* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Transport of amino acids (L-valine, L-lysine, L
Survey
Document related concepts
Point mutation wikipedia , lookup
Fatty acid synthesis wikipedia , lookup
Proteolysis wikipedia , lookup
Plant nutrition wikipedia , lookup
Lipid signaling wikipedia , lookup
Metalloprotein wikipedia , lookup
Chemical synapse wikipedia , lookup
Fatty acid metabolism wikipedia , lookup
Magnesium in biology wikipedia , lookup
Oxidative phosphorylation wikipedia , lookup
Genetic code wikipedia , lookup
Vesicular monoamine transporter wikipedia , lookup
Western blot wikipedia , lookup
Biosynthesis wikipedia , lookup
Amino acid synthesis wikipedia , lookup
Biochemistry wikipedia , lookup
Transcript
Journal of Experimental Botany, Vol. 51, No. 351, pp. 1663±1670, October 2000 Transport of amino acids (L-valine, L-lysine, L-glutamic acid) and sucrose into plasma membrane vesicles isolated from cotyledons of developing pea seeds A. de Jong1 and A.C. Borstlap2 Transport Physiology Research Group, Department of Plant Ecology and Evolutionary Biology, Utrecht University, Sorbonnelaan 16, NL-3584 CA Utrecht, The Netherlands Received 22 May 2000; Accepted 24 May 2000 Abstract Introduction Transport of the amino acids L-valine, L-lysine, and L-glutamic acid and of sucrose was studied in plasma membrane vesicles isolated from developing cotyledons of pea (Pisum sativum L. cv. Marzia). The vesicles were obtained by aqueous polymer twophase partitioning of a microsomal fraction and the uptake was determined after the imposition of a Hq-gradient (DpH, inside alkaline) anduor an electrical gradient (Dy, inside negative) across the vesicle membrane. In the absence of gradients, a distinct, time-dependent uptake of L-valine was measured, which could be enhanced about 2-fold by the imposition of DpH. The imposition of Dy stimulated the influx of valine by 20%, both in the absence and in the presence of DpH. Uptake of L-lysine was more strongly stimulated by Dy than by DpH, and its DpHdependent uptake was enhanced about 6-fold by the simultaneous imposition of Dy. In the absence of gradients the uptake of L-glutamic acid was about 2-fold higher than that of L-valine, but it was not detectably affected by DpH or Dy. Although the transport of sucrose was very low, a stimulating effect of DpH could be clearly demonstrated. The results lend further support to the contention that during seed development cotyledonary cells employ Hq-symporters for the active uptake of sucrose and amino acids. Cotyledons are the bulkiest part of the legume seed in its later stages of development and account for most of the nutrient absorption and synthesis of storage compounds in the embryo. Developing seeds are supplied by a stream of phloem sap that carries along sucrose and amino acids serving as their main organic nutrients. The current view is that the contents of the terminal sieve elements pass symplastically to the seed coat parenchyma cells, from which they are released into the apoplastic space and are ®nally taken up by the cotyledons (for review: Patrick, 1997). In contrast with sucrose, amino acids seem to undergo appreciable processing in the seed coat parenchyma, for the composition of the amino acid mixture released by seed coats is quite different from that in phloem exudates (Rochat and Boutin, 1991; Lanfermeijer et al., 1992). Generally, the concentration dependence of the in¯ux of solutes into plant cells can be analysed into one or more saturable components and a linear component (Borstlap, 1983). This also applies to the uptake of sucrose and amino acids into cotyledons of developing legume seeds. Cotyledons of soybean (Lichtner and Spanswick, 1981b; Thorne, 1982), pea (Lanfermeijer et al., 1991), and broad bean (McDonald et al., 1996) all take up sucrose by a saturable component (Km~5 15 mM; Vmax~3 9 mmol g 1 FW h 1) and a linear component (k~30 80 mmol g 1 FW h 1 M 1). Similarly, uptake of the neutral amino acid L-valine by pea cotyledons displayed a saturable component (Km~5 mM) and a linear component (k~100 200 mmol g 1 FW h 1 M 1). The saturable component of amino acid uptake did not Key words: Amino acids, cotyledons, plasma membrane, proton symport, sucrose. 1 Present address: Academic Medical Center, Department of Cardiac Catherization B2-115, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. 2 To whom correspondence should be addressed. Fax: q31 30 2518366. E-mail: [email protected] ß Society for Experimental Biology 2000 1664 de Jong and Borstlap appear before a rather advanced stage of development at which the water content of the cotyledons had decreased to about 65% (Lanfermeijer et al., 1990). At this stage the cotyledons are full-grown and about one-third of the ®nal amount of storage proteins has already been deposited. In soybean cotyledons (water content ;80%), the uptake of 2-aminoisobutyric acid and L-glutamine is also dominated by a linear component. Nitrogen starvation of the isolated cotyledons apparently led to derepression of a saturable system with a Km for glutamine of 96 mM (Bennett and Spanswick, 1983). It is not known whether this system is comparable to that appearing during pea seed development. Evidence has accumulated that saturable sucrose uptake by developing cotyledons is effected by a Hq-symporter. The uptake is attended by a transient membrane depolarization (Lichtner and Spanswick, 1981a), depends strongly on the external pH and is sensitive to protonophores (Lichtner and Spanswick, 1981b; Thorne, 1982; Lanfermeijer et al., 1991; McDonald et al., 1996). In addition, more recent work with faba bean and pea has shown that transcripts of a gene encoding a Hqusucrose symporter localize to the outer cell layers of the cotyledons (Harrington et al., 1997; Weber et al., 1997; Tegeder et al., 1999). Proton symporters are probably also involved in amino acid uptake by cotyledons since the saturable uptake component is sensitive to protonophores and shows a distinct pH-dependency (Lanfermeijer et al., 1990). In developing seeds of Arabidopsis the Hquamino acid symporter AAP1 has been found to be expressed in embryo and endosperm (Hirner et al., 1998). Aqueous polymer two-phase partitioning of the microsomal fraction from tissue homogenates can be used to obtain a fraction that is enriched in plasma membrane vesicles (Larsson et al., 1987). Hq-gradients or membrane potentials can be imposed across the vesicle membranes to drive Hq-coupled anduor electrogenic transport of solutes against their concentration gradients. In this way Hq-symport of sucrose anduor amino acids has been demonstrated in plasma membrane vesicles from various plant tissues (for review: Bush, 1993). In the present paper the isolation and characterization of plasma membrane vesicles from cotyledons of developing pea seeds is described, and Hq-symport of sucrose, L-valine and L-lysine in these vesicles is demonstrated. Transport studies with plasma membrane vesicles isolated from pea seed coats are presented in an accompanying paper (De Jong and Borstlap, 2000). Materials and methods Plant material Pea plants (Pisum sativum L. cv. Marzia) were grown from seeds (Nunhems Zaden B.V., Haelen, The Netherlands) in a growth chamber as described (De Jong and Wolswinkel, 1995) but no ¯owers were removed. Seeds were harvested towards the end of seed ®lling (water content of the cotyledons 54±58%). Isolation of plasma membrane vesicles Isolation of microsomal membranes and plasma membrane puri®cation were performed at 0±4 8C. Seeds were taken from the pods and after the removal of seed coat and embryonic axis the cotyledons (100±140 g FW) were blended with 200 ml of homogenization medium (50 mM MOPS-KOH, pH 7.5, 330 mM sucrose, 5 mM EDTA, 1 mM DTT, 0.75% (wuv) insoluble polyvinylpyrrolidone) in a Braun kitchen homogenizer. After three consecutive 20 s bursts at maximal speed, the resulting slurry was ®ltered through a layer of 250 mm mesh Perlon polyamide gauze and BSA was added to the ®ltrate to a ®nal concentration of approximately 2 g l 1. To get rid of the large amounts of starch, the ®ltrate was ®rst centrifuged at 4200 g for 10 min in a Sorvall RC-5 centrifuge with an SS 34 rotor. The pellet was discarded and the supernatant was centrifuged at 20 000 g for 15 min. The microsomal fraction was obtained by recentrifugation of the supernatant at 50 000 g for 90 min, and the microsomal pellet was taken up in resuspension medium (330 mM sucrose, 5 mM potassium phosphate, pH 7.8, 10 mM KCl) to a ®nal volume of 10 ml. Isolation of plasma membrane vesicles was carried out by aqueous polymer two-phase partitioning (essentially as described by Larsson et al., 1987). Nine ml of the microsomal fraction was added to 27 g of phase mixture to form a 36 g phase system with a ®nal concentration of 5.7% (wuw) dextran T500, 5.7% (wuw) polyethyleneglycol 4000, 330 mM sucrose, 5 mM potassium phosphate, pH 7.8, 10 mM KCl, 1 mM DTT, and 0.1 mM EDTA. After mixing thoroughly, the phases were separated by centrifugation at 1500 g for 5 min. The upper phase was then repartitioned twice with fresh lower phases giving U3 and the three lower phases were sequentially re-extracted with a fresh upper phase giving U39. The upper phases U3 and U39 were each diluted 2-fold in washing medium (330 mM sorbitol, 50 mM HEPES-KOH, pH 7.0, 39 mM KCl, 1 mM DTT, 0.1 mM EDTA). After centrifugation at 100 000 g for 60 min in a Beckman L60 ultracentrifuge with an SW 28 rotor the supernatant was discarded and the combined pellets were resuspended in 40 ml of washing medium and centrifuged again at 100 000 g for 60 min. The ®nal plasma membrane pellet was resuspended in 300±900 ml pH7K-medium (330 mM sorbitol, 50 mM HEPES-KOH, pH 7, 39 mM KCl, 0.1 mM DTT) to a ®nal concentration of 0.3±1 mg protein ml 1, frozen in liquid nitrogen and stored at 80 8C until use. Protein assay Protein was assayed with the bicinchoninic acid reagent (Pierce) following the instructions of the manufacturer. Bovine serum albumin in 0.9% sodium chloride and 0.05% sodium azide was used as a standard, and the absorbance was read at 562 nm. ATPase assay ATPase activity was measured as the release of inorganic phosphate from ATP after 30 min of incubation at 30 8C. The reaction medium used was adopted from Larsson et al. (Larsson et al., 1988) with some minor modi®cations and contained 330 mM sucrose, 50 mM MES-TRIS, pH 6.5, 0.1 mM sodium molybdate, 1 mM sodium azide, 0.1 mM Na2EDTA, 25 mM K2SO4, 3 mM disodium ATP, 3 mM MgCl2, and 40 ml of vesicle suspension in a ®nal volume of 500 ml. Molybdate and azide Plasma membrane vesicles from cotyledons were included in the reaction medium to inhibit acid phosphatases and mitochondrial ATPase, respectively (Gallagher and Leonard, 1982). ATPase activity was determined in the absence and in the presence of 0.01% (wuv) Brij 58 ( polyoxyethylene 20 cethylether). This detergent exposes all latent ATP-binding sites without otherwise affecting the ATPase activity (Palmgren et al., 1990; Johansson et al., 1995). Accordingly, the percentage of latent ATPase activity was calculated as activity with Brij 58 activity without Brij 58 100%3 activity with Brij 58 Orthovanadate (1 mM) and ba®lomycin A1 (0.02 mmol mg 1 protein) were used to inhibit P-type and V-type ATPases, respectively (Gallagher and Leonard, 1982; Bowman et al., 1988). Inorganic phosphate was determined colorimetrically according to Coccuci and MarreÁ (Coccuci and MarreÁ, 1984). One ml of reagent consisting of 3% (wuv) (NH4)6Mo7O24.4H2O, 7.5% (vuv) of concentrated H2SO4, 3% (wuv) FeSO4.7H2O and 0.75% (wuv) sodium dodecyl sulphate was added to the ATPase incubation mixture. After 10 min the reaction was terminated by adding 0.5 ml of 7% (wuv) Na3-citrate. KH2PO4 in the presence of 3 mM Mg-ATP was used as a standard. The absorbance was measured at 750 nm. Cytochrome c oxidase assay Cytochrome c oxidase activity was determined according to Hodges and Leonard (Hodges and Leonard, 1974) with a few modi®cations: 0.01% (wuv) Triton X-100 was used as a detergent, the ®nal concentration of cytochrome c was 45 mM, and the reference cuvet also contained 0.83 mM K3Fe(CN)6 to allow the complete oxidation of cytochrome c. Activities were calculated using a molar extinction of coef®cient for the reduced cytochrome c of 18.5 mM 1 cm 1. Glucan synthase II assay The activity of glucan synthase II (1,3-b-glucan synthase) was determined using UDP-w3Hx-glucose as a substrate and measuring the incorporation of w3Hx-glucose into polyglucan (essentially as described by Fredrikson and Larsson, 1989). The reaction mixture contained 0.012% (wuv) digitonine and DTT was omitted. Transport assays Uptake was measured after diluting the vesicle suspensions 15-fold into an uptake medium that contained about 10 kBq ml 1 of the 14C-labelled substrate and 0.1 mM valinomycin. To impose a transmembrane pH gradient (DpH, inside alkaline) and a membrane potential (Dy, inside negative) across the membrane, the vesicle suspension was diluted in pH5Namedium (330 mM sorbitol, 50 mM MES-NaOH, pH 5.0, 47 mM NaCl, 0.1 mM DTT). Control assays, determining the uptake in the absence of gradients, were carried out by diluting the vesicles in pH7K-medium (330 mM sorbitol, 50 mM HEPES-KOH, pH 7.0, 39 mM KCl, 0.1 mM DTT). The pH5K-medium (330 mM sorbitol, 50 mM MES-KOH, pH 5.0, 47 mM KCl, 0.1 mM DTT) was used as the uptake medium to impose DpH alone, and the pH7Na-medium (330 mM sorbitol, 50 mM HEPES-NaOH, pH 7.0, 39 mM NaCl, 0.1 mM DTT) to impose Dy alone. The uptake experiments were started by adding the vesicle suspension to the uptake medium and were run at 20 8C in a water bath. At speci®ed times, 200 ml aliquots of the incubation mixture were ®ltered through ®lter membranes (Schleicher & Schuell, ME25, pore size 0.45 mm), prewetted 1665 with uptake medium and placed on a ®lter holder (Schleicher & Schuell, type DN 025u4). The ®lters were washed four times, each time with 600 ml of uptake medium, transferred to scintillation vials, and air-dried. Radioactivity was determined after addition of 6 ml of Emulsi®er Scintillator Plus (Packard) in a Tri-Carb 2200CA liquid scintillation analyser (Packard). In some cases initial rates of uptake were estimated by polynomial regression of the uptake-time curve (Dorando and Crane, 1984), and will be given with the ®tting-derived standard errors. Results Plasma membrane purity About 0.6% of the amount of protein in the microsomal fraction was recovered in the plasma membrane fraction U3qU39. Recovery of plasma membrane markers was very poor. As little as 3±4% of glucan synthase II and vanadate-inhibitable ATPase activity in the microsomal fraction was recovered in the plasma membrane fraction, whereas >60% of the activities remained in the ®rst lower phase L1 (Table 1). Almost colourless plasma membrane preparations were obtained from the deep green microsomal fractions and the speci®c activity of the mitochondrial marker cytochrome c oxidase in U3qU39 was >50-fold lower than in the microsomal fraction (Table 2). About 90% of the ATPase activity in the plasma membrane fraction could be inhibited by vanadate. The speci®c activities of the plasma membrane markers were increased by 7-fold and 10-fold, respectively. A minor part (12%) of the ATPase activity in the microsomal fraction could be inhibited by ba®lomycin and this activity was enriched 3.5 times in (U3qU39). About 50% of the plasma membrane vesicles had the right-side-out orientation, as judged from the latency of the vanadateinhibitable ATPase activity (Table 2). Transport of L-valine Uptake of L-valine by plasma membrane vesicles was determined under four conditions: after the simultaneous imposition of DpHqDy, after imposition of DpH alone or Dy alone, and when no gradient was present at all. As can be seen in Fig. 1a, even in the absence of gradients a distinct, time-dependent uptake of valine was measured. This uptake could be largely removed by washing the vesicles with an osmoticum-free medium (Fig. 2a). The imposition of DpH stimulated the valine in¯ux about 2-fold, whereas the imposition of Dy increased the in¯ux by about 20%, both when imposed alone or in combination with DpH. In the presence of DpHqDy, valine uptake reached levels of about 230 pmol mg 1 protein after 10 min of incubation (Fig. 1a). Assuming an intravesicular volume of 5 ml mg 1 protein this uptake corresponds with an accumulation ratio of ;40. The DpH-stimulated uptake was abolished when CCCP was included in the uptake medium. However, CCCP also reduced the uptake in the absence of gradients (Fig. 2b). 1666 de Jong and Borstlap Table 1. Protein and activities of marker enzymes in the microsomal fraction, the first lower phase (L1), and the combined upper phases (U3qU39) Unless indicated otherwise, the data represent the mean"SE of four preparations, and the enzyme activities are expressed in mmol min 1. Protein (mg) Cytochrome c oxidase Glucan synthase II a ATPase, total activity b Vanadate-inhibitable Ba®lomycin-inhibitable a b (U3qU39) Microsomal fraction L1 73"7 86"11 79 5.86"0.36 4.38"0.37 0.71"0.15 55"7 70"9 82 3.76"0.27 2.73"0.10 0.61"0.04 0.42"0.11 0.009"0.002 2.4 0.21"0.08 0.19"0.07 0.02"0.01 Recovery (%) L1 (U3qU39) 75 81 103 64 62 86 0.57 0.01 3.1 3.6 4.3 2.4 Activities (nmol min 1) of a single preparation. Determined in the presence of molybdate and azide. Table 2. Specific activities (nmol mg 1protein min 1) of marker enzymes in the microsomal fraction, the first lower phase (L1), and the combined upper phases (U3qU39) Unless indicated otherwise, the data represent the mean"SE of four preparations. Cytochrome c oxidase Glucan synthase GSII a ATPase, total activity b Vanadate-inhibitable Ba®lomycin-inhibitable Vanadate inhibition (%) Ba®lomycin inhibition (%) Latency (%) d Microsomal fraction L1 (U3qU39) Enrichment in (U3qU39) 1244"24 0.86 82"9 62"10 10"1 75"4 c 12"2 33"2 c 1346"232 1.12 70"7 51"5 12"2 73"3 17"2 27"2 23"6 8.67 474"47 429"39 35"7 91"1 7"1 51"2 0.018 10.1 5.78 6.92 3.5 a Single preparations. Determined in the presence of molybdate and azide. c Mean"SE of five preparations. d Latency of the vanadate-inhibitable ATPase activity. b Fig. 1. Uptake of amino acids by plasma membrane vesicles from pea cotyledons after the imposition of DpHqDy (k), DpH (m), Dy (I) or no gradients (j). (a) Uptake of L-valine supplied at a concentration of 1.12 mM. Symbols represent the mean values "SE of four preparations. Estimated initial in¯uxes (pmol mg 1 protein min 1) were 46"2 (k), 38"2 (m), 21"2 (I), and 17"1 (j). (b) Uptake of L-lysine suplied at a concentration of 0.85 mM. Symbols represent the mean values"SE of three (k;j) or two (m;I) preparations. (c) Uptake of L-glutamic acid supplied at a concentration of 1.04 mM. Symbols represent the mean values of three preparations. For clarity the standard errors are only shown for the upper most and the lower most symbols. In microsomal vesicles and in vesicles from the ®rst lower phase L1 uptake of valine was much lower than in the plasma membrane vesicles (Fig. 2c). After 10 min of incubation the uptake was about 10-fold less, and the initial in¯uxes were about 5-fold lower than in the plasma membrane vesicles. Routinely, the uptake experiments were carried out with vesicles that had been stored at 80 8C, i.e. after one Plasma membrane vesicles from cotyledons Fig. 2. Uptake of L-valine by plasma membrane vesicles from pea cotyledons. Uptake was determined after the imposition of DpHqDy by diluting the vesicle suspension in pH5Na-medium (k), or in the absence of gradients by diluting the vesicle suspension in pH7K-medium (j). L-valine was supplied at a concentration of 1.12 mM. Initial in¯uxes are expressed in pmol mg 1 protein min 1. (a) Effect of osmotic shock. Aliquots of the incubation mixture were washed on the ®lter membrane with pH7K-medium (j), or pH7K-medium without sorbitol (e). (b) Effect of CCCP. Uptake was also determined when 10 mM CCCP was included in the pH5Na-medium (\), or pH7K-medium (%). Initial in¯uxes: 34"3 (k), 13.0"0.6 (j), 8.9"0.7 (\), and 5.3"1.0 (%). (c) Valine uptake in plasma membrane vesicles (k), microsomal vesicles (m), and vesicles from the L1-phase (^). Uptake was measured in pH5Na-medium. (d) Uptake by freshly prepared plasma membrane vesicles. Initial in¯uxes: 87"4 (k), and 37"4 (j). freezeuthaw treatment. In freshly prepared vesicles, the uptake of valine was about twice as high, both when DpHqDy were imposed and in the absence of gradients (Fig. 2d). Transport of L-lysine and L-glutamic acid The uptake-time curves for L-lysine were less regular than those for L-valine (Fig. 1b). Since extrapolation of the curves to time zero is problematic, no attempt was made to estimate the initial in¯uxes. In the absence of gradients a time-dependent uptake was discernible, but it occurred at a lower rate than that of valine. The uptake was slightly enhanced by the imposition DpH, and could be further increased by a factor of ;6 when Dy was imposed simultaneously. When imposed alone, Dy had a greater effect than DpH. In the absence of gradients the uptake of glutamic acid was very pronounced, being about twice as high as that of valine. But in contrast to the uptake of valine and lysine 1667 Fig. 3. Transport of sucrose into plasma membrane vesicles from pea cotyledons. The uptake of sucrose, supplied at a concentration of 0.74 mM, was determined after the imposition of DpHqDy, DpH alone, Dy alone, or in the absence of gradients by diluting vesicle suspensions in pH5Na-medium, pH5K-medium, pH7Na-medium or pH7K-medium, respectively. For clarity, and because Dy had no detectable effect, the data for the pH5-media were combined (k) as were the data for the pH7-media (j). Symbols represent mean values"SE of two measurements with ®ve vesicle preparations. The lines drawn resulted from a quadratic (k) or linear regression analysis (j), giving intercepts on the ordinate of 3.24"0.93 and 2.78"0.58, respectively. The estimated initial in¯uxes (pmol mg 1 protein min 1) were: 0.61"0.42 (k) and 0.091"0.087 (j). that of glutamic acid was not signi®cantly affected by the imposition of DpH anduor Dy (Fig. 1c). Transport of sucrose Uptake of sucrose by plasma membrane vesicles was very low (Fig. 3). When DpH or DpHqDy were imposed, the uptake amounted to ;6 pmol mg 1 protein after 10 min of incubation. After correction for the different substrate concentrations this uptake turns out to be about 25 times lower than for valine. A considerable part of the uptake appeared to be independent of the incubation time, as indicated by the intercepts of the uptake-time curves on the y-axis. Most likely, this time-independent uptake represents extravesicular label that was not removed during the washing of the vesicles on the ®lter membranes. It represents approximately 0.03% of the amount of label that was present in the 0.2 ml samples of the incubation mixture from which the vesicles were collected. Even though the uptake of sucrose could not be measured very accurately a signi®cant increase in the uptake after the imposition of DpH could be clearly demonstrated (Fig. 3). The imposition of Dy, either alone or in combination with DpH, had no detectable effect (not shown). After 10 min of incubation the DpH-dependent uptake of sucrose amounted to ;3 pmol mg 1 protein. 1668 de Jong and Borstlap Discussion Enrichment and recovery of plasma membranes by two-phase partitioning Preliminary experiments had shown that the use of twophase systems that contained 5 mM KCl and polymer concentrations of 5.3% or 5.5% resulted in plasma membrane fractions that were contaminated with chloroplast membranes. Yields of plasma membranes became extremely low when polymer concentrations were raised to 5.9%. Therefore, a two-phase system has been chosen with polymer concentrations of 5.7%, in which the KCl concentration was enhanced to 10 mM to reduce contamination with chloroplast membranes further. The plasma membrane fraction consisted of the combined upper phases (U3qU39) that were obtained by the partitioning of a microsomal fraction against three lower phase (L1 to L3). As compared with the microsomal fraction, the two plasma membrane markers were enriched in this fraction by 7- and 10-fold, respectively (Table 2). This is close to the enrichment factors for these marker enzymes (10 and 8, respectively) in highly puri®ed membrane vesicles from oat leaves (Larsson et al., 1987). Contamination of mitochondrial membranes in these preparations was very low. Only 0.01% of cytochrome c oxidase was recovered in (U3qU39), whereas Larsson et al. recovered 0.8% (Larsson et al., 1987). Unfortunately, the recovery of plasma membranes in (U3qU39) was poor (Table 1). As little as 3±4% of the plasma membrane markers in the microsomal fraction was recovered in (U3qU39), whereas >60% was retained in the ®rst lower phase L1. For comparison, Larsson et al. recovered ;75% of the marker enzymes in (U3qU39), and ;15% in L1 (Larsson et al., 1987). It may be concluded that with a two-phase system containing 5.7% of the polymers and 10 mM KCl low yields of highly puri®ed plasma membranes can be isolated from developing pea cotyledons. Hquamino acid symport During the development of the pea seed, a saturable system (Kmf5 mM) for amino acid (L-valine) uptake appears in the cotyledons when their water content has decreased to ;65%. The valine in¯ux by this system has been found to be enhanced about four times when the external wHqx was raised from 0.1 to 10 mM and to be inhibited by CCCP (Lanfermeijer et al., 1990). It seems very likely, therefore, that the DpH-dependent uptake of valine in the plasma membrane vesicles is effected by the same transporter as the saturable system identi®ed in experiments with cotyledons. Probably, this transport is electrogenic, for it was enhanced by the imposition of Dy (inside negative) across the vesicle membrane. Since Dy was imposed as a Kq-diffusion potential by a 15-fold dilution of a vesicle suspension containing 50 mM Kq into a potassium-free medium, its magnitude is expected to amount to ±58310log 15~ 69 mV. The imposition of Dy increased the valine in¯ux by about 20%. This is less than has been observed for low-af®nity amino acid transporters from Arabidopsis (AAP1uNAT2 and AAP5) expressed in Xenopus oocytes (Boorer et al., 1996a; Boorer and Fischer, 1997) which were stimulated about 2-fold when the membrane was hyperpolarized from 0 to 69 mV. The cationic amino acid L-lysine was also transported into the vesicles by a Hq-symport mechanism as evidenced by the stimulating effect of DpH, particularly when imposed together with Dy (Fig. 1b). It is possible that valine and lysine share the same transporter (Heremans et al., 1997). Low-af®nity Hq-symporters have been identi®ed that accept neutral as well as cationic amino acids as substrates. (Fischer et al., 1995; Boorer and Fischer, 1997). This would imply that transport of lysine across the membrane is accompanied by the movement of two positive electrical charges, which might explain that it was much more strongly stimulated by Dy than the transport of valine. However, the possibilty that lysine is partially transported by a uniporter cannot be excluded. Evidence that uniporters for cationic amino acids may occur in the plant plasma membrane has already been presented (Weston et al., 1995) when the uptake of lysine and arginine in vesicles from Ricinus roots was observed to be stimulated by Dy but not by DpH. Amino acid transport under non-energized conditions Even in the absence of DpH or Dy the uptake of L-valine in the vesicles was quite considerable amounting to ;50% of that measured after the imposition of DpHqDy. In freshly prepared vesicles the valine in¯ux driven by the proton motive force as well as the in¯ux under the non-energized condition was about twice as high as in vesicles after one freezeuthaw treatment. This may be taken as evidence that both ¯uxes are effected by the same transporter. Inhibition by CCCP of the valine in¯ux under the non-energized condition indicates that the protonophore also had a direct inhibiting effect on the transporter as has been described for a HquCl -symporter (Alvarado and Vasseur, 1998). The time-dependent uptake of labelled substrate in vesicles under non-energized conditions is generally thought to result from binding of the substrate to material inside the vesicles, or to the trapping of a product of enzymatic conversion (Dorando and Crane, 1984). An alternative explanation could be that the labelled substrate is exchanged against unlabelled substrate that was entrapped in the vesicles during their isolation. Developing pea cotyledons with a water content of Plasma membrane vesicles from cotyledons 1 ;55% contain about 50 mmol g FW of amino acids (Lanfermeijer et al., 1989). Since the cotyledons were homogenized in 2 vols of medium it is not unlikely that the total concentration of amino acids entrapped in the vesicles was in the order of 10 mM. Transport of L-glutamic acid Puzzling results were obtained in uptake experiments with glutamic acid (Fig. 1c). Under non-energized conditions, at an external pH 7, the uptake was very pronounced and approximately as high as the valine uptake after the imposition of DpHqDy. Obviously, glutamic acid can be accumulated in the vesicles by a DpH-independent transport mechanism. Glutamic acid is probably transported in its anionic form which, at pH 7, is by far the predominant ionic species with an abundance of 99.6%. But then a lower uptake is to be expected after the imposition of DpH, where the external pH was 5 and, consequently, the abundance of the anionic species has decreased to 85%. Perhaps this decrease was compensated by some DpH-dependent uptake. An attempt to test this supposition by using CCCP to eliminate possible DpH-dependent uptake failed, however, because the DpH-independent uptake of glutamate was also strongly inhibited by the protonophore (data not shown). Hqusucrose symport In isolated cotyledons (water content ;55%) the in¯uxes of L-valine and sucrose by the saturable systems, calculated for an external substrate concentration of 1 mM are approximately 30 and 15 pmol g 1 FW min 1, respectively. This contrasts with the results obtained with the plasma membrane vesicles, in which the sucrose in¯ux was about 50-fold lower than that of valine. The sucrose in¯ux into the vesicles may be so low because the media used in the ®rst steps of their isolation contained a high concentration (0.33 M) of sucrose. It can be envisaged that sucrose entrapped in the vesicles transinhibits the in¯ux of the labelled sucrose. Alternatively, the discrepancy between the valine- and sucrose in¯uxes in the vesicles and those in the isolated cotyledons may be due to a different distribution of the transporters in the cotyledonary tissue. The sucrose transporter is known to be restricted to the outer cell layers of the cotyledon (McDonald et al., 1996; Weber et al., 1997; Tegeder et al., 1999). If uptake of labelled substrates by isolated cotyledons is mainly brought about by the outer cell layers, and if the amino acid transporter is also present in the storage parenchyma, the amino acid transport activity in vesicles may be much higher than anticipated from uptake experiments with cotyledons. Because transport by a Hqusucrose symporter is electrogenic it will be more or less enhanced by Dy. An increase in the activity of the Hqusucrose symporter from 1669 pea cotyledons in response to Dy was probably too low to be detected in these experiments. This contrasts with sucrose transporters from leaves which have been clearly shown to be stimulated by Dy (Lemoine and Delrot, 1989; Boorer et al., 1996b). Sucrose transporters in cotyledons of developing legume seeds could be functionally somewhat different from those in leaves. Another indication for this is that their Kms (5±15 mM) are considerably higher than the Kms (0.5±1 mM) of sucrose transporters from leaves. Concluding remarks Aqueous polymer two-phase partitioning was used to obtain plasma membrane vesicles from developing pea cotyledons. If slight contaminations with other membranes are tolerated, the very low recovery (3%) of plasma membranes can probably be greatly enhanced by using a two-phase system in which the polymer concentrations are lowered from 5.7% to 5.5%. The demonstration of the activity of Hq-symporters for L-valine, L-lysine and sucrose provides additional evidence that cotyledonary cells take up sucrose and amino acids from the seed apoplasm by means of Hq-symporters. Further work is required to clarify the relatively low activity of the Hqusucrose transporter, the speci®city of the amino acid transporter(s), and the mechanism of glutamate transport. Acknowledgements The authors wish to thank Jolanda Schuurmans for her support and for carrying out some of the uptake experiments. References Alvarado F, Vasseur M. 1998. Direct inhibitory effect of CCCP on the Cl ÐHq symporter of the guinea pig ileal brushborder membrane. American Journal of Physiology 274, C481±491. Bennett AB, Spanswick RM. 1983. Derepression of amino acidHq cotransport in developing soybean embryos. Plant Physiology 72, 781±786. Boorer KJ, Fischer WN. 1997. Speci®city and stoichiometry of the Arabidopsis Hquamino acid transporter AAP5. The Journal of Biological Chemistry 272, 13040±13046. Boorer KJ, Frommer WB, Bush DR, Kreman M, Loo DDF, Wright EM. 1996a. Kinetics and speci®city of a Hquamino acid transporter from Arabidopsis thaliana. The Journal of Biological Chemistry 271, 2213±2220. Boorer KJ, Loo DDF, Frommer WB, Wright EM. 1996b. Transport mechanism of the cloned potato Hqusucrose cotransporter StSUT1. Journal of Biological Chemistry 271, 25139±25144. Borstlap AC. 1983. The use of model-®tting in the interpretation of `dual' uptake isotherms. Plant, Cell and Environment 6, 407±416. Bowman EJ, Siebers A, Altendorf K. 1988. Ba®lomycins: a class of inhibitors of membrane ATPases from microorganisms, 1670 de Jong and Borstlap animal cells, and plant cells. Proceedings of the National Academy of Sciences, USA 85, 7972±7976. Bush DR. 1993. Proton-coupled sugar and amino acid transporters in plants. Annual Review of Plant Physiology and Plant Molecular Biology 44, 513±542. Cocucci MC, MarreÁ E. 1984. Lysophosphatidylcholineactivated, vanadate-inhibited, Mg2q-ATPase from radish microsomes. Biochimica et Biophysica Acta 771, 42±52. De Jong A, Borstlap AC. 2000. A plasma membrane-enriched fraction isolated from the coats of developing pea seeds contains Hq-symporters for amino acids and sucrose. Journal of Experimental Botany 51, 1671±1677. De Jong A, Wolswinkel P. 1995. Differences in release of endogenous sugars and amino acids from attached and detached seed coats of developing pea seeds. Physiologia Plantarum 94, 78±86. Dorando FC, Crane RK. 1984. Studies of the kinetics of Naqgradient-coupled glucose transport as found in brushborder membrane vesicles from rabbit jejunum. Biochimica et Biophysica Acta 772, 273±287. Fischer WN, Kwart M, Hummel S, Frommer WB. 1995. Substrate speci®city and expression pro®le of amino acid transporters (AAPs) in Arabidopsis. Journal of Biological Chemistry 270, 16315±16320. Fredrikson K, Larsson C. 1989. Activation of 1,3-b-glucan synthase by Ca2q, spermine and cellobiose. Localization of activator sites using inside-out plasma membrane vesicles. Physiologia Plantarum 77, 196±201. Gallagher SR, Leonard RT. 1982. Effect of vanadate, molybdate, and azide on membrane-associated ATPase and soluble phosphatase activities of corn roots. Plant Physiology 70, 1335±1340. Harrington GN, Franceschi VR, Of¯er CE, Patrick JW, Tegeder M, Frommer WB, Harper JF, Hitz WD. 1997. Cell speci®c expression of three genes involved in plasma membrane sucrose transport in developing Vicia faba seed. Protoplasma 197, 160±173. Heremans B, Borstlap AC, Jacobs M. 1997. The rlt11 and raec1 mutants of Arabidopsis thaliana lack the activity of a basicamino-acid transporter. Planta 201, 219±226. Hirner B, Fischer WN, Rentsch D, Kwart M, Frommer WB. 1998. Developmental control of Hquamino acid permease gene expression during seed development of Arabidopsis. The Plant Journal 14, 535±544. Hodges TK, Leonard RT. 1974. Puri®cation of a plasma membrane-bound adenosine triphosphatase from plant roots. Methods in Enzymology 32, 392±406. Johansson F, Olbe M, Sommarin M, Larsson C. 1995. Brij 58, a polyoxyethylene acyl ether, creates membrane vesicles of uniform sidedness. A new tool to obtain inside-out (cytoplasmic side-out) plasma membrane vesicles. The Plant Journal 7, 165±173. Lanfermeijer F, Koerselman-Kooij JW, Borstlap AC. 1990. Changing kinetics of L-valine uptake by immature pea cotyledons during development: an unsaturable pathway is supplemented by a saturable system. Planta 181, 576±582. Lanfermeijer F, Koerselman-Kooij JW, Borstlap AC. 1991. Osmosensitivity of sucrose uptake by immature pea cotyledons disappears during development. Plant Physiology 95, 832±838. Lanfermeijer FC, Koerselman-Kooij JW, KolloÈffel C, Borstlap AC. 1989. Release of amino acids from cotyledons of developing seeds of pea (Pisum sativum L.). Journal of Plant Physiology 134, 592±597. Lanfermeijer FC, Van Oene MA, Borstlap AC. 1992. Compartmental analysis of amino-acid release from attached and detached pea seed coats. Planta 187, 75±82. Larsson C, Widell S, Kjellbom P. 1987. Preparation of highpurity plasma membranes. Methods in Enzymology 148, 558±568. Larsson C, Widell S, Sommarin M. 1988. Inside-out plant plasma membrane vesicles of high purity obtained by aqueous two-phase partitioning. FEBS Letters 229, 289±292. Lemoine R, Delrot S. 1989. Proton-motive-force-driven sucrose uptake in sugar beet plasma membrane vesicles. FEBS Letters 249, 129±133. Lichtner FT, Spanswick RM. 1981a. Electrogenic sucrose transport in developing soybean cotyledons. Plant Physiology 67, 869±874. Lichtner FT, Spanswick RM. 1981b. Sucrose uptake by developing soybean cotyledons. Plant Physiology 68, 693±698. McDonald R, Fieuw S, Patrick JW. 1996. Sugar uptake by the dermal transfer cells of developing cotyledons of Vicia faba L. 1. Experimental systems and general transport properties. Planta 198, 54±63. Palmgren MG, Sommarin M, Ulvskov P, Larsson C. 1990. Effect of detergents on the Hq-ATPase activity of inside-out and right-side-out plasma membrane vesicles. Biochimica et Biophysica Acta 1021, 133±140. Patrick JW. 1997. Phloem unloading: sieve element unloading and post-sieve element transport. Annual Review of Plant Physiology and Plant Molecular Biology 48, 191±222. Rochat C, Boutin JP. 1991. Metabolism of phloem-borne amino acids in maternal tissues of fruit of nodulated or nitrate-fed pea plants (Pisum sativum L.). Journal of Experimental Botany 42, 207±214. Tegeder M, Wang XD, Frommer WB, Of¯er CE, Patrick JW. 1999. Sucrose transport into developing seeds of Pisum sativum L. The Plant Journal 18, 151±161. Thorne JH. 1982. Characterization of the active sucrose transport system of immature soybean embryos. Plant Physiology 70, 953±958. Weber H, Borisjuk L, Heim U, Sauer N, Wobus U. 1997. A role for sugar transporters during seed development: molecular characterization of a hexose and a sucrose carrier in fava bean seeds. The Plant Cell 9, 895±908. Weston K, Hall JL, Williams LE. 1995. Characterization of amino-acid transport in Ricinus communis roots using isolated membrane vesicles. Planta 196, 166±173.