cell wall sleeve around plasmodes-mata. American Journal of Botany 81, 1422–1427.

Balachandran S., Xiang Y., Schobert C., Thompson G.A. & LucasW.J. (1997) Phloem sap proteins from Cucurbita maxima andRicinus communis have the capacity to traffic cell to cell throughplasmodesmata. Proceedings of the National Academy of Sci-ence, USA 94, 14150–14155.

Balus̆ka F., Cvrckova F., Kendrick-Jones J. & Volkmann D. (2001)Sink plasmodesmata as gateways for phloem unloading. MyosinVIII and calreticulin as molecular determinants of sinkstrength? Plant Physiology 126, 39–46.

Balus̆ka F., Samaj J., Napier R. & Volkmann D. (1999) Maizecalreticulin localizes preferentially to plasmodesmata in rootapex. Plant Journal 19, 481–488.

Barclay G.F., Peterson C.A. & Tyree M.T. (1982) Transport offluorescein in trichomes of Lycopersicon esculentum. CanadianJournal of Botany 60, 397–402.

von Bargen S., Salchert K., Paape M., Piechulla B. & KellmannJ.-W. (2001) Interactions between the tomato spotted wilt virusmovement protein and plant proteins showing homologies tomyosin, kinesin and DnaJ-like chaperones. Plant Physiologyand Biochemistry 39, 1083–1093.

Bassel G. & Singer R.H. (1997) mRNA and cytoskeletal elements.Current Opinion in Cell Biology 9, 109–115.



Beebe D.U. & Turgeon R. (1991) Current perspectives on plas-modesmata: structure and function. Physiologia Plantarum 83,194–199.

Beffa R.S., Hofer R.-M., Thomas M. & Meins Jn. F. (1996)Decreased susceptibility to viral disease of b-1,3-glucanase-deficient plants generated by antisense transformation. PlantCell 8, 1001–1011.

van Bel A.J.E. & Knoblauch M. (2000) Sieve element and com-panion cell: the story of the comatose patient and the hyperac-tive nurse. Australian Journal of Plant Physiology 27, 477–487.

van Bel A.J.E. & van Rijen H.V.M. (1994) Microelectrode-recorded development of the symplasmic autonomy of the sieveelement / companion cell complex in the stem phloem ofLupinus luteus L. Planta 192, 165–175.

Belitser N.V., Zaalishvili G.V. & Sytnianskaja N.P. (1982) Ca2+-binding sites and Ca2+-ATPase activity in barley root tip cells.Protoplasma 111, 63–78.

van den Berg C., Willemsen V., Hage W., Weisbeek P. & ScheresB. (1995) Cell fate in the Arabidopsis root meristem determinedby directional signalling. Nature 378, 62–65.

van den Berg C., Willemsen V., Hendriks G., Weisbeek P. &Scheres B. (1997) Short-range control of cell differentiation inthe Arabidopsis root meristem. Nature 390, 287–289.

Blackman L.M. & Overall R.L. (1998) Immunolocalisation of thecytoskeleton to plasmodesmata of Chara corallina. Plant Journal14, 733–741.

Blackman L.M. & Overall R.L. (2001) Structure and function ofplasmodesmata. Australian Journal of Plant Physiology 28, 709–727.

Blackman L.M., Boevink P., Santa Cruz S., Palukaitis P. & OparkaK.J. (1998) The movement protein of cucumber mosaic virustraffics into sieve elements in minor veins of Nicotiana clevelan-dii. Plant Cell 10, 525–537.

Blackman L.M., Harper J.D.I. & Overall R.L. (1999) Localizationof a centrin-like protein to higher plant plasmodesmata.European Journal of Cell Biology 78, 297–304.

Boevink P., Oparka K.J., Santa Cruz S., Martin B., Betteridge A.& Hawes C. (1998) Stacks on tracks: the plant Golgi apparatustraffics on an actin/ER network. Plant Journal 15, 441–447.

Botha C.E.J. & Cross R.H.M. (2000) Towards reconciliation ofstructure with function in plasmodesmata – who is the gate-keeper? Micron 31, 713–721.

Botha C.E.J. & Evert R.F. (1988) Plasmodesmatal distribution andfrequency in vascular bundles and contiguous tissues of the leafof Themeda triandra. Planta 173, 433–441.

Botha C.E.J., Cross R.H.M., van Bel A.J.E. & Peter C.I. (2000)Phloem loading in the sucrose-export-defective (SXD-1) mutantmaize is limited by callose deposition at plasmodesmata in bun-dle sheath–vascular parenchyma interface. Protoplasma 214, 65–72.

Botha C.E.J., Hartley B.J. & Cross R.H.M. (1993) The ultrastruc-ture and computer-enhanced digital image analysis of plas-modesmata at the Kranz mesophyll–bundle sheath interface ofThemeda triandra var. imberbis (Retz) A. Camus in convention-ally-fixed leaf blades. Annals of Botany 72, 255–261.

Bowman J.L. (2000) The YABBY gene family and abaxial cell fate.Current Opinion in Plant Biology 3, 17–22.

Boyko V., Ferralli J., Ashby J., Schellenbaum P. & Heinlein M.(2000b) Function of microtubules in intercellular transport ofplant virus RNA. Nature Cell Biology 2, 826–832.

Boyko V., van der Laak J., Ferralli J., Suslova E., Kwon M.-O. &Heinlein M. (2000a) Cellular targets of functional and dysfunc-tional mutants of tobacco mosaic virus MP fused to green fluo-rescent protein. Journal of Virology 74, 11339–11346.

Brand U., Fletcher J.C., Hobe M., Meyerowitz E.M. & Simon R.(2000) Dependence of stem cell fate in Arabidopsis on a

feedback loop regulated by CLV3 activity. Science 289, 617–619.

Brown S.M., Oparka K.J., Sprent J.I. & Walsh K.B. (1995)Symplastic transport in soybean root nodules. Soil Biology andBiochemistry 27, 387–399.

Bucher G.L., Tarina C., Heinlein M., Di Serio F., Meins F. Jr &Iglesias V.A. (2001) Local expression of enzymatically activeclass I b-1, 3-glucanase enhances symptoms of TMV infection intobacco. Plant Journal 28, 361–369.

Canto T., Prior D.A.M., Hellwald K.-H., Oparka K.J. & PalukaitisP. (1997) Characterization of cucumber mosaic virus. IV. MPand coat protein are both essential for cell-to-cell movement ofcucumber mosaic virus. Virology 237, 237–248.

Cantrill L.C., Overall R.L. & Goodwin P.B. (1999) Cell-to-cellcommunication via plant endomembranes. Cell Biology Interna-tional 23, 653–661.

Cantrill L.C., Overall R.L. & Goodwin P.B. (2001) Changes insymplastic permeability during adventitious shoot regenerationin tobacco thin cell layers. Planta 214, 206–214.

Carr D.J. (1976) Historical perspectives on plasmodesmata. InIntercellular Communication in Plants: Studies on Plasmodes-mata (eds B.E.S. Gunning & A.W. Robards), pp. 291–295.Springer-Verlag, Berlin, Germany.

Carrington J.C., Kasschau K.D., Mahajan S.K. & Schaad M.C.(1996) Cell-to-cell and long-distance transport of viruses inplants. Plant Cell 8, 1669–1681.

Chaffey N. & Barlow P. (2001) The cytoskeleton facilitates a three-dimensional symplasmic continuum in the long-lived ray andaxial parenchyma cells of angiosperm trees. Planta 213, 811–823.

Chen M.-H., Sheng J., Hind G., Handa A.K. & Citovsky V. (2000)Interaction between the tobacco mosaic virus MP and host cellpectin methylesterases is required for viral cell-to-cell move-ment. EMBO Journal 19, 913–920.

Chisholm S.T., Mahajan S.K., Whitham S.A., Yamamoto M.L. &Carrington J.C. (2000) Cloning of the Arabidopis RTM1 gene,which controls restriction of long-distance movement of tobaccoetch virus. Proceedings of the National Academy of Science, USA97, 489–494.

Chisholm S.T., Parra M.A., Anderberg R.J. & Carrington J.C.(2001) Arabidopsis RTM1 and RTM2 genes function in phloemto restrict long-distance movement of tobacco etch virus. PlantPhysiology 127, 1667–1675.

Choi C.W. (1999) Modified plasmodesmata in Sorghum (Sorghumbicolor L. Moench) leaf tissues infected by maize dwarf mosaicvirus. Journal of Plant Biology 42, 63–70.

Citovsky V. & Zambryski P. (1993) Transport of nucleic acidsthrough membrane channels: snaking through small holes.Annual Review of Microbiology 47, 167–197.

Citovsky V., Knorr D., Schuster G. & Zambryski P. (1990) TheP30 MP of tobacco mosaic virus is a single-strand nucleic acidbinding protein. Cell 60, 637–647.

Citovsky V., Knorr D. & Zambryski P. (1991) Gene I, a potentialcell-to-cell movement locus of cauliflower mosaic virus, encodesan RNA-binding protein. Proceedings of the National Academyof Science, USA 88, 2476–2480.

Citovsky V., McLean B.G., Zupan J.R. & Zambryski P. (1993)Phosphorylation of tobacco mosaic virus cell-to-cell MP by adevelopmentally regulated plant cell wall-associated proteinkinase. Genes and Development 7, 904–910.

Citovsky V., Wong M.L., Shaw A.L., Venkataram Prasad B.V. &Zambryski P. (1992) Visualisation and characterisation oftobacco mosaic virus MP binding to single-stranded nucleicacids. Plant Cell 4, 397–411.

Cleland R.E., Fujiwara T. & Lucas W.J. (1994) Plasmodesmal-mediated cell-to-cell transport in wheat roots is modulated byanaerobic stress. Protoplasma 178, 81–85.



Cook M.E., Graham L.E., Botha C.E.J. & Lavin C.A. (1997) Com-parative ultrastructure of plasmodesmata of Chara and selectedbryophytes: toward an elucidation of the evolutionary origin ofplant plasmodesmata. American Journal of Botany 84, 1169–1178.

Crawford K.M. & Zambryski P.C. (2000) Subcellular localizationdetermines the availability of non-targeted proteins to plas-modesmatal transport. Current Biology 10, 1032–1040.

Deom C.M., Lapidot M. & Beachy R.N. (1992) Plant virus move-ment proteins. Cell 69, 221–224.

Deom C.M., Oliver M.J. & Beachy R.N. (1987) The 30-kilodaltongene product of tobacco mosaic virus potentiates virus move-ment. Science 237, 389–394.

Derrick P.M., Barker H. & Oparka K.J. (1990) Effect of virusinfection on symplastic transport of fluorescent tracers in Nic-otiana clevelandii leaf epidermis. Planta 181, 555–559.

Di Serio F., Schöb H., Iglesias A., Tarina C., Bouldoires E. &Meins F. Jr (2001) Sense- and antisense-mediated gene silencingin tobacco is inhibited by the same viral suppressors and isassociated with accumulation of small RNAs. Proceedings of theNational Academy of Science, USA 98, 6506–6510.

Ding B., Haudenshield J.S., Hull R.J., Wolf S., Beachy R.N. &Lucas W.J. (1992a) Secondary plasmodesmata are specific sitesof localization of the tobacco mosaic virus movement protein intransgenic tobacco plants. Plant Cell 4, 915–928.

Ding B., Turgeon R. & Parthasarathy M.V. (1992b) Substructureof freeze-substituted plasmodesmata. Protoplasma 169, 28–41.

Ding B., Kwon M.-O. & Warnberg L. (1996) Evidence that actinfilaments are involved in controlling the permeability of plas-modesmata in tobacco mesophyll. Plant Journal 10, 157–164.

Ding B., Li Q., Nguyen L., Palukaitis P. & Lucas W.J. (1995)Cucumber mosaic virus 3a protein potentiates cell-to-celltrafficking of CMV RNA in tobacco plants. Virology 207, 345–353.

Ding B., Turgeon R. & Parthasarathy M.V. (1991) Plasmodesmatalsubstructure in cryofixed developing tobacco leaf tissue. InRecent Advances in Phloem Transport and Assimilate Partition-ing (ed. J.L. Bonnemain, S. Delrot, J.L. Dainty & W.T. Lucas),pp. 317–323. Ouest Editions, Nantes, France.

Ding D.-Q. & Tazawa M. (1989) Influence of cytoplasmic stream-ing and turgor pressure gradient on the transnodal transport ofrubidium and electrical conductance in Chara corallina. PlantCell Physiology 30, 739–748.

Dorokhov Y.L., Mäkinen K., Frolova O.Y., Merits A., SaarinenJ., Kalkkinen N., Atabekov J.G. & Saarma M. (1999) A novelfunction for a ubiquitous plant enzyme pectin methylesterase:the host-cell receptor for the tobacco mosaic virus movementprotein. FEBS Letters 461, 223–228.

Drake G.A., Carr D.J. & Anderson W.P. (1978) Plasmolysis, plas-modesmata, and the electrical coupling of oat coleoptyle cells.Journal of Experimental Botany 29, 1205–1214.

Duckett C.M., Oparka K.J., Prior D.A.M., Dolan L. & Roberts K.(1994) Dye-coupling in the root epidermis of Arabidopsis isprogressively reduced during development. Development 120,3247–3255.

Ehlers K. & Kollmann R. (2000) Synchronization of mitotic activ-ity in protoplast-derived Solanum nigrum L. microcalluses iscorrelated with plasmodesmal connectivity. Planta 210, 269–278.

Ehlers K. & Kollmann R. (2001) Primary and secondary plas-modesmata: structure, origin and functioning. Protoplasma 216,1–30.

Ehlers K., Binding H. & Kollmann R. (1999) The formation ofsymplastic domains by plugging of plasmodesmata: a generalevent in plant morphogenesis? Protoplasma 209, 181–192.

Ehlers K., Schulz M. & Kollmann R. (1996) Subcellular localiza-

tion of ubiquitin in plant protoplasts and the function of ubiq-uitin in selective degradation of outer-wall plasmodesmata inregenerating protoplasts. Planta 199, 139–151.

Enkerli K., Hahn M.G. & Mims C.W. (1997) Immunogold local-ization of callose and other plant cell wall components in soy-bean roots infected with the oomycete Phytophthora sojae.Canadian Journal of Botany 75, 1509–1517.

Epel B.L. (1994) Plasmodesmata: composition, structure and traf-ficking. Plant Molecular Biology 26, 1343–1356.

Epel B.L., Kuchuck B., Kotlizky G., Shurtz S., Erlanger M. &Yahalom A. (1995) Isolation and characterization of plasmodes-mata. Methods in Cell Biology 50, 237–253.

Erwee M.G. & Goodwin P.B. (1983) Characterisation of the Ege-ria densa Planch. leaf symplast. Inhibition of the intercellularmovement of fluorescent probes by group II ions. Planta 158,320–328.

Erwee M.G. & Goodwin P.B. (1984) Characterization of the Ege-ria densa leaf symplast: response to plasmolysis, deplasmolysisand to aromatic amino acids. Protoplasma 122, 162–168.

Erwee M.G. & Goodwin P.B. (1985) Symplast domains inextrastelar tissues of Egeria densa Planch. Planta 163, 9–19.

Erwee M.G., Goodwin P.B. & van Bel A.J.E. (1985) Cell-cellcommunication in the leaves of Commelina cyanea and otherplants. Plant Cell and Environment 8, 173–178.

Esau K., Cronshaw J. & Hoefert L.L. (1967) Relation of beetyellows virus to the phloem and to movement in the sieve tube.Journal of Cell Biology 32, 71–87.

Esau K. & Thorsch J. (1985) Sieve plate pores and plasmodesmata,the communication channels of the symplast – ultrastructuralaspects and developmental relations. American Journal of Bot-any 72, 1641–1653.

Evert R.F., Eschrich W. & Heyser W. (1977) Distribution andstructure of the plasmodesmata in mesophyll and bundle-sheathcells of Zea mays L. Planta 136, 77–89.

Fisher D.B. (1999) The estimated pore diameter for plasmodesmalchannels in the Abutilon nectary trichome should be about 4nm,rather than 3nm. Planta 208, 299–300.

Fletcher J.C., Brand U., Running M.P., Simon R. & MeyerowitzE.M. (1999) Signaling of cell fate decisions by CLAVATA3. Ara-bidopsis shoot meristems. Science 283, 1911–1914.

Foster T., Veit B. & Hake S. (1999) Mosaic analysis of the domi-nant mutant, Gnarley1-R, reveals distinct lateral and transversesignaling pathways during maize leaf development. Develop-ment 126, 305–313.

Franceschi V.R., Ding B. & Lucas W.J. (1994) Mechanism of plas-modesmata formation in characean algae in relation to evolu-tion of intercellular communication in higher plants. Planta 192,347–358.

Fujiwara T., Giesman-Cookmeyer D., Ding B., Lommel S.A. &Lucas W.J. (1993) Cell-to-cell trafficking of macromoleculesthrough plasmodesmata potentiated by the red clover necroticmosaic virus MP. Plant Cell 5, 1783–1794.

Gamalei Y.V. (1989) Structure and function of leaf minor veins intrees and herbs. Trees 3, 96–110.

Ghoshroy S., Lartey R., Sheng J. & Citovsky V. (1997) Transportof proteins and nucleic acids through plasmodesmata. AnnualReview of Plant Physiology and Plant Molecular Biology 48, 27–50.

Gilbertson R.L. & Lucas W.J. (1996) How do viruses traffic on thevascular highway? Trends in Plant Science 1, 260–268.

Gillespie T., Boevink P., Haupt S., Roberts A.G., Toth R., Valen-tine T., Chapman S. & Oparka K.J. (2002) Functional analysisof a DNA-shuffled movement protein reveals that microtubulesare dispensable for cell-to-cell movement of tobacco mosaicvirus. Plant Cell 14, 1207–1222.

Gisel A., Barella S., Hempel F.D. & Zambryski P.C. (1999) Tem-



poral and spatial regulation of symplastic trafficking duringdevelopment in Arabidopsis thaliana apices. Development 126,1879–1889.

Golecki B., Schulz A., Carstens-Behrens U. & Kollmann R. (1998)Evidence for graft transmission of structural phloem proteins ortheir precursors in heterografts of Cucurbitaceae. Planta 206,630–640.

Golecki B., Schulz A. & Thompson G.A. (1999) Translocation ofstructural P proteins in the phloem. Plant Cell 11, 127–140.

Goodwin P.B. (1983) Molecular size limit for movement in thesymplast of the Elodea leaf. Planta 157, 124–130.

Grabski S., de Feijter A.W. & Schindler M. (1993) Endoplasmicreticulum forms a dynamic continuum for lipid diffusionbetween contiguous soybean root cells. Plant Cell 5, 25–38.

Gunning B.E.S. & Overall R.L. (1983) Plasmodesmata and cell-to-cell transport in plants. Bioscience 33, 260–265.

Hake S. (2001) Transcription factors on the move. Trends in Genet-ics 17, 2–3.

Hamilton A.J. & Baulcombe D.C. (1999) A species of small anti-sense RNA in posttranscriptional gene silencing in plants. Sci-ence 286, 950–952.

Heinlein M., Epel B.L., Padgett H.S. & Beachy R.N. (1995) Inter-action of tobamovirus movement proteins with the plant cytosk-eleton. Science 270, 1983–1985.

Helariutta Y., Fukaki H., Wysocka-Diller J., Nakajima K., Jung J.,Sena G., Hauser M.-T. & Benfey P.N. (2000) The SHORT-ROOT gene controls radial patterning of the Arabidopsis rootthrough radial signaling. Cell 101, 555–567.

Hofius D., Herbers K., Melzer M., Omid A., Tacke E., Wolf S. &Sonnewald U. (2001) Evidence for expression level-dependentmodulation of carbohydrate status and viral resistance by thepotato leafroll virus movement protein in transgenic tobaccoplants. Plant Journal 28, 529–543.

Holdaway-Clarke T.L., Walker N.A., Hepler P.K. & Overall R.L.(2000) Physiological elevations in cytoplasmic free calcium bycold or ion injection result in transient closure of higher plantplasmodesmata. Planta 210, 329–335.

Holdaway-Clarke T.L., Walker N.A., Reid R.J., Hepler P.K. &Overall R.L. (2001) Cytoplasmic acidification with butyric aciddoes not alter the ionic conductivity of plasmodesmata. Proto-plasma 215, 184–190.

Huang Z., Andrianov V.M., Han Y. & Howell S.H. (2001) Iden-tification of Arabidopsis proteins that interact with the cauli-flower mosaic virus (CaMV) MP. Plant Molecular Biology 47,663–675.

Hughes J.E. & Gunning B.E.S. (1980) Glutaraldehyde-induceddeposition of callose. Canadian Journal of Botany 58, 250–258.

Hull R. (1989) The movement of viruses in plants. Annual Reviewof Phytopathology. 27, 213–240.

Hull R. (1992) Down the tube. Tubules extending from protoplastsinfected with cowpea mosaic virus may reveal how some virusesmove from cell to cell in infected plants. Current Biology 2, 224–226.

Iglesias V.A. & Meins F. Jr (2000) Movement of plant viruses isdelayed in a b-1,3-glucanase-deficient mutant showing a reducedplasmodesmatal SEL and enhanced callose deposition. PlantJournal 21, 157–166.

Imlau A., Truernit E. & Sauer N. (1999) Cell-to-cell and long-distance trafficking of the green fluorescent protein in thephloem and symplastic unloading of the protein into sink tissues.Plant Cell 11, 309–322.

Itaya A., Liang G., Woo Y.M., Nelson R.S. & Ding B. (2000)Nonspecific intercellular protein trafficking probed by green-fluorescent protein in plants. Protoplasma 213, 165–175.

Itaya A., Woo Y.-M., Masuta C., Bao Y., Nelson R.S. & Ding B.(1998) Developmental regulation of intercellular protein traf-

ficking through plasmodesmata in tobacco leaf epidermis. PlantPhysiology 118, 373–385.

Jackson D. (2001) The long and the short of it: signaling develop-ment through plasmodesmata. Plant Cell 13, 2569–2572.

Jackson D., Veit B. & Hake S. (1994) Expression of maizeKNOTTED1 related homeobox genes in the shoot apicalmeistem predicts patterns of morphogenesis in the vegetativeshoot. Development 120, 405–413.

Jenik P. & Irish V. (2001) The Arabidopsis floral homeotic geneAPETALA3 differentially regulates intercellular signalingrequired for petal and stamen development. Development 128,13–23.

Jones M.G.K. (1976) The origin and development of plasmodes-mata. In Intercellular Communication in Plants: Studies on Plas-modesmata (eds B.E.S. Gunning & A.W. Robards), pp. 81–105.Springer-Verlag, Berlin, Germany.

Jones D.L., Gilroy S., Larson P.B., Howell S.H. & Kochian L.V.(1998) Effect of aluminium on cytoplasmic Ca2+ homeostasis inroot hairs of Arabidopsis thaliana (L.). Planta 206, 378–387.

Jorgensen R.A., Atkinson R.G., Forster R.L.S. & Lucas W.J.(1998) An RNA-based information superhighway in plants. Sci-ence 279, 1486–1487.

Jørgensen K.E. & Møller J.V. (1979) Use of flexible polymers asprobes of glomerular pore size. American Journal of Physiology236, F103–F111.

Kauss H. (1987) Some aspects of calcium-dependent regulation inplant metabolism. Annual Review of Plant Physiology 38, 47–72.

Kim M., Canio W., Kessler S. & Sinha N. (2001) Developmentalchanges due to long-distance movement of a homeobox fusiontranscript in tomato. Science 293, 287–289.

Kim I., Hempel F.D., Sha K., Pfluger J. & Zambryski P.C. (2002a)Identification of a developmental transition in plasmodesmatalfunction during embryogenesis in Arabidopsis thaliana. Devel-opment 129, 1261–1272.

Kim J.Y., Yuan Z., Cilia M., Khalfan-Jagani Z. & Jackson D.(2002b) Intercellular trafficking of a KNOTTED1 green fluores-cent protein fusion in the leaf and shoot meristem of Arabidop-sis. Proceedings of the National Academy of Science, USA 99,4103–4108.

Kitajima E.W. & Lauritis J.A. (1969) Plant virions in plasmodes-mata. Virology 37, 681–685.

Knight A.E. & Kendrick-Jones J. (1993) A myosin-like proteinfrom a higher plant. Journal of Molecular Biology 231, 148–154.

Kotlizky G., Katz A., van der Laak J., Boyko V., Lapidot M.,Beachy R.N., Heinlein M. & Epel B.L. (2001) A dysfunctionalmovement protein of tobacco mosaic virus interferes with tar-geting of wild-type movement protein to microtubules. Molecu-lar Plant–Microbe Interactions 14, 895–904.

Kragler F., Lucas W.J. & Monzer J. (1998a) Plasmodesmata:dynamics, domains and patterning. Annals of Botany 81, 1–10.

Kragler F., Monzer J., Shash K., Xoconostle-Cázares B. & LucasW.J. (1998b) Cell-to-cell transport of proteins: requirement ofunfolding and characterisation of binding to a putative plas-modesmatal receptor. Plant Journal 15, 367–381.

Kragler F., Monzer J., Xoconostle-Cázares B. & Lucas W.J. (2000)Peptide antagonists of the plasmodesmal macromolecular traf-ficking pathway. EMBO Journal 19, 2856–2868.

Lachaud S. & Maurousset L. (1996) Occurrence of plasmodesmatabetween differentiating vessels and other xylem cells in Sorbustorminalis L. Crantz and their fate during xylem maturation.Protoplasma 191, 220–226.

Lalonde S., Boles E., Hellmann H., Barker L., Patrick J.W., From-mer W.B. & Ward J.M. (1999) The dual function of sugar carri-ers: transport and sugar sensing. Plant Cell 11, 707–726.

Lartney R.T., Ghoshroy S. & Citovsky V. (1998) Identification of



an Arabidopsis thaliana mutation (vsm1) that restricts systemicmovement of tobamoviruses. Molecular Plant–Microbe Interac-tions 11, 706–709.

Lee J.-Y., Yoo B.-C. & Lucas W.J. (2000) Parallels betweennuclear-pore and plasmodesmal trafficking of information mol-ecules. Planta 210, 177–187.

Leisner S.M. & Turgeon R. (1993) Movement of virus and photo-assimilate in the phloem: a comparative analysis. Bioessays 15,741–748.

Lew R.R. (1996) Pressure regulation of the electrical properties ofgrowing Arabidopsis thaliana L. root hairs. Plant Physiology112, 1089–1100.

Lucas W.J. & Gilbertson R.L. (1994) Plasmodesmata in relationto viral movement within leaf tissues. Annual Review of Phyto-pathology 32, 387–411.

Lucas W.J. & Wolf S. (1993) Plasmodesmata: the intercellularorganelles of green plants. Trends in Cell Biology 3, 308–315.

Lucas W.J., Bouché-Pillon S., Jackson D.P., Nguyen L., Baker L.,Ding B. & Hake S. (1995) Selective trafficking of KNOTTED1homeodomain protein and its mRNA through plasmodesmata.Science 270, 1980–1983.

Lucas W.J., Ding B. & van der Schoot C. (1993) Tansley Review,58. Plasmodesmata and the supracellular nature of plants. NewPhytologist 125, 435–476.

Madore M.A. & Lucas W.J. (1986) Characterization of the sourceleaf symplast by means of lucifer yellow CH. In PhloemTransport (eds J. Cronshaw, W.J. Lucas & R.T. Giaquinta), pp.129–133. Alan R. Liss, Inc., New York, USA.

Mahajan S.K., Chisholm S.T., Whitham S.A. & Carrington J.C.(1998) Identification and characterisation of a locus (RTM1)that restricts long-distance movement of tobacco etch virus inArabidopsis thaliana. Plant Journal 14, 177–186.

le Maire M., Aggerbeck L.P., Monteilhet C., Andersen J.P. &Moller J.V. (1986) The use of high-performance liquid chro-matography for the determination of size and molecular weightof proteins: a caution and a list of membrane proteins suitableas standards. Analytical Biochemistry 154, 525–535.

Mallory A.C., Ely L., Smith T.H., Marathe R., Anandalakshmi R.,Fagard M., Vaucheret H., Pruss G., Bowman L. & Vance V.B.(2001) HC-Pro suppression of transgene silencing eliminates thesmall RNAs but not transgene methylation or the mobile signal.Plant Cell 13, 571–583.

Martindale V.E. & Salisbury J.L. (1990) Phosphorylation of algalcentrin is rapidly responsive to changes in the external milieu.Journal of Cell Science 96, 395–402.

Mas P. & Beachy R.N. (1999) Replication of tobacco mosaic viruson endoplasmic reticulum and role of the cytoskeleton and virusmovement protein in intracellular distribution of viral RNA.Journal of Cell Biology 147, 945–958.

Mas P. & Beachy R.N. (2000) Role of microtubules in the intrac-ellular distribution of tobacco mosaic virus movement protein.Proceedings of the National Academy of Science, USA 97, 12345–12349.

Matsushita Y., Deguchi M., Youda M., Nishiguchi M. & NyunoyaH. (2001) The tomato mosaic tobamovirus movement proteininteracts with a putative transcriptional coactivator KELP. Mol-ecules and Cells 12, 57–66.

Matsushita Y., Hanazawa K., Yoshioka K., Oguchi T., KawakamiS., Watanabe Y., Nishiguchi M. & Nyunoya H. (2000) In vitrophosphorylation of the movement protein of tomato mosaictobamovirus by a cellular kinase. Journal of General Virology81, 2095–2102.

Matsushita Y., Miyakawa O., Deguchi M., Nishiguchi M. &Nyunoya H. (2002a) Cloning of a tobacco cDNA coding for aputative transcriptional coactivator MBF1 that interacts with

the tomato mosaic virus movement protein. Journal of Experi-mental Botany 53, 1521–1532.

Matsushita Y., Yoshioka K., Shigyo T., Takahashi H. & NyunoyaH. (2002b) Phosphorylation of the movement protein of Cucum-ber mosaic virus in transgenic tobacco plants. Virus Genes 24,231–234.

Maule A.J. (1991) Virus movement in infected plants. CriticalReviews in Plant Sciences 9, 457–473.

Maule A.J. (1994) Plant virus movement: de novo process or rede-ployed machinery. Trends in Microbiology 2, 305–306.

McLean B.G., Hempel F.D. & Zambryski P.C. (1997) Plant inter-cellular communication via plasmodesmata. Plant Cell 9, 1043–1054.

McLean B.G., Waigmann E., Citovsky V. & Zambryski P. (1993)Cell-to-cell movement of plant viruses. Trends in Microbiology1, 105–109.

McLean B.G., Zupan J. & Zambryski P.C. (1995) Tobacco mosaicvirus MP associates with the cytoskeleton in tobacco cells. PlantCell 7, 2101–2114.

Mezitt L.A. & Lucas W.J. (1996) Plasmodesmal cell-to-cell trans-port of proteins and nucleic acids. Plant Molecular Biology 32,251–273.

Mezitt Provencher L., Miao L., Sinha N. & Lucas W.J. (2001)Sucrose Export Defective1 encodes a novel protein implicated inchloroplast-to-nucleus signaling. Plant Cell 13, 1127–1141.

Moore A.E. & Stone B.A. (1972) Effect of infection with TMVand other viruses on the level of b-1,3-glucan hydrolase in leavesof Nicotiana glutinosa. Virology 50, 791–798.

Morvan O., Quentin M., Jauneau A., Mareck A. & Morvan C.(1998) Immunogold localization of pectin methylesterases in thecortical tissues of flax hypocotyl. Protoplasma 202, 175–184.

Münch E. (1930) Die Stoffbewegungen in der Pflanze. G. Fisher,Jena, Germany.

Murillo I., Cavallarin L. & San Segundo B. (1997) The maizepathogenesis-related PRms protein localizes to plasmodesmatain maize radicles. Plant Cell 9, 145–156.

Nakajima K., Sena G., Nawy T. & Benfey P.N. (2001) Intercellularmovement of the putative transcription factor SHR in root pat-terning. Nature 413, 307–311.

Nelson R.S. & van Bel A.J.E. (1998) The mystery of virus traffick-ing into, through and out of vascular tissue. Progress in Botany59, 476–533.

Noueiry A.O., Lucas W.J. & Gilbertson R.L. (1994) Two proteinsof a plant DNA virus coordinate nuclear and plasmodesmaltransport. Cell 76, 925–932.

Olesen P. (1979) The neck constriction in plasmodesmata. Planta144, 349–358.

Olesen P. & Robards A.W. (1990) The neck region of plasmodes-mata: General architecture and some functional aspects. In Par-allels in Cell to Cell Junctions in Plants and Animals (ed. A.W.Robards, W.J. Lucas, J.D. Pitts, H.J. Jongsma & D.C. Spray),pp. 145–170. Springer-Verlag, Berlin, Germany.

Oparka K.J. & Prior D.A.M. (1992) Direct evidence for pressure-generated closure of plasmodesmata. Plant Journal 2, 741–750.

Oparka K.J. & Roberts A.G. (2001) Plasmodesmata. A not soopen-and-shut case. Plant Physiology 125, 123–126.

Oparka K.J. & Santa Cruz S. (2000) The great escape: phloemtransport and unloading of macromolecules. Annual Review ofPlant Physiology and Plant Molecular Biology 51, 323–347.

Oparka K.J. & Turgeon R. (1999) Sieve elements and companioncells – traffic control centres of the phloem. Plant Cell 11, 739–750.

Oparka K.J., Murphy R., Derrick P.M., Prior D.A.M. & SmithJ.A.C. (1991) Modification of the pressure-probe technique per-mits controlled intracellular microinjection of fluorescentprobes. Journal of Cell Science 98, 539–544.



Oparka K.J., Roberts A.G., Boevink P., Santa Cruz S., RobertsI.M., Pradel K.S., Imlau A., Kotlizky G., Sauer N. & Epel B.(1999) Simple, but not branched, plasmodesmata allow the non-specific trafficking of proteins in developing tobacco leaves. Cell97, 743–754.

Ormenese S., Havelange A., Deltour R. & Bernier G. (2000) Thefrequency of plasmodesmata increases early in the whole shootapical meristem of Sinapsis alba L. during floral transition.Planta 211, 370–375.

Overall R.L. & Blackman L.M. (1996) A model of the macromo-lecular structure of plasmodesmata. Trends in Plant Science 1,307–311.

Overall R.L., Wolfe J. & Gunning B.E.S. (1982) Intercellular com-munication in Azolla roots. I. Ultrastructure of plasmodesmata.Protoplasma 111, 134–150.

Owens R.A., Blackburn M. & Ding B. (2001) Possible involvementof the phloem lectin in long-distance viroid movement. Molec-ular Plant–Microbe Interactions 14, 905–909.

Palauqui J.-C. & Vaucheret H. (1998) Transgenes are dispensablefor the RNA degredation step of cosuppression. Proceedings ofthe National Academy of Science, USA 95, 9675–9680.

Palevitz B.A. & Hepler P.K. (1985) Changes in dye coupling ofstomatal cells of Allium and Commelina demonstrated by micro-injection of Lucifer yellow. Planta 164, 473–479.

Pennazio S., D’Agostino G., Appiano A. & Redolfi P. (1978)Ultrastructure and histochemistry of the resistant tissue sur-rounding lesions of tomato bushy stunt virus in Gomphrenaglobosa leaves. Physiological Plant Pathology 13, 65–171.

Perbal M.-C., Haughn G., Saedler H. & Schwarz-Sommer Z.(1996) Non-cell-autonomous function of the Antirrhinum floralhomeotic proteins DEFICIENS and GLOBOSA is exerted bytheir polar cell-to-cell trafficking. Development 122, 3433–3441.

Peterson C.A. & Rauser W.E. (1979) Callose deposition and pho-toassimilate export in Phaseolus vulgaris exposed to excesscobalt, nickel and zinc. Plant Physiology 63, 1170–1174.

Petty I.T.D. & Jackson A.O. (1990) Mutational analysis of barleystripe mosaic virus RNA b. Virology 179, 712–718.

Pfluger J. & Zambryski P.C. (2001) Cell growth: the power ofsymplastic isolation. Current Biology 11, 436–439.

Poethig R.S. & Sussex I.M. (1985) The cellular parameters of leafdevelopment in tobacco: a clonal analysis. Planta 165, 170–184.

Price W.C. (1966) Flexuous rods in phloem cells of lime plantsinfected with citrus tristeza virus. Virology 29, 285–294.

Prüfer D., Schmitz J., Tacke E., Kull B. & Rohde W. (1997) In vivoexpression of a full-length cDNA copy of potato leafroll virus(PLRV) in protoplasts and transgenic plants. Molecular GeneralGenetics 253, 609–614.

Quader H., Hofmann A. & Schnepf E. (1987) Shape and move-ment of the endoplasmic reticulum in onion bulb epidermis cells:possible involvement of actin. European Journal of Cell Biology44, 17–26.

Radford J.E. & White R.G. (1998) Localization of a myosin-likeprotein to plasmodesmata. Plant Journal 14, 743–750.

Reichelt S. & Kendrick-Jones J. (2000) Myosins. In Actin: aDynamic Framework for Multiple Plant Cell Functions (ed. C.J.Staiger, F. Balus̆ka, D. Volkmann & P. Barlow), pp. 29–44.Kluwer Academic Publishers, Dordrecht, The Netherlands.

Reichelt S., Knight A.E., Hodge T.P., Baluska F., Samaj J., Volk-mann D. & Kendrick-Jones J. (1999) Characterization of theunconventional myosin VIII in plant cells and its localization atthe post-cytokinetic cell wall. Plant Journal 19, 555–567.

Reid R.J. & Overall R.L. (1992) Intercellular communication inChara: factors affecting transnodal electrical resistance and sol-ute fluxes. Plant, Cell and Environment 15, 507–517.

Rinne P.L.H., Kaikuranta P.M. & van der Schoot C. (2001) Theshoot apical meristem restores its symplastic organisation during

chilling-induced release from dormancy. Plant Journal 26, 249–264.

Rinne P.L.H. & van der Schoot C. (1998) Symplasmic fields in thetunica of the shoot apical meristem coordinate morphogeneticevents. Development 125, 1477–1485.

Robards A.W. (1971) The ultrastructure of plasmodesmata.Protoplasma 72, 315–323.

Robards A.W. (1976) Plasmodesmata in higher plants. In Intercel-lular Communication in Plants: Studies on Plasmodesmata (edsB.E.S. Gunning & A.W. Robards), pp. 15–57. Springer-Verlag,Berlin, Germany.

Robards A.W. & Lucas W.J. (1990) Plasmodesmata. AnnualReview of Plant Physiology 41, 369–419.

Roberts D.A. (1952) Independent translocation of sap-transmiss-able viruses. Phytopathology 42, 381–387.

Roberts I.M., Boevink P., Roberts A.G., Sauer N., Reichel C. &Oparka K.J. (2001) Dynamic changes in the frequency andarchitecture of plasmodesmata during the sink-source transitionin tobacco leaves. Protoplasma 218, 31–44.

Roberts A.G., Santa Cruz S., Roberts I.M., Prior D.A.M., TurgeonR. & Oparka K.J. (1997) Phloem unloading in sink leaves ofNicotiana benthamiana: comparison of a fluorescent solute witha fluorescent virus. Plant Cell 9, 1381–1396.

Robinson-Beers K. & Evert R.F. (1991) Fine structure of plas-modesmata in mature leaves of sugarcane. Planta 184, 307–318.

Rojo E., Sharma V.K., Kovaleva V., Raikhel N.V. & Fletcher J.C.(2002) CLV3 is localised to the extracellular space, where itactivates the Arabidopsis CLAVATA stem cell signaling path-way. Plant Cell 14, 969–977.

Ruan Y.-L., Llewellyn D.J. & Furbank R.T. (2001) The control ofsingle-celled cotton fiber elongation by developmentally revers-ible gating of plasmodesmata and coordinated expression ofsucrose and K+ transporters and expansin. Plant Cell 13, 47–60.

Ruiz-Medrano R., Xoconostle-Cázares B. & Lucas W.J. (1999)Phloem long-distance transport of CmNACP mRNA: implica-tions for supracellular regulation in plants. Development 126,4405–4419.

Ruiz-Medrano R., Xoconostle-Cázares B. & Lucas W.J. (2001)The phloem as a conduit for inter-organ communication. Cur-rent Opinion in Plant Biology 4, 202–209.

Russin W.A., Evert R.F., van der Veer P.J., Sharkey T.D. & BriggsS.P. (1996) Modification of a specific class of plasmodesmata andloss of sucrose export ability in the sucrose export defective1maize mutant. Plant Cell 8, 645–658.

Ryabov E.V., Robinson D.J. & Taliansky M.E. (1999) A plantvirus-encoded protein facilitates long-distance movement ofheterologous viral RNA. Proceedings of the National Academyof Science, USA 96, 1212–1217.

Samuel G. (1934) The movement of tobacco mosaic virus withinthe plant. Annals of Applied Biology 21, 90–111.

Santa Cruz S. (1999) Perspective: phloem transport of viruses andmacromolecules – what goes in must come out. Trends in Micro-biology 7, 237–241.

Santa Cruz S., Roberts A.G., Prior D.A.M., Chapman S. &Oparka K.J. (1998) Cell-to-cell and phloem-mediated transportof potato virus X: the role of virions. Plant Cell 10, 495–510.

Satoh H., Matsuda H., Kawamura T., Isogai M., Yoshikawa N. &Takahashi T. (2000) Intracellular distribution, cell-to-cell traf-ficking and tubule-inducing activity of the 50 kDa MP of applechlorotic leaf spot virus fused to green fluorescent protein. Jour-nal of General Virology 81, 2085–2093.

Scholthof H.B., Scholthof K.-B.G., Kikkert M. & Jackson A.O.(1995) Tomato bushy stunt virus spread is regulated by twonested genes that function in cell-to-cell movement and host-dependent systemic invasion. Virology 213, 425–438.

Schulz A. (1995) Plasmodesmal widening accompanies the short-



term increase in symplasmic phloem unloading in pea root tipsunder osmotic stress. Protoplasma 188, 22–37.

Schulz A. (1999) Physiological control of plasmodesmal gating. InPlasmodesmata. Structure, Function, Role in Cell Communica-tion (eds A.J.E. van Bel & W.J.P. van Kesteren), pp. 173–204.Springer, Berlin, Germany.

Sessions A., Yanofsky M.F. & Weigel D. (2000) Cell-cell signalingand movement by the floral transcription factors LEAFY andAPETALA1. Science 289, 779–781.

Sharma V.K. & Fletcher J.C. (2002) Maintenance of shoot andfloral meristem cell proliferation and fate. Plant Physiology 129,31–39.

Sivaguru M., Fujiwara T., Samaj J., Baluska F., Yang Z., OsawaH., Maeda T., Mori T., Volkmann D. & Matsumoto H. (2000)Aluminium-inducedÆ3-b-D-Glucan inhibits cell-to-cell traffick-ing of molecules through plasmodesmata. A new mechanismof aluminium toxicity in plants. Plant Physiology 124, 991–1005.

Sokolova M., Prüfer D., Tacke E. & Rohde W. (1997) The potatoleafroll virus 17K movement protein is phosphorylated by amembrane-associated protein kinase from potato with biochem-ical features related to protein kinase C. FEBS Letters 400, 201–205.

Solovyev A.G., Stroganova T.A., Zamyatnin A.A., Fedorkin O.N.,Schiemann J. & Morozov S.Y. (2000) Subcellular sorting ofsmall membrane-associated triple gene block proteins: TGBp3-assisted targeting of TGBp2. Virology 269, 113–127.

Storms M.M.H., Kormelink R., Peters D., van Lent J.W.M. &Goldbach R.W. (1995) The nonstructural NSm protein oftomato spotted wilt virus induces tubular structures in plant andinsect cells. Virology 214, 485–493.

Sundell C.L. & Singer R.H. (1991) Requirement of microfilamentsin sorting of actin messenger RNA. Science 253, 1275–1277.

Susi P. (2000) Dye-coupling in tobacco mesophyll cells surround-ing growing tobacco mosaic tobamovirus-induced local lesions.Journal of Phytopathology 148, 379–382.

Sussex I.M. (1951) Experiments on cause of dorsiventrality inleaves. Nature 167, 651–652.

Sussex I.M. (1989) Developmental programming of the shoot mer-istem. Cell 56, 225–229.

Szymkowiak E.J. & Sussex I.M. (1992) The internal meristem layer(L3) determines floral meristem size and carpel number intomato periclinal chimeras. Plant Cell 4, 1089–1100.

Szymkowiak E.J. & Sussex I.M. (1993) Effect of lateral suppressoron petal initiation in tomato. Plant Journal 4, 1–7.

Terry B.R. & Robards A.W. (1987) Hydrodynamic radius alonegoverns the mobility of molecules through plasmodesmata.Planta 171, 145–157.

Thomas C.L. & Maule A.J. (1995) Identification of the cauliflowermosaic virus MP RNA-binding domain. Virology 206, 1145–1149.

Tilney L.G., Cooke T.J., Connelly P.S. & Tilney M.S. (1991) Thestructure of plasmodesmata as revealed by plasmolysis, deter-gent extraction, and protease digestion. Journal of Cell Biology112, 739–747.

Tomenius K., Clapham D. & Meshi T. (1987) Localization byimmunogold cytochemistry of the virus-coded 30K protein inplasmodesmata of leaves infected with tobacco mosaic virus.Virology 160, 363–371.

Tucker E.B. (1982) Translocation in the staminal hairs of Set-creasea purpurea. I. A study of cell ultrastructure and cell-to-cellpassage of molecular probes. Protoplasma 113, 193–201.

Tucker E.B. (1987) Cytoplasmic streaming does not drive intercel-lular passage in staminal hairs of Setcreasea purpurea. Proto-plasma 137, 140–144.

Tucker E.B. (1990) Calcium-loaded 1,2-bis (2-aminophenoxy)

ethane-N,N,N¢,N¢-tetraacetic acid blocks cell-to-cell diffusion ofcarboxyfluorescein in staminal hairs of Setcreasea purpurea.Planta 182, 34–38.

Tucker J.E., Mauzerall D. & Tucker E.B. (1989) Symplastic trans-port of carboxyfluorescein in staminal hairs of Setcreasea pur-purea is diffusive and includes loss to the vacuole. PlantPhysiology 90, 1143–1147.

Turner A., Wells B. & Roberts K. (1994) Plasmodesmata of maizeroot tips: structure and composition. Journal of Cell Science 107,3351–3361.

Unseld S., Höhnle M., Ringel M. & Frischmuth T. (2001) Subcel-lular targeting of the coat protein of African cassava mosaicgeminivirus. Virology 286, 373–383.

Vaucheret H., Elmayan T., Elmayan T., Feuerbach F., Godon C.,Morel J.-B., Mourrain P., Palauqui J.-C. & Vernhettes S. (1998)Transgene-induced gene silencing in plants. Plant Journal 16,651–659.

Vogel H., Grieninger G.E. & Zetsche K.H. (2002) Differentialmessenger RNA gradients in the unicellular alga Acetabulariaacetabulum. Role of the cytoskeleton. Plant Physiology 129,1407–1416.

Voinnet O., Pinto Y.M. & Baulcombe D.C. (1999) Suppression ofgene silencing: a general strategy used by diverse DNA andRNA viruses of plants. Proceedings of the National Academy ofScience, USA 96, 14147–14152.

Voinnet O., Vain P., Angell S. & Baulcombe D.C. (1998) Systemicspread of sequence-specific transgene RNA degradation inplants is initiated by localized introduction of ectopic promoter-less DNA. Cell 95, 177–187.

Waigmann E. & Zambryski P. (1995) Tobacco mosaic virus move-ment protein-mediated protein transport between trichomecells. Plant Cell 7, 2069–2079.

Waigmann E., Chen M.-H., Bachmaier R., Ghoshroy S. & Cit-ovsky V. (2000) Regulation of plasmodesmal transport by phos-phorylation of tobacco mosaic virus cell-to-cell MP. EMBOJournal 19, 4875–4884.

Waigmann E., Lucas W.J., Citovsky V. & Zambryski P. (1994)Direct functional assay for tobacco mosaic virus cell-to-cellmovement protein and identification of a domain involved inincreasing plasmodesmatal permeability. Proceedings of theNational Academy of Science, USA 91, 1433–1437.

Waigmann E., Turner A., Peart J., Roberts K. & Zambryski P.(1997) Ultrastructural analysis of leaf trichome plasmodesmatareveals major differences from mesophyll plasmodesmata.Planta 203, 75–84.

Weintraub M., Ragetli H.W.J. & Leung E. (1976) Elongated virusparticles in plasmodesmata. Journal of Ultrastructural Research56, 351–364.

Wells A.L., Lin A.W., Chen L.-Q., Safer D., Cain S.M., Hasson T.,Carragher B.I., Milligan R.A. & Sweeney H.L. (1999) MyosinVI is an actin-based motor that moves backwards. Nature 401,505–508.

White R.G., Badelt K., Overall R.L. & Vesk M. (1994) Actinassociated with plasmodesmata. Protoplasma 180, 169–184.

Whitham S.A., Anderberg R.J., Chisholm S.T. & Carrington J.C.(2000) Arabidopsis RTM2 gene is necessary for specific restric-tion of tobacco etch virus and encodes a small heat shock-likeprotein. Plant Cell 12, 569–582.

Wille A.C. & Lucas W.J. (1984) Ultrastructural and histochemicalstudies on guard cells. Planta 160, 129–142.

Williams L.E., Lemoine R. & Sauer N. (2000) Sugar transportersin higher plants – a diversity of roles and complex regulation.Trends in Plant Science 5, 283–290.

Wolf S., Deom C.M., Beachy R.N. & Lucas W.J. (1989) MP oftobacco mosaic virus modifies plasmodesmatal SEL. Science246, 377–379.



Wolf S., Deom C.M., Beachy R. & Lucas W.J. (1991) Plasmodes-matal function is probed using transgenic tobacco plants thatexpress a virus MP. Plant Cell 3, 593–604.

Wright K.M. & Oparka K.J. (1997) Metabolic inhibitors inducesymplastic movement of solutes from the transport phloem ofArabidopsis roots. Journal of Experimental Botany 48, 1807–1814.

Wu J.H. & Dimitman J.E. (1970) Leaf structure and callose for-mation as determinants of TMV movement in bean leaves asrevealed by UV irradiation studies. Virology 40, 820–827.

Wu X., Weigel D. & Wigge P.A. (2002) Signaling in plants byintercellular RNA and protein movement. Genes and Develop-ment 16, 151–158.

Xoconostle-Cázares B., Xiang Y., Ruiz-Medrano R., Wang H.-I.,Monzer J., Yoo B.-C., McFarland K.C., Franceschi V.R. & LucasW.J. (1999) Plant paralog to viral MP that potentiates transportof mRNA into the phloem. Science 283, 94–98.

Yahalom A., Lando R., Katz A. & Epel B.L. (1998) Acalcium-dependent protein kinase is associated with maizemesocotyl plasmodesmata. Journal of Plant Physiology 153,354–362.

Zambryski P.C. & Crawford K.M. (2000) Plasmodesmata: gate-keepers for cell-to-cell transport of developmental signals inplants. Annual Review of Cell and Developmental Biology 16,393–421.

Zee S.-Y. (1969) Fine structure of the differentiating sieveelements of Vicia faba. American Journal of Botany 17, 441–456.

Zhang W.H. & Tyerman S.D. (1997) Effect of low oxygen concen-tration on the electrical properties of cortical cells of wheatroots. Journal of Plant Physiology 150, 567–572.

Received 26 April 2002; received in revised form 19 July 2002;accepted for publication 21 July 2002

Plasmodesmata and the control of symplastic transport

Documents

Transcript of Plasmodesmata and the control of symplastic transport