Xanthomonas campestris overcomes Arabidopsis stomatal innate immunity through a DSF cell-to-cell signal-regulated virulence factor

Abstract

Pathogen-induced stomatal closure is part of the plant innate immune response. Phytopathogens using stomata as a way of entry into the leaf must avoid the stomatal response of the host. In this article, we describe a factor secreted by the bacterial phytopathogen Xanthomonas campestris pv campestris (Xcc) capable of interfering with stomatal closure induced by bacteria or abscisic acid (ABA). We found that living Xcc, as well as ethyl acetate extracts from Xcc culture supernatants, are capable of reverting stomatal closure induced by bacteria, lipopolysaccharide, or ABA. Xcc ethyl acetate extracts also complemented the infectivity of Pseudomonas syringae pv tomato (Pst) mutants deficient in the production of the coronatine toxin, which is required to overcome stomatal defense. By contrast, the rpfF and rpfC mutant strains of Xcc, which are unable to respectively synthesize or perceive a diffusible molecule involved in bacterial cell-to-cell signaling, were incapable of reverting stomatal closure, indicating that suppression of stomatal response by Xcc requires an intact rpf/diffusible signal factor system. In addition, we found that guard cell-specific Arabidopsis (Arabidopsis thaliana) Mitogen-Activated Protein Kinase3 (MPK3) antisense mutants were unresponsive to bacteria or lipopolysaccharide in promotion of stomatal closure, and also more sensitive to Pst coronatine-deficient mutants, showing that MPK3 is required for stomatal immune response. Additionally, we found that, unlike in wild-type Arabidopsis, ABA-induced stomatal closure in MPK3 antisense mutants is not affected by Xcc or by extracts from Xcc culture supernatants, suggesting that the Xcc factor might target some signaling component in the same pathway as MPK3.

Figure 1.

Reversal of bacteria-induced stomatal closure by Xcc depends on an intact rpf/DSF signaling system. A and B, Promotion of stomatal closure in epidermal peels from Arabidopsis Ler (A) or Col-0 (B) after incubation with bacterial strains during 1 or 3 h. C, Promotion of stomatal closure after 3 h by wild-type Xcc and rpf mutants, and by coincubation with two different strains. Datasets marked with an asterisk are significantly different from controls (E. coli 3 h in A and B; rpfC and rpfF in C) as assessed by Student's t test: *, P < 0.001. D, Bacterial growth after infection of Col-0 plants by dipping with Xcc strains (10⁷ cfu/mL). The mean and se of three independent experiments are given.

Figure 2.

Xcc, but not rpfC or rpfF, interferes with promotion of stomatal closure by ABA. Promotion of stomatal closure by ABA for 2 h in the presence of wild type of Xcc or rpf mutants. The dataset marked with an asterisk is significantly different from control (ABA) as assessed by Student's t test: *, P < 0.001. The mean and se of three independent experiments are given.

Figure 3.

Ethyl acetate extracts of Xcc strain culture supernatants affect stomata in a similar way as living bacteria. Promotion of stomatal closure by bacteria for 1 h (A) or ABA for 2 h (B) in the presence of extracts from Xcc strains. C, Promotion of stomatal closure in V. fava by ABA for 2 h in the presence of an extract from Xcc. D, Inhibition of stomatal opening by ABA in Arabidopsis for 2 h in the presence of an extract from Xcc. Datasets marked with an asterisk are significantly different from controls (no extract [A]; ABA [B and C]) as assessed by Student's t test: *, P < 0.001. The mean and se of three (A and B) or two (C and D) independent experiments are given.

Figure 4.

The ability of Xcc to migrate through stomata in isolated epidermis is dependent on a functional rpfF/DSF system. Confocal images of the inner side of detached Arabidopsis epidermis were recorded after floating them for 3 h, with the cuticle side in contact with bacterial suspensions of Xcc, rpfF, or rpfC in the presence or not of extracts of supernatants from Xcc, rpfF, or rpfC cultures. An extract from Xcc, but not from rpfF or rpfC, restored the ability of rpfF and rpfC to migrate through the epidermis. An extract from rpfC, which contains a high concentration of DSF, reduced the ability of Xcc to migrate through stomata. The experiment was repeated with similar results.

Figure 5.

Ethyl acetate extracts of Xcc strains culture supernatants can complement coronatine deficiency in Pst. Arabidopsis plants were infected by dipping with wild-type Pst DC3000 or coronatine-deficient Pst DC3118 coronatine-deficient mutant strains in the presence of wild-type Xcc or rpf mutant extracts. Bacterial growth was measured after 4 d. The datasets marked with an asterisk are significantly different from controls (Pst cor⁻) as assessed by Student's t test: *, P < 0.001. The mean and se of two independent experiments are given.

Figure 6.

Guard-cell-specific MPK3 antisense mutants are impaired in their response to bacteria, LPS, and stomatal inhibiting factor from Xcc. Promotion of stomatal closure in epidermal peels from Col-0 and two MPK3 antisense lines (AS3a and AS3c) for 1 h by bacteria (A) or by LPS (B). MPK3 antisense mutants are insensitive to the inhibitory effect of Xcc (C) or of an Xcc extract (D) on ABA-induced promotion of closure. Datasets marked with an asterisk are significantly different from controls (Col-0 controls in A, B, C, and D) as assessed by Student's t test: *, P < 0.001. The mean and se of two independent experiments are given.

Figure 7.

Inhibition of MPK3 expression in guard cells eliminates the requirement of coronatine for Pst penetration through stomata. Bacterial growth in wild-type Col-0 and MPK3 antisense mutant (AS3c) plants 4 d after infection by dipping with wild-type Pst DC3000 or the Pst DC3118 cor⁻ mutant. The datasets marked with asterisks are significantly different from controls (Pst cor⁻ infiltrated in wild-type Col-0) as assessed by Student's t test: *, P < 0.001.The mean and se of two independent experiments are given.

Figure 8.

Proposed model of the mode of inhibition of stomatal closure by Xcc. The DSF (diamonds), which mediates cell-to-cell communication, interacts with the cell surface receptor RpfC, which in turn triggers signaling that leads to transcription of genes required for the synthesis of the stomatal inhibitory compound (triangles). This molecule diffuses to the extracellular space and gets contact with stomatal guard cells, where it blocks stomatal closure induced by ABA or PAMPs by interfering with some signaling component within a putative signaling pathway branch including H₂O₂ and MPK3. MPK3 is absolutely required for bacterial or PAMP-induced promotion of closure pathway, as shown by the failure of MPK3 antisense plants to close stomata in response to these stimuli. For ABA, instead, other pathways, including ABA-induced cytosolic pH increase, would compensate for the absence of MPK3. The failure of the stomatal inhibitor to interfere with promotion of closure by ABA in MPK3 antisense plants could be explained by compensatory adjustments within the guard-cell-signaling network, indicated by thick arrows, that would make up for the absence of MPK3. WT, Wild type; LRR, Leu-rich repeat receptor.