STABILITY AND PROBLEMS IN RECOVERY OF FUMONISINS ADDED TO CORN-BASED FOODS

Because the natural occurrence of fumonisins is so far known almost exclusively in corn, we have limited our investigations on their stability to corn-based foods. In these studies, distinction must be made between real losses, binding, and any matrix-related method problems. Fumonisins B-1 (FB1) and B-2 (FB2) were about 40% recovered when heated in corn meal at 190 degrees C, about 20-30% recovered when heated in moist corn meal at 190 degrees C, and completely unstable in corn meal at 220 degrees C. Average recoveries of FB1 and FB2 added to blank heated matrixes were 69-107% in control experiments. Baking corn meal muffins spiked with 2.5 mu g FB1 and FB2/g corn meal at 220 degrees C also resulted in losses of fumonisins. Little or no fumonisins were recovered from corn bran flour when methanol-water (3 + 1) was used as extraction solvent. However, when methanol-borate buffer (pH 9.2) (3 + 1)was used, recoveries averaged 91 +/- 17 and 84 +/- 9%, respectively, for FB1 and FB2; and natural contamination of the corn bran flour with FB1 and FB2 at levels of 1.9 and 0.95 mu g/g, respectively, was revealed. Comparable recoveries were observed for 1 brand of a corn bran breakfast cereal, but the binding effect was not seen with a second brand, for which methanol-water (3 + 1)alone was a good extraction solvent. Recoveries of FB1 and FB2 from a mixed cereal for babies were only about 50% with either extraction solvent mixture. Additives present in the mixed cereal-calcium carbonate, dibasic calcium phosphate, and reduced iron (iron filings)-did not affect recoveries of FB1 and FB2 when added to ground corn or corn flakes, although hydrated ferrous sulfate did.