Enzyme-linked immunosorbent assay of total aflatoxins B1, B2, and G1 in corn: follow-up collaborative study.

Abstract

A direct competitive enzyme-linked immunosorbent assay screening method for aflatoxins at 20 ng/g in corn was studied by 15 collaborating laboratories. Test samples of corn were extracted by blending with methanol-water (8 + 2). The extracts were filtered and the filtrates were diluted with buffer to a final methanol concentration of < 30%. Each diluted filtrate was applied to a test device containing a filter with immobilized polyclonal antibodies specific to aflatoxins B1, B2, and G1. Aflatoxin B1-peroxidase conjugate was added, the test device was washed with water, and a mixture of hydrogen peroxide and tetramethylbenzidine was added. A test sample was judged to contain > or = 20 ng aflatoxins/g when, after exactly 1 min, no color was observed on the filter; if a blue or gray color developed, the test sample was judged to contain < 20 ng aflatoxins/g. All laboratories correctly identified naturally contaminated corn test samples. Only one false positive was found for controls containing no aflatoxins. The correct responses for positive test samples spiked at levels of 10, 20, and 30 ng aflatoxins/g (the ratio of B1:B2:G1 was 15:1:3) were 67, 97, and 100%, respectively. This method was adopted first action by AOAC INTERNATIONAL as a change in method for 990.34 for screening for aflatoxins B1, B2, and G1 in corn at total aflatoxin concentrations of > or = 20 ng/g.