Early detection of spoilage moulds in bread using volatile production patterns and quantitative enzyme assays

There is significant interest in methods for the early detection of quality changes in cereal grains. The development of electronic nose technology in recent years has stimulated interest in the use of characteristic volatiles and odours as a rapid, early indication of deterioration in grain quality. This review details the current status of this area of research. The range of volatiles produced by spoilage fungi in vitro and on grain are described, and the key volatile groups indicative of spoilage are identified. The relationship between current grain quality descriptors and the general classes of off-odours as defined in the literature, e.g. sour, musty, are not very accurate and the possible correlation between these for wheat, maize and other cereals, and volatiles are detailed. Examples of differentiation of spoilage moulds and between grain types using an electronic nose instrument are described. The potential for rapid and remote grain classification and future prospects for the use of such technology as a major descriptor of quality are discussed.

Fungal growth leads to spoilage of food and animal feeds and to formation of mycotoxins and potentially allergenic spores. Fungi produce volatile compounds, during both primary and secondary metabolism, which can be used for detection and identification. Fungal volatiles from mainly Aspergillus, Fusarium, and Penicillium have been characterized with gas chromatography, mass spectrometry, and sensory analysis. Common volatiles are 2-methyl-1-propanol, 3-methyl-1-butanol, 1-octen-3-ol, 3-octanone, 3-methylfuran, ethyl acetate, and the malodorous 2-methyl-isoborneol and geosmin. Volatile sesquiterpenes can be used for taxonomic classification and species identification in Penicillium, as well as to indicate mycotoxin formation in Fusarium and Aspergillus. Developments in sensor technology have led to the construction of "electronic noses" (volatile compound mappers). Exposure of different nonspecific sensors to volatile compounds produces characteristic electrical signals. These are collected by a computer and processed by multivariate statistical methods or in an artificial neural network (ANN). Such systems can grade cereal grain with regard to presence of molds as efficiently as sensory panels evaluating grain odor. Volatile compound mapping can also be used to predict levels of ergosterol and fungal colony-forming units in grain. Further developments should make it possible to detect individual fungal species as well as the degree of mycotoxin contamination of food and animal feeds. Copyright 1999 Academic Press.

Volatile profiles and hydrolytic enzyme production by one non-mycotoxigenic and three mycotoxigenic strains of Fusarium moniliforme and F. proliferatum, grown in vitro for up to 96 h on a grain medium at 25 degrees C/0.95 water activity, were examined for differentiation of isolates. After spore lawn inoculation, measurements were made after 48, 72 and 96 h by sampling the head space above cultures with an electronic nose system using a 14 sensor surface polymer array, and by extraction and quantification of hydrolytic enzymes. There was good reproducibility of volatile patterns between replicates of the same treatment. Principal component analysis indicated that discrimination could be achieved between the uninoculated controls, the non-mycotoxigenic strain and the mycotoxin-producing strains for both species after 48 h. The total and specific activity of three out of seven enzymes (beta-D-glucosidase, alpha-D-galactosidase and N-acetyl-beta-D-glucosaminidase) were found to increase significantly in the non-mycotoxigenic when compared with the toxigenic strains of both species after 72 h. Activities of the others (beta-D-fucosidase, alpha-D-mannosidase, beta-D-xylosidase and N-acetyl-alpha-D-glucosaminidase) were not significantly different between strains. The study has shown for the first time that it is possible to differentiate between mycotoxigenic and non-mycotoxigenic strains of such spoilage fungi based on their volatile production patterns using an electronic nose system. These results have significance in the development of methods for the early detection of toxin-producing spoilage moulds in the food industry.