Optimization and validation of a quantitative liquid chromatography–tandem mass spectrometric method covering 295 bacterial and fungal metabolites including all regulated mycotoxins in four model food matrices

An LC–MS/MS “dilute and shoot” method for the determination of 295 fungal and bacterial metabolites was optimized and validated according to the guidelines established in the Directorate General for Health and Consumer Affairs of the European Commission (SANCO) document No. 12495/2011. Four different types of food matrices were chosen for validation: apple puree for infants (high water content), hazelnuts (high fat content), maize (high starch and low fat content) and green pepper (difficult or unique matrix). Method accuracy and precision was evaluated using spiked samples in five replicates at two concentration levels. Method trueness was demonstrated through participation in various proficiency tests. Although the method covers a total number of 331 analytes, validation data were acquired only for 295 analytes, either due to the non-availability of analytical standards or due other reasons described in this paper. Concerning the apparent recovery, the percentage of 295 analytes matching the acceptable recovery range of 70–120% lied down by SANCO varied from 21% in green pepper to 74% in apple puree at the highest spiking level. At the levels close to limit of quantification only 20–58% of the analytes fulfilled this criterion. The extent of matrix effects was strongly dependent on the analyte/matrix combination. In general, the lowest matrix effects were observed in apple puree (59% of analytes were not influenced by enhancement/suppression at all at the highest validation level). The highest matrix effects were observed in green pepper, where only 10% of analytes did not suffer from signal suppression/enhancement. The repeatability of the method was acceptable (RSD ≤ 20) for 97% of all analytes in apple puree and hazelnuts, for 95% in maize and for 89% in green pepper. Concerning the trueness of the method, Z-scores were generally between −2 and 2, despite a broad variety of different matrices. Based on these results it can be concluded that quantitative determination of mycotoxins by LC–MS/MS based on a “dilute and shoot” approach is also feasible in case of complex matrices.