Mobile genetic elements provide evidence for a bovine origin of clonal complex 17 of Streptococcus agalactiae

Abstract

We sought an explanation for epidemiological changes in Streptococcus agalactiae infections by investigating the link between ecological niches of the bacterium by determining the prevalence of 11 mobile genetic elements. The prevalence of nine of these elements differed significantly according to the human or bovine origin of the isolate. Correlating this distribution with the phylogeny obtained by multilocus sequence analysis, we observed that human isolates harboring GBSi1, a clear marker of the bovine niche, clustered in clonal complex 17. Our results are thus consistent with the emergence of this virulent human clone from a bovine ancestor.

FIG. 1.

Prevalence of the MGEs within the human (filled bars) and bovine (open bars) S. agalactiae isolates. Significant differences (without correction for bias due to multiple comparisons) are indicated by one asterisk for P values below 0.05 and two asterisks for P values below 0.001.

FIG. 2.

Relatedness of the 98 isolates of S. agalactiae analyzed by the NJ method, using MEGA v3.1 software and the Kimura two-parameter mutation model of genetic distance. The nucleotide differences between the supergene sequences obtained by multilocus sequence analysis method of Jones et al. (11) were used to infer phylogenetic relationships between isolates. We used eBURST v3 software to group the 98 isolates into CCs. For each isolate, the ST, beta-hemolysis, serotype, and presence (+) or absence (−) of the 11 MGEs are indicated. CCs are also indicated on the right of the figure (several isolates were not attributed a CC number by the Start software and are considered “singletons”). The prominent position of isolate H14 is indicated by a black arrow. The human isolates are named L or LJ (meningocerebral fluid), H (adult), V or VV (vaginal carriage), or G (gastric fluid); the bovine isolates are named N (bovine isolates are underlined). NT, nontypeable.

FIG. 3.

Factorial analysis of correspondence of 98 S. agalactiae isolates based on the MGE data. Eleven active variables (the MGEs [▴] IS1381, IS861, IS1548, GBSi1, ISSa4, ISSag2, ISSag1, SAG0448, SAG1241, SAG0610, and ISSag9), five illustrative variables (the main clonal complexes CC19 [red diamond], CC17 [pink diamond], and CC67 [green diamond] and the human [circles] and bovine [squares] origins of the isolates), and the 98 isolates are projected onto the F1/F2 plane. This plane, obtained by computation, is defined by the two principal axes of the analysis; the first axis, F1, explains most of the variance, and the second axis, F2 (orthogonal to F1), explains most of the variance not explained by F1. Together, F1 and F2 accounted for 40.97% of the total variance. Isolates of CC17 (n = 13) are shown in red, isolates of CC67 (n = 15) are shown in green, and isolates of subCC19 (n = 21) are shown in pink.