Professional Documents
Culture Documents
n higher plants, CO2 fixation occurs predominantly in mesophyll cells of mature leaves. These are net exporters of sugars and are known as carbon sources. Heterotrophic cells in roots, reproductive structures, storage and developing organs rely on a supply of sugars for their nutrition; these are known as carbon sinks (net importers). Mechanisms must exist to ensure that all sink tissues receive an adequate supply of sugars for growth and development, and sugar transporters play a pivotal role in the membrane transport of sugars and their distribution throughout the plant (Fig. 1). Long-distance transport of carbohydrates between sources and sinks occurs in specific cells of the vascular system, the phloem sieve elements (Fig. 1). During their development, sieve elements extend longitudinally, degrade most of their organelles, including vacuoles, ribosomes and nuclei, and form large, plasma membrane-lined pores in their terminal cell walls, the sieve plates. The loss of organellar structures in sieve elements is paralleled by an extreme increase in organelle density in the phloem companion cells. Sieve elements and companion cells are intimately connected by numerous branched plasmodesmata and form the socalled sieve elementcompanion cell complex (SECCC). An important function of companion cells is the supply of energy and proteins to the sieve elements. Phloem sap in sieve elements moves by bulk flow with rates of up to 1 m per h. At the source, accumulation of sugar inside the SECCC (at a concentration of several hundred mM to 1 M) results in the osmotic uptake of water, forcing sap to flow along the sieve tube. Unloading of sugar at the sink results in loss of water from the sieve tube and thus the gradient of pressure is maintained. Sucrose represents the main form of reduced carbon transported in sieve elements, although some plants can also transport raffinose, stachyose and polyols. These solutes are less subject to metabolism than glucose. After its synthesis, sucrose can pass the entire route from the mesophyll cells to the sieve
elementcompanion cell in the symplast, moving from cell to cell via plasmodesmata (symplastic loading; Fig. 1). However, frequently, sucrose is released from the mesophyll cells and actively loaded from the apoplast into the SECCC (apoplastic loading). Plasma membrane sucroseH symporters are fundamental to this process. Unloading of sucrose at the sinks might also occur either symplastically or apoplastically. The route depends on species (both pathways might operate in the same organism), organ or tissue, and developmental stage. Sink cells either import sucrose from the apoplast directly via sucrose transporters or, alternatively, sucrose can be hydrolysed to glucose and fructose by cell-wallbound invertases and taken up via monosaccharide transporters. Plants appear to have several monosaccharide and disaccharide transporters to coordinate sugar transport in diverse tissues, at different developmental stages and under varying environmental conditions. The nomenclature in the literature for plant sugar transporters has become rather confusing with SUT and SUC being used for disaccharide transporters, and STP, MST, HEX and ST being used for monosaccharide transporters. Here we will refer to disaccharide transporters as DSTs and refer to monosaccharide transporters as MSTs. Some of the physiological functions of these sugar transporters, as suggested from work at the molecular level, will be discussed. Other proteins that have been implicated in the transport of sugars [e.g. sugar-binding protein, SBP (Ref. 1)] will not be considered here.
Multi-gene families for sugar transporters
In Arabidopsis, 14 putative MST genes have been identified to date2 and at least seven DST genes can be found in Arabidopsis data libraries. Heterologous expression in yeast or Xenopus oocytes has been carried out for some of the family members, and evidence suggests that they function as proton symporters38. Generally the DSTs are highly specific for sucrose, although some have been shown to transport maltose9. Their Km values are in the
July 2000, Vol. 5, No. 7
1360 - 1385/00/$ see front matter 2000 Elsevier Science Ltd. All rights reserved. PII: S1360-1385(00)01681-2
283
P P
Sieve elements
Water
Companion cell
ATP H H+ S
+
Source ADP+Pi S
S H+
In apoplastic phloem loading, a DST protein at the plasma membrane of phloem cells accuS mulates sucrose in the SECCC to drive longdistance transport (Figs 1 and 4). In some Path plants these sucrose transporters have been S S detected in sieve elements (Ref. 12), whereas + H others have been observed in companion G+F cells14,15. This might be explained by species + + H G S H differences, but recent work in Plantago has shown that there are sucrose transporters speInvertase cific to either companion cells or sieve elS S ements (R. Stadler et al., unpublished). All of these transporters result from gene transcripP tion in the companion cell, and it will be Sink Heterotrophic important to determine the signals and mechsink cell anisms that regulate their final destination. In Solanaceae, SUT1 (DST) mRNA is localized in the sieve elements, suggesting that tranWater scription occurs in the companion cell and Trends in Plant Science might be followed by the transportation of mRNA through the plasmodesmata and transFig. 1. The contribution of sugar transporters to long-distance sugar transport between sources lation within the sieve elements12. and sinks. Long-distance transport of sucrose (S) between sources and sinks occurs in sieve In Solanaceous species, antisense studies elements of the phloem, which constitute part of the vascular system (path). Sucrose is transported in the phloem sap by bulk flow. In the source leaf, sucrose can pass the entire route from have shown that SUT1 (DST) is indeed the mesophyll cells (source cell) to the sieve elementcompanion cell in the symplast, moving essential for phloem loading and long-disfrom cell to cell via plasmodesmata (P). However, in many species, sucrose leaves the symplast tance transport16,17. Potato plants with drastiat some point, possibly via sucrose efflux carriers (red circle), and is actively accumulated from cally lowered expression of SUT1 (DST) the apoplast into sieve elements and/or companion cells by plasma membrane protonsucrose had a reduced tuber yield (size and number), symporters (purple and dark-blue circles, respectively). Energization is via the proton-motive and mature leaves accumulated sugars force generated by the plasma membrane H-ATPase (black square). Passive influx of water because of impaired export16,17. This was into the sieve tubes presumably occurs via water channels. Expression of sucrose carriers confirmed in tobacco, where strong anti(light-green circle) along the path has been related to a function for retrieving sucrose leaked sense plants had a dwarf phenotype, and from the phloem. Unloading of sucrose into sink cells might occur symplastically via plassugar export from leaves was also drastimodesmata or sucrose might be delivered to the apoplast via a sucrose efflux carrier (pink circle). Here, it is either taken up directly by plasma membrane sucrose transporters (yellow cally impaired18. In wild-type plants, circle), or first hydrolysed to glucose (G) and fructose (F) by the cell-wall invertase (dark-green lengthy dark exposure leads to starch degraelipse) then taken up via plasma membrane monosaccharide transporters (light-blue circle). dation in leaves and subsequent exportation of sucrose to sink organs. However, in antisense plants, starch remains in the leaves 0.31.0 mM range9. The MSTs have broader substrate specificity, and this is related to impaired sucrose export. Phloem-localized transporting a range of hexoses and pentoses with Km values for the SUT1 (DST) homologues might play an additional role in both the preferred substrate being typically 10100 M (Ref. 2). MSTs and retrieval of sucrose leaking from sieve elements along the transloDSTs share little homology at the amino acid level, although there cation pathway and possibly in catalysing the export of sucrose are structural similarities (Fig. 2). They are thought to be members of from the phloem in sinks19 (Fig. 1). Moreover, the same transporters the major facilitator superfamily10,11. Whereas DST genes appear to are not always confined to the phloem but are also present at lower be specific for plants, numerous genes that are homologous to the levels in other leaf cells where they might function in retrieval17,20. MSTs are found in bacteria, fungi and mammals (Fig. 3).
Sugar transporters in roots
Diverse locations for sugar transporters
Plant cells appear to possess sugar transporters tailored to meet their specific requirements at particular stages of development
284
July 2000, Vol. 5, No. 7
Roots represent heterotrophic carbon sinks and can be divided into zones of cell division, elongation and maturation. AtSTP4 (MST), a monosaccharide transporter from Arabidopsis, is found
(a)
Disaccharide-H+-symporter (PmSUC2)
470
170 90 270
350
440
510
Cytoplasmic side
(b)
50
Monosaccharide-H+-symporter (NtMST1)
20
The developing embryo is symplastically iso410 lated from maternal tissues and is reliant on 480 230 the import of sugars from the surrounding apoplastic space. Sugars are required for embryo development and for the deposition 1 of storage compounds necessary for germi270 nation. DSTs and MSTs have been identified Cytoplasmic side in Vicia faba, Pisum sativum and Hordeum vulgare (Refs 2426). In Vicia, VfSTP1 523 (MST) is expressed in the embryo cotyledon epidermal cells. Expression is confined to epiTrends in Plant Science dermal regions covering the mitotically active Fig. 2. Topological models of the disaccharideHsymporter, (a) PmSUC2 [disaccharide parenchyma, indicating that this transporter 24 transporter (DST)] and (b) the monosaccharideH symporter, NtMST1 [monosaccharide serves to supply substrate for metabolism . transporter (MST)]. MSTs and DSTs have predicted molecular masses of ~55 kDa and During later stages of embryo development, hydrophobicity analysis indicates that they are highly hydrophobic integral membrane proVfSTP1 (MST) is replaced by the sucrose teins. They share little homology at the amino acid level, although there are structural simi24 transporter VfSUT1 (DST) . VfSUT1 (DST) larities and they are thought to be members of the major facilitator superfamily. In this family, expression is highest in epidermal cells with it is proposed that the twelve transmembrane -helical structure has evolved from an ancestransfer-cell morphology and with storage tral six-transmembrane-segment transporter gene following gene duplication and fusion10. activity in the underlying parenchyma, indiFamily members generally possess a large central loop between transmembrane domains six and seven (the length of this loop varies in members of the DST and MST families). cating that the encoded transporter is responsible for providing substrate for the synthesis of storage compounds24. Certain seeds accumulate storage reserves in the endosperm dur- is transferred symplastically to the phloem region. The high expresing development. Upon germination, these are hydrolysed and the sion of RcSUT1 (DST) in the phloem indicates that sugars are products, including sugars, are released into the apoplastic space released into the apoplast before active loading by this transporter. (Fig. 5). In germinating seeds of Ricinus communis, a sucrose carrier, RcSUT1 (DST), is expressed at high levels in both the coty- A variety of roles for sugar transporters in floral organs ledon epidermal cells that are situated adjacent to the endosperm Developing pollen grains are strong sinks that require carbohyand also in the phloem13. This suggests that the epidermal cells ini- drate import from the apoplast during maturation, germination tially accumulate sucrose from the apoplast and from here sucrose and growth. The Arabidopsis monosaccharide transporter gene,
July 2000, Vol. 5, No. 7
285
(a)
EcXyIE SspGTR
100
(b)
EcAraE Prokaryotes EcGaIP AtERD6 HsGLUT1 Mammal ScHXT2 100 46 ScGAL2 98 ScHXT1 100 99 Fungi ScHXT3 AmMST1 96 ScSNF3 100 ScRGT2 CkHUP2 97 CkHUP1 Lower plants 100 CkHUP3 NtMST1 VvMST1 42 48 RcSTC 30 AtSTP1 100 18 AtSTP12 VfMST1 99 95 MtMST1 AtSTP4 76 PhPMT1 100 64 AtSTP11 78 99 AtSTP10 100 AtSTP9 RcHEX6 100 77 AtSTP3 SspSGT2 100 RcHEX1 90 AtSTP7 89 97 PaMST1 100 AtSTP14 81 AtSTP13 100 LeMST1 AtSTP2 75 AtSTP6 100 AtSTP8 EcMeIB AtSUC3
100 100 54 99 97
100
Arabidopsis
Monocots
20
HvSUT1 OsSUT1
71
Arabidopsis
PmSUC1
100
40
21 84 95 56 29 100
PmSUC2
VfSUT1
Higher plants
35 71 100
AtSUC2 AtSUC13
100 84 66
Arabidopsis
100
DcSUT2 AgSUT1
Fig. 3. Phylogenetic analysis for (a) monosaccharide transporters (MSTs) and (b) disaccharide transporters (DST). (a) Sugar-transport proteins homologous to lower and higher plant monosaccharide transporters are found in prokaryotes, fungi and mammals, indicating that the corresponding genes form an ancient gene family. Monosaccharide transporters from one species, such as the Arabidopsis AtSTPs, do not form a single subcluster within the higher plant subfamily, suggesting that several independent genes already existed at the beginning of or early in higher plant evolution. The less closely related Arabidopsis AtERD6 protein (Accession no. D89051), described as a putative sugar transporter, clusters with the human glucose facilitator. (b) Proteins closely related to higher plant sucrose transporters are not found in bacteria, fungi or mammals, indicating that plant sucrose-transporter genes evolved comparatively late. Sucrose transporters from Arabidopsis cluster together with sucrose transporters from monocots and from plants with symplastic phloem loading. A functionally uncharacterized transporter from Schizosaccharomyces pombe (Accession no. Z99165) shows some similarity, but only in the N-terminal half of the protein. EcMelB (the outgroup) is extremely distantly related to plant sucrose transporters45. Bootstrap values given at internal nodes indicate the percentage of the occurence of these nodes in 100 replicates (maximum parsimony) of the data set. The sequence of the E. coli xylose permease (EcXylE; black) and of the E. coli melibiose transporter (EcMelB; black) were used as outgroups in (a) and (b), respectively. Accession numbers for nonplant monosaccharide transporter sequences are AmMST1 (Amanita muscaria monosaccharide transporter), Z83828; EcAraE (E. coli arabinose transporter), J03732; EcGalP (E. coli galactose permease), AE000377; EcMelB (E. coli melibiose transporter), K01991; EcXylE (E. coli xylose transporter), P09098; HsGlut1 (Homo sapiens glucose transporter), 87529; ScHXT1 (Saccharomyces cerevisiae hexose transporter), M82963; ScHXT2 (S. cerevisiae hexose transporter), M33270; ScHXT3 (S. cerevisiae hexose transporter), L07080; ScRGT2 (S. cerevisiae hexose transporter-like glucose sensor), Z74186; ScSNF3 (S. cerevisiae hexose transporter-like glucose sensor), J03246; SspGTR (Synechocystis sp. glucose transporter), X16472. Maximum parsimony (MP) bootstrap analyses of aligned sequences were conducted with the PHYLIP package 3.572c. Addition of taxa was jumbled and repeated three times for each individual bootstrap replication in the MP analysis. Output treefiles were combined with the CONSENSE program and converted into a graphical representation by the TREEVIEW program. Accession numbers of plant sucrose transporters used in this tree are AbSUT1 (Asarina barclaiana, AF191024); AgSUT1 (Apium graveolens, AF063400); AmSUT1 (Alonsoa meridionalis, AF191025); AtSUC1 (Arabidopsis thaliana, X75365); AtSUC2 (A. thaliana, X75382); AtSUC3 (A. thaliana, AC004138); AtSUC4 (A. thaliana, AC000132); AtSUC5 (A. thaliana, AJ252133); AtSUC8 (A. thaliana, AAC69375); AtSUC13 (A. thaliana, AB016875); BvSUT1 (Beta vulgaris, X83850); DcSUT1 (Daucus carota, Y16766); DcSUT2 (D. carota, AB036758); HvSUT1 (Hordeum vulgare, Accession no. not yet available); HvSUT2 (H. vulgare, Accession no. not yet available); NtSUT1 (Nicotiana tabacum, X82276); OsSUT1 (Oryza sativa, D87819); PmSUC1 (Plantago major, X75764); PmSUC2 (P. major, X75764); SoSUT1 (Spinacea oleracea, X67125); StSUT1 (Solanum tuberosum, X69165); VfSUT1 (Vicia faba, Z93774). Accession nos of plant monosaccharide transporters used in this tree are AtSTP1 (A. thaliana, X55350); AtSTP2 (A. thaliana, AJ001362); AtSTP3 (A. thaliana, AJ002399); AtSTP4 (A. thaliana, X66857); AtSTP6 (A. thaliana, AC013454); AtSTP7 (A. thaliana, AF001308); AtSTP8 (A. thaliana, AF077407); AtSTP9 (A. thaliana, AC007980); AtSTP10 (A. thaliana, AB025631); AtSTP11 (A. thaliana, AB007648); AtSTP12 (A. thaliana, AL022603); AtSTP13 (A. thaliana, AF077407); AtSTP14 (A. thaliana, AC004260); CkHUP1 (Chlorella kessleri, Y07520); CkHUP2 (C. kessleri, X66855); CkHUP3 (C. kessleri, X75440); LeMST1 (Lycopersicum esculentum, AJ010942); MtMST1 (Medicago truncatula, U38651); PaMST1 (Picea abies, Z83829); PhPMT1 (Petunia hybrid, AF061106); RcHEX1 (Ricinus communis, L08197); RcHEX6 (R. communis, L08188); RcSTC (R. communis, L08196); SspSGT2 (Saccharum hybrid, L21753); VfMST1 (V. faba, Z93775); VvMST1 (Vitis vinifera, AJ001061).
286
287
(a)
(b)
Lipids Sucrose
Co
tyl co po y H
t yle
ns do Proteins
Endosperm
Amino acids
(c) (i)
LE
(ii) LE
(iii) LE
Sugars can act as regulatory signals that affect gene expression and hence plant development. In relation to resource allocation, the ability to sense altered sugar concentrations could have important advantages by allowing the plant to tailor its metabolism in source tissues to meet the demands in sinks. There is great interest in Fig. 5. Localization of the sucrose carrier in a germinating Ricinus seedling. Seeds accudissecting the signalling processes that are mulate storage reserves in the endosperm during development. Upon germination, these are involved in sugar sensing and response hydrolysed and the products, including sugars, are released into the apoplastic space. Sevenpathways. There are three possible sugarday-old dark-grown germinating Ricinus seedlings are shown in (a) with (left) and without sensing systems in plants32: (right) an endosperm. (b) Breakdown of lipids and protein in the endosperm and the release of A hexokinase-sensing system. sucrose and amino acids into the apoplast for uptake by the cotyledon. (c) Expression pattern A sucrose-transport-associated sensor. of the sucrose transporter, RcSUT1 (DST)13,46, in the germinating Ricinus seedling (three-day A glucose-transport-associated sensor. old). RcSUT1 transcripts were localized by in situ hybridization using an antisense RcSUT1 cRNA probe. Dark staining, indicating RcSUT1 expression, is seen in both the cotyledon Based on sugar signal transduction processes lower epidermal cells (LE) that are situated adjacent to the endosperm (parts iiii) and also in in Saccharomyces, it has been proposed that the phloem (red arrows in iii). No equivalent staining was observed in sections probed with members of the sugar transporter family in the sense probe (results not shown). The expression pattern indicates that the epidermal cells plants might play a role in sugar sensing33. In initially accumulate sucrose from the apoplast and from here sucrose is transferred symplastiyeast, it has been shown that membranecally to the phloem region. The high level of expression of RcSUT1 in the phloem indicates bound receptors closely related to sugar that sugars are released into the apoplast before active loading by this transporter. Scale bars transporters are used to sense external sugar 100 m in (i) and (ii); 50 m in (iii). (c) reproduced, with permission, from Ref. 13. concentrations and activate a signal transduction pathway leading to the regulation of approximately fourfold in Arabidopsis seedlings infected with transporter gene expression and insertion or degradation of proteins Alternaria brassicicola or Fusarium oxysporum, and also in sus- at the plasma membrane. This allows yeast to respond to a rapidly pension-cultured cells of Arabidopsis treated with bacterial or changing external environment. The yeast glucose sensors, ScSNF3 fungal elicitors21. It has been concluded that AtSTP4 (MST) plays (senses low glucose levels) and ScRGT2 (senses high glucose leva role in the transportation of monosaccharides into the sink els) are homologous to yeast and plant glucose transporters (Fig. 3) tissues and that it responds to an increased demand for carbohy- but do not mediate significant glucose transport34. The large Cdrate by cells under stress conditions21. Induction of AtSTP4 terminal extensions of these proteins are required for the trans(MST) expression is also observed during plant and fungal mission of the glucose signal. To date, there is no direct evidence for biotrophic interaction, for example, Arabidopsis infected with a role of plant monosaccharide and disaccharide transporters as the powdery mildew Erysiphe cichoracearum31. Expression of sensors. Nevertheless, the finding that several members of the plant AtSTP4 (MST) is induced dramatically in source leaves infected MST and DST families have unusual cytoplasmic extensions, which with mildew and, although the fungus is confined to epidermal are predicted to be cytosolic, has initiated studies in relation to the cells, AtSTP4 (MST) expression is induced in many cell types possible role of these proteins in sugar sensing33. within the leaf following infection (M. Gilbert et al., unpublished). The signalling pathway by which this occurs is unknown, and the Regulation of sugar transporters role of this transporter in this interaction is unclear. It might help As sugar transporters play such a key role in sourcesink interacto increase carbohydrate import into infected tissues, in order to tions, it is likely that they are tightly controlled. There are several
288
July 2000, Vol. 5, No. 7
affecting sourcesink relations (e.g. following application of phytohormones and elicitors, pathogen attack and wounding)44 but further work is required to investigate the coordination of events. In sink cells, it is important to understand why, in some cases, sucrose is cleaved by a cell-wall invertase and therefore hexose carriers are involved, whereas in other cases sucrose is taken up and cleaved inside the cells. In germinating pollen, both monosaccharide and disaccharide transporters are present2830, although sucrose is the preferred carbon source for in vitro germination and hexoses are inhibiting. Clearly, work is needed for a full comprehension of the respective role of each carrier type in such cells. We need to clarify which mechanisms exist for the efflux of sugars from cells, for example, for efflux from the mesophyll or from apoplastically unloading phloem cells. Under nonenergized conditions, influx carriers might be operating in efflux, allowing sugars to be transported down their concentration gradient; alternatively, different transporters might be involved. In addition, the identification of tonoplast sugar transporters is important for understanding storage and remobilization from the vacuole. We must also consider that entirely different mechanisms might exist for sugar transport, such as those involving ABC transporters. More evidence is required on the physiological function of sugar transporters to support the information from expression studies. In particular, no antisense studies have been reported for any of the monosaccharide transporters. We eagerly await studies on knockout mutants to dissect the transport processes in greater detail. The phenotype of such plants will provide clues concerning the physiological role and importance of particular transporters. In addition, plants with transporter genes under the control of cellspecific promoters will enable us to learn more about sourcesink interactions and pathways of sugar movement. Sensing functions are essential in allowing plants to adapt to changing environments. One of the major challenges will be to identify sugar-sensing mechanisms and the signal transduction pathways. It will be important to ascertain whether members of the sugar-transporter families are involved in this function.
Acknowlegements
Our work has been supported by a grant from the European Commission DG XII Biotechnology programme (ContractBIO4-CT960583). L.E.W. is grateful to the Royal Society for support. R.L.s work is supported by the CNRS, the French Ministry for Education and Research and by the Rgion Poitou Charentes. N.Ss work was supported by the Deutsche Forschungsgemeinschaft. The help of Volker Huss with Figure 3 and the help of Cyril Maingourd with Figure 4 is gratefully acknowledged.
References 1 Overvoorde, P.J. et al. (1996) A soybean sucrose binding protein independently mediates nonsaturable sucrose uptake in yeast. Plant Cell 8, 271280 2 Bttner, M. and Sauer, N. (2000) Monosaccharide transporters in plants: structure, function and physiology. Biochim. Biophys. Acta. 1465, 263272 3 Sauer, N. et al. (1990) Primary structure, genomic organisation and heterologous expression of a glucose transporter from Arabidopsis thaliana. EMBO J. 9, 30453050 4 Sauer, N. and Stolz, J. (1994) SUC1 and SUC2: two sucrose transporters from Arabidopsis thaliana; expression and characterization in bakers yeast and identification of the histidine-tagged protein. Plant J. 6, 6777 5 Boorer, K.J. et al. (1994) Steady-state and presteady-state kinetics of the Hhexose cotransporter (STP1) from Arabidopsis thaliana expressed in Xenopus oocytes. J. Biol. Chem. 269, 2041720424 6 Boorer, K.J. et al. (1996) Transport mechanism of the cloned H/sucrose cotransporter StSUT1. J. Biol. Chem. 271, 2513925144
To comprehend sourcesink regulation fully and to be in a position to manipulate the complex interactions for agricultural benefit, it is vital that the underlying membrane transport processes are thoroughly characterized. This includes determining precisely the range of sugar transporters that are involved, their structurefunction relationships, and defining their cellular and temporal expression pattern. A particular goal would be the ability to manipulate competition between sinks. For this we need to find out how sugar carriers are regulated in metabolic and storage sinks and to relate this to the information available on the activity of the sucrose metabolizing enzymes, invertase and sucrose synthase. Extracellular invertase is up-regulated in several situations
289
Lorraine Williams* is at the School of Biological Sciences, University of Southampton, Bassett Crescent East, Southampton, UK SO16 7PX; Remi Lemoine is at the Laboratoire de Biochimie et Physiologie Vgtales, ESA CNRS 6161, Btiment Botanique, 40 Avenue du Recteur Pineau, F-86022 Poitiers Cedex, France (tel 33 5 49454183; fax 33 5 49454186; e-mail remi.lemoine@campus.univ-poitiers.fr); Norbert Sauer is at the Dept of Botany II, Molecular Plant Physiology, University of Erlangen, Staudtstrasse, D-91058 Erlangen, Germany (tel 49 9131 85 28212; fax 49 9131 85 28751; e-mail nsauer@biologie.uni-erlangen.de). *Author for correspondence (tel 44 2380 594278; fax 44 2380 594319; e-mail lew@soton.ac.uk)
Students
Did you know that you are entitled to a 50% discount on a subscription to Trends in Plant Science? See the subscription order card in this issue for details.
290