Molecular networking as a novel approach to unravel toxin diversity of four strains of the dominant Dinophysis species from French coastal waters.

ABSTRACT

Some species of the genus Dinophysis contain Diarrhetic shellfish Poisoning (DSP) toxins and are the main threat to shellfish farming in Europe including France. Dinophysis species are known to produce two families of bioactive lipophilic toxins (i) okadaic acid (OA) and their analogues dinophysistoxins (DTXs) and (ii) pectenotoxins (PTXs). Only six toxins (OA, DTX1, DTX2, DTX3, PTX1 and PTX2) regulated by the European Union Legislation (EC No. 15/2011; 3) are routinely monitored using targeted chemical analysis by liquid chromatography coupled to mass spectrometry (LC-MS/MS) while toxic species of Dinophysis produce many other analogues. To tentatively identify unknown toxin analogues, a recent approach (Molecular Networking, MN) was used based on fragmentation data obtained by untargeted high resolution mass spectrometry (HRMS). An optimization of the data-dependent LC-HRMS/MS acquisition conditions was conducted to obtain more informative networks. The MN was applied to provide an overview of the chemical diversity of four strains belonging to three major Dinophysis species isolated from French coastal waters (D. acuta, D. caudata and the "D. acuminata complex" species D. acuminata and D. sacculus). This approach highlighted species-specific chemical patterns and also that Dinophysis chemical diversity is largely unexplored. Using MN allowed to identify directly known toxins and their relationship between species of Dinophysis, leading to the discovery of five new putative PTX analogues.