Combination of DNA barcoding, targeted metabolite profiling and multispectral imaging to identify mold species and metabolites in sliced bread

Abstract

Mould is a ubiquitous threat to safe food. In this work, a combination of three reliable methods to identify mould present in food at the species level is presented, to estimate the safety of the apparently non-moldy slices of bread and therefore contribute to the prevention of food waste. In overstored sliced bread, three mould species on the first slice of the stack (Chaetomium globosum, Penicillium chrysogenum and Scopulariopsis brevicaulis) were identified by DNA barcoding. The obtained sequences of the PCR products matched 100% in BLAST alignments with an e-value of 0. Targeted metabolite profiling confirmed the presence of the mycotoxin producers Chaetomium globosum and Penicillium chrysogenum. Known mycotoxins from these moulds (chaetoglobosin A and meleangrin, respectively) were quantitated (8.2 µg/g chaetoglobosin A and 2.2 µg/g meleangrin) in bread spot A. The second slice was less infected, which was revealed by only two mouldy spots, the lower amounts of determined mould metabolites and by normalized canonical discriminant analysis (nCDA) after multispectral imaging. On the third slice of the stack, no mould traces were detected by either metabolite profiling, or comparative multispectral imaging. These results suggest that it might be safe to consume bread slices in lower layers of bread stacks, even if mould growth is visible on the outer slices.