Pseudomonas syringae Effector Avirulence Protein E Localizes to the Host Plasma Membrane and Down-Regulates the Expression of the NONRACE-SPECIFIC DISEASE RESISTANCE1/HARPIN-INDUCED1-LIKE13 Gene Required for Antibacterial Immunity in Arabidopsis

Abstract

Many bacterial pathogens of plants and animals deliver effector proteins into host cells to promote infection. Elucidation of how pathogen effector proteins function not only is critical for understanding bacterial pathogenesis but also provides a useful tool in discovering the functions of host genes. In this study, we characterized the Pseudomonas syringae pv tomato DC3000 effector protein Avirulence Protein E (AvrE), the founding member of a widely distributed, yet functionally enigmatic, bacterial effector family. We show that AvrE is localized in the plasma membrane (PM) and PM-associated vesicle-like structures in the plant cell. AvrE contains two physically interacting domains, and the amino-terminal portion contains a PM-localization signal. Genome-wide microarray analysis indicates that AvrE, as well as the functionally redundant effector Hypersensitive response and pathogenicity-dependent Outer Protein M1, down-regulates the expression of the NONRACE-SPECIFIC DISEASE RESISTANCE1/HARPIN-INDUCED1-LIKE13 (NHL13) gene in Arabidopsis (Arabidopsis thaliana). Mutational analysis shows that NHL13 is required for plant immunity, as the nhl13 mutant plant displayed enhanced disease susceptibility. Our results defined the action site of one of the most important bacterial virulence proteins in plants and the antibacterial immunity function of the NHL13 gene.

Figure 1.

Basal expression of DEX:His:avrE in transgenic Arabidopsis complemented the growth of the ∆CEL mutant. A, Western-blot analysis of His:AvrE proteins in transgenic Arabidopsis plants 6 h after 30 µm DEX treatment. A polyclonal AvrE antibody was used to detect AvrE. The red arrow indicates the predicted size of AvrE. B, Representative images of 4-week-old Col-0 and transgenic DEX:His:avrE plants. C, Basal expression of DEX:His:avrE in transgenic Arabidopsis lines L1 and L6 is sufficient to substantially enhance the growth of the ∆CEL mutant. DC3000, ∆CEL, and hrcC were hand infiltrated at 1 × 10⁶ colony-forming units (cfu) mL⁻¹. Bacterial numbers were counted 3 d after infiltration. Asterisks indicate significant differences between Col-0 and L1 or L6: **, P < 0.01; and *, P < 0.05.

Figure 2.

AvrE is localized at the PM and PM-associated vesicle-like structures in the plant cell. A, Subcellular fractionation of His:AvrE expressed in transgenic Arabidopsis plants. C, Crude plant extract; T, total plant protein; S, soluble protein fraction; M, total membrane protein fraction. The red arrow indicates the His:AvrE band. B and C, Confocal images of His:YFP:AvrE expressed in leaf cells of transgenic Arabidopsis plants. The yellow arrow indicates the PM. C is a z-stacked image. D, Confocal image (z-stack) of His:YFP:AvrE expressed in leaf cells of transgenic Arabidopsis plants after plasmolysis using 0.5 m NaCl. The red arrow indicates PM Hechtian strands, and the yellow arrows indicate the PM. E, Confocal image (z-stack) of AvrE:VFP expressed in N. benthamiana leaf cells. F and G, Confocal images of His:YFP:AvrE-ee (F) and His:YFP:AvrE-kk (G) expressed in leaf cells of transgenic Arabidopsis plants. Bars = 10 µm.

Figure 3.

Serial deletion analysis reveals a putative two-domain structure of AvrE. A, Diagram showing the AvrE deletion constructs used in this study. B and C, Necrosis-inducing activity of AvrE proteins transiently expressed in tobacco leaves. Numbering is as follows: 1, empty vector; 2, AvrE_1-200aa; 3, AvrE_1-400aa; 4, AvrE_1-596aa; 5, AvrE_1-995aa; 6, His:AvrE; 7, AvrE_1-1200aa; 8, AvrE_1-1400aa; 9, AvrE_1-1600aa; 10, AvrE_200-1795aa; 11, AvrE_400-1795aa; 12, AvrE_600-1795aa; 13, AvrE_800-1795aa; 14, AvrE_990-1795aa; 15, AvrE_1200-1795aa; 16, AvrE_1400-1795aa; and 17, AvrE_1600-1795aa. Necrosis is characterized by tissue collapse and discoloration (e.g. gray color) in the infiltrated area. D, Combinations of AvrE deletion derivatives used for transient coexpression experiments in tobacco. For each combination, two Agrobacterium tumefaciens strains (optical density at 600 nm [OD₆₀₀] = 0.1) were mixed at a 1:1 ratio and infiltrated into tobacco leaves. E, Necrosis (indicated by tissue collapse with gray color) induced by combinations of truncated AvrE proteins. Photographs were taken 2 d after 10 µm DEX spray. Numbering is the same as shown in D.

Figure 4.

The two domains of AvrE physically interact, and the N-terminal domain contains a PM localization signal. A, Coimmunoprecipitation shows that AvrE-N and AvrE-C interact in N. benthamiana. The affinity-purified AvrE-N antibody was used to pull down AvrE-N. AvrE-N and AvrE-C were detected by AvrE-N and AvrE-C antibodies, respectively. IP, Immunoprecipitate. B to E, Confocal images of GFP:AvrE-N in N. benthamiana leaf cells (B), GFP:AvrE-C in N. benthamiana leaf cells (z-stack; C), GFP:AvrE-N coexpressed with AvrE-C in N. benthamiana leaf cells (D), and GFP:AvrE-C protein coexpressed with AvrE-N in N. benthamiana leaf cells (z-stack; E). Expression of GFP:AvrE-N and GFP:AvrE-C was under the control of the cauliflower mosaic virus 35S promoter (pSITEII vector). A. tumefaciens containing GFP:avrE-N/C and DEX:avrE-C/N constructs was adjusted to OD₆₀₀ = 0.1 and mixed at a ratio of 1:2 (GFP:avrE-N/C versus DEX:avrE-C/N) for infiltration. Twenty-four hours after A. tumefaciens infiltration, 10 µm DEX was sprayed to induce the expression of untagged AvrE-N and AvrE-C proteins. Confocal images were collected 5 h after DEX treatment. Bars = 20 µm. F, A simplified model showing the two physically interacting domains of AvrE and their localization in the plant cell. C, AvrE-C; N, AvrE-N.

Figure 5.

Microarray analysis of the virulence effects of bacterium-delivered AvrE and HopM1 on Arabidopsis gene expression in leaves. A, Heat map of global gene expression changes in response to the different bacterial strains indicated: Pst DC3000 (avrE⁺ and hopM1⁺), the avrE mutant (hopM1⁺), the hopM1 mutant (avrE⁺), and the avrE hopM1 double mutant. Bacteria (1 × 10⁸ cfu mL⁻¹) were vacuum infiltrated into Arabidopsis leaves, and leaf samples were collected 7 h after inoculation to allow sufficient time for type III effector delivery and, at the same time, to avoid bacterial population differences between the strains, especially between the avrE hopM1 double mutant and Pst DC3000 (Nomura et al., 2011). B, Venn diagram showing the numbers of genes up- or down-regulated by AvrE (comparing infection by the hopM1 mutant with infection by the avrE hopM1 mutant), HopM1 (comparing infection by the avrE mutant with infection by the avrE hopM1 mutant), or both.

Figure 6.

NHL13 is required for Arabidopsis immunity. A, The nhl13-1 mutant plant showed enhanced susceptibility to the avrE hopM1 mutant. Five-week-old Col-0 and nhl13-1 plants were dip inoculated with Pst DC3000, the avrE hopM1 mutant, or the hrcC mutant at 1 × 10⁸ cfu mL⁻¹. The bacterial population was counted 4 d after inoculation. Asterisks indicate significant differences in bacterial multiplication when compared with that in Col-0 plants: **, P < 0.01; and *, P < 0.05. B, Results of qRT-PCR to measure the levels of the NHL13 transcript in response to inoculation with Pst DC3000, the avrE mutant, the hopM1 mutant, the avrE hopM1 mutant, or water. The level of NHL13 was normalized to that of PROTEIN PHOSPHATASE 2A SUBUNIT A3 (PP2AA3; At1G13320).