Effect of cleaning, milling, and baking on deoxynivalenol in wheat.

Samples of wheat naturally infected by Fusarium graminearum Schwabe were obtained from mills in Oklahoma, Missouri, Kansas, and Minnesota and fields in Nebraska and Kansas in 1982; they were analyzed for deoxynivalenol (DON). The wheat was milled, and DON was found throughout all the milling fractions (bran, shorts, reduction flour, and break flour). The DON recoveries for each mill run ranged from 90 to 98%. These samples, regardless of DON concentration, also gave similar fractional distributions of DON. The greatest (21 ppm [21 micrograms/g]) concentration of DON was found in the bran, and the smallest (1 ppm) was found in the break flour. Cleaning and milling were not effective in removing DON; DON was not destroyed in the bread baked from the naturally contaminated whole wheat flour, but the effect on its concentration in the samples analyzed varied, the reduction ranging from 19 to 69%. The percent reduction found in the cleaned wheat ranged from 6 to 19%. DON concentrations in the following commercially made breads, caraway rye, seedless rye, and pumpernickel, were 45 ppb (ng/g), 39 ppb, and 0 ppb, respectively. The limits of detection by gas chromatography-mass spectrometry and high-pressure liquid chromatography for DON were 0.5 and 10 ng, respectively.