This article about cacao fermentation in Brazil presents some very interesting and useful data concerning microbial succession, as well as some of the challenges in the experimental design itself. Specifically, I thought worthy of note (1) the growth media’s effect on the quantification of microbial varieties; (2) the strong predominance of a single strain each of lactic acid (LAB) and acetic acid bacteria (ABB) in the fermentation environments; (3) the identification of various enterobacterial strains (of unknown effects); and finally (4) the authors’ conclusion that:
apparently neither the farm nor the fermentation method influences the bacterial species diversity and metabolite course of cocoa bean fermentation processes and hence the quality of fermented dry cocoa beans and chocolates produced thereof, provided that local agricultural and operational practices on the farm, encompassing crop cultivation, fermentation, and post-fermentation handling, are carried out with care (emphasis added).
More details below.
As stated by the authors, the study sought to investigate microbial growth and bacterial species diversity of traditional Brazilian organic cocoa bean box fermentations both “culture-dependently and -independently,” and to analyze the overall metabolism throughout fermentation and evaluate its impact on fermented dry cocoa bean and chocolate quality.
In general, a succession of yeast, enterobacterial, LAB, and AAB activities took place, which was accompanied with a temperature and pH increase as a result of carbohydrate fermentation, citric acid assimilation, and ethanol oxidation. These microorganisms origi- nated from the environment, in particular boxes, shovels, and sacks, which seemed to be contaminated, despite their washing after each use. The potential involvement of Enterobacteriaceae in cocoa bean fermentation has been suggested recently and was confirmed during the present study.
For the culture-dependent part of the study, fresh samples, including swab samples, were plated on various agar media: malt extract agar (MEA), plate count agar (PCA), de Man-Rogosa-Sharpe agar (MRS), deoxycholate mannitol sorbitol agar (DMS), and acetic acid medium agar (AAM). As it turned out, the choice of culture medium had a measurable impact on the reported results, particularly for AAB.
Concerning LAB, 473 isolates were identified. Sequence analysis of selected genes of representative isolates of each cluster showed that Lb. fermentum was by far the most prevalent species during both box fermentations, representing 91.5% of all LAB isolates.
Similarly, if not quite as dramatically, concerning AAB, classification and identification of 287 isolates revealed two main groups, namely A. pasteurianus and Acetobacter senegalensis, representing 73% and 27% of all isolates, respectively. Notably, “[t]he latter species was mostly isolated from the initial phases of both box fermentations.”
Both species [Lb. fermentum and A. pasteurianus] have been isolated from cocoa bean fermentations, often as dominating species, before. Despite their mesophilic character, these species show close adaptation to the microenvironment of the fermenting cocoa pulp-bean mass, allowing their isolation even at high temperatures (>40 C) and low pH (3.9-4.5). This can be ascribed to an adapted substrate catabolism and ethanol, acid, and heat tolerance. In the past, Lb. fermentum has been isolated from maize-based fermented foods, vegetable fermentations, and cassava fermentations, and A. pasteurianus from wine, vinegar, and rice fermentations. Strictly heterofermentative Lb. fermentum (converting citric acid, fermenting glucose, and using fructose as alternative external electron acceptor to convert it into mannitol) and ethanol-oxidizing A. pasteurianus are desirable during cocoa bean fermentation to obtain volatile acetic acid out of carbohydrates and/or citric acid and ethanol, respectively (internal citations omitted – click to embiggen).
Given these facts, the authors observed that “[a] wide LAB species diversity and an onset of AAB growth too early during cocoa bean fermentation results in fermented dry beans with a too acidic flavor, the reason for which has to be sought for in the fermentation practices, as has been shown for Brazilian cocoa bean box fermentations in the past. (emphasis added)” In light of this, they concluded with the suggestion that it would be reasonable to use Lb. fermentum and A. pasteurianus “as starter cultures for homogeneous, fast, and successful controlled fermentation processes in the future.”