Effects of sample preparation procedures on non-targeted NMR analysis of tomatoes

In the last years the application of nuclear magnetic resonance spectroscopy (NMR) to the metabolomics has being experienced an increasing number of applications [1], due to the development of new mathematical algorithms for the statistical analysis, and more sensitive and highly reproducible spectrometers. In metabolomic analysis the sample stability is a crucial aspect, being the prerequisite for assuring high reproducibility of the results obtained by different laboratories and, thus, a valuable inter-laboratory comparison. The stability of a sample is largely affected by its preparation protocol, which should include all the operations aimed at ensuring an efficient extraction of the metabolites and a lasting stability of the metabolic profile over the time [2]. As part of a collaborative research project (Istituto Poligrafico Zecca dello Stato – Politecnico di Bari) aimed at discriminating the geographical origin of Italian tomatoes, we disclosed the crucial aspects of the sample preparation affecting the metabolic profiles of the tomato extracts, and, importantly, we shed light on the practical procedures able to reveal the presence of some unknown metabolites and to control their transformations over the time. Moreover, we demonstrated that depending on the sample preparation the metabolic profile can suffer a significant variation. The results reported in this communication should be of general interest for the scientific community involved in food analysis