Determination of Fumonisins B1 and B2 in Corn and Corn Flakes by Liquid Chromatography with Immunoaffinity Column Cleanup: Collaborative Study

Abstract

A liquid chromatographic (LC) method for the determination of fumonisins B1 (FB1) and B2 (FB2) in corn and corn flakes was collaboratively studied by 23 laboratories, which analyzed 5 blind duplicate pairs of each matrix to establish the accuracy, repeatability, and reproducibility characteristics of the method. Fumonisin levels in the corn ranged from <0.05 (blank) to 1.41 μg/g for FB1 and from <0.05 to 0.56 μg/g for FB2, whereas in the corn flakes they ranged from <0.05 to 1.05 μg/g for FB1 and from <0.05 to 0.46 μg/g for FB2. The method involved double extraction with acetonitrile–methanol–water (25 + 25 + 50), cleanup through an immunoaffinity column, and LC determination of the fumonisins after derivatization with o-phthaldialdehyde. Relative standard deviations for the within-laboratory repeatability (RSDr) of the corn analyses ranged from 19 to 24% for FB1 and from 19 to 27% for FB2; for the corn flakes analyses, RSDr ranged from 9 to 21% for FB1 and from 8 to 22% for FB2. Relative standard deviations for the between-laboratories reproducibility (RSDR) of the corn analyses ranged from 22 to 28% for FB1 and from 22 to 30% for the FB2; for corn flakes analyses, RSDR ranged from 27 to 32% for FB1 and from 26 to 35% for FB2. Mean recoveries of FB1 and FB2 from corn spiked with FB1 at 0.80 μg/g and with FB2 at 0.40 μg/g were 76 and 72%, respectively; for corn flakes spiked at the same levels recoveries were 110 and 97% for FB1 and FB2, respectively. HORRAT ratios for the analyses of corn ranged from 1.44 to 1.53 for FB1 and from 0.96 to 1.48 for FB2, whereas for corn flakes they ranged from 1.60 to 1.82 for FB1 and from 1.39 to 1.68 for FB2.

Development of a liquid chromatography/tandem mass spectrometry method for the quantification of fumonisin B1, B2 and B3 in cornflakes

“…Mean recoveries (n = 12), irrespective of the fortifying level for FB 1 , FB 2 and FB 3 , were 84 ± 10, 78 ± 7 and 85 ± 9%, respectively. The extraction mixture ACN/MeOH/H 2 O (25:25:50, v/v/v) gave slightly inferior recovery yields, although results were comparable with those obtained in a collaborative study by Visconti et al 34. Against expectation, when acidified ACN/MeOH/H 2 O (25:25:50, v/v/v; pH 4) was used as the extraction solvent, the recoveries of FB 1 , FB 2 and FB 3 fell to 44 ± 8 for FB 1 (n = 3), 41 ± 19 (n = 3) for FB 2 and 72 ± 7 (n = 3) for FB 3 .…”

“…Cornflakes were ground using a grinder (type ZM1; Retsch GmbH & Co.KG, Germany), thoroughly mixed and stored at room temperature prior to analysis. FB 1 , FB 2 and FB 3 were extracted from the cornflakes matrix using a method derived from that of Visconti et al 34. The following extraction solvents were evaluated for their recovery yields: ACN/MeOH/H 2 O (25:25:50, v/v/v), ACN/MeOH/H 2 O (25:25:50 v/v/v) adjusted to pH 4 with 0.1 M HCl, and MeOH/H 2 O (70:30, v/v) adjusted to pH 4 with 0.1 M HCl.…”

“…However, this method showed low recoveries and high variability when applied to more complex maize‐based food matrices, such as cornflakes, muffins and other processed maize products 30, 31. Other reported approaches to overcome matrix‐related extractability problems include the use of extraction solvents such as acidified aqueous methanol,32, 33 acetonitrile/water (1:1, v/v)30 or acetonitrile/methanol/water (1:1:1, v/v/v),34–36 elevated extraction temperature,37 and base hydrolysis of food products 19. For cleanup, it has been proved that the use of immunoaffinity columns for the purification of cornflakes, muffins and infant formula extracts was more effective than purification by SAX columns 31–33…”

Exposure, Occurrence, and Chemistry of Fumonisins and their Cryptic Derivatives

“…Because fumonisins neither have UV‐absorbing chromophores, nor do they exhibit fluorescence, they have to be derivatized prior to HPLC (Visconti, Solfrizzo, & De Girolamo, ). For example, the method adopted by the Association Of Analytical Communities (AOAC) for corn analysis includes a pre‐column derivatization step resulting in fluorescent fumonisin derivatives which can be detected using liquid chromatography and a fluorescence detector (HPLC‐FLD; Visconti et al., ). This method consists of sample extraction with acetonitrile–methanol–water, immunoaffinity column clean‐up, and is followed by derivatization using o ‐phthalaldehyde and LC analysis (Visconti et al., ).…”

“…For example, the method adopted by the Association Of Analytical Communities (AOAC) for corn analysis includes a pre‐column derivatization step resulting in fluorescent fumonisin derivatives which can be detected using liquid chromatography and a fluorescence detector (HPLC‐FLD; Visconti et al., ). This method consists of sample extraction with acetonitrile–methanol–water, immunoaffinity column clean‐up, and is followed by derivatization using o ‐phthalaldehyde and LC analysis (Visconti et al., ). However, this method is associated with drawbacks having to do with the stability of the derivatives, and the precise determination of toxin contents might be disturbed by auto‐fluorescence of food matrices (Joint FAO/WHO Expert Committee on Food Additives, ).…”