The role of phytochelatins in arsenic tolerance in the hyperaccumulator Pteris vittata

Authors

  • F. J. Zhao,

    Corresponding author
    1. Agriculture and Environment Division, Rothamsted Research, Harpenden, Hertfordshire AL5 2JQ, UK;
      Author for correspondence: F. J. Zhao Tel: +44 1582763133 Fax: +44 1582760981 Email: Fangjie.Zhao@bbsrc.ac.uk;
    Search for more papers by this author
  • J. R. Wang,

    1. Agriculture and Environment Division, Rothamsted Research, Harpenden, Hertfordshire AL5 2JQ, UK;
    Search for more papers by this author
  • J. H. A. Barker,

    1. Crop Performance and Improvement Division, Long Ashton Research Station, Long Ashton, Bristol, BS41 9AF, UK; and
    Search for more papers by this author
  • H. Schat,

    1. Department of Ecology and Physiology of Plants, Faculty of Earth and Life Sciences, Vrije Universiteit, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands
    Search for more papers by this author
  • P. M. Bleeker,

    1. Department of Ecology and Physiology of Plants, Faculty of Earth and Life Sciences, Vrije Universiteit, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands
    Search for more papers by this author
  • S. P. McGrath

    1. Agriculture and Environment Division, Rothamsted Research, Harpenden, Hertfordshire AL5 2JQ, UK;
    Search for more papers by this author

Author for correspondence: F. J. Zhao Tel: +44 1582763133 Fax: +44 1582760981 Email: Fangjie.Zhao@bbsrc.ac.uk;

Summary

  • • Pteris vittata was the first identified arsenic (As) hyperaccumulator. Here we investigated whether phytochelatins (PCs) are involved in the hypertolerance of arsenic by P. vittata.
  • • P. vittata was exposed to 0–500 µm arsenate for 5 d, or to 50 µm arsenate for 0–7 d. In addition, l-buthionine-sulphoximine (BSO), an inhibitor of γ-glutamylcysteine synthetase, was used in combination with different arsenate exposures. The relationships between As accumulation and the concentrations of PCs and glutathione (GSH) were examined.
  • • PC synthesis was induced upon exposure to arsenate in P. vittata, with only PC2 detected in the plant. The As concentration correlated significantly with PC2 concentration in both roots and shoots, but not with GSH. The molar ratio of PC-SH to As was c. 0.09 and 0.03 for shoots and roots, respectively, suggesting that only a small proportion (1–3%) of the As in P. vittata can be complexed with PCs. In the presence of arsenate, addition of BSO decreased PC2 concentrations in roots and shoots by 89–96% and 30–33%, respectively. BSO alone was found to inhibit root growth of P. vittata markedly.
  • • The results suggest that PCs play a limited role in the hypertolerance of As in P. vittata.

Introduction

The brake fern Pteris vittata is the first identified arsenic (As) hyperaccumulator (Ma et al., 2001). It can accumulate > 20 000 mg As kg−1 in the shoot (frond) dry weight (d. wt). Furthermore, the bioconcentration factor, defined as the ratio of shoot As concentration to soil As concentration, was greater than 10. P. vittata is also hypertolerant to As, with a toxicity threshold value of about 10 000 mg As kg−1 d. wt in shoots (Tu & Ma, 2002; Wang et al., 2002). This compares with threshold values of 10–100 mg As kg−1 d. wt for As nontolerant plants (Kabata-Pendias & Pendias, 1992). More recently, several other fern species in the order Pteridales have been identified to be As hyperaccumulators (Visoottiviseth et al., 2002; Zhao et al., 2002). These plants may have potential for phytoremediation of As-contaminated soils or waters. Moreover, these plants offer an interesting model for studying As metabolism in plants.

Inorganic As species are generally highly phytotoxic. Arsenate is a phosphate analogue and can disrupt phosphate metabolism, whereas arsenite reacts with sulfydryl groups of enzymes and tissue proteins, leading to inhibition of cellular function and death (Meharg & Hartley-Whitaker, 2002). In As nonhyperaccumulating plant species, two mechanisms of As tolerance have been identified. First, tolerant plants have a decreased uptake of arsenate due to a suppression of the high-affinity phosphate uptake system (Meharg & Macnair, 1991, 1992; Meharg & Hartley-Whitaker, 2002). Second, arsenate taken up by plants is detoxified through reduction to arsenite, which is subsequently complexed with thiols, particularly phytochelatins (PCs). There are several lines of evidence for an important role of PCs in As tolerance. First, exposure to As induces a large response in the synthesis and accumulation of PCs in a number of plant species (Grill et al., 1987; Maitani et al., 1996; Sneller et al., 1999; Schmöger et al., 2000). Second, As-tolerant populations of Holcus lanatus were found to accumulate more PCs than nontolerant ones under the same degree of As stress (Hartley-Whitaker et al., 2001; Hartley-Whitaker et al., 2002). Third, inhibition of PC synthesis by a treatment with l-buthionine-sulphoxime (BSO), a potent inhibitor of γ-glutamylcysteine synthetase, leads to hypersensitivity to As in both As nontolerant and tolerant plants (Schmöger et al., 2000; Hartley-Whitaker et al., 2002; Schat et al., 2002). Fourth, an Arabidopsis thaliana mutant lacking functional PC synthase was significantly more sensitive to As toxicity than the wild type (Ha et al., 1999). Fifth, X-ray absorption spectroscopy reveals that the majority of As in the leaves of Brassica juncea and A. thaliana was coordinated to thiols (Pickering et al., 2000; Dhankher et al., 2002). Furthermore, intact PCs-As complexes have been isolated from plant tissues (Sneller et al., 1999; Schmöger et al., 2000).

The tolerance to As in P. vittata is at least an order of magnitude greater than in As nonhyperaccumulating plants. The mechanisms of As hypertolerance in P. vittata have not been elucidated. Arsenic extracted from the shoots of P. vittata appeared to be present mainly as inorganic arsenite (Francesconi et al., 2002; Wang et al., 2002). However, there is a possibility that the extraction procedure and subsequent speciation analysis may not preserve thiol-As complexes. Thus, the objective of this study was to investigate whether PCs are involved in the hypertolerance to As in the hyperaccumulator P. vittata.

Materials and Methods

Plant culture

Spores of P. vittata L. (produced from adult plants of the Florida accession, kindly provided initially by Dr Lena Ma, University of Florida, Gainesville, FL, USA) were germinated on moist potting compost. At the two-frond stage, sporelings were transplanted individually into 8-cm pots containing a general-purpose potting compost, and allowed to grow to the three–four frond stage. Plant roots were then washed carefully with tap water to remove adhering compost and transferred to 250-ml pots (one plant per pot) containing a nutrient solution, which was a modified Hoagland solution with 1/2 strength major nutrients and full strength micronutrients (except that Fe was supplied as Fe-EDTA at 100 µm) (Hewitt, 1966). Solution pH was buffered at around 6.0 using 5 mm MES (2-morpholino-ethanesulphonic acid). The nutrient solution was aerated continuously and renewed every 5 d. Plants were grown in the hydroponic culture for 3 wk before experiments started. The hydroponic culture and all experiments were conducted inside a controlled environment growth chamber with the following conditions: 16 h light period with a light intensity of 350 µmol m−2 s−1, 25°C/20°C day/night temperatures and 60% relative humidity.

Experiment 1

P. vittata plants were exposed to increasing concentrations of arsenate (0, 10, 50, 100 and 500 µm) for 5 d. Arsenate was supplied as Na2HAsO4, and the composition of nutrients was the same as in the preculture. Each treatment was replicated three-fold. After 5 d of exposure, plant roots and shoots were rinsed thoroughly with deionized water, blotted dry and separated. Roots and shoots were frozen in liquid nitrogen and freeze-dried at −60°C for 3 d. Samples were ground and stored in vials at −80°C until analysis.

Experiment 2

P. vittata plants were exposed to 50 µm arsenate (Na2HAsO4) for 0, 1, 3 and 7 d. This concentration was chosen because it was not toxic to P. vittata in preliminary experiments. The composition of nutrients was the same as in the preculture. On each sampling occasion, plants from three replicate pots were harvested and treated as described above.

Experiment 3

This experiment was conducted to investigate the effect of l-buthionine-sulphoximine (BSO) and arsenate exposure on root growth and PC synthesis. After 3 wk of preculture under hydroponic conditions, plants in 18 pots were treated with 500 µm BSO for 5 d. Another group of 18 plants was not treated with BSO. Thereafter, plants were exposed to 0, 500 and 1000 µm arsenate (Na2HAsO4), with or without 500 µm BSO, for a further 5 d. Each As and BSO treatment was replicated in six pots. Root elongation after 5 d exposure was measured in three pots of each treatment using the charcoal staining method (Schat & Ten Bookum, 1992). The remaining three pots for each treatment, unstained with charcoal, were used for the analysis of As and PCs. The plants were harvested and treated as described for Experiment 1.

Analysis

Ground plant materials were digested with a mixture of HNO3/HClO4 (85/15, v/v) and the concentrations of As were determined using inductively coupled plasma-atomic emission spectroscopy (ICP-AES, Fisons-ARL Accuris, Ecublens, Switzerland). Blanks and internal standards were included for quality assurance.

Freeze-dried plant material (20 mg) from Experiment 3 was extracted with 5 ml 30 mm ammonium phosphate buffer at pH 5.6 in a sonication bath for 30 min. The suspension was centrifuged and filtered through a Whatman No. 42 filter paper. As(III) and As(V) were determined in the extracts using automated hydride-generation atomic absorption spectrometry (Glaubig & Goldberg, 1988).

Extraction and analysis of PCs and glutathione (GSH) were performed using a slightly modified method described by Sneller et al. (2000). Freeze-dried plant material (10 mg) was ground in a mortar and pestle in 2 ml 6.3 mm DTPA (diethylenetriaminepentaacetic acid) with 0.1% (v/v) trifluoroacetic acid at 4°C. N-acetyl cysteine was added during grinding as an internal standard. The contents were centrifuged at 13 000 g for 10 min at 4°C and the supernatant filtered through a Costar Spin-X centrifuge tube with a 0.22-µm nylon filter. The thiols in the extract (250 µl) were derivatized with 10 µl 25 mm monobromobimane, together with 450 µl 200 mm 4-(2-hydroxyethyl)-piperazine-1-propanesulphonic acid buffer at pH 8.2 and 6.3 mm DTPA. Derivatization was carried out at 45°C for 30 min. The reaction was stopped by the addition of 300 µl of 1 m methanesulphonic acid. GSH and PCs were separated on two tandemly arranged Nova-Pak C18 columns (6 nm, 4 µm, 3.9 × 150 mm, Waters, Milford, MA, USA) at 37°C, using a slightly concave gradient of 12% to 25% (v/v) methanol for 15 min and then a linear gradient from 25% to 50% (v/v) methanol from 15 to 40 min. Fluorescence was monitored using a Waters 474 fluorescence detector. HPLC peaks were identified by mixing some of the P. vittata samples with a Silene vulgaris sample, which has a known PC composition established previously through amino acid analysis of the purified peak fraction. Reduced GSH was used as an external standard. PC concentrations were calculated as µmol GSH equivalent g−1 d. wt and corrected for derivatization efficiency (Sneller et al., 2000).

Analysis of variance was performed on all data. Least significant difference was used to compare treatment means.

Results

Experiment 1: effects of arsenate exposure concentration

There were no signs of phytotoxicity in P. vittata after 5 d exposure to arsenate up to 500 µm. The concentrations of As in both roots and shoots increased significantly (P < 0.05) with increasing As concentration in the nutrient solution (Fig. 1a). In the 100 and 500 µm As treatments, the concentration of As in the shoots exceeded 1000 µg g−1 d. wt (13.3 µmol g−1), which is the threshold value used to define As hyperaccumulation. Furthermore, the concentrations of As in the shoots were greater than those in the roots, indicating an efficient root to shoot translocation.

Figure 1.

Effects of arsenate exposure for 5 d on the concentrations of As (a), PCs (b) and GSH (c) in Pteris vittata. Bars represent ± SEs.

In the control treatment (no As), the concentrations of PCs in both roots and shoots were negligible. Arsenate exposure induced PC synthesis in both roots and shoots of P. vittata (P < 0.05; Fig. 1b). Only PC2 was detected in the plants. In the HPLC chromatograph (result not shown), PC2 in P. vittata coeluted exactly with the S. vulgaris PC2. Since the latter is a glycine-PC (Sneller et al., 2000), this result infers that the P. vittata PCs possess a glycine C-terminal. The concentrations of PC2 were 3–10 fold higher in the shoots than in the roots. The concentrations of glutathione were not significantly (P = 0.42 and 0.97 for shoot and root GSH, respectively) influenced by the As treatments (Fig. 1c).

Experiment 2: effects of arsenate exposure time

In this experiment, P. vittata was exposed to 50 µm arsenate for up to 7 d. Arsenic accumulation by P. vittata increased with exposure time (Fig. 2a). By day 3, the concentration of As in the shoots was more than double that in the roots. The synthesis of PC2 was induced by As exposure (Fig. 2b). In the shoots, PC2 concentration appeared to peak on day 3 and remained at a similar concentration on day 7. The concentration of PC2 in the roots was much smaller than that in the shoots. The concentration of GSH in the roots was not significantly (P = 0.64) influenced by the duration of As exposure (Fig. 2c). Shoot GSH level appeared to decrease during the first day of arsenate exposure and then recovered by day 3 (P = 0.048).

Figure 2.

Effects of arsenate exposure time on the concentrations of As (a), PCs (b) and GSH (c) in Pteris vittata. Bars represent ± SEs.

Fig. 3 shows the relationships between As, PC and GSH concentrations in roots and shoots of P. vittata. The data represent individual replicates for all treatments in Experiments 1 and 2. In both roots and shoots, PC concentration correlated positively and significantly (P < 0.001) with As concentration (Fig. 3a). The slopes, which represent the molar ratio of the PC-SH to As in the plants, were 0.09 and 0.03 for the shoot and root tissues, respectively. By contrast to PC, there was no significant correlation between GSH and As concentrations (Fig. 3b).

Figure 3.

Relationships between the concentrations of As and PCs (a) and GSH (b) in Pteris vittata. Data are from Experiments 1 and 2.

Experiment 3: effects of BSO

Without BSO in the nutrient solution, root elongation during the 5-d treatment period was not significantly affected by 500 µm arsenate, but was inhibited by 58% when 1000 µm arsenate was present (Fig. 4). Addition of 500 µm BSO inhibited root elongation by 73% in the absence of arsenate. In the presence of BSO, addition of 500 µm arsenate decreased root elongation by 50%, suggesting the possibility of an increased sensitivity of P. vittata to As toxicity. However, the severe effect of BSO on root elongation masked the effect of arsenate exposure. Phytotoxicity symptoms appeared in the shoots in the 1000 µm arsenate treatment, but the degree of phytotoxicity appeared similar with or without BSO.

Figure 4.

Effects of l-buthionine-sulphoximine (BSO) and arsenate exposure on root elongation in Pteris vittata. Bars represent ± SEs.

Fig. 5 shows the concentrations of As, GSH and PC2 in the roots and shoots of P. vittata in Experiment 3. The BSO treatment decreased shoot As concentration significantly (P < 0.05; Fig. 5d), but had no significant effect on root As concentration (Fig. 5a). The effect of BSO on the As accumulation in shoots may be a result of the inhibitory effect of BSO on root growth. Addition of BSO decreased the concentrations of GSH in the roots by 83% (P < 0.001, Fig. 5b). The BSO treatment also decreased the concentration of PC2 in the roots by 89–96% in the + As treatments (P < 0.001; Fig. 5c). In comparison, the effects of BSO on the concentrations of GSH and PC2 in the shoots were much smaller (30–43% reduction; P < 0.05 and P = 0.13 for GSH and PC2, respectively; Fig. 5e,f).

Figure 5.

Effects of arsenate and l-buthionine-sulphoximine (BSO) treatments on the concentrations of As, GSH and PCs in the roots (a–c) and shoots (d–f) of Pteris vittata. Bars represent ± SEs.

The root and shoot tissues from the As-exposed treatments were extracted with ammonium phosphate buffer at pH 5.7 for the determination of As speciation. This extraction procedure was found to extract on average 98% of the total As. In the shoot extracts, As(III) accounted for 85–90% of the total As, with the remaining proportion as As(V) (Table 1). In the roots, As(V) was the predominant species, accounting for 81–97% of the total As in the extracts, whilst As(III) represented only 3–19%. The BSO treatment decreased the proportion of As in the roots present as As(III) significantly (P < 0.001), but had no significant effect on As speciation in the shoots.

Table 1.  Arsenic speciation in the shoots and roots of Pteris vittata from Experiment 3
As treatment (µM)BSO treatment (µM)ShootRoot
Total As (µmol g−1 d. wt)*% as As(III)% as As(V)Total As (µmol g−1 d. wt)*% as As(III)% as As(V)
  • *

    Total As concentration extracted with ammonium phosphate buffer and determined by hydride-generation atomic absorption spectrometry (HG-AAS). Values are means ± SEs. BSO, L-buthionine-sulphoximine.

500  069.1 ± 14.584.7 ± 2.515.3 ± 2.536.8 ± 4.913.0 ± 2.087.0 ± 2.0
50040.3 ± 8.889.7 ± 1.010.3 ± 1.032.8 ± 7.4 4.2 ± 0.795.8 ± 0.7
1000  071.6 ± 16.989.9 ± 3.310.1 ± 3.334.2 ± 6.118.6 ± 3.481.4 ± 3.4
50040.0 ± 5.190.5 ± 2.7 9.5 ± 2.739.1 ± 12.0 3.2 ± 1.496.8 ± 1.4

Discussion

The results from all three experiments confirm the extraordinary ability of Pvittata to take up and transport As to the shoots. Wang et al. (2002) showed that arsenate is taken up via the phosphate uptake systems in P. vittata, similarly to As nonaccumulating plants (Asher & Reay, 1979; Ullrich-Eberius et al., 1989; Meharg & Hartley-Whitaker, 2002). Clearly, P. vittata does not employ the As resistance mechanism commonly found in many As tolerant plants, that is a decreased arsenate uptake through suppressed high affinity phosphate transporters (Meharg & Macnair, 1992; Meharg & Hartley-Whitaker, 2002). In agreement with previous studies (Ma et al., 2001; Francesconi et al., 2002; Wang et al., 2002), we found that As(III) was the predominant species of As in the shoots of P. vittata, whereas in the roots As was present mainly as As(V). Whether As(III) was present as aqueous arsenite species or as complexed arsenite cannot be determined by the speciation method used in this study. However, evidence from studies using more sophisticated methodologies, for example HPLC-ICP-MS and X-ray absorption spectroscopy, indicates that As(III) in the leaves of P. vittata is present primarily as aqueous arsenite species (Wang et al., 2002; Webb et al., 2003). The difference between roots and shoots in As speciation suggests that leaves may be the main location of arsenate reduction. Reduction of arsenate to arsenite is mediated by the arsenate reductase enzyme in a glutathione-dependent reaction in both Escherichia coli (Gladysheva et al., 1994) and Saccharomyces cerevisiae (Mukhopadhyay et al., 2000). A similar mechanism may operate in P. vittata, although this has not been established. Higher concentrations of GSH were found in the shoots of P. vittata than in the roots (Figs 1, 2, 3 and 5), which would favour arsenate reduction in the leaves. The reduction of arsenate to arsenite in roots appeared to be inhibited by the BSO treatment (Table 1), probably as a result of a marked decrease in the concentration of GSH in the BSO-treated roots.

Arsenite has a high affinity to thiols and complexation of arsenite by PCs may be expected. Exposure to As has been shown to elicit synthesis and accumulation of PCs in a number of plant species that are As nonhyperaccumulators (Grill et al., 1987; Maitani et al., 1996; Sneller et al., 1999; Schmöger et al., 2000; Hartley-Whitaker et al., 2002; Schat et al., 2002). These studies also demonstrate that PCs play an important, probably essential, role in the detoxification of arsenite in As nonhyperaccumulating plants. The results from this study show that exposure to arsenate also induced the synthesis of PC2 in the roots and shoots of the As hyperaccumulator P. vittata. The concentration of PC2 was higher in the shoots than in the roots of P. vittata, which may be related to a higher concentration of As in the shoots than in the roots, and the predominance of arsenite in the shoots in this plant. However, P. vittata differs from As nonhyperaccumulators in two aspects. First, only PC2 was found in P. vittata, whereas other plant species investigated so far also contain PCs of longer chain length. Second, the concentrations of PCs in P. vittata determined in this study (up to 7 µmol g−1 d. wt) are considerably lower than the values reported for Silene vulgaris (Sneller et al., 1999; Schat et al., 2002), Holcus lanatus (Hartley-Whitaker et al., 2001) and Rauvolfia serpentina cell cultures (Schmöger et al., 2000), under comparable concentrations of exposure to arsenate. In Experiment 3, 1000 µm arsenate was found to inhibit root elongation by more than 50%. However, the concentrations of PC2 in the 1000 µm arsenate treatment were similar to those in the 500 µm arsenate treatment, which did not inhibit root elongation. The results suggest that P. vittata has a rather limited capacity to accumulate PCs in response to As exposure.

In R. serpentia (Schmöger et al., 2000), S. vulgaris (Sneller et al., 1999; Schat et al., 2002) and in tolerant populations of H. lanatus (Hartley-Whitaker et al., 2002), the molar ratio of PCs-SH to As is ≥ 3 : 1. This ratio is consistent with the structural model of one arsenite being coordinated with three thiol groups, which was observed in a reconstituted As-PC2 complex using electrospray ionization mass spectroscopy (Schmöger et al., 2000). By contrast, we observed very low molar ratios of PCs-SH to As in the shoots (c. 0.09) and roots (c. 0.03) of P. vittata. These ratios imply that only a small proportion of the total As accumulated by the fern, c. 1% and 3% for roots and shoots, respectively, may be complexed with PCs, assuming a stoichiometry of three PCs-SH to one arsenite. These estimates are consistent with previous reports showing that inorganic arsenite was by far the most dominant species of As in the shoots of P. vittata (Francesconi et al., 2002; Wang et al., 2002). Furthermore, analysis using X-ray absorption spectroscopy showed that only a small proportion (c. 6%) of As in the leaves of P. vittata was coordinated to S (Webb et al., 2003). By contrast, almost all of the As in the shoots of Brassica juncea (Pickering et al., 2000) and Arabidopsis thaliana (Dhankher et al., 2002) was found to be complexed by thiols.

Schmöger et al. (2000) showed that an inhibition of GSH and PC synthesis by BSO led to an increased sensitivity of R. serpentia cell cultures to As toxicity. Even As-tolerant populations of H. lanatus, S. vulgaris and Agrostis castellana were rendered hypersensitive to As toxicity when BSO was added to the nutrient solutions (Hartley-Whitaker et al., 2002; Schat et al., 2002). In those studies, BSO alone had no toxic effect on root growth. In the present study, however, BSO (0.5 mm) alone was found to inhibit root growth of P. vittata markedly. Thus, it is difficult to judge whether a markedly decreased synthesis of GSH and PC as a result of the BSO treatment led to a higher sensitivity of P. vittata to As toxicity. The reason for the inhibitory effect of BSO on root growth of P. vittata remains unclear. One possible explanation is that a sufficient level of GSH is essential for root growth of P. vittata. An A. thaliana mutant (rml1) that lacks the first enzyme of the GSH biosynthesis pathway exhibits defective root cell division, because GSH is required for the phase transition of the cell cycle (Vernoux et al., 2000). These authors also found that the addition of 2.5 mm BSO to growth media completely inhibited root growth of wild type A. thaliana. Xiang et al. (2001) provided further evidence for a role of GSH in root growth. They showed that root growth in A. thaliana was decreased by 40% in the antisense plants with a low-level expression of γ-glutamylcysteine synthetase, compared with the wild type plants.

Because arsenite can be highly disruptive to metabolic processes in cytoplasm, it has to be detoxified. This may be achieved by complexation and/or vacuolar sequestration (Ghosh et al., 1999). In the leaves of P. vittata, As appears to be localized mainly in the vacuoles of epidermal cells (Lombi et al., 2002). Vacuolar sequestration is likely to be the main mechanism of As detoxification in this hyperaccumulator. Furthermore, the results from this study and other recent reports (Francesconi et al., 2002; Lombi et al., 2002; Wang et al., 2002; Webb et al., 2003) suggest that the bulk of the As that is sequestrated in the vacuoles is present as uncomplexed arsenite. What then is the role of PCs in the detoxification of As in P. vittata, considering the positive induction of PC synthesis by As exposure? It is possible that any arsenite present in the cytoplasm is complexed by PCs, thus preventing toxic effects. Cytosolic arsenite would only account for a small proportion of the total As accumulated by this species. It is not known whether PCs-As complexes, formed in the cytoplasm, are transported across tonoplast to the vacuoles. If so, PCs-As would have to dissociate in the vacuoles and PCs or their breakdown products efflux out of the vacuoles rapidly, to account for the large accumulation of As and the overall small ratios of PCs-SH to As observed. This scenario is not very likely, considering that PCs-arsenite complexes should be more stable in the vacuoles (pH c. 5.5) than in the cytoplasm (pH c. 7.5).

In conclusion, this study shows that, although arsenate exposure induces PC synthesis in P. vittata, the role of PCs in the detoxification of As appears to be limited. An analogy to our conclusion here is that PCs are not responsible for Cd hypertolerance in the hyperaccumulator Thlaspi caerulescens (Ebbs et al., 2002; Schat et al., 2002).

Acknowledgements

We thank Miss Sarah Dunham and Mr Adrian Crosland for technical assistance. J. Wang thanks the China Fellowship Council for a one-year fellowship. Rothamsted Research receives grant-aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom.

Ancillary