ELECTRONIC NOSE AND FAST GC FOR DETECTION OF VOLATILES FROM ESCHERICHIA COLI O157:H7, ESCHERICHIA COLI AND SALMONELLA IN LETTUCE

Timely and accurate determination of the presence and nature of foodborne bacterial pathogens is very important in food safety. It is desirable to have methods that can quickly discriminate pathogens from non-pathogens whilst classifying correctly different isolates of the same pathogen. In this work we evaluated the potential of an electronic nose (DiagNose^TM) and a fast GC (zNose^TM) to detect and distinguish pathogens (Escherichia coli O157:H7 and Salmonella spp.) from non-pathogens (E. coli) in lettuce. For the cross-species discrimination study, tryptic soy broth supplemented with an amino acid mixture plus D-sorbitol was used. For the within-species study, a similar medium was used and three isolates were evaluated separately for each group of bacteria. All samples were incubated at 37°C for 8 hours. After incubation, sample headspace was analyzed with: (1) DiagNose^TM a metal oxide sensor-based electronic nose (5 min analysis time) and (2) zNose^TM a fast-GC with a 1 m long DB-5 column and a surface acoustic wave sensor (30 s analysis time). Principal component analysis showed that E-nose could discriminate pathogenic Salmonella from E. coli O157:H7 but failed to discriminate Salmonella from non-pathogenic E. coli. Fast GC could discriminate pathogenic E. coli O157:H7, non-pathogenic E. coli, Salmonella and a cocktail of all three bacteria from each other. Also E-nose discriminated each of the pathogenic E. coli O157:H7 strains from control but could not distinguish the pathogenic strains from each other. Similarly, fast GC was able to distinguish each of the non-pathogenic E. coli strains from control. Neither instrument could discriminate any Salmonella isolate from the control.