Characterization of Salmonella enterica subspecies I genovars by use of microarrays

Abstract

Subspecies 1 of Salmonella enterica is responsible for almost all Salmonella infections of warm-blooded animals. Within subspecies 1 there are over 2,300 known serovars that differ in their prevalence and the diseases that they cause in different hosts. Only a few of these serovars are responsible for most Salmonella infections in humans and domestic animals. The gene contents of 79 strains from the most prevalent serovars were profiled by microarray analysis. Strains within the same serovar often differed by the presence and absence of hundreds of genes. Gene contents sometimes differed more within a serovar than between serovars. Groups of strains that share a distinct profile of gene content can be referred to as "genovars" to distinguish them from serovars. Several misassignments within the Salmonella reference B collection were detected by genovar typing and were subsequently confirmed serologically. Just as serology has proved useful for understanding the host range and pathogenic manifestations of Salmonella, genovars are likely to further define previously unrecognized specific features of Salmonella infections.

FIG. 2.

Phylogenetic tree of S. enterica subspecies I isolates. The tree was constructed with PAUP software (Sinauer Assoc. & Co.) by using the presence-absence predictions for the regions as described in Table 2. The following conditions were applied: maximum parsimony, weighting against repeated gains of genes, 10,000 bootstraps. Serogroups are indicated, and notable deviations from the expected clustering by serogroup are depicted with red arrows. Me isolates are serogroup E1, and Pa1 is serogroup A.

FIG. 3.

Relationship matrix depicting the numbers of different absence-presence calls for genomic Typhimurium LT2 and Typhi CT18 regions between strains. Phage regions are excluded. The numbers of differences in gene content are illustrated as shaded squares on a linear scale from white (no differences) to black (maximal number of differences in the matrix [117, for S. enterica subsp. indica, or VI, versus Typhi Tp1]). Strains are grouped by serovars and in the order of similarity of the most common MLEE type of the respective serovar to Typhimurium Tm1. Within the serovars, strains are ordered as in Table 1. Polyphyletic serovars are marked with black squares. Similarity between Dublin Du2 and Java JaA1 is highlighted with arrows.

FIG. 1.

S. enterica serovar Typhimurium LT2 and Typhi CT18 gene homologues with heterogeneous distribution patterns in S. enterica subspecies I serovars. Gene status is color-coded as follows: blue, present; purple, uncertain; red, absent. The strains are depicted, from left to right, in order of appearance in Table 1. (A) Serovar Typhimurium LT2 chromosomal genes. Only ORFs that are absent from at least one subspecies I strain are shown. (B) Plasmid pSLT. (C) Plasmid R46. (D) Genes present in serovar Typhi CT18, but absent from serovar Typhimurium LT2. In panels B and C, only genes that were predicted to be present in at least one subspecies I isolate outside serovar Typhimurium are shown. In panel D, only genes detected in at least one subspecies I isolate outside serovar Typhi are depicted.