Dynamics of defense responses and cell fate change during Arabidopsis-Pseudomonas syringae interactions

Abstract

Plant-pathogen interactions involve sophisticated action and counteraction strategies from both parties. Plants can recognize pathogen derived molecules, such as conserved pathogen associated molecular patterns (PAMPs) and effector proteins, and subsequently activate PAMP-triggered immunity (PTI) and effector-triggered immunity (ETI), respectively. However, pathogens can evade such recognitions and suppress host immunity with effectors, causing effector-triggered susceptibility (ETS). The differences among PTI, ETS, and ETI have not been completely understood. Toward a better understanding of PTI, ETS, and ETI, we systematically examined various defense-related phenotypes of Arabidopsis infected with different Pseudomonas syringae pv. maculicola ES4326 strains, using the virulence strain DG3 to induce ETS, the avirulence strain DG34 that expresses avrRpm1 (recognized by the resistance protein RPM1) to induce ETI, and HrcC(-) that lacks the type three secretion system to activate PTI. We found that plants infected with different strains displayed dynamic differences in the accumulation of the defense signaling molecule salicylic acid, expression of the defense marker gene PR1, cell death formation, and accumulation/localization of the reactive oxygen species, H2O2. The differences between PTI, ETS, and ETI are dependent on the doses of the strains used. These data support the quantitative nature of PTI, ETS, and ETI and they also reveal qualitative differences between PTI, ETS, and ETI. Interestingly, we observed the induction of large cells in the infected leaves, most obviously with HrcC(-) at later infection stages. The enlarged cells have increased DNA content, suggesting a possible activation of endoreplication. Consistent with strong induction of abnormal cell growth by HrcC(-), we found that the PTI elicitor flg22 also activates abnormal cell growth, depending on a functional flg22-receptor FLS2. Thus, our study has revealed a comprehensive picture of dynamic changes of defense phenotypes and cell fate determination during Arabidopsis-P. syringae interactions, contributing to a better understanding of plant defense mechanisms.

Figure 1. Dynamic changes in SA accumulation and PR1 expression during PTI, ETS, and ETI.

The fourth to sixth leaves of 30-day-old Col-0 plants were infected with Pseudomonas syringae pv. maculicola ES4326 strains, DG3 at OD₆₀₀ 0.01 or 0.001, DG34 at OD₆₀₀ 0.01 or 0.001, or HrcC^- at OD₆₀₀ 0.1 or 0.01. The infected leaves were collected at the indicated times for SA and RNA analysis. (A) SA quantitation by HPLC analysis. Statistical analysis was performed with one-way ANOVA Fisher’s PLSD tests (StatView 5.0.1). Different letters indicate significant difference among the samples at the same time point (P<0.05). (B) Northern blotting for PR1 expression. Image of rRNA was used for a loading control.

Figure 2. Dynamic ROS accumulation and localization during PTI, ETS, and ETI.

The fourth to sixth leaves of 30-day-old plants were infiltrated with P. syringae strains at OD₆₀₀ 0.01, using 10 mM MgSO₄ treatment as a control. The infiltrated leaves were collected at the indicated times and were further cut into 1x2 mm sections. The sections were incubated with freshly prepared 5 mM CeCl₃ in 50 mM MOPS at pH 7.2 or MOPS without CeCl₃ for 1 h. The samples were then fixed and processed for TEM imaging. At least six different leaf samples for each treatment were fixed, and six sections were observed in each sample. (A) Cell morphology at 0 h. (B-C) H₂O₂ localization at 6 h (B) and 24 h (C) after DG34 inoculation. (D-F) H₂O₂ localization at 6 h (D), 18 h (E) and 24 h (F) after DG3 inoculation. (G-H) H₂O₂ localization at 18 h (G) and 48 h (H) after HrcC^- inoculation. Arrows indicate electron-dense cerium deposits. Asterisks indicate bacteria. Ch, chloroplast; CW, cell wall; M, mitochondrion; P, peroxisome.

Figure 3. The abnormal growth is mainly induced during PTI.

The fourth to sixth leaves of 30-day-old Col-0 plants were infected with P. syringae strains and observed for leaf morphology. (A) Pictures of leaf cross sections. Infected leaves were collected at 4 dpi and fixed for embedding with LR White resin. One-micron sections were cut and stained with 1% toluidine blue O for photographing. Leaves infected with DG3 (0.01) were mostly dead at 4 dpi and thus no data is available. Arrows indicate enlarged cells. The size bar represents 200 μm and applies to all images. Each growth (or a protrusion) has multiple enlarged cells. (B) Quantitation of abnormal growths. The number of abnormal growths, appearing to be transparent protrusions on the treated leaves, was counted at 4 dpi with a dissecting microscope. At least 25 leaves from each treatment were used in the counting. Statistical analysis was performed with one-way ANOVA Fisher’s PLSD tests (StatView 5.0.1). Different letters indicate significant difference among the samples (P<0.05).

Figure 4. The enlarged cells have increased nuclear DNA content.

P. syringae- infected leaves were embedded in paraplast and cut into 15 μm sections for DAPI staining. Images of leaf cross-sections stained with DAPI to show nuclei were captured with a constant exposure time, using a Nikon DS cooled camera attached to a compound microscope. The images are from the following treatments: (A) 10 mM MgSO4, (B) DG34 0.01, (C) DG34 0.001, (D) HrcC^- 0.1, and (E) HrcC^- 0.01. (F) Typical nuclei of guard cells from a mock-treated leaf. Guard cells from leaves infected with DG34 0.01, DG34 0.001, HrcC^- 0.1, or HrcC^- 0.01 show visually similar nuclei (data not shown). (G) A typical nucleus of a normal mesophyll cell from a mock-treated leaf. Normal sized mesophyll cells from leaves infected with DG34 0.01, DG34 0.001, HrcC^- 0.1, or HrcC^- 0.01 show visually similar nuclei (data not shown). (H) A typical nucleus of an enlarged mesophyll cell induced by HrcC^- 0.01. Enlarged cells from leaves infected with DG34 0.01, DG34 0.001, or HrcC^- 0.1 also show large nuclei (data not shown). Arrows indicate the large nuclei in (A) to (E). The size bar in (A) represents 100 μm and applies to panels (A) to (E) while the size bar in (F) represents 5 μm and applies to panels (F) to (H). (J) Relative nuclear DNA content. The average fluorescence of nuclei of guard cells from mock-treated leaves (1) was set as 2C and was used to quantify relative nuclear DNA content of normal mesophyll cells from mock-treated leaves (2) and the enlarged mesophyll cells induced by HrcC^- (0.01) (3). At least 60 nuclei were used for each data point. Nuclear DNA contents of guard cells and normal mesophyll cells from HrcC^- (0.01)-infected leaves are similar to those of their corresponding cells from mock-treated leaves (data not shown). Statistical analysis was performed with one-way ANOVA Fisher’s PLSD tests (StatView 5.0.1). Different letters indicate significant difference among the samples (P<0.05).

Figure 5. FLS2-mediated signaling but not SA induces cell enlargement in Arabidopsis leaves.

Leaves infiltrated with HrcC^- (OD₆₀₀ 0.1), flg22 (1 μM or 10 μM), or BTH (10 μM or 300 μM) were quantified for the formation of abnormal growths 4 days post treatment, using a dissecting microscope. (A) Flg22-induced cell enlargement is FLS2-dependent. (B) HrcC^- partially requires FLS2 to induce large cells. (C) BTH treatment does not induce abnormal growth. (D) HrcC^--induced abnormal growth requires SID2 and NPR1. For quantification of abnormal growths, at least 25 leaves from 12 plants were used for each BTH treatment or bacterial infection and 18 leaves from 7 plants were used for each flg22 treatment. Statistical analysis was performed with one-way ANOVA Fisher’s PLSD tests (StatView 5.0.1). Asterisks in (A), (B), and (D) indicate significant difference between treatments of the same genotype (P<0.05).