Meeting Report
Plasmodesmata 2004. Surfing the Symplasm1 Christine Faulkner, Jeri Brandom, Andy Maule, and Karl Oparka* Faculty of Science, University of Sydney, New South Wales 2006, Australia (C.F.); Section of Plant Biology, University of California, Davis, California 95616 (J.B.); Department of Disease and Stress Biology, John Innes Centre, Norwich NR4 7UH, United Kingdom (A.M.); and Unit of Cell-Cell Communication, Scottish Crop Research Institute, Dundee DD2 5DA, United Kingdom (K.O.) Communication between cells is necessary to coordinate plant development and physiology, and is modulated in response to environmental signals and pathogen attack. Plasmodesmata (PDs; singular plasmodesma), plasma membrane-lined channels that cross the cell wall, are key components of this intercellular communication network. Historically, PDs were largely viewed as little more than channels that allowed the passive movement of small molecules such as water and metabolites between cells. Research over the past few decades has revealed that PDs are structurally and functionally dynamic and complex. It is now recognized that PDs are a central hub for intercellular transportation of a wide range of solutes, proteins, signaling molecules, and ribonucleoprotein complexes. Rapid progress has been made in many facets of plasmodesmal research, as reported at the recent 5th International Plasmodesmata meeting held at Asilomar, Pacific Grove, California, in August 2004. The Asilomar meeting is the first dedicated solely to PD research in the past 3 years, a period that has shown an expanding interest in cell-to-cell communication in plants. Although, traditionally, the PD field has brought together cell biologists and virologists, in recent years the subject has expanded to include developmental biologists and experts in gene silencing. Indeed, the topic of PD research has gained increasing relevance, as an increasing number of plant proteins and endogenous RNAs have been shown to act in a non-cell autonomous manner. Of particular topical interest is the movement of gene silencing signals. In one of the plenary talks, Olivier Voinnet (IPMB-CNRS, Strasbourg, France) elegantly described how the small RNAs (sRNAs) of 21 and 24 nucleotides associated with RNA silencing were differentially involved in local (cell-to-cell) transmission of the silencing signal. Mutant plants defective in the accumulation of the 21-nucleotide but not the 24-nucleotide sRNAs showed no movement of the silencing phenotype. Many research groups are now tackling the problem of how these and other macromolecules are transported between cells and throughout the 1 C.F. and J.B. were recipients of the two awards at the conference for outstanding postgraduate presentations. * Corresponding author; e-mail
[email protected]; fax 01382– 568575. www.plantphysiol.org/cgi/doi/10.1104/pp.104.057851.
plant. This resurgence and expansion of interest is being assisted by a major technological shift toward the use of ‘‘omics’’-based tools and genetic resources.
MACROMOLECULAR COMPONENTS OF PD
A major constraint in advancing our understanding of PDs has been a lack of knowledge of their constituent components. Various indirect approaches have previously implicated components of the cytoskeleton, proteins that interact with viral MPs, and molecular chaperones in PD function. Until recently, potential plasmodesmal components have been difficult to identify, in part because the location of PD embedded in the cell wall makes them intractable to biochemical analysis. At this meeting, several groups reported new, more direct approaches to identifying PD proteins. In an exciting development, Pat Zambryski and Insoon Kim (University of California, Berkeley) reported the use of a genetic screen for altered intercellular trafficking patterns in Arabidopsis embryos. One mutant, increased size exclusion 1 (isel1), was identified as a DEAD-box RNA helicase. Appropriately, as a green fluorescent protein (GFP) fusion protein, ISEL1 colocalized with tobacco mosaic virus (TMV) movement protein (MP) as punctate spots on the cell wall. Three groups (Bernie Epel, Tel Aviv University, Israel; Andy Maule, John Innes Centre, Norwich, UK; and Robyn Overall, Sydney University, Australia) reported the application of proteomic technologies for identifying PD proteins from purified PDs or cell wall fractions enriched for PDs. Again, these candidates were assessed by their localization as GFP fusions. One protein that appeared in all these analyses was a REVERSIBLY GLYCOSYLATED POLYPEPTIDE 2. This Golgi-associated protein appears to be targeted to PDs. An important milestone in this field will be the identification of the unique structural components of PDs. By analogy with the nuclear pore complex, we might expect there to be several tens of structural components. One intriguing possibility was a tropomyosin-like protein identified from the cell walls of Chara corallina (Christine Faulkner and Robyn Overall) that localized along the length of the PD. The abundance of putative PD-associated proteins identified in these recent studies may mean that the long drought in the isolation of PD components is over.
Plant Physiology, February 2005, Vol. 137, pp. 607–610, www.plantphysiol.org Ó 2005 American Society of Plant Biologists
607
Faulkner et al.
Particularly encouraging was the appearance in the lists of candidate proteins of proteins found previously to associate with the PD and the isolation of the same proteins from different studies, for example, the identification of a DEAD-box RNA helicase by both proteomic and genetic screens. A complementary and valuable approach to the challenge of identifying PD functional components is to ask which plant proteins interact with viral or endogenous proteins known to move through PDs. One of the first successful examples of this approach was the identification of pectin methyl esterase interacting with TMV MP, and it was comforting to hear that pectin methyl esterase was identified in the lists of proteomics-derived candidates. From reports at the meeting, we can now add the DnaJ-like protein CPIP1 (Daniel Hofius, Institute of Plant Genetics and Crop Plant Research, Gatersleben, Germany), eIF4E (Andy Maule), and FICa (Takuji Wada, RIKEN, Yokohama, Japan) as proteins that interact with the potato virus Y capsid protein, pea seed-borne mosaic virus VPg, and the non-cell autonomous protein CAPRICE, respectively.
MODES OF PD TRANSPORT
Two modes of operation have been suggested to explain patterns of intercellular trafficking through PD. One is nontargeted trafficking in which the size exclusion limit of the PD is sufficient to allow trafficking of macromolecules not specifically targeted elsewhere within the cell. This could be considered to be equivalent to diffusion. Alexis Maizel (Max Planck Institute for Developmental Biology, Tuebingen, Germany) asked the following question: Is nontargeted movement from cell to cell the default state for many macromolecules? Maizel addressed this by creating nonoverlapping deletion mutants of the transcription factor LEAFY. When fused to GFP, none of these deletion mutants lost the ability to move from cell to cell, suggesting that movement of LEAFY is nontargeted. Further analysis of various plant proteins expressed as GFP fusions showed that all but those expected to form large protein complexes could move from cell to cell, further supporting the idea that movement is the default state. A second model of trafficking is targeted, or selective, trafficking in which interaction of translocated macromolecules with PD components mediates a change in the size exclusion limit of the pore. Viral MPs are typical examples of targeted movement. One proposed means of regulating selective trafficking is phosphorylation. Jung-Youn Lee (Delaware Biotechnology Institute, Newark, Delaware) and colleagues have screened for a potential PD-associated protein kinase by testing fractionated PD-enriched cell wall preparations from BY-2 cells for phosphorylation of TMV MP. A putative PD-associated protein kinase belonging to the CASEIN KINASE 1 gene family was isolated that phosphory608
lates endogenous proteins and selectively phosphorylates viral movement proteins in vitro. Localization studies of Arabidopsis CASEIN KINASE 1 isoforms demonstrate various subcellular localization patterns, including a group that label punctae at the cell periphery, a pattern consistent with PD localization. Macromolecules that traffic by a selective or targeted pathway might be expected to contain sequences necessary for translocation. Kimberley Gallagher (Duke University, Durham, NC) was able to show that, while movement of the transcription factor SHR is dependent upon a single residue in the protein and is therefore representative of targeted and active movement, SHR cannot be trafficked from cell to cell when localized to the nucleus. This allows the speculation that chaperones that facilitate movement of SHR are localized in the cytoplasm and that SHR (which contains a nuclear localization signal) needs to interact with exportins to localize to the cytoplasm. Jae-Yean Kim (Gyeongsang National University, Jinju, Korea) used a trichome rescue assay to identify a trafficking motif of the endogenous transcription factor KNOTTED1 (KN1). Fusion of the KN1 homeodomain to the cell-autonomous GL1 conferred trichome rescue in gl1 mutants. Fritz Kragler (University of Vienna, Austria) demonstrated that MPB2C, previously described as a negative regulator of TMV MP movement, is able to block movement of KN1. Cell-to-cell movement of KN1 is not impaired simply because of the presence of a large amount of interacting protein, as the KN1-interacting protein KNB36 has no effect on KN1 trafficking in microinjection experiments. Furthermore, interaction of KN1 with MPB2C in yeast two-hybrid assays was dependent upon the KN1 homeodomain. One of the questions posed at the meeting related to the mechanism by which macromolecules traffic from cell to cell. Recent work by Karl Oparka’s group (Scottish Crop Research Institute, Dundee, UK) identified a Rab protein that localizes to plasmodesmata (Medina Escobar et al., 2003). This family of proteins is involved in the targeting of vesicles to specific locations in the cell, and thus the following question was raised: Are vesicles targeted to the plasmodesmata? This theme was continued by Alison Roberts (Scottish Crop Research Institute), who provided evidence that two viral MPs of potato mop top virus interact with components of the endocytic machinery, notably a protein involved in receptor-mediated endocytosis (RME-8). The intriguing concept of endocytosis was raised again in Alexis Maizel’s investigation of the secretion and internalization of homeodomain proteins fused to GFP in COS-7 and primary neuron co-cell culture. He was able to demonstrate that, while the full-length KN1 protein fused to GFP could not move from the COS-7 cells into the neurons, the KN1 homeodomain fused to GFP was able to move. If the movement of this protein from cell to cell in animals is analogous to the movement from cell to cell in plants, this raises the following intriguing question: Are homeodomain proteins, rather than being transported through Plant Physiol. Vol. 137, 2005
Plasmodesmata 2004: Surfing the Symplasm
plasmodesmata, trafficked by a mechanism that more closely resembles the secretion and internalization mechanism adopted by animal cells? This question is likely to be debated hotly in the near future. A common view of PD trafficking is that molecules exploit the cytoplasmic sleeve between the plasma membrane lining of the PD and the central rod-like appressed ER component, often called the desmotubule. Delivery to this location might involve targeting motifs on vesicles, while directionality to and through PDs is provided by components of the cytoskeleton. Talks by Manfred Heinlein (IPMB-CNRS) and Petra Boevink (Scottish Crop Research Institute) examined the role of the microtubule (MT) cytoskeleton in delivering MP-RNA complexes to plasmodesmata. Although both groups agreed that the decoration of MT later in the infection process is not involved in cellto-cell transport, debate continues as to whether the targeting of MP-RNA to plasmodesmata requires the actin or MT cytoskeleton. Traditionally, we have assumed that the lumen of the ER is inert with respect to molecular movement between cells. Both Robyn Overall and Bernie Epel addressed the question of whether the ER forms a functional transport pathway through the PD. Epel presented data that suggest that, in the presence of TMV MP, the desmotubule dilates and that proteins can be transported from cell to cell through the ER lumen. Overall described intricate experiments that showed that microinjection of a 3-kD dye into the ER lumen of tobacco trichomes moved into the adjacent cells, first being seen in the nucleus of the neighboring cells. However, FRAP analysis of ER-targeted GFP suggested that ER itself was not moving from cell to cell.
TISSUE CONNECTIVITY AND LONG DISTANCE TRANSPORT
Plasmodesmata are ubiquitous throughout plant tissues and, except for the few examples of symplasmically isolated cells (e.g. stomatal guard cells), provide symplasmic continuity throughout the body of the plant. Although crucial in establishing the routes for symplasmic transport, intercellular flow does not appear to be regulated by the abundance of PDs. In an exhaustive study of plasmodesmata frequency and distribution in mature stems, Aart van Bel (JustusLiebig-University, Giessen, Germany) determined that plasmodesmatal density does not appear to control the transport capacity of the dye lucifer yellow from cell to cell. This was similar to the findings of Bob Turgeon (Cornell University, Ithaca, NY), who, by comparing phloem loading in plants that translocate raffinosefamily oligosaccharides with those that translocate Suc, concluded that plasmodesmata frequency in the minor vein phloem has little to do with phloem loading. The phloem has emerged as a crucial element in defining the long-distance routes for signaling within the plant. The concept of the phloem as Plant Physiol. Vol. 137, 2005
a superhighway for the delivery of systemic informational molecules (Bill Lucas, University of California, Davis) has gained considerable support from the study of movement of viruses and endogenous macromolecules. Recent work establishing that non-cell autonomous macromolecules are able to function at a supracellular level has, as Lucas stated, raised more questions than answers about the roles of endogenous macromolecules translocated in the phloem, and the cellular components mediating this process. Some of the issues the Lucas lab is addressing include the following. Are proteins and transcripts selectively transported in the phloem? If so, what are the mechanisms that establish selectivity? Also, is the RNA present in the phloem involved in signaling and/or transmission of phenotype? RNA-binding proteins (RBPs) are likely to be one of the components involved in the transmission of RNA molecules in the phloem. Eriko Miura of the Lucas lab has identified a pumpkin phloem-localized protein, orthologous to eIF-5A, that interacts with exportin 4 (Lipowsky et al., 2000) and is involved in RNA shuttling (Bevec et al., 1996) in animal cells. Investigation into the specific roles of these RBPs will help to elucidate the mechanisms of long-distance RNA transport in the phloem. Ultimately, the purpose of local and remote signaling is to maintain a tightly regulated program for growth and development, plant defense, and physiological control. The breadth of impact of these processes is being addressed by Shmuel Wolf (The Institute of Plant Sciences in Agriculture, Rehovot, Israel) and colleagues, who are using both genomic and proteomic approaches to study the composition of phloem exudate. Microarrays will be used to compare gene expression profiles in flowering and nonflowering melon. Comparison of proteins isolated from phloem exudate of these plants identified several proteins present only in flowering plants, including three protein kinases. Interestingly, analysis of 1,200 clones of a cDNA phloem library returned 81% singletons, confirming the large population of transcripts present in phloem exudate.
THE WAY FORWARD
Speculation on the application of new technologies for plasmodesmal research was also considered at the meeting. Karl Oparka described the many different approaches to image the trafficking between cells in his opening address at the meeting. The vast array of fluorescent technologies available, such as FlAsH, fluorescent proteins, and Q-Dots, has opened up the potential for detailed analysis of the transport capacity of PDs. Further, Wolf Frommer (Carnegie Institution of Washington, Stanford, CA) presented the development of nanosensors for the imaging of metabolites. Some of these techniques might answer the following key questions that arose at the meeting. Do homeodomain proteins really traffic through PDs? What is the role of vesicle-mediated trafficking with 609
Faulkner et al.
respect to PD function? What are the roles of the expanding list of putative PD-associated proteins identified? What function do RBPs have with respect to short- and long-distance transport of mRNA? How are macromolecules targeted to PD? This is an important and exciting area of research that will have impacts across the whole of plant biology, but there are many challenges ahead. Bill Lucas and David Jackson (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY) called for collaboration between researchers to meet these challenges. The Plasmodesmata meetings are opportunities for the initial sharing of information and ideas, and Plasmodesmata 2004 was no exception. We regret that, due to space restrictions, we are unable to detail many of the interesting presentations at the meeting, and apologize to colleagues for omissions. The organizers, David Jackson, Rick Nelson, Pat Zambryski, and Bob Tur-
610
geon, did a wonderful job of putting together a fascinating scientific program in a beautiful location by the Pacific. Received December 8, 2004; returned for revision December 16, 2004; accepted December 16, 2004.
LITERATURE CITED Bevec D, Jaksche H, Oft M, Wohl T, Himmelspach M, Pacher A, Schebesta M, Koettnitz K, Dobrovnik M, Csonga R, et al (1996) Inhibition of HIV-1 replication in lymphocytes by mutants of the Rev cofactor eIF-5A. Science 271: 1858–1860 Lipowsky G, Bischoff FR, Schwarzmaier P, Kraft R, Kostka S, Hartmann E, Kutay U, Gorlich D (2000) Exportin 4: a mediator of a novel nuclear export pathway in higher eukaryotes. EMBO J 19: 4362–4371 Medina Escobar N, Haupt S, Thow G, Boevink P, Chapman S, Oparka K (2003) High-throughput viral expression of cDNA-green fluorescent protein fusions reveals novel subcellular addresses and identifies unique proteins that interact with plasmodesmata. Plant Cell 15: 1507–1523
Plant Physiol. Vol. 137, 2005
