Polar licit and illicit ingredients in dietary supplements: chemometric optimization of extraction and HILIC-MS/MS analysis

Abstract Many dietary supplements claim the ability to enhance sports performance and to improve the fitness of the consumers. Occasionally, along with legal ingredients, illicit compounds may be added without being labelled, leading to unintended doping. Hence, the aim of this study was to develop an analytical method to determine a set of 12 polar (logDpH=7 from −2.0 to +0.3) compounds including diuretics, stimulants, β2-agonists, methylxanthines, and sweeteners. Hydrophilic interaction liquid chromatography was chosen as separation strategy, coupled with tandem mass spectrometry. The instrumental method was optimized using a two-step design of experiments (DoE). Firstly, a Plackett–Burman (PB) DoE was performed to identify the more influencing variables affecting peak areas and chromatographic resolution among temperature, water percentage in the mobile phase, and flow rate, as well as type and concentration of buffers. Secondly, a D-optimal DoE was set, considering only the most significant variables from the PB-DoE results, achieving a deeper understanding of the retention mechanism. Sample processing by salt-assisted liquid–liquid extraction was studied through DoE as well, and the whole method showed recoveries in the range 40–107% and procedural precision ≤11% for all analytes. Finally, it was applied to real samples, in which the four methylxanthines and two artificial sweeteners were detected and quantified in the range of 0.02–192 mg g−1. These values were compared to the quantities declared on the DS labels, when possible. Furthermore, a sequence of MS/MS scans allowed detection of a signal in one of the samples, structurally similar to the β2-agonist clenbuterol. Graphical Abstract