Olive oil authenticity studies by target and nontarget LC–QTOF-MS combined with advanced chemometric techniques

Natasa P. Kalogiouri, Nikiforos A. Alygizakis, Reza Aalizadeh, Nikolaos S. Thomaidis

Food analysis is continuously requiring the development of more robust, efficient, and cost-effective food authentication analytical methods to guarantee the safety, quality, and traceability of food commodities with respect to legislation and consumer demands. Hence, a novel reversed-phase ultra high performance liquid chromatography-electrospray ionization quadrupole time of flight tandem mass spectrometry analytical method was developed that uses target, suspect, and nontarget screening strategies coupled with advanced chemometric tools for the investigation of the authenticity of extra virgin olive oil. The proposed method was successfully applied in real olive oil samples for the identification of markers responsible for the sensory profile. The proposed target analytical method includes the determination of 14 phenolic compounds and demonstrated low limits of detection ranging from 0.015 μg mL(-1) (apigenin) to 0.039 μg mL(-1) (vanillin) and adequate recoveries (96-107 %). A suspect list of 60 relevant compounds was compiled, and suspect screening was then applied to all the samples. Semiquantitation of the suspect compounds was performed with the calibration curves of target compounds having similar structures. Then, a nontarget screening workflow was applied with the aim to identify additional compounds so as to differentiate extra virgin olive oils from defective olive oils. Robust classification-based models were built with the use of supervised discrimination techniques, partial least squares-discriminant analysis and counterpropagation artificial neural networks, for the classification of olive oils into extra virgin olive oils or defective olive oils. Variable importance in projection scores were calculated to select the most significant features that affect the discrimination. Overall, 51 compounds were identified and suggested as markers, among which 14, 26, and 11 compounds were identified by target, suspect, and nontarget screening respectively. Retrospective analysis was also performed and identified 19 free fatty acids. Graphical Abstract Development of a novel RP-LC-ESI-QTOFMS analytical method employing target, suspect and non-target screening strategies coupled to advanced chemometric tools for the investigation of markers responsible for the sensory profile of extra virgin olive oil and guarantee authenticity.