Discriminative Long-Distance Transport of Selenate and Selenite Triggers Glutathione Oxidation in Specific Subcellular Compartments of Root and Shoot Cells in Arabidopsis

Abstract

Selenium is an essential trace element required for seleno-protein synthesis in many eukaryotic cells excluding higher plants. However, a substantial fraction of organically bound selenide in human nutrition is directly or indirectly derived from plants, which assimilate inorganic selenium into organic seleno-compounds. In humans, selenium deficiency is associated with several health disorders Despite its importance for human health, selenium assimilation and metabolism is barely understood in plants. Here, we analyzed the impact of the two dominant forms of soil-available selenium, selenite and selenate, on plant development and selenium partitioning in plants. We found that the reference plant Arabidopsis thaliana discriminated between selenate and selenite application. In contrast to selenite, selenate was predominantly deposited in leaves. This explicit deposition of selenate caused chlorosis and impaired plant morphology, which was not observed upon selenite application. However, only selenate triggered the accumulation of the macronutrient sulfur, the sister element of selenium in the oxygen group. To understand the oxidation state-specific toxicity mechanisms for selenium in plants, we quantified the impact of selenate and selenite on the redox environment in the plastids and the cytosol in a time-resolved manner. Surprisingly, we found that selenite first caused the oxidation of the plastid-localized glutathione pool and had a marginal impact on the redox state of the cytosolic glutathione pool, specifically in roots. In contrast, selenate application caused more vigorous oxidation of the cytosolic glutathione pool but also impaired the plastidic redox environment. In agreement with the predominant deposition in leaves, the selenate-induced oxidation of both glutathione pools was more pronounced in leaves than in roots. Our results demonstrate that Se-species dependent differences in Se partitioning substantially contribute to whole plant Se toxicity and that these Se species have subcellular compartment-specific impacts on the glutathione redox buffer that correlate with toxicity symptoms.

Keywords: compartmentation; oxidation; roGFP2; selenium; toxicity.

Figure 1

Impact of selenite and selenate treatments on the roots and leaves of Arabidopsis plants (A,B) Impact of increasing concentrations of selenite and selenate (0–200 μM) on root length (A) and shoots fresh weight (B) of Arabidopsis. Six-days-old seedlings were challenged for additional 15 days on At-medium (Control, black bars) or At-medium containing either selenite (gray bars) or selenate (white bars). Different lowercase letters indicate individual groups identified by pairwise multiple comparisons with a Holm-Sidak one-way ANOVA (p = 0.05, n = 7). (C) Top view of six-week-old hydroponically grown Arabidopsis plants that were either challenged for seven days with ½ Hoagland media supplemented with selenite (50 μM Na₂SeO₃) or selenate (50 μM Na₂SeO₄). The control plants were grown under the same conditions (100 μE light for 8 h, 50% humidity and 22°C day/18°C night) on ½ Hoagland medium lacking selenium. Images were digitally extracted for comparison. Scale bar = 1 cm. FW, fresh weight.

Figure 2

Impact of selenite and selenate application on the partitioning of selenium, phosphorus, and sulfur in Arabidopsis. (A–C) Deposition of selenium (A), phosphorus (B), and sulfur (C) in leaves and roots of seven-week-old hydroponically grown plants challenged for one week with 50 μM selenite (gray), 50 μM selenate (white) or no additional selenium (Control, black). Different lowercase letters indicate individual groups identified by pairwise multiple comparisons with a Holm-Sidak one-way ANOVA (p = 0.05, n = 3). n.d., not detectable, DW, dry weight.

Figure 3

Organ-specific impact of selenite and selenate application on key metabolites of the reductive sulfur assimilation pathways. (A–D) Steady-state levels of sulfate (A), cysteine (B), glutathione (C) and the carbon-nitrogen backbone for sulfide/selenide incorporation, OAS (D) in leaves and roots of seven-weeks-old hydroponically grown Arabidopsis plants challenged for one week with 50 μM selenite (gray), 50 μM selenate (white) or no additional selenium (control, black). Different lowercase letters indicate individual groups identified by pairwise multiple comparisons with a Holm-Sidak one-way ANOVA (P, 0.05, n = 7). If the power of the α-test was too low in the one-way ANOVA, we tested the statistical difference between control and the single selenium treatment by a paired students t-test (^*p = <0.05). FW, fresh weight.

Figure 4

Only selenite can be reduced by GSH in a cell-free system. (A–C) In vitro assay for determining the reduction of selenite (A,B) or selenate (C) by glutathione resulting in glutathione disulfide (GSSG). The assay allows quantification of selenium reduction by coupling the reduction of the byproduct GSSG to GSH by glutathione disulfide reductase (GR) under consumption of NADPH, which can be quantified at 340 nm. (A) The assay (0.5 μM GR, 2 mM GSH, 100 μM NADPH) was preincubated at the reaction temperature, and the reaction was started (dashed line) by the addition of 20 μM selenite (gray circles) or water (Control, black circles). (B) As controls for the specificity of the NADPH-dependent reduction of GSSG by GR, GSH (white triangles) or GR (gray triangles) was omitted from the assay. In contrast to selenite application, (C) application of selenate (white circles) did not result in GSSG formation and was indistinguishable from control (black circles). The selenate assay was performed under the identical conditions described in (A) for selenite. All assays were repeated in triplicates at two individual time points and showed similar results.

Figure 5

Time-resolved impact of selenium species on the cytosolic and the plastid glutathione redox state in roots of Arabidopsis. (A,B) Representative false-color images of the Grx1-roGFP2 signal ratio in the cytosol (A) or plastids (B) of roots treated with selenite, selenate or without additional selenium supply for indicated time points. Arabidopsis plants expressing the Grx1-roGFP2 sensor in the cytosol or plastids were treated for up to 2 days in ½ Hoagland medium (control), supplemented with either 50 μM Na₂SeO₃ (selenite) or 50 μM Na₂SeO₄ (selenate). roGFP2 was excited at 405 nm and 488 nm, respectively, and fluorescence was collected at 505–530 nm. The 405/488-nm fluorescence ratio is shown on a false-color scale spanning full reduction of roGFP2 (blue) to full oxidation (red). Images were digitally enhanced for comparison. Scale bar = 20 μm. (C,D) The roGFP2 fluorescence ratio is indicating relative changes in the glutathione redox potential in the cytosol (C) or the plastid stroma (D) of roots treated for indicated times with ½ Hoagland medium (control, black bars), supplemented with selenite (gray bars) or selenate (white bars). Data are shown as mean values ± SD. Different lowercase letters indicate individual groups identified by pairwise multiple comparisons with a Holm-Sidak one-way ANOVA (p = 0.05, n = 5–15).

Figure 6

Time-resolved impact of selenium species on the cytosolic and the plastid glutathione redox state in leaves of Arabidopsis. (A,B) Representative false-color images of the Grx1-roGFP2 signal ratio in the cytosol (A) or plastids (B) of leaves from hydroponically grown plants treated with selenite, selenate or without additional selenium supply for indicated time points via the root system. Arabidopsis plants expressing the Grx1-roGFP2 sensor in the cytosol or plastids were treated for up to 2 days in ½ Hoagland medium (control), supplemented with either 50 μM Na₂SeO₃ (selenite) or 50 μM Na₂SeO₄ (selenate). roGFP2 was excited at 405 nm and 488 nm, respectively, and fluorescence was collected at 505–530 nm. The 405/488-nm fluorescence ratio is shown on a false-color scale spanning full reduction of roGFP2 (blue) to full oxidation (red). Images were digitally extracted for comparison. Scale bar = 20 μm. (C,D) The roGFP2 ratio signal is indicating relative changes of the glutathione redox potential in the cytosol (C) or the plastid stroma (D) of leaves treated for the indicated times with ½ Hoagland medium (control, black), supplemented with selenite (gray) or selenate (white). Data are shown as mean values ± SD. Different lowercase letters indicate individual groups identified by pairwise multiple comparisons with a Holm-Sidak one-way ANOVA (p = 0.05, n = 8–20).