RESUMO
Rapid identification of non-tuberculous mycobacteria (NTM) species is important in clinical laboratories to stipulate the appropriate therapy and to offer a comprehensive infection control. We applied Fourier-Transform Infrared microspectroscopy to evaluate, whether the most frequent species of NTM can be rapidly and uniformly identified by this method using microcolonies of NTM growing on solid nutrient agar plates. To establish a standardized protocol, the heterogeneity of cell growth within the microcolonies and the reproducibility of measuring the IR spectra from whole mycobacterial microcolonies were first studied. Hierarchical cluster analysis applied to spectra obtained by linear mapping across microcolony imprints from fast- and slow-growing NTM revealed only little spectral variance between the various microcolony zones. In parallel, when repetitive measurements were performed on independently grown whole single microcolonies with diameters of 80 and 140 mum, excellent reproducibility could be achieved, verifying that mycobacterial microcolonies are well suited for FT-IR-based identification. Twenty-eight different and well-defined strains, comprising the most frequent species of NTM isolated in clinical laboratories, were used to create a classification system based on FT-IR spectra from single microcolonies. Hierarchical cluster analysis allowed the assignment of all isolates measured in replicates to their correct species-specific clusters. Additionally, a clear separation of all strains into strain-specific sub-clusters was observed. These results demonstrate the potential of FT-IR microspectroscopy to rapidly differentiate NTM at the species and strain level. The data so far obtained suggest that an extended spectral database, containing more NTM strains and covering a broader biological variance, may provide a practical solution to rapidly identify unknown NTM isolates in routine clinical-microbiological laboratories with the additional possibility to type these microorganisms at the sub-species level.
