Identificação das espécies do Complexo Burkholderia cepacia em pacientes com fibrose cística: comparação entre as técnicas de Matrix-Assisted Laser Desorption-Ionization - Time of Flight Mass Spectrometry e o sequenciamento do gene recA

Abstract

Burkholderia cepacia Complex (CBc) comprises 20 closely related species that stand out as opportunistic pathogens in Cystic Fibrosis (CF) and are associated with a considerable impact on the morbimortality of these patients. Phylogenetic species and subtypes within the Burkholderia cenocepacia species vary in relation to transmissibility and virulence. Due to inconsistencies in the correct identification of CBc species by phenotypic methods, molecular methods are recommended. However, there is no universal molecular methodology that can be applied to all microorganisms, especially in clinical laboratories. In this sense, the methodology based on mass spectrometry appears as an alternative. The aim of this study was to evaluate the performance of the Matrix Associated Laser Desorption-Ionization (MALDI-TOF MS) compared to Bcc species identification by recA gene sequencing in CF patients. We analyzed 43 samples obtained from different clinical specimens from January 2014 to February 2016. A total of 18 (41.86%) samples of Burkholderia vietnamiensis, 14 (32.55%) Burkholderia cenocepacia subtype IIIB, eight (18.6%) Burkholderia cenocepacia subtype IIIA, two (4.65%) Burkholderia contaminans and one (2.32%) Burkholderia cepacia were identified by sequencing the recA gene. Identification by MALDI-TOF MS was evaluated considering prior culture in Sheep Blood Agar (SBA) and Columbia Agar Base (CAB). 81.39% of the SBA samples were identified at the genus level (log ≥1.70-1.99) and 41.86% at the species level (log ≥2.00), corresponding to 11 samples of B. vietnamiensis (61.11%), four of B. cenocepacia IIIB (28,57%) and two of B. cenocepacia IIIA (25%). Only 39.53% of the samples cultured in CAB were identified at the genus level and 16.27% in species: B. vietnamiensis (n = 4, 22.22%) and B. cenocepacia IIIB (n = 2; 14,28%). B. cepacia sample was identified from both media and the two B. contaminans samples were not identified. Performance of MALDI-TOF MS varied with the use of different culture media and the identification at the species level was less than 50% considering the cut-off points established by the manufacturer. Thus, we evaluated whether the reduction of cut-off points would increase the accuracy of the identification. With the reduction of the escore to ≥1.70-1.99 (considered as reliable identification at genus level), another three samples of B. vietnamiensis, four of B. cenocepacia IIIA and seven of B. cenocepacia IIIB were identified, corresponding to an increase of 32.55% (in relation to the escore ≥ 2.00). However, the number of samples analyzed does not allow this criterion to be used and additional studies are required. Despite the limitations of MALDI-TOF MS as a typing method, the use of bioinformatics tools to analyze the mass spectra obtained in MALDI-TOF MS allowed the clustering of 96% of the phylogenetic subtypes IIIA and IIIB of B. cenocepacia strains. We also observed the proximity of samples related to the same patient through the nodes formed in the phylogenetic tree, but the absence of comparison with a reference typing method does not allow to infer their use for epidemiological purposes.