Linking ligand perception by PEPR pattern recognition receptors to cytosolic Ca2+ elevation and downstream immune signaling in plants

Abstract

Little is known about molecular steps linking perception of pathogen invasion by cell surface sentry proteins acting as pattern recognition receptors (PRRs) to downstream cytosolic Ca(2+) elevation, a critical step in plant immune signaling cascades. Some PRRs recognize molecules (such as flagellin) associated with microbial pathogens (pathogen-associated molecular patterns, PAMPs), whereas others bind endogenous plant compounds (damage-associated molecular patterns, DAMPs) such as peptides released from cells upon attack. This work focuses on the Arabidopsis DAMPs plant elicitor peptides (Peps) and their receptors, PEPR1 and PEPR2. Pep application causes in vivo cGMP generation and downstream signaling that is lost when the predicted PEPR receptor guanylyl cyclase (GC) active site is mutated. Pep-induced Ca(2+) elevation is attributable to cGMP activation of a Ca(2+) channel. Some differences were identified between Pep/PEPR signaling and the Ca(2+)-dependent immune signaling initiated by the flagellin peptide flg22 and its cognate receptor Flagellin-sensing 2 (FLS2). FLS2 signaling may have a greater requirement for intracellular Ca(2+) stores and inositol phosphate signaling, whereas Pep/PEPR signaling requires extracellular Ca(2+). Maximal FLS2 signaling requires a functional Pep/PEPR system. This dependence was evidenced as a requirement for functional PEPR receptors for maximal flg22-dependent Ca(2+) elevation, H(2)O(2) generation, defense gene [WRKY33 and Plant Defensin 1.2 (PDF1.2)] expression, and flg22/FLS2-dependent impairment of pathogen growth. In a corresponding fashion, FLS2 loss of function impaired Pep signaling. In addition, a role for PAMP and DAMP perception in bolstering effector-triggered immunity (ETI) is reported; loss of function of either FLS2 or PEPR receptors impaired the hypersensitive response (HR) to an avirulent pathogen.

Fig. 1.

Pep and flg22 effects on cytosolic Ca²⁺. Cytosolic Ca²⁺ level was recorded before and after ligand addition at time 0 and used to calculate the change in cytosolic Ca²⁺ from the minimum level recorded at or near time 0. Results are presented as means ± SE (replicate no. in parentheses) calculated every minute. (A) Effect of Pep1 on WT-aeq and pepr2-aeq leaves. (B) Effect of Pep3 (on WT-aeq leaves) added after pretreatment of leaves for 10 min with or without the GC inhibitor LY83583 (control). (C–F) Comparisons of peptide-dependent changes in WT-aeq plants with either dnd1-aeq or ip5-ptase-aeq plants. Recordings were made from WT-aeq (Col) plants (black lines). Also shown are changes in Ca²⁺ calculated for dnd1-aeq plants (C and D) and ip5-ptase-aeq plants (E and F). Pep3 (20 nM) was used in C and E. Flg22 (1 µM) was used in D and F.

Fig. 2.

Effect of Pep1 on in planta cytosolic [cGMP] in the root tip of FlincG seedlings. After application of Pep1 (or water as a control) to seedlings mounted on the stage of a confocal microscope, fluorescence images were taken over time. Images of the root tip at 0, 5, 10, and 15 min after application of Pep1 are shown in A–D, respectively. No changes were evident to the eye in corresponding images of water-treated roots. (E) Quantitative analysis of mean (n = 4) ± SE change in fluorescence (F/F₀) over time is shown for Pep1- and water-treated root tips. Overlay of fluorescence and bright field images (Fig. S2A) indicated that cells of the root tip columella and epidermis were responsive to the applied ligand. Application of PAMPs such as flg22 (21) or DAMPs such as Peps (22) to Arabidopsis roots evoke pathogen defense signaling cascades. Results were recorded from root tips, as they are typically considered the best tissue for live-cell fluorescence imaging (23). Immune responses evoked by flg22 occur mainly at the root elongation zone but are excluded from the root tip (21); nothing is known about Pep-responsive zones of the root. Therefore, as the FlincG assay was developed, further studies (Fig. S2B) were done focusing on the root elongation zone, which displayed a similar response to Pep1 application as was observed at the root tip, i.e., an increase in F/F₀.

Fig. 3.

Ca²⁺ effects on flg22 and Pep3 signaling (A) and PEPR1 GC domain mutation effects on Pep3 signaling (B) monitored as WRKY33 expression in isolated protoplasts. (A) Protoplasts isolated from WT plants were exposed to water (control), Pep3, or flg22 in the presence or absence of external Ca²⁺. WRKY33 expression level was normalized to that occurring in the absence of added Ca²⁺ or peptide. (B) Protoplasts isolated from leaves of pepr1/2 double null mutant plants were transformed with a construct encoding either an HA-fusion protein of WT PEPR1 (Left four bars) or an HA fusion of mutant PEPR1 (PEPR1^m) (Right four bars); in both cases PEPR1 expression was controlled by the cauliflower mosaic virus 35S promoter. For B, only Pep3 was tested as an activating ligand. Results were normalized to WRKY33 expression level of protoplasts transformed with WT PEPR1 in the absence of both added Ca²⁺ and Pep3 ligand. Results are shown as means of biological replications (n = 3 ± SE); i.e., different protoplast preparations were prepared from different plants. For A and B, a Tukey–Kramer multiple comparisons test ANOVA was used to evaluate means separation; bars with different letters are significantly different (P < 0.05). Results shown here (data compiled from three different protoplasts preparations) were repeated a total of at least two times with similar results.

Fig. 4.

Relative expression of WRKY33 in response to application of Pep3, flg22, and water (control) to leaves of WT, pepr1, pepr2, fls2, and pepr1/2 plants. Quantitative real-time PCR analysis of WRKY33 transcript levels was undertaken; results shown are means ± SE (n = 6) of transcript levels normalized to the level in WT tissue after application of water.

Fig. 5.

Signaling mediated by the PEPR receptors and FLS2 have interdependent effects on growth of a virulent pathogen and are required for unimpaired HR. (A) Proliferation of (virulent) Pst DC3000 on leaves of WT, pepr1/2, and fls2 mutant plants pretreated with water, flg22, or Pep3. Plants were pretreated 1 d before inoculation with Pst and bacterial growth was evaluated 3 d after inoculation. Results shown (note log scale of ordinate axis) are mean values of Pst recovered from leaves (n = 4) ± SE. * above a bar represents bacterial growth in leaves of either pepr1/2 or fls2 genotypes, indicating that for that genotype, bacterial growth in leaves subjected to the flg22 (or Pep3) pretreatment was significantly different (at P < 0.01) than the level found in WT leaves (for that pretreatment). Separately, a Tukey–Kramer multiple comparisons test was used to evaluate means separation for plants of the three genotypes subjected to the water pretreatment; bars with different letters above them were significantly different (at P < 0.05). (B) Hypersensitive response, evaluated quantitatively as increasing ion leakage (i.e., conductivity), in response to inoculation of leaves of WT, pepr1/2 double mutant, or fls2 mutant plants with Pst avrRpt2. Results are presented as means (n = 4) ± SE. This experiment was repeated with another set of plants; ion leakage differences between WT and mutant genotypes were similar as shown here. In addition, visual observation of HR-related tissue necrosis at 30 h postinoculation (30) indicated similar differences as the quantitative analysis shown here, i.e., that in leaves of WT plants, HR had commenced, whereas in leaves of pepr1/2 and fls2 mutant plants, no visual evidence of HR was observed.