An inositolphosphorylceramide synthase is involved in regulation of plant programmed cell death associated with defense in Arabidopsis

Abstract

The Arabidopsis thaliana resistance gene RPW8 triggers the hypersensitive response (HR) to restrict powdery mildew infection via the salicylic acid-dependent signaling pathway. To further understand how RPW8 signaling is regulated, we have conducted a genetic screen to identify mutations enhancing RPW8-mediated HR-like cell death (designated erh). Here, we report the isolation and characterization of the Arabidopsis erh1 mutant, in which the At2g37940 locus is knocked out by a T-DNA insertion. Loss of function of ERH1 results in salicylic acid accumulation, enhanced transcription of RPW8 and RPW8-dependent spontaneous HR-like cell death in leaf tissues, and reduction in plant stature. Sequence analysis suggests that ERH1 may encode the long-sought Arabidopsis functional homolog of yeast and protozoan inositolphosphorylceramide synthase (IPCS), which converts ceramide to inositolphosphorylceramide. Indeed, ERH1 is able to rescue the yeast aur1 mutant, which lacks the IPCS, and the erh1 mutant plants display reduced ( approximately 53% of wild type) levels of leaf IPCS activity, indicating that ERH1 encodes a plant IPCS. Consistent with its biochemical function, the erh1 mutation causes ceramide accumulation in plants expressing RPW8. These data reinforce the concept that sphingolipid metabolism (specifically, ceramide accumulation) plays an important role in modulating plant programmed cell death associated with defense.

Figure 1.

Isolation and Characterization of erh1. (A) Soil-grown seedlings of the parental line (ST8) and the erh1/ST8 mutant. A single representative leaf (indicated by red lines) stained with trypan blue is shown beside each seedling. The inset shows a cluster of dead cells at five times magnification. (B) erh1-enhanced cell death is suppressed in plants grown on MS agar medium. Plants were grown in MS agar medium for 4 weeks and then transplanted into autoclaved soil. Photographs were taken at 0 or 5 d posttransplanting (dpt). (C) RPW8 expression is enhanced by erh1. Four-week-old plants of the indicated genotypes were transplanted from MS agar medium to autoclaved soil. Total RNA was extracted from fully expanded mature leaves at 0 or 5 dpt, gel-blotted, and probed with a ³²P-labeled cDNA mixture of RPW8.1, RPW8.2, and PR1 sequentially. rRNA is shown as a loading control. (D) erh1-induced cell death is largely RPW8-dependent. The erh1 mutation from erh1/ST8 was introduced into Col-gl and S5 (carrying a single copy of RPW8) by crossing. Photographs were taken at 5 weeks old. Note the SHL (indicated by the red arrow) and reduced stature of erh1/S5. (E) erh1 in Col-gl results in reduced susceptibility to powdery mildew. Five-week-old, soil-grown plants of Col-gl and erh1/Col-gl were inoculated with G. cichoracearum UCSC1. Representative infected leaves from the indicated genotypes at 10 dpi are shown. (F) Infected leaves of six plants from each genotype used in (E) were subjected to quantitative measurement of disease susceptibility (see Methods for details) at 10 dpi. Data represent means ± se of three replicate experiments (P < 0.05 based on Student's t test). (G) erh1/ST8 has more pronounced HR in response to powdery mildew infection. Five-week-old plants were transplanted from MS agar medium to autoclaved soil and maintained at >90% RH. Plants were inoculated with G. cichoracearum UCSC1 immediately after transplanting. At 2 dpi, inoculated leaves were stained with trypan blue. Arrows indicate cell death induced by the fungus. Bars = 100 μm.

Figure 2.

Involvement of SA in the erh1-Mediated Cell Death Phenotype. (A) The erh1 mutation results in SA accumulation. Leaves of 6-week-old plants of the indicated genotypes were assayed for free and total SA. The amount of conjugated SA is the subtraction of free SA from total SA. Data represent means ± se from three duplicate samples in one experiment. Asterisks indicate that the value of the erh1 mutant is significantly different (P < 0.01) from that of the respective wild-type lines based on Student's t test. This experiment was repeated twice with similar results. (B) erh1-activated SHL is SA-dependent. erh1 and erh1 plus RPW8 (from line S5) were introduced into the NahG background by crossing, and respective F3 lines homozygous for erh1 or erh1/RPW8 and NahG from the respective crosses were examined for SHL by trypan blue staining at 5 weeks old. [See online article for color version of this figure.]

Figure 3.

Cloning of ERH1 and Subcellular Localization of the Protein. (A) Schematic gene structure of ERH1 (At2g37940) and the position of the T-DNA insertion. Two short red bars indicate the positions of the primers used for RT-PCR in (B). (B) RT-PCR with the parental line and two erh1/ST8 siblings for ERH1 and ACT2. (C) Genetic complementation of erh1 with ERH1 under the control of the native promoter (NP). (D) Predicted protein structure of ERH1. PAP2, phosphatidic acid phosphatase–related2; TM, transmembrane. D3 and D4 are two conserved domains found in functional homologs. (E) Alignment of two conserved motifs, D3 and D4, from predicted plant ERH1-like protein sequences, a kinetoplastid (L. major) IPCS, human SMS1, and yeast (Saccharomyces cerevisiae) AUR1p. Identical residues are highlighted in red, and conservative residues are highlighted in black or gray. Underlined in the consensus are three residues that form a catalytic triad required for enzymatic function (Huitema et al., 2004). (F) Subcellular localization of ERH1. Plasmid DNA of 35S:ERH1-DsRed was mixed in equal amounts with that of NP:RPW8.2-YFP or 35S:GFP-SYP61 (a cis-Golgi marker) or other constructs (see Supplemental Figure 4C online) and introduced into Col-0 leaf epidermal cells by bombardment (see Methods). Expression of the fluorescence-tagged proteins was imaged with a laser confocal microscope at 24 h after bombardment. Bars = 10 μm.

Figure 4.

ERH1 Encodes an IPCS. (A) Complementation of the yeast aur1 mutation by ERH1. ERH1 was cloned in the pADH-LEU2 plasmid. The parental and the recombinant plasmids were introduced into three yeast strains lacking AUR1p alone (aur1Δ), AUR1p and ceramide-associated fatty acyl chain C2-hydroxylase (aur1Δ/scs7Δ), or AUR1p along with sphinganine hydroxylase (aur1Δ/sur2Δ). The respective yeast cells were grown on SD medium (−Leu) containing FOA. (B) Yeast cells expressing ERH1 possess IPCS activities. Microsomes were prepared from aur1Δ yeast cells harboring pAUR1 or pERH1 and assayed for IPCS activities using the indicated ceramide substrates. Note that IPCS activities are reflected by the production of IPC (bottom bands) from the ceramide substrate (strong top bands). Cer, ceramide; Sph, sphingosine. (C) In vitro assays for plant IPCS activity. Leaf lysates were prepared from 6-week-old plants and subjected to IPCS assays using dodecanoyl-NBD-d-erythro-DHS as a substrate. IPCS activity was measured using relative fluorescence units. Data are means ± se (P < 0.015 based on Student's t test of the values of the two genotypes; n = 4). (D) erh1 results in ceramide accumulation in plants expressing RPW8. Sphingolipids were extracted from leaves of 8-week-old plants of the indicated genotypes and measured essentially as described previously (Markham and Jaworski, 2007). Values are means ± se (n = 5). Asterisks indicate significance at P < 0.01 compared with S5 based on Student's t test. dw, dry weight.

Figure 5.

Induction of ERH1. (A) RPW8 overexpression leads to enhanced ERH1 expression. Expression levels of ERH1 in 6-week-old plants (except S24, which was grown in MS agar for 5 weeks and transplanted to soil for 5 d) were measured by quantitative RT-PCR using TaqMan chemistry. The relative mRNA levels in RPW8-expressing lines were calculated relative to that of Col-gl (1.0). Data represent means ± se from three replicate experiments. Asterisks indicate significance at P < 0.01 compared with Col-gl based on Student's t test. (B) ERH1 is induced to higher levels by powdery mildew. Six-week-old plants were inoculated with G. cichoracearum UCSC1. cDNA was synthesized from mRNA extracted from leaf tissues of the indicated genotypes at 0, 3, or 5 dpi and used for transcript quantification by real-time RT-PCR. The relative mRNA levels were calculated relative to that of Col-gl at 0 dpi (1.0). Data represent means ± se from three replicate experiments. Asterisks indicate significance at P < 0.01 compared with Col-gl based on Student's t test. (C) and (D) ERH1 is induced by avirulent P. syringae. Total RNA was extracted from leaves of 6-week-old Col-0 plants infiltrated with the virulent P. syringae pv maculicola ES4326 strain (OD₆₀₀ = 0.0002) (C) or the avirulent strain carrying AvrRpm1 (D) at the indicated time points, gel-blotted, and probed with ³²P-labeled ERH1 or PR1 cDNA. hpi, hours postinoculation.

Figure 6.

Additive/Synergistic Effects of acd5 and erh1 on Cell Death. (A) acd5 has enhanced resistance to powdery mildew. Five-week-old plants were inoculated with G. cichoracearum UCSC1, and disease phenotypes at 8 dpi were shown by two representative leaves for each genotype. Powdery mildew infection can be seen as white powdery coating on Col-0 leaves. (B) Six-week-old, short-day-grown plants of the indicated genotypes. Note the leaf cell death and reduced plant stature of acd5/S5. (C) Four-week-old, short-day-grown plants of the indicated genotypes. Note the leaf cell death and reduced plant stature of erh1/acd5. (D) Two-week-old, short-day-grown plants of the indicated genotypes. Note the collapsed cotyledons and stunted growth of erh1/acd5/S5. (E) Silencing a rice ERH1 homolog leads to cell death and reduced plant stature. Two-and-one-half-months-old wild-type Nipponbare rice and one representative rice line transgenic for an RNAi construct targeting a rice ERH1 homolog (Os01g0850100) are shown. Arrows indicate dead leaves. (F) RT-PCR analysis of the indicated genes from four rice transgenic lines along with the parental wild type.