Michelle Czaja (Dortmund / DE; Biberach / DE), Stefanie N. Richter (Bremen / DE), Ilka Davina Nix (Bremen / DE), Katrin Sparbier (Bremen / DE), Boris Oberheitmann (Bremen / DE), Markus Kostrzewa (Bremen / DE), Arthur B. Pranada (Dortmund / DE)
Introduction: Precise species identification is crucial, especially for infection control. Due to its accuracy, MALDI-TOF MS has become a standard identification method in many microbiology laboratories. Nevertheless, for species with very similar proteomic spectra differentiation capability is limited, e.g. between Acinetobacter baumannii and A. nosocomialis. Using MALDI in negative ion mode with lipids as target might overcome this limitation.
Goals: Here, we investigated lipid extracts for the discrimination of A. baumannii from A. nosocomialis using a MALDI-TOF MS-based rapid workflow in negative ion mode.
Materials & Methods: Heat-inactivated isolates of genetically characterized A. baumannii (n=30) and A. nosocomialis (n=25) including each type strain, were processed with the MBT Lipid Xtract Kit (Bruker Daltonics, Bremen, Germany) MALDI spectra of the obtained lipid extracts were acquired with an MBT Sirius mass spectrometer (Bruker Daltonics) in negative ion mode and analysed using FlexAnalysis software.
Results: The lipid spectra of A. baumannii and A. nosocomialis reproducibly showed each specific peaks at m/z 1728, 1881 and 1910 with different intensities, corresponding to lipid A molecules of the outer membrane. A. baumannii showed high intensities for the two peaks at m/z 1728 and 1910 while A. nosocomialis showed high intensities at m/z 1881 instead. Three specific peaks at m/z 2004, 2018 and 2032 were only present in A. nosocomialis spectra, which could be used for species discrimination (Fig. 1).
In total, 55 spectra were analyzed. The discriminating peaks showed a consistent appearance amongst the different strains of both species. The presence of the peaks at m/z 2004, 2018 and 2032 and simultaneously high intensities of the peak at m/z 1881 discriminated A. nosocomialis from A. baumannii. In total 54/55 (98%) strains were correctly identified using these specific lipid markers.
Summary: Differentiating lipid markers could be used in a decision tree analysis to discriminate the two species A. baumannii and A. nosocomialis reliably and quickly. Further studies will have to be carried out to establish an algorithm for clinical diagnostics.