Untargeted Metabotyping Reveals Population-Level Variation in Plant Flavonoids and Alkaloids.

Metabolomics provides a powerful platform to characterize plants at the biochemical level, allowing a search for underlying genes and associations with higher level complex traits such as yield and nutritional value. Efficient and reliable methods to characterize metabolic variation in economically important species are considered of high value to the evaluation and prioritization of germplasm and breeding lines. In this investigation, a large-scale metabolomic survey was performed on a collection of diverse perennial ryegrass ( L.) plants. A total of 2,708 data files, derived from liquid chromatography coupled to high resolution mass spectrometry (LCMS), were selected to assess the effectiveness and efficiency of applying high throughput metabolomics to survey chemical diversity in plant populations. The data set was generated from 23 ryegrass populations, with 3-25 genotypes per population, and five clonal replicates per genotype. We demonstrate an integrated approach to rapidly mine and analyze metabolic variation from this large, multi-batch LCMS data set. After performing quality control, statistical data mining and peak annotation, a wide range of variation for flavonoid glycosides and plant alkaloids was discovered among the populations. Structural variation of flavonoids occurs both in aglycone structures and acetylated/malonylated/feruloylated sugar moieties. The discovery of comprehensive metabolic variation among the plant populations offers opportunities to probe into the genetic basis of the variation, and provides a valuable resource to gain insight into biochemical functions and to relate metabolic variation with higher level traits in the species.