Development, optimization and validation of a multimethod for the determination of 36 mycotoxins in wines by liquid chromatography-tandem mass spectrometry.

PMID 25127606


A fast and efficient multimethod for the determination of 36 mycotoxins in wine, using ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS), was developed, optimized, validated and implemented in routine analysis. A simplified, quick extraction was performed with acetonitrile, derived from the QuEChERS (quick, easy, cheap, effective, rugged and safe) approach, which was traditionally developed for pesticides analysis. This study aimed at a single extraction and chromatographic separation for 36 mycotoxins. Optimization tests were performed to find the proper ratio of wine: water and extraction solvent and the need for an additional buffering step with ammonium formate/formic acid and a dispersive SPE cleanup with various sorbents. The dSPE steps did not show significant improvement in analysis results, therefore, it was not applied in the final method to be validated. The mycotoxins were separated and detected on a UPLC-MS/MS system, used in the ESI positive ionization mode. The various mycotoxins were divided in three different concentration level groups, according to their sensitivity in UPLC-MS/MS. The validation was performed by analyzing recovery samples at three different spike levels with six replicates (n=6) at each level. Linearity (r(2)) of calibration curves, accuracy (recovery %), instrument limits of detection and method limits of quantification (LOD and LOQ), precision (RSD%) and matrix effects (%) were determined for each individual mycotoxin. From the 36 mycotoxins analyzed by UPLC-MS/MS (ESI+), 35 showed average recoveries in the range 70-120%, and 86% of these with a RSD≤20% at the lowest spike level (for Group I, II and III, respectively, 1, 50 and 10 µg kg(-1)). The higher spike levels showed even better results. Only nivalenol could not be quantified at any concentration level. The method LOQ for 86% of the mycotoxins studied was the lowest spike level tested. The matrix effect observed was low for most mycotoxins analyzed and had no significant influence on the analytical results obtained. The developed procedure was applied successfully in routine analysis in a survey of wine samples originating from different countries.