A specific sheep polyclonal antiserum was applied to the competitive direct enzyme-linked immunosorbent assay (CD-ELISA) of fumonisins in Fusarium cultures, fresh corn, animal feed, and human foods, and the results were related to fumonisin B1 (FB1) levels determined by liquid chromatography (LC). The limit of detection of FB1 for the CD-ELISA was 0.1 ng/ml. Cross reactivity (defined as [ng/ml FB1 required for 50% inhibition]/[ng/ml of analogue required for 50% inhibition] × 100) for the antiserum in the ELISA was 100, 24, and 30%, for FB1 FB2, and FB3, respectively. FB1 was detected in F. moniliforme cultures grown on corn (≤760 μg/g) but not in those for F. graminearum. Fumonisin estimates were 2.8-fold higher by CD-ELISA than FB1 estimates by LC. Upon using the antiserum in a commercial CD-ELISA format (Veratox), mean recovery of FB1 from spiked corn (0.1 to 3 μg/g)was 85.2 ± 25.3%, whereas the mean recovery by LC was 74.1 ± 6.2%. Analyses of 43 fresh corn, 28 animal feed, and 14 human food samples revealed that estimates of total fumonisins by the commercial ELISA were 2.9-, 2.2-, and 1.3-fold higher than FB1 values determined by LC. For each sample type, the ELISA significantly correlated with LC (r > 0.9; P < 0.05). Taken together, the results suggest that the polyclonal antibody-based CD-ELISA could be a reliable first-tier screening procedure for fumonisins in corn, animal feed, and human food when used in conjunction with LC confirmation. The differences between the two methods suggests the further presence of FB2 and FB3 or structurally related compounds.
† Current address: Department of Food Science and Technology, The University of Lampung, Jln. Soemantri Brojonegoro No.1, Bandar Lampung, Indonesia.