Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

I. Introduction

The long-distance transport, or translocation, of assimilates from source tissues to sink tissues occurs in the phloem of the vascular bundles. In source leaves, sucrose is synthesized in the cytoplasm of mesophyll cells as the major product of photosynthesis (see Chapter 27). For translocation, sucrose moves into the phloem sieve element (phloem loading), and travels through phloem conduits to other parts of the plant (long-distance transport). In sink tissues such as shoot apices, roots and storage tissues, sucrose moves out of the phloem (phloem unloading) and provides the carbon source for development, growth and storage (summarized in Fig. 28.1).

Fig. 28.1.
figure 1

Long-distance transport (Translocation) of sucrose from source leaves to sink tissues. (a) Collection phloem (Loading). Sucrose is synthesized in the cytosol of mesophyll cell (MC) by photosynthesis. For temporal storage, sucrose is imported into and exported from vacuole (V), via unidentified sucrose porter(s). For translocation, sucrose diffuses symplasmically to the bundle sheath cell (BSC) and the phloem parenchyma (PP) through plasmodesmata. Both symplasmic and apoplasmic routes exit to load the sieve element/companion cell (SE/CC) complex. In the latter case, sucrose is released to phloem apoplasm by unascertained mechanisms, and is then taken up actively into the SE/CC complex via a sucrose/H+ symporter which localizes to the SE or the CC, depending on plant species (see Kühn, 2003; Lalonde et al., 2004). (b) Transport phloem. Sucrose is translocated to sink tissues through the sieve tube (ST) continuum. Sucrose/H+ symporters localize to the SE/CC complex of transport phloem and play a role in retrieving sucrose leaked to the apoplasm of phloem conduits. (c) Release phloem (Unloading). Sucrose moves out of the phloem in order to provide carbon source to sink tissues. While unloading of sucrose from phloem is believed to occur symplasmically, both symplasmic and apoplasmic routes have been proposed to exist in the post-phloem transport pathway in sink tissues. In the apoplasmic route, sucrose is released to the apoplasm via unidentified porters (possibly via sucrose/H+ symporter), and then retrieved into sink tissues via sucrose/H+ symporter, or via hexose/H+ symporter following cleavage by cell wall invertase. VP, Vascular parenchyma.

As for glucose in the animal kingdom, higher plants utilize only a few defined sugar molecules for the long-distance transport of photoassimilate. Sucrose is known as the major translocating sugar in angiosperms (see Section II.B). Also sucrose is the major solute found in the phloem (sieve tube) sap of most plant species, providing not only the carbon source for growth and metabolism but also the driving force for bulk flow of the sap, namely, translocation of all other constituents in the phloem sap, including signaling molecules such as phytohormones (see Turgeon and Wolf, 2008 for further details).

Recent intensive research has made considerable progress in elucidating the mechanism of translocation at the molecular level, particularly in the area of proton-coupled sucrose transport systems across the plasma membrane, and their role in the long-distance transport of sucrose. By way of introduction, we briefly address below some key features of the translocation of sucrose in plants then detail the specific aspects of the transport on a tissue-by-tissue basis.

A. Phloem and Sieve Element/Companion Cell Complex

The phloem of angiosperms comprises four kinds of tissues, i.e., the sieve elements (SEs), the companion cells (CCs), the phloem parenchyma and the phloem fiber (see Van Bel, 2003 for further details). SEs are longitudinally interconnected via sieve pores at their terminal ends, and serve as the conducting tube or sieve tube, for the phloem sap. Whereas SEs and CCs are formed from a common mother cell by cell division, the two types of cells undergo contrasting differentiation processes during their maturation. SEs lose a large part of the cytoplasmic content such as the nucleus, tonoplast and ­ribosomes. CCs, on the other hand, retain dense cytoplasm and numerous ribosomes and mitochondria. Since SEs and CCs are closely linked by many plasmodesmata, one important task of CCs is considered to be the supply of compounds necessary for SEs to remain active over the entire growth period. These close ontogenic and functional relationships of the two cells have led to the expression: sieve element/companion cell complex (SE/CC complex), to describe their association.

B. Mass Flow Model

To account for the mechanism of long-distance transport of solutes through the phloem continuum, a pressure-driven mass flow model offers the simplest and most widely accepted explanation (see Thorpe and Minchin, 1996; Minchin and Lacointe, 2005). According to this model, solutes (sugars, amino acids, inorganic ions, etc.) move from a source to a sink along an osmotic turgor pressure gradient in the sieve tube. Within source regions, solutes are actively taken up into sieve tubes (phloem loading). This active loading process creates a more negative osmotic potential in the sieve tube. At the sink, solutes are released from sieve tubes (phloem unloading), resulting in a less negative osmotic potential in the sieve tube, allowing water to move out. This hydrostatic pressure drives a bulk flow of phloem sap from source to sink.

C. Sucrose/Proton Symporters (Sucrose Transporters)

Since the cloning by yeast complementation of the first plant sucrose transporter (SUT, Riesmeier et al., 1992), and aided by the sequencing of the Arabidopsis thaliana and rice genomes, more than 85 putative SUT sequences from at least 35 species of higher plants have been lodged in the NCBI GenBank (http://www.ncbi.nlm.nih.gov/). All of the SUTs cloned thus far fall into the major facilitator super-family (Marger and Saier, 1993; Saier, 2000) characterized by 12 predicted plasma membrane-spanning helices, arranged in two groups of 6, separated by a cytoplasmic loop (Fig. 28.2). A histidine residue located in the first extracellular loop is conserved in all plant SUTs so far identified and has been shown to be involved in the sucrose transport activity by site-directed mutagenesis (Lu and Bush, 1998). A number of these transporters have been expressed in vitro using yeast and oocyte expression systems (see below) with K m values for sucrose ranging from 0.2 to 2 mM (see Lemoine, 2000; Kühn, 2003). There is strong evidence that the mechanism of transport is via a 1:1 sucrose/H+ co-transport, driven by a transmembrane pH gradient supported by the proton pumping ATPase (Boorer et al., 1996). Some SUTs (in particular the SUT2 class cloned by Barker et al., 2000) are characterized by large cytoplasmic loop regions and are reputed to play a role in sugar sensing (see Lalonde et al., 1999; Barker et al., 2000). The role of sucrose transporters in sugar sensing will be reviewed later in this chapter.

Fig. 28.2.
figure 2

Predicted structural model of plant sucrose/H+ symporters. The twelve membrane-spanning helices are a typical feature of the major facilitator super-family transporter proteins, with both C and N termini predicted to be intracellular, as is the hydrophilic loop between helices 6 and 7.

Many reports have shown the transcriptional regulation of SUT genes in various developmental processes, by environmental factors such as light and salt stress, and by endogenous signals such as sugars and hormones (reviewed in Lemoine, 2000; Williams et al., 2000; Kühn, 2003; Lalonde et al., 2004; Sauer, 2007). Furthermore, SUT proteins can be phosphorylated (Roblin et al., 1998) and oligomerized in vitro (Reinders et al., 2002; Schulze et al., 2003) and in vivo in a redox-dependent manner (Krügel et al., 2008). It is likely that SUTs are subject to complex regulation at different levels.

For most dicot species, two or more SUT genes have been reported (reviewed by Lemoine, 2000; Williams et al., 2000; Sauer, 2007). In A. thaliana, for example, nine SUT genes are found in the genome and their properties have been examined by functional expression in yeast cells (Sauer and Stolz, 1994; Barker et al., 2000; Meyer et al., 2000; Schulze et al., 2000; Weise et al., 2000; Ludwig et al., 2000; Sauer et al., 2004). Intriguingly, two of the A. thaliana SUT genes, AtSUC6 and AtSUC7, have been reported to encode aberrant proteins and seem to be pseudogenes (Sauer et al., 2004). In potato and tomato, three different SUT genes, SUT1, SUT2 and SUT4 have been cloned and characterized (Riesmeier et al., 1993; Barker et al., 2000; Weise et al., 2000). In monocot species, five SUT genes appear to be present in the genome sequence of rice and are expressed in a variety of tissues (Aoki et al., 2003) while in barley, two genes have been identified (Weschke et al., 2000), three homeologous genes encoding SUT1 have been cloned from wheat (Aoki et al., 2002) and one SUT1-type transporter has been cloned and characterized from maize (Aoki et al., 1999) and sugarcane (Rae et al., 2005). Heterologous expression of monocot SUT genes has proven problematic with low levels of activity achieved in yeast cells for OsSUT1 and OsSUT3 (Hirose et al., 1997; Aoki et al., 2003), barley HvSUT1 and HvSUT2 (Weschke et al., 2000), and sugarcane ShSUT1 (Rae et al., 2005); however, recently maize ZmSUT1, as well as HvSUT1 and ShSUT1, have been expressed successfully in oocytes (Carpaneto et al., 2005; Sivitz et al., 2005; Reinders et al., 2006). The reason for the difficulty in obtaining good expression of the monocot SUTs in yeast is unknown and not due to the lack of sustained effort (R.T. Furbank, N. Aoki and G.N. Scofield, unpublished observations; N. Sauer, personal communication).

Based on phylogenetic analysis of deduced peptide sequences, SUTs from both dicots and monocots can be classified into three groups (Fig. 28.3; also see Sauer, 2007; Braun and Slewinski, 2009). Type-I SUTs encompass the classic SUT1-type dicot transporters which have often been assigned a role in phloem loading (see Section II.C). Type-II includes all the monocot SUT1 genes, OsSUT3, 4 and 5 and the dicot SUT2 class (the putative “sugar sensors”). Although it is plausible that each SUT group possesses distinctive biochemical properties (see Kühn, 2003; Lalonde et al., 2004; Sauer, 2007), the three SUT groups do not appear, so far, to be functionally distinct, except possibly only on the basis of Km values for sucrose, measured in heterologous expression systems (see Lemoine, 2000; Kühn, 2003; Sauer, 2007). It is very difficult to assign physiological function based on such phylogenetic relationships, as evidenced by the common observation of expression of each of these SUT genes in a variety of tissues during plant development both in dicots and in cereals (reviewed in Kühn, 2003; Sauer, 2007). There also seem to be differences in the structure of the SUT gene family between monocots and dicots; a Type-I SUT sequence is unlikely to exist in the rice genome as it has not been found in any monocots, including expressed sequence tags in public database. The absence of a monocot gene in the classic dicot SUT1 cluster raises interesting questions about phloem loading in cereal species. The physiological role of these SUTs will be the topic of further discussion in the following sections.

Fig. 28.3.
figure 3

A phylogenetic tree of plant sucrose transporter family, based on deduced amino acid sequences. The CLASTALW program was used to show the relationship between members of this family, and a Neighbor-Joining method was used to construct this unrooted dendrogram. The scale represents 0.1 substitutions per site (one change per ten amino acids). SUT or SUT-like sequences from 38 plant species were retrieved from the GenBank and used for this analysis: Dicot species, Alonsoa meridionalis AmSUT1 (accession number AF191025), Apium graveolens AgSUT1 (AF063400), AgSUT2A (AF167415), AgSUT3 (DQ286433), Arabidopsis thaliana AtSUC1 (At1g71880), AtSUC2 (At1g22710), AtSUC3/SUT2 (At2g02860), AtSUC4/SUT4 (At1g09960), AtSUC5 (At1g71890), AtSUC6 (At5g43610), AtSUC7 (At1g66570), AtSUC8 (At2g14670), AtSUC9 (At5g06170), Asarina barclaiana AbSUT1 (AF191024), Beta vulgaris BvSUT1 (X83850), Brassica oleracea BoSUC1 (AY065839), BoSUC2 (AY065840), Citrus sinensis CsSUT1 (AY098891), CsSUT2 (AY098894), Cucumis melo CmSUT4 (FJ231518), Datisca glomerate DgSUT4 (AJ781069), Daucus carota DcSUT1A (Y16766), DcSUT2 (Y16768), Eucommia ulmoides EuSUT2 (AY946204), Euphorbia esula EeSUT1 (AF242307), Glycine max GmSUT1 (AJ563364), Hevea brasiliensis HbSUT1 (DQ985466), HbSUT2A (DQ985467), HbSUT2B (DQ985465), HbSUT3 (EF067334), HbSUT4 (EF067335), HbSUT5 (EF067333), HbSUT6 (AM492537), Ipomoea batatas IbSUT2 (AY830138), Juglans regia JrSUT1 (AY504969), Lotus japonicus LjSUT4 (AJ538041), Lycopersicon esculentum LeSUT1 (X82275), LeSUT2 (AF166498), LeSUT4 (AF176950), Fig. 28.3. (continued) Malus x domestica MdSUT4 (AY445915), Manihot esculenta MeSUT1 (DQ138374), MeSUT2 (DQ138373), MeSUT4 (DQ138371), Nicotiana tabacum NtSUT1 (X82276), NtSUT3 (AF149981), Pisum sativum PsSUT1 (AF109921), PsSUF1 (DQ221698), PsSUF4 (DQ221697), Plantago major PmSUC1 (X84379), PmSUC2 (X75764), PmSUC3 (AJ534442), Phaseolus vulgaris PvSUT1 (DQ221699), PvSUT3 (DQ221701), PvSUF1 (DQ221700), Ricinus communis RcSCR1 (Z31561), RcSUC4 (AY725473), Solanum tuberosum StSUT1 (X69165), StSUT2 (AY291289), StSUT4 (AF237780), Spinacia oleracea SoSUT1 (X67125), Vicia faba VfSUT1 (Z93774), Vitis Vinifera VvSUC11 (AF021808), VvSUC12 (AF021809), VvSUC27 (AF021810), VvSUT2 (AF439321); Monocot species, Ananas comosus AcSUT4 (EF460878), Asparagus officinalis AoSUT1 (DQ273271), Bambusa oldhamii BoSUT1 (DQ020217), Hordeum vulgare HvSUT1 (AJ272309), HvSUT2 (AJ272308), Oryza sativa OsSUT1 (X87819), OsSUT2 (AB091672), OsSUT3 (AB071809), OsSUT4 (AB091673), OsSUT5 (AB091674), Saccharum hybrid ShSUT1 (AY780256), Triticum aestivum TaSUT1A (AF408842), TaSUT1B (AF408843), TaSUT1D (AF408844), Zea mays ZmSUT1 (AB008464), ZmSUT2 (AY639018), ZmSUT4 (AY581895), As for maize SUTs, the TIGR Maize Genome Database (http://maize.tigr.org/) was searched, and seven SUT or SUT-like sequences were found and retrieved. For each gene, intron/exon borders could be predicted based on maize or rice SUT cDNA sequences lodged in the GenBank, and peptide sequences were deduced from the probable mRNAs; ZmSUT3 (accession number AZM4_91110), ZmSUT5 (AZM4_71084), ZmSUT6 (AZM4_12743), ZmSUT7 (AZM4_112588). In addition, six SUT-like sequences found in the genome of Physcomitrella patent, a non-vascular plant, were also included for comparison: PpSUT1 (XM_001752913), PpSUT2 (XM_001778945), PpSUT3 (XM_001777404), PpSUT4 (XM_001768246), PpSUT5 (XM_001766929), PpSUT6 (XM_001777602). Note that PpSUT1 and PpSUT2, which are closely related to the Type-II clade, possess a long central cytoplasmic loop similar to dicot SUT2 sequences (N. Aoki, unpublished observation).

II. Phloem Loading

Phloem loading, often defined as the movement and transport of assimilates from the mesophyll cell into the sieve tube, has primary importance as the first step in the translocation of assimilates. Mesophyll cells are closely interconnected by plasmodesmata, and the sucrose synthesized there is believed to move symplasmically via plasmodesmata toward the phloem region of minor veins (see Van Bel, 1993; Lalonde et al., 2003). In the phloem of minor veins, sucrose (or its derivatives) is transported into the SE/CC complex for translocation.

A. Phloem Anatomy and the Mode of Phloem Loading

Usually, SEs have no symplasmic connection with surrounding cells except CCs, while the frequency of plasmodesmata between CCs and surrounding cells vary greatly depending on plant species. Gamalei (1989) has published an extensive survey of phloem anatomy of leaf minor veins in as many as 700 species, and found that they can be classified into two general groups mostly in a family dependent manner: those with numerous plasmodesmata between CCs and the adjacent cells were designated as type 1 (majority of woody plants and some of herbaceous plants, e.g., Cucurbitaceae, Lamiaceae), while those with few or very rare plasmodesmata at this interface as type 2 (mostly herbaceous plants, e.g., Asteraceae, Fabaceae). This suggests that in type 1 species the assimilate could move entirely through symplasm from the mesophyll cells to the SE/CC complex (symplasmic loading; Fig. 28.4), and, alternatively, in type 2 species apoplasmic, carrier-mediated transport is likely to occur around the SE/CC complex (apoplasmic loading; Fig. 28.4). Several lines of evidence in fact support this view: for example, in type 2 species, but not in type 1 species, accumulation in the minor vein and efflux from the cut end of the petiole of 14C-photoassimilate are both severely affected by para-chloromercuribenzen sulphonic acid (PCMBS), a potential inhibitor of carrier-mediated sugar uptake from the apoplasm (Van Bel et al., 1994). In addition, molecular biological studies have shown a crucial role of SUTs located in the plasma membrane of CCs or SEs in some type 2 species (see Section II.C). However, the symplasmic continuity presented by type 1 vein structure does not necessarily exclude the co-existence of both loading modes in the SE/CC complex. It has also been reported that some plants have transitional vein anatomy between type 1 and type 2 (type 1-2a, Gamalei, 1991), but their mode(s) of phloem loading remains unclear. In addition, it is feasible that the two modes of phloem loading work within single veins or single SE/CC complexes (Van Bel et al., 1992).

Fig. 28.4.
figure 4

Phloem anatomy and the mode of phloem loading. In type 1 phloem anatomy, there are numerous plasmodesmata between phloem parenchyma (PP) and companion cells (CC), allowing solutes to move entirely through symplasm from mesophyll cells to sieve element (SE)/CC complex (arrows). In type 2, there are few or very rare plasmodesmata between PP and CC, and thus solutes have to move out to apoplasm prior to subsequent uptake into SE/CC complex via plasma membrane carrier protein (white circles).

B. Sucrose as Translocating Sugar

Sucrose is contained in the phloem sap of every plant species so far examined and in many of them it is the only detectable sugar molecule. Depending on what kinds of sugars are transported in the phloem, higher plants can be divided into three major groups (Zimmermann and Ziegler, 1975). The first group, most common among higher plants, translocates only sucrose. The second group includes some species in the Cucurbitaceae and Lamiaceae for example, which translocate mainly raffinose-family oligosaccharides (RFOs) such as raffinose, stachyose and verbascose in addition to sucrose. The third group translocates sugar alcohols besides sucrose, for example, mannitol in Apiaceae species and sorbitol in Rosaceae species. Only recently it was demonstrated that a considerable amount of hexoses (e.g., glucose and fructose) could be translocated in the phloem sap of some herbaceous plants (e.g., Paraveraceae, Ranunculaceae); this had hardly been investigated in previous studies (Van Bel and Hess, 2008). Although re-evaluation of a wide-spectrum of plant species is awaited to further address this issue, this new insight suggests that hexose can be regarded as phloem mobile transport sugar, equivalent to sucrose, RFOs, or sugar alcohols, and that a transport mechanism exists for accumulating hexose into sieve tubes.

Intriguingly, this current classification based on the combination of translocating sugars partially correlates with the mode of phloem loading or anatomy of minor vein phloem described above. The RFO-translocating species typically have type 1 minor vein structure, suggesting that RFOs favor symplasmic phloem loading. On the other hand, species with type 2 vein structure generally transport only sucrose, implying apoplasmic sucrose loading.

The hypothesis of symplasmic phloem loading in the RFO-translocating type 1 species became more compelling in light of the following physiological studies. In Coleus blumei, a typical RFO-translocating type 1 species, the concentration of solute was estimated to be much higher in SE/CC complexes than in surrounding parenchyma cells (Fisher, 1986), and accumulation of 14C-photoassimilate into minor veins occurred apparently against the concentration gradient and was insensitive to PCMBS (Turgeon and Gowan, 1990). If this is the case, a simple question arises how a high concentration of photoassimilates is maintained in the SE/CC complex, irrespective of frequent plasmodesmatal connections which can facilitate diffusion of solutes toward the surrounding cells. A model to explain this, known as ‘polymer trap theory’, has been proposed by Turgeon (1991). Sucrose moves symplasmically to CCs where it is converted into RFOs which are too large in size to diffuse back through plasmodesmata toward surrounding parenchyma cells. Several lines of evidence support this hypothesis: RFO synthesis takes place in CCs (Beebe and Turgeon, 1992), and sugar concentrations in micro-dissected leaf tissues are consistent with diffusion of substrates for RFO synthesis (sucrose and galactinol) into CCs, but not with diffusion of RFOs toward the surrounding cells (Haritatos et al., 1996). Furthermore, the importance of RFO synthesis in phloem loading has been demonstrated recently, by RNAi suppression of genes for galactinol synthase (McCaskill and Turgeon, 2007). However, the size exclusion limit of the plasmodesmata connecting the CCs with the adjacent parenchyma cells is not known although it is a key factor of this hypothesis.

It should be noted that all the type 1 species do not necessarily transport RFOs. It has been suggested that transport of RFOs is restricted to the type 1 plants having CCs with a specialized anatomy known as intermediary cells, and other type 1 plants with ordinary CCs transport mostly sucrose (Turgeon and Medville, 2004). It has also been reported as an extreme case that a type 1 species willow translocates only sucrose, and, surprisingly, the sucrose seems to move simply following the concentration gradient from the mesophyll into the transport phloem, suggesting the absence of an active loading mechanism (Turgeon and Medville, 1998). In addition, the loading mechanism of sugar alcohols and the possible relationship with phloem anatomy still remains unclear (see Noiraud et al., 2001), however, genes encoding sugar alcohol transporters have been identified in some plant species (see Juchaux-Cachau et al., 2007; and references therein). In general, it is not possible to predict the mode of phloem loading and/or sugar transport simply from vein anatomy without more comprehensive studies of the physiological and anatomical heterogeneity of the SE/CC complex.

C. Sucrose Transporter as a Phloem Loader

Since the early 1970s, it has been inferred from the following observations that energy-dependent transport is necessary for the apoplasmic loading of sucrose into the SE/CC complex (reviewed in Giaquinta, 1983; Van Bel, 1993). First, as mentioned above, only a small number of sugars including sucrose are found in the phloem sap, implying selective uptake into the SE/CC complex. Second, sucrose concentrations in the phloem sap as high as 200 to 1,600 mM suggest an active transport of sucrose against the concentration gradient. Third, the activity of H+-ATPase localized in the plasma membrane of SE/CC complex rises during the sink-to-source transition of developing leaves. Fourth, phloem loading of sucrose is markedly inhibited by the thiol group-modifying reagent PCMBS, implying the involvement of membrane protein(s) in the active transport of sucrose. Detailed biochemical studies using plasma membrane vesicles have reinforced this view, and researchers came to infer the presence of a sucrose/H+ symporter on the plasma membrane of SE/CC complexes by the late 1980s (see Bush, 1993; Van Bel, 1993). The first cDNA encoding a sucrose transporter protein was isolated from the source leaves of spinach and designated SoSUT1, S pinacia o leracea su crose t ransporter 1 (Riesmeier et al., 1992). In this study a yeast complementation system was used: a yeast strain which cannot absorb sucrose from external media but can metabolize it within the cell was transformed with a spinach cDNA library and selected on the medium containing sucrose as the sole carbon source. The yeast cells expressing SoSUT1 were shown to take up sucrose (and maltose) selectively from media with an apparent K m for sucrose of 1.5 mM. The uptake activity is pH-dependent (higher activity in lower pH) and severely inhibited by either thiol modifying reagents or uncouplers. All these biochemical properties of SoSUT1 strongly suggest it to be a H+-symporter and give striking agreement with the in planta properties of the sucrose/H+ symporter that had been assumed to be involved in phloem loading. After SoSUT1, genes for SUTs have been isolated from many other plant species such as potato (Riesmeier et al., 1993), A. thaliana (Sauer and Stolz, 1994), common plantain (Gahrtz et al., 1994) and rice (Hirose et al., 1997), and most of them have been shown to have similar biochemical and structural characteristics.

The first attempt to localize the expression of SUT genes was carried out in the source leaves of potato by in situ hybridization and demonstrated that StSUT1 mRNA is located in the phloem (Riesmeier et al., 1993). Similarly, the proteins of A. thaliana AtSUC2 (Sauer and Stolz, 1994) and common plantain PmSUC2 (Stadler et al., 1995) have been shown to exist in the plasma membrane of CCs by immunolocalization. In addition, antisense suppression of StSUT1 led to decreased leaf photosynthesis and sugar export, accumulation of soluble sugars and starch in leaves and retardation in growth (Riesmeier et al., 1994; Kühn et al., 1996; Lemoine et al., 1996). Similar phenomena were also observed in the insertion mutant lines of AtSUC2 (Gottwald et al., 2000) as well as antisense suppression lines of tobacco NtSUT1 (Bürkle et al., 1998). All these data indicate that the SUTs have a crucial role in the apoplasmic phloem loading of sucrose in the source leaves.

However, it should be noted that phloem loading is not necessarily the sole function of SUTs (see Sections III and IV). Even in the case of a typical “phloem loader” described above the expression of the SUTs has been observed not only in source leaves but also in some sink tissues where sucrose is delivered from the phloem (see Section IV). At the same time, localization of a SUT to the SE/CC complex is not the hallmark of a phloem loader (see Section III.A). Whereas rice OsSUT1 was expressed in SE/CC (Matsukura et al., 2000; Scofield et al., 2007b), antisense suppression of OsSUT1 gave no phenotypic changes supporting its role in phloem loading (Ishimaru et al., 2001; Scofield et al., 2002).

D. Phloem Loading in Grass Species

Most of the discussion on phloem loading described above is based almost exclusively on data from dicot species. To date, our knowledge on the phloem anatomy and the loading mechanism of monocot species is still limited compared to those of dicot species. The following observations from cereal species support the likelihood that apoplasmic loading of sucrose occurs in these agriculturally important crops. (1) In rice, maize, wheat and barley, sucrose is the major sugar found in their phloem sap (Fukumorita and Chino, 1982; Hayashi and Chino, 1986; Ohshima et al., 1990; Winter et al., 1992). (2) In rice, maize and barley, leaf minor veins possess the type 2a anatomy, namely, the CCs are connected with the adjacent cells by only a few plasmodesmata (Evert et al., 1978, 1996; Kaneko et al., 1980; Botha and Cross, 1997). (3) In barley and wheat, symplasmic isolation of source leaf SE/CC complexes has been shown by fluorescent dye transport experiments (Haupt et al., 2001; Aoki et al., 2004). (4) In wheat and maize, assimilate transport from leaves is inhibited by PCMBS (Thompson and Dale, 1981).

As mentioned in Section I.C, involvement of SUTs in phloem loading remains unclear in monocot species. Only a little information is available on cellular localization and physiological function of SUTs in monocot leaves, while SUT mRNAs have been found in source leaves of rice (Hirose et al., 1997; Aoki et al., 2003), maize (Aoki et al., 1999), barley (Weschke et al., 2000), wheat (Aoki et al., 2002) and sugarcane (Rae et al., 2005). Involvement of SUT in assimilate transport in source leaves was previously suggested in maize as the level of ZmSUT1 mRNA increases during leaf maturation and with accumulation of carbohydrates in the mature leaf (Aoki et al., 1999). This hypothesis that ZmSUT1 functions in phloem loading in mature maize leaves, has recently been supported by characterizing a knockout mutant in this gene (Slewinski et al., 2009). In mature wheat leaf blades, TaSUT1 protein localizes to the SE of small and medium veins, implying its direct involvement in phloem loading (Aoki et al., 2004). On the other hand, in sugarcane leaves, ShSUT1 protein is not found in the phloem but in the surrounding vascular parenchyma cells and mestome sheath cells (Rae et al., 2005), suggesting its role in retrieval, or possibly release, of sucrose (see Sections II.E and III.A). Furthermore, unlike dicot Type-I SUTs, antisense repression of OsSUT1 expression in transgenic rice does not appear to affect phloem loading (Ishimaru et al., 2001; Scofield et al., 2002). Although these monocot SUT1 orthologs all fall into the same cluster in the phylogenetic tree of plant SUTs (Fig. 28.3) and appear to possess a similar affinity for sucrose (Weschke et al., 2000; Rae et al., 2005; Carpaneto et al., 2005), their localization and role(s) in the phloem loading pathway may differ depending on species. Further studies are obviously needed in order to evaluate the contribution of SUT-mediated membrane transport to phloem loading in grass species, considering the expression and localization of different SUT isogenes. In fact, all five SUT isogenes appear to be expressed in rice source leaves (Aoki et al., 2003).

Interestingly, a maize mutant sxd1 (sucrose export defective1) exhibits severely impaired growth and accumulates a large amount of starch in its leaves (Russin et al., 1996). This symptom has been attributed to occlusion of plasmodesmata at the bundle sheath/vascular parenchyma interface and the resultant blockage of assimilate export from the source leaves (Russin et al., 1996; Botha et al., 2000). Although neither are directly relevant to the SE/CC complex nor SUTs, this provides valuable information on the physiological role of plasmodesmata in the intercellular transport of assimilates.

Later on, the maize SXD1 gene has been isolated (Provencher et al., 2001) and surprisingly found to be highly similar to an A. thaliana gene VTE1, which encodes a tocopherol cyclase, an enzyme essential for vitamin E synthesis (Porfirova et al., 2002). Subsequently, Sattler et al. (2003) have confirmed a tocopherol deficiency in maize sxd1 leaves and tocopherol cyclase activity with recombinant SXD1 protein. However, an A. thaliana tocopherol-deficient line vte1 having a mutation in VTE1 did not show any sucrose export phenotype analogous to the maize sxd1 (Porfirova et al., 2002). In contrast, RNAi-mediated silencing of an SXD1 ortholog in potato led to a tocopherol deficiency in leaves and impaired photoassimilate export from leaves concomitant with vascular-specific callose deposition in source leaves (Hofius et al., 2004). It appears that the response of assimilate export phenotypes to a deficiency in tocopherol cyclase differs depending upon plant species.

Recently, another interesting maize mutant tdy1 (tie-dyed1) was reported and characterized. The tdy1 mutant plants develop variegated yellow and green leaf sectors and accumulate starch and sucrose in the yellow sectors (Braun et al., 2006). The starch accumulation of tdy1 leaves seems to be attributable to impaired sucrose export without reduction of starch breakdown (Slewinski et al., 2008) but by different mechanisms from sxd1 (Ma et al., 2008). Subsequently, TDY1 gene was cloned and found that TDY1 encodes a novel transmembrane protein and is exclusively expressed in phloem of both sink and source tissues (Ma et al., 2009). Fluorescent dye movement experiments revealed that source leaves of tdy1 plants showed retarded phloem loading, but phloem transport and unloading in sink tissues were unaffected, suggesting that TDY1 functions in carbon partitioning by promoting phloem loading although a detailed mechanism remains unclear. Notably, TDY1 orthologous proteins are found only in grass species to date. This may be a clue to understand the mode and the regulation of phloem loading in grasses.

E. Releasing Sucrose to Phloem Apoplasm

One important issue that has remained unresolved in the apoplasmic phloem loading pathway is where and how sucrose is released to the apoplasm prior to its active uptake into SE/CC complexes. There is no clear answer to this question, although the importance of the apoplasmic sucrose pool in phloem loading has been clearly demonstrated in vivo in transgenic solanaceous and A. thaliana plants expressing a yeast acid invertase in the cell wall (reviewed in Schobert et al., 2000). The overexpression of extracellular invertase led to a large accumulation of carbohydrates, a reduction of photoassimilate export and an inhibition of photosynthesis in the source leaves, resulting from cleavage of sucrose in the leaf apoplasm followed by retrieval of hexoses into mesophyll cells (see Schobert et al., 2000).

Based on physiological studies using leaf tissues and discs and structural observations on plasmodesmatal connectivity in leaf vascular bundles, phloem parenchyma cells (or bundle sheath cells in some cases) are considered to be the most probable site for the sucrose release (see Van Bel, 1993; Beebe and Russin, 1999). In some plant species, the minor vein phloem is surrounded by ‘thick-walled’ cells, whose cell walls are sometimes heavily lignified or suberized (e.g., bundle sheath cells in C4 plants). These thickened cell walls surrounding the phloem may be a physical barrier to membrane transport between the two apoplasmic compartments, vascular phloem tissue and non-vascular mesophyll tissue (cf. Section III.B.2). If this is the case, photoassimilate has to move symplasmically via plasmodesmata, to reach phloem (parenchyma) cells. A series of reports on the maize sxd1 mutant mentioned in Section II.D provides the most promising evidence for this hypothesis (see Schobert et al., 2000).

Mesophyll cells have been demonstrated to be able to release and take up sucrose in a number of studies (see Schobert et al., 2000). Whereas the mechanism of the release (or leakage) still remains unclear, the uptake has been suggested to occur via a H+-coupled plasma membrane carrier protein (Bush, 1993; Schobert et al., 2000). This release/retrieval in mesophyll is considered to regulate the levels of sugars in the mesophyll apoplasm and/or to achieve an effective channeling to actual releasing cells in the phloem. However, localization of SUT to the plasma membrane of source leaf mesophyll cells has not been shown to date (but cf. Lemoine et al., 1989).

For sucrose release from the cytosol to the apoplasm, a carrier-mediated transport system has been suggested to exist in the plasma membrane of source leaves (see Van Bel, 1993; Schobert et al., 2000; Lalonde et al., 2004). This plasma membrane efflux protein is thought to localize to sucrose-releasing cells in the phloem, most likely in phloem parenchyma cells, and facilitate the export of sucrose from the cells to the phloem apoplasm by an unascertained mechanism, presumably H+-antiport or uniport, considering acidic pH and low sucrose concentrations in the apoplasm. To date, however, no such efflux protein has been identified in plant cells. A. thaliana AtSUC3 and sugarcane ShSUT1 have been shown to localize to the cells adjacent to the phloem in leaves (Meyer et al., 2000; Rae et al., 2005), providing a possible mechanism for the release of sucrose toward the phloem if they were to function bi-directionally. This notion has gained support following the demonstration that the sucrose/H+ symporter can mediate the efflux of sucrose in Vicia faba leaf discs (M’Batchi and Delrot, 1988), in tobacco leaf plasma membrane vesicles (Borstlap and Shuurmans, 2004), and in oocytes when heterologously expressed (Carpaneto et al., 2005). The existence of sucrose effluxers can also be inferred in the post-phloem unloading pathway in sink tissues such as developing seeds (see Section IV.C). In addition, it should be noted that vesicle-mediated transport might possibly be another mechanism for sucrose efflux to the apoplasm when the sucrose stored temporally in vacuoles is remobilized (see Echeverría, 2000). In this case, vacuolar sugar porters may play roles in transporting sucrose across the tonoplast membrane (see Echeverría, 2000; Lalonde et al., 2004). In fact, localization of SUT proteins to the tonoplast membranes of mesophyll cells has recently been reported for barley HvSUT2, A. thaliana AtSUC4 and Lotus japonicus LjSUT4, all of which are members of Type-III SUTs (Endler et al., 2006; Reinders et al., 2008). However, metabolic and physiological functions of these vacuolar sucrose/H+ symporters remain unclear.

III. Long-Distance Transport (Translocation) of Sucrose

Following phloem loading in minor veins, sucrose is accumulated in major vein phloem through the leaf vein network, and then translocated to the other parts of plant. The phloem in major veins, petioles, stems and roots functions as pipes to connect source leaves with terminal sinks such as root tips, shoot apices, flower tissues and developing seeds, and is often termed the ‘transport phloem’ (see Van Bel, 2003). At present it is generally accepted that not only loading and unloading control the mass flow of phloem sap, but that the transport phloem also plays a role. (see Willenbrink, 2002; Van Bel, 2003; Minchin and Lacointe, 2005)

A. Retrieval Role of Sucrose Transporters in Transport Phloem

As mentioned in Section I, sucrose is the major osmotically active molecule in the phloem sap of most plant species. However, as sieve tubes are essentially leaky pipes, sucrose and other solutes are readily lost from the phloem sap during translocation (reviewed in Van Bel, 2003). In this context, one can presume the necessity of sucrose-retrieving mechanism(s) in the SE/CC complex of transport phloem, in order to return sucrose leaked to the phloem apoplasm (Willenbrink, 2002; Van Bel, 2003).

In fact, Grimm et al. (1990) have reported that the efflux of sucrose in the petioles of Cyclamen persicum is compensated by carrier-mediated influx, suggesting the existence of sucrose/H+ symporter in the phloem. Ayre et al. (2003) compared the distribution of endogenous translocating sugars (sucrose and/or RFOs) to that of exotic solutes (galactinol and octopine) loaded in leaf minor veins, using tobacco, which translocates predominantly sucrose, and Coleus blumei, an RFO-translocating species. While either the endogenous solutes or the exotic solutes were efficiently exported from the leaf lamina to the petioles through the sieve tubes, only the exotics were found to accumulate in substantial amount to the petiole apoplasm. From the results, the authors have proposed the existence of retention and retrieval mechanism(s) specific for endogenous translocating sugars in the transport phloem (Ayre et al., 2003).

Despite the fact that mRNAs of SUTs have been found in petioles, stems and/or roots in most plants examined, much less attention has been paid until recently to SUTs in the transport phloem than to those involved in loading and unloading. Localization of SUT proteins in the SE/CC of transport phloem has been reported in a number of plant species. In wheat, for example, TaSUT1 protein has been shown to localize to SEs along the transport phloem of leaf blades, leaf sheaths and peduncles (Aoki et al., 2003). Similar localization patterns of SUT protein in the phloem along the long distance transport pathway has been reported in rice for OsSUT1 (Scofield et al., 2007b). In tomato petioles and stems, the co-localization of three different SUT proteins in the same SE has been shown (Reinders et al., 2002). These three tomato SUTs, LeSUT1, SUT2 and SUT4, possess distinctive structures and show different affinities for sucrose in vitro, suggesting different physiological roles in the transport phloem (Riesmeier et al., 1993; Barker et al., 2000; Weise et al., 2000). It seems likely that the SE/CC complex of transport phloem deploys different SUTs in order to avoid critical losses of sucrose from the phloem sap and to maintain an efficient rate of translocation (Willenbrink, 2002; Van Bel, 2003; Ayre et al., 2003). SUT expression in ‘non-loading’ phloem has also been shown in the stem and tuber of potato (Kühn et al., 1997, 2003), in the petiole of celery, a mannitol-translocating species (Noiraud et al., 2000, 2001), and in the transport phloem of Alonsoa meridionalis, an RFO-translocating, symplasmic phloem loader (Knop et al., 2001, 2004). It is therefore reasonable to assume that ‘retrieving apoplasmic sucrose’, rather than ‘phloem loading’, is a more general role for the SUTs in the SE/CC among higher plants.

Moreover, studies using promoter-reporter constructs and specific antibodies have revealed that SUT genes are expressed in a variety of non-phloem cells in both source and sink tissues, including distal cells such as root tips, pollen grains, fiber cells, trichomes and guard cells (Meyer et al., 2000, 2004; Barth et al., 2003; Rae et al., 2005; also see Section IV) and more recently, vessel-associated cells in the xylem of walnut, a woody species (Decourteix et al., 2006). While one can speculate a role of SUT in ‘utilizing extracellular sucrose’, this may possibly extend the physiological significance of SUT genes beyond translocation and carbon partitioning. This possible diversity in physiological function of SUT may explain why higher plants possess a multi-gene family encoding sucrose/H+ symporters.

B. Carbohydrate Storage in Stems

Petioles, stems and roots of the plant are also known to function as lateral sinks (or axial sinks) for storing assimilates, as well as pipes for assimilate translocation from source leaves to terminal sinks. Here we discuss the transport of sucrose between phloem and storage cells in stems, focusing on some agriculturally important grass species, in which the stem storage of assimilates has a significant influence on yield.

1. Stem Carbohydrate Reserves in Cereal Species

In some cereal crops, temporal storage of carbohydrate in the stem has been known to contribute substantially to final grain yield (see Cock and Yoshida, 1972; Schnyder, 1993). The stem of grass species comprises the leaf sheaths, the internodes and the peduncle, connecting the nodes. These stem parts save excess photoassimilates when photosynthetic activity is high in source leaves, and accumulate these carbon reserves in the form of polysaccharides such as starch (in rice) or fructan (in wheat and barley). In this ‘accumulation’ phase, the storage parenchyma cells synthesize and store polysaccharides (starch or fructan) from sucrose in particular organelles (plastid or vacuole, respectively). The stored polysaccharides are remobilized and translocated to filling grains later when the leaf photosynthesis declines due to senescence. In this ‘remobilization’ phase, the stored polysaccharides are degraded within the organelle, exported into the cytosol, and converted into sucrose. Not surprisingly, changes in carbohydrate metabolism enzymes during this sink-to-source transition in the stem of cereal species have been shown at the biochemical and molecular levels (Hirose et al., 1999, 2006; Wardlaw and Willenbrink, 2000; Van den Ende et al., 2003; Hirose and Terao, 2004; Takahashi et al., 2005; Scofield et al., 2009; and references therein). However, further studies are necessary to clarify the synthesis and degradation pathway of polysaccharides, particularly metabolite transport processes across organelle membranes in storage cells.

Apart from the changes in metabolic pathway within storage cells, intercellular transport of sucrose between phloem SEs and storage cells could also change during the sink-to-source transition; phloem unloading in the sink phase along a sucrose gradient, and reloading in the source phase, presumably against a sucrose gradient. However, little information is available to date on solute transport in mature cereal stems. In wheat internodes, unlike in stems of some dicots (reviewed in Van Bel, 1995), fluorescent dyes can move symplasmically from the phloem to the storage parenchyma, via plasmodesmata (Aoki et al., 2004). Further studies are required on intercellular connections via plasmodesmata, and on expression and localization of transporter proteins, as well as sucrose-related enzymes. Hirose et al. (1999) have reported that OsSUT1 mRNAs appear to increase during the sink-to-source transition in rice leaf sheaths, in parallel with mRNAs for sucrose biosynthetic enzymes. This SUT may be involved in the transport path to reload sucrose into the phloem in the remobilization phase.

The sink-to-source transition in stems of cereal species, unlike that in developing leaves (see Section IV.A), occurs in the mature tissues that have already completed their development. Thus cereal stems are of particular use for studying the mechanism of this drastic change in carbon partitioning. It would be very interesting to investigate this phenomenon intensively from the perspectives of tissue and cellular anatomy, enzyme equipment for carbohydrate metabolism, symplasmic connections for solute transport, and/or transcript and protein profiles.

2. Sucrose Accumulation in Sugarcane Stems

Sugarcane, another grass species, accumulates a great quantity of sucrose in the internode and thus is the major source of commercial sugar. In maturing sugarcane internodes that are actively accumulating sucrose, it has been suggested that solutes can move symplasmically from the phloem to the storage parenchyma cells (Jacobsen et al., 1992; Rae et al., 2005). In the maturing internodes, cells at the periphery of the vascular bundle have been found to possess lignified and suberized walls, potentially forming a barrier to apoplasmic movement between phloem and storage cells, and these thick-walled cells employ a SUT protein, ShSUT1, presumably in a retrieval role in non-phloem cells (Rae et al., 2005). As well as the symplasmic pathway of sucrose transport, an apoplasmic pathway has also been proposed where sucrose is released to the internode apoplasm and cleaved by a cell wall invertase, followed by uptake via a hexose transporter and synthesis of sucrose within the cell (Walsh et al., 1996). Moreover, Casu et al. (2003) have reported that maturing internodes express a number of putative sugar transporters, as well as known sucrose and hexose transporters.

IV. Import of Sucrose to Sink Tissues

A. Sink Leaves

It has been elegantly shown by Oparka et al. (1999) that in the young dicot sink leaves but not source leaves, functional symplasmic connections exist between the phloem SEs and the adjacent mesophyll cells. These experiments indicated that the plasmodesmatal connections between the phloem SEs and mesophyll tissues became branched and ceased to pass green fluorescent protein- (GFP-) tagged virus or fluorescent dyes following the developmental transition from sink to source status which occurs in a wave from the base of the leaf to the tip (Oparka et al., 1999). It is not clear how the structural changes in plasmodesmata relate to the gating of these channels or what signal transduction pathway is responsible for the closure of the plasmodesmata, however, it appears that there is a symplasmic pathway present for post-phloem delivery of sucrose to the sink cells of the young leaf. Experiments with tracer dyes (Haupt et al., 2001; Aoki et al., 2004) indicate that a similar symplasmic connection also exists in sink leaf tissue of cereals but not in mature source leaves. Patrick and Offler (1996) support the thesis that in both young leaves and roots, sucrose is delivered symplasmically. These authors point out that by comparing measured sugar transport rates between plant tissues (such as phloem sieve tube to companion cell and seed maternal tissue to filial tissue) it appears that two orders of magnitude higher fluxes are possible through plasmodesmatal connections as compared to apoplasmic active transport (Patrick and Offler, 1996). The existence of a symplasmic pathway for sugar import to source leaves seems at odds, however, with the observation that hexose transporters and cell wall invertase are expressed at high levels in young sink leaf tissue and in roots (reviewed in Roitsch and Tanner, 1996). The presence of these proteins suggests that sucrose could be delivered into the apoplasm from the phloem where it is cleaved by cell wall invertase then taken up actively into the sink cells by hexose transporters. Paradoxically, all 5 OsSUT genes in rice are also expressed in young sink leaves (Aoki et al., 2004). The presence of sucrose synthase and SUTs, invertase and hexose transporters and functional plasmodesmata in sink leaves makes it difficult to make a clear pronouncement as to the route of post-phloem carbon import to sink leaf tissues. A likely scenario, however, is that in a similar model to that described above for long-distance transport and storage in stem tissue, symplasmic import maybe the main route for sugar flux, with transporters acting in a retrieval mode to rescue sugars from the apoplasm. This model is supported by the observation that ectopic expression of invertase in the apoplasm of solanaceous species did not appear to affect early leaf development but had profound effects on source leaf gene expression, photosynthesis, carbohydrate composition and senescence (Sonnewald et al., 1991).

B. Floral Tissues

Sugar transport plays a pivotal role at three major stages of flowering and pollen development. Firstly, the developing pollen grain must be nourished with sugars from the maternal tissues of the anther (Dorion et al., 1996; Roitsch et al., 2003), secondly sugar transporters are thought to be important for anther dehiscence (Stadler et al., 1999) and lastly, sugars and SUTs are of pivotal importance in pollen tube growth during pollen germination (Stadler et al., 1999; Hackel et al., 2006; Sivitz et al., 2008). Recently, the pathway of sugar import to pollen grains during development has been studied intensively using a molecular/biochemical approach, driven by a desire to understand the processes underlying environmental stress effects on pollen fertility (Koonjul et al., 2005) and the physiological role of sugar transporters in pollen and anthers (Imlau et al., 1999; Stadler et al., 1999; Lemoine et al., 1999). A pivotal study by Imlau and co-workers has demonstrated the locations of the apoplasmic barriers to solute delivery during pollen development (Imlau et al., 1999). These authors reported that in A. thaliana transformed with GFP driven by the AtSUC2 promoter, GFP could traffic in the phloem and be unloaded symplasmically in sink tissues such as the anther. This evidence of symplasmic connections between phloem SEs and anther tissue is consistent with dye feeding experiments which suggest that the major barrier to symplasmic transport between the phloem and the pollen occurs at the level of the tapetum. These observations are also consistent with the effects of temperature and drought stress on pollen fertility in rice (Sheoran and Saini, 1996; Oliver et al., 2005) and wheat (Dorion et al., 1996; Koonjul et al., 2005). In cereals, under stress conditions, import of sugars to the pollen grain is impaired due to down regulation of a cell wall invertase and presumably a hexose transporter (Koonjul et al., 2005; Oliver et al., 2005). Sugars and starch build up in the anther and pollen is starved of sugars for starch formation, resulting in low fertility (Sheoran and Saini, 1996). Pollen fertility seems particularly sensitive to this stress around pollen microspore stage, i.e., at meiosis (Oliver et al., 2005). Interestingly, SUTs have also been reported to express in anther tissues of rice and in pollen (Takeda et al., 2001; T. Hirose, G.N. Scofield, N. Aoki and R.T. Furbank, unpublished observations). Whether a switch from the cell wall invertase / hexose transporter couple to the SUT / sucrose synthase sugar import system occurs when pollen develops (similar to that seen in developing seed, see Section IV.C), remains to be elucidated.

C. Developing Seeds

While there still may be controversy concerning the pathway of phloem loading in leaves, there is little doubt that post-phloem transport of sucrose into filial tissue of both monocot and dicot seeds must occur apoplasmically. In all higher plants, the filial and maternal tissues of seeds are separated by the plasma membrane of both tissues with no functional plasmodesmatal connections remaining post-fertilization (see Patrick and Offler, 1995).

1. Dicot Seeds

In dicots, there is strong support for the model that in early seed development, sucrose is delivered from the phloem to the inner seed coat where it passes by processes unknown to the apoplasm and is cleaved by cell wall invertase. Hexoses are subsequently imported to the embryo via hexose transporters (reviewed by Weber et al., 1997; summarized in Fig. 28.5). Following contact of the cotyledons of the embryo with the inner seed coat, invertase and hexose transporter expression are markedly reduced and a SUT is highly expressed in the cotyledonary epidermis, coinciding with the high level expression of sucrose synthase within the cotyledon (reviewed in Weber et al., 1997; summarized in Fig. 28.5). While evidence for this developmental change in sugar import pathways has been generated from the model system Vicia faba, it appears from works in common plantain (Gahrtz et al., 1996), canola (King et al., 1997), cotton (Ruan et al., 2001), A. thaliana (Ruuska et al., 2002; Baud et al., 2005; Fallahi et al., 2008), that this may be an ubiquitous mechanism in dicot seeds. This transition in sugar import pathways is thought to herald the onset of storage product accumulation and the cessation of cell division. While the other (possibly hormonal) signaling compounds involved in this developmental switch remain unresolved, it has been proposed that sugar signaling, in particular the hexose to sucrose ratio, is pivotal in controlling this process (see Weber et al., 1997). Evidence for the importance of sugar signaling in the transition from cell division to cell expansion has also been reported from measurements of local sugar levels across sections of developing V. faba embryos (Borisjuk et al., 1998, 2002) where areas of active cell division correlated with high hexose to sucrose ratios. Also, in transgenic Vicia narbonensis seeds, expressing a yeast invertase targeted to the apoplasm, sugar levels were seriously affected and development of the embryo disrupted (Weber et al., 1998; Neubohn et al., 2000). In transgenic chickpeas harboring a similar gene construct, cell number was increased, starch and protein deposition were greatly reduced in high invertase expressing lines and germination was also disrupted, presumably by the persistence of this enzyme activity through desiccation and after rehydration (R.T. Furbank, P. Gremigni, D. Büssis and N. Turner, unpublished observations). It is interesting to note, however, that the developmental switch from hexose import to sucrose import which occurs in legumes in concert with the onset of storage phase may not be pertinent to oilseeds with a persistent endosperm such as tobacco (Tomlinson et al., 2004). In this case, 30-fold overexpression of an invertase in the embryo and seed coat apoplasm, with associated large changes in hexose levels, had no deleterious effects on seed filling (Tomlinson et al., 2004).

Fig. 28.5.
figure 5

Post-phloem transport of sucrose in developing seeds. Sucrose can move symplasmically from the SE/CC complex to the border cells of maternal tissue (nucellus or inner seed coat), but has to be released to the apoplasm prior to being taken up into filial tissues (endosperm or cotyledon). Sucrose/H+ symporter expressed in the maternal border cells may be able to act as a sucrose effluxer. In the filial tissues, two different mechanisms have been proposed for the uptake of apoplasmic sucrose; (1) sucrose is cleaved by cell wall invertase (INV), followed by the uptake via hexose/H+ symporter, and (2) sucrose is directly taken up via sucrose/H+ symporter and metabolized via sucrose synthase (SUS) within the filial cells. Symbols and abbreviations are same as in Fig. 28.1.

In the post-phloem sugar transport pathway of dicot seeds, the step which remains to be resolved is the mechanism of export of sucrose from the maternal seed coat. Sucrose is believed to move symplasmically from the phloem SEs to the cells of the inner seed coat. Sucrose must then move into the apoplasm surrounding the embryo and the inner seed coat cells to be cleared by cell wall invertase or to be actively taken up by the cotyledonary epidermal cells via SUTs (see Fig. 28.5). From the sucrose gradients between maternal tissues and the apoplasm calculated for Phaseolus vulgaris and V. faba (Patrick and Offler, 1995) simple diffusion could only support 40–50% of the flux in this transport step and carrier mediated efflux has been proposed (Patrick and Offler, 1995). A carrier mediated efflux is also supported by the inhibition of sucrose efflux from seed coats by the sulfhydryl inhibitor PCMBS and the sensitivity of this transport process to inhibition of respiration (reviewed in Patrick and Offler, 1995). Presumably, for sucrose export into the apoplasm, a sucrose/H+ antiporter would be required. No such protein has been cloned or characterized to date. Intriguingly, SUT protein and RNA have been detected at high levels in the maternal seed coat tissues of a range of dicot seeds (in pea, Tegeder et al., 1999; Zhou et al., 2007; in V. faba, Weber et al., 1997; in P. vulgaris, Zhou et al., 2007; in common plantain, Lauterbach et al., 2007; and in canola, R.T. Furbank, unpublished observations). It has been postulated (see Patrick and Offler, 1995) that the sucrose/H+ symporters could operate in the reverse direction as sucrose effluxers in seed coat transfer cells, either actively or as a passive pore. Recently this notion has gained more credence as the maize ZmSUT1 gene product has been shown to be bidirectional when expressed in oocytes (Carpaneto et al., 2005). More recently, Zhou et al. (2007) discovered novel proteins being classified into either Type-I or Type-II SUTs in seed coats of developing seeds of pea and bean. When expressed in yeast, these SUT-like proteins exhibited kinetic and biochemical properties of pH-independent facilitator, and thus were named SUFs (sucrose facilitators). The SUFs are expressed in cells that are considered responsible for sucrose efflux from seed coats, suggesting that SUFs can play a role in releasing sucrose from seed coat cells by facilitated diffusion according to an outward-directed gradient of sucrose across the plasma membrane (Zhou et al., 2007).

2. Cereal Grains

In contrast to dicot seeds, sugar import to cereal grains received only limited attention until relatively recently. The proposed pathway of import of sugars to cereal endosperm is shown in Fig. 28.5. It has long been known that sugar import from the maternal nucellus tissue to the endosperm requires transport across the apoplasm and that uptake of sugars across the aleurone layer surrounding the endosperm is energy dependent (see Patrick and Offler, 1995). The sugar transporters responsible for this process, however, were not cloned and localized until the last decade. OsSUT1, the first cereal SUT cloned (see above and Hirose et al., 1997) has been shown to be highly expressed during grain filling and to be localized to the aleurone/subaleurone and nucellus tissue around the top of the grain, closer to the nucellar projection (Furbank et al., 2001). Antisense suppression of this gene caused carbon starvation of the endosperm and a shriveled grain phenotype (Scofield et al., 2002). Similar expression patterns have been found for HvSUT1 in barley (Weschke et al., 2000), and for TaSUT1 in wheat (Bagnall et al., 2000; Aoki et al., 2002).

Whether the proposed dicot model for a switch from an invertase / hexose uptake mechanism early in development to a sucrose uptake mechanism during starch accumulation holds for cereal seeds is still under investigation. It appears from measurements of transcript levels of cell wall invertases and hexose transporters during development of rice grains (Hirose et al., 2002) and barley grains (Weschke et al., 2003) that hexoses are indeed taken up by the endosperm early in grain development. Transcript profiling in barley however, shows that sucrose synthase is also important in early development for providing UDP-glucose required in cellularization of the endosperm (Sreenivasulu et al., 2004).

The pathway of sugar import to maize kernels is less well characterized than in the other cereals and may deviate from the model for rice, wheat and barley. Early in development, hexoses produced by the action of cell wall invertase in the maternal pedicel tissues are taken up by the endosperm basal transfer cells, presumably by a hexose transporter (see Patrick and Offler, 1995). This pathway of early sugar import is supported by analysis of the maize miniature mutant, a cell wall invertase knock out, which has smaller grains, lower hexose levels, lower cell numbers and reduced endosperm mitotic activity (Vilhar et al., 2002). Curiously, the maize sh1 mutant, a sucrose synthase (SS1) knock out with obviously shrunken grain, shows only a mild reduction in starch content but early degradation of endosperm cell integrity (see Chourey et al., 1998; and references therein). The mutant Sus1-1 (a mutation in sucrose synthase isoform SS2, believed to be responsible for starch biosynthesis), has no obvious phenotype while a double mutant in both isoforms has less than 0.5% of typical sucrose synthase activity in the endosperm (Chourey et al., 1998) but still synthesizes approximately 50% of normal levels of starch (Chourey et al., 1998). This observation implies that the invertase / hexose transport pathway may be active throughout maize kernel development and both sucrose itself and hexoses may be imported during starch biosynthesis.

As with developing dicot seeds, the post-phloem efflux of sucrose from maternal tissues is also somewhat of a mystery in cereal seeds. While the cells of the nucellus in wheat appear to be symplasmically connected and continuous with the phloem (see Patrick and Offler, 1995), the efflux of sucrose from the nucellar projection is sensitive to sulfhydryl inhibitors and uncouplers (Bagnall et al., 2000). Consistent with these observations, SUT1 transcript and protein are detected in the nucellus tissues of both wheat (Bagnall et al., 2000) and rice (Furbank et al., 2001). Once again, in the absence of the discovery of any candidate genes for sucrose effluxers, it is tempting to speculate a role for SUTs in the efflux of sugar from the maternal tissue into the filial apoplasm where it is taken up by the same protein located in the aleurone acting in uptake mode.

Besides carrier-mediated sugar transport, it has been reported recently that inhibitors for endocytosis have been shown to prevent the conversion of 14C-sucrose into starch in developing barley endosperm, suggesting that vesicle-mediated transport of sucrose may contribute to apoplasmic transport of sucrose in grains during starch accumulation (Baroja-Fernandez et al., 2006). The presence of endocytic transport system for sucrose uptake into sink tissues has also been proposed in potato tubers and citrus fruits (Etxeberria et al., 2005; Baroja-Fernandez et al., 2006). Further studies are required to evaluate whether sugar transport by endocytosis contributes in vivo to the uptake of translocated assimilates into sink cells.

D. “Fleshy” Fruits

The discussion above has focused primarily on the import of photosynthate to the embryo of dicot seeds and the endosperm of cereal seeds: the primary site of storage in many agriculturally important species. Also of interest is the pathway of sugar import to fruiting structures where maternal tissue is the major carbon storage organ. The systems most studied are tomato fruit, grape berry and cotton seed. In tomato fruit, there is a developmental transition from symplasmic continuity between storage parenchyma and phloem SEs early in fruit formation to symplasmic isolation and the expression of a hexose transporter and cell wall invertase in parenchyma tissue (Ruan and Patrick, 1995). This change in transport mechanism may coincide with the transition from starch storage to accumulation of high concentrations of hexoses in the fruit. In grape berry (Coombe, 1992), it appears that sucrose can move symplasmically from the phloem to the vascular parenchyma cells and is then delivered to the apoplasm where it is taken up by the mesocarp cells (see Patrick and Offler, 1996). Comprehensive developmental studies have not been carried out but it appears that this active uptake takes place via cell wall invertase coupled to hexose transporters in the mesocarp flesh cells (Davies and Robinson, 1996; Fillion et al., 1999). However, high-level expression of a SUT in the same tissue is also observed during berry ripening, suggesting that sucrose itself may also be actively transported (Ageorges et al., 2000).

Cotton fiber is an important example of a crop plant where the major storage tissue is maternal in origin. Cotton fiber initiates as a single cell from the ovule epidermis at flowering then elongates to a length of up to 3 cm in the space of 15 to 20 days after flowering (see Ruan et al., 2001). The major storage compound in cotton fiber is cellulose and mature fiber can be comprised of over 90% cellulose by dry weight. Apart from the importance of this fiber in textile production, it has been shown to be a valuable model for cell elongation and solute transport in rapidly elongating tissue (Ruan et al., 2000, 2001). Sugars move into the cotton fiber symplasmically from the seed coat during early development then during the rapid elongation phase, plasmodesmata joining the fiber cells to the underlying seed coat are gated, no longer passing solutes or the fluorescent dye 6-carboxyfluorescein (6-CF), from the seed coat vascular bundle to the fiber (Fig 28.5). Sucrose and potassium transporters are highly expressed at this stage, causing turgor to increase dramatically in the fibers, driving the rapid elongation phase (Ruan et al., 2001). Plasmodesmata then reopen at the end of elongation, followed by a phase of rapid cellulose deposition. The appearance of callose around the neck of the plasmodesmata correlates with this closure and the reopening correlates with the expression of a fiber-specific β-glucanase, thought to degrade the callose constriction of the pore (Ruan et al., 2004, 2005). This is an important example of how the symplasmic pathway of solute transport can be restricted and then resumed during development (see Oparka and Roberts, 2001; Roberts and Oparka, 2003).

E. Roots and Tubers

Sugar import to roots and tubers has been extensively studied in the species potato and sugar beet which store substantial carbon reserves in below ground tissues (see Patrick and Offler, 1995; Bell and Leigh, 1996). Viola et al. (2001) reported evidence that the pathway of post-phloem transport of sugars into the stolon of potato changed with development. These authors observed an apoplasmic step early in stolon swelling and expansion but symplasmic connectivity with the phloem SEs later, during starch deposition (Viola et al., 2001). Localization of cell wall invertase expression suggests that this enzyme may be important early in development and in the apical bud region of the tuber. These observations may explain why overexpression of a yeast invertase in the apoplasm of potato increased tuber size (Sonnewald et al., 1997) if developmental expression patterns of this enzyme are important in determining cell number through sugar signaling early in stolon development and in termination of apical growth.

Post-phloem movement of sucrose into root tissues of non-tuber forming species has been studied in less detail, but elegant experiments using laser confocal microscopy and the phloem mobile fluorescent dye, 6-CF have shown that in root tips, a symplasmic domain extends from the phloem SEs to the meristematic tissue (Oparka et al., 1994, 1995).

F. Germinating Seeds

Seed germination is an interesting system for the study of sugar transport as the machinery which was involved in carbon storage during seed development becomes source tissue to fuel shoot and root growth in the next generation. The process of seed germination has been extensively studied in cereals from the focus of starch remobilization and hormonal control of gene expression (Bewley, 1997). Until recently, however, little was known about the pathway of sugar movement from the cereal endosperm to the shoot and root or from embryo storage tissues to the embryonic axis and growing tissues of the dicot germinating seed. Relevant to the apoplasmic transport pathway, SUT transcripts had been detected in phloem tissues of castor bean (Bick et al., 1998) and rice (Matsukura et al., 2000), suggesting that sucrose is synthesized from storage reserves and translocated in the phloem of seedling shoots. The role of non-phloem tissues in sucrose transport during seed germination is less certain. In germinating castor bean seeds, epidermal transfer cells of cotyledons abutting the endosperm contain abundant SUT mRNAs (Bick et al., 1998). However, the function of these SUTs in sugar retrieval from the endosperm is unclear as the starch hydrolysis products are generally known to be glucose and maltose. In cereal species, rice OsSUT1 (Hirose et al., 1997; Matsukura et al., 2000), wheat TaSUT1 (Aoki et al., 2002) and maize ZmSUT1 (Aoki et al., 1999) have been found to be highly expressed in germinating seeds. In rice, OsSUT1 appeared to be expressed dominantly in germinating seeds compared with the other four SUT isogenes (Aoki et al., 2003), and anti-sense suppression lines for OsSUT1 showed retarded germination (Scofield et al., 2002). These results strongly suggest that expression of the SUT1 ortholog is very important for seed germination in cereal species. However, cellular localization of SUT1 expression in germinating cereal seeds has not been fully investigated. In rice, the expression of OsSUT1 gene and the localization of OsSUT1 protein in the scutellum phloem of germinating seeds have been reported (Matsukura et al., 2000; Scofield et al., 2007a). Recent evidence from 6-CF feeding suggests that in wheat, the major barrier to symplasmic movement of sugars from the endosperm to the shoot and root is the scutellar epidermis (Aoki et al., 2006). A single symplasmic domain is then present from the scutellar ground cells to the phloem, shoot and root. TaSUT1 is expressed in the scutellar epidermis and ground cells, in addition to the protophloem/phloem SEs. There is good evidence that sucrose is synthesized in the scutellum from glucose released from starch degradation in the endosperm (Edelman et al., 1959). A model of sugar transport during cereal seed germination is shown in Fig. 28.6. Glucose or maltose are transported actively across the scutellar epidermis, then sucrose synthesized in the scutellum can move either symplasmically or apoplasmically (via SUT) to the phloem tissues. Unloading in the root and shoot is assumed to be symplasmic. It is intriguing that TaSUT1 is expressed in the scutellum epithelial cells which are in contact predominantly with glucose and maltose during the germination process (Aoki et al., 2006). In many species, SUTs will also support maltose transport (see Lemoine, 2000), raising the possibility that SUT proteins are also responsible for maltose uptake into the embryo from the endosperm.

Fig. 28.6.
figure 6

Schematic model of the cellular route of transport in the scutellum of germinating cereal seeds, highlighting potential location of sugar transporters and symplasmic transport. The products of starch degradation in the endosperm, glucose (Glc) and maltose (Mal), are taken up into scutellum via sugar porters, and then converted into sucrose in scutellar tissues. A hexose/H+ symporter and a sucrose/H+ symporter, located in the plasma membrane of scutellar epidermis and ground cells, may be engaged for the uptake of glucose and maltose, respectively. Sucrose (Suc) synthesized can be loaded into SE/CC complex either symplasmically or apoplasmically, for translocation to shoot and root apices. In the scutellar vascular bundles, a sucrose/H+ symporter localizes to SE/CC complex and may play a role in the apoplasmic loading of sucrose, or retrieving sucrose leaked to phloem apoplasm. Symbols are same as in Fig. 28.1. SPS, sucrose-phosphate synthase; Hex-P, hexose phosphate.

V. Sucrose Transporters as Sucrose Sensors?

There has been a great deal of controversy over the potential “dual function” of SUTs as sugar sensors and sugar transporters (Chiou and Bush, 1998; Lalonde et al., 1999; Barker et al., 2000; Schulze et al., 2000; Barth et al., 2003). Claims that SUTs could act as sucrose sensors gained impetus with the cloning of the dicot SUT2 type transporters, located in SEs of tomato, A. thaliana and common plantain (Barker et al., 2000; Schulze et al., 2000; Barth et al., 2003). These SUTs are characterized by extended cytoplasmic domains which show some similarity to the SNF3 and RGT2 yeast sugar sensors (Barker et al., 2000). The solanaceous SUT2 proteins were capable of little or no sucrose transport after heterologous expression in yeast (Barker et al., 2000; Reinders et al., 2002). These observations, while rather circumstantial in nature, were later supported by evidence that SUT2 co-localized with SUT1 and SUT4 in phloem SEs in potato and SUT1, 4 and 2 show some ability to form homo-oligomers and interact with each other in an in vitro system (Reinders et al., 2002). Doubt has been cast on these conclusions, however, as in common plantain, not only did the SUT2 homolog PmSUC3 not co-localize with the putative phloem loading sucrose porter PmSUC2, but carried out high rates of sucrose transport in yeast cells (Barth et al., 2003). Whether the Solanaceae are unique or data have been subject to overenthusiastic interpretation is yet to be determined.

VI. Concluding Remarks

Sucrose transport in higher plants has been and remains to be a “hot” area of research in plant biology, characterized by a great deal of controversy and debate. Perhaps a more interesting way of dealing with concluding comments from this chapter is to frame some questions, answers and unresolved conundrums. A few are set out below.

How is the phloem loaded? Apoplasmically by SUTs in the easily transformed dicot species but which class of SUT is responsible is still unclear. Loading mechanisms in graminaceous species are still unresolved and the relative role of plasmodesmata in this process remains a mystery. Why are SUTs so ubiquitous throughout plant tissues, both source and sink, which both export and import sucrose, when SUTs are traditionally regarded as sucrose importers? How is the phloem unloaded? Once again, there is often a clear symplasmic pathway in many sink tissues but SUTs are still expressed in these tissues and presumably can function in retrieval to keep sucrose “on track” and out of the apoplasm. How is sucrose released to the apoplasm, via SUTs or specific effluxers? What does the tonoplast sucrose importer/exporter look like? Do vesicle-mediated transport systems, namely exo- and endocytosis, contribute to apoplasmic transport of sucrose in vivo? Do SUTs really form homo- and hetero-oligomeric complexes in vivo and can they act as sensors analogous to the yeast glucose sensors? Why do so many A. thaliana sugar transport knock-outs (including the putative sucrose sensor mutants) show no phenotype and potentially a high level of redundancy? How do higher plants control SUTs and the bulk-flow of phloem sap?

Rarely has an area of research progressed so quickly from a molecular viewpoint but thrown up so many challenges to our understanding of such an important process in higher plants.