INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

P is one of the most important plant nutrients that significantly affect growth and metabolism. Although the total amount of P in soil may be high, it is often present in unavailable forms such as phytic acid (Richardson, 1994), or Ca, Fe, and Al phosphates (Holford, 1997). Low availability of P is a major constraint for crop production in many low-input systems of agriculture worldwide, especially in the highly weathered soils of the humid tropics and subtropics, in many sandy soils of the semiarid tropics, and in calcareous soils of the temperate regions, where crop productivity is severely compromised through lack of available P (Raghothama, 1999). Also, after application of P to the soil the recovery of applied P by crop plants in a growing season is very low, because in the soil more than 80% of the P becomes immobile and unavailable for plant uptake due to adsorption on Al or Fe oxides/hydroxides, precipitation with Ca, or conversion to organic forms (Holford, 1997).

Higher plants have developed various strategies of acquiring sparingly soluble nutrients from soil. In response to P deficiency, various species from different families develop so-called proteoid roots. These are bottlebrush-like clusters of rootlets of limited growth with an average length of 0.5 to 1 cm. The rootlets are closely arranged along lateral roots and are usually covered with long and dense root hairs (Purnell, 1960; Dinkelaker et al., 1995; Watt and Evans, 1999b). The name, proteoid roots, derives from the fact that most species in the family Proteaceae can develop such root clusters when they are grown in infertile soils (Purnell, 1960; Dinkelaker et al., 1995). One of the most remarkable characteristics of proteoid roots is that they are able to strongly acidify the rhizosphere soil. In a calcareous soil (20% CaCO₃), proteoid roots of white lupin (Lupinus albus) acidified rhizosphere from pH 7.5 to 4.8 (Dinkelaker et al., 1989). In some cases, the pH of rhizosphere soil can be decreased by proteoid roots as low as pH 3.6 (Li et al., 1997). This pH decrease in the rhizosphere can dissolve P from Ca phosphate and increase P availability in calcareous soils. It has been well established that this acidification by proteoid roots is attributed to the release of a huge amount of organic acids, predominantly as citric and malic acids, into rhizosphere under P deficiency conditions (Gardner et al., 1983; Dinkelaker et al., 1989; Li et al., 1997; Keerthisinghe et al., 1998; Neumann et al., 1999). A release rate of organic acids in the range of 2.4 to 7.4 µmol (h g fresh weight)¹ has been reported for proteoid roots of white lupin (Keerthisinghe et al., 1998; Neumann et al., 1999). In addition, because the rootlets are closely arranged, the released organic acids can accumulate to a high concentration in the rhizosphere of proteoid roots. About 0.1 mmol citric acid per g soil has been reported in the rhizosphere soil of proteoid roots of white lupin (Dinkelaker et al., 1989; Gerke et al., 1994; Li et al., 1997). This concentration was sufficient to release P from sparingly soluble Fe and Al phosphate (Gerke et al., 1994) by mechanisms of ligand exchange or chelation of metal ions (Hinsinger, 1998). Besides organic acids, proteoid roots also release large amounts of acid phosphatases into rhizosphere (Dinkelaker et al., 1995; Gilbert et al., 1999; Neumann et al., 1999; Miller et al., 2001), which can efficiently release P from organic P compounds in an acidified rhizosphere (Li et al., 1997; Neumann et al., 2000). Because of such efficient adaptation ability, all species forming proteoid roots can grow in soils with poorly available P in natural ecosystems. Of the species that form proteoid roots, white lupin is the only one currently used in agriculture and the one that has been most intensively studied (Watt and Evans, 1999b). In fact, in the investigations on adaptation mechanisms of higher plants to P deficiency, white lupin has become a model plant (Johnson et al., 1996b; Neumann et al., 1999; Watt and Evans, 1999b).

It has been repeatedly demonstrated that citric acid is the predominant acid released by proteoid roots of white lupin under P-deficient conditions (Dinkelaker et al., 1989; Johnson et al., 1996a, 1996b; Li et al., 1997; Neumann et al., 1999). The amount of released citric acid can represent as much as 11% (Gardner et al., 1983) to 23% (Dinkelaker et al., 1989) of the total plant dry weight, depending on physiological development stages and the severity of the P stress. The highest exudation activity of citric acid is related to the mature root clusters, whereas young and old clusters release only a limited amount of acids (Keerthisinghe et al., 1998; Neumann et al., 1999; Watt and Evans, 1999a). Biochemical studies reveal that increased release of citric acid by proteoid roots is related to the enhanced synthesis of this organic acid in proteoid root cells (Johnson et al., 1994; Neumann et al., 1999). In addition, the activity of phosphoenolpyruvate carboxylase has been demonstrated to be related to the accumulation and release of citric acid by proteoid roots (Johnson et al., 1994, 1996a, 1996b; Keerthisinghe et al., 1998; Neumann et al., 1999; Watt and Evans, 1999a). About 30% of released carbon in the form of organic acids, mainly citric acid, originated from dark CO₂ fixation by phosphoenolpyruvate carboxylase in roots of P-deficient white lupin (Johnson et al., 1996a).

The understanding of the mechanisms underlying the release of organic acids by plant roots is limited. Because of high cytosolic pH (7-7.5), citric acid dissociates into citrate and H⁺ in cytosol. This implies that the release of citric acid by proteoid roots must be attributed to at least two structurally separated plasma membrane transport processes: citrate export and proton export (Fig. 1). There is indication for an anion channel-mediated citrate export out of proteoid root cells (Neumann et al., 1999). However, so far there is no investigation into the mechanisms of enhanced proton export out of proteoid root cells. It is generally accepted that plasma membrane H⁺ ATPase (EC 3.6.1.35) is responsible for the export of protons out of plant cells (Serrano, 1989). This enzyme acts as a primary transporter by pumping protons out of the cell, thereby creating pH and electric potential differences across the plasma membrane. In this way, a large amount of protons produced during the synthesis of organic acids may be removed out of the cells. In addition, it is well known that the transport of many solutes such as ions or metabolites involves secondary transporters that are driven by the proton motive force created by the H⁺ ATPase. This may be also true for the release of organic anions such as citrate out of proteoid root cells. If this holds true, an adaptation of plasma membrane H⁺ ATPase in active proteoid roots can be expected. In fact, it has been demonstrated that plasma membrane H⁺ ATPase responds to a number of environmental factors, such as saline stress (Niu et al., 1993), low solution pH (Yan et al., 1998), nutrient supply (Schubert and Yan, 1997), and Fe deficiency (Dellórto et al., 2000). It is feasible to hypothesize that plasma membrane H⁺ ATPase of proteoid root cells may be involved in the release of citric acid and possibly plays a central role in the adaptation of white lupin to P deficiency.

View larger version (16K): [in this window] [in a new window]   Figure 1.   Hypothetical mechanism for the release of organic acids by proteoid root cells of white lupin.

In this study, we tested our hypothesis of involvement of plasma membrane H⁺ ATPase in rhizosphere acidification and investigated its contribution to the adaptation to P deficiency in proteoid root cells of white lupin. Plasma membrane was isolated from proteoid roots actively releasing protons. In vitro, enzyme kinetics of plasma membrane H⁺ ATPase was compared among active proteoid roots of P-deficient plants, proteoid roots of P-sufficient plants, and lateral roots of P-sufficient and -deficient plants. The aim of our investigations was to reveal the mechanisms involved in rhizosphere acidification by active proteoid roots of white lupin in response to P deficiency.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Development and Acidifying Activity of Proteoid Roots of P-Deficient White Lupin

The operational definition of a proteoid root is a secondary root with densely clustered rootlets (Fig. 2), following the morphological description of previous authors (Purnell, 1960; Gardner et al., 1981; Dinkelaker et al., 1995; Watt and Evans, 1999b). After about 10 d of cultivation, proteoid roots became visible in both P-sufficient and -deficient plants. In comparison with P-sufficient plants, the induction of proteoid roots in P-deficient plants occurred 1 to 2 d earlier. After 21 d, the average number of proteoid roots per plant was 61 ± 8 (means ± SE in four independent experiments) for P-deficient plants and 21 ± 2 for P-sufficient plants. Typically, P-sufficient plants produced only one long proteoid root on a primary lateral root, whereas P-deficient plants showed three or more sequentially clustered proteoid roots on a primary lateral root (Fig. 2). As a consequence, P deficiency resulted in an increase of root weight with a simultaneous decrease of shoot weight. After 21 d of cultivation, 7.4 ± 1.2 g plant¹ of fresh root weight for control plants and 10.3 ± 2.1 g plant¹ for P-deficient plants were determined. The shoot fresh weights were 8.2 ± 0.4 g plant¹ for control plants and 6.3 ± 0.1 g plant¹ for P-deficient plants. The P concentrations of dry shoot matter were 6.2 ± 0.3 mg g¹ for control plants and 1.4 ± 0.2 for P-deficient plants. Despite lower P concentration in shoots of P-deficient plants, no typical symptom of P deficiency was observed. H⁺ release by plant roots during the whole cultivation period was recorded by means of a pH stat system. Up to 12 d, plants of both treatments showed comparable H⁺ release.

View larger version (96K): [in this window] [in a new window]   Figure 2.   Proteoid roots produced by white lupin with (A) and without (B) phosphate. Plants were grown in a solution culture for 3 weeks. Plasma membrane was isolated from different types of roots: proteoid roots of P-sufficient plants, marked as proteoid (+P); lateral roots of P-sufficient plants, marked as lateral (+P); active proteoid roots (the youngest, fully developed proteoid root) of P-deficient plants, marked as proteoid (P); lateral roots of P-deficient plants, marked as lateral (P).

However, after 12 d of cultivation, P-sufficient plants showed increasing consumption of H⁺, whereas P-deficient plants continued to release H⁺ with a higher rate than before (Fig. 3). This concurred with an extensive production of proteoid roots by P-deficient plants during this period. At the harvest, the H⁺ release activity of proteoid roots was determined using an agar sheet with bromocresol purple as pH indicator. Only the youngest, fully developed proteoid roots showed strong rhizosphere acidification (Fig. 4A) and therefore were defined as active proteoid roots. There was no acidification either for old proteoid roots from P-deficient plants or for the proteoid roots of P-sufficient plants (not shown). The acidification of active proteoid roots was completely inhibited by 1 mM vanadate (Fig. 4B), indicating the involvement of active H⁺ pumping of the plasma membrane H⁺ ATPase in vivo. For the investigation of plasma membrane H⁺ ATPase in vitro, plasma membrane was isolated from four types of roots (Fig. 2): active proteoid roots from P-deficient plants [the youngest, fully developed, acidifying proteoid roots, marked as proteoid (P)], lateral roots from P-deficient plants [4- to 5-cm apical zone of lateral roots, marked as lateral (P)], proteoid roots from P-sufficient plants [marked as proteoid (+P)], and lateral roots from P-sufficient plants [4- to 5-cm apical zone of lateral roots, marked as lateral (+P)].

View larger version (20K): [in this window] [in a new window]   Figure 3.   H+ release by white lupin roots during a 3-week cultivation period. Plants were grown in nutrient solution, pH of which was kept constant at pH 6 by means of a pH stat system (Schott, Mainz, Germany). The amount of NaOH (or H2SO4) used for maintaining pH 6 was recorded daily and used for the calculation of H+ release. Values represent means ± SE of four independent experiments.

View larger version (59K): [in this window] [in a new window]   Figure 4.   Identification of active proteoid roots (A) and inhibitory effect of vanadate on rhizosphere acidification (B). Plants were grown in nutrient solution without phosphate for 3 weeks. After washing with deionized water, roots were carefully spread on the surface of agar sheet (0.75% [w/v] agar, 0.006% [w/v] bromocresol purple, 1 mM CaSO4, and 2.5 mM K2SO4, pH 6) and gently pressed into the agar sheet and incubated for 5 h under light in a growth chamber. For study of the inhibitory effect of vanadate (B), 1 mM vanadate was included on the right side of the agar sheet.

Effect of Different Root Types of White Lupin on the Isolation of Plasma Membrane

A summary of ATPase-specific activities associated with phase-partitioned plasma membrane from different root types of white lupin is presented in Table I. To avoid an underestimation of contamination by endoplasmic membranes, we analyzed azide- and nitrate-sensitive ATPase activity at pH 8.0. The inhibitor-sensitive ATPase hydrolytic activity of each membrane fraction was calculated by subtracting the ATPase hydrolytic activity in the presence of the inhibitor from the activity of the control at each assay pH. The ATPase activity of obtained membrane fractions from different types of white lupin roots showed a negligible sensitivity to azide and nitrate. On the other hand, about 95% of ATPase activity was sensitive to 0.1 mM vanadate. However, the membrane fractions showed a molybdate-sensitive ATP hydrolase activity, which indicates the presence of unspecific acid phosphatases (Widell and Larsson, 1990). This unspecific acid phosphatase activity was higher in roots from P-deficient plants than those from P-sufficient plants. Several attempts including homogenizing root tissues with 0.25 M KI and freezing-thawing the isolated membrane vesicles with following washing proved to be less effective to eliminate the unspecific acid phosphatase activity in the membrane fractions (data not shown). Therefore, in all analyses for ATPase activity, besides 1 mM azide and 50 mM nitrate, 1 mM molybdate was included to suppress the unspecific acid phosphatase activity. This assay medium warrants the determination of plasma membrane H⁺ ATPase activity and a direct comparison of its activity between membrane fractions derived from different types of roots of white lupin.

View this table: [in this window] [in a new window]   Table I.   Inhibitor-sensitive ATPase-hydrolytic activity associated with plasma membranes of different types of white lupin roots Membranes were isolated from different root types of 3-week-old white lupin: active proteoid roots of P-deficient plants [proteoid (P)], lateral roots of P-deficient plants [lateral (P)], proteoid roots of P-sufficient plants [proteoid (+P)], and lateral roots of P-sufficient plants [lateral (+P)]. Assays were conducted at 30°C. The inhibitor-sensitive activity was calculated by substrating the ATP-hydrolytic activity in the presence of inhibitor from the activity of the control. The values represent means ± SE (percentage relative to control) of four independent experiments.

Increase of Plasma Membrane H⁺ ATPase Activity in Active Proteoid Roots

In an assay (pH range 5.6-7.0), the plasma membrane derived from active proteoid roots of P-deficient plants [proteoid (P)] showed a more than 2 times higher ATPase activity than those from other types of roots. At its optimum pH (pH 6), a 3 times higher ATPase activity was determined in the plasma membrane of active proteoid roots in comparison with those of other membrane fractions (Fig. 5). On the contrary, the plasma membrane of lateral roots of P-deficient plants [laterals (P)] showed lower ATPase activity as compared with those of plasma membranes of P-sufficient plants. In addition, there was no difference in ATPase activity between proteoid roots and lateral roots when plants had been grown with 0.25 mM P. Figure 5 also indicates a variation of pH optimum for ATPase activity in the active proteoid roots. Plasma membrane ATPase of active proteoid roots showed a narrow pH optimum at pH 6.0, whereas a broader pH optimum between 6.2 and 6.4 was evident for ATPase activity of other types of roots. The pH optimum around 6.5 was reported for plasma membrane ATPase of maize (Zea mays) roots (Yan et al., 1998), oat (Avena sativa; Palmgren and Sommarin, 1989), and Arabidopsis (Luo et al., 1999).

View larger version (24K): [in this window] [in a new window]   Figure 5.   Comparison of ATPase activity of plasma membranes derived from different types of white lupin roots. Plants were grown in nutrient solution at pH 6 for 3 weeks. Plasma membrane was isolated from proteoid roots of P-sufficient plants [proteoid (+P)], lateral roots of P-sufficient plants [lateral (+P)], active proteoid roots (the youngest, fully developed proteoid root) of P-deficient plants [proteoid (P)], and lateral roots of P-deficient plants [lateral (P)]. Plasma membrane ATPase activity was analyzed in the presence of 1 mM molybdate, 1 mM azide, and 50 mM nitrate at 30°C. Values represent means ± SE of four independent experiments.

Increase in V_max, K_m, and Vanadate Sensitivity of Plasma Membrane H⁺ ATPase from Active Proteoid Roots

To gain a deeper insight into the enzyme properties of plasma membrane ATPase of active proteoid roots, we analyzed the kinetic characteristics of ATPase at ATP concentrations from 50 to 4,000 µM with an ATP regenerating system (Sekler and Pick, 1993). In this concentration range, plasma membrane ATPase revealed typical Michaelis-Menten kinetics for all membrane fractions (Fig. 6A), as found for maize (Yan et al., 1998) and Arabidopsis (Palmgren and Christensen, 1994). After transformation of data according to Eadie-Hofstee, a linear relationship (r > 0.98) was found between ATPase activity and the ratio of ATPase activity to ATP concentration (Fig. 6B). V_max and K_m were obtained from the intercept and slope of the straight line, respectively. At assay pH 6.0 and 6.5, the ATPase of active proteoid roots showed a significantly higher V_max than those of other three root types (Table II). In addition, the V_max of plasma membrane ATPase from lateral roots of P-sufficient plants was significantly higher than those of lateral roots of P-deficient plants or the proteoid roots of P-sufficient plants. For active proteoid roots, there was also a significant increase in K_m in comparison with other three types of roots, indicating a significant decrease in substrate affinity of the plasma membrane ATPase. Among lateral roots of P-deficient plants and lateral roots or proteoid roots of P-sufficient plants, there was no difference for K_m at assay pH 6.0. When assay pH was changed from 6.0 to 6.5, an increase in K_m by 2, 1.5, 1.9, and 1.8 times was recorded for active proteoid roots and lateral roots from P-deficient plants and proteoid roots and lateral roots from P-sufficient plants, respectively (Table II).

View larger version (25K): [in this window] [in a new window]   Figure 6.   Comparison of the kinetic characteristics of plasma membrane ATPase from different types of white lupin roots (see legend of Fig. 5). A, Dependence of ATPase activity on ATP concentration. Plants were grown in nutrient solution at pH 6 for 3 weeks. Plasma membrane ATPase activity was analyzed in the presence of 1 mM molybdate, 1 mM azide, and 50 mM nitrate at 30°C. The concentration of ATP was kept constant in the range of 50 to 4,000 µM. Values represent means ± SE of four independent experiments. B, Eadie-Hofstee plot of the data presented in A (r > 0.98).

View this table: [in this window] [in a new window]   Table II.   Effect of assay pH on the kinetic characteristics and vanadate sensitivity of plasma membrane ATPase of different white lupin roots Membranes were isolated from different root types of 3-week-old white lupin: active proteoid roots of P-deficient plants [proteoid (P)], lateral roots of P-deficient plants [lateral (P)], proteoid roots of P-sufficient plants [proteoid (+P)], and lateral roots of P-sufficient plants [lateral (+P)]. Vmax and Km were determined using ATP concentrations from 50 to 4,000 µM at 30°C. An ATP-regenerating system (5 mM PEP and 5 units of pyruvate kinase) was used to keep constant ATP concentrations. The sensitivity of ATPase activity to vanadate was determined within a concentration range of 0.5 to 500 µM. The values represent means ± SE of four independent experiments. Significant differences (P < 5%) between treatments are indicated by different letters.

For plasma membrane H⁺ ATPase, besides kinetic parameters such as K_m and V_max, another important parameter is the sensitivity of ATPase to vanadate. There is considerable variation for vanadate sensitivity among various plasma membrane ATPases (Regenberg et al., 1995; Morsomme et al., 1996). As for the determination of K_m and V_max, the inhibition of ATPase activity by vanadate was determined in a vanadate concentration range from 0.5 to 500 µM. After transformation of data according to Eadie-Hofstee, the concentration of vanadate for 50% inhibition of ATPase activity (I₅₀) was estimated in a similar way as for K_m. At assay pH 6, a significantly higher I₅₀ was estimated for the ATPase in the plasma membranes derived from P-deficient plants in comparison with those from P-sufficient plants (Table II). This indicates an induction of lower vanadate sensitivity of plasma membrane ATPase in white lupin roots by phosphate deficiency. In addition, at assay pH 6.0, the sensitivity of ATPase to vanadate was significantly lower for active proteoid roots than for lateral roots of P-deficient plants. It is interesting to note that the pH dependence of I₅₀ was totally different as for K_m. When assay pH was increased from 6.0 to 6.5, there was a decrease of I₅₀ for plasma membrane ATPase by 2-fold for active proteoid roots of P-deficient plants and 1.4 and 1.6-fold for proteoid roots and lateral roots of P-sufficient plants, respectively. Conversely, there was no change for lateral roots of P-deficient plants (Table II).

Increase of H⁺ ATPase Enzyme Concentration in the Plasma Membrane of Active Proteoid Roots

Although an increase in plasma membrane H⁺ ATPase activity has been reported under a number of environmental conditions, in most cases it was not clear whether the observed changes in H⁺ ATPase activity reflect modulation of the amount or the turnover rate of hydrolysis of the enzyme (Serrano, 1989). Identifying these two different components is a prerequisite for understanding plant response to environmental conditions on a molecular basis. In the present study, we solved this problem by determining the activation energy of ATP hydrolysis and measuring ATPase enzyme concentration in plasma membrane by means of western-blot analysis using a specific antibody for plasma membrane H⁺ ATPase.

For the determination of activation energy of ATPase, which is, according to Arrhenius, related to the turnover rate of hydrolysis of the enzyme, we analyzed the V_max at two different assay temperatures at optimum pH level for each membrane fraction (Yan et al., 1998). For all four membrane fractions, about 1.8 times increase in V_max was caused by increasing assay temperature from 25°C to 30°C, from which a comparable activation energy of 92 kJ mol¹ was calculated for all four membrane fractions analyzed. This indicates that the ATPase in different membrane fractions has a comparable turnover rate of ATP hydrolysis. As a consequence, the higher hydrolytic activity of ATPase found in the plasma membrane of active proteoid roots must be attributed to a higher ATPase enzyme concentration. ATPase enzyme concentration in plasma membrane was analyzed by means of western-blot technique. Membrane proteins (4 µg) were separated by SDS-PAGE on 10% (w/v) acrylamide gel. At a molecular mass of 97 kD, the band of plasma membrane derived from active proteoid roots of P-deficient plants was more intensive than those from proteoid roots of P-sufficient plants or from lateral roots grown either with or without phosphate (Fig. 7). Immunoblotting with a polyclonal antibody specific for the central part of plant H⁺ ATPase showed higher intensity for the plasma membrane of active proteoid roots of P-deficient plants (Fig. 7), as compared with the proteoid roots of P-sufficient plants or lateral roots from both P-sufficient and P-deficient plants. In addition to the strongly staining 97-kD region, there was also a weak cross-reaction with a minor band for the membrane fraction of proteoid roots from P-sufficient plants. For a quantitative comparison, the integrated evaluation of the intensity and area of signals was carried out by setting control (lateral roots of P-sufficient plants) as 100% in three independent experiments. In comparison with the control, the increase was 422% (±75%) for active proteoid roots of P-deficient plants, 142% (±12%) for the lateral roots from P-deficient roots, and 110% (±11%) for the proteoid roots of P-sufficient plants, respectively. These data reveal that the higher steady-state level of the plasma membrane H⁺ ATPase in the active proteoid roots is responsible for higher ATPase hydrolytic activity. In addition, there was also a small increase in the ATPase enzyme concentration in the plasma membrane of lateral roots from P-deficient plants in comparison with those from P-sufficient plants, although the converse was true for ATPase hydrolytic activity and V_max for these two membrane fractions (Figs. 5 and 6; Table II).

View larger version (58K): [in this window] [in a new window]   Figure 7.   Separation of plasma membrane proteins by SDS-PAGE (above) and immunodetection of plasma membrane H+ ATPase by western-blotting technique (below). M, Standard markers for molecular mass (Sigma, St. Louis); 1, plasma membrane from lateral roots of P-sufficient plants; 2, plasma membrane from active proteoid roots of P-deficient plants; 3, plasma membrane from lateral roots of P-deficient plants; 4, plasma membrane from proteoid roots of P-sufficient plants. Arrows indicate plasma membrane H+ ATPase. For separation of plasma membrane proteins, membrane vesicles (4-µg membrane proteins) were loaded onto polyacrylamide gel. After separation, the obtained gel was stained with Coomassie Brilliant Blue (above). For western-blot analysis (below), after separation on the gel the membrane proteins including molecular mass markers were transferred to polyvinylidene difluoride (PVDF) membrane filter (0.2 µm). For staining of the obtained blot, the lane of molecular mass markers was separated from other lanes of membrane proteins. The former was stained with Coomassie Brilliant Blue. The remaining blot with the lanes of plasma membrane proteins was incubated with a polyclonal antibody raised against the central portion of AHA2 (amino acids 340-650) and visualized with a secondary antibody (alkaline phosphatase-conjugated anti-rabbit IgG, Sigma). After separate staining, the blot of standard marker for molecular mass and the blot of plasma membrane proteins were combined.

Increase in H⁺-Pumping Activity of the Plasma Membrane of Active Proteoid Roots

Plasma membrane H⁺-pumping activity was monitored by A₄₉₂ of acridine orange (AO). After initiation of H⁺ pumping by addition of Mg-ATP, there was rapid quenching and it eventually reached a constant level. This established pH gradient was completely collapsed by 5 µM gramicidin (Fig. 8A). Furthermore, this H⁺-pumping activity was completely inhibited by 100 µM vanadate (data not shown). Compared with lateral roots either grown with or without phosphate and the proteoid roots of P-sufficient plants, absorbance quenching of AO caused by plasma membrane vesicles from active proteoid roots of P-deficient plants was more rapid at the beginning and reached a higher level after 40 min. Despite comparable absorbance quenching of AO at the beginning, membrane vesicles from P-deficient lateral roots and from P-sufficient proteoid roots showed a higher level of quenching of AO after 40 min than those from P-sufficient lateral roots. Two parameters, initial rate and maximum quenching (pH gradient), were used to characterize the plasma membrane H⁺ pumping. The initial rate of H⁺ pumping was determined according to the quenching rate within the 1st min, which reflects active H⁺ influx into plasma membrane vesicles (Yan et al., 1998). Maximum quenching was measured 40 min after initiation of the H⁺ pump. At this time, net H⁺ transport across the plasma membrane was zero and H⁺ influx due to active pumping and H⁺ efflux because of leakage reached equilibrium. This parameter indicates the steepest pH gradient that can be created by H⁺-pumping activity. At assay pH 6.5, the initial rate of H⁺ pumping by plasma membrane ATPase from active proteoid roots of P-deficient plants was increased by about 3 times in comparison with those from the other three types of roots. There was no difference in initial rate of H⁺ pumping among the other three membrane fractions (Table III). Furthermore, the plasma membrane from active proteoid roots of P-deficient plants created a 1.4 to 1.8 times steeper pH gradient than proteoid and lateral roots of P-sufficient plants. For plants grown with phosphate, there was a significantly higher pH gradient for proteoid roots than those for lateral roots. However, the difference in pH gradient between active proteoid roots and lateral roots from P-deficient plants was less pronounced.

View larger version (22K): [in this window] [in a new window]   Figure 8.   Comparison of active H+ transport driven by plasma membrane H+ ATPase (A) and passive H+ transport by leakage (B) across the plasma membranes. Membrane vesicles were isolated from different types of roots of white lupin (see legend of Fig. 5). For the comparison of active H+ transport (A), the pH gradient formation across vesicle membranes was monitored by A492 of AO. At assay pH 6.5, intravesicular acidification was initiated by addition of 5 mM Mg-ATP. The established pH gradient was completely collapsed by 5 µM gramicidin (Gram.). For the comparison of passive H+ transport (B), the intravesicular acidification was initiated by addition of 5 mM Mg-ATP to create a pH gradient across plasma membrane vesicles. For a reliable comparison, ATPase activity was stopped by addition of 500 µM vanadate after quenching had reached 0.0300 A units for all four membranes. The resting pH gradient was collapsed by gramicidin (Gram., 5 µM).

View this table: [in this window] [in a new window]   Table III.   Effect of different root types of white lupin on the active H-pumping of ATPase and passive H leakage of plasma membrane Membranes were isolated from different root types of 3-week-old white lupin: active proteoid roots of P-deficient plants [proteoid (P)], lateral roots of P-sufficient plants [lateral (+P)], and lateral roots of P-deficient plants [lateral (P)]. The assay was conducted at 25°C at pH 6.5 using 50 µg of membrane proteins. The values represent means ± SE of four independent experiments. Significant differences (P < 5%) between treatments are indicated by different letters.

For inside-out vesicles, the establishment of a pH gradient is determined by both active influx (pumping) and passive efflux (leakage). The apparent discrepancy between the increase of H⁺-pumping activity (3 times higher initial rate for vesicles from active proteoid roots than those from P-sufficient lateral roots) and the increase of pH gradient (only 1.8 times steeper pH gradient) is probably due to a difference in passive H⁺ transport between the membrane fractions. To determine H⁺ efflux from plasma membrane vesicles, we measured the degradation of the pH gradient after stopping H⁺ pumping by addition of 500 µM vanadate. Because the degradation of pH gradient depends on the gradient itself, a comparison between treatments should be made at the same pH gradient. Therefore, we stopped H⁺ pumping when the pH gradient of plasma membrane vesicles reached 0.0300 A units (Fig. 8B). After addition of vanadate, the established pH gradient was degraded quickly and then reached a relative constant level. This resting pH gradient was completely collapsed by gramicidin (Fig. 8B). It is evident from Figure 8B that the pH gradient degradation was faster for active proteoid roots than for other three types of roots. The degradation rate within the 1st min after the addition of vanadate was measured as the initial rate to describe how rapidly the pH gradient degradation starts. Compared with two lateral roots and proteoid roots of P-sufficient plants, the initial degradation rate for plasma membrane vesicles from active proteoid roots of P-deficient plants was approximately 2 times higher (Table III). This indicates that the plasma membrane from active proteoid roots of P-deficient plants was more permeable for H⁺ than those from proteoid roots of P-sufficient plants and lateral roots either from P-sufficient or -deficient plants. A similar result was reported for maize roots adapted to low solution pH (Yan et al., 1998). The difference in initial degradation rate among proteoid roots of P-sufficient plants and two lateral roots was less pronounced (Table III).

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Isolation of Plasma Membrane from Different Types of White Lupin Roots

The membrane fractions isolated from different types of white lupin roots showed a high sensitivity to vanadate (0.1 mM, pH 6.5) and negligible sensitivity to nitrate (50 mM, pH 8) and azide (1 mM, pH 8), indicating a plasma membrane fraction with low contamination by membranes of tonoplast and mitochondrial origins (Table I). However, the ATP hydrolysis by obtained membrane fractions, especially those from P-deficient plants, showed a considerable sensitivity to molybdate (1 mM, pH 6.5), indicating an unspecific acid phosphatase activity (Widell and Larsson, 1990). Although the localization of unspecific acid phosphatase activity is in most cases uncertain, the fact that the ATPase hydrolytic activity in the obtained membrane fractions was not sensitive to nitrate suggests that the unspecific acid phosphatase activity was not related to the tonoplast. In plant cells, most of the unspecific acid phosphatases exist in the vacuole as soluble proteins (Duff et al., 1994). These proteins can be trapped in membrane vesicles during homogenization of plant tissues. In addition, white lupin roots release a large amount of acid phosphatases in response to P deficiency (Li et al., 1997; Gilbert et al., 1999; Neumann et al., 2000; Wasaki et al., 2000; Miller et al., 2001). After synthesis, acid phosphatases are transported via secretory vesicles to plasma membranes and are released into apoplast. We treated the isolated membrane with freezing-thawing-cycling followed by washing. However, this treatment did not reduce the unspecific acid phosphatase activity. This result suggests a binding of phosphatases to plasma membrane. In white lupin, cDNA clones of two isoforms of acid phosphatases, lasap1 and lasap2, have been characterized (Wasaki et al., 1999, 2000). lasap2 is specifically expressed in roots exposed to P deficiency and is supposed to encode secretory acid phosphatase. The expression of lasap1 evidently not only occurs in roots but also in leaves and responses to P deficiency. According to cDNA analysis, it was suggested that acid phosphatase encoded by lasap1 might be anchored to plasma membrane by a glycosylinositolphospholipid (Wasaki et al., 2000). In conclusion, despite a considerable activity of unspecific acid phosphatases, the isolated membrane fractions can be considered as plasma membrane with low contamination of other membrane origins. To warrant a characterization of plasma membrane H⁺ ATPase, 1 mM molybdate was included in all analyses to suppress acid phosphatase activity.

Quantitative Adaptation of Plasma Membrane H⁺ ATPase in Active Proteoid Roots to P Deficiency

The quantitative adaptation of plasma membrane H⁺ ATPase of active proteoid roots to P deficiency is supported by the following results obtained in vitro: Compared with lateral roots from both P-sufficient and -deficient plants and proteoid roots of P-sufficient plants, (a) the hydrolytic activity of plasma membrane H⁺ ATPase of active proteoid roots was about 3-fold higher (Fig. 5), (b) V_max of H⁺ ATPase in plasma membrane of active proteoid roots was about 3 times higher (Fig. 6; Table II), (c) western blotting showed a about 4 times higher H⁺ ATPase enzyme concentration in plasma membrane of active proteoid roots (Fig. 7), (d) the initial rate of H⁺-pumping activity of active proteoid roots was 3-fold higher (Fig. 8A; Table III), (e) plasma membrane H⁺ ATPase of active proteoid roots created and maintained a higher pH gradient (Fig. 8A; Table III), and (f) this higher pH gradient was not explained in terms of reduced H⁺ permeability of this membrane; rather, the plasma membrane of active proteoid roots showed about 2 times higher H⁺ permeability (Fig. 8B; Table III).

Of all characteristics for the quantitative adaptation, the high steady-state H⁺ ATPase enzyme concentration in the plasma membrane is the fundamental one that is responsible for overall increase in hydrolytic ATPase activity and H⁺-pumping activity recorded for active proteoid roots. In fact, for all membrane fractions analyzed, there was no difference in hydrolytic efficiency for H⁺ ATPase, which is indicated by comparable activation energy of hydrolytic reaction of this enzyme. In addition, there is no difference in coupling efficiency between ATP hydrolysis and H⁺ pumping for plasma membrane H⁺ ATPase derived from different types of roots. This conclusion is supported by the fact that the increase rate for both hydrolytic ATPase activity and its H⁺-pumping activity for active proteoid roots was similar (Figs. 5 and 8; Tables II and III).

A higher steady-state enzyme concentration of H⁺ ATPase in the plasma membrane of active proteoid roots can result either from a higher expression rate (transcriptional level) of genes encoding plasma membrane H⁺ ATPase or an enhanced synthesis rate of the enzyme proteins (translational level), or from a reduced degradation rate of H⁺ ATPase (Sze et al., 1999). It is suggested that the translational regulation of H⁺ ATPase controls the amount of the proteins only within a short range of concentration. A 2-fold variation in translation efficiency has been suggested for pma1 and pam3, the plasma membrane H⁺ ATPases of Nicotiana plumbaginifolia (Lukaszewicz et al., 1998). If this holds true for white lupin, it implies that the 4-fold higher H⁺ ATPase enzyme concentration in the plasma membrane of active proteoid roots can only be partly explained in terms of translational regulation. In the coleoptile of maize, auxin stimulated not only gene expression of plasma membrane H⁺ ATPase but also the membrane flow from endoplasmic reticulum to plasma membrane, resulting in a higher steady-state H⁺ ATPase concentration in the plasma membrane (Hager et al., 1991). In addition, this study also showed that the half-life time of plasma membrane H⁺ ATPase was only 12 min. Therefore, if the degradation rate of the plasma membrane H⁺ ATPase can be slowed down in active proteoid roots, this may also significantly contribute to the higher steady-state H⁺ ATPase concentration in the plasma membrane. So far, it remains an open question: Which regulation possibilities are responsible for the higher steady-state H⁺ ATPase enzyme concentration in the active proteoid roots and thereby for the adaptation of white lupin to P deficiency?

Qualitative Adaptation of Plasma Membrane H⁺ ATPase in Active Proteoid Roots to P Deficiency

Besides quantitative adaptation, a qualitative adaptation of plasma membrane H⁺ ATPase of active proteoid roots to P deficiency is also evident. This conclusion can be supported by following results: Compared with lateral roots from both P-sufficient and -deficient plants and proteoid roots of P-sufficient plants, (a) plasma membrane H⁺ ATPase of active proteoid roots showed a more acidic pH optimum (Fig. 5), (b) the K_m value of the plasma membrane H⁺ ATPase of active proteoid roots was significantly higher (Fig. 7; Table II), and (c) at assay pH 6.0, the vanadate sensitivity of the plasma membrane H⁺ ATPase of active proteoid roots was lower (higher I₅₀ for vanadate, Table II).

At enzyme level, it is well established that a part of the C-terminal region of H⁺ ATPase constitutes an auto-inhibitory domain. Removal of this domain by different effectors such as fusicoccin (a fungal toxin) in vivo, lysophospholipids, and trypsin in vitro can activate the enzyme (Palmgren et al., 1991; Johansson et al., 1993). The question arises whether this mechanism is involved in the regulation of plasma membrane H⁺ ATPase of active proteoid roots and thereby responsible for the qualitative adaptation to P deficiency. The removal of the auto-inhibitory domain of H⁺ ATPase is characterized by (a) a higher degree of stimulation in H⁺-pumping than in hydrolytic ATPase activity, (b) an increase in V_max and a decrease in K_m, (c) a shift of pH optimum toward more alkaline values, and (d) sensitivity of the enzyme to vanadate remains unchanged (Palmgren et al., 1988; Palmgren and Sommarin, 1989). All our data determined for the H⁺ ATPase in plasma membrane of active proteoid roots are not consistent with the above-mentioned changes in H⁺ ATPase caused by removal of auto-inhibitory C terminus. Compared with lateral roots and the proteoid roots of P-sufficient plants, the plasma membrane H⁺ ATPase of active proteoid roots showed (a) the same degree of stimulation for both H⁺-pumping and hydrolytic activity (Figs. 5 and 8; Table III); (b) not only a higher V_max, but also a higher K_m (Fig. 6; Table II); (c) a more acidic pH optimum (Fig. 5); and (4) a lower sensitivity to vanadate at assay pH 6.0 (Table II). In addition, the higher activity of H⁺ ATPase can be explained in terms of higher H⁺ ATPase enzyme concentration in the plasma membrane of active proteoid roots (Fig. 7).

The qualitative changes of the enzyme may be explained by differential expression of H⁺ ATPase isoforms in the plasma membrane of active proteoid roots. The plant plasma membrane H⁺ ATPase is encoded by a multigene family (Michelet and Boutry, 1995). Using heterologous expression of functional plant plasma membrane H⁺ ATPase in yeast (Saccharomyces cerevisiae), distinct differences in enzyme kinetics, pH optimum, and vanadate sensitivity have been reported for different isoforms of H⁺ ATPase in Arabidopsis (Palmgren and Christensen, 1994). A 10-fold higher K_m was found for AHA3 as compared with AHA1 or AHA2. In addition, AHA2 showed 3 times lower sensitivity to vanadate than AHA1 or AHA3 (Palmgren and Christensen, 1994). Therefore, it is likely that in active proteoid roots, an enhanced expression of a special isoform (or isoforms) of plasma membrane H⁺ ATPase is induced to meet the high demand of the cells to extrude H⁺. Such a tissue-, organ-, or cell-specific expression of isoforms of H⁺ ATPase has been reported by a number of investigations (DeWitt et al., 1991; Michelet et al., 1994; Moriau et al., 1999) and summarized as a "tissue-specific expression model" in which enzymatically equivalent enzymes are expressed in different cells by promoters providing controlled expression levels ideally suited for a cell's particular needs (Palmgren and Harper, 1999).

So far, it is not clear how the enhanced expression of the proteoid root-specific isoform(s) of plasma membrane H⁺ ATPase is triggered in active proteoid roots. In the present study, it was shown that in vivo, strong rhizosphere acidification was associated only with the youngest, fully developed proteoid roots from P-deficient white lupin. On the contrary, neither the proteoid roots produced by P-sufficient plants nor the old proteoid roots from P-deficient plants showed rhizosphere acidification. In addition, the apical zone of lateral roots from P-deficient plants also failed to acidify rhizosphere significantly (Fig. 4). These in vivo data are closely related to the activity of plasma membrane H⁺ ATPase in vitro (Fig. 5). Our data suggest that the enhanced expression of H⁺ ATPase in active proteoid roots of white lupin is under the control of a combination of root structures, P deficiency, and physiological development of proteoid roots. A similar conclusion for the release of citrate by white lupin roots was drawn by a number of investigators (Johnson et al., 1996 a, 1996b; Keerthisinghe et al., 1998; Neumann et al., 1999, 2000; Watt and Evans, 1999a; Massonneau et al., 2001).

In conclusion, our data clearly demonstrate an involvement of plasma membrane H⁺ ATPase in the adaptation of white lupin to P deficiency. The repeatedly reported strong acidification associated with active proteoid roots of white lupin in response to P deficiency may be attributed to enhanced expression of tissue-specific H⁺ ATPase isoform(s) in proteoid roots. Our data support the model that the release of organic acids consists of two separate transport processes: an active H⁺ efflux driven by plasma membrane H⁺ ATPase and a passive efflux of organic anions mediated by channel-like transporters (Fig. 1). It becomes evident that plasma membrane H⁺ ATPase plays an essential role for the enhanced H⁺ release by plant roots under P deficiency and, therefore, is one of essential enzymes involved in the adaptation of plants to P deficiency.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Plant Cultivation

Seeds of white lupin (Lupinus albus L. cv Amiga, kindly supplied by Dr. Römer, Südwestsaat GbR, Rastatt, Germany) were soaked in aerated 1 mM CaSO₄ for 1 d and germinated at 22°C in the dark between two layers of filter paper moistened with 1 mM CaSO₄. After 4 d, seedlings were transferred to a container with 50 L of one-fourth-strength concentrated nutrient solution. Plants were grown in a growth chamber under controlled conditions. Lamps (Powerstar HQI-T 400W/D, Osram, Frankfurt) gave a light intensity of approximately 400 µE m² s¹ at shoot height, with a day/night cycle of 16 h/8 h at 22°C/15°C. Relative humidity was 60%. After 2 and 4 d of cultivation, the concentration of the nutrient solution was increased to one-half and full strength concentration, respectively. The full-strength nutrient solution had the following composition: 0.5 mM Ca(NO₃)₂, 1.75 mM K₂SO₄, 0.25 mM KCl, 1.25 mM MgSO₄, 25 µM H₃BO₃, 1.5 µM MnSO₄, 1.5 µM ZnSO₄, 0.5 µM CuSO₄, 0.025 µM (NH₄)₆Mo₇O₂₄, and 20 µM Fe(III)-EDTA. For the control plants, the nutrient solution received an additional 0.25 mM KH₂PO₄. The complete solution in the containers was changed every 3 d. Solution pH was kept constant at 6.0 by continuous titration with 0.1 M NaOH or 0.05 M H₂SO₄ using a pH stat system (Schott). Plants were harvested at the age of 3 weeks for the isolation of plasma membrane.

Measurement of Plant Growth, Number of Proteoid Roots, Phosphate Concentrations of Plant Shoots, and Detection of Rhizosphere Acidification by Intact Proteoid Roots

After 3 weeks of cultivation, plant shoots and roots were separated. After determination of fresh weights, plant shoots were dried at 80°C for 2 d and then the dry weight of shoots was determined. For determination of fresh weights, roots of white lupin were thoroughly washed with deionized water three times and blotted dry with tissue paper. The number of proteoid roots per plant was determined. For determination of phosphate concentration of shoot, plant matter was dry ashed and the resulting ash was dissolved in HNO₃. The phosphate concentration was then determined photometrically by using molybdate-vanadate reagents (yellow method for P determination).

The exudating activity of proteoid roots was determined according to Römheld et al. (1984). The roots of 3-week-old plants were thoroughly washed with deionized water and spread on the agar sheet containing 0.75% (w/v) agar, 0.006% (w/v) bromocresol purple, 2.5 mM K₂SO₄, and 1 mM CaSO₄, pH 6. The roots were carefully pressed into agar without being damaged. The incubation for visualization of rhizosphere acidification was conducted in a growth chamber under light for 5 h. For the determination of vanadate sensitivity of rhizosphere acidification of intact proteoid roots, two agar sheets were separately prepared. The agar sheet for the control had the same composition as mentioned above. For vanadate treatment, the agar sheet contained an additional 1 mM Na₃VO₄.

Plasma Membrane Isolation

Plasma membrane was isolated from four different types of roots of 3-week-old white lupin (see also Fig. 2): lateral roots of P-sufficient plants [4- to 5-cm apical zone of lateral roots marked as lateral (+P)], proteoid roots of P-sufficient plants [marked as proteoid (+P)], lateral roots of P-deficient plants [4- to 5-cm apical zone of lateral roots marked as lateral (P)], and active (the youngest, fully developed) proteoid roots of P-deficient plants [marked as proteoid (P)]. Plasma membrane of roots was isolated according to Yan et al. (1998) with some modifications. Roots were cut and washed three times with chilled, deionized water and ground in ice-cold homogenization buffer with a mortar and pestle. The homogenization buffer contained 250 mM Suc, 250 mM KI, 2 mM EGTA, 10% (v/v) glycerol, 0.5% (w/v) bovine serum albumin, 2 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 5 mM 2-mercaptoethanol, and 50 mM 1,3-bis(tris[hydroxymethyl]methylamino) propane (BTP), adjusted to pH 7.8 with MES. The homogenate (adjusted to a grinding medium/tissue ratio of 4 mL g¹ fresh weight) was filtered through two layers of Miracloth (Calbiochem-Novabiochem, San Diego) and centrifuged in a swinging bucket rotor at 11,500g (AH 629 rotor, 36 mL, Sorvall Products, Newtown, CT) for 10 min at 0°C. The supernatants were centrifuged at 87,000g for 35 min. The microsomal pellets were resuspended in phase buffer (250 mM Suc, 3 mM KCl, and 5 mM KH₂PO₄, pH 7.8).

The microsomal membrane preparation was fractionated by two-phase partitioning in aqueous dextran T-500 (Sigma) and polyethylene glycol (Sigma) according to the method of Larsson (1985). Phase separations were carried out in a series of 32-g phase systems that contained 6.1% (w/w) of each polymer dissolved in phase buffer (see above). Stock solutions of polymers were prepared with concentrations of 20% and 40% (w/w) for dextran and polyethylene glycol, respectively. The concentration of dextran stock solution was determined by optical rotation (Larsson, 1985). The phase stock was weighed and diluted to 6.1% (w/w, each polymer) with phase buffer to a final weight of 32 g. Polymers in "start tubes" were, however, diluted to 26 g. Six grams of microsomal resuspension (in phase buffer) was added to the upper phase of each start tube. The tubes were sealed with Parafilm (American National Can, Greenwich, CT) and mixed by inversion (30 times). Phase separation was achieved at 4°C by centrifugation at 720g (Sorvall AH-629 rotor, 36 mL) for 23 min followed by two washing steps in identical phases. Centrifugation times for the second through fourth phases were 15, 10, and 5 min, respectively. The upper phases obtained after four separations were diluted with phase buffer (see above) and centrifuged at 151,200g for 40 min. The pellets were washed with resuspension buffer (250 mM Suc, 3 mM KCl, and 5 mM BTP/MES, pH 7.8) and pelleted again. The pellets were resuspended in resuspension buffer, divided into aliquots, and immediately stored in liquid nitrogen. Protein was quantified according to the method of Bradford (1976) using bovine serum albumin (Sigma) as a standard.

Enzyme Assay

Hydrolytic ATPase activity was determined in 0.5 mL of 30 mM BTP/MES buffer containing 5 mM MgSO₄, 50 mM KCl, 50 mM KNO₃, 1 mM Na₂MoO₄, 1 mM NaN₃, 0.02% (w/v) Brij 58 (Sigma), and 5 mM disodium-ATP. Reaction was initiated by the addition of 1 to 2 µg of membrane protein, proceeded for 30 min at 30°C, and stopped with 1 mL of stopping reagent [2% (v/v) concentrated H₂SO₄, 5% (w/v) SDS, and 0.7% (w/v) (NH₄)₂MoO₄)] followed immediately by 50 µL of 10% (w/v) ascorbic acid. After 10 min, 1.45 mL of arsenite-citrate reagent (2% [w/v] sodium citrate, 2% [w/v] sodium m-arsenite, and 2% [w/v] glacial acetic acid] was added to prevent the measurement of phosphate liberated because of ATPase activity from ATP hydrolysis under acidic conditions (Baginski et al., 1967). Color development was completed after 30 min and A₈₂₀ was measured by means of a spectrophotometer. ATPase activity was calculated as phosphate liberated in excess of boiled-membrane control. The kinetic characteristics of plasma membrane H⁺ ATPase were studied in the presence of an ATP-generating system that included 5 units of pyruvate kinase (Sigma) and 5 mM PEP (Boehringer Mannheim/Roche, Basel; Sekler and Pick, 1993). V_max and K_m were determined by means of a regression analysis. Activation energy of ATPase was calculated using the Arrhenius sp. equation from V_max values determined at 25°C and 30°C, respectively.

pH Gradient

The formation of a pH gradient across the plasma membrane of inside-out vesicles was measured as the quenching of A₄₉₂ by AO. The change of the quenching was continuously monitored by a spectrophotometer (Carry 4 Bio, Varian Australia Pty Ltd., Mulgrave, Victoria, Australia). The assay mixture contained 5 mM BTP/MES (pH 6.5), 12 µM AO, 300 mM KCl, 250 mM Suc, 0.5 mM EGTA (adjusted to pH 6.5 with BTP), 9 µM valinomycin, 1 mM NaN₃, 1 mM Na₂MoO₄, 50 mM KNO₃, 0.05% (w/v) Brij 58, and 50 µg of membrane protein in a final volume of 1.5 mL. Brij 58 was used to create inside-out vesicles (Johansson et al., 1995). After equilibration of the membrane vesicles with the reaction medium (about 20 min), the reaction was initiated by the addition of Mg-ATP (mixture of MgSO₄ and disodium-ATP, adjusted to pH 6.5 with BTP) to give a final concentration of 5 mM. The reaction temperature was 25°C.

Gel Electrophoresis and Immunodetection of Plasma Membrane H⁺ ATPase

Plasma membrane proteins were separated by SDS-PAGE using the system of Laemmli (1970). Membrane vesicles (4-µg membrane proteins) were solubilized in SDS-loading buffer containing 0.125 mM Tris-HCl, pH 7.4; 10% (w/v) SDS; 10% (v/v) glycerol; 0.2 M dithiothreitol, 0.002% (w/v) bromocresol blue, 5 mM phenylmethylsulfonyl fluoride, and 0.05% (w/v) trasylol. The standard markers for protein molecular mass were purchased form Sigma. After a 30-min shaking at room temperature (22°C), samples were loaded on a discontinuous SDS-polyacrylamide gel (6% [w/v] acrylamide stacking gel and 10% [w/v] acrylamide separating gel). The resulting gels were stained with 0.1% (w/v) Coomassie Brilliant Blue R-250 in 40% (v/v) ethanol overnight. To enhance the image, the gels were then incubated for 10 min with a solution containing 8% (v/v) acetic acid and 25% (v/v) ethanol. Then the gels were destained with a 10% (v/v) acetic acid and 30% (v/v) ethanol solution.

For the western-blot analysis, after separation by SDS-PAGE samples were transferred to PVDF membrane filters (0.2 µm, Pall Specialty Materials, Port Washington, NY) using a semidry blotting system with a buffer containing 10 mM 3-cyclonexylamino-1-propane sulfonic acid (pH 11, adjusted with NaOH) and 20% (v/v) methanol for 1.5 h at room temperature and at a current intensity of 0.8 mA cm². For staining of the obtained blot, the lane of the standard markers of molecular mass was separated from the lanes of the membrane proteins. The former was stained with Coomassie Brilliant Blue R-250, as described above. For the identification and quantification of plasma membrane H⁺ ATPase, the remaining blot with plasma membrane proteins was incubated with a polyclonal antibody (kindly supplied by Dr. Michael G. Palmgren, Royal Veterinary and Agricultural University, Copenhagen) specific for the central part of plant H⁺ ATPase (amino acids 340-650 of AHA2). The antiserum was diluted 1:3,000 in TBS-T buffer (1 mM Tris-HCl [pH adjusted to 8.0 with NaOH], 15 mM NaCl, and 0.1% [v/v] Tween 20) and incubation was carried out for 1 h at room temperature followed by an incubation at 4°C overnight. After rinsing in TBS-T, PVDF membrane filters were incubated at room temperature for 2 h with a 1:30,000 (v/v) diluted secondary antibody (alkaline phosphatase-conjugated anti-rabbit IgG, Sigma). After rinsing in TBS-T, the filters were incubated for 5 min in a buffer containing 100 mM Tris-HCl (pH 9.5, adjusted with NaOH), 100 mM NaCl, and 5 mM MgCl₂. After separate staining, the part of the blot with the lane of standard molecular mass and the part of the blot with lanes of plasma membrane proteins were combined (see Fig. 7B). For quantification of plasma membrane H⁺ ATPase, the blots were scanned by using a videocamera (Module CCD, LTF-Labortechnik, Wasserburg-Bodensee, Germany), and the H⁺ ATPase immunoreactive bands were quantified densitometrically with software (TINA, Raytest Isotopenmessgeräte, Straubenhardt, Germany).

Statistical Treatment

Variation is indicated by ±SE (if bars exceed symbols in figures). Significant differences between treatments were calculated by using the Student's t test.