Immunoassays of selected mycotoxins in hay, silage and mixed feed

Immunoassays, including radioimmunoassay (RIA) and direct competitive enzym e-linked immunosorbent assays (dc-ELISAs) were used for the analysis of the mycotoxins, produced by Alterneria alternata, AAL TA toxin (AAL), cyclopia zonic acid (CPA), deoxynivalenol (DON), fumonisin B1 (FmB1), PR toxin (PRT) and zearalenone (ZEA) in hay, silage and mixed feed. In the analysis, 10 g well-ground feed samples in 100 ml of 84% acetonitrile in water were shake n in an orbital shaker for 1 h. After filtration, the filtrates were dilute d with buffer and directly subjected to dc-ELISA for AAL toxin, CPA, FmB1, PRT and ZEA. For total DON, including 15-acetyldeoxynivalenol (15-AcDON) an d 3-acetyldeoxynivalenol (3-AcDON), the extracts were air-derived, acetylat ed, and subjected to both RIA and dc-ELISA using the Ridascreen DON enzyme immunoassay (EIA) kit. The average recoveries of toxins added to the hay at 0.3 and 1.0 ppm levels were found to be 77.5, 111, 95.8, 85.7, 95.5 and 13 8% for AAL toxin, DON, CPA, FmB1, PRT and ZEA, respectively. When the sampl e extracts were subjected to a C-18 reversed-phase cartridge treatment prio r to dc-ELISA, the recovery of ZEA at 0.1 and 1.0 ppm levels was found to b e 111 and 75%, respectively. Analysis of 63 samples (25 hay and 38 others, including corn silage and mixed feed) showed that the frequencies of AAL to xin, DON, CPA, FmB1, PRT and ZEA in the Seed were 96.8, 100, 87.3, 36.5, 76 .2, and 3.2%, respectively. The over-all average concentrations of toxins i n the feed samples, ex-eluding the nondetectable, were 0.56, 0.73, 0.34, 0. 28, 0.13 and 0.22 mu g g(-1) (ppm) for AAL toxin, DON, CPA, FmB1, PRT and Z EA, respectively. For hay alone, the frequencies were 100% (average level o f 0.72 ppm), 100% (average level of 0.61 ppm), 80% (0.39 ppm), 52% (0.12 pp m) and 80% (0.15 ppm) for AAL, DON, CPA, FmB1 and PRT, respectively. ZEA wa s not found in the hay samples. Lineal I-egression analysis between the res ults of DON in 30 feed samples analyzed simultaneously with RIA and EIA sho wed agreement, with a correlation coefficient of 0.80 and a slope of 1.34.