Phosphate promotes Arabidopsis root skewing and circumnutation through reorganisation of the microtubule cytoskeleton

Abstract

Phosphate (P_i) plays a key role in plant growth and development. Hence, plants display a range of adaptations to acquire it, including changes in root system architecture (RSA). Whether P_i triggers directional root growth is unknown. We investigated whether Arabidopsis roots sense P_i and grow towards it, that is whether they exhibit phosphotropism. While roots did exhibit a clear P_i-specific directional growth response, it was, however, always to the left, independent of the direction of the P_i gradient. We discovered that increasing concentrations of KH₂PO₄, trigger a dose-dependent skewing response, in both primary and lateral roots. This phenomenon is P_i-specific - other nutrients do not trigger this - and involves the reorganisation of the microtubule cytoskeleton in epidermal cells of the root elongation zone. Higher P_i levels promote left-handed cell file rotation that results in right-handed, clockwise, root growth and leftward skewing as a result of the helical movement of roots (circumnutation). Our results shed new light on the role of P_i in root growth, and may provide novel insights for crop breeding to optimise RSA and P-use efficiency.

Keywords: Arabidopsis; circumnutation; microtubule cytoskeleton; phosphate; root skewing.

Fig. 1

Arabidopsis root growth deviates to the left in response to increased P_i levels, independent of where the concentration gradient comes from. (a) Schematic diagram of P_i‐tropism assay, with low‐P_i concentration in the agar medium and high P_i on a piece of filter paper (blue), placed at the left‐hand corner, 3.5 cm away from the seedlings. (b) Visualisation of P_i diffusion using radioactive ³²P_i. The ³²P_i was spotted onto the filter paper and the radioactivity imaged 97 h later. (c) Phenotype of Col‐0 seedlings grown on low P_i‐agar medium with filter paper containing high P_i or MQ water, either on the left‐ or right side of the agar plate, 6 d after transfer. (d) Quantification of root angles. The middle line represents the median, and the dotted line represents quartiles. Different letters indicate significant differences (P < 0.05) among treatments (One‐way ANOVA followed by Tukey HSD post hoc analysis, F _3,133 = 48.48, P < 0.001). n = 25–49. Bar, 1 cm.

Fig. 2

Increasing KH₂PO₄ concentrations promote root skewing in Arabidopsis ecotypes Col‐0 (a–d) and Ws‐4 (e–h). (a, e) Seedling phenotype after 9 d of growth on ½MS medium containing indicated KH₂PO₄ concentrations. (b, f) Projections of primary roots of 34–40 seedlings grown at indicated P_i concentration. (c, g) Same as panels b and f, but with the overlap indicated in purple. (d, h) Quantification of root skewing using horizontal growth index (HGI). Data represent means ± SE (n = 34–40). Different letters indicate significant differences (P < 0.05) among treatments using One‐way ANOVA followed by Tukey HSD post hoc analysis: F _12,479 = 189.5, P < 0.0001 (d) or F _12,484 = 415, P < 0.0001 (h). Bars, 1 cm.

Fig. 3

Phosphate is responsible for the induced root skewing in Arabidopsis. (a, c) Primary root projections and overlap (purple) of 9‐d‐old seedlings of Ws‐4 (a) and Col‐0 (c), grown on normal ½‐strength Murashige & Skoog medium (½MS) (containing standard 625 μM P_i) or ½MS supplemented with 625 μM P_i of different cationic composition, that is K⁺, Na⁺, or NH₄ ⁺. (b, d) HGI for Col‐0 (b) and Ws‐4 (d). The middle line represents the median and box limits indicate the 25^th and 75^th percentiles; whiskers span down to the minimum and up to the maximum values, and individual data points are represented by dots. One‐way ANOVA was performed to identify significant differences of Ws‐4 (F _3,144 = 64.39, P < 0.0001) and Col‐0 (F _3,147 = 155.3, P < 0.001) between treatments. Different letters indicate a statistically significant difference (P < 0.05) by Tukey HSD post hoc analysis (n = 31–40). Bars, 1 cm.

Fig. 4

Time‐lapse of root growth and skewing responses in Col‐0 and Ws‐4 seedlings at 300 or 1250 μM P_i. The growth of Arabidopsis Col‐0 and Ws‐4 seedlings were followed over a period of 6 d in a 16 h : 8 h, L : D regime. Images were captured every 10 min. Left panels (a, c, e) 300 μM P_i, right panels (b, d, f) 1250 μM P_i. (a, b) Primary root length, (c, d) growth rate, and (e, f) primary root angles. Values represent the means of five biological replicates of five seedlings. Blue symbols, Col‐0; red symbols, Ws‐4. See also Supporting Information Videos S1 and S2.

Fig. 5

Increasing P_i concentrations promote skewing of lateral roots. Root system architecture of Arabidopsis thaliana seedlings, ecotypes Col‐0 (a) and Ws‐4 (d) grown at 300‐, 625‐, or 1250 μM P_i for 10 d. The first two lateral roots on the left (indicated in red) and on the right (indicated in yellow) were analysed for growth directions. (b, e) Projections of multiple left (red) or right (yellow) lateral roots of Col‐0 (b) or Ws‐4 (e). (c, f) Box plot analyses of HGI values for left‐ and right lateral roots of Col‐0 (c) or Ws‐4 (f). The middle line represents the median and box limits indicate the 25^th and 75^th percentiles; whiskers span down to the minimum and up to the maximum values, and individual data points are represented by dots. The one‐way ANOVA and Kruskal–Wallis test were performed to identify significant differences of Col‐0 (F _5,445 = 297, P < 0.0001) and Ws‐4, respectively. Different letters indicate significant differences (P < 0.05) among treatments. Values represent lateral roots of 30–41 seedlings. Bars, 1 cm.

Fig. 6

P_i promotes right‐handed root growth and left‐handed epidermal cell file rotation (CFR). (a, b) Stereomicroscopic images of root tips of Arabidopsis Col‐0 (a) and Ws‐4 (b) at 300‐, 625‐, or 1250 μM of P_i. (c, d) Zoom‐in of elongation zone of Col‐0 (c) and Ws‐4 (d) root grown at 1250 μM P_i. Black arrows indicate root growth direction; red dotted lines indicate CFR direction. (e, f) Confocal images of root elongation zone of propidium iodide‐stained Col‐0 (e) or Ws‐4 (f) seedlings, grown at indicated P_i concentrations. (g) Quantitative analysis of epidermal CFR angle at elongation zone of Col‐0 or Ws‐4 seedlings. Different letters indicate significant differences (P < 0.05) between treatments (One‐way ANOVA followed by Tukey HSD post hoc analysis, F _5,139 = 73.92, P < 0.0001). Values represent three biological replicates with 6–11 seedlings analysed for each replicate. The middle line represents the median and box limits indicate the 25^th and 75^th percentiles; whiskers span down to the minimum and up to the maximum values, and individual data points are represented by dots. (h) Skewing direction of seedling (front view). (i, j) Stereomicroscopic image showing right‐handed (clockwise) root growth (i) and left‐handed (counterclockwise) epidermal cell file rotation (j) in the same image, captured from a Col‐0 seedling grown vertically for 4 d and was then placed horizontally for another 2 d, which causes roots to coil into their growth direction. Red dotted lines indicate the direction of the epidermal cell file (i). All images are front view. Bars: (a–f) 100 μm; (h–j) 1 mm.

Fig. 7

P_i‐dependent root skewing is enhanced by microtubule‐stabilisation drug, Taxol. (a) Phenotype of Arabidopsis Col‐0 seedlings grown on indicated P_i concentrations with and without 1 μM Taxol after 7 d. Bar, 1 cm. (b) Quantification of HGI value of seedlings growing at different P_i concentrations with or without Taxol. Two‐way ANOVA was used to determine the significant differences between different P_i treatments (F _2,151 = 300.8, P < 0.0001) and Taxol treatments (F _1,151 = 3557, P < 0.0001). (c) Stereomicroscope images of the surface of roots exposed to different treatments. (d) Quantitative analysis of epidermal CFR angle at the base of the elongation zone at different P_i concentrations. Two‐way ANOVA was used to determine the significant differences between different P_i treatments (F _2,137 = 179.8, P < 0.0001) and Taxol treatments (F _1,137 = 393.7.1, P < 0.0001). (e) Confocal microscope images of propidium iodide‐stained elongation zone of root grown on different P_i concentrations. The data shown are means ± SE. Different letters indicate significant differences (P < 0.05) between treatments. All values represent two biological replicates with 11–15 seedlings analysed for each replicate. Bars, 100 μm.

Fig. 8

Effect of P_i on PDS and microtubule orientation in FP‐reporter lines. (a) Quantification of HGI value in different microtubule reporter lines, grown on different P_i concentrations. All values represent the means of two biological replicates with 6–15 seedlings analysed for each replicate. (b) Original confocal images (left, green) and ImageJ‐processed images (right, coloured) of 5‐d‐old UBQ::GFP‐TUB6 seedlings. (c) Quantification of microtubule orientation by measuring multiple ROIs in cells grown at different P_i concentrations in Arabidopsis UBQ::GFP‐TUB6. (d) Distribution of microtubule orientation. Angles ranging from 22.5 to −22.5° were considered ‘Transverse’, while angles ranging from 67.5 to 90° and −67.5 to −90° were considered ‘Longitudinal’; all others were designated as ‘Random’. Values represent the mean of five biological replicates with 10–15 seedlings per condition. Data shown are means ± SE. Asterisks indicate a statistically significant difference (**, P < 0.01; ***, P < 0.001, ns, not significant) by two‐way ANOVA. Differences between 300 and 1250 μM were statistically significant (P < 0.001) for each line except for UBQ1::GFP‐MBD, which did not respond at all. Bar, 10 μm.

Fig. 9

P_i affects root penetration power due to 3D skewing. To test whether P_i promotes PDS in a 3D‐agar environment, the bending of primary roots was analysed in a vertical two‐layered agar system. The first layer of 0.7% allows seed germination and root growth, while the second layer contains either 0.7% or 1.6% agar to score the penetration power of the roots. (a) Schematic drawing of the vertical two‐layered agar system. (b) Examples of Arabidopsis bending and nonbending roots. Nonbending roots directly penetrate the lower layer medium while bending roots bounch‐off the lower agar medium and grow along the interface. Left panel shows typical nonbending root at 300 μM P_i while right panels shows typical bending/skewing roots at 1250 μM (c) Quantification of roots showing a bending response at 300, 625 and 1250 μM P_i. X‐axis indicates agar concentration of the lower layer (0.7% or 1.6%). Data shown are means ± SE. The Kruskal–Wallis test was performed to identify significant differences. Different letters indicate significant differences (P < 0.05) among treatments. All values represent two biological replicates. n = 91–126. Bars, 1 mm.

Fig. 10

Model explaining right‐ and left‐handed directional movements of microtubules, cells and root during PDS in Arabidopsis. (a–c) Mechanism involves the rearrangement of microtubules affecting the CFR of epidermal cells in the elongation zone and rotational growth direction (circumnutation) of the root tip. Increased P_i concentrations promote a right‐handed shift of cortical microtubules (cMT) (a), which is followed by the right‐handed twisting of cellulose microfibrils (CMF) (b). This creates a biophysical torsion during cell elongation that triggers an increase in left‐handed CFR of epidermal cells in the root elongation zone (b), and promotes the subsequent rightward growth of the root (as seen on horizontal agar plates) and the leftward skewing on vertical plates (c). Figure is adapted from Zhou (2021).