H+ antiporter with a chloroplast transit peptide

Biochimica et Biophysica Acta 1818 (2012) 2362–2371

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbamem

Arabidopsis KEA2, a homolog of bacterial KefC, encodes a K +/H + antiporter with a chloroplast transit peptide María Nieves Aranda-Sicilia a, 1, 2, Olivier Cagnac a, 2, Salil Chanroj b, 3, Heven Sze b, María Pilar Rodríguez-Rosales a, Kees Venema a,⁎ a b

Dpto de Bioquímica, Biología Celular y Molecular de Plantas, Estación Experimental del Zaidín, Consejo Superior de Investigaciones Científicas, c/. Profesor Albareda 1, 18008 Granada, Spain Department of Cell Biology and Molecular Genetics and Maryland Agricultural Experiment Station, University of Maryland, College Park, MD 20742, USA

a r t i c l e

i n f o

Article history: Received 6 February 2012 Received in revised form 10 April 2012 Accepted 16 April 2012 Available online 21 April 2012 Keywords: Potassium transport Arabidopsis K+/H+ antiport Yeast Reconstitution

a b s t r a c t KEA genes encode putative K+ efflux antiporters that are predominantly found in algae and plants but are rare in metazoa; however, nothing is known about their functions in eukaryotic cells. Plant KEA proteins show homology to bacterial K+ efflux (Kef) transporters, though two members in the Arabidopsis thaliana family, AtKEA1 and AtKEA2, have acquired an extra hydrophilic domain of over 500 residues at the amino terminus. We show that AtKEA2 is highly expressed in leaves, stems and flowers, but not in roots, and that an N-terminal peptide of the protein is targeted to chloroplasts in Arabidopsis cotyledons. The full-length AtKEA2 protein was inactive when expressed in yeast; however, a truncated AtKEA2 protein (AtsKEA2) lacking the N-terminal domain complemented disruption of the Na +(K +)/H + antiporter Nhx1p to confer hygromycin resistance and tolerance to Na + or K+ stress. To test transport activity, purified truncated AtKEA2 was reconstituted in proteoliposomes containing the fluorescent probe pyranine. Monovalent cations reduced an imposed pH gradient (acid inside) indicating AtsKEA2 mediated cation/H + exchange with preference for K+ = Cs+ > Li+ > Na+. When a conserved Asp721 in transmembrane helix 6 that aligns to the cation binding Asp164 of Escherichia coli NhaA was replaced with Ala, AtsKEA2 was completely inactivated. Mutation of a Glu835 between transmembrane helix 8 and 9 in AtsKEA2 also resulted in loss of activity suggesting this region has a regulatory role. Thus, AtKEA2 represents the founding member of a novel group of eukaryote K+/H+ antiporters that modulate monovalent cation and pH homeostasis in plant chloroplasts or plastids. © 2012 Elsevier B.V. All rights reserved.

1. Introduction An essential feature of all organisms is the ability to regulate internal ion concentrations and pH in response to changing environmental conditions or during development. Prokaryotes rely on a diverse array of transporter proteins at the plasma membrane to accomplish these tasks. The development of intracellular membrane systems and compartments has led to a considerable increase in the number of ion transporters in eukaryote cells [1,2]. As a result, plants contain a large number of sequences encoding proteins that share homology to monovalent cation/proton antiporters (CPAs) [2]. The CPA superfamily can be subdivided into the CPA1 family and the CPA2 family.

⁎ Corresponding author at: Dpto de Bioquímica, Biología Celular y Molecular de Plantas, CSIC, Estación Experimental del Zaidín, C/ Profesor Albareda No. 1, E-18008 Granada, Spain. Fax: + 34 958 129 600. E-mail address: [email protected] (K. Venema). 1 Present address: Plant Genetic Engineering Laboratory, School of Agriculture and Food Sciences, University of Queensland, Brisbane, Qld 4072, Australia. 2 These authors contributed equally. 3 Present address: Department of Biotechnology, Faculty of Science, Burapha University, Chonburi 20130, Thailand. 0005-2736/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamem.2012.04.011

Eukaryotic CPA1 transporters are related to bacterial NhaP proteins and are ubiquitous in animals, plants and fungi. In plants, it comprises the intracellular NHX and plasma membrane SOS1-like antiporters [3]. Plant NHX and SOS antiporters have been shown to be important for pH and cation homeostasis and are involved in extrusion of Na + from the cells, the accumulation of Na + and K + in vacuoles or regulation of pH and ionic composition in endosomal compartments [4-7]. The CPA2 family includes genes encoding plant CHX and KEA transporters. These sequences share high homology especially to bacterial antiporters, but are rare in fungi or animals [3,8]. The number of CHX genes has increased dramatically from Charophyte alga to flowering plants. Their abundant expression in the male gametophyte as well as the roles determined so far in guard cell movement (AtCHX20) or pollen tube guidance (AtCHX21/23) indicate that CHX genes are required for reproductive and vegetative processes related to survival on land [3,9,10]. CHX genes were also suggested to play a role in salt resistance or potassium nutrition [11–13]. The intracellular isoforms AtCHX16–20 complement disruption of homologous yeast transporter Kha1p and restore tolerance to basic pH at low K + concentration, suggesting a role in pH homeostasis [14,15]. pH regulation in endomembrane compartments by CHX17 and CHX20 was suggested to be related to membrane trafficking and sorting of

MN. Aranda-Sicilia et al. / Biochimica et Biophysica Acta 1818 (2012) 2362–2371

cargo. Yeast complementation and tracer studies in bacteria indicate that CHX proteins mediate monovalent cation transport with preference for K +, though their activities depend on different pH's [13,15]. In contrast to the multiplicity of CHX genes, KEA genes have remained relatively constant at three to seven genes from singlecelled algae to flowering plants. This observation indicates that KEA transporters play conserved roles in cellular processes specific to plants ranging from Chlamydomonas to poplar trees [3]. So far, no information is available about the role and functions of plant KEA transporters. KEA proteins constitute an ancient group of transporters with homology to bacterial transporters that harbour a C-terminal KTN domain. The presence of a C-terminal KTN domain is a feature shared by many bacterial potassium channels and transporters, like Kch, TrkA, Ybal and KefB or KefC [16]. The Arabidopsis KEA proteins can be divided into two separate groups, KEA1–3 and KEA4–6, each with distinct bacterial homologs [3]. AtKEA1–3 proteins are closely related to the bacterial glutathioneregulated K + efflux transporters KefB and KefC, that are important for bacterial survival during exposure to toxic metabolites [17]. Electrophiles react with glutathione to form glutathione adducts, thereby releasing KefC and KefB from inhibition by GSH, thus eliciting K + efflux and H + uptake. Apart from the presence of the regulatory KTN domain, these bacterial proteins are further activated by the binding of ancillary proteins KefF and KefG [18]. Binding of KefF and GSH was shown to induce conformational changes in KefC KTN dimers that affect antiporter activity [19-21]. The bacterial KefC protein was originally proposed to function as a K + channel, but recent studies indicate a K +/H + antiporter mode of action [22]. KEA4–6 proteins from Arabidopsis thaliana form a separate group relatively distant to the KEA1–3 group [3]. AtKEA4–6 proteins share high homology to cyanobacterial sequences but have lost the C-terminal KTN domain, similar to the related TMCO3 proteins from metazoa [3]. To date, none of the KEA or KefC-like genes from eukaryotes has been functionally characterized. In this paper, we provide the first functional study of a KEA protein from plants. AtKEA2 shows high expression levels in aerial parts and the N-terminus is targeted to chloroplasts. As the N-terminal domain of AtKEA2 protein was highly toxic when expressed in bacteria, we tested a truncated AtKEA2 that consisted of a conserved transmembrane Na–H domain and KTN domain similar in size to Escherichia coli KefC. We show by heterologous complementation that the short AtsKEA2 complements sensitivity to hygromycin and high K + or Na + concentrations caused by disruption of ScNHX1 in yeast. When wild-type AtsKEA2 was purified and reconstituted in lipososmes, we demonstrated electroneutral cation/H + antiport activity, whereas protein mutated in two acidic residues within transmembrane span 6 or in a putative regulatory loop close to transmembrane span 9 had no activity. These results demonstrate that AtKEA2 encodes a cation/proton exchanger with preference for K +. 2. Material and methods 2.1. Yeast strains and growth Conditions All Saccharomyces cerevisiae strains used were derivatives of W3031B (Matα leu2-13 112, ura3-1, trp1-1, his3-11 15, ade2-1, can1-100). Strains WX1 (nhx1Δ::TRP1), ANT3 (ena1-4Δ::HIS3, nha1Δ::LEU2) and AXT3 (ena1-4Δ::HIS3, nha1Δ::LEU2, nhx1Δ::TRP1) were gifts from Drs. F.J. Quintero and J.M. Pardo (IRNA-CSIC, Sevilla, Spain). A Kha1 disruption was made in strain AXT3 as described in Maresova et al. [23], giving rise to strain AXT4K (ena1-4Δ::HIS3, nha1Δ::LEU2, nhx1Δ:: TRP1, kha1Δ::KanMX6). The strain OC05 (nhx1Δ::HIS3, vnx1Δ:: KanMX6, vcx1Δ::HphNT1) was used for transport experiments as described in Cagnac et al. [24]. Cells were grown at 30 °C in YPD (1% yeast extract, 2% peptone, 2% glucose), APD (10 mM arginine, 8 mM H3PO4, 2% glucose, 2 mM MgSO4, 1 mM KCl, 0.2 mM CaCl2, trace

2363

minerals and vitamins) [25] or SD medium (2% glucose, 0,7% yeast nitrogen base without amino acids (Difco), 50 mM succinic acid–Tris pH 5,5) with the required amino acids and supplements as indicated. For induction of antiporter expression in cells carrying plasmids with antiporter genes under control of the GAL1 promoter, glucose (2%) was replaced by galactose (2%) in the growth media. 2.2. Growth test For growth assays, a preculture was made by growing the strains for 1 day in selective SD medium to an optical density (A660 nm) of approximately 1. Next, the cultures were diluted 1/6 with the same medium without glucose and grown for 18 h at 30 °C. The preculture was washed twice in sterile water, and all cell suspensions were adjusted to an optical density of 0.2, after which 5 μl of serial tenfold dilutions were spotted on YPD or APD plates supplemented with salts or hygromycin B. Yeast strains were transformed using the standard lithium acetate method [26]. 2.3. Cloning of AtKEA2 The AtKEA2 gene was originally annotated to correspond to locus At4g00630 (old gene accession: AAC13638.1). We obtained cDNA for this gene from the laboratory of Dr. J.M. Ward (University of Minnesota). The cDNA was sequenced and transferred by PCR and TOPO cloning to the pENTR /SD/D-TOPO vector (Invitrogen), using forward primer 5'-CACCATGTTCCCTCAGCAAGAGGT -3' and reverse primers 5'-GATAGCGAGTGTGCCTTCAAT‐3' (no stop codon) or 5'-TTAGATAGCGAGTGTGCCTTCAAT-3' (with stop codon). The short AtKEA2 sequence (AtsKEA2), corresponding to AAC13638.1, was transferred by LR recombination to vector pYESDEST52 (Invitrogen). The full-length AtKEA2 gene was obtained using homologous recombination in yeast as described by Fusco et al. [27]. The missing part of the gene, corresponding to locus At4g00640 (old gene accession: AAC13619.1) of the AAC13638.1 gene was amplified by PCR from Arabidopsis cDNA using forward primer 5'-AGCTATCAAACAAGTTTGTACAAAAAAGCAGGCTCCGCGGCC GCCCCCTTCATGGATTTTGCGTCTAGCG-3' and reverse primer 5'-GTAACACGACCACAGCCA-3'. The PCR product and the pYES-AtsKEA2 linearized by NotI and HindIII were used to transform the yeast strain BYa by standard LiAc transformation [26]. Yeast was grown for 2–3 days on SC medium without uracil. The correct insertion of the PCR product in the pYES-AtsKEA2 was first checked by PCR on colonies using the above primers. Plasmids from independent positive colonies were extracted and sequenced as described by Singh et al. [28]. 2.4. Determination of the expression profile for AtKEA2 in Arabidopsis Total RNA from 30‐day-old plants was isolated with TriReagent according to manufacturer's specifications (Sigma). cDNA was then synthesized from 1 μg of total RNA using the first strand cDNA synthesis kit (Roche Applied Science). RT-PCR was performed as described previously by Leterrier et al. [29]. The KEA2 gene was amplified using forward primer 5'-ATGTTCCCTCAGCAAGAGGT-3' and reverse primer 5'GTAACACGACCACAGCCA-3', which amplifies a 500‐bp fragment of cDNA or a >1500‐bp fragment of genomic DNA. As a control ACTIN 2 was amplified using forward primer 5'-CGGTATTGTGCTGGATTCTG-3' and reverse primer 5'-ATCTCCTGCTCGTAGTCAAC-3'. 2.5. Determination of the subcellular localization of AtKEA2 The first 300 bp of the AtKEA2 coding sequence was amplified by PCR using forward primer 5'-CACCATGGATTTTGCGTCTAGCG-3' and reverse primer 5'-ACTTTGACAAAGTAACCTAAACC-3' and then cloned in vector pENTR-SD-TOPO. The 300‐bp fragment was introduced via LR reaction into the binary vector pB7FWG2 fused to the eGFP coding sequence [30]. Transient expression of KEA2-eGFP in Arabidopsis seedlings was performed as described by Li et al. [31]. Five‐day-old

2364

MN. Aranda-Sicilia et al. / Biochimica et Biophysica Acta 1818 (2012) 2362–2371

Arabidopsis seedlings were soaked in 20 ml cocultivation medium containing 0.25 × MS, 1% sucrose, 100 μM acetosyringone, 0.005% Silwet L-77 and GV3101 A. tumefaciens cells transformed with the pB7FWG2 plasmid. Cocultivation was carried out in darkness at 22 °C for 40 h. Prior to fluorescence microscopy seedlings were washed in 20 ml of medium containing 0.25× MS with 1% sucrose. 2.6. Site‐directed mutagenesis The single point mutants AtsKEA2-D721A and AtsKEA2-E835K were generated using the Quick Change® II site-directed mutagenesis kit (Stratagene) according to modifications described by Wang and Malcolm [32]. The complementary primers used to create AtsKEA2-D721A were as follows: AtsKEA2-D721A-For (5'-CGGTCTTGCTTTTTCAAGCTTTGGCTGTGGTCGTGT TAC-3'), AtsKEA2-D721A-Rev (5'-GTAACACGACCACAGCCAAAGCTTGAA AAAGCAAGACCG-3'). The complementary primers used to create AtsKEA2-E835K were as follows: AtsKEA2-E835K-For (5'-CTTGCAGAGACTGAATTCTCCTTG CAAGTTAAATCAGATATCGCTCC-3'), AtsKEA2-E835K-Rev (5'-GGAGCGATA TCTGATTTAACTTGCAAGGAGAATTCAGTCTCTGC-3'). Underlined letters indicate the changes in nucleotide sequence, and bold letters the introduced silent restriction sites HindIII (D721A) or EcoRI (E835K). 2.7. Purification of AtsKEA2 protein by Ni 2 + affinity chromatography The AtsKEA2 protein was purified from yeast strain AXT3 expressing the gene, essentially as described for the AtNHX1 protein [33] with some modifications. Cells were grown in 5 L of APG medium to an A660 nm of 1.0. Microsomal membranes were isolated as described [33] after which the microsomal membrane fraction (5 ml, 10 mg protein/ml in 100 mM Tris–HCl, pH 7.5, 20% glycerol, 0.1 mM DTT and 0.1 mM EDTA) was mixed with 45 ml solubilization buffer (50 mM KH2PO4, 500 mM NaCl, 10 mM imidazole–PO4, 20% glycerol, 0.5% n-dodecyl-β-D-maltoside, pH 6.5) supplemented with 200 μl of protease inhibitor cocktail (Sigma) and incubated for 30 min at 4 °C under gentle shaking. Unsolubilized material was removed by centrifugation for 30 min at 37,000 rpm (Beckman 70Ti rotor). The supernatant was mixed with 1 ml of Ni-NTA resin (Qiagen) and incubated overnight at 4 °C. The resin was then poured into a polypropylene column and washed with 16 and 4 ml of solubilization buffer with 10% glycerol and 0.075% n-dodecyl-β-D-maltoside containing 20 and 50 mM of imidazole, respectively. Finally, bound protein was eluted slowly in 8 ×175 μl of 50 mM BTP-Mes pH 7.5, 200 mM imidazole (pH 7.4) and 20% glycerol supplemented with 2 μg/ml pepstatin, 0.2 mM PMSF, 0.1 mM DTT and 0.1 mM EDTA. The purified protein that normally eluted in fractions 3, 4 and 5 was frozen in liquid nitrogen and stored at −80 °C, until use for subsequent experiments. 2.8. Reconstitution of AtsKEA2 proteins and measurement of cation/H + exchange in vitro Reconstitution was performed by elimination of detergent using Sephadex G-50 (fine; Amersham Biosciences, Inc.) spin columns and Biobeads (SM-2; Bio-Rad) as previously described [33] using 5 μg of protein and in the presence of ammonium sulfate. In order to encapsulate pyranine during the reconstitution, it was essential to preload the G-50 spin column with 200 μl of 2.5 mM pyranine (BTP salt) in reconstitution buffer, as the pyranine present in the liposome-protein-detergent mixture was eliminated >99% in the top of the column. This shows that the reconstitution takes place along the passage of the sample through the column, allowing also an efficient elimination of imidazole or traces of KCl and NaCl in the protein sample. An inside acid pH gradient was created by diluting the proteoliposomes 50 fold in (NH4)2SO4 free medium. Cation/H+ exchange activity was monitored by measuring the rise in fluorescence of encapsulated pyranine in the presence of various monovalent cations as previously described [33] To calibrate pyranine

fluorescence with pH, proteoliposomes were diluted in the presence of (NH4)2SO4 to avoid creation of a pH gradient and 1 μM gramicidin, after which the medium was adjusted to different pH with NaOH or H2SO4. A plot of pH vs. log((Rmax − R)/(R − Rmin)) yielded a straight line with slope of 1 between pH 5 and 8.5, and pH could be estimated from the ratio of pyranine fluorescence according to equation pH =pKa´+ log((Rmax − R)/(R − Rmin)), in which R is the ratio of pyranine fluorescence excited at 467/404 nm, Rmin the minimal ratio observed at acidic pH and Rmax the maximal ratio observed at alkaline pH [34]. 2.9. Isolation of intact vacuoles from yeast and transport assays Intact vacuoles were isolated as described by Cagnac et al. [35]. In vitro transport assays using intact vacuoles were performed as described by Cagnac et al. [24]. 2.10. Gel electrophoresis and immunoblotting Membrane fractions or purified proteins were separated by SDSPAGE on 7.5% or 10% acrylamide using the system of Laemmli [36]. Immunoblotting was performed as described [37]. After protein electrotransfer to a polyvinylidene difluoride membrane (Pall Gelman), the blot was incubated with a monoclonal antibody (1:5000) raised against the V5 epitope (AbCam SV5-Pk1). Anti-mouse IgG-alkaline phosphatase conjugate (Sigma) was used at a dilution of 1:10,000 to reveal the antigen-antibody reaction. Total proteins from yeast were extracted as described in Cagnac et al. [35]. 2.11. Fluorescence microscopy A Nikon C-1 confocal laser scanning microscope was used to visualize plant cells. GFP-tagged protein was detected using a standard argon 488-nm laser excitation and a BA515/30 filter for emission. Autofluorescence of chlorophyll was observed with a BA605/75 filter for red emission. All confocal images were taken at 40 × magnification. A Nikon Eclipse 50i fluorescent microscope was used to visualize yeast cells. GFP and eYFP fused proteins were visualized by using B-2A filters (excitation filter: 450–490 nm; dichromatic mirror cut-on wavelength: 500 nm; barrier filter wavelength: 515 nm longpass). All images were taken at 100× magnification, with a Nikon DS-Fi1 CCD camera. 2.12. Protein determination Protein was determined by the method of Bradford [38] with the Bio-Rad protein assay reagent and bovine serum albumin as standard. 3. Results 3.1. Sequence analysis The AtKEA2 protein shares 77% identity and 88% similarity to the AtKEA1 protein suggesting they are a result of a recent gene duplication and could have redundant functions. More variation is observed in the N-terminal domain, indicating it might perform isoform-specific regulatory roles (Fig. 1). For both proteins, 10 transmembrane helices are predicted in the complete sequence, while 13 transmembrane helices are predicted in the E. coli KefB and KefC sequences (Fig. 1) [39]. Aligning KefC with AtKEA1 and AtKEA2 proteins indicates a most probable number of 12 transmembrane helices, in accordance with the number determined from the crystal structure of the E. coli NhaA protein (Fig. 1) [40]. A short negatively charged cytoplasmic regulatory loop at positions 258–266 in KefC (sequence HALESDIEP) is also conserved in plant sequences. This region is critical for the sensitivity of KefC to GSH and GSH adducts (GSX) [41] and is proposed to interact with the KTN domain [19,20]. The N-terminal region of the KTN domain, including the typical glycine motif of the Rossman fold involved in NAD+

MN. Aranda-Sicilia et al. / Biochimica et Biophysica Acta 1818 (2012) 2362–2371

binding, also shows a high degree of homology between the plant and bacterial sequences, while the C-terminal part of the domain that contains most residues involved in GSH binding, notably starting from the α6 helix that is exchanged between subunits in the bacterial proteins is less conserved [19]. 3.2. Cloning short and full-length AtKEA2 Recently, AtKEA2 is predicted to encode a protein of 1174 or 1185 amino acids (gene accessions AT4G00630.1 and AT4G00630.2) that includes a soluble N-terminal domain of 571 amino acids, a Na–H domain of 376 amino acids, and a C-terminal KTN domain of 116 amino acids. For many years, the N-terminal domain was thought to be a separate gene and thus missed in initial gene annotation. The two genes corresponding to the N-terminus and the conserved Na–H domain are now combined in locus At4g00630 to give the full-length AtKEA2 sequence. A short version of the AtKEA2 cDNA sequence (AtsKEA2) including the conserved Na-H domain was amplified from Arabidopsis cDNA and cloned by in vivo cloning in yeast (J.M. Ward, personal communication). We have detected the long AtKEA2 cDNA (Fig. S1) by PCR of Arabidopsis cDNA; however, attempts to directly clone the full-length cDNA or the DNA sequence coding the soluble N-terminal domain alone (part 1) in a bacterial cloning vector were unsuccessful. We therefore cloned the full-length AtKEA2 gene by in vivo cloning in yeast strain BYa, fusing the N-terminal domain (part1) DNA to the short AtKEA2 (AtsKEA2) sequence. Plasmids were

2365

isolated from various yeast colonies and then inserted into bacteria to facilitate their amplification for further applications. Surprisingly, after 2 days incubation only a limited number of ampicillin-resistant colonies were obtained. The only clones recovered consistently contained large deletions or insertions of bacterial DNA in the sequence encoding the soluble N-terminal region. We encountered similar problems in our attempts to clone the N-terminal domain (part1) DNA region, while we could successfully re-amplify pYES and pYESAtsKEA2 using the same procedures (Fig. S2). As amplification of AtKEA2 in bacteria was not possible in our conditions, plasmids were isolated from independent yeast colonies and sequenced. The cDNA sequence corresponded to gene accession number At4G00630.1, which is slightly shorter than At4G00630.2. Neither AtKEA2 nor the sequence coding for the N-terminal domain (part 1) could be amplified from the Lacroute [42] or Invitrogen A. thaliana cDNA library (Fig. S1). 3.3. Expression and subcellular localization of AtKEA2 in Arabidopsis We determined the presence of AtKEA2 transcripts by semiquantitative PCR in different organs. AtKEA2 is abundant in leaves, stems and flowers (Fig. 2A). AtKEA2 was not detected in roots under our RT-PCR conditions (27 cycles), whereas a faint band was observed after 40 cycles of amplification (data not shown). The N-terminal regions of AtKEA2 and AtKEA1 are predicted by ChloroP to contain respectively a 57 and 49 amino acid chloroplast transit

AtKEA2 AtKEA1 EcKefC

1 M D F A S S V Q R Q S M F H G G A D F A S Y C L P N R M I S A K L C P K G L G G T R F WD P M I D S K V R S A I R S K R N V S Y R S S L T L - N A D F N G R F Y - G H L L P A K P - - - - - - - - Q N V P L G F R L L C Q S S D S V G D L V G N M E YA S T F QR P I L F H GGD G - A S YC F P N R L I S - - - - P K G I S I T - - - - S - GD S K V H S C F R L R R N V A QS G T L N LMN A C F S GR F YS GH L H S T K S I L GN GH QA K R I P F G F R L R C QGH ES L GN A D S N - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 111 D R N L E F A E G S D D R E V T F S K E E K D T R E Q D S A P S L E E L R D L L N K A T K E L E V A S L N S T M F E E K A Q R I S E V A I A L K D E A A S A WN D V N Q T L N V V Q E A V D E E S V A K E A V Q K A T M A L S L A E A R L Q V A D H R I - - G E S S E S S D E T E A T D L K D A R V E N D T D S L E E L K E L L H K A I K E L E V A R L N S TM F E E K A QR I S E R A I A L K D E A A T AWL D V N K T L D V I R D T V Y E E A L A K E A V Q T A TM A L S L A E A R L QV I - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 231 L ES L EA - - - EG YN T - S E ES EV R D GV K D K E EA L L S A K A D I K EC Q EN L A S C E EQ L R R L QV K K D E L QK EV D R L N EA A ER A Q I S A L K A E ED V A N I MV L A EQA V A F E L EA T QR V N D A E I A L QR A E V ES L EA GA GN D I P H V S E E T E E T I D V N D K E EA L L A A K D D I K EC QV N L D N C ES Q L S A L L S K K D E L QK EV D K L N E F A E T I Q I S S L K A E ED V T N I MK L A EQA V A F E L EA T QR V N D A E I A L QR A E - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 347 K T L F GS Q T Q E T T QGK V L D GK N T I V G ED EV L S - E I V D V S H QA ER - - - - - - D L V V V GV S S - - - - D - V - - - G - - - T QS Y ES - D N EN GK P T A D F A K EA EG EA EK S K N V V L T K K Q EV QK D L P R ES K S L S I S Q T P E E T QGQ - L S D E E T S Q E D AM V L S GN V E D V T H QV E K E S P K D GD L P V V Q I T A E L V P D I V GQR N K K L T QP Y E S S D H E N GK P S V E S S K V V E A D S E K P K I N V Q T K K Q E T QK D L P K E G - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

AtKEA2 AtKEA1 EcKefC

M556 448 S S H N G T K T S L K K S S R F F P A S F F S S N G D G T A T V F E S L V E S A K Q QWP K L I L G F T L L G A G V A I Y S N G V G R N N Q L P Q Q P N I V S T S A E D V S S S T K P L I R QM Q K L P K R I K K L L EM F P Q Q E V N E E E A S S L N S P K A S F N K S S R F F S A S F F S S N P D G T A T V F G S L V G S V K Q QWP K L V L G L A L L G A G L T L Y S N G V G G N N Q L L Q Q P D V T S T S T E D V S S S T K P L I R Q V Q K L P K R I K K L L EM I P H Q E V N E E E A - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - MDS H

AtKEA2 AtKEA1 EcKefC AtKEA2 AtKEA1 EcKefC AtKEA2 AtKEA1 EcKefC

AtKEA2 AtKEA1 EcKefC

AtKEA2 AtKEA1 EcKefC

Na_H exchanger Domain TM1 TM2 568 S L L D V LWL L L A S V I F V P L F QK I P GGS P V L G Y L A A G I L I GP YG L S I I R N V H G T K A I A E F GV V F L L F N S L F D F LWL L L A S V I F V P L F QK I P GGS P V L G Y L A A G I L I GP YG L S I I R N V H G T R A I A E F GV V F L L F N T L I Q A L I Y L G S A A L I V P I A V R L G L G S - V L G Y L I A G C I I G P WG L R L V T D A E S I L H F A E I G V V L M L F I 5 TM5 TM6 D721 Na_H exchanger Domain 688

TM8

E835

TM8

*

Na_H exchanger Domain

TM9

TM10

TM11

E262 regulatory loop Rossman fold X

TM12

** *

X

KTN domain X

*

S S L L F L V V G I SMA L T PWL A A GGQ L I A S R F E L QD V R S L L P V ES E T D D L QGH I I I C G F GR I GQ I I A Q L L S ER L I P F V A L D V S S D R V A I GR S L D L P V Y F GD A GS R EV L H K I GA D R A C A A A I A L S S L L F L V V G I SMA I T PWL A A GGQ L I A S R F E L H D V R S L L P V ES E T D D L QGH I I I C G F GR V GQ I I A Q L L S ER L I P F V A L D V S S D R V T I GR S L D L P V Y F GD A GS K EV L H K I GA GR A C A A V V A L A K S L T L A V A L S M A A T P I L L V - - - - I L N R L E Q S S T E E A R - E A D E I D E E Q P R V I I A G F G R F G Q I T G R L L L S S G V K M V V L D H D P D H I E T L R K F GM K V F Y G D A T R M D L L E S A G A A K A E V L I N A I 357 1048

AtKEA2 AtKEA1 EcKefC

TM7

A G L S M A L G A F L A G L L L A E T E F S L Q V E S D I A P Y R G L L L G L F F M T V GM S I D P K L L L A N F P L I M G T L G L L L V G K T I L V V I I G K L F G I S I I S A V R V G L L L A P G G E F A F V A F G E A V N Q G I M T P Q L A G L S M A L G A F L A G L L L A E T E F S L Q V E S D I A P Y R G L L L G L F F M T V GM S I D P K L L L S N F P V I V G T L G L L I V G K T M L V V I M G K L F G I S I I S A I R V G L L L A P G G E F A F V A F G E A V N Q G I M S P Q L V G L S M A M G A F L A G V L L A S S E Y R H A L E S D I E P F K G L L L G L F F I G V GM S I D F G T L L E N P L R I V I L L L G F L I I K I A M L W L I A R P L Q V P N K Q R R W F A V L L G Q G S E F A F V V F G A A QM A N V L E P EW

928

AtKEA2 AtKEA1 EcKefC

TM5

*

237

AtKEA2 AtKEA1 EcKefC

TM4

A L S S T A V V L QV L Q ER G ES T S R H GR A T F S V L L F QD L A V V V L L I L I P L I S P N S S K GG I G F QA I A EA L G L A A I K A A V A I T G I I A GGR L L L R P I YK Q I A EN R N A E I F S A N T L L V I L G T S L L T A R A L S S T A V V L QV L Q ER G ES T S R H GR A S F S V L L F QD L A V V V L L I L I P L I S P N S S K GG I G F QA I A EA L G L A A V K A A V A I T A I I A GGR L L L R P I YK Q I A EN R N A E I F S A N T L L V I L G T S L L T A R A L S S T A I A M Q A M N E R N L M V T QM G R S A F A V L L F Q D I A A I P L V A M I P L L A T S S A S T T M G - - - - - - A F A L S A L K V A G A L V L V V L L G R Y V T R P A L R F V A R S G L R E V F S A V A L F L V F G F G L L L E E 123 808

AtKEA2 AtKEA1 EcKefC

TM3

I G L E L S V ER L S SMK K YV F G L GS A QV L V T A A V I G L I T H YV A GQA GP A A I V I GN G L I G L E L S V ER L S SMK K YV F G L GS A QV L V T A A V V G L L A H YV A GQA GP A A I V I GN G L I G L E L D P Q R L WK L R A A V F G C G A L QM V I C G G L L G L F C M L L G - L R WQ V A E L I GM T L

KTN domain

X XX

X

X

D T P G A N Y R C V WA L S K Y F P N V K T F V R A H D V D H G L N L E K A G A T A V V P E T L E P S L Q L A A A V L A Q A K L P T S E I A T T I N E F R S R H L S E L A E L C E A S G S S L G - Y G F S R S T S K P K P P S P S E T S D D N Q D A P G A N Y R C V WA L S K F Y P N V K T F V R A H D V V H G L N L E K A G A T A V V P E T L E P S L Q L A A A V L A Q A K L P T S E I A N T I N E F R T R H L S E L T E L C E A S G S S L G - Y G Y S R - T S K P K P Q - P S D A S G D N Q D D P Q T N L Q L T EM V K E H F P H L Q I I A R A R D V D H Y I R L R Q A G V E K P E R E T F E G A L K T G R L A L E S L G L G P Y E A R E R A D V F R R F N I QM V E EM A M V E N D T K A R A A V Y K R T S A M L S E I I T E D R E H L S 472 1167 I I EG - T L A I - - - - - - - - - - - - - - - - - - - I I EGG T V V I - - - - - - - - - - - - - - - - - - - L I Q R H GWQ G T E E G K H T G N M A D E P E T K P S S 592

Fig. 1. Alignment of AtKEA 1, AtKEA2 and EcKefC. AtKEA1 (AT1G01790.1), AtKEA2 (AT4G00630.2) and E. coli KefC (AAC73158) were aligned with ClustalW. The N-terminal peptide of 100 residues used to show chloroplast localization is indicated by a gray line above the sequence. The predicted transit peptides for AtKEA1 and AtKEA2 are indicated by the dented black line. Residues D721 (cation binding) and E835 (regulation) corresponding to D166 and E278 of the AtsKEA2, or D156 and E262 of E. coli KefC sequence are indicated. The AtsKEA2 starts at M556 and ends at I1174. The pfam00999 Na–H domain and KTN domains are labeled with a black line. The Rossman fold glycine motive is indicated by *. Residues that were shown to be important for glutathione binding in the KefC KTN structure are indicated by X. Predicted transmembrane helices in AtKEA1,2 and EcKefC are marked by gray boxes.

2366

MN. Aranda-Sicilia et al. / Biochimica et Biophysica Acta 1818 (2012) 2362–2371

peptide (score 0.497 for AtKEA2, 0.501 for AtKEA1, http://www.cbs. dtu.dk/services/ChloroP/; Fig. 1) [43]. As the full-length AtKEA2 protein could not be cloned in bacteria, due to toxicity, it was not possible to generate an AtKEA2-GFP fusion protein for subcellular localization in plants. Therefore, we generated a fusion of the first 300 base pairs of the N-terminus. In this fragment, the transit peptide is predicted with the same probability and length as in the full‐length protein. In Arabidopsis seedlings transformed with this construct, GFP fluorescence is detected in cotyledons of distinct compartments that colocalized with red fluorescence characteristic of chlorophyll (Fig. 2B). The results suggest that the full length protein AtKEA2 is most likely localized to chloroplasts. Our results would complement several proteomic studies, which detected AtKEA2 in the chloroplast inner envelope fraction [44–46]. 3.4. Complementation in yeast We first tested whether AtKEA2 behaved similarly to AtCHX17 or CHX20, as CHXs and KEAs are both classified within the CPA2 family. It was shown before that disruption of KHA1 in a yeast strain with disruptions in transport systems ena1–4, nha1 and nhx1 causes sensitivity to alkaline pH in low K + media [23] and that Arabidopsis CHX17 and CHX20 can complement this sensitivity [9,14,15]. We expressed the full-length AtKEA2 protein, as well as the short version lacking the N-terminal soluble domain in yeast strain AXT4K, harbouring disruptions in genes coding for Ena1-4p, Nha1p, Nhx1p and Kha1p transporters. None of the AtKEA2 proteins could fully complement the KHA1 disruption, although expression of AtsKEA2 marginally improved growth at 5 mM KCl and pH 7.5 compared to the kha1Δ disrupted strain (Fig. 3). Next, we checked whether the AtKEA2 protein could complement the disruption of NHX1, the only CPA1 protein in yeast. The most characteristic feature of yeast Nhx1p is that it confers strong resistance to hygromycin B, a phenotype that is related to the role of Nhx1p in endosomal protein sorting and prevacuolar pH regulation [8,47]. Expression of plant NHX isoforms complements this sensitivity. Furthermore, NHX1 disruption causes NaCl and KCl sensitivity, related to vacuolar or prevacuolar K +/Na + accumulation. Surprisingly, expression of AtsKEA2, but not the full-length protein complemented the NaCl, KCl and hygromycin sensitivity of the NHX1 disruption (Fig. 4). To ensure that lack of complementation by the full-length protein was not caused by an expression defect, we tagged AtKEA2 at its C-terminus with a V5–6 epitope and a 6× histidine tag (V5-

H6). Both short and full-length AtKEA2 proteins were detected by immunoblotting (Fig. 5), indicating the proteins were expressed. 3.5. Transport activity of reconstituted AtsKEA2 A V5-H6 tag was inserted at the C-terminus of AtsKEA2 to enable its detection and purification. The tagged AtsKEA2 sequence complemented the hygromycin sensitivity of a yeast strain with disruption in the NHX1 gene, indicating the fusion protein was functionally active (Figs. 4 and 6). Although the AtsKEA2-V5-H6 polypeptide could be detected in yeast vacuolar preparations by immuno-staining (Fig. 5), we could not measure its activity in isolated vacuoles from strains expressing the tagged or untagged protein (Fig. S3), indicating that most of the protein is not present at the main vacuole but perhaps at prevacuolar compartment, similar to yeast Nhx1p [24]. As an alternative approach, we purified the short AtKEA2 protein by Ni2 + affinity purification from yeast microsomes. The tagged polypeptide was readily detected using a monoclonal antibody raised against the V5 epitope in microsomes of cells grown on galactose, expressing AtsKEA2, but not in control microsomes from cells grown on glucose (data not shown). The protein was purified as previously described [33]. However, due to relatively low levels of expression, as well as low affinity for the Ni2 + column of the His-tagged AtsKEA2 protein, the washing to remove contaminating proteins from the Ni-NTA column could not be very stringent, resulting in a partial purification of the protein in the final eluted fractions (Fig. 7A). Nevertheless, a major band between 50 and 75 kDa markers was observed in the purified fractions that crossreacted with the monoclonal antibody (Fig. 7B). The purified protein was reconstituted in liposomes containing the pH indicator pyranine. After creating an inside acid pH gradient, the effect of monovalent cations to dissipate the pH gradient was assayed (Fig. 8A). Addition of K + or Cs + caused internal alkalinization. Less alkalinization was observed in the presence of Na + or Li +. As compared to results previously obtained with NHX proteins, the exchange reaction was rather inefficient, probably due to low protein yield of the purification [33,48]. These results suggested AtsKEA2 transported K + and Cs + preferentially relative to Na + or Li +. 3.6. Mutagenesis of D 721 and E 835 inactivated AtsKEA2 To verify whether the exchange activity was specifically due to AtsKEA2 protein, two point mutations were introduced in the short

Fig. 2. AtKEA2 is expressed in aerial parts and localizes to chloroplasts in cotyledons. (A) Expression of AtKEA2 determined by semi-quantitative RT-PCR. Total RNA was isolated from organs of 30 d plants and reverse-transcribed. AtKEA2 was amplified with 27 cycles. Actin 2 gene was amplified as a loading control. (B) Localization of AtKEA2 to chloroplasts of Arabidopsis cotyledons. The first 300 bp of AtKEA2 fused to eGFP was transiently expressed in seedlings. (a) GFP fluorescence from a cotyledon; (b) chlorophyll autofluorescence of the same field as (a); and (c) merged image of (a) and (b). Scale bar of 30 μm.

MN. Aranda-Sicilia et al. / Biochimica et Biophysica Acta 1818 (2012) 2362–2371

A

4.5

5.7

6.5

7.0

2367

7.5 kha1+sKEA2

pH

kha1+KEA2 kha1 + vector Control (

B

0

5

1

)

50

25

KCl ( mM)

kha1+sKEA2 kha1+KEA2 kha1+vector Control (

)

Fig. 3. AtKEA2 does not restore growth at alkaline pH and low K+ in KHA1 disrupted strains. AtKEA2 and AtsKEA2 were introduced in yeast shuttle vector pYES-DEST52 as described under “Experimental Procedures”. Each construct was introduced in mutant strain AXT4K (ena1-4Δ, nha1Δ, nhx1Δ, kha1Δ). Each strain was grown and serial dilutions were made as described. Five microliters of each dilution were spotted onto APG medium at different pH without additional K+ (A) or APG medium at pH 7.5 with the indicated KCl concentrations (B). As a positive control, strain AXT3 (ena1-4Δ, nha1Δ, nhx1Δ, ΚHA1) was used. Plates were incubated at 30 °C for 2–5 days. Only strain AXT3 harbouring the endogenous KHA1 gene is able to grow at alkaline pH and low K+ concentrations.

AtKEA2 sequence. First, we mutated an aspartate in transmembrane span 6, corresponding to D 721 in the full-length AtKEA2 sequence (Fig. 1), to alanine. Alignments showed that this residue corresponds to D 164 in E. coli NhaA sequence, which was shown to be involved in cation binding and transport [49]. Mutation of this conserved Asp residue thus would most likely inactivate the protein. Secondly, we mutated a glutamate corresponding to E 835 in the full-length sequence (Fig. 1) close to the predicted TM helix 9. This glutamate is homologous to E. coli KefC E 262. In KefC E 262 is located in a short regulatory loop, partially conserved in the AtKEA2 sequence (Fig. 1). Mutation of this residue in KefC results in a constitutively activated protein independent of regulation by glutathione [41]. The possibility thus existed that mutation would yield a more active AtKEA2 protein. Expression of either AtsKEA2-D721A or AtsKEA2-E835K in nhx1Δ strains failed to rescue the hygromycin sensitivity, indicating that both proteins were inactive (Fig. 6). The proteins could be expressed and purified with similar efficiency as the wild-type protein (Fig. 7C, D). After reconstitution in liposomes, antiport activity was however 0

50

3.7. Transport mode of AtKEA2 The E. coli KefC protein was suggested to have a channel-like transport activity [17]. However, recent results indicated that the KefC functions as an electroneutral cation/H + antiporter [22]. In the case of channel‐like activity, vesicular alkalinization in response to addition of K + is the result of the electrical coupling of K + and H + fluxes, with the rate of proton transport limited by the proton permeability of the membrane. In that case, addition of a protonophore is expected to enhance the rate of K +/H + exchange. We therefore tested the antiport activity in the presence of the protonophore FCCP. We only observed very small variations in alkalinization even at very high FCCP concentrations up to 10 μM FCCP, indicating that the AtKEA2 protein catalyzes electroneutral K +/H + exchange (Fig. 8D). 100 control (NHX1)

NaCl (mM)

A

not detected (Fig. 8), supporting the idea that both D 721 and E 835 residues were essential to transport.

nhx1 nhx1+KEA2 nhx1+sKEA2 0

800

900

1000 control (NHX1)

KCl (mM)

B

nhx1 nhx1+KEA2 nhx1+sKEA2 0

Hygromycin( g/ml)

C

10

20

30 control (NHX1) nhx1 nhx1+sKEA2 nhx1+KEA2

Fig. 4. AtsKEA2 but not full-length AtKEA2 complements NHX1 disruption. AtKEA2 and AtsKEA2 in vector pYES-DEST52 were introduced in yeast strain AXT3 (ena1-4Δ, nha1Δ, nhx1Δ), as described for Fig. 3, and spotted on YPG medium supplemented with the indicated NaCl (A), KCl (B) or hygromycin (C) concentrations. Yeast strain ANT3 (ena1-4Δ, nha1Δ, NHX1) was used as a positive control. AtsKEA2 but not full-length AtKEA2 complements sensitivity to high NaCl, KCl or hygromycin concentrations of the nhx1Δ strain.

150

sta nd ar d Co ntr ol (-) s-A tK EA 2

(+ )

MW

kDa 250 150 100 75

M kDa 250

co ntr ol

tK E

B

s-A

A2 At KE

Co ntr ol

A

A2

MN. Aranda-Sicilia et al. / Biochimica et Biophysica Acta 1818 (2012) 2362–2371

(-)

2368

1

M

4

3

100 75

50 37

50 37

25 20

25 Fig. 5. Immunodetection of V5-tagged AtKEA2 protein. (A) AtKEA2 in yeast extract. Total proteins were isolated from the strain BYa nhx1Δ transformed with empty plasmid as negative control (1), pYES-AtKEA2-V5-6xHis (2) or pYES-AtsKEA2-V5-6xHis (3) and separated by SDS-Gel electrophoresis on a 7.5% acrylamide gel. (B) AtKEA2 in yeast vacuolar membrane. Vacuoles were isolated from strain OC05 (vnx1Δ, vcx1Δ, nhx1Δ) [24] transformed with integrative vector pRS306-Gal1-AtsKEA2-V5-6xHis grown on glucose as negative control (3) or galactose (4) as described in experimental procedures, and proteins were separated by SDS-gel electrophoresis on a 10% acrylamide gel. As a positive control, AtsKEA2 protein purified as described in experimental procedures was run in lane 1. After electro-transfer, protein was detected with a monoclonal antibody against the V5 epitope.

4. Discussion 4.1. AtKEA2 protein behaves like a cation/proton antiporter A gene family encoding homologs of E. coli KefC is conserved in plants from unicellular chlorophytes to embryophytes; however, the biological functions of plant KEA homologs are not known. In this paper we provide the first in vitro evidence that a plant KEA protein, AtKEA2, mediates monovalent cation/proton exchange. It is important to point out that the functionally active AtsKEA2 we characterized is

truncated and not full-length; though it includes the entire conserved Na/H antiport domain found in all CPA proteins as shown by multiple alignment with bacterial KefC-like homologs and other eukaryote CPA2 homologs [3]. At present we do not know why the full-length protein is inactive. We suggest that the N-terminal region has a regulatory or auto-inhibitory role as seen before with Ca2 +/H+ antiporter, CAX1 [50] and Ca2 + pumping ATPase, ACA2 [51]. In those examples, the

A) 250 150 100 75

Hygromycin ( g/ml) 0 15

B)

Protein 2

3

4

5

6

2

anti-V5 blot 3

4

5

6

50 Control+vector

37

nhx1+vector 25

nhx1+sKEA2 nhx1+sKEA2 D721A

K

A

35 E8

wt

21

K 35 E8

21 D7

anti-V5 blot

150 100 75 50

KCl (mM) 800

wt

nhx1+sKEA2 E835K nhx1+sKEA2

0

D)

A

nhx1+vector

D7

C) Protein

Control+vector

37

900 Control+vector nhx1+vector

25

nhx1+sKEA2 E835K nhx1+sKEA2 nhx1+sKEA2 D721A Fig. 6. Residues D721 and E835 are essential for AtsKEA2 activity. AtsKEA2 mutants and fusion proteins in vector pYES-DEST52 were expressed in yeast strain AXT3 (ena1–4Δ, nha1Δ, nhx1Δ), and sensitivity to hygromycin and KCl was assayed as described for Figs. 3 and 4. (A) Strains expressing AtsKEA2 with a C-terminal V5-H6 epitope complement hygromycin sensitivity of nhx1Δ. (B) Strains expressing AtsKEA2 mutated at D721A and E835K failed to complement hygromycin or KCl sensitivity.

Fig. 7. Wild-type or mutated AtsKEA2 is synthesized and can be partially purified. Microsomal membranes of yeast cells expressing wild-type AtsKEA2-V5-H6 (A and B) or mutated proteins AtsKEA2-D721A and E835K (C and D) were solubilized after which total proteins were bound to Ni-NTA and washed as described under experimental procedures. In A and B, 5 μl of the eluted fractions 2–6 containing the purified proteins were subjected to SDS-PAGE on a 10% acrylamide gel (lanes 2–6). The gel was stained with SYPRO-Ruby protein gel stain, or used for western blot analysis using a monoclonal antibody against the V5 epitope. In C and D, only the eluted fraction used for subsequent reconstitution experiments is shown.

MN. Aranda-Sicilia et al. / Biochimica et Biophysica Acta 1818 (2012) 2362–2371

A)

B)

s-AtKEA2

0.6

2369

s-AtKEA2 D835K

0.6

0.5

0.5 K, Cs Li Na

0.4

pH 0.3

0.4 0.3

TMA 0.2

0.2

0.1

0.1

0

0

200

400

600

800

1000

0

0

200

400

600

800

1000

Time (s)

C)

D)

s-AtKEA2 D721A

0.6

0.6

0.5

0.5

0.4

0.4

pH 0.3

0.3

0.2

0.2

0.1

0.1

0

0

200

400

600

800

1000

0

FCCP 5 M 10 M 0 1 M

0

200

400

600

800

1000

Time (s) Fig. 8. pH gradient-dependent cation transport in reconstituted AtsKEA2V5-H6 but not in mutated protein. AtsKEA2-V5-H6 and mutant proteins were purified and reconstituted in liposomes containing pyranine as described in experimental procedures. An acid inside pH gradient was created in proteoliposomes by ammonium sulfate dilution as described in experimental procedures. Sulfate salts of the indicated cations at 100 mM were added (arrow) to start the reaction. The fluorescence signal was calibrated with pH as described in experimental procedures. (A) Liposomes with AtsKEA2. K+ and Cs+ caused the largest fluorescence change relative to Li+ and Na+. The organic cation TMA+ had little or no effect. (B, C) Liposomes reconstituted with mutated proteins D721A or E835K. Addition of cations caused no change in rate of alkalinization. (D) Effect of FCCP on AtsKEA2 activity. K2SO4 at 50 mM was added (arrow) to liposomes containing AtsKEA2 in the presence or absence of the protonophore FCCP. Final [FCCP] was 1, 5 or 10 μM. The vesicular alkalinization is not enhanced by the addition of the protonophore, indicative of directly coupled electroneutral exchange.

transport properties of CAX1 or ACA2 were revealed from studies using the truncated and active proteins [50,51], as the N-terminal hydrophilic region acted in an autoinhibitory manner. Here we emphasize that even though AtsKEA2 lacks its N-terminal domain, the fundamental properties of the transporter is very likely retained in the short protein for the following reasons. Firstly, the short AtKEA2 includes the entire conserved Pfam domain of cation/H+ exchanger and all the transmembrane spans predicted for members of the KEA family. Secondly, homologs in E. coli, KefB or KefC, are active K+ efflux transporters that do not have a long N-terminal domain. Our results of AtsKEA2 expressed in yeast or reconstituted in liposomes would support this idea. 4.2. AtsKEA2 is an electroneutral K +/H + antiporter We tested the cation specificity and mode of transport activity using the purified and reconstituted AtsKEA2 protein. We observed more alkalinization of liposomes in the presence of K + or Cs +, than by Li + or Na +, suggesting the cation transported by AtsKEA2 is K + in exchange with H +. We had previously reported that AtNHX1 showed high exchange activity in the presence of Na + or K +, followed by lower activities in the presence of Li + and Cs +, while LeNHX2 showed high exchange activity in the presence of K +, intermediate activity with Na +, and much lower activity in the presence of Li + and Cs + [33,48]. Therefore, the cation specificity of AtsKEA2 is clearly different, and indicates it functions as a K + transporter in vivo. The cation specificity of AtsKEA2 is similar to what was reported for the bacterial KefC transporter in the presence of activating protein KefF, which had relatively broad substrate specificity in the order Rb+ = K+ > Li+ > Na+ [22]. The overall activity for AtsKEA2 is relatively

low, compared to that of NHX proteins. This could be due to the absence of regulatory proteins, as in the case of the E. coli KefC protein. Most likely, however, the low activity is due to the low protein recovered due to reconstitution of partially-purified protein preparation. As a consequence, many lipid vesicles do not contain the antiporter protein, leading to an underestimation of ΔpH. Based on the final fluorescence intensity obtained after K + addition in the absence (only antiporter containing vesicles alkalinize) or presence of nigericin (all vesicles alkalinize), we estimated that approximately 10–20% of vesicles contained active antiporter protein, compared to 50–60% in the case of the NHX proteins (data not shown) [33]. Correction for this discrepancy yields similar exchange activity for AtsKEA2 or NHX proteins. We verified that the observed cation exchange activity in proteoliposomes was dependent on the AtsKEA2 protein by testing mutants. We constructed a mutant protein AtsKEA2-D721A, which removes the conserved D721 residue that is predicted to be the cation or H+ binding site based on alignments (Fig. 1) [49]. The mutated protein could be purified with similar recovery as the wild type protein, but after reconstitution we could no longer detect cation/H+ exchange. We tested the mode of K + transport mediated by AtsKEA2. Based on patch-clamp studies and rates of K + efflux, KefC transporter was previously suggested to operate as a K + channel [17]. In vitro, antiport activity by KefC was dependent on KefF, but the electrogenicity of the reaction was not studied, leaving open the possibility that in these assays H+ efflux was electrically coupled to K + influx in a channel-like mechanism [22]. Also in vivo, KefC can mediate K+ efflux in the absence of KefF, indicating KefC could have channel like activity in the absence of KefF [22]. To test whether AtsKEA2 functions as electrogenic K+ channel or electroneutral K +/H+ antiporter, we assayed alkalinization of

2370

MN. Aranda-Sicilia et al. / Biochimica et Biophysica Acta 1818 (2012) 2362–2371

liposomes in the presence of the protonophore FCCP. In our assay, K + influx could be electrically coupled to H+ efflux, depending on the relative membrane permeability to other charged species, such as SO42- and relatively impermeant buffers. Therefore, electrogenic K+ influx via a K + channel would be limited by electrogenic H+ efflux and SO42 − influx across the lipid bilayer, and determined by the relative conductance of the membrane to these ionic species. In the presence of a protonophore, H+ transport across the lipid bilayer is no longer limiting and would result in increased K+/H+ exchange. This can be illustrated artificially using liposomes where K +/H+ exchange induced by the K+ ionophore, valinomycin, is strongly increased in our experimental conditions by the presence of FCCP, while K +/H+ exchange catalyzed by the electroneutral exchanger, nigericin, is insensitive to the addition of the protonophore FCCP (Fig. S4). We did not observe significant differences in alkalinization of proteoliposome with AtsKEA2 upon addition of FCCP. Therefore, we propose that the observed alkalization upon addition of K+ is most likely due to a direct coupling of K+ to H+ flux by an electro-neutral antiport mechanism.

activity in isolated vacuoles, indicating it is localized perhaps also in prevacuolar compartments and present in the vacuolar fraction as a contaminating protein. Alternatively, kinetical properties of AtsKEA2 might not be compatible with K+ uptake in vacuoles, although electroneutral K+/H+ exchangers could also be fully reversible. In yeast, we could not determine localization of short AtKEA2 fused to GFP. Short AtKEA2 proteins with C-terminal GFP tags failed to complement the NHX1 disruption and were mislocalized to the ER membrane (Fig. S5), while no fluorescence or expression was observed for short AtKEA2 with N-terminal GFP tag. 4.5. Function of plastid AtKEA2

The relatively low activity of the AtsKEA2 protein could be due to absence of unknown regulatory compounds. Therefore we tried to construct a more active version of the AtsKEA2 protein, based on data available for the KefC protein. In KefC, E 262 is located in a short regulatory loop HALESDIEP (Fig. 1). This region was shown to be involved in regulation of gating by glutathione, via a direct interaction with the C-terminal part of the KTN domain [20,41]. Mutation E262K results in spontaneous K + leak in the absence of GSH. The regulatory loop is partially conserved in the plant AtKEA2 sequence (Fig. 1). We thus tested if mutation of the corresponding residue in AtsKEA2, E835K, would yield a more active protein. Both yeast complementation studies and in vitro activity measurements however indicated that the mutated plant protein is devoid of activity. These data indicate that plant AtsKEA2 is not regulated in the same way as the bacterial KefC transporter. Indeed, the C-terminal part of the KTN domain is poorly conserved between the plant and bacterial antiporters (Fig. 1). It is this part of the domain that was shown to be involved in glutathione binding in KefC [20]

Bioinformatic predictions and proteomic studies had hinted at a plastid localization of AtKEA1, KEA2, or KEA3: (i) AtKEA1 or AtKEA2 is predicted to have an organellar targeting sequence, specifically a chloroplast transit peptide (see http://www.cbs.dtu.dk/services/ ChloroP/); (ii) phylogenetic studies showed that genes encoding AtKEA1–3 type homologs are found in a variety of photosynthetic organisms, ranging from cyanobacteria, green plants, red algae to secondary endosymbionts [3]; and (iii) AtKEA1–3 proteins were previously detected in proteomic studies of the chloroplast inner envelope, [44–46]. The expression pattern determined in this study of AtKEA2 in aerial parts, but not in roots, as well as the targeting of an Nterminus-GFP fusion protein to chloroplasts in Arabidopsis cotyledons would confirm AtKEA2 as a genuine chloroplast protein. Chloroplast localization and electroneutral K +/H+ activity suggest that AtKEA2 and its plant homologs may be important for monovalent cation and pH homeostasis of plastids. Potassium plays an important role in chloroplast electrical balance. Light induced proton pumping into the thylakoid lumen is electrically compensated by K + efflux from the thylakoids [54]. Maintenance of K+ and H+ homeostasis in the stroma was shown to depend on proton coupled K + efflux at the inner envelope membrane [55]. Photosynthesis reactions are furthermore sensitive to the chloroplast volume changes that can be induced by light, osmotic stress or water deficit, and chloroplasts rapidly regulate osmotic potential by exchange of organic solutes and K+ [56-58]. Possibly, the AtKEA2 N-terminus can sense chloroplast volume changes through interaction of the coiled coil regions with membranes or other structural elements resulting in regulation of antiporter activity.

4.4. The catalytic domain of AtKEA2 is functionally similar to NHX

5. Conclusions

As AtKEA2 clusters with the CPA2 superfamily, it is surprising that AtsKEA2 in yeast behaves like NHX1, a member of the CPA1 family. Predicted cation transporters were classified as two separate families, CPA1 and CPA2, according to the extent of amino acid sequence homology [52]; however, the functional distinction between KEA and NHX is less clear. In yeast the CPA2 protein Kha1 is a Golgi localized protein. Disruption of the gene results in sensitivity to alkaline pH at low K+ concentrations [23]. The CPA1 protein Nhx1 however is localized in prevacuolar compartments, and disruption causes hygromycin, NaCl and KCl sensitivity [53]. Based on our findings using yeast as the heterologous host cell, AtsKEA2 resembles intracellular Nhx1 of the CPA1 family more closely than Kha1 of the CPA2 family (Figs. 3 and 4), as AtsKEA2 complements NHX1 disruption phenotypes, but not KHA1 disruption phenototypes. These results suggest that AtsKEA2 has similar transport activity compared to NHX and that yeast complementation may not depend on fundamental differences between CPA1 or CPA2 transporters alone, but rather on the subcellular location of the antiporters. We localized AtsKEA2-GFP to intracellular compartments of leaf protoplast (not shown) after transient expression, similar to the prevacuolar localization of yeast NHX1. Due to its localization in prevacuolar compartments, yeast NHX activity cannot be determined in isolated vacuoles [24,35]. Although the AtsKEA2 protein was detected by immunoblotting in isolated vacuolar fractions of yeast (Fig. 5), we could not detect AtsKEA2 antiporter

Here we have shown that KEA2 from Arabidopsis constitutes a chloroplast alkali cation/proton antiporter. A partially purified and reconstituted short version of the antiporter, lacking the N-terminal domain was shown to mediate K +/H + exchange. This first functional study of a KEA gene from plants suggests that AtKEA2 is involved in K + and pH homeostasis of chloroplasts or plastids.

4.3. Regulatory mechanisms for AtKEA2 differ from bacterial KefC transporters

Acknowledgments This work was supported by grants BIO2008-01691 from the Spanish Plan Nacional de Biotecnología to M.P.R.-R., a CSIC I3P fellowship and a Ramón Areces Foundation Postdoctoral Fellowship to. M.N.A.-S. and a CSIC JAE-Doc fellowship to O.C. Authors gratefully acknowledge Dr John M. Ward (University of Minnesota) for the gift of the short AtKEA2 cDNA, and NSF grant IBN-209792 to J.M.W. Work was partially supported by National Science Foundation Grant IBN0209788 and the Division of Chemical Sciences, Geosciences, and Biosciences, Office of Basic Energy Sciences of the U.S. Department of Energy through Grant DE-FG0207ER15883 to H.S., and a Royal Thai Government Graduate Fellowship to S.C. Authors gratefully acknowledge María Isabel Gaspar Vidal and Elena Sánchez Romero for technical assistance and the Scientific and Microscopy Services at the EEZ-CSIC for DNA sequencing and confocal imaging.

MN. Aranda-Sicilia et al. / Biochimica et Biophysica Acta 1818 (2012) 2362–2371

Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.bbamem.2012.04.011.

[30] [31]

References [32] [1] A.B. Chang, R. Lin, W. Keith Studley, C.V. Tran, M.H. Saier Jr., Phylogeny as a guide to structure and function of membrane transport proteins, Mol. Membr. Biol. 21 (2004) 171–181. [2] T. Nagata, S. Iizumi, K. Satoh, S. Kikuchi, Comparative molecular biological analysis of membrane transport genes in organisms, Plant Mol. Biol. 66 (2008) 565–585. [3] S. Chanroj, G. Wang, K. Venema, M.W. Zhang, C.F. Delwiche, H. Sze, Conserved and Diversified Gene Families of Monovalent Cation/H+ Antiporters from Algae to Flowering Plants, Front. Plant Sci. 3 (2012) 25. [4] M.P. Rodriguez-Rosales, X. Jiang, F.J. Galvez, M.N. Aranda, B. Cubero, K. Venema, Overexpression of the tomato K+/H+ antiporter LeNHX2 confers salt tolerance by improving potassium compartmentalization, New Phytol. 179 (2008) 366–377. [5] E.O. Leidi, V. Barragan, L. Rubio, A. El-Hamdaoui, M.T. Ruiz, B. Cubero, J.A. Fernandez, R.A. Bressan, P.M. Hasegawa, F.J. Quintero, J.M. Pardo, The AtNHX1 exchanger mediates potassium compartmentation in vacuoles of transgenic tomato, Plant J. 61 (2010) 495–506. [6] E. Bassil, M.-A. Ohto, T. Esumi, H. Tajima, Z. Zhu, O. Cagnac, M. Belmonte, Z. Peleg, T. Yamaguchi, E. Blumwald, The Arabidopsis intracellular Na +/H+ antiporters NHX5 and NHX6 are endosome associated and necessary for plant growth and development, Plant Cell 23 (2011) 224–239. [7] F.J. Quintero, J. Martinez-Atienza, I. Villalta, X. Jiang, W.-Y. Kim, Z. Ali, H. Fujii, I. Mendoza, D.-J. Yun, J.-K. Zhu, J.M. Pardo, Activation of the plasma membrane Na/H antiporter salt-overly-sensitive 1 (SOS1) by phosphorylation of an autoinhibitory C-terminal domain, Proc. Natl. Acad. Sci. 108 (2011) 2611–2616. [8] C.L. Brett, M. Donowitz, R. Rao, Evolutionary origins of eukaryotic sodium/proton exchangers, Am. J. Physiol. Cell Physiol. 288 (2005) C223–C239. [9] S. Padmanaban, S. Chanroj, J.M. Kwak, X. Li, J.M. Ward, H. Sze, Participation of endomembrane cation/H + exchanger AtCHX20 in osmoregulation of guard cells, Plant Physiol. 144 (2007) 82–93. [10] Y. Lu, S. Chanroj, L. Zulkifli, M.A. Johnson, N. Uozumi, A. Cheung, H. Sze, Pollen tubes lacking a pair of K+ transporters fail to target ovules in Arabidopsis, Plant Cell 23 (2011) 81–93. [11] F. Cellier, G. Conejero, L. Ricaud, D.T. Luu, M. Lepetit, F. Gosti, F. Casse, Characterization of AtCHX17, a member of the cation/H + exchangers, CHX family, from Arabidopsis thaliana suggests a role in K + homeostasis, Plant J. 39 (2004) 834–846. [12] D. Hall, A.R. Evans, H.J. Newbury, J. Pritchard, Functional analysis of CHX21: a putative sodium transporter in Arabidopsis, J. Exp. Bot. 57 (2006) 1201–1210. [13] J. Zhao, N.H. Cheng, C.M. Motes, E.B. Blancaflor, M. Moore, N. Gonzales, S. Padmanaban, H. Sze, J.M. Ward, K.D. Hirschi, AtCHX13 is a plasma membrane K+ transporter, Plant Physiol. 148 (2008) 796–807. [14] L. Maresova, H. Sychrova, Arabidopsis thaliana CHX17 gene complements the kha1 deletion phenotypes in Saccharomyces cerevisiae, Yeast 23 (2006) 1167–1171. [15] S. Chanroj, Yongxian Lu, S. Padmanaban, Kei Nanatani, Nobuyuki Uozumi, R. Rao, Heven Sze, Plant-specific cation/H+ exchanger 17 and its homologs are endomembrane K+ transporters with roles in protein sorting, J. Biol. Chem. 286 (2011) 33931–33941. [16] S. Choe, Potassium channel structures, Nat. Rev. Neurosci. 3 (2002) 115–121. [17] I.R. Booth, Bacterial ion channels, Genet. Eng. (N Y) 25 (2003) 91–111. [18] S. Miller, L.S. Ness, C.M. Wood, B.C. Fox, I.R. Booth, Identification of an ancillary protein, YabF, required for activity of the KefC glutathione-gated potassium efflux system in Escherichia coli, J. Bacteriol. 182 (2000) 6536–6540. [19] T.P. Roosild, S. Miller, I.R. Booth, S. Choe, A mechanism of regulating transmembrane potassium flux through a ligand-mediated conformational switch, Cell 109 (2002) 781–791. [20] T.P. Roosild, S. Castronovo, S. Miller, C. Li, T. Rasmussen, W. Bartlett, B. Gunasekera, S. Choe, I.R. Booth, KTN (RCK) domains regulate K+ channels and transporters by controlling the dimer-hinge conformation, Structure 17 (2009) 893–903. [21] T.P. Roosild, S. Castronovo, J. Healy, S. Miller, C. Pliotas, T. Rasmussen, W. Bartlett, S.J. Conway, I.R. Booth, Mechanism of ligand-gated potassium efflux in bacterial pathogens, Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 19784–19789. [22] M. Fujisawa, M. Ito, T.A. Krulwich, Three two-component transporters with channel-like properties have monovalent cation/proton antiport activity, Proc. Natl. Acad. Sci. U. S. A. 104 (2007) 13289–13294. [23] L. Maresova, H. Sychrova, Physiological characterization of Saccharomyces cerevisiae kha1 deletion mutants, Mol. Microbiol. 55 (2005) 588–600. [24] O. Cagnac, M.N. Aranda-Sicilia, M. Leterrier, M.P. Rodriguez-Rosales, K. Venema, Vacuolar cation/H + antiporters of Saccharomyces cerevisiae, J. Biol. Chem. 285 (2010) 33914–33922. [25] A. Rodriguez-Navarro, J. Ramos, Dual system for potassium transport in Saccharomyces cerevisiae, J. Bacteriol. 159 (1984) 940–945. [26] R.D. Gietz, R.H. Schiestl, A.R. Willems, R.A. Woods, Studies on the Transformation of Intact Yeast-Cells by the Liac/S-DNA/Peg Procedure, Yeast 11 (1995) 355–360. [27] C. Fusco, E. Guidotti, A.S. Zervos, In vivo construction of cDNA libraries for use in the yeast two-hybrid system, Yeast 15 (1999) 715–720. [28] M.V. Singh, P.A. Weil, A method for plasmid purification directly from yeast, Anal. Biochem. 307 (2002) 13–17. [29] M. Leterrier, F.J. Corpas, J.B. Barroso, L.M. Sandalio, L.A. del Rio, Peroxisomal monodehydroascorbate reductase. Genomic clone characterization and functional

[33]

[34] [35]

[36] [37]

[38]

[39]

[40]

[41]

[42] [43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

[54] [55]

[56]

[57] [58]

2371

analysis under environmental stress conditions, Plant Physiol. 138 (2005) 2111–2123. M. Karimi, D. Inze, A. Depicker, GATEWAY vectors for Agrobacterium-mediated plant transformation, Trends Plant Sci. 7 (2002) 193–195. J.F. Li, E. Park, A.G. von Arnim, A. Nebenfuhr, The FAST technique: a simplified Agrobacterium-based transformation method for transient gene expression analysis in seedlings of Arabidopsis and other plant species, Plant Methods 5 (2009) 6. W. Wang, B.A. Malcolm, Two-stage polymerase chain reaction protocol allowing introduction of multiple mutations, deletions, and insertions, using QuikChange site-directed mutagenesis, Methods Mol. Biol. 182 (2002) 37–43. K. Venema, F.J. Quintero, J.M. Pardo, J.P. Donaire, The Arabidopsis Na+/H+ exchanger AtNHX1 catalyzes low affinity Na+ and K+ transport in reconstituted liposomes, J. Biol. Chem. 277 (2002) 2413–2418. G. Grynkiewicz, M. Poenie, R.Y. Tsien, A new generation of Ca2+ indicators with greatly improved fluorescence properties, J. Biol. Chem. 260 (1985) 3440–3450. O. Cagnac, M. Leterrier, M. Yeager, E. Blumwald, Identification and characterization of Vnx1p, a novel type of vacuolar monovalent cation/H+ antiporter of Saccharomyces cerevisiae, J. Biol. Chem. 282 (2007) 24284–24293. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. L. Kerkeb, K. Venema, J.P. Donaire, M.P. Rodriguez-Rosales, Enhanced H+/ATP coupling ratio of H+-ATPase and increased 14–3–3 protein content in plasma membrane of tomato cells upon osmotic shock, Physiol. Plant 116 (2002) 37–41. M.M. Bradford, A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72 (1976) 248–254. D.O. Daley, M. Rapp, E. Granseth, K. Melen, D. Drew, G. von Heijne, Global topology analysis of the Escherichia coli inner membrane proteome, Science 308 (2005) 1321–1323. C. Hunte, E. Screpanti, M. Venturi, A. Rimon, E. Padan, H. Michel, Structure of a Na +/H + antiporter and insights into mechanism of action and regulation by pH, Nature 435 (2005) 1197–1202. S. Miller, R.M. Douglas, P. Carter, I.R. Booth, Mutations in the glutathione-gated KefC K+ efflux system of Escherichia coli that cause constitutive activation, J. Biol. Chem. 272 (1997) 24942–24947. M. Minet, M.E. Dufour, F. Lacroute, Complementation of Saccharomyces cerevisiae auxotrophic mutants by Arabidopsis thaliana cDNAs, Plant J. 2 (1992) 417–422. O. Emanuelsson, H. Nielsen, G. von Heijne, ChloroP, a neural network-based method for predicting chloroplast transit peptides and their cleavage sites, Protein Sci. 8 (1999) 978–984. A. Bräutigam, A.P.M. Weber, Proteomic analysis of the proplastid envelope membrane provides novel insights into small molecule and protein transport across proplastid membranes, Mol. Plant 2 (2009) 1247–1261. M. Ferro, S. Brugiere, D. Salvi, D. Seigneurin-Berny, M. Court, L. Moyet, C. Ramus, S. Miras, M. Mellal, S. Le Gall, S. Kieffer-Jaquinod, C. Bruley, J. Garin, J. Joyard, C. Masselon, N. Rolland, AT_CHLORO, a comprehensive chloroplast proteome database with subplastidial localization and curated information on envelope proteins, Mol. Cell. Proteomics 9 (2010) 1063–1084. B. Zybailov, H. Rutschow, G. Friso, A. Rudella, O. Emanuelsson, Q. Sun, K.J. van Wijk, Sorting signals, N-terminal modifications and abundance of the chloroplast proteome, PLoS One 3 (2008) e1994. C.L. Brett, D.N. Tukaye, S. Mukherjee, R. Rao, The yeast endosomal Na+K+/H+ exchanger Nhx1 regulates cellular pH to control vesicle trafficking, Mol Biol Cell 16 (2005) 1396–1405. K. Venema, A. Belver, M.C. Marin-Manzano, M.P. Rodriguez-Rosales, J.P. Donaire, A novel intracellular K+/H+ antiporter related to Na+/H+ antiporters is important for K+ ion homeostasis in plants, J. Biol. Chem. 278 (2003) 22453–22459. I.T. Arkin, H. Xu, M.O. Jensen, E. Arbely, E.R. Bennett, K.J. Bowers, E. Chow, R.O. Dror, M.P. Eastwood, R. Flitman-Tene, B.A. Gregersen, J.L. Klepeis, I. Kolossvary, Y. Shan, D.E. Shaw, Mechanism of Na+/H+ antiporting, Science 317 (2007) 799–803. J.K. Pittman, K.D. Hirschi, Regulation of CAX1, an Arabidopsis Ca(2 +)/H + antiporter. Identification of an N-terminal autoinhibitory domain, Plant Physiol. 127 (2001) 1020–1029. J.F. Harper, B. Hong, I. Hwang, H.Q. Guo, R. Stoddard, J.F. Huang, M.G. Palmgren, H. Sze, A novel calmodulin-regulated Ca2+−ATPase (ACA2) from Arabidopsis with an N-terminal autoinhibitory domain, J. Biol. Chem. 273 (1998) 1099–1106. P. Maser, S. Thomine, J.I. Schroeder, J.M. Ward, K. Hirschi, H. Sze, I.N. Talke, A. Amtmann, F.J. Maathuis, D. Sanders, J.F. Harper, J. Tchieu, M. Gribskov, M.W. Persans, D.E. Salt, S.A. Kim, M.L. Guerinot, Phylogenetic relationships within cation transporter families of Arabidopsis, Plant Physiol. 126 (2001) 1646–1667. R. Nass, R. Rao, Novel localization of a Na +/H + exchanger in a late endosomal compartment of yeast. Implications for vacuole biogenesis, J. Biol. Chem. 273 (1998) 21054–21060. Z. Fang, F. Mi, G.A. Berkowitz, Molecular and physiological analysis of a thylakoid K + channel protein, Plant Physiol. 108 (1995) 1725–1734. X. Wang, G.A. Berkowitz, J.S. Peters, K+−conducting ion channel of the chloroplast inner envelope: functional reconstitution into liposomes, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 4981–4985. S.P. Robinson, Osmotic adjustment by intact isolated chloroplasts in response to osmotic stress and its effect on photosynthesis and chloroplast volume, Plant Physiol. 79 (1985) 996–1002. D.C. McCain, Combined effects of light and water stress on chloroplast volume regulation, Biophys. J. 69 (1995) 1105–1110. A.S. Gupta, G.A. Berkowitz, Chloroplast osmotic adjustment and water stress effects on photosynthesis, Plant Physiol. 88 (1988) 200–206.