Metabolite quantification by NMR and LC-MS/MS reveals differences between unstimulated, stimulated, and pure parotid saliva

Saliva is a readily available biofluid that is sensitive to metabolic changes and can be collected through rapid and non-invasive collection procedures, and it shows great promise for clinical metabolomic studies. This work studied the metabolite composition of, and the differences between, saliva samples collected by unstimulated spitting/drooling, paraffin chewing-stimulated spitting, and parotid gland suction using targeted nuclear magnetic resonance (NMR) spectroscopy and liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) for metabolite quantification. As applied here, these two analytical techniques provide complementary metabolite information and together extend the metabolome coverage with robust NMR quantification of soluble metabolites and sensitive targeted LC-MS/MS analysis of bioactive lipids in specific metabolic pathways. The NMR analysis was performed on ultrafiltrated (3 kDa cutoff) saliva samples and resulted in a total of 45 quantified metabolites. The LC-MS/MS analysis was performed on both filtered and unfiltered samples and resulted in the quantification of two endocannabinoids (AEA and PEA) and 22 oxylipins, which at present is the most comprehensive targeted analysis of bioactive lipids in human saliva. Important differences in the metabolite composition were observed between the three saliva sample collection methods, which should be taken into consideration when designing metabolomic studies of saliva. Furthermore, the combined use of the two metabolomics platforms (NMR and LC-MS/MS) proved to be viable for research and clinical studies of the salivary metabolome. (C) 2017 Elsevier B.V. All rights reserved.