Overall, the project was a success, and we achieved our main objective. Some details are listed below.
Work package 1 focussed on deciphering the plants’ odorous vocabulary to create a complete inventory of “odour-prints” for a wide range of herbivore-plant and pathogen-plant combinations. We first evaluated the efficiency of various sorbents in trapping volatile organic compounds (VOCs) with the traditional dynamic headspace volatile collection system, followed by analysis with GC-MS. With the most efficient trapping filter selected, we collected and compared the emissions induced by various plant pests and diseases. Our model plant is maize, but we were also including cotton and bean plants in some of the studies. We were not able to obtain all insects and pathogens that we had planned to test, but we were able to create a nice overview of the variation in volatiles emissions that are released by important crop plants in response to different antagonists. This work continues in the context a follow-up Horizon Europe project named “PurPest”.
Work package 2 concerned the evaluation of sensor technologies for the detection of specific plant volatile mixtures. We mainly used two different approaches, one based on membrane-type surface stress sensors (MSS) and the other one on proton-transfer-reaction time-of-flight mass spectrometry (PTR-TOF-MS). In a laboratory assay, these sensors were astonishingly well at distinguishing the odorous signals from maize plants subjected to two caterpillar pests and one pathogen. Particularly impressive were the results obtained with the state-of-the-art PTR-TOF-MS, which we purchased for the project. It allows for real-time detection and quantification of volatiles blends and it was readily able to distinguish between the odours of plants that are under attack by different pests. Importantly, it was not only able to make this distinction under laboratory conditions but also outside, using deep learning, it could distinguish between healthy maize plants and maize plants with simulated caterpillar damage with high accuracy.
Work package 3 aimed to genetically manipulate maize plants to release a unique blend of easy-to-detect volatiles (aldoximes and nitriles) upon herbivory on the roots. We identified two maize genes that are specifically expressed in the leaves when the roots are damaged by rootworms. The promotors of these genes were coupled to two poplar genes involved in the biosynthesis of aldoximes and nitriles. The process of genetic transformation was successful, but it did not result in the envisioned odour emissions from the leaves of plants that were attacked by rootworms. However, we did find that some maize varieties, without genetic transformation, would emit volatiles from their leaves when their roots were subjected to insect damage. This implies that our sensor approach could also work to detect an important root pest that is currently invading Europe. We will be part of another European initiative to use odour sensors to detect pest and hope to further explore this possibility of odour-based root pest detection.
