Studies thus far suggest that metal sequestration in most hyper accumulators mainly occurs in non-photosynthetic cells of the epidermis and surface structures, namely trichomes, as a mechanism to reduce accumulation in the metal sensitive photosynthetic apparatus . In S. alfredii however, Tian et al. reported that Cd concentrations were highest in stem parenchyma and leaf mesophyll tissues, suggesting that compartmentalization of Cd in parenchyma cells may be important for the hyper accumulation and detoxification of Cd. A better understanding of the mechanisms involved in the uptake and storage of Cd by the parenchyma cells is therefore necessary to elucidate the physiological mechanisms of Cd hyper accumulation in S.alfredii and to provide insights into the diversity of strategies present in Cd hyper accumulators. The study of the uptake and sequestration of metals into plant cells is much more difficult to carry out than in animal and bacterial cells, which are devoid of cell walls . Hence there is only limited information available on absorption and sequestration of heavy metals into specific storage cells of hyper accumulators. The aim of the present study is to understand the characteristics of cellular uptake and sequestration of Cd into the terminal storage sites of the Cd hyper accumulator S. alfredii in comparison with its non-hyper accumulating ecotype . To reach this goal, we investigated: the sequestration kinetics of Cd by plant cells in aerial parts of HE S.alfredii using micro X-ray fluorescence , which is a powerful tool for spatial elemental imaging in biological systems ; Cd accumulation in individual leaf cells, namely protoplasts, isolated from the two contrasting ecotypes; Cd tolerance in the two kinds of leaf cells in terms of cell viability and membrane integrity, as well as hydrogen peroxide production,berry pots under Cd stress.Plants of intact 4-week old seedlings of HE and NHE S.alfredii were selected for mesophyll protoplast isolation. Protoplasts were prepared based on the method developed by Ma et al. and Robert et al. , with some modification. Briefly, uniform leaves were selected and peeled to remove abaxial sides.

They were then sliced into 1 to 2 mm pieces and suspended in 30 ml cell digesting medium composed of 1.5% cellulase, 0.4% macerozyme, 0.4 M mannitol, 20 mM KCl, 20 mM MES and 10 mM CaCl2. After collection by vacuum filtration, the suspension was shaken at 60 rmp at 28 o C for 2 h. Afterwards, the suspension was filtered through a 75 μm cell strainer and rinsed twice with W5 buffer, which comprised 154 mM NaCl, 125 mM CaCl2, 5 mM KCl and 2 mM MES at pH 5.8. The collected protoplasts were then centrifuged at 80 g for 20 min in a J2-HS centrifuge . The pellets were suspended in 30 ml W5 buffer and centrifuged twice under the same conditions to eliminate the enzymes. Isolated mesophyll protoplasts were suspended in W5 buffer for use in the following experiments.Protoplast density was measured via a viability test using fluorescein diacetate dye with a hemocytometer under fluorescence microscopy . The viable protoplasts were spheroid and showed bright green fluorescence, whereas dead protoplasts were ruptured. A 3 ml protoplast solution was suspended in W5 buffer, supplemented with or without Cd. At each time interval , a 10 μl suspension was collected for viability analysis. The percentage viability in the Cd exposed samples was calculated by normalization to the number of viable protoplasts from the control.leaves were incubated in W5 buffer. To protect the protoplasts from light, the samples were kept in the dark before the following experiments. An aliquot of a concentrated solution of CdCl2 was added to the W5 buffer to achieve a final Cd concentration of 10 μM. At each time interval , the solution was filtered through a 10 μm filter membrane, with the protoplasts retained on the membrane due to their diameter being larger than the aperture of filter membrane. The protoplasts were then gently rinsed three times with 4 ml W5 buffer to remove any Cd2+ on their surfaces. Each treatment was replicated four times. The protoplasts on the filter membranes were broken and dissolved using 15 mL deionized water and then filtered through a 0.45 μm membrane. The experiments were conducted at two different temperatures, 25 o C and 4 °C. The concentrations of Cd in the filtrates were analysed using inductively coupled plasma mass spectroscopy.

The integrity of mesophyll protoplasts was determined as described above at each time interval, in order to precisely calculate the Cd accumulation rate in the protoplasts.Mesophyll protoplasts isolated from young leaves of the two S. alfredii ecotypes were compared in regards to the concentration-dependent kinetics of Cd accumulation. Collected protoplasts were incubated in W5 buffer in the dark and then supplemented with a series of concentrations of different CdCl2, ranging from 0 to15 μM, at two different temperatures, 25 °C and 4 °C.After 30 min for uptake, Cd concentrations in the protoplasts were analyzed as described above, with each treatment replicated four times. The Cd accumulation rate for each replicate was calculated using Cd concentration and protoplast integrity data.The Cd-specific probe LeadmiumTM Green AM dye was used to investigate Cd localization in protoplasts. Protoplasts were incubated in W5 buffer with the addition of 0.4 μg mL−1 Leadmium TM Green AM stock solution. The stock solution of LeadmiumTM Green AM was prepared by adding 50 μl of dimethyl sulphoxide to one vial of the dye and then diluting this with 1:10 of 0.85% NaCl . After incubation for 30 min in the dark on a shaker set to 60 r min-1, the protoplasts were washed four times with fresh W5 buffer to remove extra Leadmium TM Green AM dye. In each washing step, protoplasts were centrifuged at 50 g for 1 min and the supernatant replaced by an equal volume of fresh W5 buffer. After the last wash, the protoplasts were carefully mixed with W5 buffer, with addition of 10 µM or 200 µM Cd. At each time interval , the fluorescence of Cd in protoplasts was detected under a fluorescence microscope using filters S450-490 for excitation and S505-520 for emission. For concentration-dependent imaging of Cd, the protoplasts were treated with different amounts of Cd for 90 min and then imaged using fluorescence microscopy, as described above.The detection of H2O2 in protoplasts in vivo was carried out using CM-H2DCFDA , according to the methods of Zhang et al. , with slight modification. The protoplasts were suspended in W5 buffer with or without 200 μM Cd. At each time interval , a 200 μl solution was collected and rinsed in W5 buffer three times to remove any Cd2+ on the protoplast surfaces.

This solution was then incubated for 10 min in the dark with the addition of CM-H2DCFDA at a final concentration of 5 μM. As negative controls, protoplasts were incubated for 30 min with a H2O2 scavenger, namely 1.0 mM ascorbate , before staining with the fluorescent dyes . H2O2 fluorescence was visualized using a 45 μl sample of the solution placed in single concave slide, using a fluorescence microscope ,hydroponic growing with excitation at 450-490 nm and emission at 505-520 nm. The relative fluorescence units were converted to percentages by normalization to the H2O2 fluorescence of the control.Mesophyll protoplasts were successfully isolated from young leaves of HE and NHE S. alfredii plants . The protoplasts were exposed to uptake solutions containing 10 μM Cd for the given time-course for accumulation of the metal. The 2 h uptake period showed no significant difference in terms of the Cd accumulation rate between the protoplasts of the two ecotypes . Cd accumulation in the mesophyll protoplasts of both ecotypes was more or less linear within the 2 h window of exposure to 10 μM Cd . Ice-cold treatment at 4 °C significantly decreased the cumulative accumulation of Cd in both HE and NHE mesophyll protoplasts within the 2 h uptake period . Concentration-dependent Cd accumulation kinetics in mesophyll protoplasts were further investigated for the two S. alfredii ecotypes. At low Cd levels in solution, concentration-dependent Cd accumulation kinetics in mesophyll protoplasts from the both ecotypes were characterized by non-saturating curves, with the accumulation rate slightly but not significantly higher in HE protoplasts . As with experiments into the time-dependent kinetics of Cd accumulation , treatments at a low temperature of 4°C significantly inhibited Cd uptake by mesophyll protoplasts of both S. alfredii ecotypes, regardless of Cd levels in the uptake solutions. Fluorescence imaging of Cd in protoplasts revealed differences between the two S.alfredii ecotypes. The Cd probe, LeadmiumTM Green AM dye, was successfully loaded into the mesophyll protoplasts of both ecotypes treated with Cd, showing clear bright green fluorescence, with a very weak signal in controls . After treatment with 10 µM Cd for 60–120 min, some mesophyll protoplasts collected from HEs were filled with Cd, as indicated by the clear bright green spheroid in the fluorescence images . Fluorescence imaging confirmed that the sequestration of Cd into vacuoles occurred in HE protoplasts treated with 10–30 µM Cd for 90 min . As shown in Fig. 6, protoplasts were spheroid with a large vacuole in the centre of the cells and chloroplasts distributed around their periphery. The merged images of Cd fluorescence plus bright field indicated that Cd was largely localized centrally in HE protoplasts after a 90 min exposure to 10 µM or 20 µM Cd .

In contrast, green fluorescence of Cd was only present at the periphery of mesophyll protoplasts from NHEs. Furthermore a very low amount of Cd was taken up into the vacuoles of NHE protoplasts within the 120 min time frame . Treatment of mesophyll protoplasts with a high concentration of Cd confirmed the efficient sequestration of Cd by the vacuoles of HE plants . After a 10 min exposure to high levels of Cd, Cd fluoresence was observed in the centre of HE protoplasts, although its intensity was slightly lower than that in the other parts of the cell . This suggests that a certain amount of Cd uptake is undertaken by mesophyll protoplasts of HEs, with Cd entering into their vacuoles within 10 min of exposure. The central localization of Cd in HE protoplasts became more clear and pronounced with longer exposure to Cd.Protoplast integrity was monitored during their uptake of Cd through both ICP-MS and fluorescence imaging. Interestingly, the results showed that mesophyll protoplasts isolated from HEs appear to be more tolerant to Cd stress than those of NHEs. Protoplast integrity, as a percentage relative to controls, in HE S.alfredii was significantly higher than that of NHEs after exposure to 10 µM Cd for 2 h . This difference in tolerance between the HE and NHE protoplasts is highly pronounced under high Cd stress . Regardless of the Cd treatment period, the percentage of intact protoplasts in HEs in the presence of Cd was consistently higher than that of NHEs. The estimated mean viability of NHE protoplasts was less than 40% after Cd exposure for only 30 min, with intact protoplasts scarcely observed at 5 h. Although the percentage of intact protoplasts of HE S.alfredii gradually decreased as Cd exposure time increased, about 10% of the protoplasts remained viable for at least 7.5 h . According to the results of two-way ANOVA, the ecotypes, Cd exposures and their interactions had a significant effect on mesophyll protoplast integrity. Protoplast integrity and viability was also determined using FDA dye and showed a similar result, namely higher tolerance of HE protoplasts to high Cd stress when compared with NHEs .Intracellular reactive oxygen species production was monitored at the single cell level and compared between mesophyll protoplasts of HEs and NHEs after Cd treatment. Fluorescence signals denoting H2O2 were much stronger in protoplasts from NHEs than in those from HEs, at each time point. The generation of H2O2 was clearly evident as bright green fluorescence. H2O2 diffused into vacuole compartments in NHE protoplasts after 30 min of Cd exposure, with H2O2 levels reaching the highest at 1 h . In the HE protoplasts, Cd-induced green fluorescence was far less pronounced and was first observed in the plasma membrane and subsequently localized largely to chloroplasts .