RIN13 is a positive regulator of the plant disease resistance protein RPM1

Abstract

The RPM1 protein confers resistance to Pseudomonas syringae pv tomato DC3000 expressing either of the Type III effector proteins AvrRpm1 or AvrB. Here, we describe the isolation and functional characterization of RPM1 Interacting Protein 13 (RIN13), a resistance protein interactor shown to positively enhance resistance function. Ectopic expression of RIN13 (RIN13s) enhanced bacterial restriction mechanisms but paradoxically abolished the normally rapid hypersensitive response (HR) controlled by RPM1. In contrast with wild-type plants, leaves expressing RIN13s did not undergo electrolyte leakage or accumulate H2O2 after bacterial delivery of AvrRpm1. Overexpression of RIN13 also altered the transcription profile observed during a normal HR. By contrast, RIN13 knockout plants had the same ion leakage signatures and HR timing of wild-type plants in response to DC3000(avrRpm1) but failed to suppress bacterial growth. The modified phenotypes seen in the RIN13s/as plants were specific to recognition of AvrRpm1 or AvrB, and wild-type responses were observed after challenge with other incompatible pathogens or the virulent DC3000 isolate. Our results suggest that cell death is not necessary to confer resistance, and engineering enhanced resistance without activation of programmed cell death is a real possibility.

Figure 1.

Characterization of the RIN13–RPM1 Interaction. (A) Full-length RPM1, the extended RPM1 NB-ARC domain (F1), and N- and C-terminal deletions of the the NB-ARC domain (F2 and F3) were cloned into the yeast two-hybrid bait vector (pEG202), and interaction with RIN13 was tested by assaying for reporter (β-galactosidase) activity. (B) The interaction specificity of RIN13 was tested using nearly matched endpoints of the related resistance gene RPP5 and matched endpoint baits from RPS2 and the Brassica napus RPM1 alleles, 1A and 9N, cloned in frame into pEG202. Only 9N interacted with RIN13. In this diagram, the two most variant substitutions in 9N are indicated by an asterisk. A single nonconserved amino acid change most likely to account for the absence of an interaction between 1A and RPM1(F1), an R-to-M substitution, is boxed. (C) RIN13 interacts with the NB-ARC domain of RPM1 in vitro. Crude bacterial extracts expressing Intein-RIN13 or an Intein control were bound to chitin beads. Crude yeast extracts expressing HA epitope-tagged F1 (left panel) or F2 RPM1 (center panel) baits (see Figure 1) were added to the beads, and the mixture was washed and binding complexes fractionated by SDS-PAGE. F1- or F2-RPM1 bound to RIN13-Intein was visualized with anti-HA antisera. The respective Intein and Intein-RIN13 Coomassie-stained loadings for each lane are shown below the immunoblot. Crude F1 and F2 fractions (1/14th of input) are shown in the right panel at right.

Figure 2.

Overexpression of RIN13 Suppresses the HR. Typical interaction phenotypes for homozygous RIN13s and RIN13as lines or wild-type (Col-5) plants after challenge (inocula, 2 × 10⁷ cfu/mL) with the following pathogens: (A) to (E), DC3000(avrRpm1); (F), DC3000; (G) to (I), DC3000(avrRpt2). Leaves were photographed at 5 h ([A] to [C]), 16 h ([G] to [I]), 20 h (D), or 24 h ([E] and [F]) after challenge. RIN13s lines fail to induce an HR after challenge with DC3000(avrRpm1), and leaves collapse coincident with those undergoing a compatible interaction (cf. [E] and [F]).

Figure 3.

RPM1-Mediated Resistance Is Enhanced by Ectopic Expression of RIN13. (A) Overexpression of RIN13 enhances resistance specified by RPM1. Bacterial growth was compared in RIN13s (black bars) with Col-5 (white bars) after challenge with DC3000(avrRpm1) resuspended to 0.8 × 10⁵ cfu/mL. Each growth measurement represents the average bacterial count derived from six plants, sampling three leaves/plant at each time point. The results were repeated three times with similar results. (B) RIN13 overexpression does not affect resistance responses elicited by unrelated R genes or the virulent carrier isolate. Leaf discs were sampled at the appropriate time after the following challenges: RIN13s with DC3000(avrRpt2) (black bars) or DC3000 (light-gray bars) and Col-5 with DC3000(avrRpt2) (white bars) or DC3000 (dark-gray bars). The experiment was repeated three times, and in all cases no difference in bacterial growth could be distinguished between the treatments.

Figure 4.

Reduction in RIN13 Expression Results in Enhanced Susceptibility to DC3000(avrRpm1) but Not DC3000 or DC3000 Carrying avrRpt2. (A) Bacterial growth was significantly enhanced in RIN13as (black bars) lines compared with Col-5 (white bars) after challenge with DC3000(avrRpm1). The results were repeated at least three times with similar results. (B) RIN13 diminution does not modify responses to bacteria carrying avrRpt2. Bacterial growth in RIN13s leaves [DC3000(avrRpt2) (black bars) or DC3000 (light-gray bars)] was identical to growth measured in Col-5 control leaves inoculated with DC3000(avrRpt2) (white bars) or DC3000 (dark-gray bars). The experiment is representative of three replicates. (C) RIN13 knockout plants phenocopy the enhanced susceptibility of RIN13as lines. Levels of bacterial growth were significantly enhanced in leaves of ΔRIN13 compared with Col-5 plants after challenge with DC3000(avrRpm1) (black and white bars, respectively). As expected, no differences were measured in response to challenge with virulent DC3000 (light-gray and dark-gray bars, respectively). These assays were repeated twice with similar results.

Figure 5.

Changes in RIN13 Expression Modify Physiological Responses after RPM1 Elicitation. (A) Electrolyte leakage is suppressed in RIN13s but not in ΔRIN13 lines after RPM1 elicitation. Ion leakage in ΔRIN13 leaves challenged with DC3000(avrRpm1) (red) was indistinguishable from a wild-type (Col-0) resistance response (blue). However, similarly challenged RIN13s lines (black) exhibited restricted ion leakage identical to DC3000-challenged Col-0 (yellow) or RIN13s (green) leaves. (B) Cell death is suppressed in RIN13s but not ΔRIN13 leaves in response to an avirulent pathogen. Lactophenol trypan blue exclusion staining was used to monitor cell viability after challenge with DC3000(avrRpm1). At 6 hpi, no cell death was evident in RIN13s leaves (panel A) in contrast with ΔRIN13 (panel B) and control leaves (panel C). Isolated microscopic patches of dying cells were detected at 14 hpi (panel D) and with increasing frequency (17 hpi, panel E) until confluent staining at 24 hpi (panel F). (C) RIN13s leaves fail to accumulate H₂O₂ during an incompatible response. H₂O₂ accumulation was measured by DAB polymerization (Thordal-Christensen et al., 1997) in leaves harvested 4 hpi with DC3000(avrRpm1). Macroscopic reddish-brown deposits characteristic of ROI generation were detected in the wild type but not in RIN13s leaves, indicative of suppression of ROI generation mechanisms.

Figure 6.

Biophoton Generation Is Abolished in RIN13s Lines after Challenge with DC3000(avrRpm1) or Conditional Expression of avrRpm1. (A) Bright-field image of RIN13s, RIN13as, and control Col-5 plants challenged with DC3000(avrRpm1) (open circles). (B) No biophoton emission was detected in inoculated RIN13s leaves, consistent with the absence of the HR (Bennett et al., 2005). (C) Bright-field image of RIN13s/DEX:avrRpm1 and DEX:avrRpm1 lines immediately after application of 10 μM DEX (open circles) or inoculation with DC3000(avrRpm1) (A₆₀₀ 0.05). (D) Strong biophoton emission is detected in DEX:avrRpm1 plants after either bacterial inoculation or DEX application but not in the RIN13s/DEX:avrRpm1 line.

Figure 7.

RPM1-Elicited RIN13s Plants Have Modified Gene Expression Patterns. The induction of RPM1-specific gene transcripts (de Torres et al., 2003) was monitored after DC3000(avrRpm1) challenge (2 × 10⁷ cfu/mL). RIN13s plants showed enhanced induction of RIPK compared with wild-type plants. By contrast, induction of the TONB transcript was significantly delayed in RIN13s lines.

Figure 8.

Model for RIN13 Function. In the nonelicited state, RPM1 adopts a conformational state in which the RIN13 binding site in the NB-ARC domain is buried (A). Upon elicitation, conformational changes, possible induced by AvrRpm1 phosphorylation of RIN4, expose the RIN13 binding site and allow RIN13 to cooperate in normal defense signaling processes in conjunction with one or more unknown interactors (?) (B). In ΔRIN13 lines, absence of RIN13 prevents elaboration of full wild-type resistance responses but does not compromise signaling through pathways that elicit hypersensitive cell death (C). By contrast, overexpressed RIN13 preferentially occupies binding sites that activate bacterial restriction mechanisms. Simultaneously, RIN13 occupation prevents or hinders signaling components that activate the HR resulting in suppression of cell death (D).