Free and hidden fumonisins in Brazilian raw maize samples

- Brazilian maize samples were found to contain high levels of hidden fumonisins.
- Hidden fumonisins increased in samples highly contaminated with free fumonisins.
- Hidden fumonisins were detected in raw maize samples that lacked free fumonisins.
- Levels of total and free fumonisins showed a strong positive correlation.

Fumonisins are secondary metabolites produced primarily by fungi strains that belong to the genera Fusarium and Alternaria, which have been shown to be highly prevalent in maize crops. Some authors have documented the presence of hidden forms of fumonisins occurring in raw maize. This purpose of this study was to determine the occurrence of free and hidden fumonisins in raw maize. The concentrations of fumonisins in 72 naturally contaminated maize samples were analyzed using liquid chromatography coupled to mass spectrometry. The performance parameters of the method to determine free fumonisins forms (FB1 and FB2) and hydrolyzed fumonisins forms (HFB1 and HFB2) were evaluated using the standards from the Commission of the European Communities (Commission, 2002). The analytical methods employed fell within the established guidelines. The amount of total fumonisins measured based on the hydrolyzed forms (HFB1 + HFB2) was 1.5-3.8 times greater than the amount of free fumonisins (FB1 + FB2). The concentration of hidden fumonisins was calculated by subtracting the levels of free fumonisins from the total fumonisin levels. The levels of hidden fumonisins were calculated to be 0.5-2.0 times greater than the level of free fumonisins. A strong positive correlation (R = 0.97) was observed between free fumonisins (FB1 + FB2) and. total fumonisins (HFB1 + HFB2). Based on this correlation, a predictive model was generated to estimate the total fumonisin level based on the measured/reported free fumonisin concentration. These results show that the risk of exposure to fumonisins is likely underestimated if only free fumonisins are considered. However, the predictive model could be a novel approach to estimating the total amount of fumonisins in maize samples without needing to perform expensive and time-consuming analytical methods.