A Calcium-Regulated Gatekeeper in Phloem Sieve Tubes

The phloem is a highly specialized long distance transport tissue in vascular plants. Despite detailed knowledge of the development and ultrastructure of phloem tissue in a wide variety of plant species, surprisingly little is known about the mechanisms and regulation of phloem transport. The main function of the phloem is transport of the immediate products of photosynthesis (e.g., sugars) from “source” tissue (actively photosynthesizing leaves) to “sink” tissue (immature leaves, growing root tips, and developing flowers, fruit, and seed). It is sometimes overlooked that a wide variety of other material is also transported through the phloem, including proteins, amino acids, solutes, viruses, and various signaling molecules.

PHLOEM STRUCTURE AND FUNCTION

All students of plant physiology are familiar with the mass flow concept of phloem transport. According to this concept, the transport of materials through phloem sieve tubes is passive, nonselective, and driven entirely by pressure gradients that are maintained by active loading of photosynthate in source tissue and unloading of materials in sink tissue.

Phloem tissue is well designed for long distance transport. The flow of materials takes place through sieve tubes, which are made up of long, tube-like, enucleate sieve elements arranged end to end and connected by specialized end walls known as sieve plates. The sieve plates contain large (up to 1 to 2 μm) pores that allow for passage of materials between sieve elements. Typically, each sieve element in angiosperms is accompanied by one or more companion cells, which interact intimately with the sieve element and play a crucial role in regulating phloem loading and unloading and in the turnover of sieve element proteins and other components (reviewed by Oparka and Turgeon, 1999). Sieve elements and companion cells are derived from unequal longitudinal division of a single “fusiform mother cell.” The cell that becomes the sieve element undergoes a highly regulated partial autolysis, not unlike that of programmed cell death, in which the central vacuole breaks down and the nucleus, ribosomes, cytoskeleton, and Golgi bodies are degraded. The result is a large, nearly empty cell that is well suited for conductance of a wide range of molecules. However, specific components and organelles are retained that may have functions in transport and/or cell–cell interactions within the sieve tubes (van Bel and Knoblauch, 2000).

P-PROTEINS AND PHLOEM TRANSPORT

Mature sieve elements retain a modified endoplasmic reticulum (ER), mitochondria, numerous types of proteins, and specialized sieve element plastids. Most of these components are arranged along the lateral walls of the sieve element and collectively constitute a system known as the parietal layer (Figure 1). A wide variety of sieve element proteins and plastids have been observed that can be distinguished by genus or family.

Figure 1.

Schematic Representation of the Histology of the Sieve Element/Companion Cell-Complex in Vicia faba.

Four sieve elements (SE) connected by sieve plates (SP) are shown with their adjacent companion cells (CC). Only the companion cells contain nuclei (N), vacuoles (V), and chloroplasts (C); mitochondria (M) and endoplasmic reticulum (ER) are also indicated. Typical features of the sieve elements in this species are the sieve element plastids (Pl), the parietal P-proteins (PP), and the P-protein crystalloids (PC). One crystalloid is shown in the dense state, in which it permits the sieve element contents to flow past it. The other is depicted in the dispersed state (DPC), in which it blocks the tube. The direction of flow is downwards, as indicated.

Behnke (1991a)(1991b) presented electron micrographs and descriptions of plastids and crystalloid proteins from the sieve elements of many different species. Behnke (1991a) distinguished two types of sieve element plastids: S-type plastids, which contain only starch inclusions, and P-type plastids, which contain mainly protein inclusions. Behnke (1991b) referred to “nondispersive” versus “dispersive” protein bodies, also called P-proteins or structural sieve element proteins. These terms arose because of changes that are observed during sieve element ontogeny. Early in the maturation process, many protein bodies are often observed, some of which disperse as the sieve element matures and some that remain unchanged in mature sieve elements.

There are many shapes of nondispersive fibrous and crystalloid protein bodies, also called P-proteins, which may be quite large and are often observed in the lumen of sieve elements (Figure 1). Electron micrographic images sometimes show masses of fibrous or amorphous protein located directly in front of or within the pores of sieve plates and that appear to block transport through the pores. To date, the large crystalloid P-protein bodies that occur exclusively in Fabaceae were classified as being nondispersive (Behnke, 1991b).

It has long been thought that some sieve element proteins may function to block the pores of injured sieve tubes to prevent the loss of assimilate. Sieve elements appear to be extremely sensitive to injury, making it difficult to isolate and fix tissues for observation without wounding them, possibly leading to an overabundance of micrographic images with blocked pores.

Some researchers have considered the possibility that sieve element proteins might regularly block mass flow through sieve tubes; thus, an additional transport mechanism might be required to assist materials through sieve tube pores. One possibility put forth was electro-osmosis (see Spanner, 1979) involving either a potassium ion or a proton pump connected to the sieve plate to maintain sufficient pressure to allow the flow of assimilates across the plate. Other theories invoked an active role for sieve element proteins. Evert (1982) reviewed various theories for phloem transport and concluded that the pressure flow mechanism remained the most likely; he also stated that it was difficult to conceive an active role for sieve element proteins, or P-proteins, in phloem transport. For one thing, the amount and structure of sieve element proteins are highly variable among different plant families. Many monocots and some dicots appear to lack crystalloid P-proteins entirely. Also, it has been shown that some “nondispersive” crystalloid P-protein can disperse and move very rapidly to plug sieve pores after injury. Evert concluded that the most likely function of P-protein is to seal the sieve plate pores of injured sieve elements as a rapid first line of defense against the loss of assimilates.

It is possible to obtain electron micrographs from well-preserved and largely uninjured sieve tubes using gentle tissue preparation techniques. Evert (1982) showed an electron micrograph of Cucurbita maxima sieve tubes with completely unoccluded pores. Ehlers et al. (2000) presented a detailed analysis of sieve elements in Vicia faba (broad bean) and Lycopersicon esculentum (tomato). Well-preserved sieve elements clearly showed open, unoccluded sieve pores. Deposits of stacked ER cisternae, plastids, and proteins were observed along the parietal walls and along the walls of sieve plates (particularly in V. faba), but gaps that were free of materials were maintained in front of the sieve pores. These authors also showed that, after injury, the sieve plate pores became occluded by dispersed P-proteins, but the sieve element organelles remained intact and in place along the parietal walls. Numerous clamp-like structures were observed that appeared to anchor the plastids, mitochondria, and ER cisternae in place along the plasma membrane of the parietal wall. The clamps appear to be specific to organelles of the sieve elements and likely function to prevent the components of sieve elements from being carried along through the sieve tubes by the mass flow turbulence.

P-PROTEIN IS AN ACTIVE PARTICIPANT IN THE REGULATION OF PHLOEM TRANSPORT

Knoblauch and van Bel (1998) used confocal laser scanning microscopy to visualize fluorescent dyes moving in sieve tubes of V. faba, which provided definitive evidence of unimpeded mass flow in intact plants. This study also reported that P-type plastids in V. faba actually exploded upon injury of sieve elements, releasing their protein contents, which, together with the dispersed crystalloid protein, rapidly occluded the sieve plate pores. In this issue of The Plant Cell, Knoblauch et al. (pages 1221–1230) extend their previous work with confocal microscopy to show that crystalloid P-proteins of V. faba rapidly disperse and occlude sieve plate pores after injury or osmotic shock. Furthermore, they make the striking observation that this process is rapidly reversible and controlled by calcium fluxes. It is shown that protein from the large crystalloid protein bodies in V. faba sieve elements dispersed to plug the sieve plate pores after injury from micropipette injection or osmotic shock induced by various osmolytes. Addition of the chelating agent EDTA completely prevented crystalloid P-protein dispersal, and repeated exchanges of Ca²⁺- and EDTA-containing media induced the alternate dispersal and re-assembly, respectively, of crystalloid P-proteins in injured sieve tubes. This work represents an important advance in our knowledge of phloem transport.

Data from the current study, together with that of Knoblauch and van Bel (1998), suggest that P-protein originating from all sources within V. faba sieve elements (e.g., P-plastids, parietal proteins, and larger crystalloid P-proteins) takes part in the occlusion of sieve plate pores after injury to cells. Repeated observations now suggest that crystalloid proteins in V. faba are much more sensitive to perturbation than are the P-type plastids. Unlike the reversibility of crystalloid protein dispersal, explosion and dispersal of the P-type plastids appears to be irreversible and presages the death of a sieve element (M. Knoblauch and A.J.E. van Bel, personal communication). Dispersal of parietal proteins after injury has also been observed and was found to be more sensitive to perturbation than was the explosion of P-type plastids (Knoblauch and van Bel, 1998).

In addition to the “structural” sieve element proteins discussed here, a wide variety of soluble proteins have been reported from phloem exudates of many plant species. Hayashi et al. (2000) reported that more than 100 polypeptides could be detected in phloem exudates from a variety of species, including wheat and rice. Unlike the protein composition of other plant tissues, low-molecular-weight proteins appeared to be the dominant proteins in the phloem. In addition, thioredoxin h, glutaredoxin, and glutathione reductase have been found in reasonably high concentrations in phloem exudates, suggesting a redox mechanism for the regulation of sieve tube function. Interestingly, some of the principal sieve element proteins from Oryza sativa, C. maxima, and Ricinus communis can interact with plasmodesmata to increase their size exclusion limit. Two C. maxima proteins, CmPP16 and CmPP36, also appear to mediate the transport of RNAs (Xoconostle-Cázares et al., 1999, 2000). It is likely that the primary function of many of these proteins is the regulation of macromolecular trafficking via the unilaterally branched plasmodesmata located at the junction of sieve elements and companion cells (van Bel and Knoblauch, 2000). As shown by Knoblauch et al. in the present work, other phloem-specific proteins appear to function in the regulation of conductance through the sieve plate pores.

A GENERAL PHENOMENON?

Large crystalloid P-proteins are a particular characteristic of the Fabaceae (legumes). Knoblauch et al. examined sieve elements of Urtica dioica (Urticaceae) and Rubus fruticosus (Rosaceae) in response to wounding and found that the P-protein bodies present in these species failed to disperse and occlude sieve pores, even when severely damaged in the presence of free Ca²⁺. P-type plastids also appear to have somewhat limited distribution. Proteinaceous P-type plastids were observed in just 64 of 382 dicot families studied by Behnke (1991a); the majority of families contained starch-filled S-type plastids.

Calcium is an important component of many signal transduction pathways, and calcium regulation has been implicated in phloem function. In the current study, Knoblauch et al. show that an influx of calcium into legume sieve elements stimulates the rapid and reversible dispersal of crystalloid P-protein to occlude sieve plate pores. It will be important to determine if this phenomenon is limited strictly to the Fabaceae or if other families have similar mechanisms, perhaps involving the parietal and/or other P-proteins or plastids. The concentration of free calcium in sieve tubes of R. communis has been found to be significantly higher than that in surrounding tissue (Brauer et al., 1998), and calcium-dependent protein kinases have been detected in rice phloem sap (Nakamura et al., 1993). Volk and Franceschi (2000) showed evidence of a calcium channel in the sieve element plasma membrane of tobacco and of the aquatic plant Pistia stratiotes using immunolabeling with antibodies to a calcium channel protein. These authors proposed that calcium channels become activated during wounding or pathogen attack, facilitating calcium influx into phloem tissues. McEuen et al. (1981) detected a calcium binding protein distinct from calmodulin in phloem exudates of C. maxima and speculated that it might be associated with P-protein function. Do any of these species have a calcium-regulated reversible mechanism for sealing sieve elements? The grasses may represent an interesting case for the investigation of sieve tube sealing after injury; they appear to lack structural sieve element proteins (i.e., crystalloid or fibrous P-proteins and parietal proteins), and sealing of injured sieve elements might be achieved only by the contents of exploded plastids and/or the slower process of callose formation.