Aluminum-induced plasma membrane surface potential and H+-ATPase activity in near-isogenic wheat lines differing in tolerance to aluminum

Authors


Author for correspondence: H. Matsumoto Tel: +81 86 4341209 Fax: +81 86 4341249 Email: hmatsumo@rib.okayama-u.ac.jp

Summary

  • • This study aims to clarify the effect of aluminum (Al) on the surface potential and the plasma membrane (PM) H+-ATPase activity, as linked to exudation of organic acid anions, using near-isogenic Al-tolerant (ET8) and Al-sensitive (ES8) wheat (Triticum aestivum) lines.
  • • We examined the surface potential (ζ potential) and H+-ATPase activity of PM vesicles obtained from the root tips (10 mm) and the region distal to the root tip of wheat treated with Al either in vivo or in vitro.
  • • After 4-h in vivo treatment with Al (2.6 µm), H+-ATPase activity and H+ transport rate were decreased and ζ potential was depolarized in PM vesicles from root tips of Al-sensitive ES8 but not of Al-tolerant ET8. In vivo treatment with 50 µm malate alleviated deleterious Al effects on H+-ATPase activity and the ζ potential in PM vesicles isolated from root tips of ES8. In PM vesicles from root tips, in vitro treatment with 10 µm Al had a greater inhibitory effect on H+-ATPase activity of ES8 than that of ET8. However, in vitro Al treatment did not influence the H+-ATPase activity in PM vesicles prepared from the region distal to the root tip in the two wheat lines.
  • • Al-induced exudation of malate, as a basis for the mechanism of Al tolerance, is accompanied by changes in PM surface potential and the activation of H+-ATPase.

Introduction

Plant productivity is greatly influenced by environmental stresses, such as freezing, drought, flooding and metal toxicity. In particular, inhibition of root growth caused by an aluminum (Al)-induced impairment of cell division and elongation is a well-known early and dramatic symptom of Al phytotoxicity in acid soils (Horst, 1995; Kochian, 1995; Taylor, 1995; Matsumoto, 2000). The root apex is the primary target and plays a central role in Al tolerance and toxicity (Ryan et al., 1993; Sivaguru & Horst, 1998; Ahn et al., 2001). Nevertheless, mechanisms of Al tolerance and toxicity in plant cells are poorly understood (Kochian, 1995; Matsumoto, 2000). Exclusion of Al by chelation with organic acid anions, such as citrate (Miyasaka et al., 1991), malate (Delhaize et al., 1993) and oxalate (Ma et al., 1997), excreted from root apices has been suggested as an Al tolerance mechanism. Many other mechanisms of Al toxicity have also been proposed (for reviews see Delhaize & Ryan, 1995; Kochian, 1995; Matsumoto, 2000): for example the disruption of cytoplasmic calcium (Ca) homeostasis (Rengel, 1992; Jones et al., 1998; Zhang & Rengel, 1999; Rengel & Zhang, 2003) and changes in surface potential (Kinraide et al., 1992; Kinraide, 1994; Kinraide et al., 1998), with the induction of callose suggested to be a characteristic marker for Al toxicity (Horst et al., 1997).

The role of plasma membrane (PM) in Al tolerance has been of interest for many years (Ahn et al., 2002). The PM is a barrier to the passive movement of ions into root cells, and several responses of plant cells to Al are related to the alteration of PM properties (Rengel, 1996; Kochian & Jones, 1997). The root zone-specific depolarization of the PM surface potential in squash (Cucurbita pepo) was recently shown to be an early symptom of Al toxicity (Ahn et al., 2001). Sivaguru et al. (1999) found that Al caused instantaneous PM depolarization in root cells of an Al-sensitive maize cultivar, and that the intensity of depolarization varied with the root growth zone. In addition, it was proposed decades ago that negative surface-charge densities might be lower in Al-tolerant genotypes (Vose & Randall, 1962). This idea was supported by Wagatsuma & Akiba (1989) who established the relationship between the surface negativity of root protoplasts and Al tolerance using five plant species. More recently, Yermiyahu et al. (1997) found that the genotypic differences in surface charge density could account only partly for the differential genotypic tolerance to Al.

The modulation of H+-ATPase activity, and thus H+ pumping, is involved in a wide range of fundamental cellular processes, such as the formation and maintenance of an electrochemical gradient that serves as the driving force for the secondary ion transport. The H+-ATPase activity plays a role in cell expansion, initiation of cell division, nutritional status, and the response to phytotoxins (Serrano, 1985; Axelsen & Palmgren, 2001). Recently, Yan et al. (2002) demonstrated a positive relationship between root exudation of organic acids and the H+-ATPase-mediated H+ extrusion in active proteoid roots of white lupin under phosphorus deficiency. Under Al stress, intact wheat seedlings have an undiminished or even enhanced capacity for proton extrusion (Kinraide, 1988). By contrast, PM vesicles isolated from Al-treated seedlings exhibit a diminished PM H+-ATPase activity (Matsumoto, 1988; Sasaki et al., 1995; Hamilton et al., 2001). Therefore, comparison of two near-isogenic wheat genotypes differing in Al tolerance may help shed more light on the mechanisms of Al tolerance. In this paper, we report on H+-ATPase activity and other properties of PM vesicles prepared from the tips and more distal parts of roots of Al-sensitive (ES8) and Al-tolerant (ET8) near-isogenic wheat lines after the treatment with Al in vivo and in vitro. The role of PM in malate exudation as the central tolerance mechanism used by a near-isogenic Al-tolerant wheat line exposed to Al is also discussed.

Materials and Methods

Growth conditions

Near-isogenic wheat (Triticum aestivum L.) lines differing in Al tolerance at the Alt1 locus were used. The Al-tolerant (ET8) and Al-sensitive lines (ES8) were derived from a cross between the Al-tolerant cultivar Carazinho and the Al-sensitive ‘Egret’, with the resulting progeny backcrossed eight times to ‘Egret’ or derivatives of ‘Egret’ (Fisher & Scott, 1987). Seeds were surface-sterilized with 1.5% (v/v) NaOCl + 1.0% SDS (sodium dodecyl sulfate) for 10 min, rinsed thoroughly with tap water, and germinated on filter paper moistened with 0.2 mm CaCl2 solution (pH 4.5) at 25°C in the dark for 2 d. Germinated seedlings were transferred to the identical solution in polyethylene pots and grown in a controlled-environment chamber with a 14 h light/10 h dark cycle under 580 µmol m−2 s−1 light for 5 d. The light/dark temperatures were set at 25°C/20°C and relative humidity was kept at 65%. The solution was renewed daily.

After the Al treatment with 2.6 µm total concentration for 0, 1 or 4 h, the roots were washed three times with double-distilled water, and the root tips (0–10 mm, as measured from the root tip including the root cap) as well as the regions distal to the tip (10–20 mm) were excised using a razor blade. Four excised segments (representing one replicate) were transferred to 1.5-ml Eppendorf (Netheler–Hinz Gmbh, Hamburg, Germany) tubes each containing 1 ml of 2 m HCl for 48 h. The Al content in the HCl digest of root tissue was determined by an atomic absorption spectrophotometer (Z-8270; Hitachi, Tokyo, Japan) after dilution. In some experiments malate (5 or 50 µm) was added to toxic Al solutions to check its ability to ameliorate the Al stress. Malate stock solutions were adjusted to pH 4.5 and were added to 0.2 mm CaCl2 solution that contained 2.6 µm Al (Delhaize et al., 1993b).

Determination of ζ potential and H+-ATPase activity using PM vesicles

The plants were grown in 0.2 mm CaCl2 solution adjusted to pH 4.5 for 5 d from germination and then treated with Al (2.6 µm) only or with Al and malate (5 or 50 µm) for 4 h in vivo before isolation of PM vesicles. The PM vesicles were isolated by the two-phase partitioning method of Palmgren et al. (1990), with slight modification as described previously (Ahn et al., 2001). The root tips (0–10 mm) and the regions distal to the tip (10–20 mm), about 5 g fresh wt each, were obtained from the plants treated with Al for isolation of PM vesicles. The H+-ATPase activity in the PM vesicles was measured following the method of Ahn et al. (2001).

Proton uptake into Brij 58 (9 polyoxyethylene acylether)-induced inside-out vesicles (5 µg of protein) (Palmgren et al., 1990) treated with and without Al in vivo was monitored as the decrease in the intensity of the acridine orange absorbance at 495 nm. The assay medium consisted of 20 µm acridine orange, 2 mm ATP-BTP (bis-tris propane) (pH 7.0), 140 mm KCl, 4 mm MgCl2, 1 mm ethylenediaminetetraacetic acid (EDTA), 1 mm dithiothreitol (DTT), 1 mg ml−1 bovine serum albumin (BSA) (essentially fatty acid free), and 5 µg PM protein in a total volume of 2 ml. A total of 5 µm gramicidin was added as indicated in Fig. 3b. After 5 min preincubation at 20°C, the reaction was initiated by addition of ATP, and the pH gradient formation was monitored until it reached steady state. The rate of H+ accumulation was estimated from the initial slope of absorbance quenching (ΔA495) of acridine orange.

Figure 3.

Effect of aluminum (Al) treatment (2.6 µm) for 4 h on the H+ transport into inside-out plasma membrane (PM) vesicles isolated from root tips (0–10 mm) (a,c) and the region distal to the tip (10–20 mm) (b,d) of ET8 (a,b) and ES8 (c,d). Wheat (Triticum aestivum) plants were grown in 0.2 mm CaCl2 solution for 5 d from germination and were then treated with 2.6 µM Al (pH 4.5) for 4 h before isolation of PM vesicles. The pH gradient formation was determined by monitoring ΔA495 of acridine orange. In the assay solution at pH 6.5, intravesicular acidification was initiated by addition of 2 mm ATP. The pH gradient formation did not occur without ATP. The established pH gradient was completely abolished by the addition of 5.0 µm gramicidin (G). A representative example of four replications from different vesicle preparations that gave similar results is presented. Asterisks show statistically different means between –Al and +Al treatment: *, P < 0.05 (Student's t-test).

To examine the effect of Al on H+-ATPase activity in vitro, 5 µg of PM vesicles isolated from plants grown without Al were exposed to Al (0, 2.5, 5 or 10 µm) for 10 min, suspended in a dilution buffer consisting of 330 mm sucrose, 5 mm MOPS (3-[N-Morpholino] propanesulfonic acid)-BTP (pH 6.5), and 5 mm KCl followed by centrifugation (100 000 g for 1 h) to remove the unbound Al. The membrane protein was determined by the Bradford method (1976) using BSA as a standard.

The ζ potential (which approximates the surface potential) was calculated using a light scattering electrophoresis apparatus (ELS-8000; Otsuka Electronics Ltd, Japan). The ζ potential of PM vesicles isolated from root tips and the region distal to the root tips treated with or without Al was calculated from the mobility in free-flow electrophoresis. The ζ potential was determined using Latex (Dow Chemical, Tokyo, Japan) as a standard for electrophoresis. The ζ potential of PM vesicles was calculated from the electrophoretic mobility (µ) using the Smoluchowski equation:

ζ = µη/ɛɛo

(η is the viscosity; ɛ is the relative dielectric constants of the buffer; ɛo is the dielectric constant of vacuum) (Gimmler et al., 1991). All other parameters were determined as described previously (Ahn et al., 2001).

Western blotting and immunolocalization of PM H+-ATPase

The PM H+-ATPase (approximately 100 kDa) was detected using a standard Western-blot protocol with an antibody following determination of protein content using the Bradford method (1976). Immunolocalization of H+-ATPase was examined as described by Ahn et al. (2001) with a slight modification. The images of H+-ATPase from roots were obtained using the 543 nm excitation line of He–Ne laser fitted to a Zeiss (Oberkochen, Germany) confocal microscope using a Ph3-Plan-Neofluar 100 × oil immersion (1.3 NA) objective. The root surface images were the overlay of 7–11 optical sections (each 0.75 µm thick); scan configurations were kept constant using the recycle option of the LSM 510 (Oberkochen, Germany) software to assess the intensity differences among treatments. All images were organized using Adobe Photoshop 4.0 J (Adobe Systems Inc., San Jose, CA, USA) and printed using a Fujix-3000 Pictography digital printer (Fuji-Film, Tokyo, Japan).

Results

Growth inhibition by Al and Al content in root apex

According to Zhang & Rengel (1999), the root growth of both ET8 and ES8 was decreased by exposure to 50 µm Al in 0.2 mm CaCl2 solution (pH 4.2). For example, after a 1-h exposure to 50 µm Al, root growth of ES8 was inhibited by 46% and that of ET8 by only 18%. In addition, root growth of ES8 was inhibited 20% and 60%, respectively, by 1-h and 4-h treatments with 2.6 µm Al in 0.2 mm CaCl2 solution (pH 4.5), while root growth of ET8 was little affected. In the present study, we quantified Al content in the root tip (0–10 mm from the tip) and the region 10–20 mm from the tip of ET8 and ES8 lines after the treatment with 2.6 µm Al. The amount of Al accumulated in ES8 (6.20 ± 1.2 nmol (root apex)−1) was 2.5-fold greater than that in ET8 (2.52 nmol ± 0.8 (root apex)−1). By contrast, the region 10–20 mm from the root tip of either genotype did not accumulate measurable amounts of Al during the first 4 h of exposure.

Impact of Al on H+-ATPase activity and ζ potential of PM vesicles in Al-tolerant and Al-sensitive genotypes

The H+-ATPase activity in PM vesicles isolated from the root tip and the region distal to the tip was increased by the addition of Brij 58, indicating that approximately 83% of the vesicles were sealed and right-side-out forms. A specific inhibitor of PM ATPase activity, vanadate, inhibited the activity by about 88%, while inhibitors of vacuolar (nitrate) and mitochondrial (azide) ATPase activities caused less than 12% inhibition (data not shown). The Al treatment depolarized the ζ potential of PM vesicles isolated from the root tips of ES8 significantly (from −18.5 to −14.3 mV; P < 0.025, Student's t-test) but hyperpolarized it in ET8 (from −15.1 to −17.7 mV; P < 0.05, Student's t-test). In the region distal to the tip, Al influenced the ζ potential of PM only slightly in both wheat lines (Fig. 1a). The treatment of plants with 2.6 µm Al for 4 h increased the H+-ATPase activity in the PM vesicles isolated from root tips of ET8 (from 26 to 31 µmol Pi mg−1 protein h−1), but decreased it in ES8 (from 37 to 27 µmol Pi (phosphate) mg−1 protein h−1) (P < 0.05, Student's t-test). Treatment with Al, however, increased the H+-ATPase activity in the region distal to the root tip in both ET8 (from 29 to 38 µmol Pi mg−1 protein h−1) and ES8 (from 34 to 42 µmol Pi mg−1 protein h−1) (Fig. 1b). In addition, a positive relationship was obtained between the ζ potential and H+-ATPase activity in PM vesicles isolated from the root tip and the region distal to the tip of both ET8 and ES8 grown under Al stress (data not shown). Inclusion of malate in the solution during the 4-h Al treatment in vivo prevented the deleterious effects of Al on the ζ potential and H+-ATPase activity in the PM vesicles isolated from the root tips of ES8 (Fig. 2).

Figure 1.

Effect of aluminum (Al) treatment (2.6 µm) for 4 h on the surface potential (ζ potential) (a) and the H+-ATPase activity (b) of plasma membrane (PM) vesicles isolated from root tips (0–10 mm) and the region distal to the tip (10–20 mm) of ET8 and ES8 near-isogenic wheat (Triticum aestivum) lines. Plants were grown in 0.2 mm CaCl2 solution for 5 d from germination and were then treated with 2.6 µm Al (pH 4.5) for 4 h before isolation of PM vesicles. The electrophoresis medium was free of Al. Values are means ± SE (n = 4) of three separate experiments. Asterisks indicate statistically different means between –Al (closed columns) and +Al (open columns) treatment: *, P < 0.05; **, P < 0.025 (Student's t-test).

Figure 2.

Amelioration of aluminum (Al)-induced depolarization of the surface potential (ζ potential) (a) and inhibition of H+-ATPase activity (b) by malate. Wheat (Triticum aestivum) plants were grown in 0.2 mm CaCl2 solution adjusted to pH 4.5 for 5 d from germination and were then treated with Al (2.6 µm) only, or with 2.6 µm Al and malate (5 µm and 50 µm), for 4 h in vivo. The ζ potential and H+-ATPase activity of PM vesicles isolated from tips (0–10 mm) of ES8 wheat line were measured. The electrophoresis medium was free of Al. Values are means ± SE (n = 3) of three separate experiments.

The pH gradient formation across PM vesicles was determined by monitoring ΔA495 of acridine orange using inside-out PM vesicles (Johansson et al., 1995). Acridine orange did not quench without ATP (Fig. 3a), and 5 µm gramicidin dissipated the ΔpH (Fig. 3b). Hence, the vesicles were not leaky to H+ and could transport H+ into the vesicles only in an ATP-dependent manner. Treatment with 2.6 µm Al for 4 h decreased the H+ transport by 20% in vesicles from ES8 root tips and increased it by 10% in vesicles from ET8 root tips compared to control (Fig. 3a,c). The H+-transport activity in the region distal to the tip increased in vesicles isolated from both wheat lines (Fig. 3b,d). Therefore, changes in the H+ transport rates were consistent with changes in H+-ATPase activity and the ζ potential (Fig. 1).

Localization of PM H+-ATPase in root apex

In the Western blot analysis the polyclonal anti-maize PM H+-ATPase antibody gave a single band at approximately 100 kDa in the PM vesicle protein isolated from wheat roots, showing a high specificity to wheat PM H+-ATPase (data not shown). The polyclonal antibody also decorated the PM of all cells along the root apex of the control plants (Fig. 4). Higher H+-ATPase fluorescence was observed in the ET8 root cells than in the ES8 cells in both the apical and basal regions. The cells in the apical (2–3 mm) (set e) and the basal region (4–5 mm) (set c) in ET8 showed more intense fluorescence compared with those in ES8 (sets d and f). The fluorescence intensity in both apical and basal ES8 root cells decreased after 4 h of Al treatment (d′ and f′). This may be caused by an Al-induced decrease in the H+-ATPase protein per unit area of the PM (Yan et al., 1998) or alteration to stoichiometric configuration of the autoinhibitory domain in the C-terminus of H+-ATPase (Sze et al., 1999) concomitant with inhibition of enzyme activity by direct interaction with Al, limiting the antibody cross reactivity with H+-ATPase.

Figure 4.

Immunolocalization of plasma membrane (PM) H+-ATPase in the root apex of wheat (Triticum aestivum) ET8 (sets c and e) and ES8 (sets d and f). Confocal images of negative control of roots incubated without antibody are also shown (sets a and b). The roots were labeled with an anti-PM H+-ATPase antibody after 0 h (c, d, e and f), 1 h (c′, d′, e′ and f′) and 4 h (c″, d″, e″ and f″) of treatment with 2.6 µm aluminum (Al). The images are from the regions 2–3 mm (sets e and f) and 4–5 mm (sets c and d) from the root tip. Note a decrease in the intensity of PM H+-ATPase in ES8 after the Al treatment. The experiment was replicated three times independently, giving similar results. Images from a representative experiment are shown. Bar, 40 µm.

Effect of the in vitro Al treatment on ζ potential and H+-ATPase activity in isolated PM vesicles

Malate chelates Al and thus ameliorates Al toxicity (see Fig. 2), possibly decreasing Al accumulation in roots. In the present study, deleterious effects of Al on the ζ potential and H+ transport were considerably greater in ES8 than ET8. The question arises whether tolerance of ET8 to Al is caused by (1) decreased accumulation of Al as a consequence of exudation of malate, or (2) by a direct response of PM to toxic Al. Therefore, we analysed the direct effects of Al in vitro on the ζ potential and H+-ATPase activity in the PM vesicles isolated from the root tips of ES8 and ET8. The PM vesicles were exposed to various concentrations of Al (0, 2.5, 5, 10, 20 or 50 µm) for 10 min in vitro, as reported previously (Ahn et al., 2001). In control (0 Al), the ζ potential of PM vesicles was more negative in ES8 than in ET8 (Fig. 5). The change in the ζ potential in response to 10 µm Al was significantly smaller in ET8 than in ES8 (P < 0.05, Student's t-test). However, the ζ potential was depolarized markedly by 20 µm Al in both ES8 (from −18.5 mV to −1.0 mV) and ET8 (from −15.5 mV to −3.0 mV) (Fig. 5). The ζ potential was more negative in ES8 compared with ET8 at Al concentrations lower than 5 µm, but less negative at higher Al concentrations. However, it should be noted here that the electrophoresis medium buffer for the assay of ζ potential was adjusted to pH 6.5 to maintain the functional integrity of the PM vesicles during measurements. At this pH, Al3+ concentration is very low, but even if the concentration is extremely low, Al3+ might be bound because it has a very high affinity to the PM.

Figure 5.

Effects of in vitro aluminum (Al) concentrations, 0, 2.5, 5, and 10 µm (a), and 20 µm and 50 µm (b) on the surface potential (ζ potential) of plasma membrane (PM) vesicles isolated from the root tip (apical 10 mm) of wheat (Triticum aestivum) ET8 (closed circles) and ES8 (open circles) grown without Al for 5 d. PM protein (5 µg, pH 7.4) was treated with Al for 10 min in vitro, and the ζ potential was measured immediately afterwards. The electrophoresis medium was free of Al. The experiment was conducted three times independently, giving similar results. Values are means ± SE (n = 3). Asterisks show statistically different means: *, P < 0.05 (Student's t-test).

The Al treatment (5–10 µm) decreased the relative H+-ATPase activity in the PM vesicles isolated from the root tip of ES8 (P < 0.05, Student's t-test) (Fig. 6b), while this activity remained generally unchanged in ET8 (Fig. 6a). The relative H+-ATPase activity in the PM vesicles isolated from the region distal to the tip was not influenced by the Al treatment in either ET8 or ES8. Furthermore, Al at higher concentrations (20 µm and 50 µm) completely inhibited the relative H+-ATPase activity in the PM vesicles isolated from either the root tip or the region distal to the tip in both genotypes (data not shown).

Figure 6.

Effects of aluminum (Al) concentrations (0, 2.5, 5, and 10 µm) in vitro on the relative H+-ATPase activity in plasma membrane (PM) vesicles isolated from the roots of wheat (Triticum aestivum) ET8 (a) and ES8 (b). Seedlings were grown in the Al-free 0.2 mm CaCl2 solution adjusted to pH 4.5 for 5 d after germination. The PM vesicles were isolated from the root tips (0–10 mm; closed circles) and the region distal to the tip (10–20 mm; open circles). PM protein (5 µg, pH 7.4) was treated with Al at various concentrations for 10 min in vitro and centrifuged at 100 000 g for 1 h to minimize carry-over of Al from the treatment solution. This experiment was conducted three times, and values from a representative experiment are means ± SE (n = 3). Asterisks show statistically different means: *, P < 0.05 (Student's t-test).

Discussion

ET8 and ES8 are near-isogenic lines of wheat, differing in Al tolerance at a single locus (Alt1). The Al-tolerant line ET8 excretes 10-fold more malate than the Al-sensitive line ES8 under Al stress (Delhaize et al., 1993b). This mechanism of Al tolerance is closely associated with the Al-induced activation of malate-permeable channels in the PM (Ryan et al., 1997; Zhang et al., 2001). Based on the role of PM in malate exudation as the central mechanism by the near-isogenic Al-tolerant wheat line exposed to Al, the main purpose of this study was to clarify the effect of Al on the surface potential and the PM H+-ATPase activity, and to examine whether Al treatment of PM vesicles in vitro, whereby the organic acid exudation mechanism is avoided, can alter the surface potential and PM H+-ATPase activity caused by inherent differences in the membrane composition.

The PM membrane potential in plant cells is maintained mainly by H+-ATPase. The activity of this enzyme plays a central role in the functional association of PM surface charge and the H+ efflux, and is a crucial factor for plant survival under various environmental stresses. For example, a decrease in PM surface potential is correlated with the decline in H+-ATPase activity in PM vesicles isolated from tomato roots grown under salt stress (Suhayda et al., 1990) and squash roots grown under Al toxicity (Ahn et al., 2001). In line with this idea, Gimmler et al. (1991) emphasized the importance of ζ potential in metal toxicity in acid-resistant and salt-resistant green algae.

Nevertheless, the relationship between Al toxicity and H+-ATPase activity remains controversial. In intact seedlings of wheat, Al did not diminish and even enhanced the capacity for H+ secretion (Kinraide, 1988, 1993). By contrast, Al diminished the H+-ATPase activity in the PM vesicles prepared from Al-treated seedlings of barley (Matsumoto, 1988; Matsumoto et al., 1992), wheat (Sasaki et al., 1995; Hamilton et al., 2001) and squash (Ahn et al., 2001). In animals, Al inhibited the lysosomal H+-pump activity in rat liver (Zatta et al., 2000). In the present study, the PM ζ potential of control roots of ES8 was more negative than that of ET8. Hence, we suggest that, upon exposure to Al, root-cell PM of ES8 attracts more Al than that of ET8. In addition, PM H+-ATPase activity was higher in control roots of ES8 than in those of ET8 (Fig. 1). In the presence of Al, the rate of depolarization of PM ζ potential in ES8 was always higher than in ET8. This is probably related to the decrease in pumping out of H+ through PM in Al-sensitive plants as we reported previously (Ahn et al., 2001). The ζ potential of PM in the root tips of ES8 plants exposed to 2.6 µm Al for 4 h was depolarized by the same treatment, accompanied by a decrease in H+-ATPase and H+ transport (Figs 1 and 3). The Al treatment (e.g. 2.6 µm Al) inhibited root growth in ES8 but not in ET8 (Zhang & Rengel, 1999) because ET8 exudes large amounts of malate under Al stress, with malate chelating Al into nontoxic form (Delhaize et al., 1993b). Aluminum binding to the PM surface occurs quickly and precedes the exudation of malate and may be enhanced by an activation of anion channels. As Al levels at the PM surface decline, the surface potential recovers to more negative values (hyperpolarization) (Fig. 1a) because of the accumulation of malate and H+-ATPase becoming less inhibited (Fig. 1b). Recently, Kinraide (2001) showed no evidence that the ζ potential has anything to do with the PM channel activity. However, anion channels activating at hyperpolarized membrane potentials have been reported (Terry et al., 1991). Therefore, it is important to emphasize that exudation of malate via Al-induced anion channels in ET8 (see Zhang et al., 2001) might be linked to malate-induced hyperpolarization of the PM ζ potential. However, such observation might be a consequence of greater Al binding to the root-cell PM in Al-sensitive than in Al-tolerant genotypes (because of more negative charges in the former). As a consequence, the root-cell PM in the Al-sensitive genotype gets depolarized and H+-ATPase activity inhibited to a greater extent than in the Al-tolerant genotype that continues to exude malate, thus hyperpolarizing the PM.

Immunofluoresence studies using the maize PM H+-ATPase antibody coupled with confocal laser scanning microscopy showed that the amount of H+-ATPase in the apical (2–3 mm) and basal (4–5 mm) cells in ES8 roots decreased slightly after 4 h of Al treatment (Fig. 4). The decrease noted in the immunolocalization of H+-ATPase may have been caused by the slight structural changes that prevented the binding of antibody to the H+-ATPase rather than the decomposition of the protein itself because we found no obvious differences by Western blotting in tips and the region distal to the tips in ET8 and ES8 plants treated with 0 µm and 2.6 µm for 4 h (data not shown). The structural alteration might have resulted from the conformational change in the autoinhibitory domain in the C-terminus of H+-ATPase (Sze et al., 1999). Haug (1984) proposed that Al could interact with proteins by either competing with other cations for negatively charged binding sites, or by inducing conformational changes, which could influence ion–protein interactions.

The ET8 wheat plants exposed to Al (in vivo treatment) were affected less than ES8 plants (taking into account H+-ATPase activity, H+ transport rate and the ζ potential). The question arises whether a lack of strong Al effect on ET8 is caused by (1) decreased accumulation of Al because of the exudation of malate, or (2) by resistant properties of PM in response to toxic Al. To answer that question, we investigated the Al effect on the H+-ATPase activity and surface potential in the PM vesicles in vitro (thus avoiding the exudation of malate). Interestingly, low concentrations of Al caused a slight depolarization of the ζ potential (Fig. 5) and a decrease in H+-ATPase activity (Fig. 6) only in the PM vesicles isolated from the ES8 root tips, while relatively high concentration of Al caused a similar type of change in the PM vesicles isolated from either the ES8 or ET8 root tips. This difference in the direct effect of Al on the PM membrane properties might be a factor contributing to the difference in Al tolerance between ET8 and ES8; nevertheless, it should be pointed out that the difference in malate exudation between these two near-isogenic lines is 10-fold compared with the smaller differences observed in PM membrane properties upon exposure to Al.

This study pointed out that ET8 and ES8 have slightly different PM properties, even in the absence of Al, indicating that Alt1 locus identified as differing between the two lines is potentially pleiotropic (i.e. causing differences in the phenotype), or that the difference between the two lines is caused by more than one gene. However, it cannot be excluded that the primary role of the Alt1 locus is to code for differences in electrical surface properties and H+-ATPase activities, which then activate anion channels to balance charges across the PM and maintain cytosolic pH within an optimal metabolic range (Lindberg & Strid, 1997; Yan et al., 2002). Further research is required to clarify these issues.

Acknowledgements

This work was supported by the Program for the Promotion of Basic Research Activities in Innovative Biosciences (PROBRAIN), Ministry of Agriculture, Forests and Fisheries, Japan (to H. M.), Grant-in-Aid for General Scientific Research (A) from the Ministry of Education, Science, Sports and Culture of Japan (to H. M.), and Ohara Foundation for Agricultural Sciences. We would like to thank Prof. Hideo Sasagawa (Okayama University) for supplying H+-ATPase primary antibody. We are indebted to Michiyo Ariyoshi for her experimental assistance in the laboratory.

Ancillary