Fig 1.
Natural variation in root hairs of 166 Arabidopsis thaliana accessions.
(A) Split-plate assay. Right panel: Col-0 root in—P. Red lines are drawn 1 mm apart (B) Correlation (R = 0.64) and 95% confidence interval of root hair density and root hair length under control conditions. Inset: phenotype of Mitt9311 (left) and Wa-1 (right) under control conditions, scaling bar 0.1 mm. Values represent mean of replicates with each 5 plants.
Fig 2.
Natural variation in root hair traits of 166 Arabidopsis thaliana accessions.
(A) Root hair density, (B) length, (C) surface under control (white bars) and scarce local phosphorus (blue bars). Error bars indicate standard error.
Fig 3.
Dependence of root hair density change with—P on root hair density under control conditions.
Weak negative correlation (R = -0.27, red line) and 95% confidence interval (grey lines) of root hair density response and density under control conditions.
Fig 4.
Root hair density and length responses to scarce local phosphorus (P) supply.
The responses were normalized to the density and length under control conditions (+P) and are given as relative values under—P (values of 1 indicate no change). (A) Root hair density response (density-P / density +P) versus relative length (length-P / length +P). (B-G) root hair phenotypes of Kondara (B,C), Col-0 (D,E) and Ga-0 (F,G) under control and local scarce P. Scaling bars: 0.5 mm.
Fig 5.
Manhattan-plots for root hair density (A) and root hair surface (B) response to local scarce phosphorus supply.
Light red dots (above the significance threshold) indicate SNPs used as cofactors in MLMM. SNPs of different chromosomes are given in different colors. Horizontal line indicates significant Bonferroni corrected p-values at 5% significance level.
Table 1.
Candidate genes as detected for root hair density response to scarce local phosphate supply by genome-wide association study.
Table 2.
Candidate genes as detected for root hair surface response to scarce local phosphate supply by genome-wide association study.
Fig 6.
Root hair length and density in loss of function mutants of candidate genes.
(A) Root hair density of wild type (Col-0), rlp48–1, aox1d-1, at1g32360–1, cyr1 (all mutants in the Col-0 background). Black bars: with P, grey bars: without P. (B) Root hair length of wild type (Col-0), rlp48–1, aox1d-1, at1g32360–1, cyr1. Black bars: control with P, grey bars: without P, error bars indicate standard error. Statistically significant data points compared to diploid individuals under the same conditions are given as ‘*’ (p<0.05).
Fig 7.
Root hair length and density in isogenic accessions differing in ploidy.
(A) Root hair density of Col-0 (2x), Col-0 (4x), Wa-1 (2x), Wa-1 (4x), 2x, diploid; 4x, tetraploid, Black bars: with P, grey bars: without local P. (B) Root hair length of Col-0 (2x), Col-0 (4x), Wa-1 (2x), Wa-1 (4x). Black bars: with P, grey bars: without P. Statistically significant data points compared to diploid individuals are given as ‘*’(p<0.05), error bars indicate standard error.