Rapid tool for identification of bacterial strains using Fourier transform infrared spectroscopy on genomic DNA.

Abstract

We developed a new rapid and reliable method for identifying bacteria using a combination of Fourier transform infrared (FT-IR) spectroscopy of bacterial genomic DNA and multivariate analysis. FT-IR spectra of genomic DNA from four type strains of Pseudomonas spp., three type strains of Escherichia spp. and two type strains of Bacillus spp. were analysed in the 4000-400cm-1 region. Spectral differences were found in the frequency regions of N-H stretching (amide I), C=O stretching vibrations (amide II) and PO2 - ionized asymmetric and symmetric stretching. Partial least squares discriminant analysis of the FT-IR spectra showed that the microbial strains could be discriminated by hierarchical clustering analysis. FT-IR spectral analysis of bacterial genomic DNA has potential for the rapid identification of bacteria at the genus and species levels. This study reports a new bacterial identification method using multivariate analysis of FT-IR spectra of bacterial genomic DNA.