Salt Enhances Disease Resistance and Suppresses Cell Death in Ceramide Kinase Mutants

Abstract

Sphingolipids act as structural components of cellular membranes and as signals in a variety of plant developmental processes and defense responses, including programmed cell death. Recent studies have uncovered an interplay between abiotic or biotic stress and programmed cell death. In a previous study, we characterized an Arabidopsis (Arabidopsis thaliana) cell-death mutant, accelerated cell death5 (acd5), which accumulates ceramides and exhibits spontaneous cell death late in development. In this work, we report that salt (NaCl) treatment inhibits cell death in the acd5 mutant and prevents the accumulation of sphingolipids. Exogenous application of abscisic acid (ABA) and the salicylic acid (SA) analog benzothiadiazole demonstrated that the effect of NaCl was partly dependent on the antagonistic interaction between endogenous SA and ABA. However, the use of mutants deficient in the ABA pathway suggested that the intact ABA pathway may not be required for this effect. Furthermore, pretreatment with salt enhanced the resistance response to biotic stress, and this enhanced resistance did not involve the pathogen-associated molecular pattern-triggered immune response. Taken together, our findings indicate that salt inhibits sphingolipid accumulation and cell death in acd5 mutants partly via a mechanism that depends on SA and ABA antagonistic interaction, and enhances disease resistance independent of pattern-triggered immune responses.

Figure 1.

Phenotypes of the wild type (WT) and the acd5 mutant after salt treatments. A–D, Three-week–old wild-type and acd5 plants were irrigated with water or the indicated concentration of NaCl once to soil capacity and then irrigated routinely with water. A, Representative images of wild-type and acd5 plants 21 d after NaCl treatments. Scale bars = 1 cm. The inset depicts the typical acd5 cell death lesion; inner scale bars = 4 mm. This experiment was repeated three times with independent samples. More than 75 plants were observed for each treatment. B, Statistical analysis of the inhibitory effects of different concentrations of NaCl on the acd5 mutant. The percentage of acd5 plants exhibiting cell death lesions was recorded and calculated 17 d after NaCl treatment. Values are means ± se from triplicate biological repeats (n ≥ 28). Significant differences were determined by ANOVA post hoc test (P < 0.05). Note that the cell-death phenotype was barely visible in acd5 plants under 300-mm NaCl treatment. C and D, Representative images of stained wild type and acd5 for detection of H₂O₂ and cell death 17 d after the 300-mm NaCl treatment. Leaves were stained with DAB to detect H₂O₂ or trypan blue to detect cell death. Scale bars = 1 mm. The brown precipitate in (C) indicates DAB oxidation at the site of H₂O₂ accumulation. The blue spots in (D) indicate the dead cells visualized by trypan blue staining. At least 12 leaves were stained in each treatment. These experiments were repeated twice using independent samples. Note that no DAB or trypan blue staining was observed in the acd5 leaves after NaCl treatment. E, After the emergence of the cell-death phenotype, 25-d–old acd5 plants were treated with water or 300-mm NaCl. The two square inset images were recorded the phenotypes at 25 d post planting. The two main and two small rectangular images were recorded leaf and stem phenotypes 12 d after the treatment. Scale bars = 1 cm. For the inset, scale bars = 2 mm. Red arrows indicate the lesions in the acd5 mutant. These experiments were done with three biological repeats. Note that no cell death lesions were observed in the stem after NaCl treatment (right).

Figure 2.

Sphingolipid profiles in the wild-type (WT) and acd5 plants after NaCl treatment. Three-week–old wild-type and acd5 plants were irrigated with water or 300-mm NaCl. Sphingolipids were extracted from rosette leaves of plants 8 d after the treatment. Values are means ± se from triplicate technical repeats. This experiment was repeated twice using independent samples. Data sets marked with different letters indicate significant differences determined by ANOVA post hoc test (P < 0.01). A, Total contents of LCBs, ceramides, hydroxyceramides, glucosylceramides, and GIPCs in wild-type and acd5 plants after NaCl treatment. B, Measurement of ceramide species with LCB moieties (left) and FA moieties (right). C, Measurement of hydroxyceramide species with LCB moieties (left) and FA moieties (right).

Figure 3.

Influence of NaCl on ABA and SA pathways in the wild-type (WT) and acd5 plants. Three-week–old wild-type and acd5 plants were irrigated with water or 300-mm NaCl. The rosette leaves were collected at the indicated time points and used for RNA isolation and phytohormone extraction. These experiments were repeated at least twice using independent samples. Values are means ± se from triplicate technical repeats. Significant differences were determined by ANOVA post hoc tests as P < 0.05 in (A) and (B) and P < 0.01 in (C–E) using different letters. A and B, Three days after NaCl treatment, the expression of the ABA biosynthesis gene NCED3 (At3g14440), ABA downstream signaling genes RAB18 (At5g66400), RD22 (At5g25610), and COR15A (At2g42540), SA-pathway–related genes SID1 (At4g39030), SID2 (At1g74710), and PAD4 (At3g52430), and the SA downstream marker gene PR1 (At2g14610) were measured by RT-qPCR. ACTIN2 (At3g18780) transcript levels were used as the internal control. The gene expression values presented are relative to average water-treated wild-type levels (set as “1”). C, Fluctuation of the ABA levels in wild-type and acd5 plants after NaCl treatment. D and E, The accumulation of free SA (D) and total SA (E) in wild-type and acd5 plants after NaCl treatment.

Figure 4.

ABA and NaCl suppress the cell-death phenotype of the acd5 mutant after induction by BTH. A and B, Three-week–old plants were sprayed with (+) or without (−) 400-μm ABA for 24 h, then subsequently sprayed with (+) or without (−) 300-μm BTH. Representative photos were recorded 9 d after the BTH treatment (A). Scale bars = 1 cm. B, Statistical analysis of the inhibitory effects of exogenous ABA on the acd5 mutant cell-death phenotype after BTH induction as shown in (A). The percentage of acd5 plants exhibiting cell death lesions in the total treated acd5 plants was recorded. At least 24 plants were tested for each treatment. Values are means ± se from triplicate biological repeats. Significant differences were determined by ANOVA post hoc test (P < 0.05). C and D, Three-week–old plants were irrigated with (+) or without (−) 300-mm NaCl for 24 h, then sprayed with (+) or without (−) 300-μm BTH. Representative photos were recorded 9 d after the BTH treatment (C). Scale bars = 1 cm. D, Statistical analysis of the inhibitory effects of NaCl on the acd5 mutant cell-death phenotype after BTH induction as shown in (C). The percentage of acd5 plants exhibiting cell death lesions in the total treated acd5 plants was recorded. At least 34 plants were tested for each treatment. Values are means ± se from triplicate biological repeats. Significant differences were determined by ANOVA post hoc test (P < 0.05). E, Total ceramides and hydroxyceramides in wild-type (WT) and acd5 plants upon the combined treatments with NaCl and BTH. The rosette leaves were collected 5 d post BTH treatment. This experiment was repeated twice using independent samples. Values are means ± se from triplicate biological repeats. Data sets marked with different letters indicate significant differences determined by ANOVA post hoc test (P < 0.01).

Figure 5.

Effect of ABA pathway mutations on the cell-death phenotype of the acd5 mutant after NaCl treatment. Three-week–old plants of wild-type (WT), acd5, abi4-t, abi2-2, and the double mutants abi4-t acd5 and abi2-2 acd5 were irrigated with or without 300-mm NaCl. Phenotypes were recorded 9 d after treatment. Scale bars = 1 cm. At least 16 plants per line were tested each time. Arrows indicate cell death lesions. This experiment was repeated three times using independent samples.

Figure 6.

NaCl treatment enhanced plant resistance to bacterial infection. Three-week–old wild-type (WT) and acd5 plants were treated with or without 300-mm NaCl. After 24 h of treatment, leaves were infiltrated with the virulent Psm DG3 at OD₆₀₀ = 0.0005. The experiment was repeated three times by using independent samples. A, Representative leaves 4 d after Psm DG3 inoculation. At least 32 plants were infiltrated for each case. Scale bars = 5 mm. B, Representative microscopic images of leaves post Psm DG3 infection. At 2 d post inoculation, at least nine infected leaves were detached and stained with trypan blue to observe cell death. Scale bars = 1 mm. C and D, Leaf discs from 3-week–old wild type, acd5, and NahG plants were harvested for quantification of bacterial growth at 3 d post infiltration (C) or 2 d post infiltration (D). Bacterial growth was measured as colony-forming units (cfu)/disc of leaf tissue. Values are means ± se from at least six biological replicates. Significant differences were determined by ANOVA post hoc tests (P < 0.05) using different letters.

Figure 7.

Effect of NaCl pretreatment on the pattern-triggered immune response. Three-week–old wild-type (WT) and acd5 plants were irrigated with or without 300-mm NaCl. A to D, At 24 h after irrigation with or without 300-mm NaCl, the plants were inoculated with the Pst DC3000 mutant hrcC (OD₆₀₀ 0.01) for 24 h (A and B) or 100 nm of the biologically active epitope of bacterial flagellin flg22 for 12 h (C and D). Scale bars = 200 μm (A and C). Images were photographed by fluorescence microscopy and the callose deposition (the white spots) was calculated by the software ImageJ (National Institutes of Health). Values in (B) and (D) = means ± se from at least eight biological repeats. These experiments were repeated three times. Significant differences were determined by ANOVA post hoc tests (P < 0.05) using different letters. E, ROS production induced by flg22. At 24 h after the water or NaCl treatment, leaf discs (diameter = 0.5 mm) from third to fifth leaves were treated with 100-nm flg22 or mock (distilled water). ROS were recorded by a luminol-based assay, presented as relative light units (RLUs). Data represent means ± se from eight biological replicates. Significant differences were determined by ANOVA post hoc tests (P < 0.05) using different letters. This experiment was repeated three times using independent samples. F, The relative transcript level of FRK1 (At2g19190) after flg22 induction. At 24 h after the water or 300-mm NaCl treatment, 100-nm flg22 was infiltrated into the third to fifth leaves. The RNA samples were collected 3-h post infiltration. Data represent means ± se from triplicate technical replicates. Significant differences were determined by ANOVA post hoc tests (P < 0.05) using different letters. This experiment was repeated twice using independent samples.

Figure 8.

Effects of NaCl on Pro, electrolyte leakage of plants, and Psm DG3 growth. Three-week–old plants were irrigated with 300-mm NaCl or water. Rosette leaves were harvested to extract free Pro or to detect relative electrolyte leakage at the indicated time points. These experiments were repeated twice using independent samples. A, The accumulation of free l-Pro in wild-type (WT) and acd5 plants. Data represent means ± se from triplicate biological repeats. B, The electrolyte leakage of wild-type and acd5 plants was detected at the indicated time points. Data represent means ± se from triplicate biological repeats. Significant differences were determined by ANOVA post hoc tests (P < 0.01) using different letters. C, Effect of NaCl on the growth of Psm DG3 in vitro. Different concentrations of NaCl were added to the bacterial liquid culture medium. The abundance of Psm DG3 was measured by a spectrophotometer at OD₆₀₀ at the indicated time points. Data represent means ± se from four biological repeats.