Metabolome-Wide Analysis of Stable Isotope Labeling—Is It Worth the Effort?

Stable isotope assisted metabolomics techniques have emerged as a valuable tool in systems biology for metabolic flux analysis and pathway discovery (Duckwall et al., 2013; Chokkathukalam et al., 2014; Niedenführ et al., 2015). Traditionally, stable isotope labeling analyses have been highly targeted, meaning that isotopic enrichment of only a small set of metabolites has been analyzed to deduce metabolic fluxes. However, in recent years, tools for the global non-targeted detection, quantification, and computational analysis of isotopic enrichment have become available In this article, we will discuss whether and how such novel non-targeted stable isotope labeling analyses can be applied for systems-biomedical research. Pathological alterations of cellular processes usually manifest in altered metabolism. Therefore, analysis of metabolism is an ideal entry-point to diagnose or analyze diseases. Metabolic fluxes are the endpoint of cellular regulation and most likely to reflect changes on the genome, transcriptome or proteome level, and hence, are a valuable read-out for biomedical research (Wegner et al., 2015). Since intracellular metabolic fluxes cannot be measured directly, they are probed using stable isotope labeling: An isotopically enriched substrate is applied and the metabolization of this tracer leads to isotopic enrichment in downstream metabolites, depending on the underlying metabolic fluxes (Buescher et al., 2015). These enrichment patterns are analyzed by mass spectrometry (MS) or nuclear magnetic resonance (NMR) and are used to deduce metabolic fluxes (Truong et al., 2014; Young et al., 2014). Because isotopic labeling patterns are a direct consequence of metabolic fluxes, changes in these patterns indicate metabolic flux changes (Sauer, 2006). Consequently, global analysis of labeling patterns would allow for the global detection of metabolic flux changes. Because isotopic enrichment can be deduced solely from mass spectra (Jennings and Matthews, 2005), differential flux analysis based on differences in these isotopic labeling patterns does not require prior compound identification or a model of the metabolic network. This way, a non-targeted stable isotope labeling analysis can also consider unexpected or unknown compounds and reactions, circumventing current limitations of compound identification. Such a data-driven metabolic flux analysis is perfectly suited to pinpoint disease-specific alterations of cellular metabolism as little information on the experimental outcome is required. However, even in cases where there are specific hypotheses in place, a data-driven analysis may identify previously overlooked features. As such, non-targeted stable isotope labeling analysis can function as a hypothesis generator, aiding the design of subsequent experiments. Other than e.g., 13 C metabolic flux …