Effects of feeding Spodoptera littoralis on lima bean leaves. II. Continuous mechanical wounding resembling insect feeding is sufficient to elicit herbivory-related volatile emission

Abstract

Herbivore feeding elicits defense responses in infested plants, including the emission of volatile organic compounds that can serve as indirect defense signals. Until now, the contribution of plant tissue wounding during the feeding process in the elicitation of defense responses has not been clear. For example, in lima bean (Phaseolus lunatus), the composition of the volatiles induced by both the insect caterpillar Spodoptera littoralis and the snail Cepaea hortensis is very similar. Thus, a mechanical caterpillar, MecWorm, has been designed and used in this study, which very closely resembles the herbivore-caused tissue damage in terms of similar physical appearance and long-lasting wounding period on defined leaf areas. This mode of treatment was sufficient to induce the emission of a volatile organic compound blend qualitatively similar to that as known from real herbivore feeding, although there were significant quantitative differences for a number of compounds. Moreover, both the duration and the area that has been mechanically damaged contribute to the induction of the whole volatile response. Based on those two parameters, time and area, which can replace each other to some extent, a damage level can be defined. That damage level exhibits a close linear relationship with the accumulation of fatty acid-derived volatiles and monoterpenes, while other terpenoid volatiles and methyl salicylate respond in a nonlinear manner. The results strongly suggest that the impact of mechanical wounding on the induction of defense responses during herbivore feeding was until now underestimated. Controlled and reproducible mechanical damage that strongly resembles the insect's feeding process represents a valuable tool for analyzing the role of the various signals involved in the induction of plant defense reactions against herbivory.

Figure 1.

GC profiles of induced VOCs emitted from lima bean leaves upon feeding of different herbivores. Headspace volatiles were collected for 24 h during the whole treatment. A, Feeding by S. littoralis larvae. B, Control without any feeding. C, Feeding by C. hortensis. Identification of compounds was performed by MS: (1) octen-3-ol; (2) Hex-Ac; (3) β-ocimene; (4) linalool; (5) DMNT; (6) C₁₀H₁₄; (7) MeSA; (8) C₁₀H₁₆O; (9) indole; (10) β-caryophyllene; (11) nerolidol; (12) TMTT. IS, Internal standard (1-bromodecane, 0.9 mm). All experiments have been conducted at least six times with similar results.

Figure 2.

MecWorm, the mechanical caterpillar. A, Design of MecWorm. I, Punching unit; II, plexiglass cabinet; III, volatile collection unit. B, Section of the punching unit of MecWorm, including the aglet treating a lima bean leaf. C, Lima bean leaf damaged by MecWorm.

Figure 3.

GC profile of induced VOCs emitted from lima bean leaves upon mechanical damage. A, Treatment with MecWorm. Headspace volatiles were collected for 24 h during the whole treatment (time = 17 h, 7 min; area = 733 mm²). B, Treatment with a pattern wheel. C, Collection of volatiles emitted from the nontreated opposite leaf. Identification of compounds was performed by MS (see Fig. 1). Experiments have been repeated nine (A), four (B), and eight times, respectively, with similar results.

Figure 4.

Relative amounts of various emitted VOCs in relation to time and extent of mechanical treatments in lima bean leaves. Volatiles have been induced by either mechanical damage of a fixed area at various or similar lengths of time combined with various areas. The relative increase or decrease in the amount of volatiles was determined by the ratio of the peak area of the particular compound (A_VOC) to the peak area of the internal standard (A_IS). Statistical evaluations were done with SPSS (version 11.5). Different letters indicate significant differences between treatments (P < 0.05) according to one-way ANOVA on rank-transformed data. The Student-Newman-Keuls test was used to determine post hoc differences. Sample sizes are n = 5 (2.9 h, 313 mm²), n = 6 (16 h, 313 mm²), and n = 10 (17 h, 733 mm²).

Figure 5.

Correlation of the relative amounts of volatiles with the damage levels of mechanical wounding in lima bean leaves. The relative amount of the volatiles was determined by the ratio of the peak area of the particular compound (A_VOC) to the peak area of the internal standard (A_IS). The damage levels have been determined with 0 (no treatment), 0.33 (2.9 h, 313 mm²), 1.80 (6.85 h, 733 mm²; 16 h, 313 mm²), and 4.52 (17 h, 733 mm²), respectively. A, Octen-3-ol (○) and Hex-Ac (•). B, Ocimene (▪), C₁₀H₁₆O (•), linalool (○), and C₁₀H₁₄ (▴). C, DMNT (▴), MeSA (•), and TMTT (▪). MeSA and TMTT did not show linear correlations. The results are the mean ± sd (n = 5–10).

Figure 6.

Scanning electron micrographs of lima bean leaf damage zones after feeding of S. littoralis larvae (A and B) and after treatment with the mechanical caterpillar (C and D).