Cell line dependence of metabolite leakage in metabolome analyses of adherent normal and cancer cell lines

Analysis of the metabolome can be sample and cell dependent. In this study, we compared two conventional pre-treatment approaches (trypsinization and cell scraping) in three adherently grown mammalian cell lines (two breast cancer cell lines MDA-MB-436, MCF-7 and an endothelial cell line—HMEC-1). We report experimental evidence, for the first time, demonstrating that metabolite leakage occurs with both treatments, and that the cell lines are differentially influenced. In addition, we examined two recently reported approaches of simultaneous quenching and extraction that showed minimal metabolome leakage. We also investigated the culture of cells on beads for rapid quenching and extraction, as a novel sample handling protocol. For metastatic breast cancer cells MDA-MB-436, the two direct quenching approaches and the bead harvesting approach showed favourable results with respect to metabolome leakage, compared to the conventional approaches. We characterised the recovery of eleven different classes of metabolites identified by gas chromatography–mass spectrometry in the cell extracts and the supernatants following quenching. Analysis of results based on metabolite classes is shown to be a useful approach aiding metabolomic interpretations. We also examined the effect of including a protein precipitation step on the metabolite classes detected. The de-proteinization step did not show significant improvement in overall recoveries. This investigation suggests that it is important to establish the level of metabolome leakage for the specific cell line investigated, irrespective of the methodology employed. Rapid approaches that combine quenching and extraction steps may be more effective in retaining valid metabolome data, with minimal metabolome leakage occurring.