Involvement of pectolytic micro‐organisms in coffee fermentation

International Journal of Food Science & Technology (Impact Factor: 1.38). 01/2002; 37(2):191 - 198. DOI: 10.1046/j.1365-2621.2002.00556.x


During the fermentation of Coffea arabica L., the most frequently found pectolytic bacteria were Erwinia herbicola and Klebsiella pneumoniae. These micro-organisms produce pectatelyase which is unable to depolymerize esterified pectins of mucilage without previous de-esterification. Furthermore, the optimal activities are observed at pH 8.5 whereas fermentation conditions are acidic (5.3–3.5). The major lactic acid bacteria, Leuconostoc mesenteroides, do not produce pectolytic enzymes. Only a Lactobacillus brevis strain, rarely isolated with a low frequency, shows a polygalacturonase activity compatible with fermentation conditions. Mucilage decomposition seems to be correlated to acidification and not to enzymatic pectolysis. Inoculation with pectolytic micro-organisms allows microbiological control of the fermentation but does not speed up the process. It would be preferable to use lactic acid bacteria so that the pH remained as close as possible to natural fermentation, where acidification is important. This practice would standardize the coffee fermentation microflora and therefore control the end product quality.

Download full-text


Available from: Sylvie Avallone, Sep 30, 2015
191 Reads
    • "In the literature , only a few studies have been published towards the use of starter cultures for coffee fermentations, although the attempt to control coffee fermentation has existed for over 40 years. A pectolytic yeast (Agate & Bhat, 1966; Avallone et al., 2002) and waste-water of a previous fermentation (Avallone et al., 2002; Butty, 1973; Calle, 1957, 1965) used as inocula were early attempts to utilize starters in coffee fermentations . In more recent studies, Silva et al. (2013) used a multicomponent starter preparation that included pectinolytic yeasts and bacteria. "
    Advances in Fermented Foods and Beverages, Edited by W.H. Holzapfel, 01/2014: chapter Coffee: fermentation and Microbiota; Woodhead Publishing., ISBN: ISBN: 978-1-78242-015-6
  • Source
    • "" Over-fermentation " (beyond that necessary to loosen the mucilage) is generally considered detrimental to coffee quality (Gibson and Butty 1975; Castelein and Verachtert 1981; Lopez and others 1989; Puerta-Quintero 1999, 2001), and its control is usually a component of quality and consistency improvement programs that are aimed at gaining access to specialty markets. Contrary to earlier suggestions (Castelein and Verachtert 1981), recent studies of the fermentation process (Avallone and others 2001a, 2001b, 2002) are interpreted to indicate that the pectin-rich mucilage is degraded neither by endogenous pectolytic enzymes nor by pectolytic bacteria. Rather, it is suggested (Avallone and others 2001a) that physicochemical changes of the carbohydrate matrix, in concert with limited pectolysis, may be occurring. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The recent “crisis” brought about by the collapse of the worldwide commodity coffee market has caused severe economic conditions for coffee producers in developing countries, including those of Central America. As a result, many coffee producers desire to improve the quality and consistency of their product to enter the specialty market. With the ultimate aim of assisting coffee producers in their quality control efforts, this study was designed to determine the feasibility of simple chemical measurements of the fermentation process on remote farms and to assess the potential of these measurements for assisting the producers in control and optimization efforts. Temperature, pH, and the concentrations of glucose, ethanol, and lactic acid were measured throughout the course of 7 coffee mucilage fermentation batches on 4 farms. In each batch, a pattern was observed in which the pH was initially in the range 5.5 to 5.7 and decreased sharply to about 4.6 as fermentation neared completion. Glucose concentration was seen to drop throughout the course of most batches, whereas either ethanol or lactic acid increased sharply near completion. The pH profile may prove useful in predicting the time of fermentation completion and in preventing over-fermentation of coffee mucilage.
    Journal of Food Science 05/2005; 70(5):C321 - C325. DOI:10.1111/j.1365-2621.2005.tb09960.x · 1.70 Impact Factor
  • Source
    • "In another study, it has been reported that yeasts isolated were non-pectinolytic (Avallone et al., 2001). In a recent study, Avallone et al. (2002) reported that the mucilage degradation seems to be due to acidification rather than to microbial pectinolytic enzymes. Cultivation and isolation of yeasts as applied in previous studies can be combined with cultureindependent methods to give a complete picture of the microbial diversity. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Samples of Coffea arabica were collected during the different stages of the fermentation from two production sites in Tanzania. The yeasts community was identified by genotyping using ITS-PCR and sequence analysis of the D1/D2 domain of the 26S rRNA gene. For confirmation, denaturating gradient gel electrophoresis (DGGE) of PCR-amplified 26S rRNA gene was performed to detect yeast directly from coffee samples without cultivation. Yeast counts were in the range 4.0 x 10(4) - 5.0 x 10(7) CFU/g with an increase during fermentation. Three yeasts species were dominant. The predominant yeast found during fermentation and drying was Pichia kluyveri. Pichia anomala was found in high numbers during drying of coffee beans. Hanseniaspora uvarum was the predominant yeast during fermentation but decreased during drying. Kluyveromyces marxianus, Candida pseudointermedia, Issatchenkia orientalis, Pichia ohmeri and Torulaspora delbrueckii occurred in concentrations of 10(3) CFU/g or below in coffee samples. Saccharomyces cerevisiae and Candida xestobii were not isolated by cultivation, but by the DGGE technique. A good agreement was found between the sequence analysis of the D1/D2 domain of the 26S rRNA gene and sequencing of the DGGE bands.
    Yeast 05/2004; 21(7):549-56. DOI:10.1002/yea.1124 · 1.63 Impact Factor
Show more