2.6. Lipid Analysis (or Analysis of Fatty Acids) by LC-MS/MS

Lipid fraction was extracted in chocolate samples by applying the Bligh & Dyer protocol [20]. Briefly, 1 g of sample was mixed with 2 mL of chloroform and 4 mL of methanol followed by shaking for 2 min. Then, 2 mL of fresh chloroform was added and the final mixture was vortexed for 2 min. Finally, 3.6 mL of water was added to the mixture and, after vigorous shaking for 2 min, it was centrifuged at 771 g for 10 min. The lower phase was collected and transferred to a clean vial. The remaining pellet was then submitted to a second extraction step by adding 4 mL of a mixture of chloroform/methanol (90:10 v/v) followed by a 1 min shaking and a final centrifugation for 10 min at 771 g. The lower phase was then collected and mixed to the first extract. Finally, 1 mL of lipid extract was dried and dissolved in 2 mL of chloroform-methanol (1:1v/v), filtered through a 0.45 µm PTFE syringe filter and analyzed by LC-MS/MS.

Lipid fraction was analyzed by LC-MS/MS equipment consisting of a Q-Exactive™ Plus Hybrid Quadrupole-Orbitrap™ High Resolution Mass Spectrometer coupled to a UHPLC pump system (Thermo Fisher Scientific, Bremen, Germany). The mixture of lipids (5 µL injection volume, n = 2 replicates) was chromatographically separated on an Acclaim Mixed Mode HILIC-1 (150 × 2.1 mm, 3 µm, 120 Å, Thermo Fisher Scientific) as follows: 0–6 min isocratic at 90% of solvent B, 6–15 min gradual decrease of solvent B down to 60%, 15–25 min constant at 60% of B, 25–30 min gradual increase of solvent B up to 90%. These conditions were kept constant for 10 min for column conditioning. The mobile phases used for chromatographic separation were: A: MeOH:H₂O 90:10 (v/v) + 1 mM Ammonium acetate; B: AcN:H₂O 90:10 (v/v) + 1 mM Ammonium acetate. The flow rate was set to 300 µl/min and the temperature of the column set at 25 °C. MS analysis was performed both in positive and negative polarity by running the instrument in FullMS/Data dependent acquisition mode (FullMS/DD2). Resolution was set to 70,000 FWHM and 35,000 FWHM for FullMS and dd-MS2 scan event respectively, while dd-setting maximum AGC target value was set at 6.00e3.

Lipid identification was performed by processing raw MS data with the software Compound Discoverer 3.0 (Thermo Fisher Scientific, Bremen Germany) by setting the same parameters as detailed in par. 2.1.6. with exception for ChemSpider node where the only database interrogated was LipidMaps. Software results were filtered according to the same criteria above reported (par. 2.1.6).