Desenvolvimento e validação de metodologia para determinação de multimicotoxinas em amostras de amêndoas de cacau por LC-MS/MS

Abstract

The cultivation and commercialization of cocoa beans and their derivatives show an important economic and social role for Brazil due to its export and domestic consumption. The quality of the final product is largely tied to the microorganisms present in the product during the fermentation process, which are essential for the sensorial characteristics. However, the presence of undesirable microorganisms as toxigenic fungi may compromise the food safety of consumers due to the synthesis of mycotoxins. Therefore, the objective of this work was to develop, validate and apply an analytical methodology for the determination of multimicotoxins in cocoa. The developed method employed liquid-liquid extraction with addition of salt and freezing step with subsequent liquid chromatography analysis coupled with sequential mass spectrometry (LC-MS/MS). The extraction and optimization was based on the extraction techniques QuEChERS and dilute and shoot, the aim of the modification was to reduce the co-extractives of the final extract and to enable an efficient extraction without a clean up step for 42 mycotoxins. The performance of the proposed method was evaluated for 34 mycotoxins in cocoa, the results obtained were discussed based on the national and international guidelines established for contaminants in food, among them: SANTE/12089/2016, SANTE/11813/2017, the decision 2002/657/EC, Regulation 401/2006/EC and the Manual de garantia da qualidade analítica. The recoveries and precision were given as appropriate, except for the ionic mycotoxins with the ammonium adduct (NH4 + ), E-cristinine and β-ZOL, which presented greater variation in the results, possibly due to the competition of Na+ ions by the ionization of the molecule. This result directly reflected the measurement uncertainty of these mycotoxins, since the accuracy and correction factor of the recovery were the factors that presented the greatest impact on the uncertainty of the method. The reproducibility of the method was evaluated for ochratoxin A (OTA) using samples of proficiency tests, in which z-score between ± 2 was considered satisfactory. The limit of quantification of each mycotoxin was determined by fortification at levels theoretically obtained, ranging from 1.0 to 75 μg kg-1 . The co-occurrence study for 34 mycotoxins was carried out on 135 samples of dried cocoa beans. The results indicated that 42% of the samples had quantifiable levels for mycotoxins: aflatoxins (AFB1, AFB2 and AFG1), OTA, citrinin (CIT), cyclopiazonic acid (CA), tenuazonic acid (TA), paxilin (PAX), esterigmatocistin STG), zearalenone (ZON), and fumonisin B2 (FB2). Of the contaminated samples, 18% presented co-occurrence between two and six mycotoxins, allowing the potentiation of toxicity and synergistic effects in the body of the consumer. In addition, to date there are no international legislation with maximum tolerated limits for mycotoxins in cocoa, so the present case study is important to compose future discussions to regulate the same.