(36b) Functional Role of Bacteria Involved in Cocoa Fermentation Processes According to a Metabolic Prediction Using 16S rRNA Reads

The fermentation of cocoa beans is a natural and spontaneous process where a succession of bacteria and yeasts transforms the plant material. The coloration, acidity, size and the content of flavor precursor metabolites, such as free amino acids, organic acids, and volatile compounds, define the quality of the cocoa bean. Previous studies have linked the presence and dynamics of microbial groups with the quality of the fermented cocoa bean. The microbial groups present in the cocoa fermentation are Lactic Acid Bacteria (LAB) and Acid Acetic Bacteria (AAB); the bacteria succession is related to the physicochemical characteristics of the ecosystem (temperature, pH, and availability of nutrients). The presence of bacteria and yeast has been studied with culture-dependent methods such as DGGE, although few studies use Next Generation Sequencing technologies to characterize bacterial communities in the fermentation process. Furthermore, a revision of the metabolic pathways during the process of cocoa bean transformation has not yet been carried out.

One approach for the detection of bacteria into system is the generation of 16S rRNA sequences. With this method, it is possible to know the bacterial groups and their relative abundance during cocoa bean fermentation. Likewise, from these data, it is possible to make a prediction of gene families into the system. In this work, we propose to recognize the changes in the abundance of genes associated with metabolic pathways in different stages o f the fermentation processes. For this, a cacao farm was selected in three agroecological regions of Colombia (Montaña Santandereana, Zona Marginal Baja Cafetera, and Valles Interandinos Secos) in two seasons of 2016. Cocoa samples were collected in the upper and middle zone of a fermenter box, during fermentation process every 12 hours until the end of the process. After sequencing with Illumina MiSeq of 16S rRNA reads, the detection of taxonomic groups was performed and PICRUST was used to predict the number and abundance of gene families for each fermentation time point. Subsequently, an analysis of the metabolic pathways to which the gene families are associated was carried out using KEGG database. This methodology is an approximation to an enrichment analysis of metabolic pathways.

Lactobacillus cacaonum, Lactobacillus plantarum, and Acetobacter pasteurianus were detected in this study and selected as starter cultures to obtain high quality cocoa. Other groups were Erwinia tasmaniensis, Komagataeibacter oboediens, and Klebsiella quasipneumoniae. We obtained 3.331 gene families of which only 19,6% (654 KEGG orthology) are associated with a metabolic pathway, in greater numbers to Global and overview maps, Metabolism of cofactors and vitamins and Amino acid metabolism. On the other hand, 2.454 reactions were found in the KEGG database. The gene families in the metabolic pathways were associated to Glycolysis, Amino acid biosynthesis and, the flavonoid pathway. No difference were detected between the gene familes found into the fermentative processes of the agro-ecological regions. There are some pathways such as Valine, leucine and isoleucine degradation, which presented a greater proportion of gene families from the 48th hour until the end of fermentation process in the three agroecological regions. This type of analysis allows to detect and quantify genes related to Carbohydrates and Energy metabolism of bacteria functional groups during a fermentation process. Metabolic pathways as Flavonoids biosynthesis are not abundantly found, this show us that this method do not have high precision to study other metabolic pathways of interest related to flavour of cocoa fermented bean.