Berberine bridge enzyme-like oxidases orchestrate homeostasis and signaling of oligogalacturonides in defense and upon mechanical damage

Abstract

Plant immunity is triggered by endogenous elicitors known as damage-associated molecular patterns (DAMPs). Oligogalacturonides (OGs) are DAMPs released from the cell wall (CW) demethylated homogalacturonan during microbial colonization, mechanical or pest-provoked mechanical damage, and physiological CW remodeling. Berberine bridge enzyme-like (BBE-l) proteins named OG oxidases (OGOXs) oxidize and inactivate OGs to avoid deleterious growth-affecting hyper-immunity and possible cell death. Using OGOX1 over-expressing lines and ogox1/2 double mutants, we show that these enzymes determine the levels of active OGs vs. inactive oxidized products (ox-OGs). The ogox1/2-deficient plants have elevated levels of OGs, while plants overexpressing OGOX1 accumulate ox-OGs. The balance between OGs and ox-OGs affects disease resistance against Pseudomonas syringae pv. tomato, Pectobacterium carotovorum, and Botrytis cinerea depending on the microbial capacity to respond to OGs and metabolize ox-OGs. Gene expression upon plant infiltration with OGs reveals that OGOXs orchestrate OG signaling in defense as well as upon mechanical damage, pointing to these enzymes as apoplastic players in immunity and tissue repair.

Keywords: Damage‐Associated Molecular Patterns; berberine bridge enzyme‐like oxidases; oligogalacturonides; plant immunity.

Figure 1

Transient expression of OGOX1‐GFP, OGOX2‐GFP, and CELLOX2‐GFP in tobacco epidermal cells. OGOX1‐GFP, OGOX2‐GFP, and CELLOX2‐GFP labeled the cell wall, colocalizing with the cell wall marker PGIP2‐RFP at 48 h post‐infiltration (hpi). Plasmolyzed cells coexpressing OGOX1‐GFP, OGOX2‐GFP, CELLOX2‐GFP, and the plasma membrane marker, pm‐rk, showed the presence of these proteins in the cell wall; the arrows evidence the green fluorescent cell wall and the retracted red fluorescent plasma membrane. Scale bars 20 μm for OGOX1‐GFP, OGOX2‐GFP, CELLOX2‐GFP, and PGIP2‐RFP; 10 μm for pm‐rk (plasmolysis).

Figure 2

OGOX1::GUS transgenic leaves show increased GUS activity in response to pathogens, wounding and elicitor treatment. GUS activity was analyzed in OGOX1::GUS # 3.2 line. Results with a second independent line (#2.3) are in Figure S1. (a) Rosette leaves were drop‐inoculated with B. cinerea conidia (5 μl, ×10⁵ spore ml⁻¹) or PDB (mock) as a control and GUS assay was performed 48 h post‐inoculation. (b) Rosette leaves were infiltrated with P. syringae pv. tomato DC3000 at OD = 0.002 or with H₂O (mock‐infiltrated). GUS assay was performed 72 hpi. (c) Rosette leaves were drop‐inoculated at punctured sites with P. carotovorum cells (5 μl, OD₆₀₀ = 0.025) or 50 mm potassium phosphate buffer pH 7.0 (mock‐inoculated). GUS assay was performed 14 hpi. (d) Untreated excised leaves analyzed at 72 h as a control. (e) Rosette leaves were wounded by crushing with knurled tweezers; GUS assay was performed 1 h after crushing. (f) Rosette leaves were infiltrated with OGs (200 μg ml⁻¹), flg22 (100 nm), and water as control. GUS assay was performed 1‐h post‐infiltration.

Figure 3

HPAEC‐PAD analyses of chelating agent‐extracted oligosaccharides (ChASS) from total cell wall preparations (AIS) of WT, ogox1/2, and OGOX1‐OE leaves. Leaves were infiltrated with (a) OGs (600 μg ml⁻¹) or (b) water, and the fractions containing OGs (ChASS) were analyzed. In the chromatographic profiles (left), OGs and oxidized OGs (ox‐OGs) are indicated by numbers corresponding to the degree of polymerization (DP). Graphs on the right show the sum of peak areas of OGs and ox‐OGs (DP 9–11) as seen in the chromatographic profiles. No oxidized OGs were detected in water‐infiltrated leaves. OG and ox‐OG preparations of DP 9–16 were used as standards. Asterisks indicate statistical significant differences according to the student t‐test (*P < 0.05).

Figure 4

Altered expression of OGOXs affects defense‐related gene expression and callose deposition induced by OGs. (a) Expression was analyzed by quantitative RT‐PCR in rosette leaves from 4‐week‐old plants of Arabidopsis Col‐0 WT, OGOX1‐OE overexpressing plants (lines #1.9 and #11.8), and CRISPR Cas‐deleted ogox1/ogox2 plants (lines #1.5 and #4.6) at 1 h (FRK1, CYP81F2, RBOHD) and 3 h (PAD3) post‐infiltration with OGs (60 μg ml⁻¹) or water as control. UBQ5 transcript levels were used for normalization. The mean of three biological replicates (±SD) is shown. To ensure accurate normalization, UBC9 (At4g27960) was used as a second reference gene for FRK1 and CYP81F2 expression (Figure S9). (b) Callose deposition in WT, ogox1/2, and OGOX1‐OE leaves after infiltration with OGs and H₂O. Callose deposits were stained with aniline blue 24 h after treatment. Each picture is representative of at least 18 pictures acquired per genotype per treatment. Callose deposits were counted using the “Analyze Particles” function of ImageJ. Values indicate fluorescent area/background ± SE. Values for callose deposition of ogox1/2 and OGOX1‐OE lines were significantly different from WT, with a P < 0.05 as determined by Student's t‐test. (c) OG‐induced H₂O₂ accumulation in the culture medium of WT, ogox1/2 and OGOX1‐OE seedlings. Measurements were taken 30 min after treatment with 30 μg ml⁻¹ OGs. Values represent the mean of at least six replicates ± standard error (SE). Asterisks in (a, c) indicate statistically significant differences of mutants compared to WT treated plants according to Student's t‐test (*P < 0.05; **P < 0.01; ***P < 0.001).

Figure 5

Plants altered in OGOX expression show altered response to pathogens. Lesion areas produced by P. carotovorum and B. cinerea were quantified at 16 and 48 hpi, respectively, using the ImageJ software. Pseudomonas syringae DC3000 (P. syringae pv. tomato) spread was quantified at 72 hpi. Asterisks indicate statistically significant differences of mutants compared to WT according to Student's t‐test (*P < 0.05; **P < 0.01; ***P < 0.001).

Figure 6

Hydrogen peroxide production in response to mechanical damage is higher in the ogox1/2 mutants. Leaves of 4‐week‐old plants were damaged with knurled‐tip tweezers, excised after 1 h, and subjected to DAB staining for 4 h. In situ dark‐brown precipitate generated by DAB oxidation in the presence of hydrogen peroxide was quantified per leaf using the “Analyze Particles” function of ImageJ. Values are the mean ± SD of at least three damaged leaves from three different plants. The experiment was repeated three times with similar results; a representative experiment is shown. Asterisks indicate statistically significant differences of mutants compared to WT according to Student's t‐test (*P < 0.05; **P < 0.01).