Effectiveness of density sorting in reducing aflatoxin b1 and fumonisins in maize grain

Abstract

Mycotoxins such as aflatoxins and fumonisins are prevalent contaminants of maize, which is a major staple food in Kenya. Aspergillus flavus, Aspergillus parasiticus and Fusarium verticilloides are the major producers of carcinogenic aflatoxins and fumonisins respectively. Currently there are no effective methods of decontaminating grains and whole consignments have to be destroyed. This study sought to determine the effectiveness of density sorting in reducing aflatoxin B1, fumonisins, Aspergillus spp. and Fusarium spp. populations in maize grains. Samples (n=206) were collected during the 2017 harvest crop from markets in eight counties in Western and Nyanza regions of Kenya. Sample numbers differed across counties ranging from 10-30 per county. All samples were analyzed for mycotoxins using an ELISA assay. Ten samples with more than 50 ppb of aflatoxin B1 and 4 ppm of fumonisins were weighed into 300 g with two replicates and sorted using a density sorter into heavy and light fractions constituting 65-75% and 25-35% of the original weight respectively. Bulk density was determined by filling a container of given weight and volume with kernels and the weights were determined for the heavy and light fractions. Kernel weight for each of the heavy and light fractions was determined by weighing 100 kernels. The effectiveness of density sorting in reducing mycotoxin-producing fungi was determined by isolation from 20 samples of the unsorted and 80 samples of the sorted heavy and light fractions. Finely ground maize flour was serially diluted and plated on PDA and Rose Bengal Modified Dichloran media. Single isolates of Fusarium spp. and Aspergillus spp. colonies were counted after five days and the number of colony forming units determined. Each fraction was analyzed for aflatoxin B1 and fumonisins by ELISA then reduction of the toxins in the heavy fractions determined in comparison to the unsorted samples. The unsorted maize samples had up to 765±0 ppb aflatoxin B1 and 16±0 ppm fumonisins. The Majority (68%) of the samples showed a co-existence of the two toxins with aflatoxin B1 being more prevalent. Bulk density and kernel weights of the fractions were higher in the heavy fractions and lower in the light fractions. Mycotoxin-producing fungi isolated from unsorted and sorted samples were Aspergillus spp., Fusarium spp. and Penicillium spp. Prevalence of Aspergillus flavus was higher in 93% of the samples followed by Penicillium spp. at 85% and Fusarium verticilloides at 67%. Population of Aspergillus flavus and Fusarium verticilloides significantly (p<0.05) varied among the unsorted, heavy and light fractions, with the light fractions exhibiting highest populations and the unsorted grains exhibiting the lowest. There was no significant (p>0.05) reduction in the populations of Aspergillus flavus and Fusarium verticilloides in the heavy fractions. Density sorting did not effectively lower the fungal populations in the heavy fractions though the light fractions had evidently higher populations than the unsorted and the heavy fractions. Density sorting reduced fumonisins in 100% of the samples with an average of 71% reduction and aflatoxin B1 in 50% of the samples while the levels increased in the rest of the samples averaging the percentage change at -12.8%. Bulk density and aflatoxin B1 levels exhibited a strong correlation. Bulk density and fumonisins levels in light fractions had a strong correlation while in the heavy fractions the correlation was weak. Density sorting can be used to reduce fumonisins and aflatoxin B1 effectively in maize grain but had no effect on mycotoxin-producing fungi. The density sorter machine should be improved for large scale use at a commercial level. Key words: Aspergillus, Aflatoxins, Density sorting, Fumonisins, Fusarium, Mycotoxin