Plant Cell, Vol. 10, 525-538, April 1998, Copyright © 1998, American Society of Plant Physiologists
The Movement Protein of Cucumber Mosaic Virus Traffics into Sieve Elements in Minor Veins of Nicotiana clevelandii
Leila M. Blackman1,a,
Petra Boevinka,
Simon Santa Cruza,
Peter Palukaitisa, and
Karl J. Oparkaa
a Unit of Cell Biology, Department of Virology, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, United Kingdom
Correspondence to:
Karl J. Oparka, kopark{at}scri.sari.ac.uk (E-mail), 44-1382-562426 (fax).
 |
ABSTRACT |
|---|
The location of the 3a movement protein (MP) of cucumber mosaic virus (CMV) was studied by quantitative immunogold labeling of the wild-type 3a MP in leaves of Nicotiana clevelandii infected by CMV as well as by using a 3a–green fluorescent protein (GFP) fusion expressed from a potato virus X (PVX) vector. Whether expressed from CMV or PVX, the 3a MP targeted plasmodesmata and accumulated in the central cavity of the pore. Within minor veins, the most extensively labeled plasmodesmata were those connecting sieve elements and companion cells. In addition to targeting plasmodesmata, the 3a MP accumulated in the parietal layer of mature sieve elements. Confocal imaging of cells expressing the 3a–GFP fusion protein showed that the 3a MP assembled into elaborate fibrillar formations in the sieve element parietal layer. The ability of 3a–GFP, expressed from PVX rather than CMV, to enter sieve elements demonstrates that neither the CMV RNA nor the CMV coat protein is required for trafficking of the 3a MP into sieve elements. CMV virions were not detected in plasmodesmata from CMV-infected tissue, although large CMV aggregates were often found in the parietal layer of sieve elements and were usually surrounded by 3a MP. These data suggest that CMV traffics into minor vein sieve elements as a ribonucleoprotein complex that contains the viral RNA, coat protein, and 3a MP, with subsequent viral assembly occurring in the sieve element parietal layer.
|
|
INTRODUCTION |
|---|
The movement of systemic plant viruses occurs in two distinct phases. In the first, the virus, or an infectious component of the virus, moves from cell to cell through plasmodesmata, the small channels that interconnect higher plant cells (reviewed in Maule 1991 
; Lucas and Gilbertson 1994 ; Carrington et al. 1996 ; Gilbertson and Lucas 1996 ). This movement entails passage of the virus through a range of interconnecting tissues and is often referred to as local movement (see Carrington et al. 1996 ). In an infected leaf, local movement involves the transport of virus from the epidermis, which is usually the first tissue to be mechanically infected, through mesophyll tissues and eventually to minor veins within which the virus gains access to the phloem. The final step in local virus movement involves the entry of the virus into the sieve element before its long-distance transport in the phloem (systemic movement). Once in the translocation stream, the movement of the virus to sink tissues is determined predominantly by source–sink relationships (Leisner et al. 1992 , Leisner et al. 1993 ; Leisner and Turgeon 1993 ; Carrington et al. 1996 ) and by the symplastic pathways available for assimilate unloading in developing sink tissues (Wang and Maule 1994 ; Roberts et al. 1997 ).
The local movement of viruses in leaf mesophyll tissues has been studied extensively. For several viruses, cell-to-cell movement through plasmodesmata is thought to be regulated by virally encoded movement proteins (MPs) that have a range of functions, including binding to single-stranded RNA (Citovsky et al. 1992 ), targeting to plasmodesmata (Tomenius et al. 1987 ; Oparka et al. 1997 ), and modifying the size exclusion limit of the plasmodesmal pore ("gating"; Wolf et al. 1989 ; Derrick et al. 1990 ; Waigmann and Zambryski 1995 ; Angell et al. 1996 ; Oparka et al. 1997 ). In addition to MPs, some viruses also require the coat protein (CP) to facilitate local spread (reviewed in Gilbertson and Lucas 1996 ). However, whether the CP acts independently or in concert with MPs is unclear.
To date, studies of MP function have been restricted to epidermal (e.g., Derrick et al. 1990 ; Oparka et al. 1996a , Oparka et al. 1997 ) or mesophyll cells (e.g., Wolf et al. 1989 ; Ding et al. 1992 , Ding et al. 1995 ; Fujiwara et al. 1993 ; Vaquero et al. 1994 ), because these tissues have proven to be readily accessible to microinjection procedures. The role of viral MPs in local movement within minor leaf veins remains unknown, due mainly to the inaccessibility of vein networks for microinjection and because of the difficulty in separating short-distance and long-distance transport components (Carrington et al. 1996 ). Recent evidence suggests that the boundary between the bundle sheath and phloem parenchyma may possess unusual plasmodesmata that require additional modification before some viruses can enter the phloem of minor veins (Ding et al. 1992 ; Wintermantel et al. 1997 ; reviewed in Nelson and Van Bel 1997 ). However, a direct demonstration that MPs play a role in virus entry into the phloem has not been reported.
The genome of cucumber mosaic virus (CMV) consists of five genes expressed from three genomic and two subgenomic RNAs (Palukaitis et al. 1992 ; Ding et al. 1994 ). RNA 3 contains an open reading frame that encodes a 30-kD MP (termed the 3a MP) that shares limited sequence similarity with the MP of tobacco mosaic virus (TMV; Melcher 1990 ). The 3a MP of CMV, when fluorescently labeled, has been shown to traffic through mesophyll cells after microinjection (Ding et al. 1995 ); in transgenic plants expressing the 3a MP, mesophyll plasmodesmata allow the nonspecific passage of 10-kD dextrans (Vaquero et al. 1994 ; Ding et al. 1995 ). More recently, the 3a MP was shown to target specifically the plasmodesmata of different cell types during cell-to-cell virus movement when expressed from a potato virus X (PVX) vector as a fusion with the green fluorescent protein (GFP) (Oparka et al. 1996b ). However, as in the case of MPs from other viruses, the exact role of the 3a MP in mediating the entry of CMV into the phloem is as yet unknown.
In this study, we compare the immunolocalization of the CMV 3a MP expressed during a wild-type infection with that of a 3a–GFP fusion expressed from a PVX vector (Oparka et al. 1996b ). We show, by using quantitative immunogold labeling of virally infected leaf tissues, that the wild-type 3a MP targets all plasmodesmata leading to the minor vein, including the specialized plasmodesmata that connect the sieve element and companion cells. More significantly, we demonstrate that the 3a MP enters the minor vein sieve elements and accumulates subsequently in elaborate formations within the sieve element parietal layer. We also show that an interaction of the 3a MP with either the CMV RNA or CMV CP is not required for MP entry into the sieve element. Finally, we provide evidence suggesting that complete virus assembly occurs within minor vein sieve elements. These results demonstrate that the CMV 3a MP is directly involved in the final entry of an infectious viral component into the minor vein phloem and implicate the 3a MP in potentiating the systemic spread of CMV throughout the plant.
|
|
RESULTS |
|---|
The Plasmodesmal Pathway to Minor Veins
To examine the potential pathways for virus movement between the epidermis and minor veins, we scored ultrathin sections of CMV-infected leaf lamina for plasmodesmata that were immunogold labeled with the 3a MP (Table 1). The data show that plasmodesmata between all cells of the mesophyll tissues and also those between cells within the bundle sheath of minor veins were heavily labeled with the viral MP. An examination of 705 plasmodesmata from inoculated source leaves revealed that between 20 and 70% of the plasmodesmata observed in thin sections were labeled with the 3a MP. The most extensively labeled plasmodesmata were those that connected sieve elements and companion cells (70%), and plasmodesmata at this interface also showed the highest density of gold labeling (Table 1). A similar examination of 703 plasmodesmata from leaf tissues infected with PVX expressing a 3a–GFP fusion (PVX.3a–GFP) showed that there was no significant difference between the total percentage of labeled plasmodesmata observed in wild-type CMV-infected leaf tissue and tissue infected by PVX.3a–GFP (Mann-Whitney test; 95% confidence level [ Freund 1988 ]; selected examples of immunolabeled plasmodesmata connecting different cell types in the leaf are shown in Figure 1A to D).
View larger version (74K):
[in this window]
[in a new window]
|
Figure 1.
Plasmodesmata Immunogold Labeled with Antibodies Raised against the CMV 3a MP.
(A) and (B) The 3a MP was expressed from CMV during a wild-type infection.
(C) and (D) The 3a MP was expressed as a 3a–GFP fusion by using a PVX vector.
b, bundle sheath; c, companion cell; m, mesophyll; p, vascular parenchyma element; s, sieve element. Bars in (A) to (D) = 200 µm.
|
|
View this table:
[in this window]
[in a new window]
|
Table 1.
Distribution of the CMV 3a MP in Plasmodesmata Connecting Different Cell Types in Infected N. clevelandii Leaves
|
|
3a–GFP Targeting of Plasmodesmata
When sections of a leaf infected with PVX.3a–GFP were examined under the confocal laser scanning microscope, punctate GFP labeling was apparent between several cell types in the leaf (Figure 2A to E). Labeled plasmodesmata connecting palisade mesophyll cells with spongy mesophyll cells were often present in discrete pit fields when seen in favorable optical sections by using the confocal laser scanning microscope (Figure 2B). Confirmation that the GFP- labeled regions of the wall were plasmodesmata was obtained by colocalization of the callose associated with the neck region of the plasmodesmal pores (Figure 2C). As in the case of the CMV-infected tissue, 3a–GFP expressed from PVX was localized to the central cavity of plasmodesmata in infected cells by immunogold labeling (Figure 1C and Figure 1D). Similarly, the 3a–GFP fusion was found in plasmodesmata at all cell interfaces. Thus, the presence of the PVX vector did not impede the ability of 3a–GFP to target plasmodesmata.
View larger version (62K):
[in this window]
[in a new window]
|
Figure 2.
Fluorescence Localization of 3a–GFP Expressed from a PVX Vector.
Figure 2. (continued).
(A) Punctate fluorescence from cell walls connecting epidermal cells. Bar = 10 µm.
(B) Plasmodesmal pit field at the junction of a palisade mesophyll cell with a spongy mesophyll cell. Bar = 10 µm.
(C) Colocalization of callose (red) with 3a–GFP (green) in a spongy mesophyll pit field. Bar = 1 µm.
(D) Intense 3a–GFP fluorescence in leaf minor vein networks. Bar = 200 µm.
(E) Higher magnification of a minor vein showing a 3a–GFP—labeled sieve element (S) running alongside a xylem element (X). Point fluorescence also occurs in mesophyll plasmodesmata (arrows). Bar = 25 µm.
(F) and (G) Double labeling of a GFP-containing sieve element (F) with aniline blue (G) reveals the position of a sieve plate (arrowheads) that appears as a dark band in (F). Bar in (F) = 10 µm for (F) and (G).
(H) to (J) 3a–GFP fluorescence from within individual minor vein sieve elements. At low gain settings on the confocal laser scanning microscope, punctate regions of fluorescence are seen to be connected by threadlike extensions (arrow in [H]). At a higher gain setting (I), the fluorescence from the punctate regions saturates but reveals an additional extensive GFP-labeled network (arrow in [I]). A confocal section of the sieve element parietal layer revealing the 3a–GFP—labeled network is shown in (J). Bar in (H) = 5 µm for (H) to (J).
|
|
The 3a MP Accumulates in Minor Vein Sieve Elements
An examination of several minor veins from CMV-infected leaf tissue by immunogold labeling showed that the mature sieve elements were heavily labeled with 3a MP antibodies (Figure 3A, Figure 3C, and Figure 3D). Immunogold labeling was particularly strong in the parietal layer of the sieve elements and also across sieve plates and lateral sieve areas (e.g., Figure 3A, Figure 3C, and Figure 3D), giving the sieve elements a stronger MP signal than that detected in any of the infected cells en route to the minor veins. In contrast to the sieve element parietal layer, the lumen of the sieve element, including dispersed P protein filaments, showed no immunogold labeling. Control sections from uninfected leaf tissues showed no labeling of the sieve elements with the 3a MP antibody (data not shown).
View larger version (117K):
[in this window]
[in a new window]
|
Figure 3.
Immunogold Labeling of Minor Vein Sieve Elements with an Antibody Raised against the CMV 3a MP.
Figure 3. (continued).
(A) Wild-type CMV infection. Gold labeling of the 3a MP is intense along the sieve element parietal layer. Labeling is intense close to lateral sieve
areas and plasmodesmata that connect the sieve element with companion cells (arrowheads). An unlabeled aggregate of virus particles bounded by a membrane (open arrowheads) occurs in the parietal layer. Bar = 250 nm.
(B) Sieve elements labeled with 3a–GFP expressed from a PVX vector. Heavy gold labeling occurs in the parietal layer of sieve elements and across an adjoining sieve plate. Bar = 500 nm.
(C) Wild-type CMV infection. 3a MP antibody labeling is intense around a viral aggregate (arrowheads) in the sieve element. Bar = 500 nm.
(D) Wild-type CMV infection. 3a MP antibody labeling (arrowheads) occurs at the sieve element side of plasmodesmata that connect sieve elements and companion cells. Bar = 500 nm.
c, companion cell; s, sieve element; sp, sieve plate; v, virus particles.
|
|
Labeling of Branched Plasmodesmata between the Sieve Element and Companion Cell
In common with the minor veins of other species, the sieve elements of Nicotiana clevelandii are connected to their companion cells by characteristic plasmodesmata that have a single pore on the sieve element wall and a branched architecture in the adjoining companion cell wall (see Van Bel and Kempers 1997 ). In comparison with plasmodesmata in infected mesophyll tissues, the sieve element–companion cell plasmodesmata were seldom labeled within the plasmodesmal cavity. Instead, 3a MP localization was most intense around the pore exterior leading into the sieve element, giving immunogold labeling distribution the appearance of a wide collar around the neck region of the pore (Figure 3A and Figure 3D). Labeling was more intense outside of the callose collars surrounding the pore, suggesting that the membrane(s) adjacent to the pore was labeled rather than the pore itself. In all cases, labeling of the plasmodesmata was heavier on the sieve element side of the pore than on the companion cell side (Figure 3D).
Fluorescence Localization of the 3a–GFP in Sieve Elements
Examination of infected tissues expressing 3a–GFP showed that the fusion protein, like the wild-type 3a MP, was strongly targeted to sieve elements. Under low magnification, highly fluorescent cells containing GFP were seen to run alongside the minor vein xylem elements (Figure 2D and Figure 2E). The intense GFP signal from the phloem "swamped" the GFP signal emanating from labeled mesophyll plasmodesmata. Double staining of the fluorescent GFP-containing cells with aniline blue revealed the callose associated with sieve plates and lateral sieve areas and clearly identified the GFP-labeled cells as sieve elements (Figure 2F and Figure 2G). In some cases, accumulations of 3a–GFP were apparent close to but not on the sieve plates (Figure 2F).
At higher magnifications, individually labeled sieve elements were seen to contain discrete clusters of intense 3a–GFP fluorescence and also an elaborate network of 3a–GFP—labeled structures surrounding the sieve element parietal layer (Figure 2H and Figure 2J). The clusters of 3a–GFP fluorescence were interconnected by threadlike structures (Figure 2H). Increasing the laser gain on the confocal laser scanning microscope caused the fluorescence arising from the punctate clusters to saturate but revealed an extensive interconnecting network of fluorescent threads not visible at lower gain settings (Figure 2I and Figure 2J). As in the case of immunogold labeling of the wild-type 3a MP (Figure 3), the fluorescence labeling of sieve elements with 3a–GFP was intense in comparison with other cell types (compare the sieve element in Figure 2E with surrounding mesophyll cells). To confirm that the fluorescent structures seen in the sieve element were derived from the 3a–GFP fusion and not from free GFP that may have entered the sieve element during the infection process, sections containing the 3a–GFP fusion were probed with an antibody raised against the 3a MP and examined in an electron microscope. As in the case of the wild-type 3a MP, the 3a–GFP fusion was shown to be strongly localized to the sieve element parietal layer, which lies close to the plasma membrane (Figure 3B), producing a localization that was indistinguishable from that of the wild-type 3a MP. Thus, we consider the fluorescent structures seen in the confocal laser scanning microscope to be an accurate reflection of the subcellular localization of the CMV 3a MP.
CMV Particles Are Not Found in Plasmodesmata
Despite an extensive search of >2000 plasmodesmata that were either labeled or unlabeled with the 3a MP, intact CMV particles were not found within plasmodesmata in infected leaf tissues. This was confirmed by labeling sections parallel to those used for 3a MP localization with antibodies raised against the intact virion. After immunolocalization, a strong signal was obtained from the cytoplasm of all infected cells due to the presence of dispersed virions. In Figure 4A, gold labeling is intense in CMV-infected companion cells but absent from the branched plasmodesmata connecting sieve elements with their companion cells. As in the case of control sections probed with the 3a MP antibody, uninfected tissue sections labeled with the antibody raised against the CMV virion showed no immunogold labeling (data not shown).
View larger version (115K):
[in this window]
[in a new window]
|
Figure 4.
Immunogold Labeling of Sieve Elements with a CMV Virion-Specific Antibody.
Figure 4. (continued).
(A) The cytoplasm of infected companion cells shows heavy gold labeling, but gold particles are absent from the plasmodesmata (arrows) that connect sieve elements and companion cells. Bar = 350 nm.
(B) Oblique section of a mature sieve element shows immunogold labeling of a viral aggregate in the parietal layer (arrow) and heavy labeling of the cytoplasm of adjoining companion cells. Bar = 500 nm.
(C) Enlargement of the arrowed region shown in (B), with gold particles overlying aggregated CMV virions. Bar = 150 nm.
(D) Virus aggregate in the sieve element parietal layer shows heavy immunogold labeling. Bar = 500 nm.
c, companion cell; s, sieve element; sp, sieve plate; v, virus particles.
|
|
Intact Virions Occur in Sieve Elements
In several mature sieve elements displaying 3a MP localization, assembled CMV–virus aggregates were detected in the parietal layer (Figure 3A, Figure 3C, and Figure 3D and Figure 4B to D). In the companion cell, the virus usually appeared as dispersed virions; however, within the sieve element, the virus formed paracrystalline aggregates attached to the parietal layer. The MP antibody was often localized close to these aggregates but not directly on them (Figure 3C). Several of the virus aggregates were surrounded by a discrete membrane that appeared to anchor them in place (Figure 3A). The viral aggregates labeled strongly with the antibody raised against CMV particles (Figure 4B to D). Serial sections of individual minor vein sieve elements showed that branched plasmodesmata between the sieve elements and the companion cells (cf. Figure 4A) failed to label with antibody to virions, despite the presence of labeled virions within the sieve element.
|
|
DISCUSSION |
|---|
To date, the mechanism by which viruses enter the sieve elements of minor veins has remained elusive (see Carrington et al. 1996 ). TMV is possibly the most extensively studied virus with respect to movement functions, although TMV, like many other viruses, requires its CP for successful long-distance transport (see Gilbertson and Lucas 1996 ). The role of the TMV MP in viral movement within minor veins is at present unclear. Genetic experiments have suggested that the TMV MP may provide both short-distance and long-distance movement functions (reviewed in Carrington et al. 1996 ). However, in a study of transgenic tobacco plants expressing the TMV MP, Ding et al. 1992 found no labeling of vascular plasmodesmata with the MP antibody. The branched plasmodesmata at the bundle sheath–phloem parenchyma interface were immunolabeled with the MP but did not support the movement of microinjected fluorescent dextrans. It was suggested that the TMV CP may be essential at this boundary to facilitate viral entry into the minor vein phloem (Ding et al. 1992 ). However, recent studies with TMV CP–deficient mutants have shown that these viruses can reach at least the phloem parenchyma cells of minor veins, indicating that the CP is necessary only after the virus has traversed the bundle sheath–phloem parenchyma plasmodesmata (Ding et al. 1996 ; reviewed in Nelson and Van Bel 1997 ).
The Pathway of CMV to Minor Vein Sieve Elements
In the case of CMV, we have shown clearly that whether localized during a wild-type infection or expressed as an MP–GFP fusion from a PVX vector, the 3a MP targets plasmodesmata of mesophyll cells and also those of vascular elements within the bundle sheath. The trafficking of the 3a MP directly into minor vein sieve elements suggests that this MP is involved in the very final stages of virus entry into the phloem and as such plays an essential role in the passage of viral RNA (vRNA) into the sieve element. Interestingly, at the sieve element–companion cell plasmodesmata, the 3a MP was not localized within the central cavity, as occurred in the mesophyll, but on the sieve element interior around the neck of the pore. In addition, the 3a MP was distributed along the sieve element parietal layer and across lateral sieve areas. Confocal microscopy showed that the 3a–GFP fusion was present as an elaborate array of fluorescent "clusters" interconnected by threadlike projections. The origin of the projections is as yet unknown. Immunogold labeling did not reveal any distinct sieve element structures to which the MP was bound. Thus, the fluorescent network observed in the confocal microscope may have been composed completely of the 3a–GFP fusion protein. However, it is likely that the intense fluorescent clusters correspond to the heavily labeled regions seen around sieve element–companion cell plasmodesmata after immunogold labeling. It is important to note that in leaf tissues of N. clevelandii infected with PVX that expresses GFP as a "free" cytosolic protein, such filamentous structures have not been observed in minor vein sieve elements (A.G. Roberts, S. Santa Cruz, and K.J. Oparka, unpublished data). Thus, their presence appears to be a unique feature of the CMV 3a MP.
The intense labeling of minor vein sieve elements with the 3a MP suggests that this MP not only enters sieve elements but accumulates there in large amounts. The reasons for this are as yet obscure. It is possible that the mature sieve element does not have the necessary machinery to degrade the protein, leading to its progressive accumulation during infection. In expanding TMV infection sites, the MP undergoes a cycle of synthesis and degradation (Padgett et al. 1996 ). This is shown clearly in infection sites that express an MP–GFP fusion (Heinlein et al. 1995 ; Padgett et al. 1996 ), giving the infection sites a distinct halo-shaped appearance. Within this fluorescent halo of cells, plasmodesmata become targeted with the MP–GFP fusion and are "gated" to allow the passage of 10-kD dextran (Oparka et al. 1997 ). However, to the interior of the fluorescent halo, plasmodesmal gating diminishes significantly, and the cells are less fluorescent due to the progressive degradation of viral MP (Padgett et al. 1996 ; Oparka et al. 1997 ). If we assume that a similar sequence of events occurs during CMV infection, then it is possible that both mesophyll cells and vascular parenchyma elements may have the capacity to degrade the viral MP after the passage of the virus. Within the sieve element, protein degradation is not possible, leading to the progressive accumulation of 3a MP during viral infection.
A second possibility is that the mature sieve element does not have the required cytoskeletal elements to traffic the 3a MP any farther, causing it to accumulate close to its points of entry into the sieve element. Recently, it has been shown that components of the cytoskeleton, particularly microtubules, may be involved in trafficking MP–RNA complexes toward plasmodesmata (Heinlein et al. 1995 ; McLean et al. 1995 ; reviewed in Lucas and Gilbertson 1994 ). The lack of cytoskeletal elements in mature sieve elements (see Van Bel and Kempers 1997 ) may inhibit any further potential transcellular movement of the viral MP. A third possibility is that the fluorescent structures are not by-products of the infection process but instead represent functional structures that provide a "scaffold" for the assembly or transport of virus within the sieve element. It will be interesting to determine whether such structures form also at sites of virus unloading from the phloem.
Absence of Virions in Plasmodesmata
It has been demonstrated that CMV requires its CP for both local (Suzuki et al. 1991 ; Boccard and Baulcombe 1993 ) and long-distance (Suzuki et al. 1991 ; Taliansky and Garcia-Arenal 1995 ) movement. Despite an extensive search of plasmodesmata, we were unable to detect intact virions within plasmodesmata, including the branched plasmodesmata that connect the sieve elements with their companion cells. Thus, our observations concur with previous suggestions that CMV does not move from cell to cell in an encapsidated form but as a ribonucleoprotein complex that is trafficked through plasmodesmata by the 3a MP (Ding et al. 1995 ; Nguyen et al. 1997 ). The complete absence of intact virions from the sieve element–companion cell plasmodesmata (as evidenced by the lack of virion-specific antibody labeling) is significant because it suggests that CMV also moves through this final boundary as a ribonucleoprotein complex rather than as a particle. Although the CP of CMV is required for local movement (Canto et al. 1997 ), it clearly does not have to interact with the 3a MP to facilitate the entry of the latter into the sieve element. This was demonstrated when the 3a–GFP expressed from a PVX vector was able to traffic into the sieve element in a manner identical to that of the wild-type 3a MP, that is, in the absence of the CMV CP. In addition, entry of the PVX-delivered 3a–GFP into sieve elements demonstrates that the 3a MP does not have to interact with CMV RNA to traffic from the companion cell to sieve element. These observations confirm previous reports in which microinjection (Ding et al. 1995 ) or biolistic (Itaya et al. 1997 ) techniques were used and indicate that the 3a MP alone has the capacity to move from cell to cell.
Does CMV Assemble on the Sieve Element Reticulum?
Previous studies of CMV movement have shown vRNA to be present in minor vein sieve elements during wild-type CMV infection (Wintermantel et al. 1997 ), although the form in which the virus was present (virion or nucleoprotein) within sieve elements was not determined. Our observations suggest that CMV encapsidation occurs in the parietal layer of mature sieve elements. Although virions were absent from plasmodesmata, they were found in large aggregates in the sieve element parietal layer, frequently bounded by a discrete membrane. Earlier studies of cucumovirus infections of parenchyma cells have depicted a similar membranous structure around virus aggregates (Martelli and Russo 1985 ). This bounding membrane is unlikely to have been the sieve element plasma membrane because this would necessitate an extracellular location of the virus, which is inconsistent with the known symplastic movement of plant viruses (Lucas and Gilbertson 1994 ).
A more likely candidate for the viral envelope is the sieve element reticulum (SER), which is an organelle that lines the parietal layer of mature sieve elements (Esau 1978 ; Sjolund and Shih 1983 ). Growth of virus aggregates on the SER may provide an anchorage site for both the CP and vRNA and prevent the "naked" vRNA and CP subunits from entering the translocation stream (see also Gilbertson and Lucas 1996 ). Release of virions from the assembly sites may occur through the eventual rupture of the SER membrane. Accumulation of virus in sieve elements may also be an adaptation for virus transmission because CMV is transmitted by aphid species that feed on the phloem (Palukaitis et al. 1992 ).
The model shown in Figure 5 is an attempt to incorporate our present observations of CMV movement with other known parameters influencing the movement of this virus. In this scheme, the CMV vRNA is trafficked through plasmodesmata by the 3a MP, probably as a nucleoprotein complex (see also Ding et al. 1995 ), although it should be stressed that some of the 3a MP may be capable of entering the sieve element without being complexed to vRNA. Virus assembly may occur within individual cells after the entry of the vRNA; however, based on microinjection studies (Ding et al. 1995 ; Nguyen et al. 1997 ), it is possible that the vRNA–MP complex may traverse several cells without the need for an intervening encapsidation stage.
View larger version (64K):
[in this window]
[in a new window]
|
Figure 5.
Schematic Model for the Movement of CMV into Minor Vein Sieve Elements.
Assembled virus particles occur within the companion cell cytoplasm (1), and these are detectable by virion-specific antibody labeling. Before movement into the sieve element, the virus disassembles in the cytoplasm of the companion cell (2). A linear ribonucleoprotein complex, which is composed of the vRNA (black strands), 3a MP (black circles), and CP subunits (gray triangles), then forms (3) before movement through the plasmodesmal pore that connects the companion cell with the sieve element (4). In the sieve element, CP and vRNA are incorporated into a membrane-bound viral assembly complex (VAC) in the parietal layer (5), while the MP from the dissociated transport complex remains outside of the VAC and lines the sieve element plasma membrane. After rupture of the VAC membrane, assembled virions are released into the translocation stream (6). ER, endoplasmic reticulum.
|
|
Because the CMV 3a MP shares sequence similarity with the TMV MP, we hypothesize that the 3a–vRNA complex utilizes the cytoskeleton as a delivery system to the plasmodesmal pore. During transit of the 3a–vRNA complex, some of the viral MP is retained within the plasmodesmal pore. The final delivery of the 3a–vRNA complex to the sieve element is via the branched plasmodesmata that connect the sieve element and companion cell. As in other models of vRNA transport (e.g., Lucas et al. 1993 ; Gilbertson and Lucas 1996 ), we have assumed an interaction between the 3a–vRNA and the desmotubule that is continuous through the sieve element–companion cell plasmodesmata (Van Bel and Kempers 1997 ). During this final transfer phase, the vRNA is trafficked to viral assembly complexes located on the SER in the parietal layer. After this ultimate stage of local movement, the 3a MP accumulates along the sieve element parietal layer. Currently, it is not known whether all of the 3a MP that enters the sieve element remains bound in this way. Conceivably, some of the MP enters the translocation stream and may be capable of exerting its effects in cells distant to its point of entry into the phloem.
Role of the Viral CP
How does the viral CP enter the minor vein sieve elements before viral assembly? One possibility is that CP subunits are trafficked as part of the same nucleoprotein complex that contains the vRNA and 3a MP (Figure 5). Once in the sieve element, detachment of the 3a MP from the movement complex may permit encapsidation of the vRNA to occur. In the model described above, the requirement of viral CP for systemic movement lies in its role in encapsidating the vRNA before long-distance movement rather than modifying plasmodesmata. A recent study on the cell-to-cell movement of the leaf sucrose transporter (SUT1) has suggested that synthesis of this protein may occur in the sieve element after the trafficking of SUT1 mRNA through sieve element–companion cell plasmodesmata (Kuhn et al. 1997 ). In our study, we found no ribosomes associated with the SER, confirming the conventional view that the SER is not functional in protein synthesis (Esau 1978 ; Sjolund and Shih 1983 ; Fisher et al. 1992 ). Our model is based on the assumption that all of the viral CP that entered the sieve element did so through plasmodesmata and not as a result of synthesis within the sieve element. We have also assumed that viral encapsidation occurred from CP subunits trafficked in tandem with the vRNA–MP complex. However, it is possible that the CP may enter the sieve element separately by an as yet undefined route.
CMV is an icosahedral cucumovirus that, unlike the comoviruses, does not move as an intact virion through plasmodesmata; neither is it dependent on the formation of tubular structures that span the plasmodesmal pore (reviewed in Carrington et al. 1996 ). Thus, icosahedral plant viruses may have evolved quite different strategies for moving through plasmodesmata. It will be interesting to determine whether the MPs from a range of viruses traffic into minor vein sieve elements and the extent to which MPs are translocated over long distances in the phloem. To date, studies have concentrated on epidermal or mesophyll tissues and have highlighted the role of MPs in plasmodesmal targeting and gating. For viruses that employ different local movement strategies, a clear challenge for the future will be to ascertain whether the same mechanisms (and genes) that control virus movement through the mesophyll tissues also determine entry into and movement within the minor vein phloem domain.
|
|
METHODS |
|---|
Construction of 3a–Green Fluorescent Protein in a Potato Virus X Vector
The 3a–green fluorescent protein (GFP) fusion was made by using overlap extension polymerase chain reaction (Higuchi et al. 1988 ). The synthetic gene was cloned into the potato virus X (PVX) vector pTXS.P3C2, as described by Boevink et al. 1996 , using standard molecular biology techniques (Sambrook et al. 1989 ).
Inoculation of Plants with 3a–GFP and Wild-Type Cucumber Mosaic Virus
Leaves from Nicotiana clevelandii seedlings were inoculated with 5 µL of infectious transcript of pTXS.3a–GFP made using a T7 in vitro transcription kit (Ambion Inc., Austin, TX), as outlined in Boevink et al. 1996 . The ability of 3a–GFP, when expressed from PVX, to target mesophyll plasmodesmata has been described previously (Oparka et al. 1996b ). To locate the 3a movement protein (MP) during a wild-type cucumber mosaic virus (CMV) infection, leaves were mechanically inoculated with 20 µL of wild-type CMV (Fny-CMV) (Palukaitis et al. 1992 ) at 100 µg mL-1 in PBS.
Immunoelectron Microscopy
The pathway of virus movement into minor vein (classes IV and V) sieve elements of leaves inoculated with wild-type CMV and 3a–GFP was determined by immunogold localization of the 3a MP (Kaplan et al. 1995 ). Leaves were harvested 10 to 14 days after inoculation with wild-type CMV or 3a–GFP expressed from the PVX vector. Small pieces of leaf tissue (2 to 3 mm2) were fixed for 2 hr in 2% paraformaldehyde and 1 or 2% glutaraldehyde in PME buffer (50 mM Pipes, 5 mM EGTA, and 2 mM MgSO4). Fixative infiltration of the tissue was aided by the application of a slight vacuum. Explants were rinsed twice for 10 min with PME buffer, dehydrated in an ethanol series, and then embedded in London Resin White (medium grade; London Resin Co., Reading, UK). Tissue from uninfected control plants of the same age was treated in the same way. Ultrathin serial sections from each type of tissue were collected on pyroxylin-coated nickel grids and blocked in 1% (w/v) BSA plus 1% (w/v) cold water–fish skin gelatin in PBS.
The 3a MP antibodies were cross-absorbed with acetone-extracted powder of healthy N. clevelandii extract before use (da Rocha et al. 1986 ). Sections were incubated on drops of 3a MP rabbit antisera diluted 1:150 in blocking buffer for 2 to 4 hr at room temperature. Grids were rinsed three times for 5 min in PBS and three times for 5 min in 20 mM Tris-Cl, pH 7.2. Grids were then incubated in goat anti–rabbit IgG conjugated to 15 nm of gold (AuroProbe, Amersham, Buckinghamshire, UK), which was diluted 1:50 in blocking buffer, at room temperature for 2 hr. Grids were rinsed twice for 5 min in PBS, twice for 5 min in 20 mM Tris-Cl, pH 7.2, and twice for 5 min in distilled water. Some sections from wild-type CMV-infected tissue were labeled with cross-absorbed antibodies raised against whole CMV virions (Ding et al. 1995 ). Sections were stained with uranyl acetate and lead citrate and examined at 80 kV in an electron microscope (model 1200 CX; JEOL UK Ltd., Watchmead, Welwyn Garden City, UK).
The pathway of the 3a MP from mesophyll cells to the sieve element in minor veins was analyzed by examining the number of plasmodesmata labeled and the density of gold particles in wild-type CMV compared with PVX-infected tissues. For this study, 21 sections from four control vascular bundles were compared with 28 sections of five vascular bundles from wild-type CMV-infected tissue and 35 sections of five bundles from leaves infected with the PVX vector expressing 3a–GFP. The statistical significance of the number of plasmodesmata labeled and also the density of label in different cell types between controls and CMV-infected and 3a–GFP—infected tissue were determined using the Mann-Whitney test of ranked sums (Freund 1988 ).
Confocal Laser Scanning Microscopy
The distribution of 3a–GFP in live tissue was examined with a Bio-Rad MRC 1000 confocal laser scanning microscope, using methods described previously (Oparka et al. 1997 ; Roberts et al. 1997 ).
Detection of Callose
In fresh sections, callose associated with the minor vein sieve elements was detected using decolorized 0.05% (w/v) aniline blue. To detect the callose associated with mesophyll plasmodesmata, the immunolocalization procedure described in detail by Oparka et al. 1997 was employed.
|
|
FOOTNOTES |
|---|
1 Current address: School of Biological Sciences, University of Sydney, Sydney, Australia. 
|
|
ACKNOWLEDGMENTS |
|---|
We are grateful to Denton Prior for expert technical assistance and to Alison Roberts for the preparation of Figure 5. K.J.O., S.S.C., and P.P. gratefully acknowledge the financial support of the Scottish Office Agriculture, Environment and Fisheries Department. L.M.B. was the recipient of a Mylnefield Research Services Fellowship.
Received December 1, 1997; accepted February 5, 1998.
|
|
REFERENCES |
|---|
Angell, S.M., Davies, C., and Baulcombe, D.C. (1996) Cell-to-cell movement of potato virus X is associated with a change in the size-exclusion limit of plasmodesmata in trichome cells of Nicotiana clevelandii.. Virology 216:197-201[CrossRef][Medline].
Boccard, F., and Baulcombe, D.C. (1993) Mutational analysis of cis-acting sequences and gene function in RNA 3 of cucumber mosaic virus. Virology 193:563-578[CrossRef][Medline].
Boevink, P., Santa Cruz, S., Hawes, C., Harris, N., and Oparka, K.J. (1996) Virus-mediated delivery of the green fluorescent protein to the endoplasmic reticulum of plant cells. Plant J. 10:935-941[CrossRef].
Canto, T., Prior, D.A.M., Hellwald, K.-H., Oparka, K.J., and Palukaitis, P. (1997) Characterization of cucumber mosaic virus. IV. Movement protein and coat protein are both essential for cell-to-cell movement of CMV. Virology 237:237-248[CrossRef][Medline].
Carrington, J.C., Kasschau, K.D., Mahajan, S.K., and Schaad, M.C. (1996) Cell-to-cell and long-distance transport of viruses in plants. Plant Cell 8:1669-1681[CrossRef][ISI][Medline].
Citovsky, V., Wong, M.L., Shaw, A.L., Prasad, B.V.V., and Zambryski, P. (1992) Visualization and characterization of tobacco mosaic virus movement protein binding to single-stranded nucleic acids. Plant Cell 4:397-411[Abstract/Free Full Text].
da Rocha, A., Ohki, S.T., and Hiruki, C. (1986) Detection of mycoplasmalike organisms in situ by indirect immunofluorescence microscopy. Phytopathology 76:864-868.
Derrick, P.M., Barker, H., and Oparka, K.J. (1990) Effects of virus infection on symplastic transport of fluorescent tracers in Nicotiana clevelandii leaf epidermis. Planta 181:555-559.
Ding, B., Haudenshield, J.S., Hull, R.J., Wolf, S., Beachy, R.N., and Lucas, W.J. (1992) Secondary plasmodesmata are specific sites of localization of the tobacco mosaic virus movement protein in transgenic tobacco plants. Plant Cell 4:915-928[Abstract/Free Full Text].
Ding, B., Li, Q., Nguyen, L., Palukaitis, P., and Lucas, W.J. (1995) Cucumber mosaic virus 3a protein potentiates cell-to-cell trafficking of CMV RNA in tobacco plants. Virology 207:345-353[CrossRef][ISI][Medline].
Ding, S.W., Anderson, B.J., Haase, H.H., and Symons, R.H. (1994) New overlapping gene encoded by the cucumber mosaic virus genome. Virology 80:204-213.
Ding, X.-S., Shintaku, M.H., Arnold, S.A., and Nelson, R.S. (1996) Invasion of minor veins of tobacco leaves inoculated with tobacco mosaic virus mutants defective in phloem-dependent movement. Proc. Natl. Acad. Sci. USA 93:11155-11160[Abstract/Free Full Text].
Esau, K. (1978) Developmental features of the primary phloem in Phaseolus vulgaris L. Ann. Bot. 42:1-13[Abstract/Free Full Text].
Fisher, D.B., Wu, Y., and Ku, M.S.B. (1992) Turnover of soluble proteins in the wheat sieve tube. Plant Physiol. 100:1433-1441[Abstract/Free Full Text].
Freund, J.E. (1988). Modern Elementary Statistics, 7th ed. (Englewood Cliffs, NJ: Prentice-Hall).
Fujiwara, T., Giesman-Cookmeyer, D., Ding, B., Lommel, S.A., and Lucas, W.J. (1993) Cell-to-cell trafficking of macromolecules through plasmodesmata potentiated by the red clover necrotic mosaic virus movement protein. Plant Cell 5:1783-1794[Abstract].
Gilbertson, R.L., and Lucas, W.J. (1996) How do viruses traffic on the `vascular highway'? Trends Plant Sci. 1:260-268[CrossRef][ISI].
Heinlein, M., Epel, B., Padgett, H.S., and Beachy, R. (1995) Interaction of tobamovirus movement proteins with the cytoskeleton. Science 270:1983-1985[Abstract/Free Full Text].
Higuchi, R., Krummel, B., and Saiki, R.K. (1988) A general method of in vitro preparation and specific mutagenesis of DNA fragments: Study of protein and DNA interactions. Nucleic Acids Res. 16:7351-7367[Abstract/Free Full Text].
Itaya, A., Hickman, H., Bao, Y., Nelson, R., and Ding, B. (1997) Cell-to-cell trafficking of cucumber mosaic virus movement protein: Green fluorescent protein fusion produced by biolistic gene bombardment in tobacco. Plant J. 12:1223-1230[CrossRef].
Kaplan, I.B., Shintaku, M.H., Zhang, L., Marsh, L.E., and Palukaitis, P. (1995) Complementation of virus movement in transgenic tobacco expressing cucumber mosaic virus 3A gene. Virology 209:188-199[CrossRef][Medline].
Kuhn, C., Franceschi, V., Schulz, A., Lemoine, R., and Frommer, W.B. (1997) Macromolecular trafficking indicated by localization and turnover of sucrose transporters in enucleate sieve elements. Science 275:1298-1300[Abstract/Free Full Text].
Leisner, S., and Turgeon, R. (1993) Movement of virus and photoassimilate in the phloem: A comparative analysis. Bioessays 15:741-748[CrossRef][ISI][Medline].
Leisner, S.M., Turgeon, R., and Howell, S.H. (1992) Long-distance movement of cauliflower mosaic virus in infected turnip plants. Mol. Plant-Microbe Interact. 5:41-47.
Leisner, S.M., Turgeon, R., and Howell, S.H. (1993) Effects of host plant development and genetic determinants on the long-distance movement of cauliflower mosaic virus in Arabidopsis. Plant Cell 5:191-202[Abstract].
Lucas, W.J., and Gilbertson, R.L. (1994) Plasmodesmata in relation to viral movement within leaf tissues. Annu. Rev. Phytopathol. 32:387-411[CrossRef][ISI].
Lucas, W.J., Ding, B., and Van der Schoot, C. (1993) Plasmodesmata and the supracellular nature of plants. New Phytol. 125:435-476[CrossRef][ISI].
Martelli, G.P., and Russo, M. (1985). Virus–host relationships: Symptomatological and ultrastructural aspects. In The Plant Viruses, Vol. 1, Polyhedral Virions with Tripartite Genomes, R.I.B. Francki, ed (New York: Plenum Press), pp. 163–201.
Maule, A.J. (1991) Virus movement in infected plants. Crit. Rev. Plant Sci. 9:457-473.
McLean, B.G., Zupan, J., and Zambryski, P.C. (1995) Tobacco mosaic virus movement protein associates with the cytoskeleton in tobacco cells. Plant Cell 7:2101-2114[Abstract].
Melcher, U. (1990) Similarities between putative transport proteins of plant viruses. J. Gen. Virol. 71:1009-1018[Abstract/Free Full Text].
Nelson, R., and Van Bel, A.J.E. (1997) The mystery of virus trafficking into, through and out of the vascular tissue. Prog. Bot. 59:476-533.
Nguyen, L., Lucas, W.J., Ding, B., and Zaitlin, M. (1997) Viral RNA trafficking is inhibited in replicase-mediated resistant transgenic tobacco plants. Proc. Natl. Acad. Sci. USA 93:12643-12647[Abstract/Free Full Text].
Oparka, K.J., Roberts, A.G., Roberts, I.M., Prior, D.A.M., and Santa Cruz, S. (1996a) Viral coat protein is targeted to, but does not gate, plasmodesmata during cell-to-cell movement of potato virus X. Plant J. 10:805-813[CrossRef].
Oparka, K.J., Boevink, P., and Santa Cruz, S. (1996b) Studying the movement of plant viruses using green fluorescent protein. Trends Plant Sci. 1:412-418[CrossRef].
Oparka, K.J., Prior, D.A.M., Santa Cruz, S., Padgett, H.S., and Beachy, R.N. (1997) Gating of epidermal plasmodesmata is restricted to the leading edge of expanding infection sites of tobacco mosaic virus (TMV). Plant J. 12:781-789[CrossRef][ISI][Medline].
Padgett, H.S., Epel, B., Kahn, T.W., Heinlein, M., Watanabe, Y., and Beachy, R.N. (1996) Distribution of tobamovirus movement protein in infected cells and implications for cell-to-cell spread of infection. Plant J. 10:1079-1088[CrossRef][ISI][Medline].
Palukaitis, P., Roossinck, M.J., Dietzgen, R.G., and Francki, R.I.B. (1992) Cucumber mosaic virus. Adv. Virus Res. 41:281-348[ISI][Medline].
Roberts, A.G., Santa Cruz, S., Roberts, I.M., Prior, D.A.M., Turgeon, R., and Oparka, K.J. (1997) Phloem unloading in sink leaves of Nicotiana benthamiana: Comparison of a fluorescent solute with a fluorescent virus. Plant Cell 9:1381-1396[Abstract].
Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, 2nd ed. (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press).
Sjolund, R.D., and Shih, C.Y. (1983) Freeze-fracture analysis of phloem structure in plant tissue cultures. I. The sieve element reticulum. J. Ultrastruct. Res. 82:111-121[CrossRef][ISI][Medline].
Suzuki, M., Kuwata, S., Katoaka, J., Masuta, C., Nitta, N., and Takanami, Y. (1991) Functional analysis of deletion mutants of cucumber mosaic virus RNA 3 using an in vitro transcription system. Virology 183:106-113[CrossRef][ISI][Medline].
Taliansky, M.E., and Garcia-Arenal, F. (1995) Role of cucumovirus capsid protein in long-distance movement within the infected plant. J. Virol. 69:918-922.
Tomenius, K., Clapham, D., and Meshi, T. (1987) Localization by immunogold cytochemistry of the virus-coded 30K protein in plasmodesmata of leaves infected with tobacco mosaic virus. Virology 160:363-371[CrossRef][ISI].
Van Bel, A.J.E., and Kempers, R. (1997) The pore/plasmodesma unit; key element in the interplay between sieve element and companion cell. Prog. Bot. 58:278-291.
Vaquero, C., Turner, A.P., Demangeat, G., Sanz, A., Serra, M.T., Roberts, K., and Garcia-Luque, I. (1994) The 3a movement protein from cucumber mosaic virus increases the gating capacity of plasmodesmata in transgenic tobacco plants. J. Gen. Virol. 75:3193-3197[Abstract/Free Full Text].
Waigmann, E., and Zambryski, P. (1995) Tobacco mosaic virus movement protein–mediated protein transport between trichome cells. Plant Cell 7:2069-2079[Abstract].
Wang, D., and Maule, A.J. (1994) A model for seed transmission of a plant virus: Genetic and structural analyses of pea embryo invasion by pea seed–borne mosaic virus. Plant Cell 6:777-787[Abstract].
Wintermantel, W.M., Banerjee, N., Oliver, J.C., Paolillo, D., and Zaitlin, M. (1997) Cucumber mosaic virus is restricted from entering minor veins of transgenic tobacco exhibiting replicase-mediated resistance. Virology 231:248-257[CrossRef][Medline].
Wolf, S., Deom, C.M., and Beachy, R.N. (1989) Movement protein of tobacco mosaic virus modifies plasmodesmata size exlusion limit. Science 246:377-379[Abstract/Free Full Text].
ABSTRACT
INTRODUCTION
