Characterization of the interaction between the plasma membrane H+-ATPase of Arabidopsis thaliana and a novel interactor (PPI1)

Authors


M. I. De Michelis, Dipartimento di Biologia ‘L. Gorini’, Università di Milano, CNR Istituto di Biofisica – Sezione di Milano, via G. Celoria 26, 20133 Milano, Italy
Fax: +39 02 50314815
Tel: +39 02 50314822
E-mail: mariaida.demichelis@unimi.it

Abstract

Proton pump interactor, isoform 1 (PPI1) is a novel interactor of the C-terminus of Arabidopsis thaliana plasma membrane H+-ATPase (EC 3.6.3.6) (Morandini P, Valera M, Albumi C, Bonza MC, Giacometti S, Ravera G, Murgia I, Soave C & De Michelis MI (2002) Plant J31, 487–497). We produced two fusion proteins consisting of, respectively, the first 88 amino acids or the entire protein deleted of the last 24 hydrophobic amino acids, and we show that the latter protein has a threefold higher affinity for the H+-ATPase. PPI1-induced stimulation of H+-ATPase activity dramatically decreased with the increase of pH above pH 6.8, but became largely pH-independent when the enzyme C-terminus was displaced by fusicoccin-induced binding of 14-3-3 proteins. The latter treatment did not affect PPI1 affinity for the H+-ATPase. These results indicate that PPI1 can bind the H+-ATPase independently of the C-terminus conformation, but is not able to suppress the C-terminus auto-inhibitory action.

Abbreviations
Brij 58

polyoxyethilene 20 cethyl ether

BTP

bis tris propane {1,3-bis[tris(hydroxymethyl)methylamino]propane}

FC

fusicoccin

GST

glutathione S-transferase

IPTG

isopropyl thio-β-d-galactoside

NTA

nitrilotriacetic acid

PM

plasma membrane

The H+-ATPase is the major electrogenic pump in the plasma membrane (PM) of plant cells. By pumping protons from the cytoplasm to the apoplast it generates an electrochemical proton gradient, which drives the transport of mineral ions and organic solutes, and plays a crucial role in cytoplasmic and apoplastic pH homeostasis [1,2]. The PM H+-ATPase participates in a variety of physiological processes such as phloem loading, stomata opening, mineral nutrition, growth of root hairs and pollen tubes, salt and osmotolerance, leaf movements, and acid growth [1,2]. In vivo, its activity is modulated by several signals such as hormones (auxin, abscisic acid), light, water potential, acid load, toxins like fusicoccin (FC) and pathogens, but a molecular description of the mediators involved is missing for most of these signals [1,2].

Plant genomes contain a large family of PM H+-ATPase genes (12 in Arabidopsis thaliana, 10 in rice and nine in Nicotiana plumbaginifolia), which can be grouped in five clusters based on sequence alignments and intron positions [3,4]. Individual isoforms exhibit tissue- and developmental-specific expression patterns and a number of quantitative differences in catalytic and regulatory properties [1–4]. Thus, the first regulation of proton pumping activity in different cells types and physiological conditions takes place at both the transcriptional and translational levels [1–4].

As to post-translational regulation, the best-known mechanism described to date involves the auto-inhibitory action of the C-terminal domain. The plant PM H+-ATPase is a P-type ATPase with an extended (approximately 100 amino acids) cytosolic C-terminus containing two inhibitory regions. Proteolytic cleavage or genetic deletion of the C-terminus has little effect on enzyme activity at the acidic pH optimum (pH 6.4–6.6), but markedly increases enzyme activity in the physiological range of cytoplasmic pH values (pH 7.0–7.5), resulting in an alkaline shift of the pH optimum ([1] and references therein). The auto-inhibitory action of the C-terminus is suppressed, besides by pH, by lysophospholipids and by 14-3-3 proteins ([1] and references therein). The latter are regulatory proteins present in all eukaryotic systems which modulate the activity of a number of target proteins, generally binding to sequence motifs including a phosphorylated Ser or Thr residue [5–8]. Phosphorylation of the highly conserved penultimate Thr residue of the PM H+-ATPase results in binding of 14-3-3 protein. 14-3-3 binding is stabilized by the fungal toxin FC which decreases the dissociation rate, thus inducing the formation of an almost irreversible complex in which the enzyme is locked in the same active conformation determined by cleavage of the C-terminus [1,9–16]. Also, blue-light activation of PM H+-ATPase in guard cells of broad beans involves protein kinase-mediated phosphorylation of Ser and Thr residues in the C-terminus of the pump and 14-3-3 binding [17]. Much less is known about in vitro and in vivo activation by other effectors: an increase of PM-associated 14-3-3s has been observed also in response to cold or osmotic stress, and their binding to the H+-ATPase is suggested by the parallel increase of the number of FC-binding sites [18–20]. As to auxin, a soluble auxin receptor has been reported to bind and activate the PM H+-ATPase, but the site of binding has not been identified so far [21].

A novel interactor of the PM H+-ATPase C-terminus was identified in a two-hybrid screening. The novel protein, named PPI1 (proton pump interactor, isoform 1), is a 612 amino acids protein rich in charged and polar residues, except for the extreme C-terminus where it presents a hydrophobic stretch of 24 amino acids forming a putative transmembrane domain. PPI1 does not resemble any protein of known function, but it is probably the first identified member of a new family of plant regulatory proteins, as at least five A. thaliana genes and many expressed sequence tags (ESTs) from different plant species encode proteins with significant similarity to PPI1 [22].

The N-terminal domain of PPI1, originally identified by the two-hybrid technique, binds A. thaliana PM H+-ATPase in overlay experiments and stimulates enzyme activity. The interaction is not suppressed by controlled tryptic cleavage of the enzyme, indicating that the PPI1 binding site in the H+-ATPase C-terminus is localized upstream of the main tryptic cleavage site and thus is different from the 14-3-3 binding site. Moreover, PPI1 further enhances FC-stimulated H+-ATPase activity [22].

Here we report a characterization of the interaction of PPI1 with the H+-ATPase in PM isolated from control and FC-treated A. thaliana cultured cells, which indicates that PPI1 is unable to suppress the auto-inhibitory action of the enzyme C-terminus, but further enhances the activity of the enzyme whose C-terminus has been displaced by low pH or by FC-induced binding of 14-3-3s.

Results

The C-terminus of isoform 1 of the PM H+-ATPase of A. thaliana (AHA1) interacts with the first 88 amino acids of PPI1 [22], indicating that the PM H+-ATPase binding site of PPI1 is localized therein. Indeed, fusion proteins containing the first 88 amino acids of PPI1, linked either to a His-tag (His6PPI188) or to GST, interact with A. thaliana H+-ATPase in the PM and stimulate its activity [22]. However, other parts of the protein may be important for regulation of the interaction. As the entire PPI1 protein was difficult to handle due to low solubility (unpublished results from the authors' laboratory), we expressed in Escherichia coli a truncated protein devoid of the hydrophobic C-tail, with a His6-tag at the C-terminal end, far away from the interaction site (PPI1588His6). The fusion protein was purified by Ni-NTA affinity chromatography and its ability to interact with the H+-ATPase C-terminus was tested against another fusion protein harboring the last 104 amino acids of AHA1 fused to GST, GST–AHA1(846–949). Figure 1 shows that PPI1588His6 and GST–AHA1(846–949) bound each other in overlay experiments both when a membrane spotted with GST–AHA1(846–949) was incubated with PPI1588His6 (Fig. 1A) and when PPI1588His6 was spotted and the membrane incubated with GST–AHA1(846–949) (Fig. 1B); the signals were specific as no signal was detected when free GST was spotted and the membrane incubated with PPI1588His6 (Fig. 1A) or when an unrelated His-tagged protein was spotted and the membrane incubated with GST–AHA1(846–949) (Fig. 1B).

Figure 1.

Interaction between PPI1588His6 and the C-terminus of A. thaliana PM H+-ATPase (AHA1). The indicated proteins were spotted and incubated with 1 µm PPI1588His6 (A) or 1 µm GST–AHA(1846–948) (B) as described in Experimental procedures. Interaction was detected by immunodecoration with antisera against the N-terminus of PPI1 (A) or the C-terminus of the H+-ATPase (B). His6–ACA8(1–116) reproduces the N-terminus of an A. thaliana PM Ca2+-ATPase [31]. Results are from one experiment, representative of three giving similar results.

The ability of PPI1588His6 to activate the H+-ATPase in PM isolated from cultured A. thaliana cells was compared to that of His6PPI188. Figure 2 shows that both proteins stimulated the H+-ATPase activity assayed at pH 6.4 in a concentration-dependent manner, but PPI1588His6 was about threefold more effective than His6PPI188. The k0.5 values evaluated from five independent experiments were 0.4 ± 0.1 µm for PPI1588His6 and 1.7 ± 0.2 µm for His6PPI188. Thus, all the following experiments were performed with PPI1588His6.

Figure 2.

Stimulation of A. thaliana PM H+-ATPase activity as a function of the concentration of His6PPI188 and PPI1588His6. PM treatment with the specified concentrations of His6PPI188 (closed triangles) or PPI1588His6 (open triangles) and H+-ATPase activity assays were performed at pH 6.4. Results are given as percentage stimulation of H+-ATPase activity which in the absence of PPI1 was 665 nmol·min−1·mg protein−1. Results are from one experiment, representative of five giving similar results.

The analysis of the effect of PPI1588His6 on the dependence of PM H+-ATPase activity on the concentration of MgATP (Fig. 3) showed that stimulation decreased with the increase of PPI1588His6 concentration. Consequently, PPI1588His6 only slightly increased Vmax but substantially lowered the apparent Km for MgATP.

Figure 3.

Effect of PPI1588His6 on the dependence of PM H+-ATPase activity on the concentration of MgATP. PM treatment with (open symbols) or without (closed symbols) 2 µm PPI1588His6 and H+-ATPase activity assays (pH 6.4) were performed as described in Experimental procedures, except that ATP concentration was varied between 0.1 and 2 mm, as indicated, in the presence of a constant excess of 2 mm MgSO4. Results are from one experiment, representative of three giving similar results. The mean Vmax and apparent Km values were, respectively, 1.20 ± 0.04 µmol·min−1·mg protein−1 and 0.35 ± 0.05 mm in the absence and 1.37 ± 0.06 µmol·min−1·mg protein−1 and 0.13 ± 0.02 mm in the presence of PPI1588His6.

Activation of the PM H+-ATPase by cleavage or by displacement of the auto-inhibitory C-terminal domain is strongly pH-dependent, increasing with the increase of pH beyond the relatively acidic pH optimum of enzyme activity [1,23–26]. The dependence of H+-ATPase activation by PPI1588His6 on the pH of the assay medium is completely different: Fig. 4 shows that the effect of PPI1588His6 on H+-ATPase activity was very high at pH 6.0, but decreased with the increase of pH, virtually disappearing above pH 7.0. As a consequence, the pH optimum for enzyme activity is slightly more acidic in the presence of PPI1588His6 than in its absence.

Figure 4.

pH dependence of the activation of A. thaliana PM H+-ATPase by PPI1588His6. PM treatment with (open symbols) or without (closed symbols) 2 µm PPI1588His6 and H+-ATPase activity assays were performed at the specified pHs. Results are from one experiment, representative of three giving similar results.

A completely different picture emerged when the effect of PPI1588His6 on H+-ATPase activity was assayed in PM isolated from FC-treated cells. FC determines a stable association of 14-3-3 proteins to the C-terminus of the H+-ATPase, locking the enzyme in an active conformation [9,10,12–16]. Consequently (Fig. 5), enzyme activity stayed high throughout the pH range examined (up to pH 7.1). Addition of PPI1588His6 further enhanced the H+-ATPase activity in a pH-independent manner.

Figure 5.

pH dependence of the activation of A. thaliana H+-ATPase in PM purified from FC-treated cultured cells by PPI1588His6. PM treatment with (open symbols) or without (closed symbols) 2 µm PPI1588His6 and H+-ATPase activity assays were performed at the specified pHs. Results are from one experiment, representative of three giving similar results.

The different conformation of the enzyme C-terminus in PM isolated from control or FC-treated cells may alter the accessibility of PPI1588His6. To test this possibility, we analyzed PPI1588His6-induced activation of the H+-ATPase in the two PM fractions as a function of PPI1588His6 concentration. Assays were performed at pH 7.0 to ensure at the same time effective auto-inhibition and reliable measurements of PPI1 effect in control PM. The results reported in Fig. 6 show that stimulation of the H+-ATPase activity in PM isolated from control or FC-treated cells similarly increased with the increase of PPI1588His6 concentration; the k0.5 values evaluated from three independent experiments were 0.24 ± 0.06 and 0.19 ± 0.02 µm, respectively, for control PM and PM from FC-treated cells.

Figure 6.

Dependence on the concentration of PPI1588His6 of the stimulation of H+-ATPase activity in PM purified from control and FC-treated cultured cells. Assays were performed at pH 7.0. Results are given as percentage stimulation of H+-ATPase activity which in the absence of PPI1 was 261 (control, open triangles) and 507 (FC-treated, closed triangles) nmol·min−1·mg protein−1. Results are from one experiment, representative of three giving similar results.

Discussion

PPI1 is a modulator of the plasma membrane H+-ATPase, which binds the enzyme C-terminus and stimulates its activity [22]. The available preliminary evidence indicates that its mechanism of action is different from that of 14-3-3 proteins, the best known modulators of the autoinhibitory action of the enzyme C-terminus ([1,22] and references therein), proposing PPI1 as a novel mechanism of regulation which could play an important role in the subtle modulation of proton extrusion in response to endogenous or exogenous signals.

The two-hybrid screen for interactors of the C-terminus of AHA1 led to the isolation of a cDNA fragment encoding the first 88 amino acids of PPI1 [22]. This result, together with the finding that fusion proteins containing the first 88 amino acids of PPI1 linked to an His-tag (His6PPI188) or to GST interact with A. thaliana H+-ATPase in the PM and stimulate its activity [22], suggested that the site of interaction with the PM H+-ATPase was localized in the N-terminus of PPI1. To further characterize the biological activity of PPI1 we produced a new fusion protein, containing the PPI1 protein devoid only of the last 24 amino acids, a putative transmembrane domain (PPI1588His6); the His-tag was fused to the protein C-terminus, to minimize its effects on the conformation of the protein N-terminus. The results reported in this paper show that this fusion protein has an affinity for the H+-ATPase threefold higher than that of His6PPI188. This result suggests that residues downstream of the first 88 amino acids of PPI1 may participate in the interaction with the H+-ATPase and makes PPI1588His6 a suitable tool to study the mechanism of action of PPI1.

The analysis of the pH dependence of PPI1-induced activation of the H+-ATPase showed that stimulation decreases dramatically with the increase of pH above pH 6.8; PPI1-induced activation of the H+-ATPase becomes pH-independent in PM isolated from FC-treated cells. At pH values above the optimum for H+-ATPase activity, the C-terminus exerts its auto-inhibitory action, presumably by binding to an intramolecular site [1,23–27]; thus, it might hamper the access of PPI1588His6. FC-induced binding of 14-3-3 displaces the C-terminus [1,9–16] and thus might facilitate the binding of PPI1. However, the k0.5 value for the PPI1–H+-ATPase interaction at pH 7.0 was at least as low as at pH 6.4 and not affected by FC-induced 14-3-3 binding, indicating that the affinity of the H+-ATPase for PPI1588His6 is not altered by the conformation of the C-terminus. These results indicate that (Fig. 7) PPI1, in response to an as yet unidentified signal, can interact with the PM H+-ATPase independently from its activation state, but is not able to suppress the auto-inhibitory action of the C-terminal domain. PPI1 can only hyper-activate H+-ATPase molecules whose C-terminus has been displaced by other factors such as low pH or 14-3-3 proteins.

Figure 7.

Schematic model of the mechanism of action of PPI1 on the PM H+-ATPase.

Experimental procedures

Strains, media and general techniques

Escherichia coli XL10 (Stratagene, La Jolla, CA, USA) was used for recombinant DNA work while BL21(DE3)pLysS (Novagen, Madison, WI, USA) and BL21(DE3) Codon plusTM pRil strains (Stratagene) were employed as hosts for protein expression. All strains were grown in Lennox broth base (Gibco BRL, Rockville, MD, USA).

Bacterial transformation was according to the protocol of [28]. Soluble proteins were assayed with the Bio-Rad protein assay (Bio-Rad, Hercules, CA, USA) with γ-globulin as a standard, while membrane proteins were assayed according to [29] with bovine serum albumin as a standard.

Plasmid construction

A DNA fragment coding for the first 588 amino acids of PPI1 protein was amplified from clones isolated previously [22] using the following primers: gatggatcccatATGGGTGTTGAAGTTGTA annealing around the start codon of the Ppi1 ORF and gactcgagATTAGTCGACTTCTTACGC annealing just before the putative transmembrane domain (capital letter in the sequence represent nucleotides matching target sequence). The PCR product was cloned into pET-32b plasmid (Novagen) deleted of the thioredoxin gene, using NdeI and XhoI restriction sites. The resulting plasmid was transferred into E. coli strain XL10 and the frame and the identity of the cloned fragment verified by sequencing. The construct with the N-terminal portion of PPI1 has been previously described [22].

The DNA fragment coding for the last 104 amino acids (Ser846–Val949) of AHA1 was amplified from EST clones 49E5 from Arabidopsis Biological Resource Center (ABRC, Ohio State University, OH, USA) using the following primers: ggatcccatatgAGCGGAAAGGCGTGG and ggatcctcaCACAGTGTAGTGA. The PCR product was cloned into pGEX-2TK vector for fusion to GST, using the BamHI restriction site. The frame and identity of the PCR product were checked by sequencing.

Protein expression and purification

The plasmid encoding the PPI1 protein truncated of its terminal 24 hydrophobic amino acids with a His6-tag at its C-terminus (PPI1588His6) was transformed into BL21(DE3) Codon plusTM pRil strain (Stratagene) and its expression was induced in liquid cultures at 37 °C (0.6–0.7 D595) with 1 mm isopropyl thio-β-d-galactoside (IPTG). After 1 h of induction, cells were cooled on ice, centrifuged and stored at −80 °C. Cells pellets were lysed in the presence of 0.4%N-lauroyl sarcosine [30] and sonicated until a clear, non viscous solution was obtained. Particulate matters were removed by centrifugation (15 min at 12 000 g) and the soluble fraction loaded onto a Ni2+-NTA agarose affinity column. Protein was purified essentially as described by the Ni-NTA supplier (Qiagen, Milan, Italy). Eluted fractions were monitored by SDS/PAGE, pooled and concentrated by centrifugation with Vivaspin 15R (cut-off 30 kDa; Vivascience AG, Hannover, Germany). Imidazole was removed by repeated cycles of concentration-dilution with 1 mm BTP-Hepes pH 8.0, glycerol 10% (w/v) Brij 58 (Aldrich, Milwaukee, WI, USA) was added to the sample (0.1 mg·mL−1) in the first concentration cycle.

The expression of the N-terminal portion of the protein (His6PPI188) was done in the same conditions of PPI1588His6, but with 3 h of induction. Protein was purified essentially as described by the Ni-NTA supplier (Qiagen). Eluted fractions were monitored by SDS/PAGE, pooled and concentrated by centrifugation with Vivaspin 6 (cut-off 5 kDa; Vivascience AG). Imidazole was removed as described above.

The C-terminus of AHA1 fused to GST, GST–AHA(1846–949) was expressed in E. coli strain BL21(DE3)Codon plusTM pRIL (Stratagene). Cells were grown at 37 °C until D595 of 0.6 was reached, then 1 mm IPTG was added and the culture grown for 2 h. GST–AHA(1846–949) was purified by affinity chromatography on Glutathione Sepharose 4B gel (Amersham Biosciences, Piscataway, NJ, USA). The purification procedure was performed under native conditions as described in the manufacturer instructions except for the addition of 0.1% (w/v) lysozyme and 0.5% (v/v) Triton X-100 during cell lysis.

His6-ACA81−116 was produced as described by Luoni et al. [31].

Plant material and isolation of PM vesicles

Cell suspension cultures of A. thaliana ecotype Landsberg were grown as described in [32]. In vivo treatment with FC was performed for 120 min by adding the phytotoxin to the culture medium at the final concentration of 10 µm. Cells were harvested by a double centrifugation at 1000 g for 5 min; highly purified PM fractions were obtained by a two-step aqueous two-phase partitioning system as described in Olivari et al. [15].

Overlay experiments

The interaction between PPI1588His6 and the C-terminus of the PM H+-ATPase was tested both by incubating with PPI1588His6 a membrane on which GST–AHA(1846–949) (5 µm) had been spotted, and, vice versa, by incubating with GST–AHA(1846–949) a membrane on which PPI1588His6 (3 µm) had been spotted. Fusion proteins were spotted (2 µL of each) onto 0.2 µm nitrocellulose and incubated for 3 h at room temperature in blocking solution [1% (w/v) bovine serum albumin (BSA), 0.2 mm EGTA, 50 mm KNO3, 2 mm MgSO4, 5 mm (NH4)2SO4, 0.1 mm ammonium molybdate, 40 mm BTP/Mes pH 6.4]. Membranes on which GST–AHA(1846–949) was spotted were incubated for 2 h at room temperature in the same blocking solution with the addition of 1 µm PPI1588His6 and interaction was detected by immunodecoration with antiserum against the N-terminus of PPI1. Membranes on which PPI1588His6 was spotted were incubated for 2 h at room temperature in the same blocking solution with the addition of 1 µm GST–AHA(1846–949) and interaction was detected by immunodecoration with antiserum against the C-terminus of the H+-ATPase. The antiserum against the N-terminus of PPI1 was raised in rabbit using His6PPI188 as antigene. Immunodecoration was performed by incubating the membrane for 2 h at room temperature with the antiserum diluted 1 : 1000 in 20 mMTris/HCl pH 7.4, 150 mm NaCl, 3% (w/v) BSA and 0.1% (v/v) Tween20. The antiserum against the C-terminus of the H+-ATPase was raised in rabbit using as antigene the highly conserved sequence Arg912–Tyr943 of A. thaliana proton pump isoform 2 (AHA2) conjugated to ovalbumin. Immunodecoration was performed by incubating the membrane for 2 h at room temperature with the antiserum diluted 1 : 1000 in 20 mMTris/HCl pH 7.4, 150 mm NaCl, 3% (w/v) BSA and 0.1% (v/v) Tween20. After several washes, signal detection was performed with an ECL anti-rabbit IgG linked to horseradish peroxidase (Amersham Biosciences) diluted 1 : 5000 in the same solution reported above.

PM H+-ATPase activity

Unless otherwise specified, PM H+-ATPase activity was assayed in 0.2 mm EGTA, 50 mm KNO3, 2.3 mm MgSO4, 5 mm (NH4)2SO4, 0.1 mm ammonium molybdate, 1 µg·mL−1 oligomycin, 100 µg·mL−1 Brij 58, 5 µm carbonyl cyanide p-trifluromethoxy-phenylhydrazone, buffered with 40 mm BTP-Mes (pH 6.4–6.8) or BTP-Hepes (pH 7–7.3), 2 units·mL−1 pyruvate kinase, 2 mm phosphoenolpyruvate and 0.3 mm ATP.

Plasma membranes (0.5–1 µg protein) were incubated at 0 °C for 15 min with or without the specified PPI1 fusion proteins in 90 µL of assay medium in absence of ATP, pyruvate kinase and phosphoenolpyruvate; all samples contained the same volume of 1 mm BTP-Hepes pH 8.0, 10% glycerol. The volume was then adjusted to 100 µL with assay medium containing ATP, pyruvate kinase and phosphoenolpyruvate and the reaction was carried out for 60 min at 30 °C. Released Pi was determined as described in De Michelis and Spanswick [33]. The PM H+-ATPase activity was evaluated as the difference between total activity and that measured in the presence of 100 µm vanadate (less than 10% of total activity at pH 7; less than 5% of total activity at pH 6.4). Reported data are the results from one experiment with three replicates representative of at least three experiments; SE of the assays did not exceed 3% of the measured value.

Acknowledgements

This project was supported by the Italian Ministry for Instruction, University and Research in the FIRB 2001 frame.

Ancillary