Discrimination and Phylogenomic Classification of Bacillus anthracis-cereus-thuringiensis Strains Based on LC-MS/MS Analysis of Whole Cell Protein Digests

摘要

Modern taxonomy, diagnostics, and forensics of bacteriarnbenefit from technologies that provide data for genomebasedrnclassification and identification of strains; however,rnfull genome sequencing is still costly, lengthy, and laborrnintensive. Therefore, other methods are needed to estimaterngenomic relatedness among strains in an economicalrnand timely manner. Although DNA-DNA hybridizationrnand techniques based on genome fingerprinting or sequencingrnselected genes like 16S rDNA, gyrB, or rpoBrnare frequently used as phylogenetic markers, analyses ofrncomplete genome sequences showed that global measuresrnof genome relatedness, such as the average genomernconservation of shared genes, can provide better strainrnresolution and give phylogenies congruent with relatednessrnrevealed by traditional phylogenetic markers. Bacterialrngenomes are characterized by a high gene density;rntherefore, we investigated the integration of mass spectrometry-rnbased proteomic techniques with statistical methodsrnfor phylogenomic classification of bacterial strains.rnFor this purpose, we used a set of well characterizedrnBacillus cereus group strains isolated from poisoned foodrnto describe a method that relies on liquid chromatography-rnelectrospray ionization-tandem mass spectrometry ofrntryptic peptides derived from whole cell digests. Peptidesrnwere identified and matched to a prototype database (DB)rnof reference bacteria with fully sequenced genomes tornobtain their phylogenetic profiles. These profiles werernprocessed for predicting genomic similarities with DBrnbacteria estimated by fractions of shared peptides (FSPs).rnFSPs served as descriptors for each food isolate and werernjointly analyzed using hierarchical cluster analysis methodsrnfor revealing relatedness among investigated strains.rnThe results showed that phylogenomic classification ofrntested food isolates was in consonance with results fromrnestablished genomic methods, thus validating our findings.rnIn conclusion, the proposed approach could be usedrnas an alternative method for predicting relatedness amongrnmicrobial genomes of B. cereus group members andrnpotentially may circumvent the need for whole genomernsequencing for phylogenomic typing of strains.