Lazarus1, a DUF300 protein, contributes to programmed cell death associated with Arabidopsis acd11 and the hypersensitive response

Abstract

Background: Programmed cell death (PCD) is a necessary part of the life of multi-cellular organisms. A type of plant PCD is the defensive hypersensitive response (HR) elicited via recognition of a pathogen by host resistance (R) proteins. The lethal, recessive accelerated cell death 11 (acd11) mutant exhibits HR-like accelerated cell death, and cell death execution in acd11 shares genetic requirements for HR execution triggered by one subclass of R proteins.

Methodology/principal findings: To identify genes required for this PCD pathway, we conducted a genetic screen for suppressors of acd11, here called lazarus (laz) mutants. In addition to known suppressors of R protein-mediated HR, we isolated 13 novel complementation groups of dominant and recessive laz mutants. Here we describe laz1, which encodes a protein with a domain of unknown function (DUF300), and demonstrate that LAZ1 contributes to HR PCD conditioned by the Toll/interleukin-1 (TIR)-type R protein RPS4 and by the coiled-coil (CC)-type R protein RPM1. Using a yeast-based topology assay, we also provide evidence that LAZ1 is a six transmembrane protein with structural similarities to the human tumor suppressor TMEM34. Finally, we demonstrate by transient expression of reporter fusions in protoplasts that localization of LAZ1 is distributed between the cytosol, the plasma membrane and FM4-64 stained vesicles.

Conclusions/significance: Our findings indicate that LAZ1 functions as a regulator or effector of plant PCD associated with the HR, in addition to its role in acd11-related death. Furthermore, the similar topology of a plant and human DUF300 proteins suggests similar functions in PCD across the eukaryotic kingdoms, although a direct role for TMEM34 in cell death control remains to be established. Finally, the subcellular localization pattern of LAZ1 suggests that it may have transport functions for yet unknown, death-related signaling molecules at the plasma membrane and/or endosomal compartments. In summary, our results validate the utility of the large-scale suppressor screen to identify novel components with functions in plant PCD, which may also have implications for deciphering cell death mechanisms in other organisms.

Figure 1. laz1-mediated suppression of cell death in acd11.

(A) Suppression of BTH-induced acd11 cell death in the four laz1 alleles in comparison to the acd11/nahG background. Two-week-old plants were sprayed with 100 µM BTH and photographed 7 days after treatment. Untreated control plants are shown in the upper panel and BTH treated plants in the lower panel. Scale bars represent 1 cm. (B) Suppression of cell death in acd11 after out-crossing the nahG transgene. acd11/laz1-1 plants were grown under SD for four weeks and compared to acd11, acd11/eds1-2, acd11/pad4-2 and wild type Ler (upper panels). Trypan blue staining (lower panels) of younger (dashed circles) and older (solid circles) leaves revealed numerous dead cell foci in acd11 compared to their marked reduction in acd11/laz1, and apparent absence in acd11/eds, acd11/pad4 and wild type Ler. Bar indicates 1 cm.

Figure 2. Transcriptome analysis of laz1.

(A) Comparison of the global expression fold change between acd11/nahG and nahG with the fold change between acd11/nahG/laz1-1 and acd11/nahG 72 hr after BTH treatment. This showed a strong opposite response with a Pearson correlation of −0.93. (B, C) Box plots of the expression of the 500 most significantly up- (B) or down-regulated (C) genes in BTH-treated acd11/nahG. The genes exhibit only moderate responses in acd11/nahG/laz1 plants.

Figure 3. Map-Based Cloning of LAZ1.

(A) Map of the LAZ1 locus on chromosome 4. Positions of markers on three BACs for mapping laz1-1 to a 200 kb interval are indicated (see Table S2 for primers) with the numbers of recombinants. Lower panel shows differential expression of genes within the interval measured by transcript profiling. This identified At4g38360 as the gene in the interval with the most reduced expression between acd11/nahG/laz1-1 and acd11/nahG (log2 fold change with expression intensity ceiling set to 0). The probe set for At4g38360 is indicated. (B) Expression profiling of At4g38360 depicted relative to Ler before treatment (0 hrs). As determined by 2-way ANOVA with the factors treatment and genotype, At4g38360 transcripts accumulated (p-value = 0.018) in acd11/nahG upon BTH treatment (72 hrs), but were not induced in acd11/nahG/laz1-1 or background controls. Similarly, At4g38360 was significantly induced in the acd11 mutant (p = 0.00059; determined by t-test, data not shown). (C) Structure of LAZ1 including positions of the laz1 mutations identified in the screen and the T-DNA integration site of laz1-5 (SALK_034193). The coding region and protein domains (DUF300, domain of unknown function 300; CC, coiled-coil) are boxed. laz1-1 is a γ-induced, one-base pair deletion in exon 6 leading to a premature stop codon (see Figure S3A). laz1-4 has an EMS-derived G->A resulting in conversion of a plant conserved Aspartate to Asparagine (D360N). The base pair changes in DEB induced laz1-2 (T->A) and EMS-induced laz1-3 (G->A) lead to mutations in splice donor sites of introns 1 and 3, causing deletions in preceeding exons (see Figure S3B–C).

Figure 4. laz1 suppression of HR cell death.

Ion leakage assays, with mean and standard errors (SE) in A–C calculated from four disks per treatment with four replicates within an experiment, after: (A) Pto DC3000 (avrRps4) infection of laz1-5 compared to Col-0, eds1-2 and rar1-28. (B) Pto DC3000 (avrRpt2) infection of laz1-5 compared to Col-0, ndr1-1 and rar1-28. (C) Pto DC3000 (avrRpm1) infection of laz1-5 compared to Col-0 and rar1-28.

Figure 5. LAZ1 topology.

(A) Six LAZ1 forms (1–6) were C-terminally fused to the dual SUC2/HIS4C reporter to determine their orientation in membranes. These LAZ1 fusions were chosen to assess the validity of predicted transmembrane regions (Figure S4). Lengths and positions of the fusions are indicated by lines and stars, respectively. (B) Growth of yeast harboring the six LAZ1 reporter fusions on histidinol containing medium indicates cytosolic localization of the reporter due to its histidinol dehydrogenase activity. YBR210w (ER localized C-terminus) and YAL007c (cytosolic localized C-terminus) served as controls . (C) The glycosylation status of six LAZ1 fusions (Fig. 5A) was determined by comparing the sizes of endoglycosidase H (endo H) treated and untreated samples followed by Western blotting using anti-HA antibody. Faster migrating (lower) bands after endo H treatment indicate that the fusion is glycosylated and the SUC2/HIS3 C-terminal reporter resides in the ER. YBR210w (ER localized C-terminus) and YAL007c (cytosolic localized C-terminus) served as controls .

Figure 6. Subcellular localisation of LAZ1.

(A) Confocal images of Arabidopsis mesophyll protoplast detect the distribution of YFP-LAZ1 between cytosol, plasma-membrane and vesicle-like structures. YFP-LAZ1 localisation at plasma membrane and vesicle compartments (arrows) was confirmed by co-staining with the membrane selective dye FM4–64. Enlargement of boxed region highlights the overlap between YFP-LAZ1 and FM4–64 staining in endosomal compartments. Size bar corresponds to 10 µm. (B) YFP-LAZ(D360N) localization is primarily cytosolic and shows no overlap with FM4–64 staining. Size bar represents 10 µm.