As mentioned above, this emphasizes the need for a standardized preparation

procedure to exclude any influence of the sample preparation procedure on the quality of the protein spectra. Other studies also showed that bacterial protein profiles may be altered by varying growing conditions and extraction solvents. For example, triflouroacetic acid can be used instead of formic acid or different matrix solutions can be applied [23, 38, 39]. To overcome this problem, all leptospiral samples included in this study were cultured and extracted under standardized conditions. Furthermore, as proposed by Welker et al. [40] to ensure the quality of an established protein reference spectra database, each genomospecies was represented by several strains. Beyond this, MSP creation was performed twice, in two self-contained laboratories. Gamma-secretase inhibitor The quality of the established database was confirmed by defined measurements. To exclude any influence of the preparation method sample protein extracts of the reference strains were spotted and measured four times in each laboratory. Reliable species identification for all used strains was successful. Only one field isolate, L. kirschneri serovar Grippotyphosa, did match with the same score value for L. kirschneri and L. interrogans. This indicates that the differentiation of closely related species

by MALDI Biotyper™ is difficult. In this Ralimetinib case, 16S rRNA sequencing revealed the correct species to be L. kirschneri. The close phylogenetic relationship of the two species was confirmed in former sequencing projects [41–43]. Nevertheless, a clear separation of the species L. borgpetersenii and L. interrogans was possible. Studies showed that the genome of the two species L. interrogans and L. borgpetersenii differ in their chromosome size and gene numbers. In comparison to the other two pathogenic species, L. borgpetersenii contains the smallest genome size with 3,931 kb. This pathogenic Etomidate species is not adapted

for the existence in the outer environment [1, 44], which may be due to the loss of genes in the evolutionary process. Differences in the bacterial genome structure followed by the transcription of different proteins in the host and under laboratory conditions can result in the loss of protein peaks in MALDI-TOF MS spectra leading to differences in the proteome profiles. This observation is well-described for other microorganisms such as Brucella spp. [37, 45]. Considering these known leptospiral genomic variations, we hypothesize that it is possible to distinguish lepotspiral strains on the basis of discriminating peaks in their protein profiles. The most critical point for successful subtyping of gram-positive and gram-negative bacteria is the rigorous control of the extraction procedure, as described for Salmonella enterica[46].