Isotopomer analysis by mass spectrometry and mathematical algorithm

In metabolic engineering, discovery of the flux distribution in a metabolic network plays a very important role. To evaluate the flux distribution, often spesifically labelled substrates are used in cultivations and positional isotopomer distributions of the metabolites (for example amino acids, sugar phosphates) are analysed. At the moment the main methods for analysing the positional isotopomer distributions are NMR and GC-MS but recently also GC-MS/MS is succesfully used [1]. An interest for LC-MS and LC-MS/MS- methods has been increased because of an easier sample pretreatment than with GC-MS or GC-MS/MS [2]. Also no derivatization of the analytes is needed with LC-MS and LC-MS/MS -techniques. In our earlier study, a computational method called PIDC (Positional Isotopomer Distribution Calculator) has been developed for calculation of positional isotopomer distribution from mass spectrometric data [3]. In current work, PIDC has been developed further and a new and more user-friendly version has been used in isotopomer calculations. In our experiments, an alanine standard solution with natural alanine, 1-13C-alanine and 3-13C-alanine was used. PIDC-calculations were done from the LC-MS and LC-MS/MS -data. In measurements, daughter ion scanning mode and MRM (multiple reaction monitoring) mode with two collision energy (15 eV and 30 eV) were tested. Repeatability (n= 5 injections) and measured isotopomer distribution/theoretical distribution relationship were compared between different methods. The best results were with measurements with MRM mode, 15 eV, where RSD:s varied between 1-10 % depending on the isotopomer and the measured isotopomer distributions were 99 106 % from the theoretical ones. Also GC-MS (MTBSTFA derivatization) experiments with the alanine solution have been tested. Acid hydrolysed Saccharomyces cerevisiae biomass sample from an anaerobic chemostat cultivation on glucose where 10 % of the carbon source molecules were uniformly 13C-labelled, has been analysed by LC-MS and LC-MS/MS. Positional isotopomer distributions of the amino acids in the sample were calculated by PIDC and the results have been compared to the distributions analysed by NMR. References [1] F.M.H. Jeffrey, J.S. Roach, C.J. Storey, A.D. Sherry, C.R. Malloy, Analytical Biochemistry 300 (2002) 192. [2] W.A. Van Winden, 13C-Labelling Technique for Metabolic Network and Flux Analysis, Theory and Applications, Doctoral Thesis (2002), DelftChemTech building, Delft University of Technology, The Netherlands. [3] A. Rantanen, J. Rousu, J.T. Kokkonen, V. Tarkiainen, R.A. Ketola, Metabolic Engineering 4 (2002), 285.