Two hundred and fourteen unprocessed corn samples (1997-98 harvest), collected at wholesale markets in different regions in Brazil, were surveyed for the occurrence of mycotoxins. The samples were analysed for aflatoxins B1, B2, G1, G2, zearalenone and fumonisin B1 using in-house validated methods. The occurrence of aflatoxin B1, zearalenone and fumonisin B1 was found in 38.3, 30.4 and 99.1% of the samples, respectively. Aflatoxin B1, zearalenone and fumonisin B1 contamination levels varied from 0.2 to 129, 36.8 to 719, and 200 to 6100 microg/kg, respectively. The co-occurrence of the two carcinogenic mycotoxins aflatoxin B1 and fumonisin B1 was observed in 100% of the aflatoxin-contaminated samples (82 samples). Co-occurrences of aflatoxin B1: zearalenone: fumonisin B1 and aflatoxin B1: aflatoxin B2: fumonisin B1 were found in 18 and 43 samples, respectively.

A method for determining fumonisin B1 (FB1) in corn was developed and the clean-up optimized in order to give an extract suitable for one-dimensional thin layer chromatographic (TLC) analysis. FB1 was extracted with a solution of methanol:water (80:20, v/v), purified through an immunoaffinity column and separated on a C18 reversed phase TLC plate. The FB1 was visualized with 0.1 mol/l sodium tetraborate, 0.40 mg/ml fluorescamine in acetonitrile and 0.01 mol/l boric acid:acetonitrile (2:3, v/v) for fluorescence detection, and quantified by densitometric analysis. Water, acetonitrile:water (1:1 v/v) and acetonitrile:water (4:1 v/v) were evaluated as TLC solvents for running both standards and samples together with derivatization procedures aimed at improving separation, resolution, sensitivity and linearity. The mean recovery for FB1 for spiked samples was found to be 85% and the linear equation of standard calibration curve by densitometric analysis gave an r2 value higher than 0.99. The maximum coefficient of variation for replicate analysis of spiked samples was 19%. The absolute amount of FB1 standard detectable on a TLC plate was 2 ng, giving a detection limit for the method of 0.1 mg/kg. The method has been shown to be robust in the application of FB1 monitoring in corn (214 samples) collected in different regions of the country. FB1 was detected in 99% of these samples in the range of 0.2 to 6 mg/kg.