Optimization and validation of a LC-MS/MS method for the determination of Alternaria toxins in tomato products

Alternaria species are fungi present in the soil as normal components of the microflora and are both saprophytic and plant pathogenic. They can infect oilseeds such as sunflower and canola, as well as tomatoes, apples, citrus fruits, olives, and many other fruits and vegetables. Alternaria species are able of producing more than 70 secondary metabolites, some of which are chemically characterized and reported as mycotoxins, Alternaria toxins (ATs). The most common Alternaria toxins in food products are alternariol (AOH), alternariol monomethyl ether (AME), tenuazonic acid (TeA) and tentoxin (TEN). Thus, this work aims to optimize and validate a LC-MS/MS method for the determination of Alternaria toxins (AOH; AME; TeA and TEN) in tomato products. The operating parameters of the mass spectrometer were optimized through the direct infusion of single solutions of the different ATs studied. During tuning, the temperature and voltage values applied to the source, as well as the flow and temperature of the desolvation gas, were optimized. Intensities of the precursor ions were monitored, evaluating both positive and negative ionization. AOH, AME, TeA and TEN gave a greater signal intensity in negative ionization and the most abundant precursor ion was found to be, for all 4 analytes, [M-H]-. Furthermore, during the infusion, for each compound studied, two characteristic fragments were identified to be used for quantification (quantifier, Q) and identification (qualifier, q) and the respective collision energies. The choice of product ions was made by preferring transitions characterized by the best intensity and repeatability of the signal, also in terms of ion ratio (IR). The tests for the optimization of the sample preparation were performed on a non-contaminated fresh tomato sample. Different extraction mixtures and different QuEChERS composition were tested. The optimized method was fully validated has been validated. In particular, the identification criteria of the molecules were defined and the linearity of the response was verified. Furthermore, the matrix effect was evaluated and the accuracy and precision were estimated. LOD and LOQ values and the method concentration range were identified. Finally, the uncertainty was estimated with a metrological approach. The values of the characteristic parameters of the method obtained were judged very positively for both accuracy and precision when compared with what is required in the EURL-MP guideline. LOQ values obtained were judged adequate, and in accordance with what is indicated in Recommendation 2022/553, which for tomato based products requires a LOQ not exceeding 4 μg/kg for AOH and AME and not exceeding 20 μg/kg for TeA. TEN is not included in the Recommendation, therefore there are no reference values, however, the LOQ value obtained is considered adequate as it is lower than that validated for AOH and AME. The validated method was applied to the analysis of 46 samples of tomato-based products, pulp and peeled tomatoes (N=17), puree (N=10), double concentrate (N=16) and triple concentrate (N=3). The indicative levels reported in Recommendation (EU) 2022/553 are exceeded for two samples in the "Puree" category for AOH, while the measured AME concentration level did not exceed the indicative levels in any of the analyzed samples. As for TeA, 4 samples showed have contamination above the indicative levels, 2 in the "Puree" category and 2 in the "Double concentrate" category. However, considering the mean values, always lower than the indicative level, the picture is not alarming.