Phloem transport of viruses

Long-distance transport of macromolecules was first seen by observing the spread of viruses that misuse the phloem for their passive translocation through the plant, leading to the spread of the infection, especially into growing areas. After infection, viruses normally spread locally from cell to cell through PD until they reach the phloem. Viruses contain special proteins, namely movement proteins (MPs), that can bind and unfold single-stranded RNAs and facilitate their intercellular translocation by building protein-RNA transport complexes [ribonucleoprotein (RNP) complexes] [3]. It is currently unknown whether MPs are also needed for phloem-dependent movement [4]. However, several studies indicate that cell-to-cell and systemic virus transport follow different mechanisms [5-7], and there seem to be different viral and endogenous plant components involved [7-10]. In addition to MP, an important viral factor for phloem movement is the coat protein that seems to be strictly required for long-distance translocation of several groups of viruses [11,12].

With regard to endogenous plant factors, a cell wall glycine-rich protein and a pectin methylesterase have been shown to influence viral spread, probably by modulating the permeability of PD [5,7,13]. An interesting recent finding is that the interaction of the groundnut rosette virus ORF3 protein with subnuclear Cajal bodies and the nucleolus is required to capture the plant protein fibrillarin to form an RNP complex capable of long-distance phloem movement [14,15]. Also, endogenous phloem proteins, namely different phloem lectins and the phloem protein CmPP16 (a protein with properties resembling those of virus MPs), have been found to interact with viral RNAs and endogenous mRNAs and thus have been suggested to be involved in virus import and/or translocation as RNP complexes [16,17], but a conclusive demonstration of their importance for viral spread is, as yet, missing.

Phloem transport of endogenous macromolecules

The occurrence of unspecified RNA species was described nearly 40 years ago, but it was long regarded as an artefact caused by sampling. The first evidence that endogenous plant mRNAs are present in SEs came from the in situ localisation of the mRNA of a sucrose transporter in SEs and the PD connecting SEs to CCs [18]. Meanwhile, a range of different mRNAs from phloem samples of various species, obtained using different sampling techniques, have been detected (reviewed in [19]) and they are nowadays regarded as being authentic components of the phloem stream. At the same time, heterografting studies performed by different research groups could show that specific transcripts can indeed move and, moreover, induce phenotypic alterations in their target tissues [20-22]. For example, in potato, overexpression of the transcription factor BEL1 mRNA induced a marked increase of tubers per plant, and this phenomenon could cross graft junctions, thus indicating phloem mobility of this mRNA [23].

Recently, small RNAs of fewer than 30 nucleotides [short interfering RNAs (siRNAs) and microRNAs (miRNAs)] were found to occur in the phloem stream [24-26]. siRNAs are involved in a process called post-transcriptional gene silencing, an innate plant defence mechanism against transposable elements and viruses [27] that can spread systemically throughout the plant [28]. While siRNAs are generally assumed to be mobile between cells and also systemic [29], most miRNAs, involved mainly in controlling developmental processes, seem to act cell-autonomously under normal growth conditions in adult tissues, as demonstrated, for example, by miR171 [30,31]. However, it was recently shown that specific nutrient starvation-responsive miRNAs (such as miR395 that responds to sulphur, miR398 to copper, and miR399 to phosphate stress) can accumulate in the phloem at high levels when plants are grown under the respective nutrient-deprived condition [26]. Moreover, different results suggested that during phosphate deprivation systemic signalling is required [32,33], and recently miR399 was shown to be indeed translocated across graft junctions in the model species Arabidopsis, indicating that phloem movement of this specific miRNA could be involved in regulating the response toward phosphate starvation systemically [34,35]. This observation, however, does not answer the question of whether miRNAs are generally mobile or immobile from cell to cell and over long distances. There might exist a small subset of specific mobile miRNAs, or miRNA mobility might be restricted to particular developmental stages and/or specific environmental conditions.

Also, the occurrence of proteins in phloem sap was noticed a long time ago and since then different studies that comprehensively identified phloem proteins from various species have been published [36-38]. However, the capability of the polypeptides to be transported in vivo has been demonstrated for only a few of them (for example, the pumpkin phloem protein CmPP16 that, as mentioned above, displays similarities to viral MPs) [39]. Moreover, the physiological functions of most of the phloem polypeptides remain unknown. For many of them, roles in signal transduction and stress responses have been suggested but not conclusively demonstrated. For several years, the Solanaceae-specific peptide systemin served as the paradigm for a phloem-mobile protein involved in long-distance signalling [40], but according to more recent results, it seems rather likely that it is jasmonic acid and not systemin that constitutes the phloem-translocated signal during wound responses [41]. To date, the only convincing example of a phloem protein involved in signalling is the flower-promoting protein flowering locus T (FT), which was recently shown to be present in the phloem at detectable concentrations when plants were sampled at the onset of flowering [38]. Subsequently, several independent studies have provided compelling evidence that FT long-distance transport can indeed contribute to flower induction in different plant species [42-46], and the FT protein could thus be a component of the long-sought ‘florigen’ [47]. Further studies are required to find out whether any of the other phloem sap proteins can act in the information transfer during other developmental processes or stress responses.

Although the transport mechanisms of endogenous macromolecules are not well understood, it is assumed that phloem import and transport of both proteins and RNAs are mediated by specific binding proteins acting as molecular chaperones. It was proposed that phloem heat shock proteins might participate in delivering polypeptides to the phloem [48,49], but other components with functions related to protein folding (for example, the abundant phloem cyclophilins [38]) could also be involved. Several phloem-sap proteins have also been suggested to transport RNAs in the form of RNP complexes since they are RNA-binding, are translocatable across graft unions, and can increase the SEL of PD [17]. In particular, the pumpkin phloem sap protein CmPP16 possesses properties similar to those of viral MPs and mediates the transport of its own and foreign mRNAs from cell to cell and over long distances through the phloem, as shown by microinjection studies and grafting experiments [39]. Small RNA transport in pumpkin seems to follow a similar route mediated by the small RNA-binding phloem protein CmPSRP1 that, however, seems to be restricted to this one plant species [24]. In other plants, one or more of the various RNA-binding phloem proteins that have been described to occur in phloem sap [16,36,38] could possibly do the job. Also, the obvious similarities between the translocation of viruses and endogenous macromolecules suggest that the underlying mechanisms are the same or at least highly similar, but both will need to be demonstrated in future experiments.