Volatile Semiochemicals Emitted by Beauveria bassiana Modulate Larval Feeding Behavior and Food Choice Preference in Spodoptera frugiperda (Lepidoptera: Noctuidae)

Abstract

Beauveria bassiana is an entomopathogenic fungus that parasitizes and kills insects. The role of volatile organic compounds (VOCs) emitted by B. bassiana acting as semiochemicals during its interaction with lepidopterans is poorly explored. Here, we studied the effect of VOCs from B. bassiana and 3-methylbutanol (as a single compound) on the feeding behavior of L2 larvae of Spodoptera frugiperda in sorghum plants. Additionally, we assessed whether fungal VOCs induce chemical modifications in the plants that affect larval food preferences. Metabolomic profiling of plant tissues was performed by mass spectrometry and bioassays in a dual-choice olfactometer. The results showed that the larval feeding behavior was affected by the B. bassiana strain AI2, showing that the insect response is strain-specific. Furthermore, 80 µg of 3-methylbutanol affected the number of bites. The larval feeding choice was dependent on the background context. Fragment spectra and a matching precursor ion mass of 165.882 m/z enabled the putative identification of 4-coumaric acid in sorghum leaves exposed to fungal VOCs, which may be associated with larval deterrent responses. These results provide valuable insights into the bipartite interaction of B. bassiana with lepidopterans through VOC emission, with the plant as a mediator of the interaction.

Keywords: UPLC-MS; fungal volatiles; fusel alcohols; insect behavior; olfactometry; semiochemicals.

Figure 1

Experimental strategy used to investigate the effect of volatiles emitted by Beauveria bassiana on the larval feeding behavior (A) and (B) food choice preference in Spodoptera frugiperda.

Figure 2

Spodoptera frugiperda larval feeding behavior in response to volatiles emitted by the AI2 (A–C) and AS5 (D–F) strains of Beauveria bassiana. Panels (A,D) correspond to the feeding area (mm²) (removed tissue after 12 h of consumption). Panels (B,E) show the percentage of herbivory (missing leaf area with respect to the total leaf area). Panels (C,F) correspond to the number of bites per leaf inflicted by larvae. Control refers to treatments with uninoculated PDA culture medium. n = 114–183, glm with gamma distribution, α = 0.05.

Figure 3

Larval dual-choice olfactometry bioassays in response to pairs of odor sources. (A) The bioassays utilized mycelium from the strains AI2 and AS5 cultured in PDA culture medium. n = 40–63. (B) Bioassays were performed on sorghum foliage unexposed and exposed to VOCs from the strains AI2 and AS5. n = 30–46. The bars indicate the frequency of insects that selected either odor source. Pie charts show proportions of responding (black) and not responding (white) individuals. Statistical significance according to Chi-square two-sides test (* p < 0.05, *** p < 0.001) at α = 0.05.

Figure 4

Larval feeding behavior of L2 larvae of Spodoptera frugiperda on Sorghum bicolor foliage exposed to 8 and 80 µg of 3-methylbutanol. (A) Feeding area (mm²) (removed tissue after 12 h of consumption). (B) Percentage of herbivory (missing leaf area with respect to the total leaf area). (C) Number of bites per leaf inflicted by larva. n = 114–183, analyzed with glm with gamma distribution, α = 0.05. Different letters indicate statiscally significant differences, ns means not statiscally significant. (D) Olfactometry dual-choice bioassay of L2 larvae exposed to 8 μg of 3-methylbutanol compared to water and artificial diet. The bars indicate the frequency of insects that selected either odor source. Pie charts show proportions of responding (black) and not responding (white) individuals. Statistical significance according to Chi-square two-sides test (*** p < 0.001) at α = 0.05.

Figure 5

(A) Representative chromatogram of the chemical composition of sorghum foliage exposed to VOCs emitted by B. bassiana. Insets correspond to the abundance of the ion signals 165.882, 353.347, and 437.402 m/z. Green, red, blue, and black colors indicate VOCs AS5, VOCs AI2, 3-methylbutanol, and control treatments, respectively. (B) Heatmap of the normalized signal intensities of the 116 ions obtained from the metabolic profiles. Hierarchical clustering with Euclidean distance.

Figure 6

(A) Biomarker ions obtained by random forest model for differentiating between the Sorghum bicolor foliage treatments, including exposure to VOCs AI2, VOCs AS5, and 8 µg of 3-methylbutanol. Random forest with 500 trees. (B) Principal component analysis from the chemical profiles of sorghum plants differentially treated with the fungal VOCs. PERMANOVA test, α = 0.05. (C) Mean intensities from the ion signals with the highest Gini index, n = 6. Different letters indicate statiscally significant differences by one-way ANOVA, Tukey post hoc, α = 0.05.