Comparison of two sampling, quenching and extraction methods for quantitative yeasts metabolomics

Knowing intracellular metabolite concentrations is important for fundamental metabolism research as well as for biotechnology. The first steps in a quantitative metabolomics workflow, i.e., sampling, quenching, and extraction, are key to obtaining unbiased quantitative data. A qualified quenching and extraction method not only needs to rapidly terminate the in vivo biochemical reaction activities to preserve the endogenous metabolite levels but also has to fully extract all metabolites from cells. Recently, two different filtration-based sampling, quenching, and extraction methods have been proposed and used for quantitative yeast metabolomics. One method integrates the quenching and extraction into one step using a methanol-acetonitrile-water solution after a filtration step, while the other --more conventional-- method quenches with cold methanol and extracts with boiling ethanol. In this study, we tested whether these two methods are equally well suited for quantitative metabolome analyses with yeast. Using isotope dilution mass spectrometry (IDMS) with GC-MS and LC-MS as analytical methods, in combination with precise quantification of cell volumes, we determined absolute concentrations of 63 intracellular metabolites covering amino acids, organic acids, phosphorylated sugars, and nucleotides in two S. cerevisiae strains with different physiology. By analyzing the data from samples generated with the two methods, we found that while both methods yielded essentially identical concentrations for amino and organic acids, the cold-solvent extraction yielded significantly lower concentrations for particularly phosphorylated sugars and nucleotides, presumably because of lower quenching or extraction efficiency of this method. Our results show that methanol-quenching combined with boiling-ethanol-extraction is the more accurate approach when aiming to quantify a broad range of different metabolites from yeast cells. The significant discrepancy observed between both common metabolite extraction methods demonstrates the importance of method optimization for quantitative studies in particular when working with microbes with rigid cell walls such as those found in yeast.