Long-distance trafficking of macromolecules in the phloem

Abstract

The fact that macromolecules such as proteins and mRNAs overcome the symplastic barriers between various tissue domains was first evidenced by the movement of plant viruses. We have recently demonstrated that viral infection disengages the symplastic restriction present between the sieve element-companion cell complex and neighboring cells in tobacco plants. As a result, green fluorescent protein, which was produced in mesophyll and bundle sheath cells, could traffic into the sieve tube and travel long distances within the vascular system. In this addendum we discuss the likely existence of a novel plant communication network in which macromolecules also act as long-distance trafficking signals. Plasmodesmata interconnecting sieve elements and companion cells as well as plasmodesmata connecting the sieve tube with neighboring cells may play a central role in establishing this communication network.

Translocation of photoassimilates from the source (site of synthesis) to various sink organs is governed, in part, by short-distance intercellular transfer of assimilates to the loading region of the phloem and long-distance transport within the plant vascular system. Sucrose, which is synthesized in the leaf mesophyll, moves cell-to-cell symplastically through plasmodesmata until it reaches the boundary of the sieve element (SE)-companion cell (CC) complex. In many plant species, the connection between phloem parenchyma (PP)/bundle sheath (BS) cells and CCs is characterized by a sparseness of plasmodesmata (e.g., Solanaceae), and sucrose is exported out of the cells to the apoplast. This type of plants (apoplastic loaders) uses sucrose proton symporters to load the sucrose into the vasculature.¹ Cucurbits are considered one of the model plants for symplastic phloem loading.² This type of plant is characterized by abundant plasmodesmata interconnecting the intermediary cells, which are specialized CCs, with the neighboring BS cells. It is generally accepted that in these plants, phloem loading includes intercellular movement of sucrose through the plasmodesmata, along the entire pathway from the mesophyll cell to the SE-CC complex.

Interestingly, the existence of plasmodesmata interconnecting the SE-CC complex and neighboring cells is evident in all plant species that are characterized by an apoplastic phloem-loading mechanism. Moreover, microinjection experiments have indicated that plasmodesmata interconnecting the PP-CC are functional, in that they allow the exchange of small membrane-impermeable fluorescent probes.³ Virus movement through plasmodesmata from the mesophyll into the SEs further supports the notion that the symplastic communication between the CC-SE complex and the neighboring cells is functional.⁴

One can assume that in apoplastic-loading plants, it would be an advantage to maintain the SE-CC complex as an isolated domain, with no functional plasmodesmata interconnecting it to the neighboring tissue. Symplastic continuity between the two domains could result in leakage of sucrose out of the vasculature and a significant reduction in the efficacy of sucrose loading. The fact that the two domains are interconnected suggests that any back-leakage of sucrose that might occur is insignificant relative to the likely efficacy of this communication route.

What might the advantage be for symplastic communication between the SE-CC complex and the neighboring tissue? Accumulated evidence suggests that at the tissue/organ level, cell-to-cell trafficking of information molecules allows for noncell-autonomous control over a range of processes, whereas at the organismal level, the phloem serves as an information superhighway, delivering a wide range of macromolecules to enable the plant to function as a whole organism.^5–8 We advanced the hypothesis that plasmodesmata interconnecting the CCs and PP/BS cells play a pivotal role in controlling the long-distance trafficking of putative signaling molecules.

Cell-to-Cell and Long-Distance Movement of Proteins

To date, there has been no experimental proof for the ability of plasmodesmata, interconnecting the phloem and the neighboring tissue, to traffic endogenous macromolecules between the two domains. In a recent study, we provided direct experimental support for the hypothesis that virus infection enables the movement of “endogenous” proteins from the mesophyll to the sieve tube.⁹ In that study, we introduced the gfp gene into transgenic tobacco plants under the FBPase promoter, resulting in localization of the protein to mesophyll and BS cells only. In cucumber mosaic virus (CMV)-infected plants, GFP expression was observed in the CCs, indicating the protein's ability to move cell-to-cell via the plasmodesmata interconnecting the BS and CCs. Moreover, expression of GFP in nontransgenic control tobacco scions grafted onto FBPase:GFP-expressing stocks infected with CMV established that the virus enables loading of the protein into the CC-SE complex and its long-distance movement to distant organs. The ability of GFP to enter the vasculature and move long distances was also evident upon infection of the grafting plants with other viruses. These results indicate that even in plants that are considered to be apoplastic loaders, plasmodesmata interconnecting the CC-SE complex and neighboring cells can be exploited by viral infection to enable nonselective trafficking of macromolecules from the mesophyll into the sieve tube. However, whether selective (or nonselective) movement of endogenous macromolecules exists without viral infection remains an open question.

The long-distance transport of endogenous macromolecules in plants also involves their trafficking through plasmodesmata that interconnect CCs to the functional sieve tube system.^10–12 It is generally assumed that SEs lack transcription and/or translation capacity and therefore, any proteins and mRNA molecules that are present in the sieve tube must have been trafficked through the plasmodesmata interconnecting the CCs and SEs. Characterization of the macromolecule profiles in phloem sap provides a useful mean to study the functioning of those plasmodesmata.

A comparison of protein profiles in the phloem sap of healthy and CMV-infected melon plants revealed 10 to 20 additional proteins in the latter, exhibiting a 5 to 10% increase in their phloem sap protein contents as compared to healthy plants. Interestingly, the protein profile of phloem sap collected from transgenic melon plants expressing the CMV-movement protein (MP) under the CaMV-35S promoter resembled the profile observed in CMV-infected plants (Malter D, Wolf S, unpublished). These results suggest that expression of the CMV-MP is sufficient to elicit the endogenous response that alters the control mechanism of selective protein trafficking into the sieve tube.

The influence of viral infection on cell-to-cell movement of proteins can be controlled by two general mechanisms: (1) alteration of plasmodesmal SEL to allow passive diffusion of molecules smaller than the altered SEL; (2) influence over a selective mechanism for the trafficking of specific proteins. Such a mechanism likely includes specific protein-protein interactions.¹³ Theoretically, one can assume that alteration in the SEL of plasmodesmata interconnecting the CCs and SEs is the cause for the diffusion of additional protein into the sieve tubes of the CMV-infected and CMV-MP-expressing melon plants. However, the molecular masses of the additional proteins were found to be in the range of 10 to over 100 kDa, with no specific cut-off size, that might indicate a simple change in SEL for the diffusion of these proteins (Malter D, Wolf S, unpublished). Moreover, the fact that only a small number of additional proteins were found in the phloem sap of infected melon plants suggests that the plant maintains some degree of selectivity, even during viral infection. We therefore suggest that the traffic of proteins from the CCs to the sieve tubes is under molecular control in virally infected plants as well, and does not necessarily occur via simple diffusion.

Identification and characterization of the additional proteins present in the phloem sap of virally infected or CMV-MP-expressing plants should shed light on their putative biological role.

mRNA as a Signal Molecule

In addition to proteins, phloem sap contains a wide range of mRNA molecules. A few recent studies have provided first evidence for the role of mRNA as a long-distance signaling molecule.^14–16 Traffic of the tomato mouse ears (Me) mutant gene from stock to a wild-type tomato scion resulted in the development of octapinnate compound leaves in the wild-type scion.¹⁶ In potato plants, trafficking of the BEL1-like transcription factor mRNA across a graft union from the leaves to the stolon tips was correlated with enhanced potato-tuber formation, suggesting that the StBEL5 gene acts as a long-distance signal molecule.¹⁴

Characterization of the RNA molecules present in the phloem sap provides another means of enhancing our understanding of the potential role of macromolecules as putative long-distance information molecules. We recently created a transcript profile of genes present in the phloem sap of melon plants.¹⁷ Functional analyses of the annotated genes revealed that over 40% of the transcripts are related to stress and defense responses, while over 15% of them are related to signal transduction. Moreover, heterografting experiments established that 6 of the 43 examined transcripts could move long distances from melon stocks to pumpkin scions. Current studies are aimed at characterizing the biological role of trafficking mRNA molecules.

Perspective

Plant growth and development are governed by sensing environmental and other external cues in various tissues that are exposed to the respective inputs, and appropriate responses in the different meristems. Trafficking of small molecules (e.g., hormones, sugars) and their action as signals that are involved in the control of the stimulation-response mechanism has long been appreciated.^18–20 Recent evidence indicates that macromolecules may also act in the plant's long-distance communication network. Of particular interest is the new finding that the protein encoded by flowering locus T (FT) moves via the phloem to the shoot apical meristem and induces flowering in Arabidopsis, rice and cucurbits.^21–23 Further studies are required to dissect the specific site at which putative signal macromolecules are transcribed/translated, to determine which sites are their targets, and to elucidate the mode by which they are delivered into and out of the sieve tubes. These studies will help determine whether long-distance movement of macromolecules is a significant component of the communication network operating at the whole plant level to regulate its development and growth.

Addendum to: Peleg G, Malter D, Wolf S. Viral infection enables phloem loading of GFP and long-distance trafficking of the protein. Plant J. 2007;51:165–172. doi: 10.1111/j.1365-313X.2007.03128.x.