Evaluation of enzyme-linked immunosorbent assay for analysis of beer for fumonisins

A recently developed sensitive indirect competitive enzyme-linked immunosorbent assay (ELISA) was applied to the determination of fumonisins in beer. Intra-assay and inter-assay recoveries averaged 98.7-102.8% at added fumonisin B1 (FB1) levels of 0.5-50 ng/ml beer, and coefficients of variation were 2.8-4.4 and 4.7-8.6%, respectively. Cross-reactivity of fumonisin B2 (FB2) compared with FB1 averaged 67% in beer. Two experiments were carried out to compare ELISA with liquid chromatography (LC) for determination of fumonisins in beer. In the first experiment, 19 samples (five previously known positive, nine other samples and five spiked samples) were passed through commercial immunoaffinity columns (ICs) and analysed by LC before conducting blind ELISA determinations on the extracts and beers directly. The known positive beers and extracts were used as blind duplicates. The second comparative experiment screened 46 beer samples by ELISA and then 22 positive and three of the negative samples were analysed by LC; the highest level found was 64.3 ng total fumonisins/ml measured by LC (24.7 ng/ml by ELISA). Regression analyses showed good correlation between ELISA and LC in the first experiment but low level interferences (equivalent to up to 5.35 ng fumonisin/ml) were observed by ELISA in the IC extracts. Five of nine beers negative by LC showed < 1 ng/ml ELISA responses on direct beer analysis. The second comparative experiment indicated underestimation by ELISA. However, there were two samples which tested positive by ELISA (0.2 ng/ml) but were found negative by LC (results close to the detection limits of both methods, which were 0.1 or 0.2 ng/ml by ELISA and 0.1-0.15 ng each fumonisin/ml by LC). There were no false negatives. It is concluded that ELISA has considerable value in rapid screening of beer directly for fumonisins.