Detection of Inverse Stable Isotopic Labeling in Untargeted Metabolomic Data

Abstract

Stable isotopic labeling is a powerful tool for determining the biosynthetic origin of metabolites and for discovering natural products that incorporate precursors of interest. When isotopically substituted precursors are not available commercially or synthetically, inverse stable isotopic labeling (InverSIL) is a useful alternative. With InverSIL, an organism is grown on an isotopically substituted medium and then fed precursors of natural isotopic abundance which can be tracked by mass spectrometry, thereby bypassing issues with precursor availability. Currently, there is no automated way to identify precursor incorporation in untargeted metabolomic data using InverSIL without specifying an expected change in the mass-to-charge ratio of metabolites that have incorporated the precursor. This makes it difficult to identify unknown natural products that may incorporate portions of precursors of interest using new biochemistry or to rapidly identify incorporation of multiple precursors into different metabolites simultaneously. To address this, we developed a new, robust workflow for the automated identification of inverse labeling in untargeted metabolomic data. We then use this method to identify metabolites that incorporate para-aminobenzoic acid and different portions of l-methionine, including in the same sample, and in the process discover the likely biosynthetic origin for the C-7 and C-9 methyl groups of the pterin portion of dephosphotetrahydromethanopterin, a C1 transfer coenzyme used by methylotrophic bacteria. This workflow can be applied in the future to streamline the use of the versatile InverSIL approach for natural product and metabolism research.