Characterization of phytochelatin synthase produced by the primitive red alga Cyanidioschyzon merolae

Abstract

Phytochelatins (PCs), non-proteinpeptides with the general structure [(γ-Glu-Cys)_n-Gly (n ≥ 2)], are involved in the detoxification of toxic heavy metals mainly in higher plants. The synthesis of the peptides is mediated by phytochelatin synthase (PCS), which is activated by a range of heavy metals. CmPCS, a PCS-like gene found in the genomic DNA of the primitive red alga Cyanidioschyzon merolae, was isolated and a recombinant protein (rCmPCS) fused with a hexahistidine tag at the N-terminus of CmPCS was produced. The finding that this protein mediated PC synthesis from glutathione in a metal-dependent way clearly establishes that rCmPCS is functional. The maximum activity was attained at a reaction temperature of 50 °C, considerably higher than the temperature required for the maximal activity of PCS isolated from the higher plant Silene cucubalus, probably due to the alga being a thermophile . CmPCS showed optimal pH in a slightly higher region than higher plant PCSs, probably due to the less effective charge relay network in the catalytic triad. In addition, the pattern of enzyme activation by metal ions was specific to rCmPCS, with Ag⁺, Cu²⁺, and Hg²⁺ showing only limited activation. In contrast to other eukaryotic PCSs, CmPCS has an extra domain in the N-terminal region from residues 1 to 109, and contains fewer cysteine residues in the C-terminal domain. These differences may be responsible for the metal specificity of the activation of CmPCS. Although the enzyme preparation lost PCS activity progressively when stored at 4 °C, the inclusion of Cd²⁺ in the preparation effectively prevented the reduction of activity. Furthermore, Cd²⁺ effectively restored the activity of the inactivated enzyme. These results indicate that Cd²⁺ ions bind the enzyme to maintain the structural integrity of the peptides .

---

Introduction

Exposure to toxic levels of heavy metals induces the production of non-proteinpeptides with the general structure (γ-Glu-Cys)_n-Gly, with n greater than or equal to 2.¹ The peptides were first discovered in the fission yeast Schizosaccharomyces pombe, and named cadystins.² The peptides have since been found in suspension-cultured cells of the higher plant Rauvolfia serpentine³ and collectively named phytochelatins (PCs). At present, PCs appear to be ubiquitous in the plant kingdom.⁴ The peptides seem to play a role in heavy metal tolerance by forming complexes with toxic heavy metals.

A phytochelatin synthase (PCS) that catalyzes PC synthesis was partially purified from S. cucubalus cells, which indicates that the peptides are synthesized through a γ-Glu-Cys dipeptidyl transfer reaction using the tripeptideglutathione (γ-Glu-Cys-Gly: GSH) or previously synthesized PC_n [(γ-Glu-Cys)_n-Gly, n ≥ 2] as substrates with heavy metal ions acting as activators.⁵ By a decade after the initial identification, genes encoding PCSs had been identified in the wheat Triticum aestivum, the fission yeast S. pombe, and Arabidopsis thaliana.^6–8 In addition to further identification of functional PCSs in higher plants,^9,10PCS expression has been found to extend to the animal kingdom. The nematode worm Caenorhabditis elegans possesses a gene with strong similarity to plant PCSs that also produces a functional product.¹¹

A recent genome sequencing project¹² revealed the presence of a PCS-like DNA sequence (CmPCS) in the red alga Cyanidioschyzon merolae (http://merolae.biol.s.u-tokyo.ac.jp/ or CMI111C in KEGG/GenomeNet [http://www.genome.jp/]). This alga, considered one of the most primitive algal species, grows in hot springs (45 °C) at a pH of 1.5.¹³ As the bioavailability of metal ions is increased in such habitats, CmPCS is anticipated to play a role in detoxification. In fact, heterologous expression of CmPCS in the Cd²⁺-sensitive Saccharomyces cerevisiae strain ycf1Δ rescued the yeast from Cd²⁺ stress due to the production of PC_n, indicating that CmPCS is functional.¹⁴ As analysis of the C. merolae genomic sequence indicated that this alga and Arabidopsis originated from a common ancestral organism,¹³ functional characterization of CmPCS may shed light on the evolution of this protein. For this purpose, recombinant CmPCS (rCmPCS) containing a hexahistidine tag at its N-terminal end was purified and its enzymatic properties were determined.

Experimental

Expression and purification of rCmPCS

The strain C. merolae 10 D was obtained from the microbial culture collection of the National Institute for Environmental Studies (Tsukuba, Japan). Cells were grown in modified Allen’s medium containing 40 mM (NH₄)₂SO₄, 4 mM MgSO₄·7H₂O, 4 mM KH₂PO₄, 1 mM CaCl₂, 0.1 mM Fe-ethylenediaminetetraacetic acid, 100 μM H₃BO₃, 30 μM MnCl₂·4H₂O, 1.5 μM ZnCl₂, 4.5 μM Na₂MoO₄, 0.6 μM CoCl₂·6H₂O, and 0.6 μM CuCl₂, pH adjusted to 2.8. The cells were cultured phototrophically with continuous illumination of 32 μmol photon m⁻² s⁻¹ at 37 °C and continuous shaking at 100 rpm.

Genomic DNA was extracted from a 30-mL C. merolae cell culture incubated for 7 days at 37 °C. The CmPCS sequence was amplified with a KOD-plus PCR kit (Toyobo, Osaka, Japan) with the primers 5′-GAGACATATGATCATTTGGCGTCCGTTGGTACGTTGCGGTTACCACG-3′ (forward primer, NdeI site in bold) and 5′-GAGAGAATTCCTACTTGTTTGCTCCTGTTGCTGGTGCTTCCGCTTGCTGTGCG-3′ (reverse primer, EcoRI in bold) using genomic DNA as the template. The resulting fragment was ligated into pColdII (Takara, Osaka, Japan) such that the resulting plasmid could express an N-terminally hexahistidine-tagged enzyme (rCmPCS), and transformed into the Escherichia coli strain XL-1 Blue (Stratagene, La Jolla, CA, USA). The construct was confirmed by sequencing with the T7 primer, and transformed into the E. coli strain BL21(DE) (Novagen, San Diego, CA, USA). The transformed cells were precultured in Luria-Bertani broth and then cultured in 1 l TB medium (Novagen) at 15 °C for 50 h.

E. coli cells expressing rCmPCS were collected by centrifugation at 8000 × g for 15 min and suspended in an extraction solution containing 1 mM 2-mercaptethanol and 10% glycerol buffered to pH 6.8 with 10 mM Tris/HCl. After disruption of the cells by sonication, the supernatant was obtained by centrifugation at 4000 × g for 15 min.

The recombinant protein was purified by Ni²⁺-affinity chromatography using FPLC (Pharmacia LKB, Uppsala, Sweden). A 5-ml HisTrap HP column (GE Healthcare Life Sciences, Piscataway, NJ, USA) was equilibrated with a binding buffer containing 20 mM HEPES/NaOH, pH 7.4, 500 mM NaCl, and 20 mM imidazole. The supernatant (10 ml) was applied to the column and washed with 10 ml of binding buffer. Then, rCmPCS was eluted with elution buffer containing 20 mM HEPES/NaOH, pH 7.4, 500 mM NaCl, and 500 mM imidazole. The homogeneity of the PCS preparation was verified using SDS-PAGE.

Determination of the PCS activity of rCmPCS

To a 100-μl reaction solution containing 0.5 mM CdSO₄, 10 mM GSH, and 10 mM 2-mercaptoethanol buffered at pH 8.0 was added 0.8 μg of rCmPCS. The solution was incubated at 40 °C for 10 min and the reaction was terminated by supplementation with 3.6 M HCl (125 μl). The PC_n concentrations were determined using a newly developed method¹⁵ based on dequenching of the Cu(I)-bathocuproine disulfonate (BCS) complex by substances capable of chelating Cu(I), such as PC_n,¹⁶ as a consequence of release of the fluorescent BCS molecule due to competition of Cu(I) chelators with BCS for Cu(I). The peptides were separated on an octadecylsilane column (7.8 mm i.d. × 150 mm, TSK-GEL ODS-80™; Tosoh, Tokyo, Japan). After equilibration with a solution containing 5% acetonitrile and 0.1% trifluoroacetic acid (TFA), aliquots of 20 μl were injected and run in a slightly concave gradient from 5% to 40% (v/v) acetonitrile with 0.1% TFA over 50 min. The eluent was merged with a post-column solution containing 0.5 μM CuSO₄, 1.1 μM BCS, and 10 μM ascorbic acid buffered at pH 10.0 with 50 mM CHES/NaOH. After the merged solution was mixed through a mixing coil, the fluorescence intensity at 770 nm with an excitation wavelength of 580 nm was monitored using an FP-920 fluorescence detector (Jasco, Tokyo, Japan). The peak area was used for quantification of PC_n. PCS activity was expressed as μmol of PC₂ synthesized per min per mg of protein.

Results

Dependence of PCS activity on the reaction conditions

rCmPCS, an N-terminally hexahistidine-tagged CmPCS, was overexpressed in E. coli BL21(DE) and purified using a Ni²⁺ metal affinity column. SDS-PAGE of the enzyme preparation showed a single band of 60 kDa. Gel filtration chromatography resulted in a peak at 55 kDa, indicating that rCmPCS is a monomeric protein. Typically, 0.47 mg of rCmPCS was obtained from 1 l of culture.

The enzyme activity of rCmPCS increased with incubation temperature, peaking at 50 °C and decreasing thereafter (Fig. 1A). An Arrhenius plot demonstrated that the value of the logarithm of enzyme activity leveled off gradually as the temperature increased from 25 °C to 50 °C (a decrease in the reciprocal of the temperature from 0.00336 to 0.00310), indicating that a considerable portion of the enzyme activity was lost at 50 °C (Fig. 1B). Therefore, although the maximum enzyme activity was observed at 50 °C, the enzyme reaction was performed at 40 °C to minimize the inactivation of rCmPCS in subsequent experiments.

Fig. 1 PCS activity of rCmPCS as a function of reaction temperature. The reaction showed maximal activity at 50 °C (A). The Arrhenius plot shows how the value of the logarithm of enzyme activity leveled off gradually as the temperature increased from 25 °C to 50 °C (a decrease in reciprocal of the temperature from 0.00336 to 0.00310), indicating that a considerable fraction of the enzyme activity was lost at 50 °C (B). PCS activity is expressed as μmol PC₂ synthesized per min per mg of protein.

Fig. 2 shows the effect of the pH of the reaction solution on rCmPCS activity. The buffer solutions employed were MES/NaOH (pH 5.5–6.5), HEPES/NaOH (pH 6.5–8.5), and CHES/NaOH (pH 8.5–9.5). It is likely that the differences in the effects of the buffers may be neglected because only negligible differences in activity were observed when two types of buffer were used (pH 6.5 and 8.5). The activity increased from almost null at pH 5.5 to the maximum value at pH 9.0, after which it dropped sharply at pH 9.5. A pH of 8.0 was employed thereafter to preclude the possible precipitation of metal ions due to the formation of metal hydroxides at pH values higher than 8.0.

Fig. 2 Effects of reaction pH on the PCS activity of rCmPCS. The buffers used were MES/NaOH (pH 5.5–6.5; closed squares), HEPES/NaOH (pH 6.5–8.5; closed circles), and CHES/NaOH (pH 8.5–9.5; closed triangles). PCS activity is expressed as μmol PC₂ synthesized per min per mg of protein.

The activation of rCmPCS by various metals and metalloids is shown in Table 1. Cd²⁺ was the most prominent activator of rCmPCS among the metals and metalloids employed, followed by Pb²⁺, As(V), As(III), Ag⁺, Cu²⁺, and Hg²⁺. No activity was detected in the presence of Ca²⁺, Co²⁺, Fe²⁺, Mg²⁺, Mn²⁺, Ni²⁺, Sb³⁺, or Zn²⁺.

Table 1 Activation of rCmPCS by various metals

Ion	Relative activity^a	Ion	Relative activity
a The activity was normalized to that of Cd^2+ (1.2 μmol PC2 synthesized min^−1 mg protein^−1). b No detectable activity.
Cd^2+	100	Co^2+	N.D.
Pb^2+	73.6	Fe^2+	N.D.
As(V)	55.5	Mg^2+	N.D.
As(III)	43.6	Mn^2+	N.D.
Ag^+	6.35	Ni^2+	N.D.
Cu^2+	5.83	Sb^3+	N.D.
Hg^2+	2.9	Zn^2+	N.D.
Ca^2+	N.D.^b	None	N.D.

---

Effects of Cd²⁺ on stability and activity restoration of rCmPCS

Although rCmPCS is stable when stored at −80 °C, the PCS activity of the enzyme preparation (32 μg ml⁻¹) diminished gradually at 4 °C, with a half-life of 4.5 days (Fig. 3). Supplementation with Cd²⁺ ions enhanced the stability. Cd²⁺ slightly prolonged the half-life to 4.8 days at a concentration of 1 μM and significantly increased the half-life to 8.2 days at a concentration of 10 μM (Fig. 3). Furthermore, the activity increased rather than decreased when 100 μM Cd²⁺ was used to maintain stability throughout the period examined, with an average activity equal to the initial activity of 140 ± 9%.

Fig. 3 Changes in PCS activity of rCmPCS (32 μg ml⁻¹) stored at 4 °C in the presence of various concentrations of Cd²⁺. rCmPCS gradually lost its activity with a half-life of 4.5 days (closed squares). Supplementation with Cd²⁺ improved the stability of the enzyme with half-lives of 4.8 and 8.2 days at concentrations of 1 μM (closed circles) and 10 μM (open squares), respectively. Cd²⁺ at a concentration of 100 μM enhanced the activity by a factor of 1.4 (open circles). PCS activity is expressed as μmol PC₂ synthesized per min per mg of protein.

The effects of Cd²⁺ ions on the restoration of PCS activity were examined for rCmPCS that had been stored for 9 days at 4 °C without any additional Cd²⁺. Supplementation with Cd²⁺ at 100 μM markedly restored the activity, in contrast to the enzyme preparation with no additional Cd²⁺, in which the activity decreased progressively (Fig. 4).

Fig. 4 Restoration of the PCS activity of rCmPCS in the presence of added Cd²⁺. An enzyme preparation that had been stored at 4 °C for 9 days and exhibited decreased activity (closed squares in Fig. 3) was incubated in the presence (squares) or absence (circles) of 100 μM Cd²⁺. Incubation of the enzyme preparation with Cd²⁺ for 10 h restored the activity to 67% that of the initial preparation. No such restoration was observed for the enzyme preparation incubated without Cd²⁺. The PCS activity is expressed as μmol PC₂ synthesized per min per mg of protein.

Discussion

Detailed analyses of the enzymatic mechanisms of Arabidopsis thalianaPCS (AtPCS1) have established that it is a bisubstrate-substituted mechanism, forming a γ-Glu-Cys acyl intermediate in a metal-independent manner.¹⁷ A catalytic triad, Cys-56, His-162, and Asp-180, has been identified in AtPCS1 (Fig. 5), with the thiol group of Cys-56 acting as the acylation site.¹⁸ The primitive red alga C. merolae possesses a PCS-like gene. Amino acid sequence alignment indicated that CmPCS also has a catalytic triad equivalent to that of AtPCS1 (Cys-164, His-283, and Asp-301; Fig. 5). The expression of CmPCS in a Cd²⁺-sensitive S. cerevisiae strain rescued the yeast from Cd²⁺ stress with concomitant synthesis of PC_n.¹⁴ These findings suggest the involvement of CmPCS in PC synthesis. It is likely that the catalytic triad in CmPCS also plays a role in PC synthesis in a manner similar to the PCSs identified to date. However, it is possible that PC_n are synthesized through other routes. S. cerevisiae synthesized low levels of PC₂ on exposure to Cd²⁺, although the synthesis was shown to be mediated by carboxypeptidases. However, an in vitro experiment showed that the enzymes were not activated by Cd²⁺,¹⁹ in contrast to the PCS-dependent PC synthesis in which the reaction itself is activated by the presence of heavy metals.

Fig. 5 Amino acid sequence alignment of PCSs from C. merolae (CmPCS, CMI111C in KEGG/GenomeNet [http://www.genome.jp/]), A. thaliana (AtPCS1, AF135155), and T. aestivum (TaPCS, AF093752). Asterisks indicate the residues in the catalytic triad. Amino acids conserved among the three proteins are highlighted in black and those conserved in two of the proteins are indicated in gray. The alignment was generated using CLUSTAL X.

rCmPCS, which possesses a hexahistidine tag at its N-terminus, was overexpressed in E. coli cells and purified by Ni²⁺ metal affinity chromatography. Gel filtration chromatography together with SDS-PAGE analysis of the enzyme preparation demonstrated the monomeric nature of the protein, with a molecular mass close to the calculated value of 61.5 kDa. Supplementation of the protein into GSH solution induced PC synthesis in the presence of Cd²⁺ but not in the absence of Cd²⁺ (Table 1). These observations clearly established that rCmPCS is functional. This is the first demonstration of functional algal PCS. The temperature at which the reaction rate was highest was 50 °C (Fig. 1), with a significantly higher value than the PCS from suspension-cultured cells of the higher plant S. cucubalus at 35 °C (ScPCS).⁵ It is reasonable to assume that rCmPCS possesses thermostability, as C. merolae is a thermophilic alga.

rCmPCS is functional over a narrow pH range, with maximum activity at pH 9.0 and half-maximal activities at pH 7.6 and 9.4 (Fig. 2). Although the maximum activity was achieved at a slightly lower pH value of 7.9, a similar narrow PCS activity profile with varying pH values was seen for ScPCS, with half-maximal activities at pH 7.4 and 8.7.⁵ In addition, the PCS activity of a crude extract of suspension-cultured tobacco cells was maximal at pH 8.0, with half-maximal activities at pH 7.2 and 8.6.²⁰ Therefore, the activities of PCSs from higher plants and algal species seem to be confined to a narrow pH range. In contrast, rCmPCS has a functional pH range higher than those of higher plant PCSs. The initial step of PC synthesis is acylation of the cysteinylthiol of the catalytic triad by a γ-Glu-Cys group, where the other amino acid residues of the catalytic triad (His and Asp) may play a pivotal role such that it enhances the nucleophilicity of the thiol function to facilitate acylation. In the case of AtPCS1, the nucleophilicity of the thiol function of Cys-56 may be enhanced by transfer of this thiol proton to the imidazole ring of His-162, the electrophilicity of which is augmented through β-carboxylate on Asp-180 by withdrawing the histidyl proton or by stabilizing the positively charged form of the histidyl residue.²¹ Although CmPCS also possesses a catalytic triad (Fig. 5), the charge relay network may be less effective in this protein than in those of higher plants. This may cause the functional pH to shift toward higher values. PCSs of higher plants may have evolved to adopt a configuration of the triad amino acid residues with an effective charge relay network.

rCmPCS is activated most prominently by Cd²⁺, followed by Pb²⁺, As(V), and As(III), whereas Ag⁺, Cu²⁺, and Hg²⁺ showed very low levels of activation (Table 1). This pattern of PCS activity, which is dependent on various metal ions, is similar to that observed for AtPCS1,⁹ except for the low activation by Ag⁺, Cu²⁺, and Hg²⁺. The catalytic triad has been shown to participate in the core reaction step of the PCS, whereas the mechanisms for metal-dependent activation are unclear. An AtPCS1 with a truncated C-terminal domain showed reduced enzymatic activity¹⁸ and altered dependency on metal ions for activation.²² In addition, although the C-terminal halves are divergent in sequence among the PCSs, the domains are rich in cysteine residues that form several motifs characteristic of metal-binding sites, such as Cys-Xaa-Yaa-Cys, Cys-Xaa-Cys, and Cys-Cys, where Xaa and Yaa are non-cysteine amino acids (Table 2). Moreover, NsPCS, the product of a PCS-like gene found in Nostoc sp. strain PCC7120, is half the length of AtPCS and shows similarity to the N-terminal domain of the eukaryotic PCSs.²³ Although NsPCS meets the requirement for γ-Glu-Cys acylation by containing the catalytic triad residues (Cys-70, His-183, and Asp-201), it does not synthesize PCs. This protein exclusively catalyzes the hydrolysis of GSH, with no appreciable²⁴ or very small²⁵ amounts of PC synthesis in a metal-independent manner. These results demonstrate the significance of the C-terminal domain in sensing metal ions to activate AtPCS. Based on the alignment of AtPCS, TaPCS, and CmPCS, CmPCS is unique in that it possesses an extra domain at the N-terminal portion (residues 1–109; Fig. 5), where all cysteine residues in this portion reside randomly. In addition, CmPCS has a relatively small number of cysteine residues in the C-terminal domain, containing five cysteine residues, two of which consist of only one Cys-Xaa-Cys motif (Table 2). These structural peculiarities of rCmPCS may result in its selective metal-sensing mechanisms that differ from those of other PCSs. C. merolae is considered one of the most primitive algal species. Again, it has been suggested that this alga and Arabidopsis originated from a common ancestor.¹³ In this context, the difference in the metal-activating patterns between rCmPCS and AtPCS1 may reflect evolutionary changes, although more data on algal PCS are required.

Table 2 Number of motifs containing cysteine residues in PCSs

Domain	Motif
	Cys-Xaa-Yaa-Cys	Cys-Xaa-Cys	Cys-Cys	Single Cys
a The extra portion of CmPCS (residues 1–109). b The C-terminal domain of CmPCS (residues 342–560). c The C-terminal domain of AtPCS1 (residues 230–485). d The C-terminal domain of TaPCS (residues 230–500).
CmPCS-X^a	0	0	0	5
CmPCS-C^b	0	1	0	3
AtPCS1-C^c	1	1	2	2
TaPCS-C^d	2	0	2	6

---

rCmPCS was inactivated gradually when stored at 4 °C, but supplementation with Cd²⁺ suppressed the inactivation in a dose-dependent manner (Fig. 3). These findings indicate that rCmPCS has at least two states, active and inactive. These states may differ only in polypeptide folding, as they can be converted easily from one to the other. rCmPCS may also tend to be in an inactive form in the absence of Cd²⁺. Cd²⁺ ion(s) hinder this process by binding to sites on the enzyme and bundling the polypeptide chains, which may maintain the integrity of the protein structure. Moreover, the addition of 100 μM Cd²⁺ restored PCS activity in an inactivated enzyme preparation (Fig. 4). Cd²⁺ ions may bind to sites on the denatured enzyme molecule, resulting in the proper rearrangement of the polypeptide by gathering the sites on the protein molecule.

List of abbreviations

BCS	bathocuproine disulfonate
GSH	glutathione
PC	phytochelatin
PC2	(γ-Glu-Cys)2-Gly
PCS	phytochelatin synthase
CmPCS	Cyanidioschyzon merolae PCS
rCmPCS	recombinant CmPCS
ScPCS	Silene cucubalus PCS
TaPCS	Triticum aestivum PCS
TFA	trifluoroacetic acid

References

1. M. H. Zenk, Gene, 1996, 179, 21–30.
2. N. Kondo, K. Imai, M. Isobe, T. Goto, A. Murasugi, C. Wada-Nakagawa and Y. Hayashi, Tetrahedron Lett., 1984, 25, 3869–3872.
3. E. Grill, E.-L. Winnacker and M. Zenk, Science, 1985, 230, 674–676.
4. M. H. Zenk, Gene, 1996, 179, 21–30.
5. E. Grill, S. Löffler, E.-L. Winnacker and M. H. Zenk, Proc. Natl. Acad. Sci. U. S. A., 1989, 86, 6838–6842.
6. S. Clemens, E. J. Kim, D. Neumann and J. I. Schroeder, EMBO J., 1999, 18, 3325–3333.
7. S. B. Ha, A. P. Smith, R. Howden, W. M. Dietrich, S. Bugg, M. J. O’Connell, P. G. Goldsbrough and C. S. Cobbett, Plant Cell, 1999, 11, 1153–1163.
8. O. K. Vatamaniuk, S. Mari, Y. P. Lu and P. A. Rea, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 7110–7115.
9. M. Oven, J. E. Page, M. H. Zenk and T. M. Kutchan, J. Biol. Chem., 2002, 277, 4747–4754.
10. S. Heiss, A. Wachter, J. Bogs, C. Cobbett and T. Rausch, J. Exp. Bot., 2003, 54, 1833–1839.
11. S. Clemens, J. I. Schroeder and T. Denkolb, Eur. J. Biochem., 2001, 268, 3640–3643.
12. M. Matsuzaki, O. Misumi, T. Shin-i, S. Maruyama, M. Takahara, S. Miyagishima, T. Mori, K. Nishida, F. Yagisawa, K. Nishida, Y. Yoshida, Y. Nishimura, S. Nakao, T. Kobayashi, Y. Momoyama, T. Higashiyama, A. Minoda, M. Sano, H. Nomoto, K. Oishi, H. Hayashi, F. Ohta, S. Nishizaka, S. Haga, S. Miura, T. Morishita, Y. Kabeya, K. Terasawa, Y. Suzuku, Y. Ishii, S. Asakawa, H. Takano, N. Ohta, H. Kuroiwa, K. Tanaka, N. Shimizu, S. Sugano, N. Sato, H. Nozaki, N. Ogasawara, Y. Kohara and T. Kuroiwa, Nature, 2004, 428, 653–657.
13. A. R. Grossman, Plant Physiol., 2005, 137, 410–427.
14. Y. Osaki, T. Shirabe, S. Tamura and E. Yoshimura, Biosci., Biotechnol., Biochem., 2008 DOI:10.1271/bbb.80505 , in press, http://www.jstage.jst.go.jp/article/bbb/advpub/0/0811061154/_pdf/-char/ja/.
15. T. Shirabe, K. Ito and E. Yoshimura, Anal. Chem., 2008, 80, 9360–9362.
16. R. N. Reese, R. K. Mehra, E. B. Tarbet and D. R. Winge, J. Biol. Chem., 1988, 263, 4186–4192.
17. O. K. Vatamaniuk, S. Mari, A. Lang, S. Chalasani, L. O. Demkiv and P. A. Rea, J. Biol. Chem., 2004, 279, 22449–22460.
18. N. D. Romanyuk, D. J. Rigden, O. K. Vatamaniuk, A. Lang, R. E. Cahoon, J. M. Jez and P. A. Rea, Plant Physiol., 2006, 141, 858–869.
19. J. Wünschmann, A. Beck, L. Meyer, T. Letzel, E. Grill and K. J. Lendzian, FEBS Lett., 2007, 581, 1681–1687.
20. R. Nakazawa, H. Kato, Y. Kameda and H. Takenaga, Biol. Plant., 2002, 45, 311–313.
21. P. A. Rea, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 507–508.
22. R. Ruotolo, A. Peracchi, A. Bolchi, G. Infusini, A. Amoresano and S. Ottonello, J. Biol. Chem., 2004, 279, 14686–14693.
23. S. Clemens, J. Plant Physiol., 2006, 163, 319–332.
24. E. Harada, E. von Roepenack-Lahaye and S. A. Clemens, Phytochemistry, 2004, 65, 3179–3185.
25. N. Tsuji, S. Nishikori, O. Iwabe, S. Matsumoto, K. Shiraki, H. Miyasaka, M. Takagi, K. Miyamoto and K. Hirata, Planta, 2005, 222, 181–191.

---