Skip to main page content
U.S. flag
See More

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Jun 21;113(25):E3577-86.
doi: 10.1073/pnas.1606322113. Epub 2016 Jun 6.

Pseudomonas syringae type III effector HopAF1 suppresses plant immunity by targeting methionine recycling to block ethylene induction

Erica J Washington  1 , M Shahid Mukhtar  1 , Omri M Finkel  1 , Li Wan  1 , Mark J Banfield  2 , Joseph J Kieber  1 , Jeffery L Dangl  3
Affiliations

Pseudomonas syringae type III effector HopAF1 suppresses plant immunity by targeting methionine recycling to block ethylene induction

Erica J Washington et al. Proc Natl Acad Sci U S A. .

Abstract

HopAF1 is a type III effector protein of unknown function encoded in the genomes of several strains of Pseudomonas syringae and other plant pathogens. Structural modeling predicted that HopAF1 is closely related to deamidase proteins. Deamidation is the irreversible substitution of an amide group with a carboxylate group. Several bacterial virulence factors are deamidases that manipulate the activity of specific host protein substrates. We identified Arabidopsis methylthioadenosine nucleosidase proteins MTN1 and MTN2 as putative targets of HopAF1 deamidation. MTNs are enzymes in the Yang cycle, which is essential for the high levels of ethylene biosynthesis in Arabidopsis We hypothesized that HopAF1 inhibits the host defense response by manipulating MTN activity and consequently ethylene levels. We determined that bacterially delivered HopAF1 inhibits ethylene biosynthesis induced by pathogen-associated molecular patterns and that Arabidopsis mtn1 mtn2 mutant plants phenocopy the effect of HopAF1. Furthermore, we identified two conserved asparagines in MTN1 and MTN2 from Arabidopsis that confer loss of function phenotypes when deamidated via site-specific mutation. These residues are potential targets of HopAF1 deamidation. HopAF1-mediated manipulation of Yang cycle MTN proteins is likely an evolutionarily conserved mechanism whereby HopAF1 orthologs from multiple plant pathogens contribute to disease in a large variety of plant hosts.

Keywords: Pseudomonas syringae; Yang cycle; ethylene; plant immune system; type III effectors.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Fig. S1.
Fig. S1.
The HopAF1 evolutionary tree. (A) Phylogenetic reconstruction of HopAF1 homologous protein sequences in sequenced genomes. Monophyletic groups belonging to the same genus are collapsed. Sequences cluster into three distinct groups: HopAF1, AvrXv3, and Xanthomonas type 2. In cases where two copies are present in the same genome, the text LOCUS 1/2 is appended to the species name. Nodes with bootstrap support >0.2 are marked. The size of the dot is proportional to the level of support. The uncollapsed tree can be viewed on iTOL (itol.embl.de/shared/HopAF1Evo). (B) Occurrence of HopAF1/AvrXv3 across the Xanthomonas phylogeny. Phylogenetic reconstruction of HopAF1-containing Xanthomonas, using a concatenated alignment of 23 single-copy marker genes. The occurrence of the HopAF1 variant and AvrXv3 is marked on the right. Marker genes used for tree construction are the same used for the P. syringae tree in C. (C) The occurrence of HopAF1/AvrXv3 across the P. syringae phylogeny. Phylogenetic reconstruction of P. syringae, using a concatenated alignment of 23 single copy marker genes. The occurrence of the HopAF1 variant and AvrXv3 is marked on the right. Marker genes used for tree construction were COG0012, COG0016, COG0048, COG0049, COG0052, COG0080, COG0081, COG0087, COG0088, COG0090, COG0091, COG0092, COG0093, COG0094, COG0096, COG0098, COG0102, COG0103, COG0185, COG0186, COG0201, COG0256, and COG0541. All three trees are available for browsing at itol.embl.de/shared/HopAF1Evo.
Fig. 1.
Fig. 1.
HopAF1 putative catalytic residues in HopAF1 transgenic lines are required for an increase in susceptibility to a disarmed pathogen and inhibition of flg22-induced defense. (A) Homology model of the HopAF1 reference allele (blue) from Pto DC3000 residues 130–284 aligned with CheD (PDB ID code 2F9Z) (silver). The side chains of the putative catalytic residues of HopAF1 D154, C159, and H186 are shown in green. The side chains of the CheD catalytic residues C27, H44, and T21 are shown in silver. The N and C termini of the proteins are labeled, as are the key structural elements (α helix 1, α helix 2, β sheet 3, and β sheet 5). PyMOL was used to generate the figure. (B) Bacterial growth of Pto DC3000D28E in Col-0 plants and two independent transgenic lines expressing either estradiol-inducible HopAF1-cerulean-HA or estradiol-inducible HopAF1C159AH186A-cerulean-HA was measured 0 and 3 d after hand-inoculation with bacteria (1 × 105 cfu/mL). Transgene expression was induced with 20 μM estradiol 24 h before bacterial inoculation. Error bars represent SEM. An ANOVA was performed among the day 3 samples, followed by Tukey’s post hoc analysis (P < 0.05). The different letters indicate groups that differ significantly. (C) Col-0, estradiol-inducible HopAF1-cerulean-HA, and estradiol-inducible HopAF1C159AH186A-cerulean-HA plants were sprayed with 20 μM estradiol for 12 h and then were infiltrated with 1 μM flg22 or water for 24 h before infiltration with 1 × 105 cfu/mL Pto DC3000. Bacterial growth was determined 0 and 3 d after bacterial inoculation. Error bars represent SEM. An ANOVA was performed among the day 3 samples, followed by Tukey’s post hoc analysis (P < 0.05). The different letters indicate groups that differ significantly.
Fig. S2.
Fig. S2.
The HopAF1 protein family. (A) Homology model of AvrXv3 from Xanthomonas vesicatoria (purple) aligned with CheD (silver). Putative catalytic residues of AvrXv3, C105, and H126, are highlighted in yellow and align with the catalytic residues of CheD (silver line above sequences). (B) Independent homology models for additional HopAF1 family members: P. syringae pv. maculicola ES4326 (HopAF1 type 1; pink), Xanthomonas oryzae pv. oryzicola MAI10 (Xanthomonas type 2; dark teal), and P. syringae pv. lachrymans pla107 (HopAF1 type 2; orange). (C) Representative protein sequences from all major groups of HopAF1 were aligned using VectorNTI AlignX. Amino acids with 80% identity are shaded in gray. Invariant putative catalytic residues are highlighted in dark green, and the conserved aspartic acid is highlighted in light green. Sites of likely posttranslocation myristoylation and palmitoylation are denoted in red and orange, respectively. In the residues underlined in blue represent the putative catalytic domain. Structural images were generated with PyMOL.
Fig. S3.
Fig. S3.
Protein expression and disease symptoms in HopAF1 transgenic lines. (A) Bacterial growth of Pto DC3000 and Pto DC3000ΔhopAF1 (46) in Col-0 was measured 0 and 3 d after hand-infiltration with bacteria (1 × 105 cfu/mL). Error bars represent SEM. An ANOVA was performed among the day 3 samples, followed by Tukey’s post hoc analysis (P < 0.05) denoted by A. (B) Immunoblot analysis with anti-HA shows a time course of the expression of HopAF1-cerulean-HA and HopAF1C159AH186A-cerulean-HA in transgenic plants after spray application of 20 μM estradiol. (C) Immunoblot analysis with anti-HA shows the expression of HopAF1-cerulean-HA and HopAF1C159AH186A-cerulean-HA in transgenic plants used in the experiment in Fig. 1B. Expression was determined 3 d after spray application of 20 μM estradiol. (D) Disease symptoms in Col-0 and two independent lines of HopAF1-cerulean-HA and HopAF1C159AH186A-cerulean-HA were photographed 3 d after hand-infiltration with 1 × 105 cfu/mL of P. syringae pv. tomato DC3000D28E. HopAF1 was induced with the application of 20 μM estradiol 24 h before bacterial inoculation. Numbers below the images denote the ratio of the number of leaves that displayed chlorotic symptoms to the number of leaves inoculated. (E) Immunoblot analysis with anti-HA shows the expression of HopAF1-cerulean-HA and HopAF1C159AH186A-cerulean-HA in the transgenic plants used in the experiment in Fig. 1C. Expression was determined 3 d after bacterial infiltration, which corresponds to 108 h (4.5 d) after estradiol induction.
Fig. S4.
Fig. S4.
AvrXv3 putative catalytic residues are required for the HR in tomato. (A) AvrXv3-4x myc and AvrXv3H126A-4x myc were transiently expressed in tomato cultivar Fla 216 using the Agrobacterium strain C58C1. Cores containing both infiltrated and uninfiltrated tissue were removed from the tomato leaves 3 d postinfiltration, and trypan blue staining was used to highlight the cell-death response associated with the HR. Numbers below the images denote the ratio of the number of leaves that displayed the HR to the number of leaves inoculated. (B) Anti-myc immunoblot analysis demonstrates the expression of AvrXv3-4x-myc and AvrXv3H126A-4x-myc in tomato cultivar Fla 216 3 d postinfiltration. Ponceau staining of the membrane indicates equal loading.
Fig. 2.
Fig. 2.
HopAF1 interacts with Yang cycle proteins MTN1 and MTN2 at the plasma membrane. (A) Estradiol-inducible HopAF1-cerulean-HA, HopAF1G2AC4S-cerulean-HA, HopAF1H186A-cerulean-HA, AvrXv3-cerulean-HA, and AvrXv3H126A-cerulean HA were expressed transiently in N. benthamiana using Agrobacterium-mediated transient transformations. Transgene expression was induced with 5 μM estradiol 3 d after bacterial infiltration. The localization of transiently expressed proteins was determined with scanning confocal laser microscopy 3 h after estradiol induction in N. benthamiana epidermal cells. PLC2-YFP is a marker for the plasma membrane localization. White arrows indicate the plasma membrane (PM), cytoplasmic streaming (CS), and nuclei (N). The zoomed images display only the cerulean channel. (Scale bars, 20 μm.) (B) Protein extracts from N. benthamiana leaves transiently expressing HopAF1-4x-myc, HopAF1H186A-4x-myc, MTN1-HA, and MTN2-HA were collected 3 d after infiltration and were subjected to immunoprecipitation using anti-myc–coupled magnetic beads; 400 μg of input and a 50× concentrated elution fraction were analyzed by anti-HA and anti-myc immunoblots. (C) HopAF1-nYFP, HopAF1H186A-nYFP, and HopAF1G2AC4S-nYFP were transiently coexpressed with MTN1-cYFP, MTN2-cYFP, and PLC2-cYFP in N. benthamiana leaves using Agrobacterium-mediated transient transformation. Confocal microscopy was used to image the reconstituted YFP signal 3 d after infiltration. (Scale bars, 50 μm.)
Fig. S5.
Fig. S5.
HopAF1 is targeted to the plasma membrane via acylation. (A) The localization of transiently expressed HopAF1-cerulean-HA and HopAF1G2AC4S-cerulean-HA was determined with scanning confocal laser microscopy 3 h after estradiol induction in N. benthamiana epidermal cells. Free YFP is a soluble protein and therefore can be used as a marker of the plant cytoplasm. White arrowheads indicate the plasma membrane (PM), cytoplasmic streaming (CS) and nuclei (N). (Scale bars, 20 μM.) (B) As in Fig. 2, estradiol-inducible HopAF1-cerulean-HA, HopAF1G2AC4S-cerulean-HA, HopAF1H186A-cerulean-HA, AvrXv3-cerulean-HA, and AvrXv3H126A-cerulean HA were expressed transiently in N. benthamiana using Agrobacterium-mediated transient transformations. Transgene expression was induced with 5 μM estradiol 3 d after bacterial infiltration. The localization of transiently expressed proteins was determined with scanning confocal laser microscopy 3 h after estradiol induction in N. benthamiana epidermal cells. Cellular fractionations into total (T), soluble (S), and microsomal (M) fractions were performed. Equal cell equivalents were loaded onto SDS/PAGE gels followed by anti-HA immunoblots to detect the fractions with the expressed proteins of interest. Anti-APX and anti–H+-ATPase were included as controls for the soluble (S) and microsomal (M) fractions, respectively. (C) Levels of HopAF1-cerulean-HA transiently expressed in N. benthamiana were titrated by varying the concentration of estradiol and the OD of Agrobacterium. Immunoblots with anti-HA were used to detect the expression of HopAF1-cerulean-HA. N. benthamiana leaves were sprayed with estradiol 3 d postinfiltration with Agrobacterium, and leaf samples were collected 3 h after estradiol application for immunoblot analysis. (D) Immunoblot analysis with anti-HA shows the expression of full-length HopAF1-cerulean-HA, HopAF1H186A-cerulean-HA, and HopAF1G2AC4S-cerulean-HA and the absence of free cerulean (black arrow) at ∼27 kDa. As above, N. benthamiana leaves were sprayed with 5 μM estradiol 3 d postinfiltration with Agrobacterium, and protein samples were collected 3 h after estradiol application. Ponceau staining indicates equal loading. The absence of free cerulean under these conditions indicates that the mislocalized HopAF1 detected in the nucleocytoplasm is not free cerulean diffusing into the nucleus.
Fig. S6.
Fig. S6.
The Yang cycle proteins MTN1 and MTN2 are found in both the plasma membrane and the cytoplasm. (A) In this figure adapted from Rzewuski et al. (94), the relationship between the Yang cycle (also known as the methionine recycling cycle) and the phytohormone ethylene in Arabidopsis is highlighted. Arabidopsis MTN1 and MTN2 are Yang cycle proteins that can cleave their substrate MTA, resulting in adenine and 5-methylthioribose (MTR) production. MTR then is phosphorylated by the kinase MTK. MTNs also function in a secondary pathway to recycle methionine. Both MTN1 and MTN2 can bind S-adenosylhomocysteine (SAH), but, perhaps as a result of steric hindrance and the rigidity of the ligand-binding pocket of MTN1 (95), only MTN2 has been shown to cleave SAH (71, 95). However, both MTN1 and MTN2 cleave MTA more efficiently than SAH (71). Methionine is a precursor for the ethylene biosynthesis pathway. It is important to note that methionine recycling supplements the continual de novo synthesis of methionine. (B) 35S:MTN1-HA and 35S:MTN2-HA were expressed transiently in N. benthamiana using Agrobacterium-mediated transient transformations. Cellular fractionations into total (T), soluble (S), and microsomal (M) fractions were performed. Equal cell equivalents were loaded onto SDS/PAGE gels followed by anti-HA immunoblots to detect the fractions with the expressed proteins of interest. Anti-APX and anti–H+-ATPase were included as controls for the soluble (S) and microsomal (M) fractions, respectively. (C) Estradiol-inducible MTN1-cerulean-HA and MTN2-cerulean-HA were transiently coexpressed with PLC2-CFP, a plasma membrane marker, in N. benthamiana and were imaged using confocal microscopy. The path of the red arrow in the merged images is reflected in the histograms, which display the intensity of citrine and the CFP signal by yellow and blue lines, respectively.
Fig. S7.
Fig. S7.
Establishing P. fluorescens as a tool to determine the effect of HopAF1 on PAMP-induced ethylene accumulation. (A) Arabidopsis leaves were hand-infiltrated with multiple concentrations of Pto DC3000 and Pto DC3000 hrcC (a type III secretion system-deficient mutant). The fresh weight of four leaves was measured 3 h postinfiltration; then these leaves were sealed in a vial for 24 h before ethylene accumulation was measured using gas chromatography. An ANOVA was performed, followed by Tukey’s post hoc analysis (P < 0.05); error bars represent SE. (B) The histogram demonstrates the increase in ethylene biosynthesis elicited by increasing concentrations of Pfo-1. (C) Immunoblot analysis of HA-tagged HopAF1 variants expressed in Pfo-1 strains induced with minimal medium for 12 h. Protein was detected with anti-HA monoclonal antibody. (D) AvrRpt2 translocation assay to monitor HopAF1 translocation into plant cells. Col-0 rpm1 RPS2 leaves were inoculated with Pfo-1 expressing HopAF1 variants fused to the ∆79AvrRpt2 reporter (HopAF1-∆79AvrRpt2, HopAF1H186A-∆79AvrRpt2, and HopAF1G2AC4S-∆79AvrRpt2) and were observed after 24 h. The ability to elicit an AvrRpt2-dependent HR in the RPS2 background, as indicated by tissue collapse, is evidence of the translocation of the fusion proteins. The ratio of the number of leaves in which the HR was evident to the number of leaves inoculated is displayed below the representative leaves for each Pfo-1 inoculation. There was no HR in the rpm1 rps2 background.
Fig. 3.
Fig. 3.
HopAF1 inhibits ethylene biosynthesis and phenocopies mtn1 mtn2 for a PTI-induced increase in ethylene biosynthesis. (A) Arabidopsis leaves were hand-infiltrated with MgCl2 (vehicle control), Pfo-1, Pfo-1 carrying the empty vector (EV), or Pfo-1 expressing HopAF1. (B) In addition to the aforementioned treatments, Arabidopsis leaves were hand-infiltrated with Pfo-1 expressing HopAF1H186A and Pfo-1 expressing HopAF1G2AC4S. (C) Leaves of Col-0 plants and Arabidopsis mutants mtn1-1, mtn1-2, mtn2-1, and mtn2-2 were hand-infiltrated with Pfo-1 carrying the empty vector (EV) or with Pfo-1 expressing wild-type, native promoter HopAF1. (D) Leaves of Col-0 and Arabidopsis mutants mtn1-2 mtn2-2 and mtk were hand-infiltrated with MgCl2, Pfo-1 carrying the empty vector (EV), or Pfo-1 expressing the wild-type native promoter HopAF1. In all experiments, the fresh weight of four leaves was measured 3 h post infiltration, and these leaves were sealed in a vial for 24 h before ethylene accumulation was measured. Letters represent treatments with significant difference according to the post hoc ANOVA Tukey’s test (P < 0.05).
Fig. S8.
Fig. S8.
HopAF1 inhibition of MTN1 activity in vitro is dependent on catalytic activity. (A and B) Circular dichroism spectra collected at 25 °C at the far-UV CD spectrum (185–260 nm) demonstrates that wild-type MTN1, 6xHis-HopAF1, and 6xHis-HopAF1H186A have a secondary structure. The H186A mutation in HopAF1 does not perturb the secondary structure. The buffer control [20 mM sodium phosphate (pH 7.4), 150 nM NaF] was set as the baseline and was subtracted from the protein-associated spectra above. (C) Michaelis–Menten plot of MTN1 activity measured in vitro using a xanthine oxidase-coupled spectrometric assay. Recombinant MTN1 was incubated for 18 h in buffer at 37 °C alone or with equimolar amounts of recombinant wild-type 6xHis-HopAF1 or GST. (D) MTN1 enzyme-specific activity measured in vitro using a xanthine oxidase-coupled spectrometric assay. MTN1 was incubated at 37 °C for 18 h with equimolar amounts of 6xHis-HopAF1, 6xHis-HopAF1 H186A, Δ147-HopAF1, Δ130-HopAF1, and GST. An ANOVA was performed, followed by Tukey’s post hoc analysis (P < 0.05). Error bars represent SE. Different letters indicate groups that differ significantly.
Fig. 4.
Fig. 4.
Arabidopsis MTN1 and MTN2 play a role in ethylene-associated innate immunity pathways. Bacterial growth of Pto DC3000D28E in Col-0 plants and mtn1 mtn2, mtk, and ein2 mutants was measured 0 and 3 d after hand-inoculation with bacteria (1 × 105 cfu/mL). Letters represent treatments with significant differences according to the post hoc ANOVA Tukey’s test (P < 0.05).
Fig. 5.
Fig. 5.
Deamidation-mimicking variants of MTN1 and MTN2 lose the ability to cleave MTA in vitro. (A) Ribbon representation of the MTN1 crystal structure (PDB ID code 2H8G). Monomers are represented in sand and silver. The ligand analog (MTT) is shown in green. The ligand-binding and catalytic residues from MTN1 are represented as blue sticks. The putative conserved targets of deamidation are labeled and are shown as red sticks. Images were generated using PyMOL. (B) MTN1 enzyme-specific activity measured in vitro using a xanthine oxidase-coupled spectrometric assay. The MTN1 catalytic dead mutant is MTN1D225N. (C) MTN2 enzyme-specific activity measured in vitro using a xanthine-oxidase coupled spectrometric assay. The MTN2 catalytic dead mutant is MTN2D212N. Letters represent treatments with significant difference according to the post hoc ANOVA Tukey’s test (P < 0.05). (D) Arabidopsis complementation lines in the mtn1 mtn2 background that express UBQ:MTN variants at the T1 stage. UBQ:MTN1N194D, UBQ:MTN1N113D, and MTN1D225N complementation lines in the mtn1 mtn2 background were not able complement the mtn1 mtn2 sterile and stunted growth phenotype, suggesting that these MTN1 deamidation-mimic variants were nonfunctional.
Fig. S9.
Fig. S9.
In vitro and in planta consequences of mutating conserved MTN asparagines. (A) Alignment of methylthioadenosine nucleosidase amino acid sequences from plants and E. coli. The conserved putative targets of deamidation, N113, N194, and N169, are highlighted in red. (B) UBQ:MTN1N113V complementation lines in the mtn1 mtn2 mutant background show a range of phenotypes in the T1 stage.
Fig. S10.
Fig. S10.
Working model of P. syringae HopAF1 function. Plant pathogens such as P. syringae and X. campestris deliver multiple T3E proteins into plant cells, including HopAF1 and AvrXv3, respectively. Recognition of PAMPs, such as flg22, a peptide derived from bacterial flagellin, by PRRs such as FLS2 triggers a signal cascade that results in multiple outputs, collectively termed PTI. Through perception of PAMPs, such as flg22, and activation of a MAPK cascade, PTI leads to an increase in ethylene biosynthesis. Our data suggest that HopAF1 inhibits PTI. Our data are consistent with the model that HopAF1 is targeted to the plant plasma membrane via acylation, where it inhibits the function of the Yang cycle proteins MTN1 and MTN2, which are required for the induced level of ethylene that occurs during PTI. As a result, the induced amount of ethylene that occurs during PTI may be an important output, making the Yang cycle an important component of the plant defense pathway and HopAF1 a critical T3E that prevents high ethylene accumulation during the plant defense response by targeting the Yang cycle. Furthermore, we have identified two asparagines in Arabidopsis MTN1 and MTN2 that are potential targets of HopAF1-mediated deamidation. Further research is required to determine whether HopAF1 deamidates or modifies Arabidopsis MTNs or itself.
See this image and copyright information in PMC

References

    1. Grant SR, Fisher EJ, Chang JH, Mole BM, Dangl JL. Subterfuge and manipulation: Type III effector proteins of phytopathogenic bacteria. Annu Rev Microbiol. 2006;60:425–449. - PubMed
    1. Lindgren PB, Peet RC, Panopoulos NJ. Gene cluster of Pseudomonas syringae pv. “phaseolicola” controls pathogenicity of bean plants and hypersensitivity of nonhost plants. J Bacteriol. 1986;168(2):512–522. - PMC - PubMed
    1. Ausubel FM. Are innate immune signaling pathways in plants and animals conserved? Nat Immunol. 2005;6(10):973–979. - PubMed
    1. Boller T, Felix G. A renaissance of elicitors: Perception of microbe-associated molecular patterns and danger signals by pattern-recognition receptors. Annu Rev Plant Biol. 2009;60:379–406. - PubMed
    1. Felix G, Duran JD, Volko S, Boller T. Plants have a sensitive perception system for the most conserved domain of bacterial flagellin. Plant J. 1999;18(3):265–276. - PubMed

Publication types

LinkOut - more resources