Graduate student Purdue University West Lafayette, Indiana
Body of Abstract: Amino acids are important precursors for numerous specialized metabolites. Comprehensive profiling of amino acid-derived metabolomes (AADMs) contributes to the identification of novel metabolites. Stable isotope labeling coupled with untargeted LC-MS analysis can be used to annotate the origin of metabolites globally. Using this strategy, a total of 1711 metabolite features were predicted to be derived from 15 amino acids in Arabidopsis (Columbia-0) leaves and stems, which represent 10% - 30% of total ion counts of metabolite features detected by untargeted LC-MS. Feeding of amino acid precursors enhanced the abundance of their cognate AADMs, as well as the levels of features derived from other amino acids, suggesting that the accumulation of amino acid-derived features (AADFs) is restricted by the availability of their direct amino acid precursor and that amino acid feeding can have long distance metabolic effects. To investigate the genetic basis of amino acid metabolism, the alignment of annotated AADFs with previous metabolic genome-wide association studies (mGWAS) led to the identification of 109,385 and 89,988 metabolite feature-SNP associations (P < 10-4) in leaves and stems, respectively. Well-characterized genes that contribute to amino acid metabolism were identified in this analysis, as well as novel associations with uncharacterized genes. Overall, this study demonstrated that the integration of isotope labeling in untargeted metabolomic analyses with mGWAS can facilitate the elucidation of genetic factors involved in metabolite biosynthesis and accumulation.