The plant cuticle is required for osmotic stress regulation of abscisic acid biosynthesis and osmotic stress tolerance in Arabidopsis

Abstract

Osmotic stress activates the biosynthesis of abscisic acid (ABA). One major step in ABA biosynthesis is the carotenoid cleavage catalyzed by a 9-cis epoxycarotenoid dioxygenase (NCED). To understand the mechanism for osmotic stress activation of ABA biosynthesis, we screened for Arabidopsis thaliana mutants that failed to induce the NCED3 gene expression in response to osmotic stress treatments. The ced1 (for 9-cis epoxycarotenoid dioxygenase defective 1) mutant isolated in this study showed markedly reduced expression of NCED3 in response to osmotic stress (polyethylene glycol) treatments compared with the wild type. Other ABA biosynthesis genes are also greatly reduced in ced1 under osmotic stress. ced1 mutant plants are very sensitive to even mild osmotic stress. Map-based cloning revealed unexpectedly that CED1 encodes a putative α/β hydrolase domain-containing protein and is allelic to the BODYGUARD gene that was recently shown to be essential for cuticle biogenesis. Further studies discovered that other cutin biosynthesis mutants are also impaired in osmotic stress induction of ABA biosynthesis genes and are sensitive to osmotic stress. Our work demonstrates that the cuticle functions not merely as a physical barrier to minimize water loss but also mediates osmotic stress signaling and tolerance by regulating ABA biosynthesis and signaling.

Figure 1.

The ced1 Mutation Decreased ABA Accumulation and the Expression of ABA Biosynthesis and Signaling Genes in Response to Osmotic Stress Treatment. (A) ABA contents in control or PEG-treated seedlings. Sixteen-day-old seedlings were transferred to plates containing 40% PEG solution for 6 h. Data are means ± se (n = 4 to 8). FW, fresh weight; WT, wild type. (B) RNA gel blot analysis of a time course of NCED3 expression in ced1 and wild-type plants after transferring to 40% PEG. Twenty micrograms of total RNA were used per lane. rRNA was used as loading control. (C) Transcript levels of ABA biosynthesis and signaling genes in response to osmotic stress in ced1 and wild-type seedlings. Sixteen-day-old plants grown on MS agar plates supplemented with 3% sucrose were transferred to plates containing PEG solutions of various concentrations and incubated for 1 h. Ethidium bromide–stained rRNAs were used as loading control.

Figure 2.

Expression of Osmotic Stress–Responsive Genes in ced1 and Wild-Type Plants. The qRT-PCR analysis was performed with total RNA from ced1 and wild-type (WT) plants under untreated control or osmotic stress treatment (40% PEG, 6 h) conditions. Real-time RT-PCR quantifications were normalized to the expression of TUB8. Error bars represent se from three independent experiments. (A) qRT-PCR analysis of representative stress-responsive genes. (B) Confirmation by qRT-PCR of the expression of genes showing significant changes between the wild type and the ced1 mutant in microarray analysis.

Figure 3.

Osmotic Stress Phenotypes of the ced1 Mutant. (A) Germination of wild-type and ced1 seeds in control or PEG-infused agar plates. Results are the average of four replicates ± se. (B) Growth of wild-type (WT) and ced1 seedlings on control (left) or −0.5 MPa PEG-infused agar plates (right). Seeds were planted on shown plates, and the pictures were taken 4 weeks after seed imbibition. (C) Wild-type and ced1 rosette plants were turgid immediately after detachment from roots, but ced1 plants became wilty 20 min later. (D) Water loss rates of detached ced1 and wild-type shoots. The snrk2.6 mutant and its wild-type background ecotype Columbia (Col-0) are shown as controls. Data are means ± se (n = 4). (E) ced1 plants were more sensitive to drought stress. Wild-type and ced1 plants were grown in soil with sufficient water for 3 weeks (Watered), and then water was withheld for 10 d (Drought) before rewatering. Plants were then allowed to recover for 2 d (Re-watered) before taking pictures. A representative picture for each treatment is shown. (F) Quantification of the survival rate of the wild type and ced1 plants in (E). Survival rates and standard deviations were calculated from the results of four independent experiments.

Figure 4.

Functional Complementation of the ced1 Mutant and Expression of CED1/BDG in Response to Osmotic Stress or ABA. (A) Complementation of ced1 by the wild-type CED1/BDG gene. The wild type (WT), ced1, and two homozygous ced1 transgenic lines (#1 and #3) expressing the wild type CED1. Plants were grown on MS (Control) or −0.5 MPa PEG-infused agar plates. The photograph was taken 4 weeks after seed imbibition. (B) RNA gel blot analysis of the CED1/BDG gene expression level in the wild type, ced1, and the two homozygous transgenic lines (#1 and #3). rRNA was used as a loading control. (C) The bdg-2 mutant was more sensitive to osmotic stress. Seeds were germinated and seedlings were grown on MS (Control) or −0.5 MPa PEG-infused agar plates for 4 weeks before taking the pictures. (D) The ced1 mutant is allelic with bdg-2. Wild-type and F1 plants (ced1 × bdg-2) were grown on MS (Control) or −0.5 MPa PEG-infused agar plates for 4 weeks before taking the pictures. (E) Regulation of CED1/BDG expression by osmotic stress and ABA. Sixteen-day-old plants were treated with 40% PEG or 100 μM ABA and incubated for the indicated time. 18S rRNA was used as a loading control.

Figure 5.

Osmotic Stress Sensitivity of Cutin Mutants. (A) Schematic diagram of a possible cutin biosynthesis pathway. Shown are the synthesis of ω-hydroxy fatty acid–based cutin monomers and the enzymes that catalyze the reactions. The relative positions of these biosynthesis steps have not been completely determined (Pollard et al., 2008). CED1/BDG is proposed to function in the polymerization of the cutin monomers and is localized extracellularly. FFA, free fatty acid. (B) Growth of wild-type (WT) and cutin mutant seedlings on MS (Control) or −0.5 MPa PEG-infused agar plates. Seeds were planted on the plates, and the pictures were taken 4 weeks after seed imbibition.

Figure 6.

Steady State Transcript Levels of ABA Biosynthesis and Signaling Genes in Cutin or Wax Mutants. Seedlings were treated with 40% PEG for the indicated time. Ten micrograms of total RNA were used per lane for the RNA gel blot analysis. rRNA was used as a loading control. Identities of probe sequences are indicated above the images, and the genotypes of different plants are indicated at left. Col-0, ecotype Columbia (wild-type plants).