Introduction

Plant cells have evolved intercellular structures, called plasmodesmata, which form membrane-lined tunnel-like channels across the cell wall and interconnect the cytoplasm of adjacent cells (Maule 2008; Brunkard et al. 2013). It is thought that this plant-specific interconnecting pathway is useful not only to share the cellular components, but also to achieve signal transduction between cells or tissues to orchestrate growth and development. The plasmodesmata have a key role in both short- and long-distance transport. A component of the plant vascular tissue, the phloem, enables long-distance transport of nutrients and signals. The phloem tube system is formed by sieve elements (SE), a highly specialized cell type forming a living phloem conduit, and adjacent companion cells (CC) that connect to the SE cells to support their function. The plasmodesmata are well developed on the SE–CC boundary to achieve active transport of photosynthetic assimilates and signal transduction. In this way, plasmodesmata facilitate intercellular communication in both local and systemic manners to integrate information and coordinate plant physiology.

The phloem tube system translocates proteins and RNAs as well as small molecules such as metabolites and phytohormones. It has been shown that some of the large molecules are used as signal agents in long-distance signaling (Lough and Lucas 2006; Kehr and Buhtz 2008). A significant example of the function of phloem-mobile proteins is a florigenic signal, which is provided by the studies on FLOWERING LOCUS T (FT) gene. The FT (or its homolog) protein is generated in leaf phloem tissues in response to photoperiod, and the transport of FT protein into target tissues can trigger flowering and the onset of bud outgrowth in the meristems above ground (Corbesier et al. 2007; Tamaki et al. 2007; Mathieu et al. 2007; Jaeger and Wigge 2007; Lin et al. 2007; Notaguchi et al. 2008; Niwa et al. 2013), and tuberization in the root stolons (Navarro et al. 2011). The phloem tube contains many types of RNA species, including siRNA, miRNA, mRNA and viral RNA. As it is now well known that RNA molecules possess a variety of activities in cells, phloem physiological functions could depend on such RNA mediators. In fact, many studies have showed that systemic silencing is achieved by siRNA function (Brosnan and Voinnet 2011). Recent studies on miRNA suggest that part of the miRNA population has a role in long-distance signaling; for example, miR399 for phosphate (Pi) homeostasis (Pant et al. 2008; Lin et al. 2008; Buhtz et al. 2010) and miR172 and miR157 for potato tuberization (Martin et al. 2009; Bhogale et al. 2014).

The transport of mRNA via the phloem has been observed in many studies. These studies have revealed that phloem transport of mRNA for some gene families is a conserved phenomenon among several plants. Phloem sap analyses have revealed that the phloem tube contains RNA-binding proteins (Balachandran et al. 1997; Xoconostle-Cázares et al. 1999; Aoki et al. 2002; Gómez et al. 2005). Some of these bind to specific sequences on the target mRNA and form RNA–protein complexes for phloem translocation (Ham et al. 2009; Cho et al. 2012). In addition, plant viruses possibly utilize host plant’s RNA transport system (Nelson and Citovsky 2005). All of these observations raise the question of whether mRNA transport via the phloem is a meaningful phenomenon. However, despite intensive studies on phloem-mobile mRNA, their biological roles remain unclear. Please see recent significant reviews for mRNAs that move over long-distance via the phloem (Kehr and Buhtz 2008; Harada 2010; Spiegelman et al. 2013). To unveil their biological function, we should examine the profiles of mRNAs that move over long-distance through the phloem tube system. This review highlights how to identify phloem-mobile mRNA, including (1) gene-expression analysis on phloem tissues or phloem sap, (2) grafting experiments to test long-distance transport, and (3) evidences of host mRNA transport into parasitic plants.

Identification of mRNA expressed in phloem tissues

The presence of mRNA in the SE–CC complex was first reported for a potato sucrose transporter, StSUC1 (Kühn et al. 1997), which is essential for sucrose export from photosynthetic leaves. In this and subsequent reports, both mRNA and protein localization in the phloem tissues were histologically tested, and the signal for SUC1 mRNA was detected near the plasmodesmata located at the cell wall boundary between the CC–SE, and the protein was localized in the CC (Kühn et al. 1997; Schmitt et al. 2008). Thus, it is assumed that a subset of mRNA/protein is tightly associated with phloem tissues for functionality. Subsequently, a series of analysis on the phloem tissues have been conducted in many plant species including monocots, cucurbits, Solanum sp., legumes and brassicas.

Expression analyses have been conducted on phloem tissues containing SE, CC and phloem parenchyma to identify genes associated with the phloem tube system. Laser capture microdissection (LCM), a powerful system for isolating targeted cells from a tissue section and analyzing gene expression profiles, has been used to isolate the phloem tissues of rice (Asano et al. 2002), apple (Kanehira et al. 2010), potato (Banerjee et al. 2006; Hannapel et al. 2013) and Arabidopsis (Deeken et al. 2008). A cDNA library was constructed for isolated phloem tissues and sequenced, and then 124 transcripts from rice and 150 transcripts from apple were identified (Asano et al. 2002; Kanehira et al. 2010). Microarray analysis of isolated phloem tissues identified 1,291 transcripts in Arabidopsis (Deeken et al. 2008). To observe the tissue-specific gene expression, the expression profiles of three sample fractions, the phloem, xylem and other cortical tissues, were compared in celery. Celery leaf petioles are good for easy isolation of such vascular strands. As a result, 73 genes were identified as predominantly or specifically expressed in the phloem or in the vascular tissues (Vilaine et al. 2003). These data sets are useful for examining the expression patterns of genes of interest in phloem tissues.

To further focus on the phloem translocation stream, phloem sap was analyzed in rice (Sasaki et al. 1998), pumpkin (Ruiz-Medrano et al. 1999), barley (Doering-Saad et al. 2002; Gaupels et al. 2008), Ricinus (Doering-Saad et al. 2006), melon (Omid et al. 2007), Arabidopsis (Deeken et al. 2008) and lupin (Rodriguez-Medina et al. 2011). Phloem sap refers to exudates from the cut edge of plant tissues such as stems and leaf petioles, or droplets from severed stylets of feeding insects as a more delicate method. Exudation from the cut surface (or feeding site) is a result of high osmotic pressure inside of the SE tube. To avoid contamination from tissues other than the phloem, the first exudates are usually discarded or washed out.

Sasaki et al. (1998) performed RT-PCR analysis of cDNA from rice phloem sap collected using insect stylets, and identified transcripts of thioredoxin h, oryzacystain I and actin, the proteins which are present in rice phloem sap (Ishiwatari et al. 1995; Schobert et al. 1995, 1998). This report was the first to detect endogenous mRNA in phloem sap. Ruiz-Medrano et al. (1999) reported ten transcripts through analysis of a pumpkin phloem sap cDNA library. The authors performed in situ RT-PCR on pumpkin tissue sections to investigate the tissue localization of candidate transcripts. First, the absence of transcripts for the Rubisco small subunit and importin α and the presence of pumpkin SUT1 mRNA in the SE cells were demonstrated. Then, the same method was applied to the candidates and signals were detected in both the CC and the SE for a total of seven transcripts (only a NAC domain protein transcript, CmNACP, was presented in the figures, the six other transcripts were mentioned in the text; see also Haywood et al. 2005 for CmGAIP). Gaupels et al. (2008) analyzed barley phloem sap collected using insect stylets by cDNA-amplified fragment length polymorphism (AFLP), and identified eight new transcripts.

Large-scale analysis was conducted on phloem sap collected by cutting stems or leaf petioles. Doering-Saad et al. (2006) analyzed Ricinus phloem exudates collected at two time points; the first 15 min and after 120 min of exudation. A total of 1,012 cDNA clones were sequenced, and 158 unique transcripts were identified from the late fraction. Similarly, Omid et al. (2007) identified 986 unique transcripts from melon phloem sap. Rodriguez-Medina et al. (2011) identified 609 unique transcripts from the phloem sap of the legume lupin, collected from cut fruits, inflorescence stalks and at the base of the stem. Deeken et al. (2008) used a model plant Arabidopsis and collected phloem exudates from 10-week-old rosette leaves for 1 h using the chelator EDTA to prevent calcium-dependent sieve tube occlusion by proteins and callose (King and Zeevaart 1974). Through microarray analysis, 2,417 transcripts were identified. As EDTA buffers usually causes cell shrinkage, contaminants from cells on the cut surface may not be negligible. However, the authors compared the phloem exudate data with two other data sets from LCM-prepared phloem tissues (presented in the same paper, see above) and ESTs constructed from CC protoplasts (Ivashikina et al. 2003), and identified 114 typical phloem transcripts.

These data sets provide useful information on the transcripts highly associated with the phloem translocation system. The data include transcripts related to metabolism, structure, DNA and RNA binding, metal homeostasis and transport, redox, stress and signal transduction. Each data set could reflect the site of collection, plant developmental stage, and the purity of the phloem sap in the respective experiments. Finally, it is important to note that the profiles obtained from these phloem exudates are not directly connected to the information for phloem-mobile mRNA. To conclude an mRNA is phloem-mobile, its transport needs to be examined (see the next section).

Evidence for long-distance transport of mRNA through grafting experiments

Grafting is a technique used to attach and connect two (or more) different plant materials to make a single plant unit. The root part is called the rootstock, and the shoot part is called the scion. As a successful graft establishes vascular connections between the stock and the scion, testing molecular transport across the graft junction provides strong evidence for long-distance transport via the phloem. The long-distance transport of several mRNAs has been demonstrated by grafting in a broad range of plants.

KNOX-BEL

Non-cell-autonomous behavior of plant mRNAs and proteins was first documented through the analysis of a maize homeobox transcription factor, KNOTTED1 (KN1). It is thought that plant KNOTTED1-like homeobox (KNOX) genes are required for meristem development and functions associated with broad organogenesis (Hay and Tsiantis 2010; Scofield and Murray 2006). Later studies on some KNOX genes implicated their expression in phloem tissues and long-distance mobility via the phloem tube system. In this section, studies on the short-distance movement of KN1 protein/mRNA are summarized and then the long-distance transport of mRNA is discussed, because these phenomena may link together in terms of the ability of KN1/KNOX to move through the plasmodesmata translocation pathway.

The maize dominant mutant Knotted (Kn) perturbs normal leaf development and extra growths, called knots, cause lobed leaves (Hake 1992). From two findings, (1) that the Kn product is required only in the inner mesophyll layer to exert the Kn effect in all leaf tissues (as demonstrated by a chimeric study; Hake and Freeling 1986), and (2) that the KN1 mRNA localization pattern in the shoot apical meristem is different from that of the KN1 protein (mRNA localization is restricted to the inner L2, L3 layers, but protein localization is found in all L1, L2, L3 layers), it was hypothesized that the KN1 protein has the capacity to translocate from cells in which it is expressed to surrounding cells/tissues (Jackson et al. 1994). Subsequent studies supported this idea (Lucas et al. 1995; Kim et al. 2005). Furthermore, the facilitation of KN1 mRNA movement from cell to cell or from layer to layer by its own protein was demonstrated in tobacco mesophyll cells and Arabidopsis leaves and shoot apical meristem, respectively (Lucas et al. 1995; Kim et al. 2005).

Similar layer-to-layer movement was found for the ANGUSTIFOLIA3 (AN3) gene. Chimeric studies and characterization using an AN3-GFP fusion reporter indicated that the AN3 protein translocates from the mesophyll to the epidermis, orchestrating cell proliferation and expansion between inner and outer tissues in the leaves (Kawade et al. 2010, 2013). Please see Kawade and Tanimoto (2015) another JPR symposium issue, for further discussion. Such trans-regulation is a reasonable mechanism for balanced growth. Thus, intercellular communication via plasmodesmata may have a fundamental role in coordinating multiple differentiated tissues into a singular plant organism.

An interesting finding was that, like the KN1 protein, the facilitation of its own mRNA transport by a protein was also reported in studies of the CmPP16 phloem RNA binding protein, identified as a plant paralog to a viral movement protein (Xoconostle-Cázares et al. 1999), and eukaryotic initiation factor 5A (Ma et al. 2010). There could be a reason explaining this behavior. For protein trafficking, the mRNA template might be trafficked close to the plasmodesmata for the efficient recruitment of the translated protein cargo to the trafficking pathway, like β-ACT mRNA translocation to the cell periphery in animal neuronal cells (Hüttelmaier et al. 2005). If that is the case, non-target specific RNA binding proteins, found in the phloem, possibly traffic template mRNA in a non-selective manner. Thus, an active protein transport system could accompany the mRNA leakage into the phloem tube.

Recent works reporting FT mRNA mobility might be explained by the above scenario (Li et al. 2009, 2011; Lu et al. 2012). In the case of FT action, several groups have provided many kinds of evidence supporting the current model; “the FT gene action to promote flowering is dependent on the transport of FT protein, not mRNA” (Mathieu et al. 2007; Jaeger and Wigge 2007; Notaguchi et al. 2008). Large-scale grafting experiments, using a synonymous FT gene that has synonymous substitutions in 171 of 175 codons, indicate that the primary sequence and secondary structure of the mRNA are not of critical importance for the long-distance action of the FT gene (Notaguchi et al. 2008). Thus, it is a good example to remind us of the need for careful validation of the biological meaning of mRNA transport.

The many facets of plant development impacted by the class I KNOX genes, which include maize KN1, have been reported one after another. One of the tomato class I KNOX genes, LeT6, is expressed throughout the entire shoot apical meristem and leaflet primordia and developing leaflet margins (Chen et al. 1997; Parnis et al. 1997). Overexpression of LeT6 causes changes in leaf morphology including higher order compounding and rounded leaflets. The Mouse ears (Me) tomato dominant mutant in which LeT6 is overexpressed as a result of gene fusion with the adjacent PFP, a glycolytic pathway enzyme gene, shows a similar phenotype to LeT6 overexpression lines (Chen et al. 1997; Parnis et al. 1997). Grafting of the Me mutant to wild-type tomato plants caused developmental changes in wild-type leaves. As PFP-LeT6 fusion transcripts were detected from the SE and the shoot apex of grafted wild-type scions, it was concluded that the transmission of PFP-LeT6 fusion transcripts across the graft union caused changes in the leaf morphology of the wild-type scions. A little caution is required in terms of which transcript sequence gave the transmissibility, as the author mentioned the mobility of both PFP and LeT6 endogenous transcripts (Kim et al. 2001). This was the first report that long-distance transport of mRNA could exert an effect on morphological development in plants. This graft-transmissible effect on leaf development was also observed when wild-type potato plants were grafted onto the Me tomato mutant stock (Kudo and Harada 2007).

mRNA localization in the CC-SE phloem complex was reported for three class I KNOX genes; LeT6 in tomato (Kim et al. 2001), CmSTMP in pumpkin (Ruiz-Medrano et al. 1999) and Potato Homeobox1 (POTH1) in potato (Yu et al. 2007; Campbell et al. 2008; Mahajan et al. 2012). The long-distance transport of CmSTMP and POTH1 mRNA was evidenced by hetero-grafting experiments in pumpkin (Ham et al. 2009) and potato (Mahajan et al. 2012). POTH1 interacts with a protein partner, StBEL5, and the transcriptional complex binds to a specific sequence motif to regulate target promoter activities (Chen et al. 2003, 2004). Transgenic potato lines overexpressing POTH1 exhibited a dwarf phenotype, altered leaf morphology and enhanced tuber formation, indicating that POTH1 regulates numerous developmental processes in potato (Rosin et al. 2003). The long-distance transport of POTH1 mRNA was evidenced by grafting experiments using potato in vitro plantlets of the wild-type and a POTH1 overexpression line, and using tobacco heterologous transgenic lines (Mahajan et al. 2012).

Interestingly, a BEL partner, StBEL5, is also transported long-distances in its mRNA form, which was demonstrated earlier than for POTH1 (Banerjee et al. 2006). StBEL5 is expressed over the entire plant body including newly formed tubers. The expression of StBEL5 increases in response to short day conditions, which are inductive for tuber formation (Banerjee et al. 2006). StBEL5 was also detected from LCM-prepared phloem tissues and phloem sap (Yu et al. 2007; Campbell et al. 2008). In potato in vitro micrografting experiments between an StBEL5 overexpressor scion and a wild-type stock, the transgene-derived StBEL5 mRNA was detected in the wild-type stocks (Banerjee et al. 2006). Thus, the phloem-based long-distance transport of mRNAs for KNOX-BEL may have a role in potato tuberization. In future studies, the necessity of mRNA transport to exert gene functions needs to be investigated to reach a clear conclusion.

As described here, the mobility of KNOX and BEL mRNAs enables plants to trans-regulate and coordinate physiological processes in local and distal manners. This finding may suggest that a common control mechanism might exist at the plasmodesmata gate system for local cell-to-cell movement and for long-distance transport.

Repressors in signaling pathways for gibberellic acid and auxin

The phytohormones gibberellin and auxin control an impressive variety of developmental processes. The long-distance transport of the transcripts for two key repressors in these hormones signaling pathways, GIBBERELLIC ACID INSENSITIVE (GAI) and Aux/IAA, have been well characterized in several plant species. Under the high hormone concentrations in the cell, GAI and Aux/IAA repressor proteins are targeted by the SCF degradation machinery, and the release of inhibition allows hormone action (Harberd et al. 2009; Parry and Estelle 2006). Mutations in a domain that is required for the normal degradation process cause a stable protein form, resulting in a hormone insensitive phenotype. In the following studies, such gain-of-function mutants were used in grafting experiments to see whether the mutant effect was transmitted to a wild-type recipient.

In a hetero-grafting experiment, pumpkin GAI (CmGAIP) mRNA was identified from a cucumber scion grafted onto a pumpkin stock (Ruiz-Medrano et al. 1999). Overexpression lines for CmGAIP and Arabidopsis GAI (AtGAI), both of which contain deletions of the DELLA domain, were generated in tomato and Arabidopsis. After grafting of these transgenic plants showing altered leaf phenotypes, the grafted wild-type scions also showed a similar leaf phenotype to the donor transgenic plants. The transport of both plant types of GAI mRNA was detected from the shoot apex of the wild-type tomato scion or the inflorescence of the wild-type Arabidopsis scion (Haywood et al. 2005). The mobility of AtGAI was also tested in Nicotiana benthamiana and apple plants heterologously (Xu et al. 2013). Furthermore, the long-distance transport of three kinds of apple GAI mRNA was detected by grafting of apple trees (Xu et al. 2010; Wang et al. 2012).

Phloem-mobile Aux/IAA transcripts were identified in melon (Omid et al. 2007). In hetero-grafting experiments between a melon donor and a pumpkin recipient, six melon transcripts out of 43 examined were identified from the pumpkin recipient scion, including two Aux/IAA transcripts (F-308 and F-571; both sequences are similar to Aux/IAA14). The transmissibility of apple IAA14 mRNA was also evidenced by grafting experiments (Kanehira et al. 2010). Our group investigated the function of phloem-mobile Aux/IAA mRNA using Arabidopsis systems (Notaguchi et al. 2012). First, six out of 29 Aux/IAA family genes were identified as preferentially expressed in vascular rich samples, isolated from mature rosette leaves by mild sonication in cellulase buffer (Endo et al. 2005). Next, hetero-grafting was conducted to test the mobility of the six candidate transcripts, and two transcripts (for IAA18 and IAA28) were identified from N. benthamiana scions grafted onto Arabidopsis donor stocks. As the loss-of-function mutants did not show any phenotype, dominant mutants were used to further address their function in micrografting experiments. The action of the IAA18 and IAA28 dominant mutants to suppress lateral root formation was transmitted to the wild-type rootstock. RT-PCR analyses confirmed their mRNA transport to the root tip of the rootstock. Using a virus vector system, it was shown that neither protein has the capacity to enter the sieve tube system. Taking these results together, we concluded that IAA18 and IAA28 mRNAs, but not their proteins, are transported long-distance via the phloem, and such mRNA transport can cause physiological changes in the destination tissues (Notaguchi et al. 2012).

In works on both GAI and Aux/IAA mRNA, only the graft-transmissible action of dominant mutants was demonstrated, therefore their biological roles remain undetermined. However, it is certain that mRNA transport can exert functions and affect physiology where the mRNA is transported. The phenotypic changes in the wild-type recipients are likely explained by the action of mutant proteins, implying the occurrence of protein translation after mRNA transport. Because these proteins are short-lived, transport in mRNA form may enhance the signaling efficiency.

Interesting coincidences are that each transcript encodes a key repressor in a hormone signaling pathway, that each of the proteins is degraded in the presence of hormones in the cell, and that each transcript has the capacity to access the plasmodesmata-based translocation system in mRNA form. As both mRNAs have the capacity to pass through the plasmodesmata in phloem tissues, cell-to-cell movement in the other tissues might be possible. Another factor is that gene expression of Aux/IAA is promoted by auxin itself. These findings, especially for those in the auxin signaling pathway, raise the hypothesis that cell-to-cell movement of Aux/IAA mRNA from the region of auxin maximum to surrounding cells provides a mechanism for lateral inhibition, and, by such a mechanism, spatially tight regulation of auxin action is achieved. Of course, the long-distance transport of these mRNAs could have a role as indicated in the original reports.

Translocation of Host mRNA into parasitic plants

Host-parasite interactions bear some analogy to the hetero-grafting situation in terms of the chimerism of two different genotypic backgrounds. Interestingly, host mRNAs are translocated into parasitic plants across the parasite haustorium. Although the phloem connection is typically not clearly structured between host and parasitic plants, mRNA that has the capacity to move across the plasmodesmata can translocate from host cells to cells in the parasite. In host-parasite interactions, mRNA might experience short cell-to-cell transport at the connecting region, and then long-distance transport via the parasite phloem tubes. Studies on the parasitic plant dodder (Cuscuta) are introduced below.

Roney et al. (2007) attempted to detect pumpkin phloem transcripts, presented in Ruiz-Medrano et al. (1999), from dodder grown on pumpkin, and identified three transcripts, CmNACP, CmSUTP1 and CmWRKYP. Further analyses were performed on dodder grown on tomato using a tomato microarray. In this case, dodder grown on other hosts—Arabidopsis, tobacco or pumpkin—were also analyzed with the tomato microarray to exclude nonspecific signals, and 474 tomato transcripts were identified. Further RT-PCR analysis showed the trafficking of 10 tomato transcripts out of 28 examined, including tomato GAI (LeGAI). In a review by the same group, the authors mentioned that RT-PCR validation confirmed the mobility of 60 % of the candidate genes, or over 250 tomato transcripts (Westwood et al. 2009). David-Schwartz et al. (2008) detected two α and β subunits of LePFP, Rubisco small subunit transcript and LeGAI from dodder grown on tomato by RT-PCR. LePFP transcripts were detected in parenchyma cells as well as in the phloem of dodder, suggesting that the mRNA is trafficked via symplasmic continuity through parenchyma cells to the dodder phloem. LeBlanc et al. (2013) performed RNA-Seq analysis of dodder grown on tomato and Arabidopsis. Although not all of their data was presented, qRT-PCR was performed for three abundant Arabidopsis transcripts, Salt-inducible Zinc Finger Protein (AtSZF1), Auxin Response Family (AtARF) and Translationally Controlled Tumor Protein (AtTCTP), and two tomato transcripts, tomato Cathepsin D Proteinase Inhibitor (SlPI) and tomato GAI (SlGAI), and confirmed their mobility. Thus, it seems that numerous mRNAs have the capacity to move from host tissues to the parasitic plant. Also, for some transcripts, long-distance mRNA transport was detected both in grafting and host-parasite interaction.

Conclusions and future perspectives

There is now much evidence for mRNA transport via the phloem tube and the ability of mRNAs to exert their molecular functions in destination tissues. Further characterization of phloem-mobile mRNAs has provided evidence for their biological roles. However, as all studies have used gene overexpression lines or gain-of-function mutants to demonstrate the transmission of gene action, so far the meaning of long-distance movement is not yet truly clarified. Thus, further identification of phloem-mobile mRNA is needed. As previous studies have demonstrated, a combination of expression analysis of phloem enriched samples and grafting experiments is useful to search for candidates and to test mobility over long-distances. Our research group has applied the RNA-Seq method to N. benthamiana/Arabidopsis hetero-grafting samples to obtain an exhaustive profile of phloem-mobile mRNA, and has identified 138 Arabidopsis transcripts from the N. benthamiana scions to date. The list includes transcripts for one homeobox protein, one BEL1-like protein and one Aux/IAA, but not for GAI, suggesting that the sequencing depth was below the saturation point. These results were further validated by RT-PCR and digital PCR. Overlap was found between our data (Notaguchi M, Higashiyama T and Suzuki T, unpublished results) and other data from host–parasite interactions (summarized in Westwood et al. 2009), implying that both strategies truly identified mobile transcripts. Thus, information on phloem-mobile mRNA can be now obtained for plants on different developmental stages or grown under specific environmental conditions.

The identification of phloem-mobile mRNA is just the beginning of such studies. Potential questions remain, such as: (1) which mRNAs are transported? Is there any selection or do all transcripts have the potential to move? (2) When are they transported; all the time whenever expressed or only under special internal/environmental conditions? (3) Where is their destination? (4) What kind of mechanism underlines for the mRNA long-distance transport? Is mRNA delivered close to the plasmodesmata inside of the originated cell first? Do plants possess any molecular system for loading into the SE tube, for transport via the tube system to block leakage from the tube or to repossess leakages from the adjacent cells, and for unloading to the surrounding tissues and subsequent cell-to-cell movement to the final destination cell/tissue? (5) What is the fate of the transported mRNA? Are they translated? If so, do the proteins have the same activity/modifications as proteins that do not experienced transport? Is there any mechanism to degrade the signal after they exert their activity to avoid continued stimulation? (6) What is their role(s) in plant growth and development? Through intensive studies for decades, we have many clues to each of these questions, but no conclusive answers yet. For furthering our understanding, obtaining phloem-mobile mRNA profiles will be useful to facilitate future studies.

In addition, a visualization system for mobile transcripts in living tissues is desired to investigate their routes and the dynamics of each process during long-distance transport. RNA imaging using genetically encoded probes is one choice (Christensen et al. 2010; Paige et al. 2011; Strack et al. 2013). Another challenge is seen in the techniques for phloem-mobile mRNA detection. Because of the low amounts of target transcripts, many groups have experienced inconsistent results between technical replicates or between different experimental methods. Thus, an easier, convincing method to test candidate transcript mobility will be a breakthrough. If a cell sorter can succeed in isolating the phloem cells from fluorescent protein tagged-SE marker lines (Thompson and Wolniak 2008; Ernst et al. 2012), such an isolation technique would be worth applying to grafting/RNA-Seq experiments to enrich target phloem-mobile mRNAs. Knowledge of phloem-mobile mRNA will not only give us a greater understanding of plant systems but also new useful skills to control agricultural traits.