Potassium (K+) gradients serve as a mobile energy source in plant vascular tissues

Abstract

The essential mineral nutrient potassium (K(+)) is the most important inorganic cation for plants and is recognized as a limiting factor for crop yield and quality. Nonetheless, it is only partially understood how K(+) contributes to plant productivity. K(+) is used as a major active solute to maintain turgor and to drive irreversible and reversible changes in cell volume. K(+) also plays an important role in numerous metabolic processes, for example, by serving as an essential cofactor of enzymes. Here, we provide evidence for an additional, previously unrecognized role of K(+) in plant growth. By combining diverse experimental approaches with computational cell simulation, we show that K(+) circulating in the phloem serves as a decentralized energy storage that can be used to overcome local energy limitations. Posttranslational modification of the phloem-expressed Arabidopsis K(+) channel AKT2 taps this "potassium battery," which then efficiently assists the plasma membrane H(+)-ATPase in energizing the transmembrane phloem (re)loading processes.

Fig. 1.

Day-length–dependent phenotype of akt2-1 plants. Phenotypical analysis of wild-type and akt2-1 knockout plants grown under three different photoperiod regimes: (A) 16 h day/8 h night, (B) 12 h day/12 h night, and (C) 8 h day/16 h night (150 μE·m⁻²·s⁻¹ in all three conditions). Photos show 5-wk-old (A), 6-wk-old (B), and 6.5-wk-old plants (C). At these time points, the Wassilewskija wild-type plants showed similar developmental stages. Time courses of number of leaves (Middle panels) and height of the main inflorescence stalk (Bottom panels) are shown as means ± SD of n ≥ 25 plants.

Fig. 2.

Inward-rectifying AKT2 channel mutants do not complement the akt2-1 knockout plant. Ten Arabidopsis lines were analyzed: WT, akt2-1, and eight akt2-1 supertransformants expressing the AKT2 wild-type channel or channel mutants under control of the native AKT2 promoter. Two independent lines of akt2-1+AKT2-K197S (inward-rectifying AKT2 mutant), two independent lines of akt2-1+AKT2-S210A-S329A (inward-rectifying AKT2 mutant), two independent lines of akt2-1+AKT2, and two independent lines of akt2-1+AKT2-S210N+S329N (preferentially nonrectifying AKT2 mutant). (A) Representative 8-wk-old plants grown under 12-h day/12-h night conditions. Note that some plants already show signs of senescence (WT, akt2-1+AKT2, and akt2-1+AKT2-S210N+S329N) whereas others have not yet started bolting. (Inset) Four-week-old plants of the representative lines akt2-1+K197S#1 and akt2-1-S210A-S329A#1 and the Wassilewskija wild type grown in long-day conditions (16 h day/8 h night). Under these conditions, all plants started flowering at the same time. (B–D) Time courses of the height of the main inflorescent stalk (Left panels) and the number of leaves (Right panels) of the plants akt2-1+AKT2-K197S (B), akt2-1+AKT2-S210A-S329A (C), and akt2-1+AKT2-S210N-S329N (D) compared with the wild type (green curves, no symbols) and the akt2-1 knockout (red curves, no symbols). Detailed analyses of complementation lines (akt2-1+AKT2 and akt2-1+pSUC2:AKT2) are presented in Fig. S3. Data are shown as means ± SD of n ≥ 20 plants.

Fig. 3.

Computational simulation of phloem (re)loading processes. (A) A SE/CC complex was modeled as a cylinder with a surface-to-volume ratio of 0.4 μm⁻¹ and placed in a three times larger environment (apoplast). Note that different cell/environment values do not qualitatively change the obtained results. The continuous flux of the phloem sap was approximated by keeping pH_SE/CC, Suc_SE/CC, and K⁺_SE/CC constant. Likewise, pH_apoplast was kept constant to reflect the buffer capacity of the apoplast. Transport of K⁺, sucrose, and H⁺ into or out of the phloem was mediated by the H⁺-ATPase, the AKT2 K⁺-channel, the H⁺/Suc cotransporter, and a sucrose leak. Additionally, K⁺ was removed from the apoplast by adjacent cells. For further details, see Fig. S5. (B–F) Simulation of the network behavior. First, AKT2 was set as an inward-rectifying channel (i). Next, AKT2 was switched from an inward-rectifying channel into a nonrectifying channel (dotted lines, arrows). Time courses of the apoplastic sucrose concentration (Suc_apoplast: B), the transmembrane sucrose flux (ΔJ_Suc: C), the membrane voltage (V_m: D), the current pumped by the H⁺-ATPase (I_H+-ATPase: E), and the electrochemical K⁺ gradient (Δμ_K: F) are shown.

Fig. 4.

Limited oxygen supply is less tolerated by plants expressing no or only inward-rectifying AKT2 channels. Two-week-old plants were transferred from the greenhouse into plastic boxes supplemented with atmospheric gas mixtures containing 21% and 10% oxygen, respectively (16-h day/8-h night). (A) Representative 32-d-old plants grown under normal (21%; Left) and reduced (10%; Right) atmospheric oxygen. (B) Time courses of number of leaves in 21% (Left) and 10% (Center) O₂ atmospheres. Symbols between panels are cross-referenced with A. (C–E) Height of the main inflorescent stalk (C), fresh weight (D), and dry weight (E) of 42-d-old plants of wild type (black bars), akt2-1 knockout (open bars), akt2-1+AKT2-K197S#1 (circles), akt2-1+AKT2-S210A-S329A#1 (triangles), and akt2-1+AKT2-S210N-S329N#2 plants (stars). (Right panels) To determine whether the reduced O₂ treatment induced significant changes in the different plants, the data were referenced to the wild type. Normalized differences were calculated as ΔΔ_norm = (X^10% − mean_WT^10%)/mean_WT^10% − (X^21% − mean_WT^21%)/mean_WT^21%, where X^10% and X^21% denote the measured values in 10% and 21% O₂, respectively, and mean_WT^10% and mean_WT^21% are the mean values obtained for the wild type in the two O₂ conditions. Data are shown as means ± SD of n ≥ 10 plants. The reaction of akt2-1 was significantly different from the wild type under O₂-reduced conditions (Student's t test, P < 2e-06), as was the response of the plants expressing only inward-rectifying AKT2 mutants: akt2-1+AKT2-K197S (P < 2e-08) and akt2-1+AKT2-S210A-S329A (P < 4e-08). In contrast, there was no significant difference between the plant akt2-1+AKT2-S210N-S329N and the wild type (B: P = 0.43; C: P = 0.98; D: P = 0.03; E: P = 0.84).

Fig. 5.

Computational simulation of a realistic scenario. First, the activity of the H⁺-ATPase was reduced to 90% of its original value (arrow 1). After equilibration, AKT2 was switched from an inward-rectifying channel into a nonrectifying channel (arrow 2). Time courses of the apoplastic sucrose concentration (Suc_apoplast) for two different sequestration rates of K⁺ from the apoplast are displayed (red and blue curves). Further details on the simulation results are provided in Fig. S8.

Fig. 6.

akt2-1 plants differ from WT plants under sufficient potassium supply but not in K⁺-limited conditions. Root growth of akt2-1 and WT plants on synthetic media supplemented with 2,500 μM (A), 100 μM (B), and 10 μM (C) potassium. (Left) Representative 22-d-old plants grown vertically on plates (16 h light, 150 μE·m⁻²·s⁻¹). (Right panels) Time course of root growth as well as fresh weight and root-to-shoot ratio of 22-d-old plants. Data are means ± SD of n = 15 plants of each genotype. Note that in normal soil akt2-1 and WT plants did not show developmental differences in long days, possibly indicating that K⁺ is not abundantly available under these conditions.