Abstract:
The evaluation of bacterial viability is an important process in the prevention of foodborne illnesses, evaluation of antimicrobial e cacy, and it is a routine laboratory task in microbial research. The standard method of bacterial enumeration is via the plate count method, which requires counting the number of colonies grown on solid growth media after a period of incubation. The plate count process often requires at least two days; it is not suitable for non-culturable cells and cannot detect dead cells. E cient, culture-independent detection of live and dead bacteria can be achieved using di erentially staining uorescent dyes SYTO 9 and propidium iodide (PI). Fluorescence microscopy and ow cytometry (FCM) has been used extensively for detection of these live/dead cell uorescence signals. However, wide-spread use of these methods is limited by the cost and size of instrumentation and operational requirement of skilled technicians. We are developing a convenient, cost-e ective and portable bre-based uorometer (the optrode) for the rapid and accurate measurement of uorescence signals to determine the viability of bacterial samples. Fluorescence emissions from mixtures of live and dead bacteria stained using SYTO 9 and PI were measured using the optrode. Optrode-measured uorescence spectra were calibrated using measurements of live and dead bacterial percentage and concentration obtained from a bead-based FCM that we optimised and characterised. We developed analysis processes that can predict the proportion and concentration of live and dead bacteria in a sample using the optrode-measured uorescence spectra. The optimised optrode method was able to predict the percentage of live bacteria in 108 bacteria/mL samples between c. 7 and 100% live, and in 107 bacteria/mL samples between c. 7 and 73% live. Furthermore, the optrode can quantify live bacteria from 108 down to 106:2 bacteria/ mL and showed the potential to detect as low as 105:7 bacteria/mL. Meanwhile, enumeration of dead bacteria was achieved between 108 and 107 bacteria/mL. Finally, the optrode method was applied to determine the e cacy of antibiotic treatments: polymyxin B, ampicillin, chloramphenicol, cipro oxacin or untreated. We obtained predictions of live bacterial percentage comparable to the reference FCM measurements for all ve treatment groups. Predictions of live bacterial concentration that was comparable to FCM measurements were also obtained for three treatment groups: chloramphenicol, cipro oxacin and untreated. The near-real time information provided by the optrode about live bacterial percentage and concentration in antibiotic-treated samples is useful in enabling the optimal use of antibiotics. The optrode is highly versatile and can be adapted for the monitoring of bacterial viability in other applications such as food safety testing.