Mycotoxin contamination in graded fractions of maize (Zea mays) in New Zealand

Bulk samples of maize (Zea mays) were collected, initially from a feed mill, and later from a drying plant, to assess the mycotoxin contamination in different graded fractions, and whether current screening practices are effective in reducing the mycotoxin content. At the feed mill, samples of incoming material, screened product, and reject material were collected. At the drying plant, samples were collected from incoming trucks and at three points in the drying process. Samples of reject screened material and plant debris were also collected. All samples were visually assessed for the presence of cob and other plant debris, and all samples were analysed for the presence of the Fusarium mycotoxins, nivalenol (NIV), deoxynivalenol (DON), and zearalenone (ZEN). Selected samples of bulk non-reject maize from both sources were fractionated into component fractions on appearance, namely-good quality maize, poor quality maize comprising small shrivelled and discoloured kernels, split maize, cob, and plant material. These fractions were weighed and analysed for mycotoxin content to determine the contribution of each subfraction to the overall mycotoxin content of the sample. it was found that most samples of non-reject maize consisted mostly of good quality maize (range 50-92%. median 86%), with from 3-24% (median 9%) rated as poor quality, 2-27% (median 4%) as split, and less than 0.4% (median 0.1%) of cob pieces. These lower quantities of lower grade material (especially poor quality and cob pieces) contributed disproportionately to the overall mycotoxin content of the bulk material. For example, in five samples of harvested maize arriving at the drying plant the mean content of poor quality maize was 7%, but it contributed a mean 46% of the NIV, 59% of the DON, and 83% of the ZEN present in the sample. Cob pieces could be highly contaminated, especially by ZEN (284 mg/kg in one sample), therefore the presence of even small amounts was significant to the overall mycotoxin content of a bulk sample. This study shows that the mycotoxin content of bulk maize could be significantly reduced by the application of more efficient screening systems that removed poor quality kernels and cob pieces.