MetaboHUB – Bordeaux Metabolome Facility

International audience

إنجليزي. Plant performance in the current context of climate change and the need to rethink agriculture in a more sustainable way is an issue of worldwide concern. Breeding varieties that are more resistant to (a)biotic stresses and produce quality crops with higher yields could be part of the solution. However, and even more so for fruit trees, breeding process is long and quite uncertain. While genomics, transcriptomics and proteomics provide a general view of plant metabolism, metabolomics reveals metabolism adjustments that occur at a given time and under given conditions, allowing the identification of biomarkers for agronomical traits of interest.By carrying out high-throughput multi-species metabolome analyses coupled with machine learning, Bordeaux Metabolome Facility takes part in the development of pioneering technologies for predicting phenotypic traits including yield, biomass composition and resistance to stress at early developmental stages (Dussarrat et al. 2022). Other innovative tools are developed, focusing on modeling metabolic pathways (Decros et al. 2023) or assessing the quality, content and authenticity of plant-derived products such as wine (Gougeon et al. 2018). The workflow starts with the ethanolic, acidic or basic extraction of the metabolites for which robotized methods have been developed, including preparation of Quality Controls performed under the same conditions as the samples. Untargeted metabolomics based on LC-MS (UHPLC-LTQ-Orbitrap HRMS, ThermoScientific) are conducted followed by data processing, annotation and statistical analyses using MSDial, PeakForest, Sirius, Metaboanalyst and GNPS. For metabolic pathway modeling, targeted metabolite quantification and enzymatic capacity assays are also performed and combined with computer modeling and model parameter optimization.High-throughput tools developed by Bordeaux Metabolome Facility from sample processing to data interpretation generate crucial results to identify biomarkers of plant performance and to build predictive computational models. These models should provide significant help for plant breeding as well as for understanding crops adaptation to climate change.