Gary Siuzdak from the Scripps Research Institute and his colleagues suspect that the heating step in gas chromatography-mass spectrometry analysis may change or destroy the compounds that are being analyzed, The Scientist reports.
Siuzdak and his colleagues report in Analytical Chemistry that they heated a set of standard samples to various temperatures for different lengths of time and then analyzed them using liquid chromatography coupled to electrospray ionization mass spectrometry — an approach that does not include a heating step.
Heating, they found, affects the molecular profiles of the samples. For instance, they report that after heating a sample to 250°C for 300 seconds, more than 40 percent of the molecular peaks were altered in the plasma metabolite analysis, as compared to unheated samples.
This, Stephen Barnes, from the Targeted Metabolomics and Proteomics Laboratory at the University of Alabama, Birmingham, tells Chemical & Engineering News, could be why whole metabolome analyses often identify less than 20 percent of cellular metabolites.
But other researchers take issue with the paper's methods. University of California, Davis' Oliver Fiehn tells C&EN that the paper is "clear nonsense." He notes that analytical chemists control for method parameters like heating when they develop GC-MS analytical methods, controls he says the Siuzdak group didn't use.
Fiehn also says that the group didn't protect hydroxyl and amino groups and sugars from being broken down, and that it analyzed trimethylsilyl-derivatized samples in a water-containing solvent — and water cleaves trimethylsilyl groups.
Still, the University of Alberta Liang Li, who develops MS methods for proteomics and metabolomics, notes that the study serves as a reminder to refrain from using high temperature or other conditions for more than is necessary.