Abstract
A gas chromatography mass spectrometry (GC-MS) metabolomics protocol was modified for quenching, harvesting, and extraction of metabolites from adherent cells grown under high (20%) fetal calf serum conditions. The reproducibility of using either 50% or 80% methanol for quenching of cells was compared for sample harvest. To investigate the efficiency and reproducibility of intracellular metabolite extraction, different volumes and ratios of chloroform were tested. Additionally, we compared the use of total protein amount versus cell mass as normalization parameters. We demonstrate that the method involving 50% methanol as quenching buffer followed by an extraction step using an equal ratio of methanol:chloroform:water (1:1:1, v/v/v) followed by the collection of 6 mL polar phase for GC-MS measurement was superior to the other methods tested. Especially for large sample sets, its comparative ease of measurement leads us to recommend normalization to protein amount for the investigation of intracellular metabolites of adherent human cells grown under high (or standard) fetal calf serum conditions. To avoid bias, care should be taken beforehand to ensure that the ratio of total protein to cell number are consistent among the groups tested. For this reason, it may not be suitable where culture conditions or cell types have very different protein outputs (e.g., hypoxia vs. normoxia). The full modified protocol is available in the Supplementary Materials.
Highlights
Reproducibility and robustness when performing metabolomics experiments are essential to guarantee that resulting biological information is both accurate and meaningful, and to reduce the likelihood of false discovery due to technical variation [1]
human umbilical vein endothelial cells (HUVECs) cultured under conditions of 20% Fetal calf serum (FCS), we discovered that this protocol was not sufficient
HUVECs cultured under conditions of 20% FCS, we discovered that this protocol was not sufficient to properly extract the cells
Summary
Reproducibility and robustness when performing metabolomics experiments are essential to guarantee that resulting biological information is both accurate and meaningful, and to reduce the likelihood of false discovery due to technical variation [1]. There are many studies focusing on sample preparation methods, and many cell-dependent methods have been optimized for the application of cell metabolomics for analyzing metabolic fingerprints [2,3,4,5,6,7,8,9,10]. No preparation method is broadly applicable for all cell types or cultivation conditions. We had previously discovered that, in our hands, certain cell extraction methods were inappropriate for the cell types we were using, typically because they did not adequately separate the protein layer from the solvents. This was assumed to be a result of inappropriate volumes and ratios of the solvents to adequately extract the cell lipid content
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
