Plasma membrane H+-ATPase-dependent citrate exudation from cluster roots of phosphate-deficient white lupin

Authors

  • NICOLA TOMASI,

    Corresponding author
    1. Dipartimento di Scienze Agrarie e Ambientali, University of Udine, Via delle Scienze 208, I-33100 Udine, Italy,
    2. Laboratory of Molecular Plant Physiology, Institute of Plant Biology, University of Zürich, Zollikerstrasse 107, CH-8008 Zürich, Switzerland,
      N. Tomasi. Fax: +39 0432 558603; e-mail: nicola.tomasi@uniud.it
    Search for more papers by this author
  • TOBIAS KRETZSCHMAR,

    1. Laboratory of Molecular Plant Physiology, Institute of Plant Biology, University of Zürich, Zollikerstrasse 107, CH-8008 Zürich, Switzerland,
    Search for more papers by this author
  • LUCA ESPEN,

    1. Dipartimento di Produzione Vegetale, University of Milan, Via Celoria, 2, I-20133 Milano, Italy,
    Search for more papers by this author
  • LAURE WEISSKOPF,

    1. Laboratory of Molecular Plant Physiology, Institute of Plant Biology, University of Zürich, Zollikerstrasse 107, CH-8008 Zürich, Switzerland,
    Search for more papers by this author
  • ANJA THOE FUGLSANG,

    1. Centre for Membrane Pumps in Cells and Disease – PUMPKIN, Danish National Research Foundation and Department of Plant Biology and Biotechnology, University of Copenhagen, Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark,
    Search for more papers by this author
  • MICHAEL GJEDDE PALMGREN,

    1. Centre for Membrane Pumps in Cells and Disease – PUMPKIN, Danish National Research Foundation and Department of Plant Biology and Biotechnology, University of Copenhagen, Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmark,
    Search for more papers by this author
  • GÜNTER NEUMANN,

    1. Institute of Plant Nutrition, University of Hohenheim, D-70593 Stuttgart, Germany and
    Search for more papers by this author
  • ZENO VARANINI,

    1. Dipartimento di Scienze, Tecnologie e Mercati della Vite e del Vino, University of Verona, Via della Pieve, 70, I-37029 San Floriano (VR), Italy
    Search for more papers by this author
  • ROBERTO PINTON,

    1. Dipartimento di Scienze Agrarie e Ambientali, University of Udine, Via delle Scienze 208, I-33100 Udine, Italy,
    Search for more papers by this author
  • ENRICO MARTINOIA,

    1. Laboratory of Molecular Plant Physiology, Institute of Plant Biology, University of Zürich, Zollikerstrasse 107, CH-8008 Zürich, Switzerland,
    Search for more papers by this author
  • STEFANO CESCO

    1. Dipartimento di Scienze Agrarie e Ambientali, University of Udine, Via delle Scienze 208, I-33100 Udine, Italy,
    Search for more papers by this author

N. Tomasi. Fax: +39 0432 558603; e-mail: nicola.tomasi@uniud.it

ABSTRACT

White lupin (Lupinus albus L.) is able to grow on soils with sparingly available phosphate (P) by producing specialized structures called cluster roots. To mobilize sparingly soluble P forms in soils, cluster roots release substantial amounts of carboxylates and concomitantly acidify the rhizosphere. The relationship between acidification and carboxylate exudation is still largely unknown. In the present work, we studied the linkage between organic acids (malate and citrate) and proton exudations in cluster roots of P-deficient white lupin. After the illumination started, citrate exudation increased transiently and reached a maximum after 5 h. This effect was accompanied by a strong acidification of the external medium and alkalinization of the cytosol, as evidenced by in vivo nuclear magnetic resonance (NMR) analysis. Fusicoccin, an activator of the plasma membrane (PM) H+-ATPase, stimulated citrate exudation, whereas vanadate, an inhibitor of the H+-ATPase, reduced citrate exudation. The burst of citrate exudation was associated with an increase in expression of the LHA1 PM H+-ATPase gene, an increased amount of H+-ATPase protein, a shift in pH optimum of the enzyme and post-translational modification of an H+-ATPase protein involving binding of activating 14-3-3 protein. Taken together, our results indicate a close link in cluster roots of P-deficient white lupin between the burst of citrate exudation and PM H+-ATPase-catalysed proton efflux.

INTRODUCTION

The availability of phosphate (P) is a widespread limiting factor for plant growth (Marschner 1995; Handreck 1997). To deal with this situation, plants have developed several strategies to improve P acquisition. The most common one is the association with mycorrhizal fungi. If plants do not engage in symbiosis, they adapt to P-poor soils by increasing the quantity and length of roots and root hairs, and in many cases by increasing the synthesis and the release of P-mobilizing root exudates such as organic acids, phenolics and acid phosphatases (Dinkelaker, Römheld & Marschner 1989; Neumann et al. 2000; Tomasi et al. 2008).

Some plants have developed special root structures in response to P starvation, such as the dauciform roots found in the Cyperaceae family (Davies, Briarty & Rieley 1973; Lamont 1974; Shane, Dixon & Lambers 2005) and the cluster or proteoid roots of many Proteaceae (Purnell 1960). Cluster roots or proteoid roots are generally bottlebrush-like root structures. Many studies have demonstrated the capacity of these roots to increase the availability of soil P (Dinkelaker et al. 1989; Neumann et al. 1999; Shane et al. 2003). P mobilization is suggested to occur predominantly via the release of large amounts of organic acid anions by the cluster roots (Gardner, Parbery & Barber 1982; Dinkelaker et al. 1989; Jones & Darrah 1994; Lambers et al. 2002).

White lupin (Lupinus albus L.) has often been used as a model plant for studying P-deficiency response and cluster-root formation (Neumann & Martinoia 2002). In cluster roots of white lupin, carboxylate release follows a spatial and temporal pattern. In the first growing phase, cluster roots release low amounts of organic acids, mainly malate. In mature proteoid roots, a burst of citrate exudation can be observed, which is accompanied by acidification of the rhizosphere (Neumann et al. 2000). Citrate is likely to be exuded via an anion channel, because exudation can be inhibited by anthracene-9-carboxylic acid (Neumann et al. 1999). In a more recent study, a citrate-specific channel from lupin root protoplasts was described (Zhang, Ryan & Tyerman 2004), which could be responsible for massive citrate exudation; however, regulatory mechanisms of carboxylate exudation still need to be elucidated. A correlation between citrate and proton extrusion has been suggested by several authors (Sas, Rengel & Tang 2001; Ohno, Koyama & Hara 2003; Ligaba et al. 2004; Shen et al. 2005; Zhu et al. 2005); however, in these studies, different or even contrasting results were obtained, possibly because of the use of different experimental approaches and culture conditions.

The involvement of the plasma membrane (PM) H+-ATPase in organic anion release via proton extrusion in white lupin was suggested by Kania et al. (2001) and Yan et al. (2002). Using isolated PM vesicles from whole proteoid roots and lateral roots grown in the presence or absence of P, these authors showed that proteoid roots exhibited an increased PM H+-ATPase activity that was correlated to changes in the amount of the corresponding protein. Ligaba et al. (2004) also observed an enhanced release of citrate, but not of malate, associated with the increased activity of the PM proton pump in cluster roots of P-deficient Lupinus pilosus. Recently, Shen et al. (2005), who investigated citrate exudation in aluminium-stressed soybean, demonstrated that citrate exudation was coupled to changes in the activity of the PM H+-ATPase, involving transcriptional and post-translational modification of the enzyme. Furthermore, it has been reported that the release of carboxylates from cluster roots of P-deficient white lupin may show diurnal variations (Watt & Evans 1999).

The goal of this work was to study the linkage between organic acids (malate and citrate) and proton exudations in fully grown cluster roots of P-deficient white lupin, highlighting the involvement and the regulatory aspects of PM H+-ATPase activity and the regulatory aspects.

MATERIAL AND METHODS

Plant material and growth conditions

White lupin plants (L. albus L. cv. Amiga; Südwestdeutsche Saatzucht, Rastatt, Germany) were grown in hydroponic conditions as described by Massonneau et al. (2001), with the exception that 48 plants were grown in 50 L containers. Seeds were soaked overnight in aerated water, and then kept for 3 d in the dark followed by 1 d in the light, on filter paper soaked in 0.2 mm CaCl2 to allow them to germinate. Seedlings were transferred to a hydroponic culture medium [0.05 mm Fe(III)-ethylenediaminetetraacetic acid (EDTA), 2.5 mm Ca(NO3)2, 0.9 mm K2SO4, 0.8 mm MgSO4, 38 µm H3BO3, 12.5 µm MnSO4, 1.25 µm CuSO4, 1.25 µm ZnSO4, 0.33 µm (NH4)6Mo7O24, 62.5 µm KCl and with 0.25 mm KH2PO4 in case of a P-sufficient condition]. Plants were grown at 22 °C and 65% relative humidity with a light period of 16 h at 200 µmol m−2 s−1 for 5 weeks. For most of the experiments, the root samples were collected from P-deficient white lupins at different times after light start (HALS). Only the fully grown parts of cluster roots were collected (see also Fig. 1), corresponding to the immature and mature parts described by Massonneau et al. (2001). For the measurement of the cytosolic and vacuolar pH under −P and +P conditions, we harvested the apical part (the last centimeter) of lateral roots.

Figure 1.

Schematic view of a white lupin plant grown for 5 weeks in a phosphate (P)-deficient nutrient solution and a detailed cluster root. This figure represents a typical cluster root and whole plant of white lupin grown for 5 weeks in nutrient solution without any external supply of P. Moreover, the different parts of the cluster roots are indicated following the nomenclature of Massonneau et al. (2001). In this work, immature and mature parts were collected together and named fully grown cluster root.

Quantification of citrate and malate released by cluster roots

Collection of root exudates was performed according to Massonneau et al. (2001). Fully grown cluster roots were collected, pre-incubated in 10 mm CaSO4 and incubated in 10 mm CaSO4 in 2-(N-morpholino) ethanesulfonic acid and 1,3-bis[tris(hydroxymethyl)-methyloamino] (MES-BTP) 15 mm (pH 6) in the absence or presence of 10 µm fusicoccin or 500 µm vanadate for 1 h at room temperature with agitation. Root weight was determined, and roots were used for the detection of root-induced change of pH (see further discussion). Sample solutions were frozen at −80 °C until processing. Citrate or malate content in the exudates was determined using the kit for citric acid or malic acid test (Boehringer, Mannheim, Germany) according to the manufacturer's instructions. Each data set represents the mean of three independent experiments run in triplicate, processed and analysed statistically (t-test, n = 3, P < 0.05) using Sigma Plot 8 (Systat software, Point Richmond, CA, USA).

Detection of root-induced change of rhizospheric pH

The acidification capability of cluster roots was determined as described by Massonneau et al. (2001). After the collection of the exudates, roots were placed on a glass plate covered by a 2 mm layer of 1% agar containing 0.04% (w/v) bromocresol purple at pH 6.0. Roots, which strongly acidify the medium, induce a colour shift of the pH indicator from purple to yellow, corresponding to a pH below 5.5. Images were captured after 30 min of incubation.

Nuclear magnetic resonance (NMR) measurements and P concentration

Roots were collected and equilibrated for 10 min in the perfusion medium (0.5 mm CaSO4, 1 mm MES-BTP, pH 6.1). 31P-NMR spectra were recorded on a standard broadband 10 mm probe on a Bruker AMX 600 spectrometer (Bruker Analytische Messtechnik GmbH, Rheinstetten-Forchheim, Germany) equipped with Indy computer running XWIN-NMR version 2.6. In vivo experiments were carried out by packing excised root segments in a 10-mm-diameter NMR tube equipped with a perfusion system connected to a peristaltic pump in which the aerated, thermoregulated (26 °C) medium [0.5 mm CaSO4, 1 mm MES-BTP (pH 6.1)] flowed at 10 mL min−1.

31P-NMR spectra were recorded at 242.9 MHz without lock, with a waltz-based broadband proton decoupling and a spectral window of 16 kHz. Chemical shifts were measured relative to the signal from a glass capillary containing 33 mm methylenediphosphonate (MDP), which is at 18.5 mg g−1 relative to the signal from 85% H3PO4. The spectra were determined using a 90° pulse angle and fast acquisition conditions (1−s recycle time). Resonance assignments were performed according to Roberts et al. (1980) and Kime et al. (1982). Intracellular pH values were calculated from the chemical shift of the cytosolic and vacuolar Pi resonance after construction of a standard titration curve (Roberts, Wadejardetzky & Jardetzky 1981).

Isolation of PMs

PM vesicles were isolated from fully grown cluster roots as following Santi et al. (1995) and Fischer-Schliebs, Varanini & Lüttge (1994). Briefly, 5 g of roots were homogenized with a mortar and pestle in a freshly prepared ice-cold extraction medium [250 mm sucrose, 2 mm MgSO4, 2 mm ATP, 10% (v/v) glycerol, 10 mm glycerol-1-phosphate, 0.16% (w/v) bovine serum albumin (BSA), 2 mm ethyleneglycoltetraacetic acid (EGTA), 2 mm dithiothreitol (DTT), 5.7% (w/v) choline-iodide, 1 mm phenylmethanesulphonylfluoride (PMSF), 20 µg mL−1 chymostatin, 10 nm okadeic acid and 25 mm MES-BTP (pH 7.6)]. Four millilitres of medium per gram fresh weight of root tissue was used.

The homogenates were filtered through four layers of cheesecloth, and the suspensions were subjected to differential centrifugation steps at 2 °C: 1500 g for 5 min (pellets discarded), 9800 g for 20 min (pellets discarded), 83 400 g for 30 min (pellets recovered) and 83 400 g for a further 30 min. Microsomes, gently resuspended in 1.2 mL of homogenization medium, were loaded onto a discontinuous sucrose gradient made by layering 2 mL of sucrose solution (1.13 g cm−3) onto a 3 mL sucrose (1.17 g cm−3) cushion, and were centrifuged at 107 600 g for 2 h. The sucrose solutions were prepared in 5 mm MES-BTP, pH 7.4, and contained all of the protectants present in the homogenization medium. Vesicles migrating to the 1.13/1.17 g cm−3 interface were collected, diluted with homogenization medium and centrifuged at 122 400 g for 30 min. The pellets were resuspended in a medium containing 250 mm sucrose, 10% (v/v) glycerol, 1 mm DTT, 50 µg mL−1 chymostatin, 10 nm okadeic acid and 2 mm MES-BTP (pH 7.0), were immediately frozen in liquid nitrogen and were stored at −80 °C until use.

Measurement of PM H+-ATPase activity and membrane protein content

PM H+-ATPase activity was measured at 38 °C in a 0.6 mL reaction [50 mm MES-BTP pH 6.5 or pH 6.2 to 8.0 for the pH dependency assay, 5 mm MgSO4, 100 mm KNO3, 600 µm Na2MoO4, 1.5 mm NaN3, 5 mm ATP-BTP pH 6.5, 0.01% (w/v) Brij 58 (polyoxyethylene 20 cetyl ether), plus or minus 100 µm V2O5; the vanadate-dependent activity was 85 ± 3%]. The reaction was started by addition of membrane vesicles containing 0.5 µg of total protein; after 30 min, the reaction was stopped and colour developed as previously described by Santi et al. (1995). Inorganic P was quantified spectrophotometrically at 705 nm as described by Forbush (1983).

Protein content was determined as described by Bradford (1976), using BSA as standard, after solubilizing membrane vesicles with 0.5 m NaOH (Gogstad & Krutnes 1982).

Western blots

Equal amounts of protein isolated at the different time points were loaded, electrophoresed in an 8% w/v sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) gel and transferred to a Protran BA 83 nylon membrane (0.2 µm; Bio-Rad, Hercules, CA, USA) with a semi-dry transfer system (Trans-blot SD; Bio-Rad). For the PM H+-ATPase blot, a polyclonal antibody against the C-terminal part of the Arabidopsis thaliana AHA3 PM H+-ATPase. The 14-3-3-protein blot was performed using an antiserum against barley 14-3-3, which was kindly provided by David B. Collinge (Royal Veterinary and Agricultural University, Frederiksberg C, Denmark). Secondary antibodies (Goat anti-rabbit IgG alkaline phosphatase conjugate; Bio-Rad) were used, and the immunodetection was performed using the standard 5-bromo-4-chloro-3-indolyl-phosphate/nitro blue tetrazolium (BCIP/NBT) protocol (Promega, Madison, WI, USA). Relative-intensity band quantifications were determined using ImageJ (1.40 g; http://rsb.info.nih.gov/ij/).

Gene expression analysis

At the harvesting times (0, 1, 3, 5, 7, 9 and 11 h after the initiation of the light period), fully grown cluster roots were collected, immediately frozen in liquid nitrogen and conserved until further processing at −80 °C. RNA extractions were performed using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer's instructions, and contaminant genomic DNA was removed with 10 U of DNase I (GE Healthcare, Munich, Germany). Total-RNA samples were cleaned up using the phenol : chloroform protocol (Maniatis, Sambrook & Fritsch 1989). One microgram of total RNA (checked for quality and quantity using a spectrophotometer, followed by a migration in an agarose gel) of each sample was retrotranscribed using 1 pmol of Oligo d(T)23VN (New England Biolabs, Inc., Beverly, MA, USA), 15 U Prime RNase Inhibitor (Eppendorf, Hamburg, Germany) and 10 U M-MulV RNase H- for 1 h at 42 °C (Finnzymes, Helsinki, Finland) following the manufacturer's instruction. After RNA digestion with 1 U RNase A (USB, Cleveland, OH, USA) for 1 h at 37 °C, gene expression analyses were performed by adding 0.1 µL of the cDNA to FluoCycleTM sybr green (20 µL final volume; Euroclone, Pero, Italy) in a DNA Engine Opticon Real-Time PCR Detection (Bio-Rad). Primers used (Tm = 58 °C) were the following: for LHA1 gene (AY989893), CCATTCATTTCTCTTTTGGGATA and GAAGACAAAGCTCAATAACCAGAA; for LHA2 gene (AY989895), GGAGACTGGCCGAAGACTT and CGGGAATTGAGGCAATACTC; for LHA3 gene (AY989894), cagggcaattttccaaagaa and acctccagagcaaggcaata; and as housekeeping gene (polyubiquitin; DQ118117), GCACCCTAGCCGACTACAAC and CCGGTAAGGGTCTTGACAAA). Triplicates were performed on three independent experiments; analyses of real-time result were performed using Opticon Monitor 2 software (Bio-Rad) and R (version 2.7.0; http://www.r-project.org/) with the qPCR package (version 1.1-4; http://www.dr-spiess.de/qpcR.html). Efficiencies of amplification were calculated following the authors' indications (Ritz & Spiess 2008): PCR efficiencies were 92.3 and 81.4% for LHA1 and LHA2 genes, respectively. Computation of the graphical representation and statistical validation (t-test) were performed using SigmaPlot 8.0 (Systat software), considering the differences in the PCR efficiency and setting up that LHA1 gene expression at time 0 is equal to 1.

RESULTS

Citrate, malate and proton exudations by cluster roots

In order to get a better settlement of the diurnal variation of citrate and malate release, we collected exudates released from fully grown white lupin cluster roots (Fig. 1), corresponding to the immature and mature stages according to the nomenclature of Massonneau et al. (2001), over a period of 11 h after starting the light period (HALS). Figure 2a shows that citrate exudation progressively increased from 1 HALS to reach a maximum at 5 HALS, where it had approximately doubled. Thereafter, citrate exudation decreased to levels lower than those recorded at the beginning of the light period. Exudation rates for malate (Fig. 2b) were always lower than those recorded for citrate and did not change significantly. These results indicate that there is a differential regulation between citrate and malate exudations.

Figure 2.

Citrate (a) and malate (b) secretion from cluster roots during the light period. Fully grown cluster roots were harvested from 5-week-old phosphate (P)-deficient white lupin at different times after the start of the light period. Exudates were collected after bathing roots for 1 h in 10 mm CaSO4, 15 mm 2-(N-morpholino) ethanesulfonic acid and 1,3-bis[tris (hydroxymethyl)-methyloamino] (MES-BTP) (pH 6) with or without 10 µm fusicoccin. Means ± SD of three independent experiments are reported. Capital letters (A and B) refer to statistically significant differences (t-test, n = 3, P < 0.05) in organic acid secretion of control roots (–fusicoccin). Asterisks indicate a statistically significant (t-test, n = 3, P < 0.05) effect of fusicoccin treatment.

To investigate whether the enhanced citrate exudation was accompanied by changes in root external medium acidification, fully grown cluster roots were placed on an agar gel containing the pH indicator bromocresol purple that turns yellow in response to acidification. The time dependence of agar acidification by cluster roots (Fig. 3) corresponded well with that of citrate exudation (Fig. 2a). The basal citrate and malate release observed at the beginning or the end of the light period (Fig. 2a,b) was not accompanied by a detectable acidification of the agar medium (Fig. 3).

Figure 3.

Acidification capacity of cluster roots during the light period. Fully grown cluster roots were harvested from 5-week-old phosphate (P)-deficient white lupin at different times after light start. After incubation for 1 h in 10 mm CaSO4, 15 mm 2-(N-morpholino) ethanesulfonic acid and 1,3-bis[tris(hydroxymethyl)-methyloamino] (MES-BTP) (pH 6) in the absence or presence of 10 µm fusicoccin or 500 µm vanadate, roots were placed for 30 min on an agar gel containing the pH indicator bromocresol purple and adjusted to pH 6. Yellow and purple colours indicate pH values respectively below and above 6.

In order to verify whether the PM H+-ATPase was involved in the development of a higher citrate exudation rate, cluster roots were treated with fusicoccin, which permanently activates the PM proton pump (Johansson, Sommarin & Larsson 1993; Palmgren 1998). Fusicoccin stimulated proton release from cluster roots (Fig. 3) and was accompanied by an increase in exudation of citrate, but not of malate (Fig. 2a,b). The potency of fusicoccin was strongest in the beginning of the light period, and negligible as citrate exudation peaked at 5 HALS (Figs 2a & 4). These results indicate that activation of the proton pump induces citrate release; however at 5 HALS, citrate exudation cannot be stimulated further suggesting that the proton pump does not respond to fusicoccin, that is, it is already operating at maximal activity.

Figure 4.

Effect of vanadate and fusicoccin on citrate secretion. Fully grown cluster roots were harvested from 5-week-old phosphate (P)-deficient white lupin at 1 and 5 h after light start (HALS). Exudates were collected after bathing roots for 1 h in 10 mm CaSO4, 15 mm 2-(N-morpholino) ethanesulfonic acid and 1,3-bis[tris(hydroxymethyl)-methyloamino] (MES-BTP) (pH 6) in the absence or presence of 10 µm fusicoccin or 500 µm vanadate. Means + SD of three independent experiments are reported; relative values in comparison with the untreated control at 1 HALS are also shown. Capital letters (A, B and C) refer to statistically significant differences (t-test, n = 3, P < 0.05).

Vanadate is a well-known inhibitor of P-type ATPases, a family of ion pumps to which the PM H+-ATPase belongs (Cocucci, Ballarin-Denti & Marrè 1980). Application of vanadate to white lupin cluster roots strongly limited medium acidification (Fig. 3). However, citrate release was only partially affected by the inhibitor (Fig. 4), the degree of inhibition being dependent on the time at which cluster roots were treated. At 1 HALS, citrate exudation was almost insensitive to vanadate (Fig. 4), whereas vanadate was able to completely inhibit the enhanced element of citrate release, reverting the exudation rate to the level measured at 1 HALS (Fig. 4). This strongly suggests an involvement of PM H+-ATPase in the increased level of citrate release in cluster roots of P-deficient white lupin.

Time course of cytosolic pH fluctuations

In order to verify whether the burst of citrate and proton release in fully grown cluster roots were reflected by changes in the cytosolic pH, an in vivo NMR spectroscopy approach was used. For this purpose, the chemical shift (δ) of phosphate (31P) resonance was followed. Cytosolic and vacuolar pH values were calculated using a standard titration curve designed for a ‘classical’ cellular composition (Roberts et al. 1981; Spickett, Smirnoff & Ratcliffe 1992). Figure 5 illustrates typical NMR profiles measured on roots of P-deficient and P-sufficient white lupin.

Figure 5.

In vivo31P-nuclear magnetic resonance (NMR) spectra of white lupin cluster roots. Apical segments of lateral roots (the last centimeter) were harvested from plants grown for 5 weeks in the absence (a) or presence (b) of Pi. Each spectrum was recorded using fast acquisition conditions with a recycle time of 1 s and represents the sum of 2700 scans. The numbered peaks correspond to (1) cytosolic phosphate (P) and (2) vacuolar P. The region including cytosolic P is shown also on an expanded scale (4×). The chemical shift of the cytosolic P peaks, reflecting the differences in pH values, is highlighted by a vertical dotted line.

Table 1 shows that the cytosolic pH changed, paralleling the release of protons and citrate (Figs 2a & 3). At the beginning of the light period, the cytosolic pH was 7.8 and subsequently increased to reach a maximum of pH 8.2 after 5 h, where citrate exudation also peaked (Fig. 2a). This alkalinization could be a consequence of proton extrusion from the cytosol to the external solution (Fig. 3). Following the maximum alkalization at 5 HALS, the cytosolic pH decreased. Addition of fusicoccin similarly resulted in an alkalinization of the cytosol (Table 1) as has also been observed previously (Espen et al. 2000).

Table 1.  Variations of cytosolic and vacuolar pH of cluster-root cells during the light period
HALS11 + FC357911
  1. Cluster roots were collected from cluster roots of 5-week-old phosphate (P)-deficient white lupin at different times after the start of the light period. Cytosolic and vacuolar pH values were derived from the chemical shift (δ) of cellular P resonance after construction of a standard titration curve with or without 10 µm fusicoccin (FC), using 31P-nuclear magnetic resonance (NMR) in in vivo experiments. Values are the means of two independent experiments. Data are presented as differences from the cytosolic pH value determined at 1 h after light start (HALS), which was 7.78. The accuracy is calculated as three times the maximum variation observed in the experiments and is equal to ±0.02 pH units for the cytoplasm and ±0.04 pH units for the vacuole.

  2. nd, not detectable.

ΔCytosolic pH0.680.130.42−0.07−0.46−0.36
ΔVacuolar pH−2.51−2.38ndndndndnd

Activity, amount and regulation of the PM H+-ATPase

To confirm that the PM H+-ATPase was activated in concomitancy with organic acid release, PM vesicles were isolated from fully grown cluster roots at different time points and their PM H+-ATPase activities were characterized (Fig. 6). Notably, the activity of the PM H+-ATPase in vesicles changed during the light period and followed a pattern similar to that observed for citrate and proton exudations (see Figs 2a & 3). Thus, it progressively increased up to 5 HALS and subsequently decreased to a basal value at 11 HALS. These results clearly indicate the concomitance of the citrate-exudation burst (Fig. 2a), the acidification of the root external medium (Fig. 3) and the increase in the ATP hydrolytic activity of the PM H+-ATPase (Fig. 6).

Figure 6.

Plasma membrane (PM) H+-ATPase activity in cluster roots during the light period. PM vesicles were isolated from cluster roots of 5-week-old phosphate (P)-deficient white lupin at different times after the start of the light period. The enzyme activity was measured at pH 6.5. Data are means ± SD of two independent experiments.

In order to determine whether the enhancement of PM H+-ATPase activity occurred at the transcriptional and/or at the post-transcriptional level, we determined (1) the expression levels of three PM H+-ATPase genes isolated from white lupin roots; (2) the amount of PM H+-ATPase protein; (3) the amount of 14-3-3 bound to the PMs; and (4) the pH dependency of the PM H+-ATPase activity, which is a measure of its regulatory state (Palmgren 1998).

Transcriptional activity of PM H+-ATPase genes was investigated by real-time RT-PCR analyses using primers designed from gene sequences of three different H+-ATPase isoforms known to be present in white lupin: LHA1 (AY989893), LHA2 (AY989895) and LHA3 (AY989894). PM H+-ATPase genes showed different expression patterns during the light period (Fig. 7). A distinct peak in transcript abundance was observed for LHA1 at 3 HALS; LHA2 was expressed at a lower level as compared with LHA1, although showing a maximal transcript abundance at 7 HALS. Expression of LHA3 was not detectable over the whole experimental period (data not shown).

Figure 7.

Expression analyses of genes coding for different plasma membrane (PM) H+-ATPase isoforms of cluster roots during the light period. Total RNA was extracted from cluster roots of 5-week-old phosphate (P)-deficient white lupin at different times after the start of the light period. Relative expression of LHA1 and LHA2 genes coding for different PM H+-ATPase isoforms of white lupin was analysed by real-time RT-PCR. Data are means ± SD of three independent experiments. Changes in gene expression were calculated on the basis of expression levels of LHA1 at 1 h after the start of the light period (HALS); relative expression of LHA2 gene was also illustrated in the insert with a higher magnification. Asterisks indicate a statistically significant (t-test, n = 3, P < 0.05) difference in expression level of each gene with respect to its expression at 1 HALS.

In order to investigate whether changes in the activity of PM H+-ATPase could be explained by changes in the amount of the corresponding protein, we performed Western blot analyses using polyclonal antibodies targeted against the C-terminal domain of the enzyme (Fig. 8). Following the onset of the light period, an increase in the amount of H+-ATPase protein was indeed observed, with a peak at 5 HALS; thereafter, the protein level progressively decreased. The changes in PM H+-ATPase protein levels followed the pattern observed for the ATP hydrolytic activity of the pump (see Fig. 6).

Figure 8.

Western blot analysis of the plasma membrane (PM) H+-ATPase and 14-3-3 proteins in cluster roots during the light period. Purified PMs isolated from cluster roots of 5-week-old phosphate (P)-deficient white lupin at different times after the start of the light period were used for immunodetection of PM H+-ATPase using antibodies raised against the C-terminal part of the Arabidopsis AHA3 PM H+-ATPase and of 14-3-3-protein using antibodies raised against the barley 14-3-3 protein. Quantification of western blot signals is expressed as percentage of that recorded at 1 h after light start (HALS). Data of a representative experiment are reported.

14-3-3 protein is involved in the activation of PM proton pump by binding to its C-terminal domain (Johansson et al. 1993; Palmgren 1998). The possible occurrence of a post-translational regulation of the PM proton pump was analysed by observing changes in the amount of activating 14-3-3 protein bound to PMs and by analysing the pH dependency of the enzyme, which moves to a slightly more alkaline maximum in response to post-translational activation (Palmgren 1998). Indeed, there was a significant increase in the amount of 14-3-3 protein bound to PMs during the first hours (Fig. 8). The pH optimum of PM H+-ATPase activity was 6.5 at the beginning of the light period (Fig. 9), which is consistent with the optimum of the down-regulated enzyme (Regenberg et al. 1995). During the light period, concomitantly with the burst of citrate exudation, the pH optimum shifted to more alkaline values (6.8 and 7.0 at 3 and 5 HALS, respectively). Taken together, these results suggest that post-transcriptional regulation of the PM proton pump had occurred at the time of the enhanced release of citrate.

Figure 9.

pH dependency of PM H+-ATPase activity of cluster roots during the light period. Plasma membrane (PM) vesicles were isolated from cluster roots of 5-week-old phosphate (P)-deficient white lupin at 1, 3, 5 and 9 h after light start (HALS). The enzyme activity was measured at different pH values in the range between 6.2 and 8.0. Data are means ± SD of two independent experiments.

DISCUSSION

Adaptation of white lupin to low-P soils is generally attributed to its capacity to form cluster roots, from which massive release of carboxylates (citrate and, to a lesser extent, malate) takes place (Gardner, Parbery & Barber 1981; Keerthisinghe et al. 1998). It has been shown that carboxylate release occurs according to a developmentally defined programme (Massonneau et al. 2001) and may show diurnal variations, with higher rates being recorded during the light period than in the dark (Watt & Evans 1999).

In the present study we have defined in detail the time course, during the illumination period, of citrate and malate release rates from immature and mature cluster roots of P-deficient white lupin (Fig. 1). The results show that the citrate exudation progressively increased from 1 to 5 h after the start of the light period (HALS), then decreased to a minimum at 11 HALS (Fig. 2a). A similar pattern was reported by Hocking & Jeffery (2004) for Lupinus luteus. Our data also show that white lupin cluster roots are able to release malate, but interestingly at lower rates and without showing any significant change during the considered period demonstrating a differential exudation pattern between citrate and malate (Fig. 2a,b).

A strong acidification of the rhizosphere has been reported to occur concomitantly with the release of carboxylates (Neumann et al. 2000). Several studies have addressed the question whether a link exists between carboxylate and proton exudations. Because of the form (anionic) of carboxylic acids at cytosolic pH values (Ma et al. 2001) and the presence of citrate transport systems at the PM of white lupin (Zhang et al. 2004), protons are likely to be needed for energization and charge compensation during release of carboxylates. Increased proton release has also been related to the increased cation/anion influx ratio of P-deficient plants (Shen et al. 2005) and to the presence of the rhizobium-legume symbiosis (Sas et al. 2001). Activation of the PM H+-ATPase has been suggested to be responsible for proton exudation associated with citrate export in white lupin (Kania et al. 2001; Yan et al. 2002). Recently, Zhu et al. (2005), studying the effect of the PM H+-ATPase effectors fusicoccin and vanadate on release of carboxylates (malate and citrate) and several ions (cations and anions) from cluster roots of P-deficient white lupin, concluded that proton extrusion may serve as charge balance for malate release, while sodium or potassium might act as counterions for citrate. Results here presented suggest that the counterions might change in function of how much citrate is released; that is, it seems there are two different mechanisms of citrate release: one might be an anion channel as shown by Zhang et al. (2004), and another might be a multidrug and toxic compound extrusion system (MATE), which transports citrate but not malate, as hypothesized by Furukawa et al. (2007).

In the present study, we show that rhizosphere acidification by immature and mature proteoid roots is concomitant with the burst of citrate exudation (Fig. 3). As expected, proton release was enhanced by fusicoccin treatment. Citrate extrusion could also be enhanced by fusicoccin, but not when the rate of release was maximal (at 5 HALS, Figs 2a & 4). This result indicates that either the proton pump was already fully activated at 5 HALS or that a maximal citrate synthesis or exudation activity has been reached at that time, which could not be further increased by activation of the PM H+-ATPase in response to fusicoccin. After 11 h of illumination, fusicoccin had only a negligible effect on citrate release while it was still able to stimulate proton efflux, suggesting that at this time, the internal supply of citrate might have become limiting. On the other hand, malate release was not affected by the fusicoccin treatment (Fig. 2b).Vanadate virtually abolished rhizosphere acidification (Fig. 3); on the other hand, the effect of the inhibitor on citrate release was dependent on the time of application, with a decrease of about 50% being observed at 5 HALS and only of 17% at 1 HALS (Fig. 4).

The marked variations in citrate and proton release from the cluster roots during the light period could conceivably influence the cytosolic pH; therefore, we performed time course measurements with NMR instrumentation (Fig. 5) and observed that during the burst of citrate and proton exudations, an alkalinization of the cytosol took place (Table 1), indicating that the cellular pH stat could not fully compensate for the release of protons, even if it is known that the capacity to regulate pH should be increased in white lupin cluster roots; in fact, phosphoenolpyruvate (PEP) carboxylase, an enzyme, which is postulated to play an important role in pH regulation, was shown to be up-regulated in these tissues (Johnson, Vance & Allan 1996). The lowest pH values were measured concomitantly with a decrease in proton and citrate efflux (from 7 to 11 HALS). Fusicoccin, applied at 1 HALS, also led to an increase in cytosolic pH (Table 1), as a consequence of PM H+-ATPase activation (Espen et al. 2000).

PM H+-ATPase activity also changed during the light period (Fig. 6), closely paralleling citrate and proton extrusion; these modifications were consistent with variations in cytosolic pH. Data on the cytosolic pH and PM H+-ATPase activity further highlight the close link between the enhanced exudation of proton and citrate in cluster roots of P-deficient white lupin. Modulation of PM H+-ATPase activity has been reported to be due either to altered gene expression or to post-translational modifications (Michelet et al. 1994; Palmgren 1998; Shen et al. 2005). The latter type of regulation is achieved by phosphorylation of the C-terminal autoinhibitory domain of the PM H+-ATPase, which enables binding of activating 14-3-3 protein (Olsson et al. 1998; Fuglsang et al. 1999). In our work, changes in PM H+-ATPase activity (Fig. 6) strictly concomitant with the burst of citrate exudation (Fig. 2a) were paralleled by similar changes in protein amount (Fig. 8); furthermore, increased transcript levels for at least one of the genes encoding the three known enzyme isoforms could be recorded (Fig. 7). These data are consistent with a transcriptional regulation of the PM H+-ATPase. On the other hand, the shift in pH optimum (Fig. 9) and the pattern of changes in 14-3-3 protein amount (Fig. 8) suggest the occurrence also of a post-translational regulation of the enzyme. Up-regulation of PM H+-ATPase due to transcriptional and post-translational modifications has been reported for aluminium-stressed soybean root tips (Shen et al. 2005); furthermore, a transcriptional regulation of PM H+-ATPase involving a specific isoform has also been demonstrated in cucumber plants exposed to Fe-deficiency (Santi et al., 2005).

The present study demonstrates for the first time the involvement of a two-component system in organic acid exudation in cluster roots of P-deficient white lupin: one, that sustains a basal rate of malate and citrate release, and another, that sustains the burst of citrate exudation, which is closely linked to active proton extrusion from the cytosol. Further, the results indicate that two different mechanisms of organic acid exudation are operating. The first component mediates a basal rate of malate and citrate release corresponding to about 2 µmol of citrate and 1 µmol of malate per hour and gram of fresh weight (Figs 2a,b & 4), and is not directly linked to the PM H+-ATPase activity. The second one corresponds to the burst of citrate release (which is about twofold higher than the basal citrate release) and is strictly PM H+-ATPase activity dependent.

ACKNOWLEDGMENTS

This work was supported by the University of Zürich, the Swiss National Foundation within the National Center of Competence in Research ‘Plant Survival’, by the Körber Stiftung (Germany), Italian Ministry for University Education and Research (MIUR) and by a grant from Italilan National Research Council (CNR) and Swiss National Science Foundation (FNS/SNF) to S.C.

Ancillary