Population structure and acquisition of the vanB resistance determinant in German clinical isolates of Enterococcus faecium ST192

Abstract

In the context of the global action plan to reduce the dissemination of antibiotic resistances it is of utmost importance to understand the population structure of resistant endemic bacterial lineages and to elucidate how bacteria acquire certain resistance determinants. Vancomycin resistant enterococci represent one such example of a prominent nosocomial pathogen on which nation-wide population analyses on prevalent lineages are scarce and data on how the bacteria acquire resistance, especially of the vanB genotype, are still under debate. With respect to Germany, an increased prevalence of VRE was noted in recent years. Here, invasive infections caused by sequence type ST192 VRE are often associated with the vanB-type resistance determinant. Hence, we analyzed 49 vanB-positive and vanB-negative E. faecium isolates by means of whole genome sequencing. Our studies revealed a distinct population structure and that spread of the Tn1549-vanB-type resistance involves exchange of large chromosomal fragments between vanB-positive and vanB-negative enterococci rather than independent acquisition events. In vitro filter-mating experiments support the hypothesis and suggest the presence of certain target sequences as a limiting factor for dissemination of the vanB element. Thus, the present study provides a better understanding of how enterococci emerge into successful multidrug-resistant nosocomial pathogens.

Figure 1

Phylogenetic analysis of the core genome of German E. faecium ST192 (a) and non-ST192 (b) clinical isolates. (a) The proportional transformed branch diagram revealed 4 different clades or subclades of ST192 isolates. (b) ST192 and non-ST192 were found to cluster according to their respective MLST type with the exception of UW7027-ST78 falling within clade III of ST192 isolates. Branch labels represent a bootstrap with 1000 permutations. Color assignment was done to visually differentiate ST192 clades as defined in this study (green: VRE clade CIa; red: VRE clade CIb; salmon: VRE clade CII; light blue: VSE/VRE clade CIII and grey VSE/VRE clade IV). Outlying brackets frame the year of isolation of the strains belonging to the respective clade. All isolates are ST192 unless stated next to the strain name. Efm_DO represents the reference E. faecium DO/TX16 (CP003583). “vanB-“ refers to vanB-negative isolates.

Figure 2. Schematic representation of Tn1549 insertion into the major insertion site HMPREF0351_10592.

Insertion and orientation of Tn1549 into HMPREF0351_10592 is depicted alongside six nucleotides framing the specific insertion site. Enumeration of nucleotides is according to the coding sequence of HMPREF0351_10592. The coupling sequence represents nucleotides which were co-transferred from the initial donor strain.

Figure 3. Phylogenetic analysis based on the Tn1549 sequences of German E. faecium ST192 and non-ST192 clinical isolates.

The proportional transformed branch diagram revealed an insertion site-specific clustering across all sequence types. Differentiation of the various insertion sites was further validated by bootstrap analysis with 1000 permutations and is indicated by branch labeling. For consistency, color coding represents the different ST192 clades as represented in Fig. 1a,b (green: VRE clade CIa; red: VRE clade CIb; salmon: VRE clade CII; light blue: VSE/VRE clade CIII and grey VSE/VRE clade IV). Unless indicated by specific ST enumeration, all isolates belonged to ST192. Insertion sites are depicted as locus_tag numbering according to the reference genome E. faecium DO (CP003583). Tn1549 represents the reference sequence used for mapping (CP013009). Insertion site “PAI” represents the reference locus_tag EFAU085_02779, as it is not present in E. faecium DO, and due to the proximity to a pathogenicity island (PAI) was termed “PAI” in the following. unk insertion site unknown.

Figure 4. S1-nuclease macrorestriction of donor, recipient and transconjugant (TC) bacteria.

Separation in PFGE disclosed the plasmid content of the strains analyzed. The red rectangle indicates the putative novel plasmid pWCF-TC1 of 66.5 kb. The red asterisks indicate strains that do not possess the respective plasmid. M marker strain S. aureus NCTC8325.