Two serine residues in Pseudomonas syringae effector HopZ1a are required for acetyltransferase activity and association with the host co-factor

Abstract

Gram-negative bacteria inject type III secreted effectors (T3SEs) into host cells to manipulate the immune response. The YopJ family effector HopZ1a produced by the plant pathogen Pseudomonas syringae possesses acetyltransferase activity and acetylates plant proteins to facilitate infection. Using mass spectrometry, we identified a threonine residue, T346, as the main autoacetylation site of HopZ1a. Two neighboring serine residues, S349 and S351, are required for the acetyltransferase activity of HopZ1a in vitro and are indispensable for the virulence function of HopZ1a in Arabidopsis thaliana. Using proton nuclear magnetic resonance (NMR), we observed a conformational change of HopZ1a in the presence of inositol hexakisphosphate (IP6), which acts as a eukaryotic co-factor and significantly enhances the acetyltransferase activity of several YopJ family effectors. S349 and S351 are required for IP6-binding-mediated conformational change of HopZ1a. S349 and S351 are located in a conserved region in the C-terminal domain of YopJ family effectors. Mutations of the corresponding serine(s) in two other effectors, HopZ3 of P. syringae and PopP2 of Ralstonia solanacerum, also abolished their acetyltransferase activity. These results suggest that, in addition to the highly conserved catalytic residues, YopJ family effectors also require conserved serine(s) in the C-terminal domain for their enzymatic activity.

Keywords: Arabidopsis thaliana; Pseudomonas syringae; YopJ family type III effectors; acetyltransferase; bacterial virulence; inositol hexakisphosphate (IP6); stomatal aperture.

Fig. 1

Identification of the autoacetylation site(s) in HopZ1a. (a) K289 is not required for the autoacetylation of HopZ1a. Tag-free HopZ1a, the catalytic mutant HopZ1a(C216A), and HopZ1a(K289R) were purified from Escherichia coli and subjected to in vitro acetylation assays using C14-acetyl CoA. The reaction was supplemented with inositol hexakisphosphate (IP6) as a co-factor. The acetylated proteins were detected by autoradiography after exposure at −80°C for 5 d. Numbers underneath the autoradiograph indicate relative acetylation levels of HopZ1a mutants compared with wild-type protein (100%). Equal loading of the proteins was confirmed by Coomassie Blue staining. (b) T346 is the prominent acetylation site in HopZ1a. Tag-free HopZ1a and HopZ1a(C216A) were incubated with IP6 and acetyl CoA before being subjected to MS/MS spectrometric analysis. The y ions are marked in the spectrum and illustrated along the peptide sequence (parent ion: m/z 847.4025, 3+). T346 acetylation was not detected in the catalytic mutant HopZ1a (C216A). (c) Substitutions of S349 and S351 greatly reduced the autoacetylation level of HopZ1a. T346 and its neighboring serine residues S344, S349 and S351 were systematically mutated and tested for autoacetylation activity using the in vitro acetylation assay with the reactions supplemented with IP6. These experiments were repeated three times with similar results.

Fig. 2

S349 and S351 are required for the acetyltransferase activity of HopZ1a. Tag-free proteins of wild-type and mutant HopZ1a were purified from Escherichia coli and examined for acetyltransferase activity using in vitro acetylation assays. MBP-AtJAZ6 was also purified from E. coli and used as a substrate. The reactions were supplemented with IP6. Numbers underneath the autoradiograph indicate relative levels of acetylation of AtJAZ6 by wild-type or mutant HopZ1a as well as the relative autoacetylation levels of HopZ1a. Equal loading of the proteins was confirmed by Coomassie Blue staining. These experiments were repeated three times with similar results.

Fig. 3

S349 and S351 are required for the hypersensitive response (HR)-triggering activity of HopZ1a in Arabidopsis thaliana ecotype Columbia (Col-0). (a) Substitutions of S349 and S351 abolished the cell death triggered by HopZ1a in A. thaliana. Pseudomonas syringae pv. tomato strain D28E (PtoD28E) expressing wild-type or mutant HopZ1a was infiltrated into leaves of 5-wk-old A. thaliana Col-0 plants at OD₆₀₀ = 0.01 (1 × 10⁷ cfu ml⁻¹). Cell death symptoms were monitored at 24 h post inoculation (hpi). PtoDC3000 carrying the empty vector pUCP20tk (EV) was used as a negative control. (b) Bacterial populations of PtoDC3000 expressing wild-type or mutant HopZ1a in A. thaliana Col-0. Arabidopsis thaliana leaves were infiltrated with PtoDC3000 carrying pUCP20tk (EV), or expressing wild-type or mutant HopZ1a at OD₆₀₀ = 0.0001 (1 × 10⁵ cfu ml⁻¹). Bacterial populations were determined at 0 and 3 d post inoculation. The average colony-forming units per square centimeter (cfu cm⁻²) and ±SD (as error bars) are presented. Different letters at the top of the bars represent data for which differences were statistically significant (two-tailed t-test; P < 0.05). The expression of HopZ1a in PtoDC3000 was confirmed by western blots (Supporting Information Fig. S3). These experiments were repeated three times with similar results.

Fig. 4

S349 and S351 are required for the virulence function of HopZ1a. (a) Mutations of S349 and S351 abolished the ability of HopZ1a to suppress flg22-mediated callose deposition. Adult leaves of 5-wk-old transgenic Arabidopsis thaliana zar1-1 plants expressing wild-type or mutant HopZ1a were infiltrated with 1 μM flg22. The infiltrated leaves were collected at 16 h post infiltration and stained with aniline blue (MP Biomedicals, Santa Ana, CA, USA). Numbers of callose deposits in the infiltrated areas were numerated using fluorescent microscopy under UV. Values are mean ±SD (as error bars) (n = 12). (b) HopZ1a with S349 and S351 mutated can no longer promote the multiplication of Pseudomonas syringae pv. syringae strain B728aΔZ3 (PsyB728aΔZ3) in Arabidopsis thaliana. Arabidopsis thaliana zar1-1 plants were infiltrated with PsyB728aΔZ3 carrying empty vector (EV), or expressing wild-type or mutant HopZ1a at OD₆₀₀ = 0.005 (5 × 10⁶ cfu ml⁻¹). Bacterial populations were determined at 0 and 3 d post inoculation. The average colony-forming units per square centimeter (cfu cm⁻²) and ±SD (as error bars) are presented. Different letters at the top of the bars represent data for which differences were statistically significant (two-tailed t-test; P < 0.05). These experiments were repeated three times with similar results.

Fig. 5

S349 and S351 are indispensable for HopZ1a to suppress stomatal defense. Leaf discs of 4-wk-old transgenic Arabidopsis thaliana zar1-1 plants expressing wild-type or mutant HopZ1a were incubated with 10 μM flg22. Two hours after flg22 treatment, stomata on the lower epidermis were observed under a microscope. (a) Micrographs of stomata were used to measure the stomatal aperture, which is expressed as the ratio of width over length. (b) The average stomatal aperture and SE (as error bars). Arabidopsis thaliana zar1-1 expressing HopZ1a(C216A) was used as a negative control. Data for which differences were statistically significant (two-tailed t-test): *, P < 0.05. The experiment was repeated twice with similar results.

Fig. 6

S349 and S351 affect the interaction of HopZ1a with inositol hexakisphosphate (IP6). (a) Analysis of the protein conformations of HopZ1a, HopZ1a(C216A) and HopZ1a(S349AS351A) in the presence or absence of IP6 by 1D ¹H nuclear magnetic resonance (NMR). Protein spectra of the aromatic region in the presence (upper traces) or absence (lower traces) of IP6 are shown. Spectral changes specifically induced by IP6 in HopZ1a and HopZ1a(C216A) are indicated by arrows. (b) In vitro acetylation assays of HopZ1a, HopZ1a(C216A) and HopZ1a(S349AS351A) in the presence or absence of IP6. Numbers underneath the autoradiograph indicate relative acetylation levels of HopZ1a mutants compared with wild-type protein in the absence of IP6 (100%). Equal loading of the proteins was confirmed by Coomassie Blue staining. The experiment was repeated twice with similar results.

Fig. 7

S349 and S351 have conserved functions in the YopJ family of acetyltransferases. (a) Amino acid sequence alignment in a conserved C-terminal region spanning residues 344–361 of HopZ1a and the corresponding region of other YopJ family effectors is shown. The accession numbers of each protein are presented in parentheses after the species name. Multiple sequence alignment was generated using C_LUSTALX (http://www.jalview.org/help/html/colourSchemes/clustal.html). A pictogram showing the relative frequency of amino acid occurrence at each position is presented above. S349 and S351 of HopZ1a are indicated by asterisks. A full-length amino acid sequence comparison is shown in Supporting Information Fig. S8. (b) In vitro acetylation assay of PopP2 produced by Ralstonia solanacerum. Tag-free PopP2, the catalytic mutant PopP2(C321A), and PopP2(S447AS449A) were purified from Escherichia coli and subjected to in vitro acetylation assay. (c) In vitro acetylation assay of HopZ3 produced by Pseudomonas syringae pv. syringae. HopZ3, the catalytic mutant HopZ3(C300A), and HopZ3(S386A) were purified from E. coli and subjected to in vitro acetylation assay. Numbers underneath the autoradiograph indicate relative acetylation levels of PopP2 or HopZ3 mutants compared with wild-type protein (100%). Equal loading of the proteins was confirmed by Coomassie Blue staining. These experiments were repeated three times with similar results.