The plant hormone ethylene restricts Arabidopsis growth via the epidermis

Abstract

The gaseous hormone ethylene plays a key role in plant growth and development, and it is a major regulator of stress responses. It inhibits vegetative growth by restricting cell elongation, mainly through cross-talk with auxins. However, it remains unknown whether ethylene controls growth throughout all plant tissues or whether its signaling is confined to specific cell types. We employed a targeted expression approach to map the tissue site(s) of ethylene growth regulation. The ubiquitin E3 ligase complex containing Skp1, Cullin1, and the F-box protein EBF1 or EBF2 (SCF^EBF1/2) target the degradation of EIN3, the master transcription factor in ethylene signaling. We coupled EBF1 and EBF2 to a number of cell type-specific promoters. Using phenotypic assays for ethylene response and mutant complementation, we revealed that the epidermis is the main site of ethylene action controlling plant growth in both roots and shoots. Suppression of ethylene signaling in the epidermis of the constitutive ethylene signaling mutant ctr1-1 was sufficient to rescue the mutant phenotype, pointing to the epidermis as a key cell type required for ethylene-mediated growth inhibition.

Keywords: Arabidopsis; EIN3 binding F-box factor EBF; auxin; ethylene; root/shoot.

Fig. 1.

Overview of the cell type-specific promoters used to drive the expression of EBF1/EBF2 F-box proteins in the different root cell types. The legend with color-coded root cell types is prepared according to Barrada et al. (26). The asterisk indicates that the promoter is ethylene-inducible in the particular cell type. Promoters marked with # were also used by Swarup et al. (16). H, trichoblasts; N, atrichoblasts.

Fig. 2.

EBF1/2 overexpression affects the ethylene signaling pathway without interfering with ethylene biosynthesis. (A) Roots of 6-d-old seedlings exhibit strong resistance to ACC treatment when EBF1/2 is overexpressed. Error bars indicate SD (n = 20, results are from one representative experiment). Absolute root length (Right) (millimeters; results obtained from a single experiment) is shown next to the relative values (Left). *P ≤ 0.05 (Welch’s t test) vs. control plants from the same genotype grown on media without ACC. (B) Bright-field micrographs of GUS-stained wild-type and F1 p35S::EBF1/2 seedlings carrying the pEBS::GUS reporter gene, grown for 6 d on 1/2 MS medium in the presence of 1 μM ACC. (Magnification: 20×.) (C) Ethylene production by Col-0, ein2-1, and p35S::EBF1/2 (expressed as picoliters of ethylene produced by a single individual over a period of 24 h). Error bars indicate SD (n ≥ 3, each independent sample contained 100 seeds grown in a sealed cuvette). *P < 0.05 vs. control (Col-0), assessed by Welch’s t test.

Fig. 3.

Targeted cell type-specific expression of EIN3-binding F-box protein EBF2 reveals that ethylene signaling in the LRC and epidermis controls root elongation. (A) Confocal images of promoter::GFP reporter lines grown in the presence of 1 μM ACC, visualizing the expression pattern of the selected root cell type-specific promoters fused to EIN3-binding F-box proteins in different root zones: cell division zone (CDZ), transition zone (TZ), elongation zone (EZ), and differentiation zone (DZ). (Magnification: 20×.) (B) Relative root length of different lines on 1/2 MS medium containing 1 μM ACC compared with medium without ACC. Error bars indicate SD (n = 3 datasets). *P ≤ 0.05 (Welch’s t test with Holm–Bonferroni sequential correction) reflects the significant differences in relative growth on ACC. (C) F1 crosses of the transgenic plants carrying constructs with cell type-specific expression of the F-box protein with the reporter pEIN3::gEIN3-3xGFP (ein3-1) confirmed the functionality of the constructs. The specific promoter-driven EBF lines used as mother plants did not contain a GFP reporter; the images of promoter::GFP depicted here are as in A and only visualize the expression pattern. Zoomed-in sectors of confocal images are shown. (Magnification: 20×.) (D) Inhibition of root cell elongation by ACC is relieved when the ethylene signal is blocked in the LRC and epidermis. Profiles of root epidermal and cortical cell expansion in wild type (Col-0), ein2-1 ethylene-insensitive mutant, and a transgenic line with epidermis- and LRC-targeted EBF2 expression (pA14::EBF2).

Fig. 4.

Epidermis-specific suppression of the ethylene signal positively influences cell expansion in leaves and complements ctr1-1. (A) Light-grown 14-d-old seedlings with compromised ethylene signaling in leaf epidermis (pML1::EBF2 and pLRC1::EBF2) are insensitive to ACC. Error bars represent SD (n = 3 datasets). **P ≤ 0.01; ***P ≤ 0.001 (Welch’s t test with Holm–Bonferroni sequential correction). Confocal images of the first true rosette leaves of pML1::GFP, pA14::GFP, and pLRC1::GFP at 14 DAG (days after germination). A wild-type Col-0 shoot was used as a control for autofluorescence correction. (Magnification: 20×.) (B) Leaf pavement cell area of ethylene-insensitive lines shows reduced response to treatment with 50 μM ACC. DIC microscopic observations are made on cleared first true leaves at 21 DAG. Three neighboring cells are artificially colored to visualize the differences between the controls and the treated samples. Error bars indicate SD (n ≥ 20). **P ≤ 0.01; ***P ≤ 0.001 (Welch’s t test with Holm-Bonferroni sequential correction). (C) Root elongation assay, phenotype, and rosette area of ctr1-1 transgenic lines carrying pLRC1::EBF2 (strong root and shoot epidermal promoter), pML1::EBF2 (weak root and strong shoot epidermal promoter), or pA14::EBF2 (strong root and weak shoot epidermal promoter) constructs. Error bars indicate SD (n = 3 datasets). Statistically significant difference with ctr1-1: *P ≤ 0.05; ***P < 0.001 (Welch’s t test with Holm–Bonferroni sequential correction).

Fig. 5.

Reduced ethylene signal in the LRC and epidermis affects auxin biosynthesis and responsiveness in the root tip. (A) Reduction of root growth by ACC in dark-grown seedlings is abolished by application of the auxin biosynthetic inhibitor l-Kyn (Kyn). Statistically significant differences between ACC and ACC + Kyn-treated individuals: ***P ≤ 0.001 (Welch’s t test). Error bars represent SD (n ≥ 20). Crosses between the translational fusion reporter line pTAA1::GFP-TAA1 and the ethylene-insensitive transgenic lines pA14::EBF2 and pLRC1::EBF2 reveal that EBF2 cell type-specific expression in the epidermis and LRC impairs ethylene responsiveness of TAA1 expression therein. (Magnification: 20×.) (B) Cells with reduced ethylene signaling have altered responsiveness to 50 nM 2,4-D short-term treatment. Error bars represent SD (n ≥ 20). Statistically different results are shown: ***P ≤ 0.001 (Welch’s t test). White arrowheads on the representative confocal images mark the epidermal cell files where the measurements are taken. (Magnification: 40×.)

Fig. 6.

Reduced ethylene signal in the LRC and epidermis affects auxin transport in the root tip. (A) Attenuation of ethylene signals in the LRC and epidermis abolishes the positive effect of ethylene on basipetal auxin transport. The reduced root elongation by ACC is at least partially dependent on functional AUX1 as evident from the F2 crosses between the recessive aux1-22 null mutant and the transgenic lines with LRC + epidermis-specific EBF2 expression (pLRC1::EBF2 and pA14::EBF2). Relative root lengths of F2 with aux1 background grown for 6 DAG in the presence of 1 μM ACC is shown. Error bars represent SD (n ≥ 4 datasets from independent crosses). Relative green intensity in the LRC of F1 crosses between the auxin transport reporter pAUX1::AUX1-YFP (confocal image at 20× magnification) and the wild type (Col-0) or transgenic line confirms that the ethylene-insensitive line has altered auxin import. (B) PIN2 is expressed in the cortex and epidermis of the root cell division zone and EZ, as it is visualized in the confocal image (20× magnification) of the reporter line pPIN2::PIN2-GFP (pin2). Relative green intensity at the middle optical section in roots of 1 μM ACC-treated seedlings vs. untreated control was measured in F1 crosses. Error bars are SD (n ≥ 10). Statistical significance between the control and the ACC-treated plants: *P ≤ 0.05; ***P ≤ 0.001 (Welch’s t test).

Fig. 7.

Integrative model of ethylene feedback on auxin homeostasis in the root tip. Ethylene perceived in the LRC and epidermis of the cell division zone regulates root growth by positive control on auxin transport and local auxin biosynthesis in the TZ. Ethylene signals perceived in the root tip relieve the suppression of CTR1 on EIN2 and EIN3/EIL1. AUX, PIN2, AUX/IAA, and TAA1 in the LRC and epidermis are indirect targets for positive regulation by EIN3/EIL1. Surface pH drops from the meristem/TZ boundary toward the zone of fast elongation (57). Low auxin concentrations stimulate and high auxin concentrations inhibit root growth, corresponding to apoplastic acidification and transient alkalinization, respectively (51), while ethylene causes growth inhibition concomitant with apoplastic alkalinization (57). Hence, through its control over auxin levels, ethylene restricts elongation growth. The symbols > and < indicate relatively higher and lower pH values compared with the adjacent zones. The model is based on findings reported elsewhere (, , –57) and in this work.