Min-Soo Choi (Seoul National University)

Youngsung Joo (Seoul National University)

Min-Soo Choi (Seoul National University)

Myungju Go (Seoul National University)

Hyeok Gyun Son (Seoul National University)

Gwanhyeong Yu (Seoul National University)

Jong-Hoon Noh (Seoul National University)

Yoonkyoung Choi (Seoul National University)

Youngsung Joo (Seoul National University)

Plants are under attack of multiple species of herbivore and defending them requires a diverse array of specialized metabolites. While ecological theories predict adaptive plant responses against multiple herbivores, the predictions have been rarely tested in metabolome-wide manner. This study employed untargeted metabolomics via liquid chromatography-tandem mass spectrometry (LC-MS/MS) to comprehensively investigate plant defense strategies in Arabidopsis thaliana exposed to four herbivore species (Spodoptera litura, Plutella xylostella, Myzus persicae, and Lipaphis pseudobrassicae). The raw features were extracted by MZmine 4 and annotated using feature-based molecular networking and SIRIUS. We observed species-specific and intensity-dependent reconfigurations of the plant metabolome. To identify ecologically relevant metabolites, we analyzed correlations between metabolite levels and herbivore performance. Correlating metabolite levels with herbivore performance revealed that induced metabolites suppressed three herbivore species but not P. xylostella. We further computed the chemical diversity of plants attacked by insect herbivores using DreaMS embedding-based similarity. The chemical diversity was increased by phloem sap feeding herbivores while decreased by chewing herbivores. These results demonstrate the complexity of plant-herbivore interactions and highlight the importance of untargeted metabolomics for understanding plant defense strategies.

​

​