Organic osmolytes are methylamines, polyols, or amino acids that are essential for cell volume regulation. Some have other important biochemical roles. For example, N,N,N-trimethylglycine (betaine) is an osmolyte that also acts as a methyl group donor in one-carbon metabolism. Previous methods for the analysis of betaine and its metabolite N,N-dimethylglycine (DMG) used derivatization and high performance liquid chromatography with ultra violet detection (HPLC-UV). They required a long sample run time, and often lacked specificity. Liquid chromatography

• tandem mass spectrometry (LC-MS/MS) methods were developed with the aim of improving the analysis of betaine and other osmolytes. The objective was to measure the osmolytes: betaine; glycerophosphorylcholine (GPC); taurine; myo-inositol; sorbitol; and trimethylamine-N-oxide (TMAO), as well as N,N-dimethylglycine (DMG), choline, sarcosine, and carnitines. The use of LC-MS/MS improved the analysis of many osmolytes (and related metabolites) compared to previous HPLC- UV methods, and the number of osmolytes that could be analyzed was increased. The analysis of polyols was challenging and no suitable method was found for the analysis of sorbitol; however, amide columns were effective at separating inositol isomers. To demonstrate applications for these methods, samples from overweight people with type 2 diabetes were analyzed to improve knowledge about how osmolytes interact with other metabolites, and their potential use as risk markers in disease. Betaine, DMG, and choline concentrations had low intra-individual variation in the plasma and urine. However, GPC, taurine, and TMAO were more variable. Most subjects with diabetes who were taking bezafibrate had extremely elevated urine betaine. A previously unknown metabolite, N,N-dimethylglycine-N-oxide (DMGO), was identified in plasma and urine. DMGO significantly correlated (p < 0.05) with DMG, betaine, and choline, suggesting that it is derived from the oxidation of DMG. The discovery of DMGO in human samples suggests that a re-evaluation of the widely accepted pathway for choline metabolism may be required.