An integrative approach to investigate natural variation in the accumulation of aliphatic glucosinolates in Arabidopsis thaliana

Abstract Glucosinolates are a fascinating class of specialised metabolites found in the plants of Brassicacea family. The variation in glucosinolate composition across different Arabidopsis ecotypes could be a result of allelic compositions at different biosynthetic loci. The contribution of methylthioalkylmalate synthase (MAM) genes to diversity of glucosinolate profiles across different Arabidopsis ecotypes has been confirmed by genetic analyses. Different MAM isoforms utilise different chain-elongated substrates for glucosinolate biosynthesis causing thus a variation in chain lengths across different Arabidopsis ecotypes. To further investigate the relationship between the genotype and the associated metabolic phenotype, we studied the diversity of genes and enzymes of glucosinolate biosynthesis. Using Shannon entropy as a measure we revealed that several genes of the pathway show a clear derivation from the expected behaviour, either accumulating non-synonymous SNPs or showing signs of purifying selection. We found that the genotype-phenotype relationship is much more complicated than inferred from the diversity of MAM synthases. We conclude therefore, that the ON/OFF feature of key QTLs is not enough to elucidate the diversity of glucosinolates across different Arabidopsis thaliana ecotypes and that glucosinolate profiles are determined also through the polymorphic residues along the coding regions of multiple metabolic genes.