Nutritional Composition and Microbial Communities of Two Non-alcoholic Traditional Fermented Beverages from Zambia: A Study of Mabisi and Munkoyo

Abstract

Traditional fermented foods and beverages are common in many countries, including Zambia. While the general (nutritional) benefits of fermented foods are widely recognised, the nutritional composition of most traditional fermented foods is unknown. Furthermore, fermentation is known to add nutritional value to raw materials, mainly by adding B-vitamins and removing anti-nutritional factors. In the case of traditional fermentation, the composition of microbial communities responsible for fermentation varies from producer to producer and this may also be true for the nutritional composition. Here, we characterized the nutrient profile and microbial community composition of two traditional fermented foods: milk-based Mabisi and cereal-based Munkoyo. We found that the two products are different with respect to their nutritional parameters and their microbial compositions. Mabisi was found to have higher nutritional values for crude protein, fat, and carbohydrates than Munkoyo. The microbial community composition was also different for the two products, while both communities were dominated by lactic acid bacteria. Our analyses showed that variations in nutritional composition, defined as the amount of consumption that would contribute to the estimated average requirement (EAR), might be explained by variations in microbial community composition. Consumption of Mabisi appeared to contribute more than Munkoyo to the EAR and its inclusion in food-based recommendations is warranted. Our results show the potential of traditional fermented foods such as Mabisi and Munkoyo to add value to current diets and suggests that variations in microbial composition between specific product samples can result in variations in nutritional composition.

Keywords: Mabisi; Munkoyo; Zambia; lactic acid bacteria; nutrient value; nutritional composition; proximate analysis; traditional fermentation.

Figure 1

Principal Components Analysis (PCA) bi-plot with Optimal Scaling of Mabisi (numbers 1 to 12) and Munkoyo (numbers 13 to 25) samples are indicated as blue circles to determine variation in nutritional parameters within and in between sample types. PC1 explained 65.6% of all variation while PC2 explained 11.8% of all variation giving a total of 77.4% variation explained by the two components.

Figure 2

Relative abundance of bacterial types that constitute the bacterial communities in traditional fermented foods Mabisi and Munkoyo; samples with the y-axis represent relative abundance and the x-axis represents the different samples. Colors in the bars show different Operational Taxonomic Units (OTUs), the legend shows the most likely BLAST identity of that OTU. Multiple OTUs can be identified as the same species.

Figure 3

Comparison of alpha diversity indices of microbial communities in traditional fermented foods Mabisi and Munkoyo samples based on Faith’s phylogenetic diversity (PD) and Chao1. The blue box plots indicate Munkoyo and red box plots indicate Mabisi. (a) shows Faith’s Phylogenetic diversity index of alpha diversity and (b) shows plots for Chao1 index of alpha diversity.