The N-terminal region of Pseudomonas type III effector AvrPtoB elicits Pto-dependent immunity and has two distinct virulence determinants

Abstract

Resistance to bacterial speck disease in tomato is activated by the physical interaction of the host Pto kinase with either of the sequence-dissimilar type III effector proteins AvrPto or AvrPtoB (HopAB2) from Pseudomonas syringae pv. tomato. Pto-mediated immunity requires Prf, a protein with a nucleotide-binding site and leucine-rich repeats. The N-terminal 307 amino acids of AvrPtoB were previously reported to interact with the Pto kinase, and we show here that this region (AvrPtoB(1-307)) is sufficient for eliciting Pto/Prf-dependent immunity against P. s. pv. tomato. AvrPtoB(1-307) was also found to be sufficient for a virulence activity that enhances ethylene production and increases growth of P. s. pv. tomato and severity of speck disease on susceptible tomato lines lacking either Pto or Prf. Moreover, we found that residues 308-387 of AvrPtoB are required for the previously reported ability of AvrPtoB to suppress pathogen-associated molecular patterns-induced basal defenses in Arabidopsis. Thus, the N-terminal region of AvrPtoB has two structurally distinct domains involved in different virulence-promoting mechanisms. Random and targeted mutagenesis identified five tightly clustered residues in AvrPtoB(1-307) that are required for interaction with Pto and for elicitation of immunity to P. s. pv. tomato. Mutation of one of the five clustered residues abolished the ethylene-associated virulence activity of AvrPtoB(1-307). However, individual mutations of the other four residues, despite abolishing interaction with Pto and avirulence activity, had no effect on AvrPtoB(1-307) virulence activity. None of these mutations affected the basal defense-suppressing activity of AvrPtoB(1-387). Based on sequence alignments, estimates of helical propensity, and the previously reported structure of AvrPto, we hypothesize that the Pto-interacting domains of AvrPto and AvrPtoB(1-307) have structural similarity. Together, these data support a model in which AvrPtoB(1-307) promotes ethylene-associated virulence by interaction not with Pto but with another unknown host protein.

Figure 1. The N-terminal region of AvrPtoB has both avirulence and virulence activities in Pseudomonas syringae pv. tomato.

(a) Disease symptoms or immunity on leaves from RG-PtoR tomato plants after inoculation with DC3000ΔavrPtoΔavrPtoB carrying either empty vector, or plasmids expressing AvrPtoB, AvrPtoB_1-307 or AvrPtoB_1-387. Leaves from a similar position on each plant were detached, de-pigmented, and photographed 4 days after inoculation. (b) Bacterial populations of leaves as in (a). P. s. pv. tomato strains were vacuum-infiltrated into RG-PtoR tomato plants with an inoculum of 10⁴ cfu ml⁻¹. Error bars represent the SE of three replicates. Experiments were repeated three times with similar results. (c) Disease symptoms on RG-prf3 tomato plants 4 days after inoculation with the P. s. pv. tomato strains as in (a). The arrows indicate severe necrosis observed on lower leaves. (d) P. s. pv. tomato populations in RG-prf3 leaves at 0 and 2 days after inoculation using a starting inoculum of 10⁴ cfu ml⁻¹. Note that older leaves were sampled and that severe necrosis of these leaves prevented sampling later than day 2. Data from three independent experiments were used for statistical analysis using the SAS general linear model procedure. Values marked by the same letters were not significantly different from each other based on a least significant difference test (P = 0.05). Error bars indicate SE. (e) Disease symptoms on RG-pto11 tomato plants 4 days after inoculation with the P. s. pv. tomato strains as in (a). AvrPtoB_1-387 was not used in this experiment because it activates Rsb immunity in RG-pto11 (Abramovitch et al., 2003). The arrows indicate severe necrosis observed on lower leaves. (f) P. s. pv. tomato populations and statistical analysis in RG-pto11 leaves as in (d).

Figure 2. A small N-terminal region of AvrPtoB is sufficient for interaction with Pto kinase

(a) Yeast two-hybrid interaction assays of AvrPtoB truncation mutants (N-terminally tagged with an hemagglutinin (HA) epitope) and Pto (upper panel). Pto was expressed as the bait and AvrPtoB truncation mutants as prey. Protein accumulation of the AvrPtoB proteins in yeast was detected by Western blotting using anti-HA antibody (lower panel). The origin of the higher molecular mass band cross-reacting with HA in the 1-190 lane is unknown. Positions of protein molecular mass markers are shown. (b) Cell death assays of AvrPtoB truncation mutants using Agrobacterium-mediated transient expression. Agrobacterium containing individual truncation mutants were syringe-infiltrated into RG-PtoR tomato leaves at a concentration (OD₆₀₀) of 0.03. Photographs were taken 3 days after infiltration. (c) Bacterial populations in RG-PtoR leaves 4 days after inoculation with DC3000ΔavrPtoΔavrPtoB strains harboring the empty vector or plasmids expressing AvrPtoB_1-307 or AvrPtoB_1-200 at inoculum of 10⁴ cfu ml⁻¹. Error bars represent the SE of three replicates. Experiments were repeated three times with similar results. (d) De-pigmented leaves from a similar position on each plant as used for the bacterial population measurements in (c). (e) Expression and secretion of AvrPtoB_1-307 and AvrPtoB_1-200 from DC3000ΔavrPtoΔavrPtoB grown in minimal medium under hrp induction conditions (Lin et al., 2006). Western blotting was performed using anti-AvrPtoB or anti-NptII antibody. Cytoplasmically localized NptII was used as a control for cell lysis. No NptII protein was detected in the culture medium using an anti-NptII antibody.

Figure 3. Phenylalanine 173 is required for interaction of AvrPtoB_1-307 with Pto

(a) Alignment of the amino acids in the AvrPtoB_121-200 regions from AvrPtoB and three homologs. Identical amino acids are shaded in black. Positions of the AvrPtoB_1-307 amino acids that were later found to be required for interaction with Pto (see Figure 5) are indicated by asterisks. (b) Yeast two-hybrid interaction between Pto and the 121–200 amino acid region from AvrPtoB_T1 and HopPmaL. Pto interacts with the wild-type 121–200 region from AvrPtoB_T1 and HopPmaL, but not F173A mutants of this region. Pto was expressed as the bait protein and the AvrPtoB_121-200 fragments were expressed as prey. Expression of the HA-tagged AvrPtoB proteins was confirmed by Western blotting using anti-HA antibody (lower panel). (c) Yeast two-hybrid interaction between Pto and AvrPtoB_1-307 mutants F173A, H172A or P174A. Pto was expressed as the bait. The three mutants were expressed as prey. Western blotting using anti-HA antibody indicates that the HA-tagged AvrPtoB_1-307 and three mutants were expressed in yeast (lower panel). (d) Disease symptoms or plant immunity of RG-PtoR tomato plants 4 days after inoculation with DC3000ΔavrPtoΔavrPtoB harboring the empty vector or plasmids expressing wild-type AvrPtoB_1-307 or the F173A mutant. (e) Bacterial populations in leaves of plants as in (d). Each P. s. pv. tomato strain was vacuum-infiltrated into RG-PtoR tomato plants with an inoculum of 10⁴ cfu ml⁻¹. Error bars represent the SE of three replicates. Experiments were repeated three times with similar results. (f) Expression and secretion of AvrPtoB_1-307 and the AvrPtoB_1-307(F173A) mutant from DC3000ΔavrPtoΔavrPtoB grown in minimal medium under hrp induction conditions. The P. s. pv. tomato secretion assay was carried out as described in Figure 2(e).

Figure 4. Four AvrPtoB_1-307 proteins with single amino acid substitutions are compromised in avirulence activity

(a) HR assay of wild-type AvrPtoB_1-307 or E165 K, F169S, G180 V, L195H mutants using Agrobacterium-mediated transient expression in RG-PtoR tomato leaves. Photographs were taken 3 days after Agrobacterium infiltration. (b) Transient expression of wild-type AvrPtoB_1-307, E165 K, F169S, G180 V or L195H mutants in RG-PtoR protoplasts. Plasmids expressing wild-type AvrPtoB_1-307 or the mutants were transformed into RG-PtoR protoplasts and cell death was monitored by Evans blue staining 24 h after transformation (upper panel). Approximately 200 randomly selected protoplast cells were counted for determination of the dead cell fraction. Protein expression was analyzed by Western blotting using an anti-AvrPtoB antibody (lower panel). (c) P. s. pv. tomato assay of AvrPtoB_1-307, E165 K, F169S, G180 V or L195H mutants. Disease symptoms or plant immunity of RG-PtoR tomato leaves 4 days after inoculation with DC3000ΔavrPtoΔavrPtoB harboring the empty vector or plasmids expressing wild-type AvrPtoB_1-307, E165 K, F169S, G180 V or L195H (upper panel). Bacterial populations in leaves of plants inoculated with the indicated strains (lower panel). Each P. s. pv. tomato strain was vacuum-infiltrated into three RG-PtoR tomato plants with an inoculum of 10⁴ cfu ml⁻¹. Error bars represent the SE of three replicates. Experiments were repeated three times with similar results. (d) Yeast two-hybrid interaction between Pto and AvrPtoB_1-307 or the AvrPtoB_1-307 mutants. Wild-type AvrPtoB_1-307 or the four mutants were expressed as prey and Pto was expressed as the bait. Protein expression of the HA-tagged AvrPtoB proteins was confirmed by Western blot using an anti-HA antibody. (e) Expression and secretion of AvrPtoB_1-307 mutants from DC3000ΔavrPtoΔavrPtoB grown in minimal medium under hrp induction conditions. The P. s. pv. tomato secretion assay was carried out as described in Figure 2(e).

Figure 5. Virulence activities of AvrPtoB_1-307 mutants

(a) Disease symptoms of tomato RG-prf3 plants inoculated with 10⁴ cfu ml⁻¹ of DC3000ΔavrPtoΔavrPtoB carrying the empty vector or plasmids expressing AvrPtoB_1-307, F173A, E165 K, F169S, G180 V or L195H. Representative de-pigmented leaves from each plant are shown in the bottom left corners. Plants were photographed 4 days after inoculation. (b) Measurement of bacterial populations in leaves of the plants in (a) at 0 and 2 days after inoculation using a P. s. pv. tomato starting inoculum of 10⁴ cfu ml⁻¹. Data pooled from three independent experiments (n = 9) were used for statistical analysis using the SAS general linear model procedure (see Experimental procedures). Values marked with the same letters were not significantly different from each other based on the least squares difference test (P = 0.05). Error bars indicate SE.

Figure 6. AvrPtoB_1-307 is sufficient to induce ethylene biosynthesis genes and promotes ethylene production during disease development in tomato plants

(a) Tomato ACC oxidase 1/2 (SlAco1/2) gene expression was induced by AvrPtoB, AvrPtoB_1-387 and AvrPtoB_1-307, but not by the virulence-deficient AvrPtoB_1-307(F173A) mutant. RG-prf3 plants were inoculated with 10⁴ cfu ml⁻¹ of DC3000ΔavrPtoΔavrPtoB carrying the empty vector, AvrPtoB, AvrPtoB_1-387, AvrPtoB_1-307 or AvrPtoB_1-307(F173A). Total RNA was isolated at the indicated time points after infiltration, and 1 μg total RNA was used to generate first-strand cDNA. The induction of SlAco1/2 was determined by 30 PCR cycles using SlAco1/2 gene-specific primers, with EF1-α gene-specific primers as a control (Cohn and Martin, 2005; Lin and Martin, 2005). The data shown are representative of two experiments with independent biological replicates that gave similar results. Lane M, 1 kb DNA ladder. (b) Ethylene production by susceptible tomato RG-prf3 plants infiltrated with 10⁴ cfu ml⁻¹ of DC3000ΔavrPtoΔavrPtoB carrying the empty vector or plasmids expressing full-length AvrPtoB, AvrPtoB_1-387, AvrPtoB_1-307 or AvrPtoB_1-307(F173A). Ethylene was measured by gas chromatography at the time points indicated (Cohn and Martin, 2005; Lin and Martin, 2005). Error bars represent the SE of six replicates. The experiment was repeated twice with similar results.

Figure 7. AvrPtoB_1-387 suppresses markers of pathogen-associated molecular patterns-induced basal defenses in Arabidopsis

(a) AvrPtoB_1-387, but not AvrPtoB_1-307, suppresses flg22-induced FRK1 promoter activity. Arabidopsis Col-0 protoplasts were co-transformed with AvrPtoB or derivatives and a FRK1-LUC reporter gene, incubated for 6 h, and then treated with 100 nM flg22 for 3 h. Luciferase activity was measured and is shown as relative promoter activity compared to a UBQ10-GUS control (He et al., 2006). Protein expression of AvrPtoB and derivatives was analyzed by Western blotting using an anti-HA antibody (right panel). (b) AvrPtoB and AvrPtoB_1-387, but not AvrPtoB_1-307, blocked flg22-induced MAP kinase activation. HA-tagged MPK3/6 and AvrPtoB or derivatives were co-expressed in Arabidopsis protoplasts for 6 h, and then 1 μM flg22 was added for 10 min. MAP kinases and effectors were immunoprecipitated with anti-HA antibody, followed by in vitro kinase assay (upper panel). Expression of MAP kinases and effectors were confirmed by Western blotting using an anti-HA antibody (bottom panel) (He et al., 2006).

Figure 8. Amino acids 308–387 encode a determinant for suppression of pathogen-associated molecular patterns-induced basal defense in Arabidopsis

(a) AvrPtoB_1-307 was unable to suppress even weak activation of MPK6 induced by one-tenth the amount (100 nM) of flg22 used in Figure 7, further supporting a requirement for amino acids 308–387. Note that flg22 at a concentration of 10 or 1 nM did not activate MPK6. (b) Five mutations (E165 K, F169S, G180 V, L195H and F173A) that compromise the avirulence or ethylene-dependent virulence activities of AvrPtoB do not affect the suppression of flg22-induced MPK6 activation by AvrPtoB_1-387.

Figure 9. Sequence alignment and structural model for AvrPtoB_121-200

(a) Sequence alignment of AvrPto_36-128, AvrPtoB_121-200 and the comparable region of AvrPtoB homologs from P. s. pv. tomato T1 and P. s. pv. maculicola ES4326. Identical residues (red) and similar residues (black) conserved across all four sequences are indicated. I96 in AvrPto and F173 in AvrPtoB are highlighted in yellow. (b) Secondary structure prediction for AvrPto and AvrPtoB sequences. Categorization into α-helix (H), β-sheet (E), turn (T) or other (x) is based on the relative probabilities of six backbone conformational states as determined by HMMSTR-R (Bystroff et al., 2000). AvrPto residues that actually adopt α-helical backbone conformation in the AvrPto NMR structure (Wulf et al., 2004) are underlined. (c) AvrPtoB homology model and conserved features of CD loop that may facilitate Pto binding. Ribbon overlay (center) of AvrPto (green) and AvrPtoB (blue), with essential residues (red) and conserved M176 (gray) shown in stick representation. The cluster of hydrophobic residues in the modeled AvrPtoB CD loop includes essential residues F169 and F173 and conserved residue M176 (right; AvrPtoB residues shown in blue). These residues overlie residues M85, I96 and M100, respectively, in the known AvrPto structure (AvrPto residues shown in green). The electrostatic surface potential of the homology model suggests an electrostatic role for essential residue E165 (left). Images were generated using DEEPVIEW (http://www.expasy.org/spdbv) (Guex and Peitsch, 1997).