Fourier transform infrared (FT-IR) spectroscopy historically is a powerful tool for the taxonomic classification of bacteria by genus, species, and strain when they are grown under carefully controlled conditions. Relatively few reports have investigated the determination and classification of pathogens such as the National Institute of Allergy and Infectious Diseases (NIAID) Category A Bacillus anthracis spores and cells (BA), Yersinia species, Francisella tularensis (FT), and Category B Brucella species from FT-IR spectra. We investigated the multivariate statistics classification ability of the FT-IR spectra of viable pathogenic and non-pathogenic NIAID Category A and B bacteria. The impact of different growth media, growth time and temperature, rolling circle filter of the data, and wavelength range were investigated for their microorganism differentiation capability. Viability of the bacteria was confirmed by agar plate growth after the FT-IR experimental procedures were performed. Principal component analysis (PCA) was reduced to maps of two PC vectors in order to distill the FT-IR spectral features into manageable, visual presentations. The PCA results of the strains of BA, FT, Brucella, and Yersinia spectra from conditions of varying growth media and culture time were readily separable in two-dimensional (2D) PC plots. FT spectra were separated from those of the three other genera. The BA pathogenic spore strains 1029, LA1, and Ames were clearly differentiated from the rest of the dataset. Yersinia rhodei, Y. enterocolitica, and Y. pestis species were distinctly separated from the remaining dataset and could also be classified by growth media. Different growth media produced distinct subsets in the FT, BA, and Yersinia spp. regions in the 2D PC plots. Various 2D PC plots provided differential degrees of separation with respect to the four viable bacterial genera including the BA sub-categories of pathogenic spores, vegetative cells, and nonpathogenic vegetative cells. This work provided evidence that FT-IR spectroscopy can indeed separate the four major pathogenic bacterial genera of NIAID Category A and B biological threat agents including details according to the growth conditions and statistical parameters.

No abstract

Infrared spectroscopy has been demonstrated as a powerful tool for taxonomic classification of bacteria when the microbes are grown and sampled under carefully controlled conditions. Infrared spectroscopy affords limited information about relative proportions of certain chemical functional groups in whole microbial cells. The objective of this work is to elucidate the ability of infrared spectroscopy to identify and speciate Bacillus spp. regardless of sample history. Spectrometers utilize different scanning methods to collect infrared absorption spectra. We employed three; transmission through a thin film, transmission infrared microscopy, and Attenuated Total Reflection (ATR). Target organisms include Bacillus anthracis, and several near neighbors. Each strain was cultured at 24°C and 35°C on three solid media. Microorganisms were incubated for up to ten days to include vegetative cells, spore formation and mature spores. Triplicate microbe samples were prepared and analyzed according to instrument requirements using the three measurement modes. Triplicate samples of BSL-3 organisms were analyzed only by the thin film transmission method. Spectral data was analyzed using the cluster analysis function of OPUS software. We report that infrared spectrometry is capable of discerning Bacillus spores from vegetative cells and the phylogenic clustering of Bacillus species according to pathogenicity levels via infrared spectral analysis.