Spectrally Accurate Quantitative Analysis of Isotope Labeled Compounds.

Calculated profile mode mass spectrometry (MS) data are fitted to lineshape calibrated LC/MS data using a Multiple Linear Regression (MLR) model to quantitate the relative concentrations of stable or radio labeled compound mixtures. This alternative approach significantly improves the precision and accuracy over existing MS methods while providing the much-needed statistical diagnostics on the goodness of the fitting model and thus reliability of the quantitative results obtained. Test compound data containing S/Cl atoms have been measured with either stable deuterium labeling or radioisotope uniform 14 C labeling onto an aromatic ring. Since the entire relative distribution of variously labeled compounds is automatically obtained through this approach, it is feasible to directly calculate the Specific Activity (SA) from such mass spectral analysis without radioactivity detection and the usual standard curve quantitation. The applicability of this approach to systematically and accurately accommodate and account for incomplete labeling chemistry or other impurities is also discussed, with wide-ranging implications including metabolic flux, HDX (Hydrogen-Deuterium Exchange), and quantitative proteomics.