Improving Perennial Ryegrass Seed Production in Northern Minnesota: Developing Metabolomics-Assisted Selection Techniques for Crown and Stem Rust Resistant Cultivar Development.

Phenotypic screening for resistance to crown rust (Puccinia coronata) in a perennial ryegrass (Lolium perenne) breeding program can be time consuming, costly, and unpredictable do to pathogen and environment variability.  Turfgrass breeders need a technique to select for rust resistance in the absence of the actual pathogen in order to be able to select for other traits simultaneously and speed cultivar development.  Metabolomics-assisted breeding has recently emerged a technique to select for desired traits based on plant chemical profiles that have been previously correlated with a trait of interest. Chemical compounds within plants, specifically secondary metabolites, are the end products of complex biochemical pathways that are regulated by genes and are often used directly as stress mediation or plant defense mechanisms making them closely associated to plant phenotypes.  The potential to use secondary metabolites as biomarkers for selecting desired traits in plants has been demonstrated in arabidopsis, grapes, wheat, and conifer. In perennial ryegrass, using reverse-phase ultra performance liquid chromatography (RP-UPLC) coupled to mass spectrometry (MS), we have developed high-throughput methods to detect a diverse mixture of secondary metabolites which could potentially be correlated with rust resistance levels.  In addition, a large perennial ryegrass population was field screened for crown rust resistance in 2009, 2010 and 2011 with 14 lines demonstrating consistent crown rust resistance rankings.  Clones from the 14 selected lines were propagated in the greenhouse and were additionally screened for rust resistance via artificial inoculation using a diverse collection of crown rust spores to control environmental variation.  The rust resistance data from the greenhouse and field will be correlated with the plant metabolic profiles to develop a model to help predict rust resistance levels in our perennial ryegrass population.