Multilocus sequence typing scheme for Enterococcus faecalis reveals hospital-adapted genetic complexes in a background of high rates of recombination

Abstract

A multilocus sequence typing (MLST) scheme based on seven housekeeping genes was used to investigate the epidemiology and population structure of Enterococcus faecalis. MLST of 110 isolates from different sources and geographic locations revealed 55 different sequence types that grouped into four major clonal complexes (CC2, CC9, CC10, and CC21) by use of eBURST. Two of these clonal complexes, CC2 and CC9, are particularly fit in the hospital environment, as CC2 includes the previously described BVE clonal complex identified by an alternative MLST scheme and CC9 includes exclusively isolates from hospitalized patients. Identical alleles were found in genetically diverse isolates with no linkage disequilibrium, while the different MLST loci gave incongruent phylogenetic trees. This demonstrates that recombination is an important mechanism driving genetic variation in E. faecalis and suggests an epidemic population structure for E. faecalis. Our novel MLST scheme provides an excellent tool for investigating local and short-term epidemiology as well as global epidemiology, population structure, and genetic evolution of E. faecalis.

FIG. 1.

Dendrogram showing the relatedness among the 55 STs of E. faecalis by use of UPGMA from the matrix of pairwise differences in the allelic profiles. The following data are included: ST; numbers of isolates with the same ST; host origin (HO, isolate from a hospital outbreak; HC, clinical sample from a hospitalized human; HF, fecal sample from a hospitalized human; CF, fecal sample from a healthy volunteer; AS, fecal sample from a healthy animal; AC, clinical sample from an animal; M, miscellaneous group, which included isolates from sewage, laboratory strains, and unknown origin); resistance (BLA⁺, beta-lactamase-producing isolate); numbers of isolates with indicated allelic profiles; and countries of origin (ARG, Argentina; BEL, Belgium; GRC, Greece; IND, India; LEB, Lebanon; NLD, The Netherlands; POR, Portugal; SP, Spain; THA, Thailand; USA, United States). Numbers in the columns under the spanner “Host origin” represent numbers of isolates. Numbers in the columns under the spanner “Resistance” represent numbers of isolates with indicated resistance patterns; letters in the same columns represent host origin abbreviations.

FIG. 2.

Clustering of 55 STs by use of the minimum spanning tree. Colors indicate isolation sources (A) or resistance phenotypes (B). Each circle represents an ST, and the type number is indicated in the circle. The area of each circle corresponds to the number of isolates. Thick, short, solid lines connect single-locus variants, thin, longer, solid lines connect double-locus variants, black dotted lines connect STs which differ in three loci, and gray dotted lines connect STs that differ in more than three loci. Pie charts indicate ST distribution. Clonal complexes are indicated.

FIG. 2.

Clustering of 55 STs by use of the minimum spanning tree. Colors indicate isolation sources (A) or resistance phenotypes (B). Each circle represents an ST, and the type number is indicated in the circle. The area of each circle corresponds to the number of isolates. Thick, short, solid lines connect single-locus variants, thin, longer, solid lines connect double-locus variants, black dotted lines connect STs which differ in three loci, and gray dotted lines connect STs that differ in more than three loci. Pie charts indicate ST distribution. Clonal complexes are indicated.