Abstract
Trichothecene mycotoxins were extracted from barley with methanol, and the extract was evaporated to dryness. Water was added, and the aqueous extract was applied to a DEAE-Sephadex A-25 column and eluted with 100ml of methanol-water (1:6). The eluates were further cleaned up on a Florisil column, then trichothecene was determined by FID-GC after trimethylsilylation. The recoveries (average) were 88% for deoxynivalenol (DN), 95% for nivalenol (N), 97% for fusarenon-X (F-X), 93% for diacetoxyscirpenol (DS), and 82% for T-2 toxin (T-2) from barley to which trichothecenes had been added at the 0.2-2.0ppm level. The detection limits were 0.04ppm for DN, 0.05ppm for N and F-X, 0.1ppm for DS, and 0.2ppm for T-2. This method was satisfactory for the quantitative and reproducible determination of trichothecenes in cereals.
By this method, trichothecenes in the range of trace-11.44ppm of nivalenol, and trace-1.54ppm of deoxynivalenol have been detected in freshly harvested barley in Ishikawa, northwest Japan, from 1977 to 1982. The TMS-derivatives of nivalenol and deoxynivalenol in naturally contaminated barley could also be identified by GC-MS.
Representative fungi, Fusarium graminearum was predominantly isolated from same barley. The infection frequency of Fusarium were 2-60%.
These data provide conclusive evidence for the natural contamination of trichothecenes and for the infection of field barley of the Northwest Japan by different strains of F. graminearum which produce either nivalenol or deoxynivalenol.
