The University of Wisconsin-Madison, Chemistry, Madison, WI 53706
Quantification by liquid chromatography-mass spectrometry (LC-MS) requires special consideration due to the potential for run-to-run variability of retention times and matrix effects such as ionization suppression during the commonly used electrospray ionization. Several different strategies have been employed for quantitative LCMS of metabolomic samples. Arguably, the most precise method involves the addition of an isotopically-labeled internal standard for every compound of interest (e.g. 2H, 13C, or 15N-labeled). This approach is expensive, but warranted in certain targeted studies. Another strategy for relative quantification, as opposed to absolute quantification, is chemical labeling, which has proven to be useful for quantification in proteomics (e.g., isotope-coded affinity tags). Though relative quantification by labeling has seen limited use for metabolomics due to the lack of a single functional group present in all metabolites, there has been none-the-less a number of recent reports of effective labeling schemes [Guo 2007, Huang 2008, Lamos 2007, Shortreed 2006 and Yang 2007]. Labeled metabolites coelute from the chromatographic separation and appear in the mass spectrum as pairs of peaks with a characteristic mass difference. The peak intensity ratio for each pair yields the relative concentration. Such labeling strategies have a number of advantages including: improved quantitative precision, increased ability for molecular identification, and enhanced detection sensitivity. A major limitation of this strategy is lack of software tools for global identification of labeled compounds and calculation of the peak intensity ratios. Here we report a new software tool for automated identification of isotopically labeled metabolites and quantification of their relative concentration.
[Abstract (DOC)]