Novel methods for identifying and quantifying metabolites in complex biological extracts by multidimensional nuclear magnetic resonance spectroscopy

Nuclear magnetic resonance (NMR) spectroscopy is one of the most powerful tools available for identifying and quantifying metabolites in simple mixtures. However, NMR's analytical utility is largely negated by signal overlap, which is inherent to NMR spectra of complex biological extracts. Consequently, NMR-based metabolomics studies are rarely able to identify the individual components of mixtures. Over the past ten years, roughly 1,000 articles have been published on NMR-based metabolomics. These studies generally rely on multivariate statistics for deciphering the overlapped spectra. I have developed an alternative strategy, bioanalytical metabolomics, which capitalizes on state-of-the-art multidimensional NMR to minimize resonance overlap. The bioanalytical metabolomics strategy allows up to 90% of the NMR observable metabolites to be identified and accurately quantified. Moreover, this approach allows hypothesis-driven research to be conducted on larger scale than was previously possible. In this thesis, I present the tools and techniques that I have developed for solving practical problems associated with the new technology and provide examples of hypothesis-driven research conducted with the bioanalytical metabolomics strategy.