A permeable cuticle is associated with the release of reactive oxygen species and induction of innate immunity

Abstract

Wounded leaves of Arabidopsis thaliana show transient immunity to Botrytis cinerea, the causal agent of grey mould. Using a fluorescent probe, histological staining and a luminol assay, we now show that reactive oxygen species (ROS), including H(2)O(2) and O(2) (-), are produced within minutes after wounding. ROS are formed in the absence of the enzymes Atrboh D and F and can be prevented by diphenylene iodonium (DPI) or catalase. H(2)O(2) was shown to protect plants upon exogenous application. ROS accumulation and resistance to B. cinerea were abolished when wounded leaves were incubated under dry conditions, an effect that was found to depend on abscisic acid (ABA). Accordingly, ABA biosynthesis mutants (aba2 and aba3) were still fully resistant under dry conditions even without wounding. Under dry conditions, wounded plants contained higher ABA levels and displayed enhanced expression of ABA-dependent and ABA-reporter genes. Mutants impaired in cutin synthesis such as bdg and lacs2.3 are already known to display a high level of resistance to B. cinerea and were found to produce ROS even when leaves were not wounded. An increased permeability of the cuticle and enhanced ROS production were detected in aba2 and aba3 mutants as described for bdg and lacs2.3. Moreover, leaf surfaces treated with cutinase produced ROS and became more protected to B. cinerea. Thus, increased permeability of the cuticle is strongly linked with ROS formation and resistance to B. cinerea. The amount of oxalic acid, an inhibitor of ROS secreted by B. cinerea could be reduced using plants over expressing a fungal oxalate decarboxylase of Trametes versicolor. Infection of such plants resulted in a faster ROS accumulation and resistance to B. cinerea than that observed in untransformed controls, demonstrating the importance of fungal suppression of ROS formation by oxalic acid. Thus, changes in the diffusive properties of the cuticle are linked with the induction ROS and attending innate defenses.

Figure 1. A: ROS production in response to wounding in leaves of A. thaliana.

(A) Fluorescence after DCF-DA staining was measured on leaf discs using fluorescence spectrophotometry. The unwounded and wounded leaf discs were infiltrated with the DCF-DA probe and the fluorescence was directly measured at intervals of 15 min during 120 min (n = 12; ±SD). The experiment was carried out 3 times with similar results. (B) Time-course of ROS production observed as DCF-DA fluorescence by fluorescence microscopy in unwounded or wounded B. cinerea (Bc)-inoculated leaves compared to mock controls. After treatment, all plants were kept under humid conditions. The experiment was carried out 5 times with similar results. (C) H₂O₂ formation observed as DAB staining and superoxide (O₂ ⁻) formation observed as NBT staining in unwounded or wounded leaves. Plants were stained immediately after wounding. The experiment was carried out 3 times with similar results. (D) Effect of DPI on ROS formation (measured as DCF-DA fluorescence) and growth of B. cinerea (determined by Trypan blue staining) in unwounded or wounded leaves. Leaves were treated for 24 h with DPI, then rinsed and either wounded, immediately stained and examined for fluorescence or inoculated with B. cinerea (Bc) and incubated in humid condition (symptoms were observed at 3 dpi). (E) The effect of catalase on ROS formation was tested by infiltration of catalase (1100 U ml⁻¹) immediately prior to wounding. Plants were stained with DAB immediately after wounding. The experiments in Fig 1D and E were carried out 5 times with similar results. (F) Production of H₂O₂ in unwounded control (Ctrl) or wounded (W) leaf discs detected by the chemiluminescence reaction with luminol immediately after wounding. The experiment was carried out 3 times with similar results. (G) Production of ROS in unwounded and wounded WT plants and in nia1nia2noa1-2, a triple mutant deficient in nitrate reductase (NIA/NR)- and Nitric Oxide-Associated1 (AtNOA1)-mediated NO biosynthetic pathways. ROS were detected as DCF-DA fluorescence immediately after wounding. The experiment was carried out 3 times with similar results.

Figure 2. Subcellular localization of ROS at wounded sites in A. thaliana leaves.

Leaves were infiltrated with DCF-DA, then wounded and ROS accumulation was observed by laser confocal microscopy immediately after wounding. The experiment was carried out 3 times with similar results.

Figure 3. The effect of humidity on resistance to B. cinerea, ROS and callose accumulation after wounding.

Wild type (WT) leaves were wounded and maintained for 1.5 h under high humidity in tightly covered well-watered trays (humid) or in uncovered trays at room conditions (dry) prior to ROS or infection with B. cinerea or callose detection. (A) Densitometric quantification of ROS production, (B) resistance to B. cinerea. W: wounded; Ctrl: unwounded control plants. (C) Callose formation. M: mock; Bc: B. cinerea-inoculated. For ROS production (n = 4; ±SD), one representative image of the fluorescent leaf surface was placed above each histogram as a visual illustration. For resistance (n = 48; ±SE) and callose formation (n = 4; ±SD), all plants were kept under humid conditions after treatment and the experiment was carried out twice with similar results. Different letters above each bar represent statistically significant differences (Dunn's test; P<0.05).

Figure 4. Wounding in mutants of ABA biosynthesis.

Leaves were wounded and maintained for 1.5 h under high humidity in tightly covered well-watered trays (humid) or in uncovered trays at room conditions (dry) prior to ROS detection or infection with B. cinerea. W: wounded; Ctrl: unwounded control plants. (A) Densitometric quantification of ROS production in unwounded and wounded aba2, aba3 mutants and in WT plants. For ROS production (n = 4; ±SD), one representative image of the fluorescent leaf surface was placed above each histogram as a visual illustration. Different letters above each bar represent statistically significant differences (Dunn's test; P<0.05). (B) Effects of wounding on resistance to B. cinerea in aba2 and aba3 mutants and WT plants. After B. cinerea inoculation, all plants were kept under humid conditions (n = 32; ±SE); the experiment was repeated twice with similar results. Different letters above each bar represent statistically significant differences (Dunn's test; P<0.05).

Figure 5. Densitometric quantification of ROS production in ABA and cuticle mutants.

ROS production at 3, 6, 12 h post inoculation with B. cinerea (Bc), mock or H₂O treatments in aba2 and aba3 as well as in bdg and lacs2.3 mutants compared to the WT. After treatment, all plants were kept under humid conditions. Low level of fluorescence density in the WT were detected only after B. cinerea at 12 h post inoculation and were not detected in response to H₂O or mock treatment. For ROS production (n = 4; ±SD), one representative image of the fluorescent leaf surface was placed above each histogram as a visual illustration. Different letters above each bar represent statistically significant differences (Dunn's test; P<0.05).

Figure 6. ABA accumulates after wounding under dry conditions and affects resistance to B. cinerea.

Leaves were wounded and maintained for 1.5 h under high humidity in tightly covered well-watered trays (humid) or in uncovered trays at room conditions (dry) prior to measurement of ABA or luciferase activity. (A) Measurement of ABA in ng mg⁻¹ fresh weight of plant tissue in unwounded or wounded plants, incubated under humid or dry conditions (n = 6; ±SD). Different letters above each bar represent statistically significant differences (Dunn's test; P<0.05). (B) Expression of pAtLTI23T::LUC or pAtHB6T::LUC in wounded leaves incubated either under humid or dry conditions compared to unwounded plants. The wounds inflicted by the forceps (arrows) show a stronger expression of the LUC gene in plants incubated under dry conditions. The experiment was repeated 3 times, one typical result is represented. (C) Effect of exogenous ABA treatment (leaf discs were floated on 100 mM ABA, 1 d prior to wounding). After wounding, ROS (measured as DCF-DA fluorescence) and growth of B. cinerea (Bc) (leaf discs were floated on water and inoculated; Trypan blue staining was carried out 2 d after inoculation) were determined (the experiment was carried out 3 times, one typical result is represented).

Figure 7. Cuticle permeability is impaired in ABA mutants and after wounding under humid conditions.

Permeability of the cuticle in ABA and cuticle mutants (positive controls) or WT plants (A) and (B). (A) Upper panels: a droplet of toluidine blue was placed on the leaf surface for 2 h in high humidity then the leaf surface was rinsed with water. The blue stain that remains attached to the cell wall is indicative of a permeable cuticle. Lower panels: leaves were bleached overnight in ethanol then stained with Calcofluor white that binds to cellulose, and viewed under UV light. Calcofluor staining to the leaf is indicative of a permeabilized cuticle (all experiments were carried out 12 times, one typical result is represented). (B) Leaves were placed in ethanol and the release of chlorophyll was followed over time. Chlorophyll leached out more rapidly in all mutants compared to WT indicating a higher cuticle permeability (n = 6; ±SD). Permeability of the cuticle in unwounded and wounded plants incubated under humid or dry conditions (C) and (D). (C) Leaves were wounded and maintained for 1.5 h under high humidity in tightly covered well-watered trays (humid) or in uncovered trays at room conditions (dry). The permeability of the cuticle was assessed using Calcofluor white. Wounding (arrow) followed by incubation under humid conditions lead to white staining visible at the wound sites and to a lesser extent in other parts of the leaf. Wounding followed by incubation in dry conditions showed no staining (the experiment was carried out 12 times, one typical result is represented). (D) Chlorophyll leached out more rapidly in plants incubated under humid conditions compared to plants incubated under dry conditions (n = 5; ±SD).

Figure 8. Expression of genes involved in cuticle formation in wounded and unwounded plants.

Leaves were wounded and maintained for 1.5 h under high humidity in tightly covered well-watered trays (humid) or in uncovered trays at room conditions (dry) prior to expression of BDG and LACS2.3 genes. Gene expression was determined 0, 15 or 30 min after wounding in plants incubated under humid or dry conditions and either mock-inoculated (M) or inoculated with B. cinerea (Bc) (n = 3; ±SD). The experiment was carried out twice with similar results. The expression levels of unwounded control plants behaved similarly under dry or humid conditions. Different letters above each bar represent statistically significant differences (Dunn's test; P<0.05).

Figure 9. ROS production and resistance to B. cinerea in response to exogenous treatments with cutinase.

A droplet containing cutinase (5 mg l⁻¹, a non-toxic concentration; see [47]) was applied to the surface of an A. thaliana leaf. After 72 h in high humidity, the drop was removed, and (A) the leaf was stained with DAB or DCF-DA (insert) to detect ROS production or (B) replaced by B. cinerea spores (6 µL; 5×10⁴ spores ml⁻¹) applied at the same location. After 72 h, lesion sizes were determined (n = 27; ±SE). The experiment was carried out twice with similar results. Different letters above each bar represent statistically significant differences (Dunn's test; P<0.05).

Figure 10. Effect of OXALATE DECRABOXYLASE over expression in A. thaliana on resistance to B. cinerea and ROS production.

(A) Resistance displayed by T3 A. thaliana lines over expressing the OXALATE DECARBOXLYLASE gene from T. versicolor (oxdec plants); all plants were previously checked for the over expression of the oxalate decarboxylase activity (n = 130; ±SE). Different letters above each bar represent statistically significant differences (Dunn's test; P<0.05). (B) Densitometric quantification of ROS production at 3, 6, 12 h post inoculation with B. cinerea (Bc) and mock in oxdec plants compared to the WT. After treatment, all plants were kept under humid conditions. Low level of fluorescence density in the WT is detected only after B. cinerea at 12 h post inoculation and not in the mock treatment. For ROS production (n = 4; ±SD), one representative image of the fluorescent leaf surface was placed above each histogram as a visual illustration. Different letters above each bar represent statistically significant differences (Dunn's test; P<0.05).

Figure 11. Model illustrating the role of the cuticle at the interface between B. cinerea and A. thaliana.

Under the action of digestive enzymes, permeabilization of the cuticle increases, allowing for early sensing and perception of elicitors (MAMPs/DAMPs) with subsequent induction of ROS and potential activation of innate immune responses. The virulent pathogen produces effector(s)/suppressor(s) (e.g. oxalic acid) that interfere with ROS build-up leading to decreased defenses and allowing colonization.