Canopy light cues affect emission of constitutive and methyl jasmonate-induced volatile organic compounds in Arabidopsis thaliana

Abstract

The effects of plant competition for light on the emission of plant volatile organic compounds (VOCs) were studied by investigating how different light qualities that occur in dense vegetation affect the emission of constitutive and methyl-jasmonate-induced VOCs. Arabidopsis thaliana Columbia (Col-0) plants and Pieris brassicae caterpillars were used as a biological system to study the effects of light quality manipulations on VOC emissions and attraction of herbivores. VOCs were analysed using gas chromatography-mass spectrometry and the effects of light quality, notably the red : far red light ratio (R : FR), on expression of genes associated with VOC production were studied using reverse transcriptase-quantitative PCR. The emissions of both constitutive and methyl-jasmonate-induced green leaf volatiles and terpenoids were partially suppressed under low R : FR and severe shading conditions. Accordingly, the VOC-based preference of neonates of the specialist lepidopteran herbivore P. brassicae was significantly affected by the R : FR ratio. We conclude that VOC-mediated interactions among plants and between plants and organisms at higher trophic levels probably depend on light alterations caused by nearby vegetation. Studies on plant-plant and plant-insect interactions through VOCs should take into account the light quality within dense stands when extrapolating to natural and agricultural field conditions.

Keywords: Arabidopsis thaliana; herbivory; light quality; phytochrome; shade avoidance; volatile organic compounds (VOCs).

Fig 1

Graphic representation of the Teflon plate used to hold plants for volatile collection. (a) Top view of the plate (inner diameter 131 mm; outer diameter 183 mm); green spots indicate the holes that were used for individual plants. (b) Top view of the plant canopy. (c) Side view of the plate (plate thickness 3 mm) placed on top of the plant pots, with plants growing through it. (d) Side view of the plant canopy.

Fig 2

Projection to latent structures–discriminant analysis (PLS-DA) of volatile emissions produced by Arabidopsis thaliana plants in four different treatments: control plants (C), plants exposed to low red (R) : far-red (FR) (R), plants under a green filter (GF) and plants that were growing in an actual canopy (CN). A score plot (a) and loading plot (b) of the first two principal components, with the explained variance in brackets, are shown. The ellipse defines the Hotelling's T2 confidence region (95%). Numbers in the loading plot (b) correspond to numbers representing individual volatile compounds as presented in Table 1.

Fig 3

Emission of different volatile organic compounds (VOCs) by undamaged Arabidopsis thaliana plants, in terms of peak area units per gram plant dry weight detected in gas chromatography–mass spectrometry (GC-MS) analysis. Bars represent mean ± SE (n = 5). Volatiles were detected in the headspace of undamaged A. thaliana plants collected for 4 h. Volatiles were collected from control plants (C), plants exposed to low red (R) : far-red (FR) (R), plants under a green filter (GF) and plants that were growing in an actual canopy (CN). Significant differences are represented by different letters (P < 0.05; one-way ANOVA; Bonferroni post hoc).

Fig 4

Ethylene emission of control, low red (R) : far-red (FR) and green filter treated Arabidopsis thaliana plants after 4 h of treatment. Bars represent mean ± SE (n = 7). Different letters indicate significant differences (one-way ANOVA; Bonferroni post hoc).

Fig 5

Peak area units of different volatile organic compounds (VOCs) emitted by Arabidopsis thaliana, corrected for plant dry weight. Volatiles were collected during 4 h from control plants (Control), plants exposed to low red (R) : far-red (FR) (R), plants sprayed with 100 μM methyl jasmonate (MJ) and plants exposed to low R : FR and treated with 100 μM methyl jasmonate (R+MJ). The text at the top left of each graph shows the main effects and interaction effect based on two-way ANOVA: * indicates a significant difference (P < 0.05); + indicates a trend (0.10 > P > 0.05); NS, not significant. Bars represent mean (n = 5) ± SE. Significantly different classes are indicated by different letters (one-way ANOVA; Bonferroni post hoc).

Fig 6

Low red (R) : far-red (FR)- and methyl jasmonate (MJ)-induced emission of ethylene from the shoot of Arabidopsis thaliana plants. Volatiles were collected from control plants (Control), plants exposed for 6 d to low R : FR (R), plants sprayed 24 h before ethylene measurements with 100 μM MJ (MJ) and plants exposed for 6 d to low R : FR and sprayed 24 h before ethylene measurements with 100 μM MJ (R+MJ). Data represent mean (n = 5) ± SE. The text at the top left of each graph shows the outcome of two-way ANOVA testing: * indicates a significant difference (P < 0.05); NS, not significant. Significantly different classes are indicated by different letters (one-way ANOVA; Bonferroni post hoc).

Fig 7

Relative expression of VEGETATIVE STORAGE PROTEIN 2 (VSP2), S-adenosylmethionine-dependent methyltransferase (BSMT1), TERPENE SYNTHASE 4 (TPS4), TPS3 and CYTOCHROME P450 72A13 (CYP72A13) in Arabidopsis thaliana. Control plants were grown under normal light conditions (red (R) : far-red (FR) 2.0), methyl jasmonate (MeJA) plants were sprayed with 100 μM MeJA 24 h before harvesting for gene expression (MJ), and double-treated plants were grown for 6 d at R : FR 0.2 and sprayed with 100 μM MeJA 24 h before harvesting for gene expression (R+MJ) (n = 5). Data represent the mean of five independent biological replicates ± SE. Different letters indicate significant differences (one-way ANOVA; Bonferroni post hoc test).

Fig 8

Preference of Pieris brassicae in two-choice tests. For each test, eight different Arabidopsis thaliana pairs were used. Plants grown at a red (R) : far-red (FR) ratio of 2.0 (Control), plants grown at an R : FR ratio of 0.2 (Low R : FR), plants sprayed with 100 μM methyl jasmonate (MeJA) and plants grown at an R : FR ratio of 0.2 and sprayed with 100 μM MeJA (Low R : FR MeJA) were tested. Bars represent the percentage of herbivores that preferred the treatment. Numbers in bars represent numbers of herbivores that made the corresponding choice. Data were analysed with a binominal test. Significant differences: *, P < 0.05; ***, P < 0.001; NS, no significant differences were found for the corresponding choice assays.