Pathogenesis-related protein 1 suppresses oomycete pathogen by targeting against AMPK kinase complex

Abstract

Introduction: During the arms race between plants and pathogens, pathogenesis-related proteins (PR) in host plants play a crucial role in disease resistance, especially PR1. PR1 constitute a secretory peptide family, and their role in plant defense has been widely demonstrated in both hosts and in vitro. However, the mechanisms by which they control host-pathogen interactions and the nature of their targets within the pathogen remain poorly understood.

Objectives: The present study was aimed to investigate the anti-oomycete activity of secretory PR1 proteins and elaborate their underlying mechanisms.

Methods: This study was conducted in the potato-Phytophthora infestans pathosystem. After being induced by the pathogen infection, the cross-kingdom translocation of secretory PR1 was demonstrated by histochemical assays and western blot, and their targets in P. infestans were identified by yeast-two-hybrid assays, bimolecular fluorescence complementation assays, and co-immunoprecipitation assay.

Results: The results showed that the expression of secretory PR1-encoding genes was induced during pathogen infection, and the host could deliver PR1 into P. infestans to inhibit its vegetative growth and pathogenicity. The translocated secretory PR1 targeted the subunits of the AMPK kinase complex in P. infestans, thus affecting the AMPK-driven phosphorylation of downstream target proteins, preventing ROS homeostasis, and down-regulating the expression of RxLR effectors.

Conclusion: The results provide novel insights into the molecular function of PR1 in protecting plants against pathogen infection, and uncover a potential target for preventing pre- and post-harvest late blight.

Keywords: AMPK kinase complex; Cross-kingdom translocation; Host-pathogen interaction; Pathogenesis-related protein 1; Phytophthora infestans; Potato late blight.

Graphical abstract

Fig. 1

Bioinformatic analysis of StPR1 in Solanum tuberosum. (A) Ten genes encode pathogenesis-related protein 1 (PR1) in the genome of S. tuberosum. StPR1.1–1.9 are clustered on chromosome 1 whereas StPR1a is located at chromosome 9. (B) Phylogenetic analysis of PR1 proteins from S. tuberosum was performed with MEGA6.0 using the neighbor-joining method. One thousand bootstrap replicates were used. The PR1 proteins in red represent secretory proteins in S. tuberosum, while those in black indicate non-secretory proteins. The identity between StPR1.8 and other PR1 proteins was analyzed using Clustal X. The amino acid sequences were collected from the following organisms: St, S. tuberosum; At, Arabidopsis thaliana; Tom, Solanum lycopersicum; Tob, Nicotiana benthamiana. (C) The analysis of conserved domains of StPR1.1–1.9 in S. tuberosum. The secretory PR1 proteins have four highly conserved α-helices (αI-IV) and β-strands (βA-D) to form a α-β-α sandwich structure, while the non-secretory PR1 proteins were absent from the amino acid sequence of αI. All the PR1 proteins have a conserved “GHYTQVVW” motif, and six conserved cysteine residues (Cys) to form three disulfide bonds. The red blank indicates a specific sequence between secretory and non-secretory PR1 proteins, which was abbreviated as TM (type-specific motif). The words in red background represent completely uniform sequences.

Fig. 2

StPR1 involved in plant defense against Phytophthora infestans in vivo. (A, B) Compared with the wild-type ‘Desiree’ (WT), the disease resistance to P. infestans in transgenic lines overexpressing StPR1.2, StPR1.3 and StPR1.8 was enhanced. 35S::GUS represent GUS in S. tuberosum under 35S promoter. (C, D) Compared with the wild-type resistant variety ‘E3′ (WT), the RNAi lines RNAi-PR1 and RNAi-SPR1 displayed enhanced susceptibility to P. infestans with higher DI, and the disease symptoms on RNAi-PR1 lines were more serious. RNAi-PR1 indicates that all the PR1 genes were knocked down, and the RNAi-SPR1 indicates that only the secretory PR1 genes were knocked down. The DIs of transgenic lines were measured after inoculating 1 × 10⁷ conidia of wild-type P. infestans for 5 days. (E, F) Heterologous overexpression of StPR1.2, StPR1.3 and StPR1.8 in P. infestans inhibited the colony growth and decreased the growth rate of P. infestans. WT, wild-type P. infestans; OE-StPR1, P. infestans mutants overexpressing StPR1. (G, H) Reduced pathogenicity of OE-StPR1.2, OE-StPR1.3 and OE-StPR1.8 on ‘Desiree’ leaves with lower DIs after 5 days of inoculation. The leaves infected by WT were used as a control. (I, J) Heterologous expression of StPR1 in P. infestans decreased the infection on tubers after 10 days of inoculation. The tubers infected by WT were used as control. CK in G-J, leaves and tubers inoculated with sterile water. Each experiment has three biological replicates. Each replication of pathogenicity tests contained at least 10 leaves and tubers, respectively. Error bars indicate standard deviation (SD). Asterisks indicate significant differences (*P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001; t-test).

Fig. 3

The GH and TM motifs are important for the anti-oomycete activity of StPR1 against Phytophthora infestans. (A) Overview of five types of StPR1 mutations (StPR1^ΔSP, StPR1^ΔTM, StPR1^ΔSPΔTM, StPR1^ΔGH and StPR1^ΔSPΔGH) that were changed at the SP, GH and TM motifs. (B) The significant inhibition of prokaryotic expressed StPR1, StPR1^ΔSP, StPR1^ΔSPΔGH and StPR1^ΔSPΔTM on sporangia germination of P. infestans. The sporangia germination rates treated with StPR1^ΔSPΔGH and StPR1^ΔSPΔTM were higher than those treated with StPR1 and StPR1^ΔSP. The sporangia germination treated with GUS was used as a control (CK). (C) Three necrosis grades are defined according to the percentage of the necrosis area developed in the pathogen inoculation area. (D) Phenotypes of plant necrosis triggered by P. infestans in Nicotiana benthamiana leaves transiently expressing PR1, PR1^ΔSP, PR1^ΔTM, PR1^ΔSPΔTM, and PR1^ΔGH indicate that the deletion of SP, GH and TM motifs decreased the anti-oomycete activity of StPR1. The necrosis caused by P. infestans in N. benthamiana expressing GFP was used as a control (CK). (E-G) The disease grades of plant necrosis triggered by P. infestans in N. benthamiana expressing StPR1 and mutations. One-tailed t-tests were used to assess statistical significance between means. Error bars indicate standard deviation (SD). In (B, E, F, and G): *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001.

Fig. 4

Cross-kingdom translocation of StPR1 from host to Phytophthora infestans in vivo and in vitro. (A, B) The GUS histochemical assays of 35S::StPR1.2^GUS transgenic plants during P. infestans infection (B). The assays in 35S::GUS were used as control (A). The GUS staining was attenuated in 35S::StPR1.2^GUS as the inoculation day progressed, while that was not in the 35S::GUS. (C) GUS histochemical assays showed that the prokaryotic expressed StPR1.2 and its mutations tagged with GUS could translocate into P. infestans cells after 12 h of co-incubation in vitro. The hyphae co-incubated with GUS were used as control. The scale bars represent 25 μm. (D, F) Western blot analysis further confirmed the presence of StPR1.2 (D), StPR1.3 (E), StPR1.8 (F) proteins and their mutations in the hyphae of P. infestans cells after 12 h of co-incubation. Purified prokaryotic expressed StPR1 and mutations are marked with red asterisks. Total proteins of P. infestans used for western blot analysis were extracted after 12 h of co-incubation with indicated proteins. β-actin was used as a loading control. (G) The specific marker protein of extracellular vesicles (EVs), TET8, was detected to accumulate significantly during P. infestans infection by western blot analysis. β-actin was used as a loading control.

Fig. 5

PiSNF1 is the target of StPR1. (A) Phylogenetic relationship among catalytic subunits of AMPK from different species. The phylogenetic tree was constructed using MEGA6.0 based on an alignment generated with ClustalX. Sl: Solanum lycopersicum, St: Solanum tuberosum, At: Arabidopsis thaliana, Pi: Phytophthora infestans, Sc: Saccharomyces cerevisiae, Vd: Verticillium dahliae, Fo: Fusarium oxysporum, GRCH: human. (B) StPR1 proteins interacts with PiSNF1 as determined by Y2H assays. PiSNF1 on pGBKT7 (BD) vector was used to confirm the interactions with secretory StPR1^ΔSP cloned into pGADT7 (AD). Yeast transformants were grown on SD/-Leu/-Trp/-His/-Ade medium with 40 g/mL X-α-gal, and the blue colony indicate interactions. Pictures were taken after 3 d of culture. (C) Y2H confirmed the interaction between PiSNF1 and StPR1^ΔSP using the target proteins as baits. (D) Co-IP assays demonstrating the interactions of StPR1.2, StPR1.3, and StPR1.8 with PiSNF1. The proteins used for Co-IP were derived from S. cerevisiae co-expressing PiSNF1^Myc with StPR1.2^HA, StPR1.3^HA, and StPR1.8^HA. (E) Co-localization of RFP-labeled StPR1.2, StPR1.3, StPR1.8 and GFP-labeled PiSNF1. The fusion constructs were transiently co-expressed in N. benthamiana and the expressions were analyzed by confocal microscopy at 3 dpi. Bar = 50 μm. The N. benthamiana infiltrated with MMA buffer was used as control.

Fig. 6

AMPK kinase complex in Phytophthora infestans is the target of StPR1. (A) Structures of the AMPK complex. AMPK exists as a heterotrimer consisting of a catalytic subunit (α) and two regulatory subunits (β and γ). (B, C) Y2H (B) and bimolecular fluorescence complementation (BiFC) (C) assays showed that B-PiSNF1, B-PiAMPKβ and PiAMPKγ interact with each other to form a heterotrimer as those in plant and animal. (D) B-PiSNF1, B-PiAMPKβ and PiAMPKγ interact with secretory StPR1^ΔSP in Y2H assays. (E) The AMPK subunits interact with StPR1.2^ΔSP, StPR1.3^ΔSP, and StPR1.8^ΔSP in BiFC assays. The StPR1.2^ΔSP, StPR1.3^ΔSP, and StPR1.8^ΔSP fused with the N terminus of YFP, and AMPK subunits fused with the C terminus of YFP. The constructs were transiently co-expressed in N. benthamiana and examined by confocal microscopy at 3 d post-infiltration (dpi). Only the nYFP halves and single proteins fused to cYFP alone were used as control. The complementation of fluorescence indicates interaction between assayed proteins.

Fig. 7

StPR1 did not change the levels of pAMPKα, while affected the AMPK phosphorylation to downstream target proteins. (A) Western blot of AMPK and phosphorylation of AMPKα (pAMPKα) in wild-type (WT) and PR1 overexpression mutants of P. infestans (OE-StPR1.2, OE-StPR1.3 and OE-StPR1.8). β-actin was used as a loading control. The AMPK expression and pAMPKα level were not different between WT and mutants. (B) Quantification of pAMPKα from WT, OE-StPR1.2, OE-StPR1.3 and OE-StPR1.8 (n = 3) through ELISA analysis. The pAMPKα levels were not significantly different among them. (C, D) Phosphorylation of AMPKα under the involvement of StPR1.2^ΔSP, StPR1.3^ΔSP, StPR1.8^ΔSP, and their mutations through phosphorylation assay in vitro. StPR1.2^ΔSP, StPR1.3^ΔSP, StPR1.8^ΔSP, and their mutations had no influence on the phosphorylation levels of AMPKα. StPR1^ΔSP: deleting the signal peptide; StPR1^ΔSPΔTM: deleting the signal peptide and TM motif; StPR1^ΔSPΔGH: deleting the signal peptide and GH motif. (E) Quantification of pACC:ACC ratio from WT, OE-StPR1.2, OE-StPR1.3 and OE-StPR1.8 (n = 3) through ELISA analysis. The pACC:ACC ratio was increased significantly in OE-StPR1.2, OE-StPR1.3 and OE-StPR1.8. (F, H) Quantification of pACC:ACC ratio after 48 h of treatment with StPR1^ΔSP, StPR1^ΔSPΔTM, and StPR1^ΔSPΔGH (n = 3) through ELISA analysis. The analysis of pACC:ACC ratio of P. infestans without PR1 treatment was used as a control. Results that were significantly different from control (CK) were indicated by asterisks. One-tailed t-tests were used to assess significances. Error bars indicate standard deviation (SD). In E-H: *, P ≤ 0.05; **, P ≤ 0.01; ***, P ≤ 0.001.

Fig. 8

Functional characterization of PiSNF1 and the influence of AMPK activation and inhibition on the colony growth and pathogenicity of Phytophthora infestans. (A) The colony phenotype of PiSNF1 overexpression mutant (OE-PiSNF1). (B) GFP fluorescence observation of OE-PiSNF1 under a fluorescence microscope. (C) Overexpression of PiSNF1 decreased the colony growth rate of P. infestans. (D, E) Reduced pathogenicity of OE-PiSNF1 on ‘Desiree’ leaves with lower DI after inoculating for 5 days. (F, G) The pathogenicity of OE-PiSNF1 on tubers was decreased. (H) Western blot analysis of pAMPKα in wild-type P. infestans (WT), wild-type P. infestans treated with DMSO (DMSO), wild-type P. infestans treated with AMPK activator A-769662 (A-769662), wild-type P. infestans treated with AMPK kinase inhibitor dorsomorphin (Dorsomorphin), and OE-PiSNF1. (I) Quantification of pAMPKα in OE-PiSNF1 mutant and under AMPK activator/inhibitor treatment. (J, K) Reduced colony growth of P. infestans following dorsomorphin treatment. The colony growth of P. infestans under the treatment of A-769662 and DMSO was not different from that of wild-type. (L, M) Reduced pathogenicity of Dorsomorphin on ‘Desiree’ leaves with lower DI after 5 days of inoculation. (N, O) Dorsomorphin decreased the infection of P. infestans on tubers. In pathogenicity tests, the leaves or tubers infected by wild-type P. infestans were used as positive control (WT). CK, leaves or tubers inoculated with sterile water. Each pathogenicity test contained at least 10 leaves or tubers, respectively. Each experiment was repeated three times. Asterisks indicate significant differences (*P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001; t-test).

Fig. 9

StPR1-PiAMPK signaling pathway is involved in regulating ROS homeostasis and RxLR expression in Phytophthora infestans. (A) The H₂O₂ content was lower in OE-StPR1.2, OE-StPR1.3, OE-StPR1.8, and OE-PiSNF1 than in wild-type P. infestans (WT). (B-D) Compared with WT, the enzyme activities of SOD, POD, and CAT were significantly increased in OE-StPR1.2, OE-StPR1.3, OE-StPR1.8, and OE-PiSNF1 mutants. (E) qRT-PCR analysis showed that compared with WT, the transcript of RxLR effectors encoding genes were down-regulated in OE-StPR1.2, OE-StPR1.3, OE-StPR1.8, and OE-PiSNF1 mutants. Y-axis indicated the relative normalized expression. Error bars represent standard error from three independent replicates. Asterisks indicate significant differences (*P ≤ 0.05; ***P ≤ 0.001; t-test).