Differentiation of closely related fungi by electronic nose analysis

In this work the potential of electronic nose analysis for differentiation of closely related fungi has been described. A total of 20 isolates of the cheese-associated species Geotrichum candidum, Penicillium camemberti, P. nordicum, and P. roqueforti and its closely related species P. paneum, P. carneum as well as the noncheese-associated P. expansum have been investigated by electronic nose, GC-MS, and LC-MS analysis. The isolates were inoculated on yeast extract sucrose agar in 20-mL headspace flasks and electronic nose analysis was performed daily for a 7-d period. To assess which volatile metabolites the electronic nose potentially responded to, volatile metabolites were collected by diffusive sampling overnight onto tubes containing Tenax TA, between the 7th and 8th day of incubation. Volatiles were analyzed by gas chromatography coupled to mass spectrometry and the results indicated that mainly alcohols (ethanol, 2-methyl-1-propanol, and 3-methyl-1-butanol) and ketones (acetone, 2-butanone, and 2-pentanone) were produced at this stage. The volatile metabolite profile proved to be species specific. Nonvolatile metabolites were collected on the 8th day of incubation and mycotoxin analysis was performed by high pressure liquid chromatography coupled to a diode array detector and a time of flight mass spectrometer. Several mycotoxins were detected in samples from the species P. nordicum, P. roqueforti, P. paneum, P. carneum, and P. expansum. Differentiation of closely related mycotoxin producing fungi incubated on yeast extract sucrose agar has been achieved, indicating that there is a potential for predicting production of mycotoxins on food and feedstuffs by electronic nose analysis.