Skip to main content
ARS Home » Research » Publications at this Location » Publication #169188


item Maragos, Chris

Submitted to: Association Official Analytical Chemists Annual Intrl Meeting & Exposition
Publication Type: Abstract Only
Publication Acceptance Date: 9/23/2004
Publication Date: 1/6/2005
Citation: Trucksess, M.W., Maragos, C.M. 2005. Mycotoxins - fumonisins [abstract]. Association Official Analytical Chemists Annual Intrl Meeting & Exposition. 88(1):314-324.

Interpretive Summary:

Technical Abstract: Fumonisin was reported to be produced from F. verticillioides in edible pine nuts. Residues of FB1 were also isolated from the tissues of pigs receiving an oral dose of 100 mg FB1/animal/day for 5-11 days. The highest concentrations were in the spleen (854 ± 2212 µg/kg), kidney (833 ± 1329 µg/kg), liver (231 ± 163 µg/kg), and lung (170 ± 311 µg/kg). The muscle also contained FB1, at an average level of 26 ± 41 µg/kg. A means of assessing human exposure to fumonisins through the analysis of hair was reported. FB1 was found at mean levels from 23 to 33 µg/kg hair and FB2 was found at mean levels from 5 to 11 µg/kg hair, depending upon the region from which they were collected. An in vitro model of mammalian toxicity indicated that N-acylation enhanced toxicity of AP1, but not FB1. Under appropriate conditions of heat and humidity, such as during certain types of cooking, fumonisins may be degraded. Cornmeal that was spiked with F. verticillioides culture material and then used to prepare baked cornbread, pan-fried corncakes, and deep-fried fritters was not significantly less toxic to rats after such cooking than before. FB1 was shown to react with glucose, methyl alpha-D-glucopyranoside, and the amino acid derivatives N-alpha-acetyl-L-lysine methyl ester and BOC-L-cysteine methyl ester. FB1 formed a stable reaction product through the tricarballylic acid (TCA) side chains after heating with methyl alpha-D-glucopyranoside. This suggests that fumonisins may bind to polysaccharides via the TCA side chains under certain conditions. Recovery of 14C-labeled FB1 from spiked cornmeal was reduced by 43% after it was used to prepare muffins. However, more 14C was eluted from an immunoaffinity column than could be accounted for as FB1, suggesting additional 14C-labeled products, related to fumonisins and recognized by the antibodies, may have been present. Total fumonisins were reduced in tortillas prepared using a traditional nixtamalization method. Fumonisins were also reduced during the baking of bread, and on boiling contaminated corn in water. A real-time PCR method was developed for detection of fumonisin-producing Fusarium based upon primers for the biosynthetic gene FUM1. The detection range was from 5 pg to 5 ng of genomic DNA per reaction. Five corn samples that tested positive for F. verticillioides by traditional microbiological methods were also positive for the FUM1 product. A PCR method based upon FUM5 was also developed for detection of fumonisin producing strains. LC with fluorescence detection of the o-phalaldehyde (OPA) derivative remains a popular method. Naphthalene 2,3-dicarboxaldehyde was used as a derivatization reagent in the detection of FB1, FB2, and FB3 in maize silage. Fumonisins were isolated from the silage using affinity columns. Of 89 samples fumonisins were found in 97%. The mean FB1, FB2, and FB3 content of positive samples were 615, 93, and 51 µg/kg respectively, suggesting fumonisins may be a frequent low level contaminant of corn silage. The extraction of fumonisins from corn-based infant foods was investigated. The best extraction solvent examined was 70% aqueous methanol at pH 4, whereas an alkaline methanolic solution at pH 9.2 was poor. An array biosensor was constructed to detect multiple analytes, including staphylococcal enterotoxin B, ricin, cholera toxin, botulinum toxins, trinitrotoluene, and fumonisins. Fluorescence polarization was combined with immunoassay to detect fumonisins, DON, and aflatoxins. A flow injection liposome immunoassay (FILIA) was used to detect FB1 in corn-based food and animal feeds, and results were compared with those of the LC method. The FILIA method detected as little as 0.1 ng FB1 in as little as 11 min. There was good correlation between the FILIA results and the LC (r2 = 0.945). A SPR biosensor for the simultaneous detection of 4 myc