The vacuolar membrane is involved in solute uptake into and release from the vacuole, which is the largest plant organelle. In addition to inorganic ions and metabolites, large quantities of protons and sugars are shuttled across this membrane. Current models suggest that the proton gradient across the membrane drives the accumulation and/or release of sugars. Recent studies have associated AtSUC4 with the vacuolar membrane. Some members of the SUC family are plasma membrane proton/sucrose symporters. In addition, the sugar transporters TMT1 and TMT2, which are localized to the vacuolar membrane, have been suggested to function in proton-driven glucose antiport. Here we used the patch-clamp technique to monitor carrier-mediated sucrose transport by AtSUC4 and AtTMTs in intact Arabidopsis thaliana mesophyll vacuoles. In the whole-vacuole configuration with wild-type material, cytosolic sucrose-induced proton currents were associated with a proton/sucrose antiport mechanism. To identify the related transporter on one hand, and to enable the recording of symporter-mediated currents on the other hand, we electrophysiologically characterized vacuolar proteins recognized by Arabidopsis mutants of partially impaired sugar compartmentation. To our surprise, the intrinsic sucrose/proton antiporter activity was greatly reduced when vacuoles were isolated from plants lacking the monosaccharide transporter AtTMT1/TMT2. Transient expression of AtSUC4 in this mutant background resulted in proton/sucrose symport activity. From these studies, we conclude that, in the natural environment within the Arabidopsis cell, AtSUC4 most likely catalyses proton-coupled sucrose export from the vacuole. However, TMT1/2 probably represents a proton-coupled antiporter capable of high-capacity loading of glucose and sucrose into the vacuole.
The vacuole can function as a sink compartment for accumulation of ions, sugars and a plethora of other metabolites. These solutes can be mobilized on demand from this large organelle in a cell type- and development-dependent fashion. Thus, the vacuolar membrane serves as an interface between the vacuolar lumen and the cytosol. Solute uptake and release are accomplished via tonoplast-localized ion channels, carriers and pumps (Martinoia et al., 2007). By pumping cytosolic protons into the vacuolar lumen, the V-type ATPase, together with V-PPase (vacuolar-pyrophosphatase), creates a proton motive force that is used to establish ion and metabolite gradients across the vacuolar membrane (Hedrich and Kurkdjian, 1988; Krebs et al., 2010).
In mesophyll cells, sugars produced by the chloroplasts in the light are exported to sink tissues via the phloem or kept in the vacuole. In the dark, sugars are remobilized to supply metabolic energy. H+ pumping activities acidify the vacuolar sap, and sucrose can enter and accumulate in the organelle via transporters powered by energy provided from protons released into the cytosol (Saftner and Wyse, 1980). Hexoses can be imported into the vacuole for dynamic long-term and temporary storage. Sugar/H+ antiport appears to be associated with the metabolic energy status of the cell as well as the proton motive force across the vacuolar membrane (Guy et al., 1979; Echeverria and Gonzalez, 2000). Sucrose uptake into the vacuole has been suggested to be catalysed by a low-affinity transporter in the tonoplast (Kaiser and Heber, 1984). In contrast, sugar unloading from the vacuole was assumed to occur via facilitated diffusion or proton-coupled symport (Kaiser and Heber, 1984; Bush, 1990; Etxeberria and Gonzalez, 2001). Electrophysiological studies with the sucrose transporters StSUT1 from Solanum tuberosum and ZmSUT1 from Zea mays in heterologous expression systems (Boorer et al., 1996; Carpaneto et al., 2005, 2010) unequivocally showed that these proteins catalyse H+-coupled sucrose symport. This transport depended on the direction of the sucrose and pH gradient as well as the membrane potential (Carpaneto et al., 2005).
In Arabidopsis, approximately 60 genes encode potential monosaccharide transporters localized to the plasma membrane as well as to endomembranes (Büttner, 2007; Neuhaus, 2007). Studies at the molecular and functional level showed that TMT (tonoplast monosaccharide transporter) and VGT (vacuolar glucose transporter) are vacuolar carrier proteins with transport capacity for both glucose and fructose (Wormit et al., 2006; Aluri and Büttner, 2007).
Within a phylogenetic tree of plant sucrose transporters, the SUC4 branch (also known as the SUT4 branch) is well separated from three other branches (Sauer, 2007). When heterologously expressed in baker’s yeast (Saccharomyces cerevisiae) or Xenopus laevis oocytes, AtSUC4 (synonym: AtSUT4) from Arabidopsis, StSUT4 from potato (Solanum tuberosum) and LjSUT4 from Lotus japonicus, three members of the SUC4 branch, appear to be associated with pH-dependent uptake of extracellular sucrose (Weise et al., 2000; Weschke et al., 2000; Reinders et al., 2008). However, proteomic studies using purified vacuolar membranes suggested localization of AtSUC4 and its barley (Hordeum vulgare) homologue HvSUT2 to the tonoplast rather than the plasma membrane (Endler et al., 2006; Reinders et al., 2008). Thus targeting of AtSUC4 apparently varies among expression systems. As AtSUC4 represents a tonoplast-localized transporter in its native environment in the Arabidopsis cell, this carrier will face neutral pH on its cytosolic side and acid pH at the luminal side. Thus, sucrose transport mediated by SUC4-type proteins could be coupled mechanistically to the proton motive force in two constellations: proton-coupled antiport or symport. Here we expressed AtSUC4 in Arabidopsis and studied the mode of AtSUC4 action using patch-clamp experiments with isolated vacuoles from wild-type plants and plants lacking the monosaccharide transporters TMT1/2 and VGT1/2. We identified TMTs as vacuolar H+/sucrose antiporters, which, like SUC4, are powered by the proton gradient but show the opposite carrier mode to the H+ symporter of the SUC family.
Resolution of carrier-mediated, sucrose-induced proton currents in Arabidopsis vacuoles
Patch-clamp studies on isolated plant vacuoles have identified ionic currents associated with transport of inorganic cations and anions. While ion channels can be detected even at the single-channel protein level, ion pumps and carriers have to be measured as assemblies in the whole-vacuolar configuration. This mode has already been used with success to characterize tonoplast-localized inositol and hexose transporters (Schneider et al., 2008; Wingenter et al., 2010) or vacuolar currents generated by PPi- or ATP-driven H+ pumps from as low as 0.2 to approximately 2 pA/pF (Hedrich and Kurkdjian, 1988; Krebs et al., 2010). These currents, with low-turnover rates, result from the combined activity of 105–106 H+ pumps. Like pumps, sugar carriers such as SGLT (sodium/glucose co-transporter) and ZmSUT1 show a turnover rate in the range of 100 sugar molecules per second (Mackenzie et al., 1996; Sze et al., 1999; Carpaneto et al., 2010). In contrast to pumps, which can be turned on and off according to metabolic energy status, proton-coupled sugar carriers appear to be sensitive to both substrates. Changes in pH at both sides of the membrane will be recognized by the surface of the proton-coupled carrier in question, and affect proton current amplitude or even direction. Up to now, electrogenic proton currents in co-transport with magnesium, chloride or nitrate have been observed (Shaul et al., 1999; De Angeli et al., 2006, 2009). It is likely that currents associated with proton-coupled K+ and Ca2+ transporters will be reported soon. Thus, a change in pH during patch-clamp experiments with vacuoles exposed to cytosolic and/or luminal media containing K+, Mg2+, Ca2+ or Cl− (physiological media) will have a feedback effect on CLC (chloride transporter/channel) and CAX(cation exchanger)-type proton currents (Barkla et al., 2008; De Angeli et al., 2009). Thus, to prevent interference with other proton-driven ion transporters, patch-clamp experiments under constant pH conditions were designed. Furthermore, we used media for the artificial cytosol and vacuole sap that were symmetric in terms of salt composition and concentration. Under these conditions, and with the membrane potential clamped to 0 mV, no driving force for ion movement across the vacuolar membrane exists. To evoke sucrose-induced proton currents across sugar/H+ antiporters or symporters, the sugar rather than protons served as the trigger in the following analyses. To identify vacuolar sucrose-induced currents, we used vacuoles isolated from wild-type Arabidopsis mesophyll cells. Vacuoles were liberated from selected protoplasts within the recording chamber by osmotic swelling (Schulz-Lessdorf and Hedrich, 1995). After the vacuolar membrane of ruptured mesophyll protoplasts was exposed, the whole-organelle patch-clamp configuration was established (Figure 1). When access to the vacuolar sap had been obtained and equilibration with the pipette solution was completed (Figure 1c–e and Movie S1), we recorded two types of sugar-induced macroscopic currents in total of 85 experiments.
Mesophyll vacuoles operate a sucrose/proton antiporter
In order to mimic physiological proton gradients across the membrane, the vacuolar (pipette) medium was buffered to pH 5.5 and the cytosolic (bath) medium was buffered to pH 7.5 (Figure 2a). When 50 mm sucrose was applied to the bath solution (Figure S1 and Movie S2), all wild-type vacuoles (n =5) responded with downward deflection of the current trace with respect to the background (Figure 2b,c). As shown in Figure 2(b), a vacuole with a capacitance of 22 pF reversibly elicited an inward current of −0.4 pA/pF during cytosolic application of sucrose. Under these experimental conditions, the inward currents probably represent proton currents generated by an antiporter that mediates sucrose influx into the vacuole in exchange for protons. This antiport was driven by both the vacuole-directed sucrose gradient and the cytosol-directed proton gradient.
To thermodynamically promote the function of the potential hidden sucrose/proton symporter at the expense of the antiporter, the pH gradient (pHcyt 5.5/pHvac 7.5) was inverted, i.e. directed into the vacuolar lumen (as for the polyol/proton symporter; Schneider et al., 2008). The driving force for sugar-induced proton transport was thereby reduced for the antiporter but increased for the symporter. Nevertheless, cytosolic sucrose treatment of the wild-type vacuoles still activated inward currents in the range from −0.5 to −2.8 pA/pF (n =3) (Figure 2c). These current responses suggest that sucrose application to the cytoplasmic membrane side can still trigger proton release from the vacuole even against a pH gradient. Thus the vacuolar membrane of mesophyll cells harbours a prominent H+-coupled sucrose antiporter activity that may overlap a potential weak sucrose/proton symporter activity. If the latter is mediated by SUC4, SUC4-mediated H+/sucrose symport should be resolved in vacuoles in which SUC4 is over-expressed and the interfering sucrose/proton antiporter activity largely reduced.
AtTMT1/2 is capable of proton-coupled sucrose and glucose antiport
We recently identified a plasma membrane carrier in the pathogenic fungus Ustilago mavdis with a moderate similarity to members of the monosaccharide transporter family and capability of sucrose transport (Wahl et al., 2010). Reasoning that a similar situation could exist in plant vacuoles, we studied two types of vacuolar monosaccharide transporter, TMT1 and TMT2 (Wingenter et al., 2010) and VGT1 and VGT2 (Aluri and Büttner, 2007). Using a physiological pH gradient, we challenged the wild-type vacuoles with 125 mm glucose from the cytosolic membrane side. Glucose-induced inward currents of approximately 2 pA/pF were elicited (Figure 3a,c). When the glucose concentration was gradually reduced from 125 to 20 mm, the sugar-induced proton currents decreased in a linear manner (Figure 3b). When using vacuoles from Arabidopsis tmt1/tmt2 double knockout plants, glucose-induced currents did not exceed 0.5 pA/pF. The results were similar for vacuoles from tmt1/tmt2/vgt1/vgt2 quadruple knockout mutants that additionally lacked VGT1 and VGT2 in the tmt1/tmt2 mutant background (Figure 3c), indicating that TMTs represent the major glucose transporter species of the Arabidopsis mesophyll vacuole. We thus examined the glucose and sucrose responses of vacuoles from Arabidopsis wild-type, AtTMT1 over-expressors (AtTMT1-OX) and Attmt1/tmt2 knockout mutants. For direct comparison of the currents triggered in response to both sugar species, individual vacuoles from wild-type, Attmt1/tmt2 knockout and AtTMT1-OX plants were consecutively exposed to glucose and sucrose (Figure 4a). The glucose-induced current responses tended to be impaired in vacuoles of tmt1/2 loss-of-function mutants compared to wild-type or AtTMT1 over-expressing plants, as previously described by Wingenter et al. (2010). Interestingly, subsequent application of sucrose always elicited current responses of similar magnitude to those obtained with glucose for the same vacuole (Figure 4a,b and inset). This was observed irrespective of the absolute sugar-induced current amplitude, which differed between the vacuole populations released from wild-type, Attmt1/tmt2 knockout and AtTMT1-OX plants (Figure 4b and inset). These results indicate that, under the experimental conditions used here, both glucose and sucrose serve as substrates for TMT1 and/or TMT2 for proton-coupled sucrose antiport across the vacuolar membrane.
AtSUC4 represents a sucrose/proton symporter
To overcome the glucose/proton and sucrose/proton antiporter activity of TMTs in the system, and to unequivocally identify AtSUC4 as a vacuolar sucrose/proton symporter in the natural environment of the Arabidopsis vacuole, we generated Attmt1/tmt2 and Attmt1/tmt2/vgt1/vgt2 knockout plants and transiently expressed an AtSUC4–GFP fusion construct in mesophyll protoplasts of these mutants. As a control, Attmt1/tmt2 knockout protoplasts were transformed with a construct encoding free GFP. Following transformation, the vacuolar membrane showed pronounced GFP fluorescence after 40–48 h with AtSUC4–GFP only, confirming the localization of AtSUC4 in the tonoplast of these mutants (Figure 5) (Endler et al., 2006). Under inverted pH gradients, application of sucrose to vacuoles over-expressing AtSUC4–GFP resulted in upward deflection of the current trace (n =6). This behaviour is in line with the characteristics of sucrose-driven proton efflux from the cytosol into the vacuole (Figure 6, normalized ΔI/Cm = 0.65 ± 0.27, n =6, mean ± SE). These results underpin our hypothesis that AtSUC4 targets the vacuolar membrane and provides the proton/sucrose symport activity for sugar release from the vacuole.
Former analyses by Endler et al. (2006) and Reinders et al. (2008) showed that SUC4-type transporters from the plant families Brassicaceae, Fabaceae and Poaceae are targeted to the vacuolar membrane. However, when AtSUC4 from Arabidopsis, StSUT4 from potato, HvSUT2 from barley and LjSUT4 from L. japonicus were expressed in baker’s yeast (Weise et al., 2000; Weschke et al., 2000) or in Xenopus oocytes (Reinders et al., 2008), they catalysed transport of sucrose across the plasma membrane, and the sucrose-induced proton currents were in agreement with the action of a plasma membrane sucrose/H+ symporter. These two lines of evidence suggest that the plasma membrane localization of SUC4-type proteins in the various heterologous expression systems could represent a mis-targeting phenomenon. Although it is debatable whether LjSUT4 in the oocyte plasma membrane works in the same way as in its native environment of the plant vacuole, LjSUT4 protein definitely shuttles sucrose in symport with protons into Xenopus oocytes. Nevertheless, direct proof of SUC4-type transporter action in planta, particularly in the Arabidopsis vacuole, the location of AtSUC4, was lacking.
To bridge this gap, we isolated vacuoles from Arabidopsis mesophyll protoplasts for patch-clamp studies and took advantage of the high resolution of the whole-vacuole configuration. However, we did not observe a major sucrose/H+ symporter activity in wild-type mesophyll vacuoles under physiological (acidic vacuolar lumen) or inverted (acidic cytosol) pH gradients across the vacuolar membrane (Figure 2). This is probably due to weak AtSUC4 expression in Arabidopsis source leaves (Weise et al., 2000), and superposition of a co-existing and dominating sucrose/H+ antiporter activity represented by TMT1/2 (Figures 3 and 4) (Wormit et al., 2006; Wingenter et al., 2010).
However, the resolution of the system was greatly improved when AtSUC4 was transiently expressed in mesophyll protoplasts lacking both TMT1/2 antiporters. In these vacuoles, the interfering sucrose/H+ antiport activity of the vacuolar monosaccharide transporter TMT1/2 was largely reduced (Figure 4). When the pH gradient was also inverted (acidic outside), sucrose application to the extra-vacuolar solution buffered to pH 5.5 elicited exclusively proton currents into the vacuolar lumen (pH 7.5) (Figure 6). In this context, and in line with the operation of a reversible thermodynamic machine, the plasma membrane-localized H+ symporter ZmSUT1 could be switched experimentally from H+-driven sucrose influx to sucrose-driven H+ efflux by inverting the sucrose gradient (Carpaneto et al., 2005). Together, these findings show that AtSUC4 in the Arabidopsis vacuolar membrane represents a sucrose/proton symporter activity.
Application of sucrose in the presence of a proper pH gradient induces proton currents that are mediated by the given sucrose carrier population (Figure 6) (Carpaneto et al., 2005; Reinders et al., 2008). When considering how to determine the number of transporters and turnover number in an expression system, it is helpful to review a recent study on ZmSUT1 (Carpaneto et al., 2010). Sucrose-dependent H+ transport via SUT1-type transporters is associated with a decrease in the membrane capacitance (Cm). In the absence of sucrose, ‘trapped’ protons move back and forth between an outer and an inner site within the transmembrane domains of the sugar/proton carrier (Carpaneto et al., 2010). This movement of protons in the electrical field of the membrane gives rise to changes in the Cm. Upon application of external sucrose, protons can pass through the membrane, turning pre-steady-state currents into transport currents. Based on parameters for the pre-steady-state current, a turnover rate of 500 molecules of sucrose per second was calculated for ZmSUT1 from Zea mays (Carpaneto et al., 2010). Given a similar turnover number for AtSUC4 and a 1:1 sucrose/H+ ratio (Carpaneto et al., 2005), a sucrose/H+ symport process of a vacuole with a capacitance of 20 pF associated with a proton release current of 20 pA requires the activity of 250 000 carrier molecules. To maintain the pH gradient across the vacuolar membrane, 1 000 000 proton pumps with a turnover rate of approximately 100 are required.
Tonoplast-localized TMTs and SUC-type transporters use the same proton gradient to transport sugars in opposite directions. To prevent futile cycles when operating the antiporter, the symporter should be turned off, and vice versa. Thus, the number and/or activation state of the TMT1/2 and SUC4 transporters must be adapted to cellular or environmental conditions. Under our experimental conditions, the sucrose/proton TMT1/2 antiporter activity greatly exceeded the sucrose/proton SUC4 symporter activity in mesophyll cell vacuoles. Therefore, TMT1/2 may be involved in accumulation of sucrose in mesophyll cell vacuoles that has been photosynthetically produced in the light. However, under salt stress, vacuolar organic osmolytes are replaced by ‘metabolically cheap’ inorganic sodium salts. Accordingly, sugar accumulation via TMT-type sugar antiporters must be suppressed and uncoupled from the proton motive force, while sugar efflux and NHX-type proton-dependent cation uptake must be promoted. Calcium-dependent and -independent phosphorylation controlling transport activity is well documented in studies with NHX-type carriers and ion channels (Mahajan et al., 2008). We have identified a protein kinase moiety binding to the central loop of TMT1 (Karina Wingenter, Oliver Trentmann, Irina Winschuh, Imke Hörmiller, Arnd Heyer, Jörg Reinders, Alexander Schulz, Dietmar Geiger, Rainer Hedrich, Ekkehard Neuhaus, unpublished data). It is thus tempting to speculate that phosphorylation is involved in controlling TMT transport features. As a result, TMT1/2 may preferentially transport mono- or disaccharides or vice versa under certain developmental or environmental conditions. Thus, future studies concerning the regulation of sugar transport via interaction with regulator proteins will very likely result in new insights into the control of TMT action in particular, and sugar transporters in general.
Patch-clamp analyses of isolated plant vacuoles
Arabidopsis thaliana wild-type (Col-0) and mutant plants were grown on soil for 5–8 weeks in a growth chamber under a 8/16 h day/night regime (22/16°C day/night temperature; light intensity 125 μmol m−2 sec−1). Isolation of mesophyll protoplasts and release of vacuoles were performed as described previously (Beyhl et al., 2009). Release of vacuoles of transformed protoplasts was achieved using W5 buffer (Yoo et al., 2007) diluted to an osmolarity of 200–300 mOsm. In line with the convention for electrical measurements on endomembranes (Bertl et al., 1992), patch-clamp experiments were performed in the whole-vacuole configuration, essentially as described by Schulz-Lessdorf and Hedrich (1995) and Ivashikina and Hedrich (2005). The vacuolar membrane was clamped to 0 mV. Patch pipettes prepared from Kimax-51 glass capillaries (Kimble Products, http://www.kimble-chase.com) were characterized by resistance in the range 1.5–3 MΩ. Macroscopic currents were recorded at a sample acquisition interval of 2 msec using an EPC-7 patch-clamp amplifier (HEKA Elektronik, http://www.heka.com) and low-pass-filtered at 30/40 Hz. Data were digitized using an ITC-16 computer interface (Instrutech Corp., http://www.instrutech.com), stored on a computer and analysed offline using the software programs Pulse (HEKA Electronik) and IGORPro (Wave Metrics Inc., http://www.wavemetrics.com). Current amplitudes for individual vacuoles were normalized to the whole-vacuolar membrane capacitance Cm to allow comparison of macroscopic current magnitudes among vacuoles. Experiments were performed under symmetrical solute conditions, with the pipette (vacuolar lumen) and bath solution (cytosol) containing 155–200 mm KCl, 1 mm CaCl2 and 2 mm MgCl2. Solutions were adjusted either to pH 5.5 or 7.5 using 10 mm Mes/Tris and 10 mm HEPES/Tris, respectively, as indicated in the figure legends. Sucrose and glucose treatment of the vacuole was achieved by manually applying high pressure to a glass pipette filled with the sugar-containing bath solution and located in front of the vacuole. Sugar application was maintained until corresponding current responses reached at least the quasi steady-state level under high time resolution.
Equilibration of the vacuolar sap with the patch pipette solution
After establishment of the whole-vacuolar configuration, equilibration of the vacuolar lumen with the patch pipette medium was monitored by loading with the fluorochrome lucifer yellow (100 μm). Fluorescence measurements were performed and analysed using MetaFluor software (Universal Imaging, http://www.moleculardevices.com). Lucifer yellow was excited at 430 nm using the VisiChrome high-speed polychromator system (Visitron Systems GmbH, http://www.visitron.de/), and fluorescence intensity was captured using a Plan-Neofluar objective 40×/0.75 W (Carl Zeiss, http://www.zeiss.com/) with a cooled charge-coupled device camera (CoolSNAP HQ; Roper Scientific, http://www.roperscientific.com/) attached to an inverse fluorescence microscope (Axiovert 135, Carl Zeiss). Fluorescence emission was filtered through a dichroic mirror of 500 nm and a 510–560 nm bandpass filter (HQ 535/25 M; AF Analysentechnik, http://www.ahf.de/). Images were recorded every fifth second. All pictures were corrected for background fluorescence recorded from reference regions that were close to the analysed vacuole using ImageJ (National Institutes of Health, http://rsbweb.nih.gov/ij/). The corresponding movie was generated with ImageJ and processed using VirtualDub (http://www.virtualdub.org). For this, the original frame rate of 7 frames per second (fps) was changed to 10 fps. The Xvid MPEG-4 Codec (http://www.xvid.org) was used for compression.
Monitoring the sugar application system
The reliability of the sugar application system was tested and monitored using the fluorochrome lucifer yellow. The application pipette was filled with 1 mm of the fluorescent dye. The fluorescence was recorded using the same set-up as described above. The corresponding movie was generated using metafluor software and processed using VirtualDub. Therefore, the frame rate was changed from 3 to 30 fps. The Xvid MPEG-4 Codec was used for compression.
Generation of homozygous Attmt1/tmt2/vgt1/vgt2 and Attmt1/tmt2 mutants
A quadruple mutant plant was created using homozygous T-DNA insertion lines for the AtTMT1 (At1g20840), AtTMT2 (At4g35300), AtVGT1 (At3g03090) and AtVGT2 (At5g17010) genes. A homozogous Attmt1/tmt2::tDNA double knockout line (Wormit et al., 2006) was crossed with an Atvgt1 T-DNA insertion line (SALK_000988; Aluri and Büttner, 2007) or an Atvgt2 T-DNA insertion line (SALK_150864). The homozygous triple mutant lines tmt1/tmt2/vgt1::tDNA and tmt1/tmt2/vgt2::tDNA were then crossed, yielding a quadruple tmt1/tmt2/vgt1/vgt2::tDNA mutant line. Homozygosity of the T-DNA insertions and lack of gene-specific transcripts, were verified for AtTMT1, AtTMT2 and AtVGT1 by PCR and RT-PCR reactions using the primers indicated by Wormit et al. (2006) and Aluri and Büttner (2007). Homozygosity of the T-DNA insertion in AtVGT2 was confirmed by PCR using primers LBb1 (5′-AGTTGCAGCAAGCGGTCCACGC-3′) and AtXYL2g+2685r (5′-AGCCACCACATCCCAATTCC-3′) and lack of AtVGT2 transcripts was verified by RT-PCR using primers AtXYL2g+2103f (5′-GCTGAGCATGGGTTGTTATCC-3′) and AtVGT2g+2685r (5′-AGCCACCACATCCCAATTCC-3′). Arabidopsis loss-of-function mutants lacking the genes encoding the vacuolar sugar carriers TMT1 and TMT2 have been genetically and physiologically described previously (Wormit et al., 2006; Wingenter et al., 2010).
Transient expression of AtSuc4–GFP in mesophyll protoplasts
For pSU4, a construct encoding an AtSUC4–GFP fusion under the control of the 35S promoter, the AtSUC4 coding sequence was amplified using the primers AtSUC4-Start-Pci (5′-ACATGTCTACTTCCGATCAAGATCGCCGT-3′) and AtSUC4-GFP-Pci (5′-ACATGTTGGCTGCTGGGAGAGGGATGGGCTTCTG-3′), which introduced PciI cloning sites at the AtSUC4 start codon and replaced the stop codon by two alanine codons and a second PciI cloning site. The PCR fragment was cloned into pCR-Blunt II-TOPO (Invitrogen, http://www.invitrogen.com/), sequenced and inserted into NcoI-digested pSB30 via its PciI sites. pSB30 is a pUC19-derived vector with a 1940 bp PacI fragment containing a 35S promoter–GFP–nos terminator box with a unique NcoI site between the 35S promoter and the GFP sequence.
pSB30 and pSU4 were used for transient transformation of mesophyll protoplasts of Attmt1/tmt2 double and Attmt1/tmt2/vgt1/vgt2 quadruple knockout mutants as described previously (Yoo et al., 2007). Vacuoles containing GFP-labelled AtSUC4 transporters were identified by GFP-emitted fluorescence and used for subsequent patch-clamp experiments. As a control, current measurements were performed on vacuoles liberated from a population of mesophyll protoplasts that were subjected to transient transformation with pSB30. Due to the cytosolic localization of GFP, released vacuoles could not be linked to GFP-containing protoplasts.
Microscopy and detection of GFP fluorescence
To confirm the vacuolar localization of AtSUC4, images of GFP-labelled vacuoles released from protoplasts after transient transformation with pSU4 were generated on a confocal laser-scanning microscope (LSM 5 Pascal; Carl Zeiss). The laser-scanning microscope was equipped with a Zeiss Plan-Neofluar 20×/0.5 objective. Images were processed (merging, brightness, contrast) using the image acquisition software ImageJ.
We are grateful to the Deutsche Forschungsgemeinschaft (Research Unit 1061) for generous financial support to R.H.