Total GSH in mature T. goesingense and A. thaliana (Wassilewskija) was previously measured and found to be 412.5 ± 80 and 268.86 ± 48.1 nmol g^-1 fwt respectively 19. The tolerance of T. goesingense to various metals including Ni, Co, Zn and Cd was investigated by measuring root growth compared to the non-tolerant relative A. thaliana Wassilewskija (WS) (Fig. 1) of plants grown on solidified MS medium containing Ni, Co, Zn or Cd. When comparing the concentration of a metal required for a 50% inhibition of root growth or (I₅₀) T. goesingense was found to be the most tolerant to Ni ≅ Zn ≅ Co, with a low level of resistance to Cd (Table 1). The relative tolerance index (RTI) was calculated as the I₅₀ [T. goesingense]/I₅₀ [A. thaliana]. By using the RTI as a comparative measure of tolerance T. goesingense was observed to have enhanced relative tolerance to Ni ≅ Zn ≅ Co, with a low level of resistance to Cd when compared to A. thaliana (WS) (Table 1). To investigate how much of the metal tolerance observed in T. goesingense is due to increased resistance of oxidative damage we measured shoot lipid peroxidation in T. goesingense and A. thaliana after 8-days growth in the presence of Ni, Zn, Co or Cd (Fig. 2). We observed that relative to A. thaliana the hyperaccumulator T. goesingense contained significantly less peroxidized lipids after growth in the presence of Ni, Co and Zn, however no difference was observed after growth in the presence of Cd.

Previously, we reported that heterologous expression of T. goesingense SAT (TgSAT_m), leading to elevated shoot GSH levels, resulted in enhanced Ni tolerance of A. thaliana 19. The total GSH present in TgSAT-m heterologous expressing lines, pooled from plates as young plants, was (S 4–9, 496 ± 20), (S 3-1, 369 ± 5), (S 5-4, 188 ± 9) nmol g^-1 Fwt, and represent the high, intermediate and low SAT expressors while the two empty vector controls with the same phenotype were (E 1–5, 172 ± 9) and (E 4–5, 142 ± 5) nmol g^-1 Fwt 19. In addition to the TgSATm Ni tolerance shown growing in the vertical plates (Fig. 3A, 30 d), we also discovered metal tolerances to Co (Fig. 3B, 18 d) and to Zn (Fig. 3C, 18 d) and resistance to Cd (Fig. 3D, 18 d). The Ni tolerance of TgSATm was previously quantified using a metal growth, zone of inhibition assay, on plates, 19 (Fig 4A Ni). Using the same disc assay here, we quantify that plants heterologously expressing TgSAT_m are also tolerant to Co (Fig. 4B) and Zn (Fig. 4C), and show enhanced resistance to Cd (Fig. 4D). By using lines with measured levels of TgSAT_m expression, enzyme activity and GSH accumulation 19, and multiple empty vector controls, quantitative data was obtained for the growth response to Ni, Co, Zn and Cd of high S 4–9(black), intermediate S 3-1(grey) and low S 5-4(white) or non-expressing lines, empty vector E 1–5(striped), of A. thaliana transformed with TgSAT_m. Though such an assay produces only a relative measurement of tolerance, quantified as the distance from a metal-soaked disc, it allows highly reproducible results and measurements of both tolerance and accumulation over a continuous gradient of metal concentrations. In this method we use different parameters to assess plant growth than just root lengths, including development of normal upright shoot cotyledons, secondary leaf formation, and formation of a root hook as the root bends on contact with the base of the Petri dish. Data is expressed as a percentage of the total number of seedlings in each ring around the metal soaked disc. This assay takes full advantage of the metal concentration gradient established as the metal diffuses from the disc, allowing resolution of differences in metal tolerance between lines and various metals. Based on estimations of shoot and root growth on both vertical plates and disk assay plates (Fig. 3 &4), heterologous expression of TgSAT_m was observed to provide significant increases in tolerance to Ni, Co and Zn along with a slight amount of Cd resistance.

Shoot material from the TgSATm lines and empty vector controls grown at 20 mm and 35 mm from the disc were harvested for ICPMS analysis to investigate the respective metal concentrations. Interestingly, although tolerance to Ni, Co and Zn was increased in lines heterologously expressing TgSAT_m, these tolerant lines had little or no increase in shoot accumulation of Ni, Co or Cd. The only exception to this was for Zn, because both the highest (S 4–9), and intermediate (S 3-1) SAT expressing and GSH accumulating TgSATm lines harvested from the 20 mm ring had a 1.3–1.5 fold higher shoot Zn concentration (8,936 ± 760 & 10,050 ± 458 μg Zn g^-1 Dwt respectively) compared to the low (S 5-4) and non-expressing line (E 1–5) at (6,704 ± 656 & 6,644 ± 94 μg Zn g^-1 Dwt respectively). While the tolerant SAT lines did not accumulate more metal than the others except for Zn the overall levels were all quite high, for Ni (~1,200), Zn (~6,000–10,000), Co (~2,000) and Cd (~2,000) μg g^-1 Dwt.

To determine if enhanced resistance to oxidative stress plays a role in the metal tolerances conferred by heterologous expression of TgSAT_mwe measured lipid peroxidation in shoots of plants after growth in the presence of toxic levels of Ni, Co, Zn and Cd (Fig. 5). These results demonstrate that under the conditions used growth in the presence of Ni, Co and Zn produce greater lipid peroxidation than Cd. They also show that lines with relatively strong expression of TgSATm (black bars), have the lowest levels of lipid peroxidation in plants grown in the presence of either Ni, Co or Zn. However, heterologous expression of TgSATm conferred no decrease in lipid peroxidation in Cd exposed plants, most likely because Cd-induced lipid peroxidation was low even in control lines (Fig. 5).

Heterologous expression of TgSAT_m was also observed to increase resistance to hydrogen peroxide (Fig. 6). This assay measures the remaining chlorophyll content of leaf discs floated on various concentrations of hydrogen peroxide under high light. Both wild type and empty vector A. thaliana were used as controls and produced the same results, data for empty vector control is shown (Fig. 6). These results establish that heterologous expression of TgSAT_m also enhances tolerance to general oxidative stress, beyond that caused specifically by growth in the presence of the metals tested. Interestingly, the hyperaccumulator T. goesingense, which is known to have constitutively elevated SAT activity and accumulate elevated GSH 19, also showed significant tolerance to hydrogen peroxide treatments (Fig. 6).

Heterologous expression of TgSAT_m was previously shown to increase both shoot GSH levels and Ni tolerance (Freeman et al., 2004). We observe here that in addition to increased Ni tolerance, elevated GSH also leads to increased tolerance to the growth inhibitory effects of Co, Zn and a low level of resistance to Cd. However, by comparing the magnitude of the tolerance to Ni, Co, Zn, and Cd resistance conferred by heterologous expression of TgSAT_m with the concentration of shoot GSH, across lines with different levels of SAT expression, we established that the effectiveness of GSH in enhancing metal tolerance varies with the metal. The percent toxicity, scored for both shoot and root, was calculated for each line and metal at the seedling ring, which had the maximum difference in tolerance when the highest SATm expressing line was compared to the control line. These tolerance values were plotted against the GSH synthesis capacity of each line, and the relationship was found to be linear, with increasing GSH conferring increasing metal tolerance to Ni, Co, Zn, and resistance to Cd (DNS). From the slope of these plots the metal tolerance vs GSH concentration value is obtained for each metal in both shoot and root (Fig. 7). The equivalent increase in shoot GSH concentrations was most effective at increasing shoot tolerance to Ni and Co, followed closely by Zn then least effective was Cd resistance. GSH also caused elevated metal tolerances in roots to Co followed by Ni and Zn, while Cd resistance in roots was least correlated to total GSH production. However, metal levels in the roots were not measured and Cd may have been much higher here, as Cd is known to accumulate in roots.

Plants were germinated and grown on plates containing 1/2 strength MS + Gamborgs B5 medium in 0.8% plant tissue grade agar alone or with the different metals under 16 hr white light (120 μmol m^-2 s^-1 of photosynthetic photon flux (PPF)). ICP-MS analysis of the different metal levels was performed on all dried shoot tissue as previously described (Lahner et al., 2003).

Metal resistance in A. thaliana was tested on agar plates (100 × 25 mm) containing 1/2 strength MS + Gamborgs B5 medium in 0.8% plant tissue grade agar. Seeds from A. thaliana heterologously expressing T. goesingense SATm and empty vector control lines were spotted in concentric rings radiating from the center of the plate where a filter paper disc was placed. Onto the disc 100 μL of 100 mM Ni(NO₃)₂, 100 mM Co (NO₃)₂, 400 mM Zn (NO₃)₂ or 50 mM Cd (NO₃)₂ was pipetted and the plates incubated at 24°C/20°C, 10 h/14 h light/dark, 120 μmol m^-2 s^-1 PPF. After 13 days growth from imbibition the percentage of upright seedlings with fully expanded cotyledons, secondary leaves and hooked roots were recorded for each ring along with the distance in mm from the metal soaked filter disc. Seedlings from both 20 and 35 mm rings were then harvested for ICP-MS analysis of metal levels. Metal analysis was performed on dried shoot tissue as previously described 38.

Hydrogen peroxide treatments involved cutting 1 cm diameter leaf discs from the center of intact mature leaves. Discs were floated on either H₂O₂ or water for 24 h at 24°C in constant light (180 μmol m^-2 s^-1 PPF). Percent total chlorophyll was measured at 654 nm following extraction in 96% ethanol 39, and normalized to chlorophyll observed in the water treated tissue. The chlorophyll contents for the control (H₂O) treatments were approximately the same per gram fresh weight for all plants used. Total Chlorophyll mg g^-1 fwt.; T. goesingense = 1.55 ± 0.32, A. thaliana = 1.57 ± 0.13, Transgenic TgSATm = 1.58 ± 0.13.

Peroxidized lipids were assayed as the presence of

t

a

r

s

T. goesingense and A. thaliana plants were germinated and grown on vertical plates containing 1/2 strength MS + Gamborgs B5 medium in 0.8% agar with various concentrations of Ni, Co, Zn and Cd for 18 days under 16 hr white light (120 μmol m^-2 s^-1 PPF). The (I⁵⁰) values were calculated by taking the metal concentrations in plates which caused 50% inhibition of root growth for the different metals used on both T. goesingense and A. thaliana. The maximum normal root growth was measured on the control plates for this experiment in the absence of any metal.