Molds are widely spread in food production and are also used in pharmacy (e.g. production of penicillin). On the other hand, they are also unwanted, very modest about growth conditions and tend to cause degradation of their environment. In fight for habitat and food they start producing mycotoxins, some of which are dangerous enough to be regulated by law in food commodities.
The main purpose of our work was to develop an analytical method for detection of some of regulated mycotoxins (aflatoxins B1, G1, B2, G2, ochratoxin A, deoxynivalenol, zearalenone and patulin) in different types of food, including starchy samples, samples rich in sugars and lipids. We tested some literature methods of sample preparation, then we simplified and optimised them to make a unified method for preparation of all sample types. We researched the influence of water and organic solvent ratio and the effect of different solvents on matrix effect and extraction efficiency. Furthermore, we inspected the effect of quantity and type of extraction salts, addition of acid and sample cleaning with dispersive solid phase extraction.
Firstly, we developed an instrumental method for multiresidue analysis of all 8 mycotoxins based on liquid chromatography tandem mass spectrometry. Because patulin differs from other mycotoxins in physico-chemical properties and also occurs in dissimilar types of food, we developed a separate sample preparation method for patulin. We validated both methods for selectivity, linearity, repeatability, accuracy, limit of quantitation, extraction efficiency and matrix effect. The method for patulin was successfully validated in three types of samples with achieved limit of quantitation low enough to analyse baby food samples. The method for other mycotoxins was also validated showing some deviations, especially for ochratoxin A, mostly because of low extraction yield, which need further development. However, we successfully developed an analytical method for detection of seven mycotoxins in concentration range within their regulation of law.
We applied the method to more than 50 food samples, at least 10 for each separate group (grains, dried fruits, nuts and apple products), which we bought in a store or were homemade. None of the store-bought sample contained mycotoxins, but we detected patulin, deoxynivalenol, ochratoxin A and zearalenone in visibly contaminated samples.