Reactive oxygen species are involved in plant defense against a gall midge

Abstract

Reactive oxygen species (ROS) play a major role in plant defense against pathogens, but evidence for their role in defense against insects is still preliminary and inconsistent. In this study, we examined the potential role of ROS in defense of wheat (Triticum aestivum) and rice (Oryza sativa) against Hessian fly (Mayetiola destructor) larvae. Rapid and prolonged accumulation of hydrogen peroxide (H(2)O(2)) was detected in wheat plants at the attack site during incompatible interactions. Increased accumulation of both H(2)O(2) and superoxide was detected in rice plants during nonhost interactions with the larvae. No increase in accumulation of either H(2)O(2) or superoxide was observed in wheat plants during compatible interactions. A global analysis revealed changes in the abundances of 250 wheat transcripts and 320 rice transcripts encoding proteins potentially involved in ROS homeostasis. A large number of transcripts encoded class III peroxidases that increased in abundance during both incompatible and nonhost interactions, whereas the levels of these transcripts decreased in susceptible wheat during compatible interactions. The higher levels of class III peroxidase transcripts were associated with elevated enzymatic activity of peroxidases at the attack site in plants during incompatible and nonhost interactions. Overall, our data indicate that class III peroxidases may play a role in ROS generation in resistant wheat and nonhost rice plants during response to Hessian fly attacks.

Figure 1.

ROS accumulation in apoplasm. A, Representative fluorescence images of leaf sheaths at the larval attack sites stained for the presence of H₂O₂. Images were derived from nonattacked plants (C) and plants at 12, 24, 48, and 72 h after the initial attack by biotype L Hessian fly larvae. The dimensions of the H₂O₂-containing zones varied due to variations in larval densities and differences in larval migration. Each leaf sheath hosted 16 to 20 larvae. Relative fluorescence units are given below each image. Newton is a susceptible line containing no resistance gene. Iris contains R gene H9. Molly contains R gene H13. Rice is a nonhost. B, Average relative fluorescence obtained from images (similar to those shown in A) stained for H₂O₂ content. Each average value was derived from three biological replicates, with each replicate including readings from four lesions. C, Average relative fluorescence obtained from images stained for O₂⁻ content. The same numbers of leaf sheaths as in B were used.

Figure 2.

ROS accumulation in cytosol. A, Representative fluorescence images of leaf sheath cells at the larval attack site stained for H₂O₂. ROS were assayed with tissues from control (C) and attacked (I) plants at 48 h (see “Materials and Methods”). WN, WI (H9), and WM (H13) represent wheat Newton, wheat Iris (containing H9), and wheat Molly (containing H13), respectively. Light stripes are cell walls between cells. The images were obtained from a fluorescent microscope with amplifications 10 × 63. B, Average values of three biological replicates from control plants (gray bars) and plants attacked by Hessian fly (black bars). The letters a and b represent two groups that are statistically different (P = 0.05).

Figure 3.

Toxicity of H₂O₂ to Drosophila larvae. The solid line represents the percentage of larval mortality at increasing concentrations of H₂O₂ on artificial diet. The dashed lines above and below the solid line represent upper and lower bounds of the 95% confidence interval (CI). LD50 is less than 0.5 μg mL⁻¹ or 1.7 μm. [See online article for color version of this figure.]

Figure 4.

qRT-PCR validation of microarray data. A and B, Spearman's rank correlation between qRT-PCR and the corresponding microarray data for wheat transcripts (A) and rice transcripts (B). C, qRT-PCR quantification of the wheat class III peroxidase transcript CK198851. Graphs show fold change values for attacked resistant (gray bars) and susceptible (white bars) wheat plants at 12, 24, 48, 72, and 120 h post egg hatch when compared with nonattacked wheat plants at the same time point. D, qRT-PCR quantification of the wheat transcript CD373657 displayed like the data in C. E, qRT-PCR quantification of two rice class III peroxidase genes, Os06g0547400 (gray bars) and Os07g0677200 (white bars), showing fold change values for attacked rice plants at 12, 24, 48, 72, and 120 h after egg hatch compared with nonattacked control plants at the 12-h time point. F, qRT-PCR quantification of a rice NADPH-dependent oxidase gene (Os01g0734200; gray bars) and a thioredoxin gene (Os04g0676100; white bars) displayed like the data in C.

Figure 5.

Changes in peroxidase activity at the attack site. A, Representative fluorescence images of leaf sheaths stained for peroxidase enzymatic activity. B, Average relative fluorescence obtained from three biological replicates. Materials and procedures were as described in Figure 1.