Rapid detection and identification of Pseudomonas aeruginosa and Escherichia coli as pure and mixed cultures in bottled drinking water using Fourier transform infrared spectroscopy and multivariate analysis

Fourier transform infrared (FT-IR) spectroscopy and multivariate analysis were used to identify Pseudomonas aeruginosa and Escherichia coli ATCC 25922 inoculated into bottled drinking water. Three inoculation treatments were examined: (i) E. coli ATCC 25922 (N = 3), (ii) P. aeruginosa(N = 3), and (iii) a 1:1 (v:v) mixed culture of both P. aeruginosa and E. coli ATCC 25922 (N = 3). The control treatment was noninoculated drinking water (N = 3). Second derivative transformation and loadings plots over the range of 1800-900 cm(-1) indicate variations in the following bacterial constituents: amide I band ca. 1650 cm(-1), amide II band ca. 1540 cm(-1), phosphodiester backbone of nucleic acids ca. 1242 and 1080 cm(-1), and polysaccharide compounds ca. 1050-950 cm(-1). Cells with the different treatments were clearly segregated from a mean centered principal component analysis. By using soft independent modeling of class analogy analysis, spectra from a given treatment could be correctly classified 83-88% of the time. These results suggest that FT-IR spectroscopy can determine whether a pure culture is present, in addition to confirming that this method can discriminate between closely related bacteria based on differences in biochemical and phenotypic characteristics that can be detected in this spectral region.