Phloem-localized, Proton-coupled Sucrose Carrier ZmSUT1 Mediates Sucrose Efflux under the Control of the Sucrose Gradient and the Proton Motive Force*

The phloem network is as essential for plants as the vascular system is for humans. This network, assembled by nucleus- and vacuole-free interconnected living cells, represents a long distance transport pathway for nutrients and information. According to the Münch hypothesis, osmolytes such as sucrose generate the hydrostatic pressure that drives nutrient and water flow between the source and the sink phloem (Münch, E. (1930) Die Stoffbewegungen in der Pflanze, Gustav Fischer, Jena, Germany). Although proton-coupled sucrose carriers have been localized to the sieve tube and the companion cell plasma membrane of both source and sink tissues, knowledge of the molecular representatives and the mechanism of the sucrose phloem efflux is still scant. We expressed ZmSUT1, a maize sucrose/proton symporter, in Xenopus oocytes and studied the transport characteristics of the carrier by electrophysiological methods. Using the patch clamp techniques in the giant inside-out patch mode, we altered the chemical and electrochemical gradient across the sucrose carrier and analyzed the currents generated by the proton flux. Thereby we could show that ZmSUT1 is capable of mediating both the sucrose uptake into the phloem in mature leaves (source) as well as the desorption of sugar from the phloem vessels into heterotrophic tissues (sink). As predicted from a perfect molecular machine, the ZmSUT1-mediated sucrose-coupled proton current was reversible and depended on the direction of the sucrose and pH gradient as well as the membrane potential across the transporter.

To ensure adequate partitioning of sucrose throughout the plant body, sucrose has to be translocated from the mesophyll cells to the sieve element-companion cell complex. Because of the energy-dependent sucrose/H ϩ symporter in apoplasmic loading plant species, the transport sugar accumulates at concentrations of several hundred mM to Ͼ1 molar in the conduct-ing vascular cells. The hydrostatic pressure difference between source and sink tissues drives the mass flow of water and nutrients in the phloem vessels (1). In sink tissues, which are dependent on carbon supply via the phloem, a symplasmic unloading of sucrose along its concentration gradient has been shown for many plant species (2). Interestingly, however, sucrose/H ϩ symporter transcripts and proteins have also been localized in sink tissues, suggesting a role in sink loading/ retrieval or unloading of sucrose via these transporters (see Ref. 2 for review). SUT1 from the potato, for example, has been detected in the sieve elements of mature source leaves as well as in developing sink leaves, roots (3), and tubers (4,5). Using a sink-specific antisense inhibition for SUT1 under the control of a tuber-specific promoter, Kü hn et al. (2003) (4) could demonstrate the involvement of SUT1 in early tuber development and, thus, phloem unloading. Further evidence for a sucrose export system was added by the localization of sucrose/H ϩ symporters expressed in symplasmically isolated tissues such as developing embryos (6,7) and growing pollen tubes (8). Furthermore Evert and Russin (1993) could show, for example, that symplasmic unloading in maize is unlikely because of the lack of plasmodesmata in the protophloem and metaphloem of developing leaves (9). Although proton-coupled sucrose carriers have been localized to the sieve tubes and companion cell plasma membranes of both source and sink tissues, the molecular representatives and mechanism of the sucrose phloem efflux is still scant.
In the present study we tested the biophysical properties and thermodynamics of ZmSUT1, a maize sucrose carrier expressed at a high level in Xenopus laevis oocytes. Specific sucrose transport inhibitors are not available, but Xenopus oocytes, like all other creatures apart from plants, do not transport sucrose. Therefore oocytes are well suited for sucrose transport studies. Using this sucrose-insensitive system we could demonstrate that this sugar carrier mediates both sucrose uptake and release. Upon a drop in membrane potential and/or pH gradient, ZmSUT1 would release sucrose from, for example, sink phloem and thus seem to represent the molecular equivalent for the sucrose efflux carrier. Based on our biophysical characterization of ZmSUT1, the "source and sink mode" of this transporter is discussed in respect to in planta phloem loading and unloading conditions. In principle, each individual transporter should be reversible. But, in contrast to the reversible transporters from the animal field and from bacteria that have been described to date, ZmSUT1 is the first that works in both directions under physiological conditions.

MATERIALS AND METHODS
Aphid Breeding-Aphids of the species Rhopalosiphum padi were bred on barley and maize grown in a climate chamber under a 14-h photoperiod.
Experimental Setup-Plant aphid cages were applied to the mature leaves of a 4-week-old potted maize. Aphids feeding on a leaf were dissected from their stylets using a laser as described previously (10). The recording electrodes were brought in contact with the phloem exudate appearing at the cut end of the stylet. The leaf was cut 15 cm proximal to the tip, and the cut end was incubated with artificial pond water containing the reference electrode (silver/silver chloride) and 1 mM NaCl, 0.1 mM KCl, 0.1 mM CaCl 2 , 100 mM sorbitol, and 1 mM MES, 1 adjusted to pH 6.0 with Tris. Sucrose pulses were applied by perfusion of artificial pond water solution. Phloem potential measurements were recorded according to (10).
Two-electrode Voltage Clamp (TEVC) Analysis in Xenopus Oocytes-ZmSUT1 cRNA was prepared using the mMESSAGE mMACHINE TM RNA transcription kit (Ambion Inc., Austin, TX). Oocyte preparation and cRNA injection have been described elsewhere (11). In TEVC studies, oocytes were perfused with a standard solution containing 30 mM KCl, 1 mM CaCl 2 , and 1.5 mM MgCl 2 based on Tris/MES buffers for pH values from 5.6 to 8.0 or based on citrate/Tris buffers for the pH values 4.5 and 5.0. The sucrose concentrations and pH values are indicated in Figs. 2-4 and 6 (and the corresponding legends) and throughout the text where noted. All solutions were adjusted to 220 mosmol kg Ϫ1 using D-sorbitol. Steady state currents were obtained by stepping the membrane potential from the holding potential of 0 mV to a series of 500-ms test pulses from 60 to Ϫ130 mV in 10-mV decrements. Difference currents were calculated by subtracting the currents in the absence of sucrose from the currents in its presence. The sucrose-induced steady state currents were measured in respect to ligand concentrations and membrane potential. At each test potential the currents were fitted to the Michaelis-Menten equation shown in Equation 1, where the substrate (S) is either [sucrose] or [H ϩ ]. These fits yielded in the maximal currents I max S for sucrose and I max H for H ϩ and the half-maximal ligand concentrations K m H for H ϩ and K m S for sucrose. Intracellular pH Measurements-PH-sensitive microelectrodes were pulled from borosilicate capillary (TW100F-3; WPI, Sarasota, FL) using a laser puller (P2000; Sutter Instruments, Novato, CA) and silanized with dimethyldichlorosilane (Fulka, Steinheim, Germany) at 200°C for 15 min. The tips of the pH microelectrodes were filled with hydrogen ionophore I mixture B (Fulka) and then back-filled with a buffer containing 40 mM KH 2 PO 4 , 23 mM NaOH, and 150 mM NaCl (pH 6.8). Only electrodes with a linear slope of 55-60 mV/pH unit over the calibration range before and after measurement were used. Signals were recorded with an electrometer (Model FD 223; WPI) in parallel to the currents in the voltage clamp mode of a TEVC amplifier (Turbo TEC 10CD; npi electronic GmbH, Tamm, Germany). On the basis of the calibration curve for the pH microelectrodes, the internal pH (pH i ) of the oocytes was calculated in consideration of the membrane potential.
14 C Sucrose Uptake Experiments-In each experiment, 10 ZmSUT1injected oocytes or 10 control oocytes were incubated in 0.05 Ci/ml 14 C sucrose with a final sucrose concentration of 5 mM in the standard solution at pH 5.6. At defined time points the oocytes were rapidly washed three times in ice-cold standard solution and transferred to liquid scintillation vials containing scintillation mixture (Emulsifier-Safe TM ; Packard, Meriden, CT). The 14 C radioactivity was counted in a liquid scintillation analyzer (Model 1900CA; Packard), and the sucrose uptake per oocyte was calculated from three independent experiments for each time point.
[ 14 C]Sucrose Release Experiments-Control oocytes and ZmSUT1injected oocytes were loaded with 0.5 Ci of radiolabeled sucrose with a final sucrose concentration of ϳ50 mM by injection (Picospritzer TM II; General Valve Co., Fairfield, NJ). After a 10-min washing period in ice-cold ND96, each single oocyte was transferred into 200 l of the standard solution at pH 5.6 or pH 5.6 in the presence of 10 mM acetate. After 2 h the 14 C radioactivity of the incubation-solution was measured in a scintillation counter. The oocytes were rapidly washed in ice-cold standard solution and transferred to the scintillation mixture for counting the 14 C radioactivity in the liquid scintillation analyzer. The percentage of sucrose release was calculated.
Patch Clamp Measurements-Giant patch recording (12) was performed in inside-out configuration on ZmSUT1 expressing Xenopus oocytes. Borosilicate glass pipettes were pulled and fire-polished to have a final tip with a diameter between 25 and 30 m. Oocytes were bathed in an external solution consisting of 30 mM KCl, 1 mM CaCl 2 , 1.5 mM MgCl 2 , 1 mM GdCl 3 , 145 mM sorbitol, and 10 mM MES/Tris, pH 5.6. After the seal was obtained, the external solution was changed (30 mM KCl, 1 mM EGTA, 2 mM MgCl 2 , 145 mM (or 500 mM) sorbitol, and 10 mM Tris/MES, pH 7.5), and the patch was excised. The recording pipette was then placed in front of a polyethylene tube in connection with the desired ionic solutions that were driven by gravity. The standard cytosolic solution contained 30 mM KCl, 1 mM EGTA, and 2 mM MgCl 2 . The cytosolic sucrose concentration ranged from 0 to 500 mM, as indicated in the remaining text where noted; sorbitol was appropriately added to each cytosolic solution to have a total sugar concentration of 500 mM. Cytosolic pH was 7.5 or 5.6 (with 10 mM Tris/MES or MES/Tris). The standard pipette solution was 30 mM KCl, 1 mM CaCl 2 , 1.5 mM MgCl 2 , 145 mM sorbitol, and 10 mM MES/Tris, pH 5.6; sucrose was added at concentrations of 0.5, 5, and 50 mM as indicated in the text where noted. Currents, filtered at 10 or 100 Hz and sampled at 200 or 400 Hz, were recorded with an EPC9 amplifier using Pulse 8.3 software (Heka Elektronic GmbH, Lembrecht, Germany). Data were analyzed by custommade programs using Igor (Wavemetrics; Lake Oswego, OR).

RESULTS
ZmSUT1 was isolated from maize and expressed in source and sink tissues such as mature leave blades and sheaths as well as pedicels and seeds (13). High sequence homologies to the rice sucrose transporter OsSUT1 (14) and to known sucrose transporters from the dicot species group ZmSUT1 into the SUT2 subfamily of sucrose transporters with high sucrose affinity (for review, see Ref. 15). ZmSUT1 is a member of a large family of membrane proteins mediating the transport of sugars, amino acids, and osmolytes across membranes. These carriers share the typical 12-transmembrane-spanning ␣-helix structure (16,17). In most eukaryotic cells these transporters couple the uptake of their substrates to electrochemical ion gradients generated by the H ϩ -or Na ϩ /K ϩ -ATPase.
Using the aphid stylet technique on maize leaf blades, it could be shown that the addition of sucrose reversibly depolarized the phloem potential (Fig. 1). To elucidate the transport characteristics of the underlying sucrose/H ϩ transporter activity with respect to sucrose affinity gradients and proton motive force, we heterologously expressed ZmSUT1 in Xenopus laevis oocytes. Functional analysis was performed using both the TEVC technique and the patch clamp technique. Oocytes expressing Zm-SUT1 efficiently imported radio-labeled sucrose with uptake rates of 6 nmol per hour and oocyte, whereas non-injected oocytes did not accumulate sucrose in detectable amounts ( Fig. 2A). To monitor the movement of protons accompanying the sucrose transport, we simultaneously recorded sucrose-induced ionic currents and changes in cytoplasmic pH i by TEVC and protonselective microelectrodes (18). Upon the addition of sucrose to the external solution, large inward currents were elicited (Fig. 2B, upper trace). Inward currents were accompanied by a decrease in pH i by up to 0.5 units within 10 min (Fig. 2B, lower trace). After the removal of sucrose from the bath medium, the inward currents returned to the pre-sucrose level again, whereas the recovery of pH i was delayed. Control oocytes showed neither sucroseinduced currents nor sucrose-dependent changes in pH i .
Stepwise increases in sucrose concentrations resulted in a gradual rise in ZmSUT1-mediated currents (Fig. 2C). In the current clamp mode, membrane depolarization in response to different sucrose concentrations could be recorded as well (Fig. 2D). Like the current response in Fig. 2C, the degree of membrane depolarization depended on the sucrose concentration applied (up to 50 mV with 10 mM sucrose). When the steady-state currents recorded in presence of extracellular sucrose concentrations between 0.5 and 50 mM were plotted against the membrane potential, ZmSUT1 currents increased upon hyperpolarization and padi sucking on maize with its stylet inserted into a sieve element of a vascular bundle. Bottom right, after the aphid is separated from its stylet by a laser pulse, the stylet stump exuded sieve tube sap to which the tip of a microelectrode was attached (400ϫ). Application of sucrose via the apoplast depolarizes phloem potential, pointing to a proton-coupled cotransporter. Upon removal of sucrose, the membrane potential repolarized. were saturated at 30 mM sucrose (not shown). Plotting the currents as a function of the sucrose concentration a single Michaelis-Menten function could be fitted to the individual, voltage-dependent sucrose saturation curves (Fig. 3A). These currentconcentration curves are hyperbolic in shape, suggesting that just one sucrose molecule binds to the transporter. The apparent affinity constant of ZmSUT1, K m S , exhibited pronounced voltageand pH-dependence (Fig. 3B; compare also Ref. 19). Hyperpolarizing voltages increased the apparent affinity to sucrose from 16 mM at 0 mV to 7.2 mM at Ϫ100 mV and pH 5.6. Upon a change to pH 6.5 the sucrose affinity was reduced. Both K m S voltage curves could be fitted with a single exponential function, allowing us to extrapolate K m S to measured phloem potentials of up to Ϫ180 mV (20). A K m S of 3.7 mM at pH 5.6 and a K m S of 12.4 mM at pH 6.5 were calculated. The maximal carrier current I max S values were found to be voltage-dependent also (not shown), decreasing linearly with negative-going membrane potentials.
To study the proton coupling of ZmSUT1-mediated sucrose transport, the steady-state currents were measured as a function of voltage and pH in the presence of 5 mM sucrose (not shown). As predicted for a proton-coupled transport process, in the pH range between 6.5 and 4.5 ZmSUT1 currents increased with increasing proton concentration and hyperpolarization. At pH values Ͼ7.0 no significant inward currents could be detected. The currents at selected voltages were plotted against the H ϩ concentration (not shown) and fitted by a single Michaelis-Menten equation to calculate K m H and I max H (not shown). The proton affinity K m H of ZmSUT1 exponentially increased with hyperpolarizing membrane potentials (Fig. 3C). This behavior is in line with the results for the sucrose affinities K m S (compare Fig. 3B). Thus, both the apparent affinity constants and the I max values for sucrose as well as for protons decrease upon hyperpolarization.
To study the inverse transport mode of ZmSUT1 and its affinity toward cytosolic sucrose, we applied the giant patch clamp technique to ZmSUT1-expressing oocytes. In the insideout configuration we varied the "cytosolic" sucrose concentration in the presence of either 0.5, 5, or 50 mM extracellular (pipette) sucrose. Upon a stepwise increase in cytosolic sucrose from 0 to 50, 100, 200, and 500 mM in the presence of 50 mM in the pipette, a progressive decrease in inward current was measured (Fig. 4A). This effect was completely reversible; inward currents reached their pre-stimulus levels after the removal of cytosolic sucrose. Non-injected oocytes, however, did not respond to variations in the cytosolic sucrose concentration. When plotting the average currents shown in Fig. 4A as a function of the cytosolic sucrose concentration, data could be fitted by a Michaelis-Menten equation (Fig. 4A, continuous line) characterized by an apparent K m of 160 mM (Fig. 4D). The inset of Fig. 4D depicts the extrapolation of the sucrose-induced currents from 2 to 3 M, a concentration range in which ZmSUT1 currents would reverse direction (I ϭ 0 at 2.38 M sucrose). When the extracellular sucrose concentration was decreased to 5 mM or even 0.5 mM, the ZmSUT1-mediated currents reversed direction at physiological cytosolic sucrose levels (Fig. 4, B and C). In the presence of 5 mM external sucrose, a K m of 278 mM was calculated (Fig. 4E). A rise in cytosolic sucrose concentration above 314 mM even inverted the current direction. Upon a further decrease in extracellular sucrose concentration to 0.5 mM and the absence of cytosolic sucrose, only very small inward currents remained (Fig. 4C). Under these conditions, however, a rise in cytosolic sucrose concentration to just 50 mM inverted the ZmSUT1 current already. From the Michaelis-Menten fit a K m of 362 mM and a zero current value at 31 mM was obtained (Fig. 4F). When plotting the K m values versus the external sucrose concentration, a decrease in K m with the rise in external sucrose concentration became evident (not shown).
Likewise, the cytosolic sucrose concentration causing the ZmSUT1 current to change direction was plotted as a function of external sucrose (Fig. 5). Under the equilibrium conditions depicted in Equation 2, where n suc and n H are the number of moles of sucrose and protons transported through the membrane, ⌬ ϭ suc cyt Ϫ suc ext is the difference between the cytosolic and external chemical poten- tial (or molar free energy) of sucrose, and ⌬ H ϩ ϭ H ϩ cyt Ϫ H ϩ ext is the difference between the cytosolic and external electro-chemical potential of protons. Therefore, Equation 3, shown here, [Suc]  In agreement with a perfectly coupled thermodynamic machine, the positive current in Fig. 4 represents the sucrose gradient-driven efflux of protons against the proton gradient. To study the two transport modes of ZmSUT1 in the absence of the proton motive force, in Fig. 6A we stepped the cytosolic sucrose concentration from 0 to 500 mM (5 mM sucrose in the  Fig. 3, is plotted against the external sucrose concentration. The continuous lines were obtained by the equilibrium equation (Equation 3) with V m ϭ 0, pH cyt Ϫ pH ext ϭ 1.9, and different values for n suc and n H (the thicker line corresponds to a 1:1 stoichiometry of the ZmSUT1 transporter). The same experimental conditions as those in Fig. 3 were used. pipette) in the absence of a pH gradient. With [Suc] cyt ϭ 0 mM and the absence of a membrane potential, we recorded an inward current as expected from the steep inward-directed sucrose gradient. Inverting the sucrose gradient by increasing [Suc] cyt to 500 mM, the carrier current reversed direction. In the presence of an inward-directed pH gradient, however, the magnitude of outward currents was smaller (compare Fig. 4). Inward currents could be re-established again upon removal of the disaccharide. Following a rise in the extracellular sucrose concentration from 5 to 50 mM and the absence of cytosolic sucrose, carrier currents remained inward (Fig. 6B). During bath perfusion to [Suc] cyt ϭ 500 mM, currents changed direction. These experiments indicate that the sucrose gradient can drive the proton flux and vice versa. In the experiment depicted in Fig. 4B, the ZmSUT1 currents were subject of a fast "rundown," most likely due to the loss of regulatory cytosolic factors. Interestingly, in Fig. 6B the decay of both inward and outward currents could be fitted by single exponential functions (dashed lines) with the same time constant. This indicates that both transport modes of ZmSUT1 are perfectly coupled via the sucrose gradient and proton motive force.
Under the conditions of the sink phloem, the sucrose gradient drives the efflux of protons and sucrose. To mimic this situation in the oocyte system in Fig. 6C, ZmSUT1-expressing oocytes were injected with [ 14 C]sucrose (final concentration of 50 mM), and the release of the radioactively labeled sucrose was measured. In ZmSUT1-oocytes, but not in water-injected control-oocytes, pronounced sucrose release was measured. As expected from our thermodynamic assumptions, the sucrose-release was enhanced when the cytosol was acidified by acetate treatment.  (32). The source site of the sieve element (SE)-companion cell (CC) complex is characterized by an outward-directed sucrose and an inward-directed H ϩ gradient. The membrane potential is hyperpolarized because of the activity of the H ϩ -ATPases localized in the companion cells. Under these conditions, sucrose is accumulated in the phloem cells by sucrose/H ϩ symporters like ZmSUT1. In the sink phloem the apoplastic concentrations of sucrose is reduced, and the membrane potential is depolarized to values around Ϫ60 mV. In this region the membrane potential mainly depends on the potassium conductance because of the reduced size (or even absence) of the energy-supplying companion cells. Thus, the proton motive force is decreased. This regime directs Zm-SUT1 into the inverse transport mode, and sucrose is released. F, fructose; G, glucose; S, sucrose.

DISCUSSION
Because of the localization of a sucrose/H ϩ transporter in sink tissues, it has previously been speculated that phloem unloading may be mediated by the same sucrose-H ϩ symporters that are responsible for phloem loading (for example, Ref. 22). The direct demonstration that ZmSUT1, a member of the phloem sucrose carrier family, acts either in the source mode or sink mode for the life-maintaining uptake, and the adsorption of sucrose is underpinned by genetic evidence. Arabidopsis mutants, which lack the ZmSUT1 homologue AtSUC2, are strongly impaired in phloem loading and unloading of sucrose, which results in stunted growth, retarded development, and sterility (23). Phloem unloading of sucrose is required for starch formation in storage tissues, such as the grains of cereals or potato tubers. When the copy number of StSUT1, a ZmSUT1 orthologue expressed in the phloem of developing tubers, is reduced by antisense repression, reduced fresh weight accumulation during tuber development was observed (4,5). Furthermore, indirect measurements with the protoncoupled monosaccharide transporter CkHUP1 from the green alga Chlorella and the SGLT1 Na ϩ /glucose transporter from human and rabbit suggest that these sugar carriers from single-celled organisms can act in the inverse transport mode to release their substrates (24 -27). To study the inverse mode of ZmSUT1, we performed patch clamp experiments in the giant inside-out configuration. Varying the cytosolic sucrose concentration, we were for the first time able to determine the cytosolic affinity constant for sucrose. Upon variation of the sucrose gradient we could reverse the direction of the proton current, e.g. by increasing the cytosolic sucrose concentration. The direction of the transport of the ZmSUT1 symporter is therefore dependent on the sum of the free energies of both the sucrose and the proton gradient across the membrane. In agreement with the above considerations, we could demonstrate that sucrose could drive protons through ZmSUT1. Recently, the reversibility of the human and rabbit Na ϩ /glucose co-transporters has been documented by measuring the reversion of the glucose-coupled Na ϩ current. Like the proton-coupled disaccharide carrier ZmSUT1, the sodium-coupled SGLT1 shows more than one order of magnitude difference between the sugar affinities of the two transport modes, indicating a functional asymmetry of both carrier types. Under physiological conditions the inverse transport mode of SGLT1 is highly improbable because of the low affinity of the sugar carrier. In the plant phloem, however, both transport modes of ZmSUT1 are probable (see model in Fig. 7). In maize source leaves, extracellular sucrose concentrations of 2.6 mM were measured (28). Assuming a pH gradient of ϳ1.5 units and a phloem membrane potential of Ϫ150 mV (29), a perfect proton-coupled ZmSUT1 would allow a theoretical phloem sucrose accumulation of up to 26 M (according to Equation 3, with n H /n suc ϭ 1). Directly measured sucrose concentrations of maize phloem sap revealed sucrose contents of ϳ0.85 M (28). In the sink phloem, however, the conditions are different. The external sucrose concentration is reduced to 1 mM or less because of the activity of cell wallbound invertases (for example, Ref. 30) and the surrounding sucrose taking up (sucking) sink tissues with hyperpolarized membrane potentials negative to Ϫ180 mV. 2 Symplasmic unloading is unlikely in maize because of the lack of plasmodesmata in the protophloem and metaphloem (9). Furthermore, in the region of the release phloem the proton motive force across the phloem membrane is less strong because of the reduced size (or even absence) of the energy-supplying companion cells (29,31). Therefore the phloem membrane potential mainly depends on the potassium conductance mediated by K ϩ channels (32). Direct measurements in these sink phloem cells revealed membrane potentials of only Ϫ60 mV. 3 At an apoplasmic sucrose concentration of 1 mM, a phloem sap sucrose concentration of 0.85 M, and a pH gradient of 1 unit of sucrose release would occur at membrane potentials positive from Ϫ115 mV (according to Equation 3 with n H /n suc ϭ 1). This regime directs Zm-SUT1 into the inverse transport mode, and sucrose is released.
The present work revealed the functional asymmetry of the phloem sucrose carrier ZmSUT1. Our data, for the first time, demonstrate the "sink mode" of this pivotal carrier type, provide for the molecular mechanism of phloem sucrose release, and explain the severe phenotype of phloem H ϩ /sucrose carrier loss-of-function mutants and antisense-repression plants. In contrast to symplasmic unloading, this sucrose/H ϩ symporterbased mechanism drives unloading of sucrose under the control of both the sucrose and the pH gradients as well as the membrane potential of the phloem and the surrounding tissues.