Advanced Structural Mass Spectrometry for Metabolomics
Goodwin, Cody Ray
:
2013-03-04
Abstract
This work explores the application of mass spectrometry-based multidimensional separations to metabolite analysis. The advent of a gas phase structure-based prioritization method for secondary metabolite prioritization and discovery is described and demonstrated. This work begins with the investigation of the gas phase structures of peptidic natural products as compared to linear peptides. Using empirical data interpreted using theoretical conformations, a structural motif is determined that is utilized for the prioritization of the tricyclic peptide siamycin II from a crude extract. This work is initially established using uniform field mobility instrument, and later expanded to the commercially available traveling wave based mobility technology. The concept of conformational uniqueness is applied to compound prioritization using a devised “inherent trendline”, resulting in the isolation and identification of a previously undiscovered set of angucyclines from a crude extract, termed lechacyclines due to their hypogean origin (Lechuguilla Cave, NM). Additionally, established multivariate statistical analysis methods are applied to the metabolomic investigation of rifampicin and streptomycin resistant mutants grown in the absence of antibiotic. This study reveals persistent aberrant metabolism, with a global increase in secondary metabolite production, resulting in the discovery of a class of previously unknown compounds, termed mutaxanthenes. Further, the development and application of a self-organizing map based metabolite analysis method is presented. This method is described in detail, and compared with the commonly used methods of principal component analysis and orthogonal partial least square-discriminant analysis. As a proof of principle, sera metabolite profiles from rat behavioral models for cocaine addiction were analyzed using the proposed Metabolite Expression Dynamics Investigator method and the aforementioned multivariate statistical methods. Prioritized features are putatively identified and metabolic inferences described. Finally, developments and applications of these methods utilized and established herein are proposed.