Salmonella enterica comprises over 2500 serovars, many of which are significant foodborne pathogens in humans. The ability to subtype these microbes is difficult due to the highly clonal nature of many Salmonella strains and a lack of congruence among traditional typing approaches. This work examines the phylogenetic utility of a multi-locus sequence typing (MLST) approach to discriminate between members of a closely related collection of salmonellae, the Salmonella reference collection A (SARA). This 72 strain collection, referred to as the 'Typhimurium' complex, consists of S. Typhimurium and its four closest serological relatives. In this analysis, nucleotide sequences from seven housekeeping genes (aroC, dnaN, hemD, hisD, purE, sucA, and thrA) were PCR amplified, sequenced, and combined into a single concatenated character matrix providing 3360bp for cladistic analysis. The resultant most parsimonious tree yielded seven clades of Salmonella strains that partitioned largely along serovar divisions within the collection except for five 'Paratyphi B' strains, two 'Saintpaul' strains, and two 'Typhimurium' strains. Convergence in the SARA tree was approximately 20% indicating that the vast majority of sequence changes were phylogenetically informative. Despite a high consistency among nucleotide substitutions, analysis of congruence identified several SARA strains with recombinant alleles in the concatenated matrix. These findings point to important differences among phylogenetic contributions made by the individual genes comprising this MLST dataset.
Published by Elsevier B.V.