Functional reconstitution and characterization of the Arabidopsis Mg2+ transporter AtMRS2-10 in proteoliposomes

Biochimica et Biophysica Acta 1818 (2012) 2202–2208

Contents lists available at SciVerse ScienceDirect

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

Functional reconstitution and characterization of the Arabidopsis Mg 2+ transporter AtMRS2-10 in proteoliposomes Sumio Ishijima ⁎, Zenpei Shigemi, Hiroaki Adachi, Nana Makinouchi, Ikuko Sagami Graduate School of Life and Environmental Sciences, Kyoto Prefectural University, Shimogamo, Sakyo-ku, Kyoto 606‐8522, Japan

a r t i c l e

i n f o

Article history: Received 3 August 2011 Received in revised form 5 April 2012 Accepted 19 April 2012 Available online 26 April 2012 Keywords: AtMRS2 Magnesium transport Proteoliposome Arabidopsis thaliana CorA superfamily

a b s t r a c t Magnesium (Mg2+) plays critical role in many physiological processes. The mechanism of Mg 2+ transport has been well documented in bacteria; however, less is known about Mg 2+ transporters in eukaryotes. The AtMRS2 family, which consists of 10 Arabidopsis genes, belongs to a eukaryotic subset of the CorA superfamily proteins. Proteins in this superfamily have been identified by a universally conserved GlyMetAsn motif and have been characterized as Mg2+ transporters. Some members of the AtMRS2 family, including AtMRS2-10, may complement bacterial mutants or yeast mutants that lack Mg2+ transport capabilities. Here, we report the purification and functional reconstitution of AtMRS2-10 into liposomes. AtMRS2-10, which contains an N-terminal His-tag, was expressed in Escherichia coli and solubilized with sarcosyl. The purified AtMRS2-10 protein was reconstituted into liposomes. AtMRS2-10 was inserted into liposomes in a unidirectional orientation. Direct measurement of Mg2+ uptake into proteoliposomes revealed that reconstituted AtMRS2-10 transported Mg2+ without any accessory proteins. Mutation in the GMN motif, M400 to I, inactivated Mg 2+ uptake. The AtMRS2-10-mediated Mg2+ influx was blocked by Co(III)hexamine, and was independent of the external pH from 5 to 9. The activity of AtMRS2-10 was inhibited by Co2+ and Ni2+; however, it was not inhibited by Ca2+, Fe2+, or Fe3+. While these results indicate that AtMRS2-10 has similar properties to the bacterial CorA proteins, unlike bacterial CorA proteins, AtMRS2-10 was potently inhibited by Al3+. These studies demonstrate the functional capability of the AtMRS2 proteins in proteoliposomes to study structure–function relationships. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Magnesium is one of the most important and abundant divalent cations within living cells. Mg2+ is as an essential cofactor in hundreds of enzymes and is indispensable for numerous cellular functions, including nucleotide utilization, protein synthesis, and maintenance of genomic stability [1]. In animals, Mg2+ is an important regulatory signal, and mediates many biochemical reactions [2,3]. In addition, Mg 2+ plays an even more prominent role in plants because it is an essential component of chlorophyll molecules, and Mg2+ regulates key enzymes that are involved in carbon fixation in chloroplasts [4,5]. Despite these critical cellular functions, the mechanisms of Mg 2+ uptake, transport, and homeostasis in eukaryotes are only slowly being elucidated. As with any cation, the cellular concentration of Mg 2+ is regulated by transmembrane pathways. These processes have been best characterized in prokaryotes, where the X-ray crystal structures of the Thermotoga maritima CorA [6–8] and Thermus thermophillus MgtE Abbreviations: IPTG, isopropyl-β-D-thiogalactopyranoside; DDM, n-dodecyl-β-Dmaltoside; NMDG-Cl, N-methyl-D-glucamine chloride; WT, wild type ⁎ Corresponding author. Tel.: +81 75 703 5674; fax: +81 75 703 5674. E-mail address: [email protected] (S. Ishijima). 0005-2736/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamem.2012.04.015

[9] Mg 2+ channels have been determined. CorA is the primary Mg 2+ uptake system in the Bacteria and Archaea domains [10]. CorA from T. maritima is a homopentamer with two transmembrane segments (TM1 and TM2) per monomer. The C-terminal periplasmic end of the pore-lining helix, TM1, contains the highly conserved “GMN” motif (Supplemental Fig. 1). Many studies have elucidated the functional properties of Mg 2+ transport systems, using genetic and electrophysiological analyses [11]. Payandeh et al. [12] recently presented the first direct evidence that CorA mediates selective Mg2+ flux without accessory proteins, using highly purified T. maritima CorA that was reconstituted into liposomes. Mutations in the GMN motif of CorA abolished Mg 2+ transport, which indicates that this motif is essential for function [12,13]. There are a number of eukaryotic proteins with weak homology to CorA in yeast, plants, and mammals. The best-studied member of the eukaryotic CorA superfamily (or 2-TM-GMN-type proteins [14]) is the yeast mitochondrial Mrs2p protein [15]. Arabidopsis possesses a large family of putative Mg 2+ transporters that are homologous to yeast Mrs2p and the CorA family in bacteria. This gene family has 10 members (and one pseudo-gene), and the family was initially named AtMRS2 [16], or alternatively, AtMGT [17] for magnesium transport. Some members of the AtMRS2 family, including AtMRS2-10, may complement yeast mutants [16–18] or bacterial mutants [17,19–21]

S. Ishijima et al. / Biochimica et Biophysica Acta 1818 (2012) 2202–2208

that lack Mg 2+ transport capabilities. Gebert et al. recently characterized the Mg 2+ transport properties of the whole gene family using heterologous expression in the yeast mrs2 mutant and observed efficient Mg 2+ uptake for AtMRS2-1, AtMRS2-7, and AtMRS2-10 [22]. Despite the results that suggest Mg 2+ transport by plant MRS2 proteins, there have been no reports that have presented the purification and molecular characterization of the plant MRS2 proteins. In this study, AtMRS2-10 was expressed in Escherichia coli cells, purified and reconstituted into liposomes. AtMRS2-10 belongs to the superfamily of 2-TM-GMN-type proteins. AtMRS2-10 possesses the signature GMN sequence; however, it has low sequence conservation with CorA (Supplemental Fig. 1). AtMRS2-10 is able to complement bacterial corA and yeast mrs2, alr1, and alr2 mutants [16–18]. Transgenic plants overexpressing AtMRS2-10 have shown an enhanced tolerance to Mg2+ deficiency and an improved tolerance to Al [23]. Although these studies indicate a functional homology to CorA, direct functional evidence on the plant protein is currently unavailable. We have now provided the first direct evidence that purified, reconstituted AtMRS2-10 mediates the transport of magnesium ions, and its transport was inhibited by Al3+. 2. Materials and methods 2.1. Plasmids The Arabidopsis sequence (stock no. ATTS4604) that encodes for AtMRS2-10 was obtained from the Arabidopsis Biological Resource Center (http://abrc.osu.edu/). AtMRS2-10 was subcloned in-frame at the NdeI site of the pET28a vector (Novagen). Site-directed mutagenesis was performed using the QuikChange protocol (Stratagene), and all plasmid sequences were confirmed by DNA sequencing. 2.2. Protein purification Proteins were expressed in E. coli BL21-Codon Plus(DE3)-RIL cells (Stratagene) grown in 3 L of TB media that contained kanamycin (30 μg/ml) and chloramphenicol (35 μg/ml), and the cells were induced at an OD600 of ~0.6 with 50 μM isopropyl-β-D-thiogalactopyranoside (IPTG) at 15 °C for 24–27 h. After harvesting, the cultures were resuspended in 15 ml buffer A (50 mM sodium phosphate buffer at pH 7.8, 300 mM NaCl, 2 mM phenylmethylsulfonylfluoride, 2 μg/ml aprotinin, 2 μg/ml leupeptin and 2 μg/ml pepstatin A) and were sonicated. The cells were centrifuged at 8000×g for 25 min at 4 °C, and 1% (w/v) Tween 20 was added to the supernatant. The supernatant was removed after centrifugation at 100,000 ×g for 60 min at 4 °C. The pellet was resuspended in 40 ml buffer A and was sonicated. Sarcosyl (N-laurylsarcosinate) was added to a final concentration of 0.3% (w/v), and the solubilized pellet was centrifuged at 100,000 ×g for 45 min at 4 °C. The supernatant was loaded onto 3 ml of Ni-NTA agarose (Qiagen) that was pre-equilibrated with buffer B (50 mM sodium phosphate buffer at pH 7.8, 300 mM NaCl, and 0.1% (w/v) sarcosyl) and was washed with buffer B that contained 10 mM imidazole. The protein was eluted in buffer B that contained 100 mM imidazole and was passed over a Sephadex G-25 column in buffer C (20 mM HEPES buffer at pH 7.0 and 100 mM KCl) that contained 0.1% sarcosyl. 2.3. Reconstitution into liposomes Unilamellar liposomes were prepared by the freeze–thaw and extrusion method. Briefly, dioleoylphosphatidylcholine and dioleoylphosphatidylserine were mixed at a molar ratio of 9:1 in chloroform and were deposited as a thin film on the interior of a flask by rotary evaporation under reduced pressure. The lipid film was suspended at 20 mM in buffer C, which resulted in a suspension of multilamellar liposomes. The suspension was subjected to 10 cycles of freezing and thawing and was stored at −80 °C until use. Unilamellar vesicles

2203

were produced by extrusion through polycarbonate membranes (pore size of 100 nm; Avanti Polar Lipids) 15 times using an Avanti Mini-Extruder and were suspended to a final concentration of 11 mM in buffer C. The liposomes were destabilized by the addition of 0.53% (w/v) n-dodecyl-β-D-maltoside (DDM) and were incubated for 3 h at room temperature with agitation. Freshly purified AtMRS2-10 was added to produce a lipid/protein ratio of 40:1 (w/w), and the mixture that contained 0.53% DDM was incubated for 2 h at 4 °C with agitation. The detergents were removed by incubating the mixture for 1 h at 4 °C with agitation with Bio-Beads SM-2 (Bio-Rad) at a bead (wet weight)/detergent ratio of 60:1 (w/w). Bio-Beads SM-2 was removed from the solution, and a second aliquot of Bio-Beads SM-2 was added. The sample was incubated for 1 h at 4 °C with agitation. The proteoliposome solution was removed from the Bio-Beads SM-2 and was centrifuged at 210,000 ×g for 30 min at 4 °C. The pellet was resuspended in potassium-free buffer D (100 mM N-methyl-D-glucamine chloride (NMDG-Cl) and 20 mM HEPES at pH 7.0) and was immediately used. 2.4. Mg 2+ transport assay by atomic absorption spectroscopy The liposomes with incorporated AtMRS2-10 were pre-incubated in buffer D for 5 min at 20 °C. Magnesium influx was initiated by adding 15 mM MgCl2 to the liposome suspensions. After incubation for 5 min at 20 °C, 1 mM Co(III)hexamine was added to block Mg 2+ flux through AtMRS2-10. The liposome suspensions were centrifuged at 100,000 ×g for 10 min at 4 °C to eliminate aggregated compounds, and the supernatant was centrifuged at 210,000 ×g for 30 min at 4 °C. The pellet was washed four times in buffer C that contained 1 mM Co(III)hexamine and was resuspended in Milli-Q water that contained 0.5% (w/v) SDS and 0.1 M HCl. The Mg2+ content was measured in an AA-6200 atomic absorption spectrophotometer (Shimadzu) using an air-acetylene flame at a wavelength of 285.2 nm. 2.5. Determination of protein and phospholipid content in the proteoliposomes The concentration of AtMRS2-10 proteins in proteoliposomes was determined by SDS-PAGE using bovine serum albumin as a standard. The phospholipid content of the proteoliposomes was assessed using a standard phospholipid assay kit (Wako Phospholipids C). 3. Results 3.1. Expression, solubilization and purification of AtMRS2-10 Expression of AtMRS2-10 was achieved using the pET28a(+) construct with an N-terminal His6-tag that was transformed into E. coli BL21-Codon Plus(DE3)-RIL cells. There was little expression detected when using the pBAD/His plasmid in E. coli or when using the pDEST10 or pDEST20 plasmids in the Bac-to-Bac baculovirus expression system (Invitrogen). The protein expression levels of the 54 kDa AtMRS2-10 were examined in E. coli BL21-Codon Plus cells grown at 15 or 37 °C in LB or TB media after induction with 0.05, 0.5 or 1 mM IPTG using SDS-PAGE (data not shown); The highest level of expression was obtained after induction with 0.05 mM IPTG when the cells were grown at 15 °C for 20 h in TB media. Coexpression of the molecular chaperone GroEL/ES did not provide a noticeable effect on the AtMRS2-10 expression level. The expressed AtMRS2-10 was efficiently solubilized with 0.3% sarcosyl (Supplemental Fig. 2); however, other detergents, including 2% DDM and 3% Tween 20 (Table 1), did not solubilize AtMRS2-10. AtMRS2-10 was also insoluble in 0.2% sarcosyl. Therefore, AtMRS2-10 was purified from the 1% Tween 20-insoluble fractions and from the 0.3% sarcosyl-soluble fractions. Once AtMRS2-

2204

S. Ishijima et al. / Biochimica et Biophysica Acta 1818 (2012) 2202–2208

A

Table 1 Detergent solubilization screening of AtMRS2-10. Detergent

% Solubilization Added efficiency

n-Dodecyl-β-D-maltoside (DDM) n-Octyl-β-D-glucoside Tween 20 Triton X-100 Nonidet P40 Briji 35 3-[(3-Cholamidopropyl)dimethylammonio]propanesulfonic acid (CHAPS) Cholic acid N-Laurylsarcosinate (sarcosyl) Cetyltrimethylammonium chloride (CTAC)

1–2 1–2 1–3 1–3 1–3 1–3 1

− − − − − − −

1–5 0.3–2 1–2

− ++ +

1

2

3

4

5

kDa 66

45

++, Good solubilization efficiency; +, low solubilization efficiency; −, no solubilizaion.

31

10 was solubilized with 0.3% sarcosyl, it stayed soluble in 0.1% sarcosyl. AtMRS2-10 was subsequently purified in the presence of 0.1% sarcosyl. 3.2. Purification of the GMN-motif mutant M400I and the N-terminaldeleted Δ16 and Δ20 variants The CorA superfamily of proteins, which includes AtMRS2, contains the conserved GMN motif [18], and this motif plays an essential role in the bacterial and yeast CorA superfamily proteins [12,13,24]. We generated the M400I mutant that targeted the conserved GMN motif. We also generated variants where the N-terminal 16 and 20 amino acids were deleted, which were named AtMRS2-10 Δ16 and Δ20, respectively (Supplemental Fig. 3). Previously, AtMRS2-10 was reported to localize to the plasma membrane when using green fluorescent protein as a reporter [17], yet some algorithms predict the possibility of localization to chloroplasts. TargetP (http://www.cbs.dtu. dk/services/TargetP/) and ChloroP (http://www.cbs.dtu.dk/services/ ChloroP/) predicted a chloroplast leader and transit peptide cleavage site at amino acid 20, and WoLF PSORT (http://wolfpsort.org/) and SignalP (http://www.cbs.dtu.dk/services/SignalP/) predict a cleavage site at amino acid 16. The mutant and variant proteins were expressed and purified using the procedures that were similar to full-length AtMRS2-10. For full-length AtMRS2-10, M400I, and the truncated Δ16 and Δ20 variants these procedures reproducibly yielded 2 mg of protein from 1 L of bacterial culture. 3.3. Reconstitution of AtMRS2-10 into proteoliposomes Purified AtMRS2-10 was reconstituted into phosphatidylcholine and phosphatidylserine-based liposomes. Purified AtMRS2-10 in 0.1% sarcosyl was added to liposomes destabilized with 0.53% DDM and detergents were removed by adsorption to Bio-Beads SM-2. To analyze the orientation of reconstituted AtMRS2-10 in the proteoliposomes, they were treated with trypsin and chymotrypsin (Fig.1A). As a control, recombinant AtMRS2-10 was subjected to the same treatment. Although protease digestion of the recombinant protein led to an almost complete digestion, >90% of reconstituted AtMRS2-10 was undegraded or degraded only near the terminus by these proteases. These results indicated that AtMRS2-10 was preferentially inserted into proteoliposomes with the most part exposed to the interior of the liposomes in our reconstituted samples. 3.4. Reconstituted AtMRS2-10 mediates Mg 2+ flux and is blocked by Co(III)hexamine The proteoliposomes were incubated with 15 mM MgCl2 in the presence and absence of 1 mM Co(III)hexamine. Co(III)hexamine is an analog of hydrated Mg 2+ and is a potent and selective inhibitor

B

C

N

Fig. 1. Orientation of AtMRS2-10 in proteoliposomes. A, protease protection assay of AtMRS2-10 in proteoliposomes. AtMRS2-10 integrated into liposomes (lanes 1 and 3) and recombinant AtMRS2-10 protein (lanes 2 and 4) were incubated with trypsin (lanes 1 and 2) and chymotrypsin (lanes 3 and 4) in a protease: AtMRS2-10 ratio of 1:50 at 20 °C for 30 min. Protease digestion was stopped by the addition of SDSPAGE sample buffer and incubation for 3 min at 98 °C. Lane 5, proteoliposomes alone. B, schematic representation of the topology of AtMRS2-10 in proteoliposomes.

of CorA superfamily members [12,17,25–27]. After incubation for 5 min with 15 mM MgCl2, 1 mM Co(III)hexamine was added to the proteoliposomes. In subsequent procedures, external Mg 2+ was removed in the presence of 1 mM Co(III)hexamine to block Mg 2+ efflux through AtMRS2-10. Reconstituted AtMRS2-10 was able to mediate Mg 2+ influx into liposomes in the presence of an inward-directed Mg 2+ gradient (Fig. 2A). A significant difference in the internal Mg 2+ content was observed with and without AtMRS2-10 incorporated into the liposomes. The AtMRS2-10-mediated Mg 2+ influx was dependent on the external Mg 2+ concentration. The activity in the proteoliposomes depended on the lipid/protein ratio when AtMRS2-10 was incorporated into the liposomes. Different lipid/protein ratios were tested, and the optimal ratio was approximately 40:1 (w/w). The Mg 2+ influx through AtMRS2-10 was reduced in the presence of 1 mM Co(III) hexamine (Fig. 2B). While the measured Mg 2+ content varied from experiment to experiment, which was likely due to concentration differences of unremoved external Mg 2+ and of Mg 2+ adsorbed to the phospholipids, the values measured in each experiment (n = 3 in

S. Ishijima et al. / Biochimica et Biophysica Acta 1818 (2012) 2202–2208

B

A 80 60 40 20 0

800

Internal Mg Content (nmol/mg of protein)

Internal Mg Content (% Maximal)

100

5

0

10

20

15

External [Mg2+] (mM)

600 400 200 0

D Mg2+ Influx (nmol/mg of protein)

Mg2+ Influx (Percent Control)

100

75

50

25

0

0.01

+ + Exp 1 Exp 2

Co(III)hexamine

C

0

2205

0.1

250 200 150 100 50 0

1.0

+ Exp 3

WT

M400I

[Co(III)hexamine] (mM)

F

E 200 150 100 50 0

WT

Δ16

(nmol/mg of protein)

400

Mg2+ Influx

Mg2+ Influx (nmol/mg of protein)

250

Δ20

300

200

100

0

5.0

6.0

7.0

8.0

9.0

External pH Fig. 2. Reconstituted AtMRS2-10-mediated Mg2+ influx. A, liposomes with (circles) and without (triangle) incorporated AtMRS2-10 were incubated (intraliposomal: 0 mM MgCl2, 100 mM KCl; extraliposomal: 100 mM NMDG-Cl) in the presence of the indicated concentrations of external Mg2+. After incubation for 5 min at 20 °C, 1 mM Co(III)hexamine was added. External Mg2+ was removed, and the Mg2+ content was determined by atomic absorption spectrophotometry as described in the Materials and methods section. The internal Mg2+ content is presented as the percentage of the maximal Mg2+ content in each experiment. Data are presented as the average value of three independent experiments, and the results are presented as the mean ± S.D. B, three independent experiments that show that Co(III)hexamine inhibited Mg2+ influx through AtMRS2-10 when 1 mM Co(III)hexamine was incubated with the proteoliposomes (intraliposomal: 0 mM MgCl2, 100 mM KCl; extraliposomal: 15 mM MgCl2, 100 mM NMDG-Cl). The experiments were performed as described in A. Data are shown as the mean± S.D., n = 3. C, the proteoliposomes were incubated (intraliposomal: 0 mM MgCl2, 100 mM KCl; extraliposomal: 15 mM MgCl2, 100 mM NMDG-Cl) in the presence of the indicated concentrations of Co(III)hexamine. After incubation for 5 min at 20 °C, 1 mM Co(III)hexamine was added. The internal Mg2+ content was measured as described in A. The difference in the Mg2+ content in the presence and absence of 1 mM Co(III)hexamine is shown as Mg2+ influx through AtMRS2-10. Data are shown as the mean ± S.D., n = 3. D, Mg2+ influx with the M400I mutant. Mg2+ influx was determined as described in C. Data are shown as the mean ± S.D., n = 3. E, Mg2+ influx of the truncated Δ16 and Δ20 variants. Data are shown as the mean ± S.D., n = 3. F, the effect of external pH on Mg2+ influx through AtMRS2-10. Data are shown as the mean± S.D., n = 3.

each experiment) were relatively constant. The measured Mg 2+ content in the presence of 1 mM Co(III)hexamine was nearly equal to the value when the Mg 2+ content was measured for liposomes without AtMRS2-10 (Fig. 2A), which indicated that 1 mM Co(III)hexamine completely inhibited Mg 2+ influx through AtMRS2-10. Therefore, it was hypothesized that the difference in the presence and absence of 1 mM Co(III)hexamine indicated the amount of Mg 2+ that was transported into the liposomes through AtMRS2-10. Mg 2+ influx through AtMRS2-10 was subsequently evaluated by the difference in the presence and absence of 1 mM Co(III)hexamine. The inhibition was dependent on the concentration of Co(III)hexamine, and 50% inhibition was observed at 0.04 mM Co(III)hexamine (Fig. 2C). The activity of the GMN-motif mutant, M400I, was reduced in this assay (Fig. 2D). These results provided direct evidence that AtMRS2-10 mediates the Mg2+

influx and indicated that AtMRS2-10-mediated Mg2+ influx was blocked by Co(III)hexamine. Mg 2+ efflux through AtMRS2-10 in the presence and absence of Co(III)hexamine could not be examined because the liposomes aggregated in the presence of Mg 2+ when the liposomes were prepared and preloaded with Mg 2+. Although the truncated Δ16 and Δ20 variants mediated Mg2+ influx into the liposomes, the amount of Mg2+ influx was reduced compared with the full-length AtMRS2-10 (Fig. 2E). Removal of the N-terminal His6-tag by thrombin digestion did not increase the Mg2+ influx. Rather, higher activity of the Mg2+ influx was observed with N-terminal His6-tagged AtMRS2-10 (data not shown). Therefore, fulllength AtMRS2-10 with an N-terminal His6-tag was used for all further studies. The AtMRS2-10-mediated Mg2+ influx was independent of the

2206

S. Ishijima et al. / Biochimica et Biophysica Acta 1818 (2012) 2202–2208

external pH, and there were no significant differences observed when using an external pH from 5 to 9 (Fig. 2F). 3.5. Effects of cations on AtMRS2-10-mediated Mg 2+ flux The inhibition of AtMRS2-10-mediated Mg2+ influx by Co(III) hexamine indicated that the transport properties of AtMRS2-10 were similar to bacterial CorA. In previous functional studies of bacterial CorA and AtMRS2 [16–22], these proteins were expressed in S. typhimurium and yeast mutants that lacked the ability to uptake Mg 2+. In these studies, the functional properties of the cations have been indicated by tracer-ion uptake inhibition assays where 63Ni2+ was used as a substitute because 28Mg2+ is not available. In this study, the effects of various cations on AtMRS2-10-mediated Mg2+ influx were directly assayed, and we examined the effects of Ca2+ and five cations to determine their effect in Mg2+ transport. Mg 2+ transport in bacterial CorA and AtMRS2 is known to be insensitive to Ca2+ and is known to be sensitive to Co2+ and Ni2+. The proteoliposomes were incubated with various cations and with 15 mM Mg 2+ in the presence and absence of 1 mM Co(III) hexamine. After incubation for 5 min, the internal Mg 2+ content was determined. Six cations were examined, and Co 2+, Ni 2+, and Al 3+ all inhibited Mg 2+ uptake (Fig. 3). The resulting inhibition by 5 mM Co 2+ and Ni 2+ was 39% and 55%, respectively, and 0.5 mM Al 3+ completely inhibited the uptake of Mg 2+. In contrast, 10 mM Ca 2+ and 15 mM Fe 2+ or Fe 3+ had no effect on the uptake of Mg 2+. 4. Discussion Here, we purified AtMRS2-10 and reconstituted it into proteoliposomes. The amount of Mg 2+ that was transported into the proteoliposomes was measured by atomic absorption spectroscopy.

A

Upon analysis of Mg 2+ transport activity, most Mg 2+ transport proteins are expressed in the mitochondria and in bacterial and yeast cells, and the mitochondria and whole cells and their membranes were assayed. The only Mg 2+ transport protein that had been previously purified and reconstituted into liposomes was T. maritima CorA [12]. AtMRS2-10 has low sequence conservation with T. maritima CorA. Nonetheless, the predicted secondary structure of AtMRS2-10 agrees with the secondary structure of the T. maritima CorA found in the three-dimensional structure (Supplemental Fig. 1). T. maritima CorA has two transmembrane segments per monomer. AtMTS2-10 was also predicted to have two transmembrane segments at similar positions using the SOSUI algorithm (http://bp.nuap.nagoya-u.ac.jp/ sosui/). T. maritima CorA was solubilized in DDM; however, AtMRS210 was insoluble in this detergent and in most other detergents that are commonly used to solubilize membrane proteins (Table 1). AtMRS2-10 was purified from the Tween 20-insoluble fraction, and AtMRS2-10 was efficiently solubilized in 0.3% sarcosyl. Sarcosyl is a harsh detergent, and has been previously used to solubilize active soybean recombinant phosphate transporter [28]. Purified AtMRS2-10 was reconstituted into liposomes using DDM. Functional reconstitution was obtained under the conditions. While AtMRS2-10 retained functional activity in the present study, the full activity may not have been evaluated. The amount of Mg 2+ in the proteoliposomes was directly determined by atomic absorption spectroscopy, and external Mg 2+ should be removed in this assay. Since 1 mM Co(III)hexamine nearly completely inhibited Mg 2+ influx through AtMRS2-10, the amount of Mg 2+ that was transported into the liposomes was evaluated by the difference in the presence and absence of 1 mM Co(III)hexamine. In addition, there are two other points that must be discussed for the Mg 2+ transport assay in AtMRS2-10, the inhibition of Mg 2+ efflux and the orientation of AtMRS2-10 in proteoliposomes. Co(III)hexamine is a potent inhibitor

B

100

100

80

80

60

60

40

40

20

20

0

0

0

10

1.0

0

1.0

10

[Ni2+] (mM)

[Co2+] (mM)

C

D 120

100

100 80 80 60 60 40

40

20 0

20 0

0.01

0.1

[Al3+] (mM)

1.0

0

Ca2+

Fe2+

Fe3+

(10 mM)

(15 mM)

(15 mM)

Fig. 3. The effect of cations on Mg2+ influx through AtMRS2-10. A–D, the proteoliposomes were incubated (intraliposomal: 0 mM MgCl2, 100 mM KCl; extraliposomal: 15 mM MgCl2, 100 mM NMDG-Cl) with the indicated concentration of cations in the presence and absence of 1 mM Co(III)hexamine. Mg2+ influx was determined as described in Fig. 2C. The data were normalized to the percentage of control (Mg2+ uptake with 15 mM Mg2+ without other cations). Data are shown as the mean ± S.D., n = 3. Aggregation of the liposomes was observed in the presence of 15 mM Co2+ and Ca2+, and Mg2+ uptake was not determined in the presence of 15 mM Co2+ and Ca2+.

S. Ishijima et al. / Biochimica et Biophysica Acta 1818 (2012) 2202–2208

of CorA superfamily proteins. In the present assay, after incubation with Mg 2+, the external Mg 2+ was removed in the presence of 1 mM Co(III)hexamine. Co(III)hexamine has also been reported to inhibit Mg 2+ efflux through the yeast mitochondrial Mrs2p protein [27]. However, the inhibition of Mg 2+ efflux through AtMRS2-10 by 1 mM Co(III)hexamine was not evaluated. The measurement of Mg 2+ uptake in this assay might be underestimated by Mg2+ efflux through AtMRS2-10 that may not be completely inhibited by external 1 mM Co(III)hexamine. The results shown in Fig. 1A indicated that the orientation of AtMRS2-10 in proteoliposomes was essentially unidirectional. Rigaud and Levy [29] indicated that when proteoliposomes were properly constructed, a unidirectional insertion of the protein into the membrane is ensured. The optimal concentration of DDM that was used to destabilize the liposomes was determined by the measurement of Mg2+ transport activity, and the determined concentration (0.53%) corresponded to calculations that were previously reported [29]. Therefore, an asymmetric orientation, rather than a random orientation, of AtMRS2-10 was expected in this study. Common features of all 2-TMGMN-type proteins are the presence of an N-terminal soluble cytoplasmic domain followed by two adjacent transmembrane helices near the C-terminus [6–8,14,15]. The hydrophobic analysis of the amino acid sequence of AtMRS2-10 also predicted the existence of two transmembrane helices, TM1 (residues 380–402) and TM2 (residues 415–434), positioned at the C-terminal end with both the N and C termini facing the inside domain (using public servers TMHMM or TMPred). The results of protease protection assay shown in Fig. 1A were consistent with the predicted topology of AtMRS2-10, indicating that the orientation of AtMRS2-10 in proteoliposomes was essentially unidirectional, with the most part exposed to the interior of the liposomes (Fig. 1B). Mg 2+ transport activity was measured in the presence of an outward potassium gradient [12]. This potassium gradient was useful for the measurement of Mg 2+ transport activity. Payandeh et al. [12] measured the Mg 2+ transport activity of T. maritima CorA reconstituted into liposomes using the Mg 2+-sensitive fluorescent dye Mag-Fura-2. We also observed the Mg 2+ influx through AtMRS2-10 with Mag-Fura-2 (data not shown), details of which will be published elsewhere. Mg 2+ transport activity of reconstituted AtMRS2-10 was inhibited by Co(III)hexamine and Al 3+, and AtMRS2-10 was moderately inhibited by Ni 2+ and Co 2+. No inhibition was observed for Ca 2+, Fe 2+, and Fe 3+. In previous studies, the transport properties of the CorA superfamily proteins including AtMRS2 proteins [17] have been analyzed in uptake experiments using 63Ni 2+ as a tracer. In these experiments, inhibition of 63Ni 2+ uptake into bacterial cells was assayed. The present results from the purified protein are mostly consistent with the results from the 63Ni tracer studies. This consistency indicated availability of the 63Ni 2+ uptake inhibition assay, which was used to study the cation transport activity of the CorA superfamily proteins, and also suggested that accessory proteins are not required for function of the CorA superfamily proteins. Although the inhibition of Mg 2+ uptake indicated an interaction of the cations with AtMRS2-10, it does not show the transport of these cations through AtMRS2-10. Bacterial CorA mediates the influx of Co 2+ and Ni 2+ with an affinity of 20–40 μΜ and 200–400 μΜ, respectively [30,31]. Recently, Xia et al. reported that the T. maritima CorA has the possible role of being the transporter of Co 2+ physiologically [32]. AtMRS2-10 may also transport Co 2+ and Ni 2+, however, the increased selectivity for Co 2+ over Ni 2+ was not observed for plant AtMRS2-10. In addition, 50% inhibition of Mg 2+ influx through AtMRS2-10 was observed at Co 2+ and Ni 2+ concentrations greater than 5 mM and 4 mM, respectively. The Mg 2+ influx through AtMRS2-10 was also inhibited by Al 3+, and 50% inhibition was observed when 50 μΜ Al 3+ was present. While there are no reports that show the Al 3+ sensitivity of the bacterial CorA family, Al 3+ inhibited Mg 2+ influx by the eukaryotic CorA homologues [33]. As was previously proposed [17,23], the MRS2 Mg 2+ transport system

2207

in plants may be a molecular target for the Al 3+-mediated inhibition of Mg 2+ uptake, which leads to Al-induced Mg 2+ deficiency in higher plants. Currently, it is not known whether Al 3+ is transported by the eukaryotic CorA homologues. Studies whether AtMRS2-10 transports Al 3+, Co 2+ and Ni 2+ are now in progress. Despite the low sequence similarity, the eukaryotic CorA homologues in yeast and mammals have been suggested to have a similar topology and overall structure as T. maritima CorA [31]. AtMRS2-10 was inhibited by Co(III)hexamine, and the GMN-motif mutant M400I was inactive for Mg 2+ transport. These results indicate that the CorA superfamily of proteins transport Mg 2+ by a similar mechanism over large phylogenetic distances [14]. Mg 2+ plays an important role in plants because Mg 2+ acts as a cofactor in enzymes that are involved in photosynthetic carbon fixation [34]. Light induces an increase in free Mg 2+ concentration in chloroplasts [4]; however, little is known about the molecular mechanisms that regulate Mg 2+ concentration in plants. Stromal alkalinization is essential for the light-induced increase of free Mg 2+ in chloroplasts, and the change in pH may not regulate AtMRS2-10, which was suggested by the results shown in Fig. 2F. Arabidopsis possesses 10 members of the AtMRS2 superfamily. These studies demonstrate the functional capability of the AtMRS2 proteins in proteoliposomes and functionally characterize the transporters. The molecular properties and regulation of AtMRS2 remain to be examined. 5. Concluding remarks We have provided the direct evidence that plant AtMRS2-10 mediates Mg 2+ transport. In addition, these results imply that accessory proteins are not required for this function. Acknowledgments We would like to thank Dr. Hidekazu Yamada (Kyoto Prefectural University) for atomic absorption spectroscopy. This study was supported in part by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science, and Technology of Japan to S. I. (22580123) Appendix A. Supplementary data Supplementary data to this article can be found online at doi:10. 1016/j.bbamem.2012.04.015. References [1] M.E. Maguire, J.A. Cowan, Magnesium chemistry and biochemistry, Biometals 15 (2002) 203–210. [2] S. Ishijima, T. Sonoda, M. Tatibana, Mitogen-induced early increase in cytosolic free Mg2+ concentration in single Swiss 3T3 fibroblasts, Am. J. Physiol. 261 (1991) C1074–C1080. [3] A.M. Romani, M.E. Maguire, Hormonal regulation of Mg2+ transport and homeostasis in eukaryotic cells, Biometals 15 (2002) 271–283. [4] S. Ishijima, A. Uchibori, H. Takagi, R. Maki, M. Ohnishi, Light-induced increase in free Mg2+ concentration in spinach chloroplasts: measurement of free Mg2+ by using a fluorescent probe and necessity of stromal alkalinization, Arch. Biochem. Biophys. 412 (2003) 126–132. [5] O. Shaul, Magnesium transport and function in plants: the tip of the iceberg, Biometals 15 (2002) 309–323. [6] V.V. Lunin, E. Dobrovetsky, G. Khutoreskaya, R. Zhang, A. Joachimiak, D.A. Doyle, A. Bochkarev, M.E. Maguire, A.M. Edwards, C.M. Koth, Crystal structure of the CorA Mg2+ transporter, Nature 440 (2006) 833–837. [7] J. Payandeh, E.F. Pai, A structural basis for Mg2+ homeostasis and the CorA translocation cycle, EMBO J. 25 (2006) 3762–3773. [8] S. Eshaghi, D. Niegowski, A. Kohl, D.M. Molina, S.A. Lesley, P. Nordlund, Crystal structure of a divalent metal ion transporter CorA at 2.9 angstrom resolution, Science 313 (2006) 354–357. [9] M. Hattori, Y. Tanaka, S. Fukai, R. Ishitani, O. Nureki, Crystal structure of the MgtE Mg2+ transporter, Nature 448 (2007) 1072–1075. [10] D.G. Kehres, M.E. Maguire, Structure, properties and regulation of magnesium transport proteins, Biometals 15 (2002) 261–270. [11] M.E. Maguire, Magnesium transporters: properties, regulation and structure, Front. Biosci. 11 (2006) 3149–3163.

2208

S. Ishijima et al. / Biochimica et Biophysica Acta 1818 (2012) 2202–2208

[12] J. Payandeh, C. Li, M. Ramjeesingh, E. Poduch, C.E. Bear, E.F. Pai, Probing structure– function relationships and gating mechanisms in the CorA Mg2+ transport system, J. Biol. Chem. 283 (2008) 11721–11733. [13] M.A. Szegedy, M.E. Maguire, The CorA Mg2+ transport protein of Salmonella typhimurium, J. Biol. Chem. 274 (1999) 36973–36979. [14] V. Knoop, M. Groth-Malonek, M. Gebert, K. Eifler, K. Weyand, Transport of magnesium and other divalent cations: evolution of the 2-TM-GxN proteins in the MIT superfamily, Mol. Genet. Genomics 274 (2005) 205–216. [15] D.M. Bui, J. Gregan, E. Jarosch, A. Ragnini, R.J. Schweyen, The bacterial magnesium transporter CorA can functionally substitute for its putative homologue Mrs2p in the yeast inner mitochondrial membrane, J. Biol. Chem. 274 (1999) 20438–20443. [16] I. Schock, J. Gregan, S. Steinhauser, R. Schweyen, A. Brennicke, V. Knoop, A member of a novel Arabidopsis thaliana gene family of candidate Mg2+ ion transporters complements a yeast mitochondrial group II intron-splicing mutant, Plant J. 24 (2000) 489–501. [17] L. Li, A.F. Tutone, R.S.M. Drummond, R.C. Gardner, S. Luan, A novel family of magnesium transport genes in Arabidopsis, Plant Cell 13 (2001) 2761–2775. [18] R.S.M. Drummond, A. Tutone, Y.-C. Li, R.C. Gardner, A putative magnesium transporter AtMRS2-11 is localized to the plant chloroplast envelope membrane system, Plant Sci. 170 (2006) 78–89. [19] L.-G. Li, L.N. Sokolov, Y.-H. Yang, D.-P. Li, J. Ting, G.K. Pandy, S. Luan, A mitochondrial magnesium transporter functions in Arabidopsis pollen development, Mol. Plant 1 (2008) 675–685. [20] D.-D. Mao, L.-F. Tian, L.-G. Li, J. Chen, P.-Y. Deng, D.-P. Li, S. Luan, AtMGT7: an Arabidopsis gene encoding a low-affinity magnesium transporter, J. Integr. Plant Biol. 50 (2008) 1530–1538. [21] J. Chen, L.-G. Li, Z.-H. Liu, Y.-J. Yuan, L.-L. Guo, D.-D. Mao, L.-F. Tian, L.-B. Chen, S. Luan, D.-P. Li, Magnesium transporter AtMGT9 is essential for pollen development in Arabidopsis, Cell Res. 19 (2009) 887–898. [22] M. Gebert, K. Meschenmoser, S. Svidova, J. Weghuber, R. Schweyen, K. Eifler, H. Lenz, K. Weyand, V. Knoop, A root-expressed magnesium transporter of the MRS2/MGT gene family in Arabidopsis thaliana allows for growth in low-Mg2+ environments, Plant Cell 21 (2009) 4018–4030.

[23] W. Deng, K. Luo, D. Li, X. Zheng, X. Wei, W. Smith, C. Thammina, L. Lu, Y. Li, Y. Pei, Overexpression of an Arabidopsis magnesium transport gene, AtMGT1, in Nicotiana benthamiana confers Al tolerance, J. Exp. Bot. 57 (2006) 4235–4243. [24] J.-M. Lee, R.C. Gardner, Residues of the yeast ALR1 protein that are critical for magnesium uptake, Curr. Genet. 49 (2006) 7–20. [25] L.M. Kucharski, W.J. Lubbe, M.E. Maguire, Cation hexaammines are selective and potent inhibitors of the CorA magnesium transport system, J. Biol. Chem. 275 (2000) 16767–16773. [26] M. Kolisek, G. Zsurka, J. Samaj, J. Weghuber, R.J. Schweyen, M. Schweigel, Mrs2p is an essential component of the major electrophoretic Mg2+ influx system in mitochondria, EMBO J. 22 (2003) 1235–1244. [27] R. Schindl, J. Weghuber, C. Romanin, R.J. Schweyen, Mrs2p forms a high conductance Mg2+ selective channel in mitochondria, Biophys. J. 93 (2007) 3872–3883. [28] R. Takabatake, S. Hata, M. Taniguchi, H. Kouchi, T. Sugiyama, K. Izui, Isolation and characterization of cDNAs encoding mitochondrial phosphate transporters in soybean, maize, rice, and Arabidopsis, Plant Mol. Biol. 40 (1999) 479–486. [29] J.-L. Rigaud, D. Levy, Reconstitution of membrane proteins into liposomes, Methods Enzymol. 372 (2003) 65–86. [30] R.L. Smith, M.E. Maguire, Microbial magnesium transport: unusual transporters searching for identity, Mol. Microbiol. 28 (1998) 217–226. [31] D. Niegowski, S. Eshaghi, The CorA family: structure and function revisited, Cell. Mol. Life Sci. 64 (2007) 2564–2574. [32] Y. Xia, A.-K. Lundback, N. Sahaf, G. Nordlund, P. Brzezinski, S. Eshaghi, Co2 + selectivity of Thermotoga maritima CorA and its inability to regulate Mg2+ homeostasis present a new class of CorA proteins, J. Biol. Chem. 286 (2011) 16525–16532. [33] C.W. MacDiarmid, R.C. Gardner, Overexpression of the Saccharomyces cerevisiae magnesium transport system confers resistance to aluminum ion, J. Biol. Chem. 273 (1998) 1727–1732. [34] S. Ishijima, A. Uchibori, S. Yamashita, M. Ohnishi, Regulation of stromal fructose 1,6-bisphosphatase and ribulose 1,5-bisphosphate carboxylase activities by free Mg2+, J. Biol. Macromol. 1 (2001) 49–52.