A MYB transcription factor regulates very-long-chain fatty acid biosynthesis for activation of the hypersensitive cell death response in Arabidopsis

Abstract

Plant immune responses to pathogen attack include the hypersensitive response (HR), a form of programmed cell death occurring at invasion sites. We previously reported on Arabidopsis thaliana MYB30, a transcription factor that acts as a positive regulator of a cell death pathway conditioning the HR. Here, we show by microarray analyses of Arabidopsis plants misexpressing MYB30 that the genes encoding the four enzymes forming the acyl-coA elongase complex are putative MYB30 targets. The acyl-coA elongase complex synthesizes very-long-chain fatty acids (VLCFAs), and the accumulation of extracellular VLCFA-derived metabolites (leaf epidermal wax components) was affected in MYB30 knockout mutant and overexpressing lines. In the same lines, a lipid extraction procedure allowing high recovery of sphingolipids revealed changes in VLCFA contents that were amplified in response to inoculation. Finally, the exacerbated HR phenotype of MYB30-overexpressing lines was altered by the loss of function of the acyl-ACP thioesterase FATB, which causes severe defects in the supply of fatty acids for VLCFA biosynthesis. Based on these findings, we propose a model in which MYB30 modulates HR via VLCFAs by themselves, or VLCFA derivatives, as cell death messengers in plants.

Figure 1.

Changes in Gene Expression in At MYB30 Plants after Inoculation with the Xcc147 Strain. (A) Venn diagrams showing the distribution of overlapping and nonoverlapping early (1.5 to 6 h after Xcc147 inoculation [hpi]) induced candidate genes in inoculated wild-type plants compared with the T₀ time point and in inoculated MYB30ox plants compared with inoculated wild type. Two independent experiments (represented by each circle) were performed. (B) Diagram of the output of analyses performed to identify candidate genes from among the 305 identified in (A). The numbers under the heading refer to the total number of genes identified by that method of analysis. The number of genes identified by all three methods is shown in the middle of the diagram. Genes identified by two but not three of the analyses are not presented. Lists of genes included in this diagram are provided in Supplemental Tables 2 to 4 online. Antisense analysis: genes (from among the 305 candidates) that were downregulated in MYB30as plants. Hierarchical clustering: unsupervised hierarchical clustering with Ward linkage and the Spearman rank correlation-based distance (2D-SOM method). (C) Consensus list of 18 genes cited above. ^aAffymetrix Probe set number. ^bArabidopsis Genome Initiative number and corresponding putative function. The final column shows the fold changes in the mean expression levels from two independent experiments in inoculated wild type compared with the wild type at T₀ (WT₁/WT₀) and in the MYB30ox line compared with the wild type after Xcc147 inoculations (Ox/WT).

Figure 2.

Effect of MYB30 Deregulation on the Expression of Pathogen-Induced, Lipid-Related Genes after Inoculation by Avirulent Bacterial Strains Xcc147 (Left) and Pst DC3000/avrRpm1 (Right). (A) Expression pattern of four genes encoding each of the four subunits of the acyl-CoA elongase complex. These patterns are representative of those of the six genes involved in VLCFA synthesis tested (see Supplemental Figure 1 online). (B) Expression pattern of two genes involved upstream of VLCFA biosynthesis. Gene expression levels (determined by quantitative RT-PCR) are relative to the expression level in the wild type at T₀. Values represent mean ± se from three independent experiments. Gray line with gray circles, the wild type; black line with black squares, MYB30ox; dotted line with open squares, MYB30ko. hpi, hours after inoculation.

Figure 3.

Transactivation of Lipid-Related Gene Promoters by At MYB30 in Transient Assays in N. benthamiana Leaves. (A) GUS activity in leaf discs infiltrated with reporter constructs from the indicated gene either alone (white bars), with the MYB30 activator construct (hatched bars), or with the activator construct containing a truncated version of MYB30 (gray bars). These results are representative of at least six independent experiments. Mean ± se from four replicates is shown. Statistical significance according to a Student's t test P value < 0.0005 is indicated by stars. MU, methylumbelliferone. (B) GFP fluorescence of leaves 48 h after agroinfiltration with pKCS1:GFP:GUS alone or with an activator construct (as indicated). (C) Immunodetection of TAP-tagged MYB30 and MYB30ΔAD 36 h after agroinfiltration. Proteins were extracted from infiltrated leaves and analyzed by protein gel blots using anti-TAP antibody. (a) Extracts from leaves after agroinfiltration with the indicated reporter construct alone; (b) extracts from leaves after agroinfiltration with the reporter and MYB30 activator constructs; (c) extracts from leaves after agroinfiltration with the reporter and MYB30ΔAD activator constructs.

Figure 4.

Alterations of Arabidposis Epicuticular Waxes by Modulation of MYB30 Expression. (A) Chlorophyll leaching assays with mature rosette leaves immersed in 80% ethanol after different times. Gray line with gray circles, the wild type; black line with black squares, MYB30ox; black line with open diamonds, fatb-ko mutant. Values are mean ± se from five replicates; one representative experiment out of two is shown. (B) GC-MS quantification (expressed as μg/dm²) of leaf epicuticular waxes from the wild type (gray bars), MYB30ox (black bars), and MYB30ko (white bars). The top right panel presents quantities of compounds grouped by molecular families. Ald., aldehyde; Alk., alkane; Ket., ketone; Alc. 2, secondary alcohol; Alc. 1, primary alcohol. Values represent means ± se from five replicates; one representative experiment out of three is shown. (C) Relative expression measured by Q-RT-PCR of two wax-related genes after inoculation by the avirulent strains Xcc147 (left panels) and Pst DC3000/avrRpm1 (right panels). Gene expression levels are quantified relative to the expression level in the wild type at time 0. Gray line with gray circles, the wild type; black line with black squares, MYB30ox; dotted line with open squares, MYB30ko. Values represent mean ± se from three independent experiments.

Figure 5.

fatb-ko Reverses Phenotypes Conferred by MYB30 Overexpression. (A) Phenotype of fatB-ko MYB30ox, fad6 MYB30ox, and ssi2 MYB30ox double mutant lines 4 d after inoculation with Pst DC3000 carrying avrRpm1 at low inoculum concentrations. Approximately 1 cm² surface of the right half of the leaves was syringe infiltrated. Selected double mutant lines showing representative symptoms are shown. (B) Relative expression level of the PR1 gene measured by quantitative RT-PCR after inoculation by Pst DC3000 carrying avrRpm1. Gene expression levels are quantified relative to the expression level in the wild type at time 0. Gray line with gray circles, the wild type; black line with black squares, MYB30ox; black line with open diamonds, fatb-ko; dotted line with black diamonds, fatb-ko MYB30ox double mutant lines. Values represent mean ± se from three independent experiments.

Figure 6.

Relationships between MYB30 and FA Desaturation and the Jasmonate Signaling Pathways. (A) Expression pattern of genes involved in plastidial FA biology: de novo PUFA (FAD3) and FA synthesis (SSI2) after inoculation by Xcc147 (left panel) or Pst DC3000/avrRpm1 (right panel). Gene expression levels (determined by quantitative RT-PCR) are relative to the expression level in the wild type at time 0. Gray line with gray circles, the wild type; black line with black squares, MYB30ox; dotted line with open squares, MYB30ko. Values represent mean ± se from three independent experiments. (B) GC-MS measurement of OPDA and JA in the wild type (gray bars), MYB30ko (white bars), and MYB30ox (black bars) plants after inoculation by the avirulent DC3000/avrRpm1 Pst strain. A representative experiment (out of two independent experiments) including means ± sd from three replicates is shown. (C) Relative expression of JA pathway marker genes associated with wounding (VSP1 and LOX3) and pathogen response (PDF1-2 and PR4) by quantitative RT-PCR. Gene expression levels are quantified relative to the expression level in the wild type at time 0. Plants were inoculated by the avirulent DC3000/avrRpm1 Pst strain. Gray line with gray circles, the wild type; black line with black squares, MYB30ox; dotted line with open squares, MYB30ko. One representative experiment out of four is shown. Data for MYB30ko were standardized with the wild-type data to appear on the same graph. (D) Phenotypes of MYB30ox jar1 double mutants, parental lines, and the wild type 4 d after inoculation by Pst DC3000/avrRpm1 at low inoculum concentrations. Approximately 1 cm² surface of the right half of the leaves was syringe infiltrated. Two double mutant lines out of five identified, showing representative symptoms, are shown.

Figure 7.

A Schematic Overview of Metabolic Pathways Regulated by MYB30 during the Incompatible Interaction between Arabidopsis and Avirulent Bacterial Pathogens. Elements strongly (maybe directly) and positively regulated by MYB30 are indicated in red (genes of the acyl-CoA elongase complex, accumulation of VLCFAs, accumulation of SA, and cell death), other elements positively but weakly regulated by MYB30 are indicated in orange (upstream steps of VLCFA synthesis, wax- and cutin-related genes, and wax accumulation), elements shown in gray are not significantly regulated by MYB30, at least during the early events of the interaction, and elements shown in blue are negatively regulated (directly or indirectly) by MYB30 (expression of genes encoding FA desaturases). We hypothesize that inoculation triggers the synthesis of new signaling molecules as indicated by the dotted arrow. These molecules, together with other lipid-derived signals, such as salicylic acid, oxylipins, JA, and phospholipids, would subsequently act for activation of the hypersensitive cell death. Synthesis of these VLCFA-derivated signals may be enhanced by MYB30. Definitely, upregulation of VLCFA synthesis in nonchallenged MYB30ox plants allows an enhanced accumulation of VLCFAs and of VLCFA metabolizing enzymes. The cell is thus predisposed to respond stronger and faster to pathogen attack. PM, plama membrane; CW, cell wall.