No abstract

No abstract

Fourier transform‐infrared spectroscopy (FT‐IR) methods enable chemical discrimination of intact bacterial cells and produce complex whole‐organism biochemical fingerprints (spectra), which are reproducible and distinct for different bacteria. Numerous researchers indicate that there is great potential for using FT‐IR methods in combination with multivariate statistics (chemometrics) to detect and identify bacteria in water, culture media and foods. This article presents a review of the FT‐IR techniques, sample preparation procedures and experimental conditions used in these studies, as well as advantages, disadvantages and challenges that remain for the development of FT‐IR detection methods.

No abstract

Fourier-transform infrared spectroscopy has been successfully used as a nondestructive method for identifying, distinguishing, and classifying pathogens. In this study, a less time-consuming Fourier-transform infrared procedure was developed to identify Escherichia coli O157:H7 and Salmonella Typhimurium. Samples containing 10(9) CFU/ml were prepared in tryptic soy broth and then serially diluted (up to eight times) to obtain bacterial solutions of 10(9) to 10 CFU/ml. These dilutions were incubated at 37 degrees C for 6 h, samples were filtered through a Metricel filter hourly (for 0 to 6 h), and spectra were obtained using a ZnSe contact attenuated total reflectance accessory on a Continu mum infrared microscope. Midinfrared spectra (4,000 to 700 cm(-1)) of Salmonella Typhimurium and E. coli O157:H7 were generated, and peak areas in the region of 1,589 to 1,493 cm(-1) were used to detect the pathogens. Initially, detection limits were between 10(6) and 10(7) CFU/ml without preenrichment, and samples starting with 500 CFU/ml were detectable following incubation for 6 h, when counts reached at least 10(6) CFU/ ml. Compared with results of previously published studies in which Fourier-transform infrared spectroscopy was used to identify select pathogens, this method is more rapid and less expensive for practical large-scale sample analysis.