Increasing the accuracy of mass isotopomer analysis through calibration curves constructed using biologically synthesized compounds

✍ Scribed by Tie Shen; Weiqun Shen; Ying Xiong; Haiyan Liu; Haoran Zheng; Hong Zhou; Bin Rui; Jianping Liu; Jihui Wu; Yunyu Shi

• Book ID: 102904754
• Publisher: John Wiley and Sons
• Year: 2009
• Tongue: English
• Weight: 556 KB
• Volume: 44
• Category: Article
• ISSN: 1076-5174
• DOI: 10.1002/jms.1583

✦ Synopsis

Abstract

Mass isotopomer analysis is an important technique to measure the production and flow of metabolites in living cells, tissues, and organisms. This technique depends on accurate quantifications of different mass isotopomers using mass spectrometry. Constructing calibration curves using standard samples is the most universal approach to convert raw mass spectrometry measurements into quantitative distributions of mass isotopomers. Calibration curve approach has been, however, of very limited use in comprehensive analyses of biological systems, mainly suffering from the lack of extensive range of standard samples with accurately known isotopic enrichment. Here, we present a biological method capable of synthesizing specifically labeled amino acids. These amino acids have well‐determined and estimable mass isotopomer distributions and thus can serve as standard samples. In this method, the bacterium strain Methylobacterium salsuginis sp. nov. was cultivated with partially ^13^C‐labeled methanol as the only carbon source to produce ^13^C‐enriched compounds. We show that the mass isotopomer distributions of the various biosynthesized amino acids are well determined and can be reasonably estimated based on proposed binomial approximation if the labeling state of the biomass reached an isotopic steady state. The interference of intramolecular inhomogeneity of ^13^C isotope abundances caused by biological isotope fractionation was eliminated by estimating average ^13^C isotope abundance. Further, the predictions are tested experimentally by mass spectrometry (MS) spectra of the labeled glycine, alanine, and aspartic acid. Most of the error in mass spectrometry measurements was less than 0.74 mol% in the test case, significantly reduced as compared with uncalibrated results, and this error is expected to be less than 0.4 mol% in real experiment as revealed by theoretical analysis. Copyright © 2009 John Wiley & Sons, Ltd.