Development of a data dependent acquisition-based approach for the identification of unknown fast-acting toxins and their ester metabolites.

Phycotoxins in the marine food-web represent a serious threat to human health. Consumption of contaminated shellfish and/or finfish poses risk to consumer safety: several cases of toxins-related seafood poisoning have been recorded so far worldwide. Cyclic imines are emerging lipophilic toxins, which have been detected in shellfish from different European countries. Currently, they are not regulated due to the lack of toxicological comprehensive data and hence the European Food Safety Authority has required more scientific efforts before establishing a maximum permitted level in seafood. In this work, a novel data dependent liquid chromatography - high resolution mass spectrometry (LC-HRMS) approach has been successfully applied and combined with targeted studies for an in-depth investigation of the metabolic profile of shellfish samples. The proposed analytical methodology has allowed: i) to discover a plethora of unknown fatty acid esters of gymnodimines and ii) to conceive a brand new MS-based strategy, termed as backward analysis, for discovery and identification of new analogues. In particular, the implemented analytical workflow has broadened the structural diversity of cyclic imine family through the inclusion of five new congeners, namely gymnodimine -F, -G, -H, -I and -J. In addition, gymnodimine A (376.5 μg/kg), 13-desmethyl spirolide C (11.0-29.0 μg/kg) and pinnatoxin G (3.1-7.7 μg/kg) have been detected in shellfish from different sites of the Mediterranean basin (Tunisia and Italy) and the Atlantic coast of Spain, with the confirmation of the first finding of pinnatoxin G in mussels harvested in Sardinia (Tyrrhenian Sea, Italy).