Restricted Gene Flow among Hospital Subpopulations of Enterococcus faecium

Abstract

Enterococcus faecium has recently emerged as an important multiresistant nosocomial pathogen. Defining population structure in this species is required to provide insight into the existence, distribution, and dynamics of specific multiresistant or pathogenic lineages in particular environments, like the hospital. Here, we probe the population structure of E. faecium using Bayesian-based population genetic modeling implemented in Bayesian Analysis of Population Structure (BAPS) software. The analysis involved 1,720 isolates belonging to 519 sequence types (STs) (491 for E. faecium and 28 for Enterococcus faecalis). E. faecium isolates grouped into 13 BAPS (sub)groups, but the large majority (80%) of nosocomial isolates clustered in two subgroups (2-1 and 3-3). Phylogenetic and eBURST analysis of BAPS groups 2 and 3 confirmed the existence of three separate hospital lineages (17, 18, and 78), highlighting different evolutionary trajectories for BAPS 2-1 (lineage 78) and 3-3 (lineage 17 and lineage 18) isolates. Phylogenomic analysis of 29 E. faecium isolates showed agreement between BAPS assignment of STs and their relative positions in the phylogenetic tree. Odds ratio calculation confirmed the significant association between hospital isolates with BAPS 3-3 and lineages 17, 18, and 78. Admixture analysis showed a scarce number of recombination events between the different BAPS groups. For the E. faecium hospital population, we propose an evolutionary model in which strains with a high propensity to colonize and infect hospitalized patients arise through horizontal gene transfer. Once adapted to the distinct hospital niche, this subpopulation becomes isolated, and recombination with other populations declines.

The Plasmidome of Firmicutes: Impact on the Emergence and the Spread of Resistance to Antimicrobials

“…Most enterococcal plasmids are able to acquire and disseminate AbR genes by different mechanisms of genetic exchange. However, the role of plasmids in the population structure and evolvability of these enterococcal species has been poorly addressed (194)(195)(196)(197) due to the overrepresentation of recent clinical and animal isolates of specific lineages commonly associated with AbR included in most studies (7,141) and due to the lack of available plasmid sequences. Similar plasmids have been found in E. faecium and other enterococcal species that may play equivalent functional roles in the gastrointestinal tract such as Enterococcus avium, Enterococcus raffinosus, E. durans, and E. hirae (195,198).…”

Nosocomial Infections and their Surveillance in Veterinary Hospitals

“…E. faecalis is the predominant species responsible for clinical infection whereas E. faecium claims the higher antibiotic resistance (Giraffa, 2002). Globally multiresistant E. faecium causing increasing number of hospital associated infections (Willems et al, 2012) and embodies upto to one third of Enterococcal infections (Willems and VanSchaik, 2009 (Metan et al, 2005. In a study conducted in Korea to investigate the possibility of MDR Enterococcus cross transmission, Enterococcus species was isolated from dogs, dog owners, veterinary personnel and five veterinary hospital environment and the study demonstrated a high prevalence of 62.5% of MDR E. faecalis and 75% of MDR E. faecium (Chung et al, 2014).…”

Worldwide Population Structure, Long-Term Demography, and Local Adaptation of Helicobacter pylori

“…We ran the DAPC analysis with 1,000 starting points and 1,000,000 iterations and found that results were consistently convergent over 10 independent trials. BAPS was run with a subset of 100,000 SNPs using the admixture model for haploid individuals and was shown to be effective to detect bacterial populations and gene flow in large-scale datasets (Tang et al 2009;Willems et al 2012). STRUCTURE was run on a subset of 100 kb, for a total of 29,242 SNPs, using 10,000 burn-in and 50,000 iterations, and we replicated 5 runs for each tested number of partitions (from 2 to 10) with the admixture model.…”