• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of plntcellLink to Publisher's site
Plant Cell. Dec 2003; 15(12): 2778–2791.
PMCID: PMC282798

Nodule Initiation Involves the Creation of a New Symplasmic Field in Specific Root Cells of Medicago Species


The organogenesis of nitrogen-fixing nodules in legume plants is initiated in specific root cortical cells and regulated by long-distance signaling and carbon allocation. Here, we explore cell-to-cell communication processes that occur during nodule initiation in Medicago species and their functional relevance using a combination of fluorescent tracers, electron microscopy, and transgenic plants. Nodule initiation induced symplasmic continuity between the phloem and nodule initials. Macromolecules such as green fluorescent protein could traffic across short or long distances from the phloem into these primordial cells. The created symplasmic field was regulated throughout nodule development. Furthermore, Medicago truncatula transgenic plants expressing a viral movement protein showed increased nodulation. Hence, the establishment of this symplasmic field may be a critical element for the control of nodule organogenesis.


Cell fate in plants is determined mainly by positional information rather than cell lineage. Plant cells are characterized by a high degree of totipotency, which means that differentiated cells can dedifferentiate, divide, and/or acquire new identities. This developmental plasticity stresses the importance of cell-to-cell communication in plant development. Plant cells are connected to their neighboring cells via plasmodesmata, small channels that span the adjoining cell walls. Such an interconnected network of cells is referred to as a “symplasmic domain” (Erwee and Goodwin, 1985) when these cells are isolated completely from other tissues and as a “symplasmic field” when this isolation is partial (Rinne and Van der Schoot, 1998). Plasmodesmata that form in new cell walls do so during cell division by fusion of the phragmoplast and are termed primary plasmodesmata, whereas those that form de novo across preexisting cell walls are referred to as secondary plasmodesmata (Ehlers and Kollmann, 2001). Various viral movement proteins (MPs), as well as a number of endogenous plant proteins, have been shown to modify the size exclusion limit (SEL) of plasmodesmata (Wolf et al., 1989; Lazarowitz, 1999; Xoconostle-Cazares et al., 1999). Mutants that affect the SEL also have been characterized in Arabidopsis and showed embryo-lethal phenotypes (Kim et al., 2002). Several studies have revealed the existence of temporal and spatial regulation of symplasmic domains in plant development (Rinne and Van der Schoot, 1998; Gisel et al., 1999, 2002; Ruan et al., 2001; Kim et al., 2002). Evidence continues to accumulate that macromolecular signaling molecules, including proteins and RNAs, can traffic from cell to cell and also are capable of long-distance trafficking through the phloem to be delivered (unloaded) into sink organs (Jorgensen et al., 1998).

The regulation of plasmodesmata permeability also is linked to organ development. Oparka et al. (1999) demonstrated a decrease in the permeability of leaf mesophyll plasmodesmata when leaves underwent the sink/source transition. This transition was accompanied by a change from simple to branched plasmodesmata. Imlau et al. (1999) demonstrated that green fluorescent protein (GFP), when expressed under the control of the companion cell–specific promoter AtSUC2, could be unloaded from the phloem into sink regions of the leaf as well as other sink organs. Tuberization of potato stolons also is associated with a plasmodesmata transition that permits symplasmic unloading of carbon compounds as a result of the creation of plasmodesmata continuity between the phloem and the newly formed storage parenchyma of the tuber (Viola et al., 2001). Furthermore, modification of plasmodesmata permeability by expressing a viral MP leads to a modification in carbon partitioning in tobacco (Olesinski et al., 1995; Hofius et al., 2001) and potato (Olesinski et al., 1996; Almon et al., 1997).

The formation of the symbiotic nodule primordium in legumes provides a model system in which to study cell dedifferentiation and the acquisition of a new identity. Nodule organogenesis is initiated by a molecular dialogue involving the host legume and the soil bacterium Rhizobium via flavonoids excreted by the plant and bacterial lipochitooligosaccharides called Nod factors. The perception of bacterial signals by the root induces a series of morphological and physiological changes that eventually lead to the formation of a new organ, the symbiotic root nodule, in which internalized bacteria convert molecular nitrogen to ammonia (Schultze and Kondorosi, 1998). In Medicago species, the first morphological change that occurs during the symbiotic interaction is the dedifferentiation of several root cell types (pericycle cells and inner cortex cells) in front of a protoxylem pole (Timmers et al., 1999). These differentiated cells are activated in response to bacteria, as seen by cytoskeletal rearrangement (Timmers et al., 1999), and then divide to form the nodule primordium. In this study, we consider the nodule primordium to be composed of pericycle, endodermis, and cortex cells that have dedifferentiated in response to Rhizobium infection. Simultaneously, bacteria penetrate the root tissue and progress toward the primordium via infection threads. The next step is the formation of a meristem at the tip of the growing nodule primordium, where most cell division activity takes place (Timmers et al., 1999). This is followed by the differentiation of nodule cells and invasion by rhizobia, leading to the development of an indeterminate nodule composed of several zones: the persistent meristematic zone at the apex (zone I), the activity of which ensures the continuous growth of the nodule; the invasion zone (zone II), in which cells differentiate and are invaded by rhizobia; and the nitrogen fixation zone (zone III), in which bacteria differentiated into bacteroids fix nitrogen (Hirsch, 1992).

Nodule initiation is controlled by a number of hormonal factors that emanate from the stele (Schultze and Kondorosi, 1998). Moreover, a systemic phenomenon called autoregulation limits the number of successful infections (Caetano-Anollés and Gresshoff, 1991a). Some hypernodulating mutants affected in this autoregulation display a shoot-dependent phenotype in grafting experiments (Delves et al., 1986; Varma Penmetsa et al., 2003), indicating a role for shoot–root communication in nodule organogenesis. Furthermore, nodule initiation depends on carbon supply, because active photosynthesis is required for nodulation (Bauer et al., 1996).

The plasmodesmata-mediated communication processes involved in the dedifferentiation of root pericycle and cortical cells, and the formation of a de novo meristem, are poorly understood. In this study, we provide evidence for the establishment of a plasmodesmata continuum between the phloem and the root cells of the nodule primordium that precedes cell division events in the developing nodule. The creation of this new symplasmic field is associated with increased plasmodesmata density in both preexisting and newly formed cell walls. We also show that GFP (27 kD) is able to move from the phloem into the nodule primordium and over long distances from the shoot, suggesting the existence of a macromolecular translocation stream between the shoot and the nodule primordium. Moreover, the pattern of GFP unloading changes during nodule development. Finally, we demonstrate that transgenic expression in Medicago truncatula of a viral MP, which is known to increase plasmodesmata permeability and able to move from cell to cell in this legume, causes a significant increase in nodule number. These results demonstrate a primary role for plasmodesmata-mediated trafficking in nodule initiation and development.


Phloem Transport and Unloading of Carboxyfluorescein into Nodule Initials

The membrane-impermeant fluorescent marker 5(6)-carboxyfluorescein (CF) was used as a symplasmic tracer by loading its ester form, 5(6)-carboxyfluorescein diacetate (CFDA), into the phloem (Roberts et al., 1997). Cotyledons of Medicago sativa seedlings inoculated with Rhizobium Sm41 were loaded with CFDA, and the plants were allowed to translocate the dye in the light. After 1 to 2 h, the phloem of the root became fluorescent. Longitudinal free-hand sections of the Sm41 spot-inoculated root area were prepared after 4 to 6 h of translocation and observed. CF was unloaded into the root apex (Figure 1A), as expected from previous studies conducted with other plants (Oparka et al., 1994). However, no fluorescence was observed in the uninoculated root cortex, indicating a symplasmic restriction between the phloem and this tissue. By contrast, CF was unloaded from the root phloem into the pericycle and cortical cells that formed the nodule primordium (Figures 1B to 1D). These results suggest that reactivation of nodule initial cells is associated with the establishment of symplasmic continuity between the phloem and specific root cells.

Figure 1.
CF Movement from the Phloem into the Nodule Primordium of M. sativa.

Formation of the Nodule Primordium Is Associated with Changes in the Plasmodesmata Network

The MP of Cucumber mosaic virus (CMV) has been shown to localize specifically to branched plasmodesmata (Blackman et al., 1998). M. truncatula plants expressing a translational fusion between the MP of CMV and GFP (3aMP-GFP plants) displayed strong punctate fluorescence on the walls of epidermal cells in mature leaves (Figure 2A). Although faint GFP fluorescence was present in root cortex cells, very little punctate labeling was observed on their cell walls, suggesting the scarcity of branched plasmodesmata in these cells. By contrast, punctate fluorescence was associated with the cell walls of vascular tissues (Figure 2B). In infected parts of the root, an increase in the punctate GFP signal from vascular tissues was observed underneath the nodule primordium (Figures 2C and 2D) and the initial dividing cortical cells, suggesting that plasmodesmata branching increases in the vascular tissue close to the primordium.

Figure 2.
GFP Fluorescence of 35S-3aMP-GFP M. truncatula Plants.

Further analysis of plasmodesmata networks in the nodule was conducted on M. sativa nodule primordia at 1 and 2 days after inoculation (DAI) using electron microscopy. At 1 DAI, cell activation (dense cytoplasm, fragmentation of the vacuole, central nucleus) was seen in the pericycle, endodermis, and the first layer of inner cortex cells (Figure 3A). Cell division in the pericycle preceded cell division in the endodermis and cortex layer. At 1 DAI, at least one division (anticlinal and periclinal) had occurred in all pericycle cells involved in nodule initiation, and in some cases two or more divisions had occurred. Only certain inner cortical cells had undergone one division at this stage (Figure 3A). At 2 DAI, the primordium was larger and the cytoplasm of activated cells was less dense than at 1 DAI. Activation and division of cells also were visible in the second cell layer of the cortex (Figure 3C). A noteworthy difference between nodule primordia at 1 and 2 DAI was the appearance of amyloplasts at 2 DAI (Figure 3H), representative of the carbon sink activity of the developing organ.

Figure 3.
Electron Microscopy of M. sativa Nodule Primordia.

Initial observation revealed the presence of plasmodesmata between the pericycle cells and the immature sieve elements of the protophloem pole or adjacent sieve elements (Figures 3D to 3G), demonstrating the existence of symplasmic continuity between the phloem and the pericycle of the nodule primordium. Branched plasmodesmata were very scarce in both noninoculated roots and nodule primordial cells, consistent with the previous observations made on transgenic 3aMP-GFP plants.

We defined the frequency of plasmodesmata as the number per unit length of shared cell wall (in micrometers) within the nodule primordium (Roberts et al., 2001). These values were compared with equivalent numbers in the uninfected area on the opposite side of the same root. A large increase in the overall frequency of plasmodesmata in the nodule primordium was found at both 1 and 2 DAI (Figure 4A). Within the nodule primordium, certain cell walls appeared to be recently formed (Figures 3A and 3C, red cell walls). Plasmodesmata frequencies in these cell walls were found to be much higher than the overall frequency in the primordium at both 1 and 2 DAI (Figure 4B). At 1 DAI, it was still possible to distinguish most of the preexisting cell walls from those formed after nodule initiation. To determine whether there was de novo plasmodesmata formation on preexisting cell walls, we measured plasmodesmata frequencies on three periclinal interfaces: the stele/pericycle interface, the pericycle/endodermis interface, and the endodermis/inner cortex interface. These interfaces are composed of preexisting cell walls, and they are indicators of the barriers to directional fluxes from the root phloem into the nodule primordium. As shown in Figure 4C, an increase in the plasmodesmata frequency of each interface compared with the root control was observed, suggesting that, in addition to the large increase observed in new cell walls, there was de novo formation of plasmodesmata on these preexisting cell walls.

Figure 4.
Quantification of Plasmodesmata Frequencies in Nodule Primordia.

GFP Unloading Is Regulated during Nodule Development

As a tracer of macromolecular phloem unloading, we expressed GFP under the control of the companion cell–specific AtSUC2 promoter (Imlau et al., 1999) in M. truncatula. In situ hybridization with a GFP antisense probe on AtSUC2-GFP nodules showed a strong signal in some cells of the phloem of nodulated and nonnodulated roots (presumably corresponding to the companion cells) (Figures 5A, 5B, 5D, and 5F to 5L). Serial sections of an AtSUC2-GFP root showed labeling of the phloem poles, particularly when the sectioning plane crossed a companion cell (Figures 5F to 5L). A strong signal also was detected in the phloem of nodules (Figures 5M and 5O). No signal was detected in nonvascular nodule tissues (Figures 5A, 5B, 5M, and 5O), whereas strong GFP fluorescence was detected in meristematic and invasion zones of nodules at a similar stage in addition to the signal present in vascular tissues (Figures 5C and 5N). This finding shows that GFP fluorescence detected in nonvascular dividing and differentiating nodule tissues was attributable to GFP unloading from the phloem. No signal was detected on the phloem of wild-type nodules hybridized with the same probe (Figure 5E). By contrast, nodules from 3aMP-GFP plants in which the expression of the GFP fusion was driven by a 35S promoter showed extensive labeling in nodule primordia as well as in the meristematic and invasion zones of mature nodules (Figure 5P). This finding confirmed that the GFP RNA was confined to the phloem when expressed from the AtSUC2 promoter in M. truncatula.

Figure 5.
In Situ Localization of the GFP RNA in AtSUC2-GFP Nodules.

In uninoculated roots of AtSUC2-GFP plants, GFP was detected in the phloem but was absent from the cortex (Figures 6A and 6B), consistent with the observation that plasmodesmata-mediated communication between the root phloem and cortex is restricted. As described previously in Arabidopsis (Imlau et al., 1999), GFP fluorescence also was detected in lateral root primordia (data not shown) and apices (Figure 6C). When AtSUC2-GFP plants were inoculated with Rhizobium Sm41, strong GFP fluorescence was detected in the nodule primordium (Figures 6D to 6F). GFP fluorescence was detected in the root cortex as soon as the first cortical cell divisions occurred (Figure 6D). Moreover, cortex cells adjacent to these dividing cells also displayed strong fluorescence (Figures 6D and 6F). This observation suggests that sink activity precedes cell division during the formation of the nodule primordium. GFP fluorescence then was distributed throughout the whole primordium (Figure 6G), as revealed by transverse sectioning. After further development, GFP fluorescence became stronger in the apical part of the developing nodule, corresponding to the meristem (Figure 6H). Thus, most of the sink activity seemed to be localized in the meristematic region of the developing nodule. Differentiation of the new vascular tissues of the nodule also was seen at this stage (Figure 6H). In mature nodules, GFP fluorescence was detected clearly in the vascular tissue and in the meristematic and invasion zones (zones I and II), whereas no fluorescence was seen in the central region corresponding to the nitrogen fixation zone (zone III) (Figure 6I).

Figure 6.
Movement of GFP Expressed in the Phloem under the Control of the AtSUC2 Promoter.

GFP Can Undergo Long-Distance Trafficking from the Shoot to the Nodule Primordium

To analyze macromolecular trafficking from the shoot to the root nodule primordium, we grafted transgenic AtSUC2-GFP scions onto wild-type rootstocks. After 1 week, grafted roots displayed GFP fluorescence in the phloem (Figure 7A), although at lower levels than in transgenic AtSUC2-GFP roots. Nonetheless, GFP was detected clearly in nodule primordia at different developmental stages (Figures 7B to 7E). These results establish the capacity for long-distance transport of GFP from the shoot to the nodule initials, in addition to its demonstrated capacity to traffic short distances from the root phloem into the primordia.

Figure 7.
Long-Distance Movement of GFP in the Nodule Primordium of Wild-Type Rootstocks Grafted on AtSUC2-GFP Scions.

Transgenic Expression of the Tobacco mosaic virus MP Increases Nodule Number

To test whether the modification of plasmodesmata function could influence nodule initiation in M. truncatula, we expressed the MP of Tobacco mosaic virus (TMV), a MP known to increase plasmodesmata permeability (Beachy and Heinlein, 2000), using a constitutive 35S promoter. Transgenic plants expressing TMV-MP were selected based on protein gel blot analysis (Figure 8A). Two weeks after inoculation with Sinorhizobium meliloti Sm41, the TMV-MP transgenic plants (lines TMV 25 and TMV 3) showed significant increase in nodule numbers (P < 0.05, Student's t test) in four independent experiments (for line TMV 25), suggesting that nodule initiation was affected in these plants (Figure 8B). Furthermore, no significant difference in nodulation was observed between R108 and various control lines (AtSUC2-GFP plants or plants transformed with an empty vector; data not shown). Hence, plasmodesmata-mediated trafficking may play a role in the control of root nodule organogenesis. No significant differences were observed in the morphology and distribution of TMV in wild-type nodules (data not shown). The functionality of the TMV MP on M. truncatula plasmodesmata was confirmed by bombardment of a construct that expresses a translational fusion between the TMV MP and GFP on M. truncatula leaves. This fusion protein was able to move from the bombarded cell into surrounding cells at 24 h after bombardment, whereas free GFP stayed confined to the bombarded cell (Figures 8C and 8D). This movement was observed in >80% of the 25 cells observed with the TMV MP–GFP fusion and never in an equivalent number of GFP-expressing cells at this time point.

Figure 8.
Nodulation of TMV-MP–Expressing M. truncatula Transgenic Plants.


Increased Plasmodesmata-Mediated Communication between the Phloem and Root Cortical Cells Occurs during Nodule Initiation

The first steps of nodule organogenesis involve the reactivation and division of pericycle cells, followed by the reactivation of endodermis and cortex cells. Timmers et al. (1999) reported that this cell activation was associated with rearrangements in the microtubule cytoskeleton. At 1 DAI, the primordium was very compact and extended only to the first layer of inner cortical cells. At 2 DAI, divisions extended into the second layer of cortex cells. At this stage, the presence of amyloplasts suggests that the flow of carbohydrates had exceeded the immediate demand.

This study describes the formation of symplasmic (plasmodesmata-mediated) continuity between the phloem and nodule initials, concomitant with the creation of a new strong carbon sink. Symplasmic phloem unloading has been reported in various carbon sink organs, such as lateral root primordia and root apices of Arabidopsis (Oparka et al., 1994, 1995) as well as sink tobacco leaves (Roberts et al., 1997). In addition, tuberization in potato involves a switch from apoplasmic to symplasmic phloem unloading of carbohydrates during development (Viola et al., 2001). Our experiments showed that uninoculated root cortex cells of Medicago species were symplasmically isolated from the root phloem, at least for molecules larger than carboxyfluorescein (376 D). When the root was infected with Rhizobium, the cells involved in the formation of the nodule primordium became symplasmically connected to the root phloem. Not only solutes but also macromolecules such as GFP (27 kD) could be unloaded from the phloem into the nodule primordium. Furthermore, the nodule primordium initials formed a symplasmic “field” that was isolated from the other root cortical cells. As the primordium enlarged, the spreading of this new symplasmic field preceded cell division, suggesting that plasmodesmata-mediated communication with neighboring cells might be necessary for the dedifferentiation of these cells. The accumulation of GFP in nodule primordia can be explained by both a constant flow from the phloem and the relatively high stability of GFP. Moreover, grafting experiments showed that a significant contribution of the GFP fluorescence detected in nodule primordia comes from the aerial part of the plant, indicating that this accumulation has two origins: local as well as distant phloem.

As described by Timmers et al. (1999), after the progression of infection moves toward the primordium, the next step in the development of the nodule is the formation of a meristem that originates from middle cortex–derived primordial cells. This meristem accumulated GFP in AtSUC2-GFP plants, indicating specialization of the sink activity in these cells. In the mature nodule, GFP fluorescence was present in the meristematic and invasion zones, whereas no GFP fluorescence was detected in the central zone, where nitrogen fixation occurs. This could be the result of a reduction in the permeability of plasmodesmata in this zone. However, it has been reported that the oxygen concentration was particularly low in this region, likely because of a combination of factors: the presence of an oxygen barrier in the cortex, the very fast consumption of oxygen by bacteroids, and the abundance of leghemoglobin, a protein with high affinity for oxygen (Witty et al., 1987). This low concentration of oxygen also could explain the absence of GFP fluorescence in this zone, because GFP fluorescence is dependent on the presence of oxygen.

The Creation of a New Sink Involves Rearrangements in the Plasmodesmata Network

Several studies have demonstrated that the root meristem and elongation zones of Arabidopsis are in symplasmic continuity with the phloem (Zhu et al., 1998; Imlau et al., 1999; Oparka et al., 1999). As root cortical cells differentiate, this symplasmic continuity is broken and plasmodesmata numbers decrease in the mature zone of the root (Zhu et al., 1998). During nodulation, differentiated root cells that had become symplasmically isolated regain their communication capacity with the phloem and reenter the cell cycle. This was shown here to be associated with an increase in the plasmodesmata frequency between these cells. The lower plasmodesmata frequency at 2 DAI could be linked to a reduction in the need for carbohydrate influx, which can be correlated with the accumulation of amyloplasts observed only at 2 DAI. Primary plasmodesmata, formed during the formation of a new cell wall (Ehlers and Kollmann, 2001), are not necessarily representative of increased communication between all cells of the primordium.

To evaluate the recreation of a symplasmic field, we analyzed plasmodesmata frequencies within and between the three different cell types of the primordium and with tissues underneath the stele during the first steps of nodule formation (stele/pericycle, pericycle/endodermis, and endodermis/cortex boundaries). At 1 DAI, these cell wall interfaces existed before cortical cell divisions had commenced. The increase in plasmodesmata-mediated communication in these boundaries allowed enhanced radial communication between the mother cells of the primordium. Hence, the establishment of the symplasmic field was attributable to the de novo formation of secondary (postcytokinetic) plasmodesmata on preexisting cell walls, possibly accompanied by an increase in the permeability of preexisting plasmodesmata.

The relative importance of these two mechanisms (frequency versus permeability) in the reestablishment of symplasmic continuity remains unclear. A number of studies have shown that plasmodesmata frequencies in stable meristems are very high, but these plasmodesmata are mainly primary in origin as a result of the very high cell division rate. The translational fusion between a TMV MP and GFP was used to monitor the evolution of plasmodesmata branching during the sink-source transition of tobacco leaves, because there was a strict correlation between the appearance of branched plasmodesmata and punctate labeling. Although it is possible that all branched plasmodesmata are not labeled by 3aMP-GFP in roots, this fusion could be used as a marker of plasmodesmata branching (Roberts et al., 2001). The absence of punctate signal in the root cortex of 3aMP-GFP plants indicated that branched plasmodesmata were scarce in this tissue, in agreement with our electron microscopy studies. By contrast, an increase in 3aMP-GFP labeling, likely corresponding to an increase in plasmodesmata branching, was observed in the root vascular tissue adjacent to the nodule primordium, suggesting that structural modifications of the plasmodesmata network between the phloem and primordium initials may have occurred concomitant with the increase in phloem unloading. For example, this effect could correspond to the formation of new pore-plasmodesmata units linking sieve elements and companion cells (Oparka and Turgeon, 1999). These pore-plasmodesmata units are branched on the companion cell side; therefore, they could be sites of localization for the 3aMP-GFP fusion.

To our knowledge, it is still unknown whether the regulation of symplasmic continuity during development is caused by an increased “gating” of preexisting plasmodesmata or the formation of new ones. During the sink-source transition of leaves, simple plasmodesmata are replaced by branched plasmodesmata. The decrease in symplasmic continuity is the result of both structural modifications and a reduction in the SEL of plasmodesmata, although it remains to be determined whether simple plasmodesmata aggregate to form branched plasmodesmata or disappear and are replaced by the latter (Oparka et al., 1999; Roberts et al., 2001).

The Role of Symplasmic Communication in Nodule Formation

In addition to carbon compounds, a number of regulatory molecules can move freely via plasmodesmata across a symplasmic field, such as phytohormones, proteins, and RNA. The cell-to-cell trafficking of some transcription factors has been shown to be essential for several aspects of plant development (Lucas et al., 1995; Sessions et al., 2000). For example, the differentiation of the endodermis layer in the Arabidopsis root depends on cell-to-cell trafficking of the SHR protein from the stele to the endodermis (Nakajima et al., 2001). During their development, plants can be seen as dynamic mosaics of symplasmic domains in which regulatory molecules can diffuse to determine a common cell fate (Pfluger and Zambryski, 2001). Temporal and spatial regulation of these symplasmic domains has been described in various developmental processes, such as the floral transition in Arabidopsis (Gisel et al., 1999, 2002), embryogenesis in Arabidopsis (Kim et al., 2002), and the elongation of single-celled cotton fibers (Ruan et al., 2001). In these organs, symplasmic isolation, or the establishment of new symplasmic connections at certain steps of development, is believed to be required to coordinate the differentiation process.

The finding that GFP can move from the shoot to the nodule through the phloem indicates that macromolecules and a plethora of small molecules can be transported by this route. Mathesius et al. (1998) reported that inoculation of the root with rhizobia induced a reorientation of the acropetal flow of auxin toward the point of inoculation. Furthermore, a feedback regulation called autoregulation that acts at the stage of nodule initiation to limit nodule number systemically has been described (Caetano-Anollés and Bauer, 1988; Caetano-Anollés and Gresshoff, 1991b). The phenotype of several supernodulating mutants affected in this regulation is dependent on the shoot genotype, as shown by grafting experiments (Delves et al., 1986; Krusell et al., 2002), which suggests that a shoot-derived mobile signal is responsible for autoregulation. Hence, the regulation of this phloem-connected symplasmic field might be critical for nodule organogenesis.

The MP of TMV has been shown in various systems to increase the SEL of plasmodesmata (Beachy and Heinlein, 2000), and our experiments provide evidence that it also is able to gate plasmodesmata in M. truncatula leaves to induce its own cell-to-cell movement. Plasmodesmata-mediated macromolecular trafficking is regulated by two different mechanisms: one is nonspecific, involving the regulation of the SEL, and the second is specific and could involve the recognition of the mobile protein by a putative plasmodesmata receptor and unfolding of the mobile molecule (Haywood et al., 2002). Our observations suggest that MP-mediated alteration of plasmodesmata permeability affected nodule initiation, because nodulation has been shown to be controlled during the onset of the cortical cell divisions (Caetano-Anollés and Gresshoff, 1991b). An explanation of this phenotype could be that nodule initials in these plants displayed increased cell-to-cell communication before their activation as a result of the higher permeability of plasmodesmata. Thus, the creation of the symplasmic field would be facilitated. Alternatively, this phenotype could be attributable to a SEL-unrelated interaction of the TMV MP with plasmodesmata that would alter macromolecular trafficking and thus modify the unloading of signal molecules into nodule initials. For example, the TMV MP could compete on plasmodesmata receptors with endogenous regulatory proteins, thereby preventing their cell-to-cell or long-distance movement and altering the crosstalk between source organs and nodule initials. In both cases, our experiments show that interacting with plasmodesmata-mediated cell-to-cell communication alters nodulation.

Our work provides a new model for analysis of the developmental regulation of symplasmic communication into a de novo–created symplasmic field and presents evidence for its role in nodule initiation. This highlights the critical role of intercellular trafficking (plasmodesmata function and cell-to-cell communication) in developmental processes.


Strains and Growth Conditions

Plants of Medicago truncatula and Medicago sativa were surface-sterilized, grown, and nodulated with Sinorhizobium meliloti strain Sm41 as described (Charon et al., 1999). For nodulation tests, plants were grown in sand for 1 week on nitrogen-containing medium, for another week without nitrogen, and inoculated with Rhizobium. Nodules were counted 2 weeks later. For each experiment, 10 to 16 plants of each line were used.

Early nodulation events for electron microscopy and carboxyfluorescein diacetate (CFDA) loading studies were monitored in M. sativa plants grown on Gibson medium for 3 days and “spot-inoculated” on the root hair emergence zone with a Sm41 culture resuspended in 0.4% agarose and 10 mM MgSO4 (OD600 = 0.4) at 40°C. This allowed us to predict the precise location where the nodule primordium would appear.

Generation of M. truncatula Transgenic Plants

M. truncatula 108-R4 plants were transformed according to Trihn et al. (1998) with the plasmids described below. AtSUC2 promoter–GFP plants were transformed with the pEPS1 plasmid (Imlau et al., 1999). Forty-four transgenic lines were selected based on the pattern and intensity of GFP fluorescence in roots. Four independent transgenic lines were analyzed further. 35S-3aMP-GFP plants were transformed with the plasmid pBP-CMV-MP-sGFP (Shalitin et al., 2002). Transgene expression was tested by RNA gel blot analysis on 12 transgenic lines (data not shown), and 3 of these independent transgenic lines were selected and analyzed further. The 35S-TMV-MP plants were transformed with the plasmid pBP-35S-TMV-MP (detailed characterization of these plants will be presented elsewhere). Eight independent transgenic lines were confirmed positive for the TMV-MP using protein gel blot analyses, and two of these lines were selected for nodulation analyses. Expression of the TMV-MP protein was analyzed on leaf extracts by protein gel blot analysis (Towbin et al., 1979) with anti-TMV-MP antibody (Almon et al., 1997).

Microscopic Detection of GFP and Carboxyfluorescein Fluorescence

Plant tissues were observed after either free-hand longitudinal sectioning of fresh material or fixation (3% paraformaldehyde) and sectioning (200 μm thick) with a vibrotome (Bio-Rad). GFP and carboxyfluorescein (CF) fluorescence was observed on a Leica TCS SP2 confocal laser scanning microscope (Wetzlar, Germany) after excitation at 488 nm. GFP and CF emission was detected between 500 and 544 nm (green channel). Remaining autofluorescence was detected between 560 and 665 nm (red channel). Overlay of the two channels allowed us to distinguish between autofluorescence (present in both channels) and GFP or CF fluorescence.

Grafting of M. truncatula Plants

After growth on agar plates for 3 days, hypocotyls sectioned at the end of the green chlorophyll zone and root scions (to be grafted) were inserted into both ends of a polyethylene capillary tube (1 cm long, 0.9 mm i.d.) and brought into close contact. Grafted plantlets then were transferred to agar plates, and the two ends of the capillary tube were sealed with 1.5% low-melting-point agarose (see the detailed protocol in the manual of the 2002 European Molecular Biology Organization course on M. truncatula at http://www.isv.cnrs-gif.fr/embo01/manuels/index.html). When grafting was successful, root growth was restored within 1 week, and grafted plants could be nodulated after another week.

CFDA Loading of M. sativa Plants

M. sativa plants were loaded with CFDA by tip-wounding the lower side of a cotyledon at 2 days after spot inoculation with Sm41 and injecting a 60-μg/mL CFDA solution with a needleless 1-mL syringe into the wounded cotyledon. Infiltrated plants then were allowed to translocate CF in the light for 4 to 8 h before imaging.

Bombardment of M. truncatula Leaves

The pJD330-GFP and the pJD330-TMV-MP:GFP plasmids was created by standard PCR protocol for each gene. The primer sequences for TMV-MP were 5′-GGATCCAAACGAATCCGATTCGG-3′ (primer 1) and 5′-GTCGACATATGGCTCTATGTTGTTAAA-3′ (primer 2); the primer sequences for GFP were 5′-GGATCCGGACCTCCTCCTGGAATGGTGAGCAAGGGCGA-3′ (primer 3) and 5′-GAGCTCTTACTTGTACAGCTCGTCC-3′ (primer 4). Primer 3 had a 5–amino acid linker sequence (Gly-Pro-Pro-Pro-Gly) to connect the two proteins that allows their free three-dimensional folding. The amplified TMV-MP gene was digested at the 5′ end with SalI and at the 3′ end with BamHI. The amplified GFP gene was digested at the 5′ end with BamHI and at the 3′ end with SacI. The two genes were fused and ligated into pBluescript SK+ (Stratagene). The fused gene was cloned as a Klenow/SacI fragment into pJD330.

The two pJD330 constructs were used for biolistic bombardment into young fully expanded leaves of M. truncatula with a Bio-Rad PDS1000/He particle gun using 1-μm gold particles and 450 p.s.i. rupture disks. They were left in a humid chamber overnight to allow GFP movement to occur.

Electron Microscopy of M. sativa Nodule Primordia

Inoculated regions of M. sativa roots spot-inoculated with Sm41 were fixed in 5% (w/v) glutaraldehyde and embedded in araldite at 1 or 2 days after inoculation (DAI) as described by Oparka et al. (1999). Although nodule primordia were invisible before fixation, poststaining with osmium tetroxide allowed us to localize them as a dark spot on fixed material before sectioning. A series of ultrathin transverse sections was taken every 5 μm to traverse the whole nodule primordium. Sections were stained with lead citrate and uranyl acetate prepared in isobutanol-saturated water (Roberts, 2002) and viewed with a JEOL 1200 EX transmission electron microscope at an accelerating voltage of 80 kV.

Counting of Plasmodesmata

In each previously described series, only one section was analyzed to avoid counting the same group of plasmodesmata on serial sections several times. An area equivalent to the nodule primordium in terms of cell types was counted on the uninfected half of the same root as a control. Only plasmodesmata extending at least to the middle lamella were counted. Branched plasmodesmata were counted as a single entity. The length of shared cell walls was measured from low-magnification pictures with an Intuos 2 Graphics Tablet (WACOM, Waukesha, WI).

Detailed analyses of the data are presented for 1- and 2-DAI primordia, although counts were determined on three primordia of both types to confirm the described pattern (data not shown). A total of 11 sections of nodule primordium and 5 sections of control root were counted for the depicted 1-DAI primordium. For the 2-DAI primordium, five sections of primordium and control roots were counted.

In Situ Hybridization

A 466-bp fragment of the GFP gene flanked by the T7 and T3 promoters was amplified on the pEPS1 plasmid with the primers GFPT3 (5′-AATTAACCCTCACTAAAGGGAGACAAGGGCGAGGAGCTGT-3′) and GFPT7 (5′-TAATACGACTCACTATAGGGAGATCTGCTTGTCGGCCATGATATAG-3′). The radioactive 35S-UTP GFP antisense probe was synthesized by in vitro transcription with the T7 polymerase. Tissue fixation, embedding in paraplast, hybridization, and development were performed as described (de Almeida Engler et al., 2001). Slides were developed after 1 week of exposure, and pictures were taken on a Reichert Polyvar microscope (New York, NY) with a DXM 1200 Nikon digital camera.

Upon request, materials integral to the findings presented in this publication will be made available in a timely manner to all investigators on similar terms for noncommercial research purposes. To obtain materials, please contact M. Crespi, rf.fig-srnc.vsi@ipserc.nitram.


We thank the cell biology imaging platform of the Institut Fédératif de Recherche 87 La Plante et son Environnement and the Conseil Général de l'Essonne, Nathalie Mansion for help with image treatments, Dror Shalitin for the 3aMP-GFP construct, and Doug Cook for providing us with the sunn seeds. This work was supported by the Association Franco-Israelienne pour la Recherche Scientifique et Technologique and the Centre National de la Recherche Scientifique–British Royal Society convention (Project 10895). A.C. and L.B. were supported by a fellowship from the Ministère de la Recherche et de l'Enseignement Supérieur.


Article, publication date, and citation information can be found at www.plantcell.org/cgi/doi/10.1105/tpc.017020.


  • Almon, E., Horowitz, M., Wang, H.L., Lucas, W.J., Zamski, E., and Wolf, S. (1997). Phloem-specific expression of the tobacco mosaic virus movement protein alters carbon metabolism and partitioning in transgenic potato plants. Plant Physiol. 115, 1599–1607. [PMC free article] [PubMed]
  • Bauer, P., Ratet, P., Crespi, M., Schultze, M., and Kondorosi, A. (1996). Nod factors and cytokinins induce similar cortical cell division, amyloplast deposition and Msenod12A expression patterns in alfalfa roots. Plant J. 10, 91–105.
  • Beachy, R.N., and Heinlein, M. (2000). Role of P30 in replication and spread of TMV. Traffic 1, 540–544. [PubMed]
  • Blackman, L.M., Boevink, P., Cruz, S.S., Palukaitis, P., and Oparka, K.J. (1998). The movement protein of cucumber mosaic virus traffics into sieve elements in minor veins of Nicotiana clevelandii. Plant Cell 10, 525–538. [PMC free article] [PubMed]
  • Caetano-Anollés, G., and Bauer, W.D. (1988). Feedback regulation of nodule formation in alfalfa. Planta 174, 385–395. [PubMed]
  • Caetano-Anollés, G., and Gresshoff, P.M. (1991. a). Plant genetic control of nodulation. Annu. Rev. Microbiol. 45, 345–382. [PubMed]
  • Caetano-Anollés, G., and Gresshoff, P.M. (1991. b). Alfalfa controls nodulation during the onset of Rhizobium-induced cortical cell division. Plant Physiol. 95, 336–373. [PMC free article] [PubMed]
  • Charon, C., Sousa, C., Crespi, M., and Kondorosi, A. (1999). Alteration of enod40 expression modifies Medicago truncatula root nodule development induced by Sinorhizobium meliloti. Plant Cell 11, 1953–1966. [PMC free article] [PubMed]
  • de Almeida Engler, J., De Groodt, R., Van Montagu, M., and Engler, G. (2001). In situ hybridization to mRNA of Arabidopsis tissue sections. Methods 23, 325–334. [PubMed]
  • Delves, A.C., Mathews, A., Day, D.A., Carter, A.S., and Gresshoff, P.M. (1986). Regulation of the soybean-Rhizobium synthesis by shoot and root factors. Plant Physiol. 82, 588–590. [PMC free article] [PubMed]
  • Ehlers, K., and Kollmann, R. (2001). Primary and secondary plasmodesmata: Structure, origin, and functioning. Protoplasma 216, 1–30. [PubMed]
  • Erwee, M.G., and Goodwin, P.B. (1985). Symplast domains in extrastelar tissues of Egeria densa Planch. Planta 163, 9–19. [PubMed]
  • Gisel, A., Barella, S., Hempel, F.D., and Zambryski, P.C. (1999). Temporal and spatial regulation of symplastic trafficking during development in Arabidopsis thaliana apices. Development 126, 1879–1889. [PubMed]
  • Gisel, A., Hempel, F.D., Barella, S., and Zambryski, P. (2002). Leaf-to-shoot apex movement of symplastic tracer is restricted coincident with flowering in Arabidopsis. Proc. Natl. Acad. Sci. USA 99, 1713–1717. [PMC free article] [PubMed]
  • Haywood, V., Kragler, F., and Lucas, W.J. (2002). Plasmodesmata: Pathways for protein and ribonucleoprotein signaling. Plant Cell 14 (suppl), S303.–S325. [PMC free article] [PubMed]
  • Hirsch, A.M. (1992). Developmental biology of legume nodulation. New Phytol. 122, 211–237.
  • Hofius, D., Herbers, K., Melzer, M., Omid, A., Tacke, E., Wolf, S., and Sonnewald, U. (2001). Evidence for expression level-dependent modulation of carbohydrate status and viral resistance by the potato leafroll virus movement protein in transgenic tobacco plants. Plant J. 28, 529–543. [PubMed]
  • Imlau, A., Truernit, E., and Sauer, N. (1999). Cell-to-cell and long-distance trafficking of the green fluorescent protein in the phloem and symplastic unloading of the protein into sink tissues. Plant Cell 11, 309–322. [PMC free article] [PubMed]
  • Jorgensen, R.A., Atkinson, R.G., Forster, R.L., and Lucas, W.J. (1998). An RNA-based information superhighway in plants. Science 279, 1486–1487. [PubMed]
  • Kim, I., Hempel, F.D., Sha, K., Pfluger, J., and Zambryski, P.C. (2002). Identification of a developmental transition in plasmodesmatal function during embryogenesis in Arabidopsis thaliana. Development 129, 1261–1272. [PubMed]
  • Krusell, L., et al. (2002). Shoot control of root development and nodulation is mediated by a receptor-like kinase. Nature 420, 422–426. [PubMed]
  • Lazarowitz, S.G. (1999). Probing plant cell structure and function with viral movement proteins. Curr. Opin. Plant Biol. 2, 332–338. [PubMed]
  • Lucas, W.J., Bouche-Pillon, S., Jackson, D.P., Nguyen, L., Baker, L., Ding, B., and Hake, S. (1995). Selective trafficking of KNOTTED1 homeodomain protein and its mRNA through plasmodesmata. Science 270, 1980–1983. [PubMed]
  • Mathesius, U., Schlaman, H.R.M., Spaink, H.P., Sauter, C., Rolfe, B.G., and Djordjevic, M.A. (1998). Auxin transport inhibition precedes root nodule formation in white clover roots and is regulated by flavonoids and derivatives of chitin oligosaccharides. Plant J. 14, 23–34. [PubMed]
  • Nakajima, K., Sena, G., Nawy, T., and Benfey, P.N. (2001). Intercellular movement of the putative transcription factor SHR in root patterning. Nature 413, 307–311. [PubMed]
  • Olesinski, A.A., Almon, E., Navot, N., Perl, A., Galun, E., Lucas, W.J., and Wolf, S. (1996). Tissue-specific expression of the tobacco mosaic virus movement protein in transgenic potato plants alters plasmodesmal function and carbohydrate partitioning. Plant Physiol. 111, 541–550. [PMC free article] [PubMed]
  • Olesinski, A.A., Lucas, W.J., Galun, E., and Wolf, S. (1995). Pleiotropic effects of TMV-MP on carbon metabolism and export in transgenic tobacco plants. Planta 197, 118–126.
  • Oparka, K.J., Duckett, C.M., Prior, D.A.M., and Fisher, D.B. (1994). Real-time imaging of phloem unloading in the root tip of Arabidopsis. Plant J. 6, 759–766.
  • Oparka, K.J., Prior, D.A.M., and Wright, K.M. (1995). Symplastic communication between primary and developing lateral roots of Arabidopsis thaliana. J. Exp. Bot. 46, 187–197.
  • Oparka, K.J., Roberts, A.G., Boevink, P., Santa Cruz, S., Roberts, I., Pradel, K.S., Imlau, A., Kotlizky, G., Sauer, N., and Epel, B. (1999). Simple, but not branched, plasmodesmata allow the nonspecific trafficking of proteins in developing tobacco leaves. Cell 97, 743–754. [PubMed]
  • Oparka, K.J., and Turgeon, R. (1999). Sieve elements and companion cells: Traffic control centers of the phloem. Plant Cell 11, 739–750. [PMC free article] [PubMed]
  • Pfluger, J., and Zambryski, P.C. (2001). Cell growth: The power of symplastic isolation. Curr. Biol. 11, R436–R439. [PubMed]
  • Rinne, P.L., and Van der Schoot, C. (1998). Symplasmic fields in the tunica of the shoot apical meristem coordinate morphogenetic events. Development 125, 1477–1485. [PubMed]
  • Roberts, A.G., Cruz, S.S., Roberts, I.M., Prior, D., Turgeon, R., and Oparka, K.J. (1997). Phloem unloading in sink leaves of Nicotiana benthamiana: Comparison of a fluorescent solute with a fluorescent virus. Plant Cell 9, 1381–1396. [PMC free article] [PubMed]
  • Roberts, I.M. (2002). Iso-butanol saturated water: A simple procedure for increasing staining intensity of resin sections for light and electron microscopy. J. Microsc. 207, 97–107. [PubMed]
  • Roberts, I.M., Boevink, P., Roberts, A.G., Sauer, N., Reichel, C., and Oparka, K.J. (2001). Dynamic changes in the frequency and architecture of plasmodesmata during the sink-source transition in tobacco leaves. Protoplasma 218, 31–44. [PubMed]
  • Ruan, Y., Llewellyn, D., and Furbank, R. (2001). The control of single-celled cotton fiber elongation by developmentally reversible gating of plasmodesmata and coordinated expression of sucrose and K+ transporters and expansin. Plant Cell 13, 47–60. [PMC free article] [PubMed]
  • Schultze, M., and Kondorosi, A. (1998). Regulation of symbiotic root nodule development. Annu. Rev. Genet. 32, 33–57. [PubMed]
  • Sessions, A., Yanofsky, M.F., and Weigel, D. (2000). Cell-cell signaling and movement by the floral transcription factors LEAFY and APETALA1. Science 289, 779–782. [PubMed]
  • Shalitin, D., Wang, Y., Omid, A., Gal-On, A., and Wolf, S. (2002). Cucumber mosaic virus movement protein affects sugar metabolism and transport in tobacco and melon plants. Plant Cell Environ. 25, 989–998.
  • Timmers, A.C., Auriac, M.C., and Truchet, G. (1999). Refined analysis of early symbiotic steps of the Rhizobium-Medicago interaction in relationship with microtubular cytoskeleton rearrangements. Development 126, 3617–3628. [PubMed]
  • Towbin, H., Staehlin, T., and Gordon, J. (1979). Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 4350–4354. [PMC free article] [PubMed]
  • Trihn, T.H., Ratet, P., Kondorosi, E., Durand, P., Kamaté, K., Bauer, P., and Kondorosi, A. (1998). Rapid and efficient transformation of diploid Medicago truncatula and Medicago sativa ssp. falcata lines improved in somatic embryogenesis. Plant Cell Rep. 17, 345–355.
  • Varma Penmetsa, R., Frugoli, J.A., Smith, L., Long, S.R., and Cook, D.R. (2003). Dual genetic pathways controlling nodule number in Medicago truncatula. Plant Physiol. 131, 998–1008. [PMC free article] [PubMed]
  • Viola, R., Roberts, A.G., Haupt, S., Gazzani, S., Hancock, R.D., Marmiroli, N., Machray, G.C., and Oparka, K.J. (2001). Tuberization in potato involves a switch from apoplastic to symplastic phloem unloading. Plant Cell 13, 385–398. [PMC free article] [PubMed]
  • Witty, J.F., Skot, L., and Revsbech, N.P. (1987). Direct evidence for changes in the resistance of legume root nodules in O2 diffusion. J. Exp. Bot. 38, 1129–1140.
  • Wolf, S., Deom, C., Beachy, R., and Lucas, W. (1989). Movement protein of tobacco mosaic virus modifies plasmodesmatal size exclusion limit. Science 246, 377–379. [PubMed]
  • Xoconostle-Cazares, B., Xiang, Y., Ruiz-Medrano, R., Wang, H.L., Monzer, J., Yoo, B.C., McFarland, K.C., Franceschi, V.R., and Lucas, W.J. (1999). Plant paralog to viral movement protein that potentiates transport of mRNA into the phloem. Science 283, 94–98. [PubMed]
  • Zhu, T., Lucas, W.J., and Rost, T.L. (1998). Directional cell-to-cell communication in the Arabidopsis root apical meristem. I. An ultrastructural and functional analysis. Protoplasma 203, 35–47.

Articles from The Plant Cell are provided here courtesy of American Society of Plant Biologists
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...