Development of a starter culture for the production of Gari, a traditional African fermented food

Abstract

Cassava, (Manihot esculenta Crantz), is used for the production of a variety of West African foods and ranks fourth in the list of major crops in developing countries after rice, wheat and maize. Gari is one of the most popular foods produced from cassava. Cassava may contain high levels of linamarin, a cyanogenic glucoside, which in its natural state is toxic to man. Therefore, some processing methods that can enhance the detoxification of cassava and lead to the improvement of the quality and hygienic safety of the food are vitally important for less toxic products to be obtained. Quality, safety and acceptability of traditional fermented foods may be improved through the use of starter cultures. There has been a trend recently to isolate wild-type strains from traditional products for use as starter cultures in food fermentation. A total of 74 bacterial strains and 21 yeast strains were isolated from a cassava mash fermentation process in a rural village in Benin, West Africa. These strains were assessed, together with 26 strains isolated at the CSIR from cassava samples sent from Benin previously, for phenotypic and technological properties. Twenty four presumptive lactic acid bacteria (LAB) were selected for further phenotypic, genotypic and technological characterization during a research visit to the BFE (now Max Rubner Institute of Nutrition and Food). After assessment, the strains VE 20, VE 36, VE 65b, VE 77 and VE 82 were chosen for further study as starter cultures. These L. plantarum strains were chosen on the basis of predominance and possession of suitable technological properties. The investigation of this study was complemented by further, similar studies on further Gari isolates in Germany by the BFE. That study was done independently from this study, but both studies served to select potential starter cultures for cassava fermentation for the production of Gari, as this was the common goal of the project. Thus, a wider final selection of potential starter cultures was decided on at the project level and this selection was further tested in fermentation experiments. A total of 17 strains were grown in optimized media in 2 L fermenters. These strains were freeze-dried and thereafter tested in lab-scale cassava mash fermentation trials. xiii The strains performed well in the small scale bucket fermentations. There was a rapid acidification evidenced by the increase in titratable acidity, ranging from 1.1 to 1.3 % at 24 hours, and 1.3 to 1.6 % at 48 hours. The effect of the starter was obvious in that it lowered the pH much faster and to lower levels than the control. It appeared that both the processing and starter culture addition played a role in the removal of cyanide during processing of the cassava into Gari. This was evident from the lower cyanide values obtained for fermentations that included starter cultures. The study also showed that especially the L. plantarum group strains could be produced as starter cultures at lower costs than compared to L. fermentum, W. paramesenteroides or L. mesenteroides strains. Overall the results of this study were crucial for the project in showing that a starter culture which is easy and economical to produce and which has the desired attributes is a feasible possibility for application in the field.