Plasmodesmal Targeting and Accumulation of TMV Movement Protein

Abstract

The interaction between tobacco mosaic virus and its host plant cells has been intensively studied as a model for macromolecular trafficking. The observation that GFP-labelled TMV movement protein localises to microtubules led to the suggestion that microtubules are required for the cell to cell movement of the virus. In a recent paper we have demonstrated that the targeting of TMV movement protein to plasmodesmata requires the actin and ER networks, which supports previous evidence from our laboratory that showed that disruption of microtubules did not prevent cell to cell movement of TMV virus, and that a mutated movement protein, which did not localise to micro-tubules, showed enhanced viral movement. In this addendum we speculate where the TMV movement protein accumulates within plasmodesmata, and the relationship of this accumulation to the cell to cell movement of the virus.

Key Words:

• actin
• endoplasmic reticulum
• microtubule
• plasmodesmata
• targeting
• tobacco mosaic virus

This article refers to:

In our latest paper,1 we have used the technique of fluoresence recovery after photo- bleaching to demonstrate that the targeting of TMV movement protein to plasmodesmata requires the actin and ER networks and not microtubules (see also ref. 2). In this work it is assumed that the mechanism by which bleached movement protein (MP) within plasmodesmata (PD) is supplemented by new fluorescent MP reflects the mechanism of MP movement to, and potentially through, PD. However, the accumulation of MP within PD may be independent from trafficking of the virus from cell to cell. This is suggested by the observation that MP linked to DsRed, in contrast to MP linked to certain other fluorescent proteins (e.g., eGFP, mRFP, PAGFP) does not target PD. It is assumed that most MP conjugates accumulate in the central cavity of the PD but the MP-DsRed tetramer is too large to penetrate into this area. Crucially, however, this lack of targeting does not prevent the cell to cell movement of the virus or appear to hinder the virus in any way, since lesions develop at similar rates when DsRed or monomeric fluorescent protein fusions are expressed.

MP also accumulates in PD where no cell to cell movement is occurring. During an infection, TMV virus is not able to enter the symplastically isolated stomatal complex. However, MP still accumulates in the nonfunctional “half” PD between guard cells and adjacent epidermal cells.

A number of groups investigating TMV-MP have observed the presence and motion of small bodies referred to as viral replication complexes (VRC).3 However, under the growth conditions employed in our experimental system, which were designed to mimic as closely as possible natural infection conditions, such motile structures have not been observed. TMV is adapted to the warm climates where tobacco production occurs and has optimal infectivity at 33°C. Our laboratory uses this temperature as standard, unlike some other groups that routinely maintain infections at lower temperatures. Instead, at the earliest stage of infection MP-GFP is first detected within the PD followed by labelling of immobile structures at the ER vertices. It has been reported that the VRC move as intact viral movement complexes (VMC) through PD and in some instances displace transgenically expressed MP from PD.3 Since we cannot see motile VRC, we cannot repeat these experiments. However, when infected with TMV.MP-mRFP, transgenic MP-GFP plants show PD labelled with both green and red proteins, both at the leading edge and at the centre of the infection. If MP is displaced from PD, surely we would expect that the PD at the centre of a lesion would be predominantly one colour; red if the viral MP replaces the plant MP, or green if the red viral MP is subsequently replaced by new MP produced by the plant.

Do the observation of motile VRC and VMC therefore relate to viral movement? Our observations using TMV.MP-DsRed and of the guard cell boundary suggest that the accumulation of MP within PD may be irrelevant to the functional virus infection. The critical requirement is targeting of MP to PD; not accumulation within PD. Indeed, the accumulation may even be a host response to sequester the MP. We propose the hypothesis that using fluorescently labelled MP it is not actually possible to observe the small amounts of protein that are functional in cell to cell movement. Due to the time required for fluorescent proteins to mature and accumulate to levels detectable by microscopy the virus may move between cells ahead of any visible indication. Therefore, whilst all MP might be transferred to PD by the same the mechanism, the actual observation of MP at the edge of an infection may be irrelevant. We therefore advise caution in the interpretation of results involving MP-fluorescent protein conjugates.

Acknowledgements

This work was supported by the Scottish Executive Environment and Rural Affairs Department (grants SCR/0611 and SCR/0612).