MALDI-ToF spectrometry for the rapid identification of Burkholderia cepacia Complex species including the novel Taxon K, recovered from patients with cystic fibrosis

ALEJANDRO MIÑÁN; PETER LASCH; ALEJANDRA BOSCH; MAREN STAEMMLER; JOSÉ DEGROSSI; DIEGO NOSEDA; PATRICIA MONTANARO; MIGUEL D´AQUINO; OSVALDO YANTORNO; DIETER NAUMANN

San Pablo, Brasil

Workshop; Spec2008. Shedding Light on Disease; 2008

Matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-ToF MS) is a useful technique for the identification of bacteria on the basis of their characteristic protein mass spectrum fingerprint (1). This new technology has emerged as a prompt identification methodology of microorganisms starting either from whole cells, cell lysates or bacterial extracts. The rapid and accurate diagnosis of respiratory infections in cystic fibrosis (CF) patients is critical not only in the early implementation of appropriate antimicrobial therapy and infection control measures, but also in epidemiological surveys. While the most common organisms isolated from sputum samples are Pseudomonas aeruginosa, Staphylococcus aureus, and Haemophilus influenzae, other non-fermenting (NF) Gram negative rods as Burkholderia cepacia complex (BCC) bacteria, Stenotrophomonas maltophilia, Ralstonia pickettii, Acinetobacter spp., and Achromobacter spp. can colonize CF patient lungs´. Particularly the BCC is an important group of organisms whose complex taxonomy has evolved dramatically in the last years. Although BCC comprised 9 distinct species (genetic types or genomovars), recently, genetic methods such as 16 S rRNA, recA gene sequence analysis and multilocus sequence typing (MLST) (2,3), have included 1 additional species to the complex and have described 5 novel BCC species and a  group of bacteria  referred to as Taxon K, which comprises at least  two species (4,5).  In routine clinical laboratories the identification of BCC members and their differentiation from the other NF rods is still a challenge due to misidentifications that often occur. The aim of the present study was to apply MALDI-ToF MS and multivariate analysis for the rapid identification of whole-cells of BCC bacteria at the species level, and their discrimination from the most common NF rods isolated from sputum samples of CF patients. We investigated 126 BCC clinical isolates (B. cepacia, B. multivorans, B. cenocepacia, B. stabilis, B. ambifaria and Taxon K bacteria), and 27 NF isolates, representing the 6 most clinically relevant NF rods (P. aeruginosa, S. maltophilia, A. xylosoxidans, Pandoraea spp, R. pickettii and Inquilinus limosus), recovered from patients hospitalized in different Cystic Fibrosis Centers and Hospitals of Argentina between 2004 and 2006. In addition 26 reference strains were analysed. The clinical isolates were phenotypically identified according to the guidelines for clinical microbiology practices for CF respiratory tract specimens. BCC bacteria were further identified by recA gene sequence analysis and the non-BCC bacteria were identified by 16S rRNA gene sequencing. MALDI-ToF spectrometry measurements were performed on 24 h-growth cultures grown on tryptone soya agar and treated with trifluoroacetic acid (TFA) as previously described by Lash et al. (6). Spectra, acquired in a Bruker Autoflex mass spectrometer (Bruker Daltonics, Germany), comprised 40-100 peaks primarily representing intracellular proteins within the m/z range masses 2 - 11 kDa. Our first step in the spectral analysis was the elaboration of ?Species-identifying biomarker proteins? (SIBPs). By means of MASS software (Matlab-based software developed in-house) a simulated gel view from sets of pre-processed mass spectra were generated. SIBPs could be evidenced from the analysis of the BCC reference strains, of well-identified Taxon K of BCC isolates, and of the non-BCC reference strains. Based on these results MALDI-ToF-MS spectra of these reference strains were subjected to hierarchical clustering using the Pearson?s correlation coefficient as the distance measure and the Ward´s algorithm for clustering the spectra. The nine BCC species and the Taxon K isolates were grouped in species-specific clusters and were also fully discriminated from the non-BCC reference strains. These results prompted us to build an intact cells MALDI-ToF-based species identification system for BCC bacteria based on similarity analysis using Mass Software. The reference database was established with spectra belonging to BCC reference strains, non-BCC reference strains, and 10 clinical isolates of Taxon K. To evaluate the identification system a total of 23 BCC clinical isolates belonging to the 5 BCC species of major incidence in CF patients, 93 isolates of Taxon K, and 27 non-BCC clinical isolate, were challenged against the reference database, as an external validation. For all the species tested, 100% of correct identification was obtained at the species level. We conclude that MALDI-ToF MS is a rapid, reproducible and reliable technique for the discrimination and identification of BCC species that cause the majority of infections in CF patients, including the novel Taxon K, discriminating them from non-BCC isolates. Therefore, whole-cell MALDDI-ToF MS seems to be a promising methodology in routine microbiological analysis of BCC.