An analytical platform for the screening and identification of pyrrolizidine alkaloids in food matrices with high risk of contamination. (16th April 2023)

• Record Type: Journal Article
• Authors: Rizzo, Serena; Celano, Rita; Piccinelli, Anna Lisa; Serio, Simona; Russo, Mariateresa; Rastrelli, Luca
• Is Part Of: Food chemistry. Volume 406(2023)
• Publication Date: 2023-04-16
• Subjects: Bee products -- Dietary supplements -- Herbal products -- Spectral library -- Suspect screening -- Target screening; Food -- Analysis -- Periodicals; Food -- Composition -- Periodicals; 664
• DOI: 10.1016/j.foodchem.2022.135058
• Languages: English
• ISSNs: 0308-8146

Abstract

Highlights: An analytical platform for the detection of PAs/PANOs in food matrices. A diagnostic product ions filtering strategy for the characterization of PAs/PANOs. An internal database of 779 known and expected unknown PAs/PANOs. A spectral library containing HRMS/MS information of 118 PAs/PANOs. The platform offers the possibility to detect both target and untarget PAs/PANOs. Abstract: An analytical platform for the detection of pyrrolizidine alkaloids (PAs) in honey, pollen, teas, herbal infusions, and dietary supplements is proposed; it includes a wide-scope suspect screening method, based on a diagnostic product ion filtering strategy for the characterization of PAs, and a target screening and identification method for the high-throughput detection of 118 PAs of a high-resolution mass spectral library. Salting-out assisted liquid–liquid extraction of aqueous extracts combined to ultra-high performance liquid chromatography–high-resolution tandem mass spectrometry was employed. The limit of identification (0.6–30 µg kg −1 ) of 28 standards were fit-for-purpose in PA-monitoring applications, with a false negative rate <1.3 % at 4 µg L –1 . The wide-scope suspect screening method allowed the tentative identification of 88 compounds. The screening of 282 commercial samples revealed a broad contamination of the studied matrices, demonstrating the effectiveness of the platform in detecting and identifying both target and untarget PAs.