Gastrointestinal Tract Colonization Dynamics by Different Enterococcus faecium Clades

The Journal of Infectious Diseases MAJOR ARTICLE Gastrointestinal Tract Colonization Dynamics by Different Enterococcus faecium Clades Maria Camila Montealegre,1,2,3,a Kavindra V. Singh,1,2,a and Barbara E. Murray1,2,3 1 Department of Internal Medicine, Division of Infectious Diseases, 2Center for the Study of Emerging and Re-emerging Pathogens, and 3Department of Microbiology and Molecular Genetics, University of Texas Graduate School of Biomedical Sciences at Houston, University of Texas Health Science Center at Houston (See the editorial commentary by Wurster, Saavedra, and Gilmore on pages 1862–5.) Enterococcus faecium, a common colonizer of the gastrointestinal tract (GIT) of healthy individuals and animals [1], has emerged as an important cause of hospital-associated infections, including bacteremias, urinary tract infections (UTIs), and even endocarditis [2]. The treatment of E. faecium infections is particularly challenging owing to the intrinsic resistance of these organisms to several antibiotics and their remarkable capacity to acquire resistance to others via mutation or horizontal gene transfer [3]. Early population biology studies of E. faecium indicated that the majority of E. faecium strains responsible for hospital-associated outbreaks and infections were genotypically different from the majority of commensal isolates [4–6]. More-recent whole-genome analyses found a deep phylogenetic split into 2 E. faecium clades, the hospital-associated clade, known as clade A, and the community-associated clade, known as clade B [7, 8]; in addition, Lebreton et al revealed a further split within clade A, into subclade A1 (associated with hospital infections) and subclade A2 (associated with animals and sporadic human infections) [7]. Received 28 August 2015; accepted 3 November 2015; published online 15 December 2015. Presented in part: American Society for Microbiology Interscience Conference of Antimicrobial Agents and Chemotherapy, San Diego, California, September 2015. a M. C. M. and K. V. S. contributed equally to this work. Correspondence: B. E. Murray, Department of Internal Medicine, Division of Infectious Diseases, The University of Texas Health Science Center at Houston, 6431 Fannin St, MSB 2.112, Houston, TX 77030 (). The Journal of Infectious Diseases® 2016;213:1914–22 © The Author 2015. Published by Oxford University Press for the Infectious Diseases Society of America. All rights reserved. For permissions, e-mail . DOI: 10.1093/infdis/jiv597 1914 • JID 2016:213 (15 June) • Montealegre et al Although studies have demonstrated the existence of large differences in the accessory [7, 9, 10] and core genome [7, 8] between the clades, very little is known about the factors that promote the predominance of subclade A1 strains in the hospital setting. Some have suggested that the transition of E. faecium from commensal to pathogen is a consequence of the enriched accessory genome of subclade A1 strains, including acquired antibiotic resistance determinants [11], genomic islands [9], and insertion sequences [12, 13]. Putative virulence factors are also found to be enriched in subclade A1 strains [14]; however, only a few of these have been experimentally proven to contribute to pathogenesis [15, 16]. GIT colonization with antibiotic-resistant enterococci generally precedes infection [17]. In addition, it has been demonstrated that, during hospitalization, ampicillin-resistant E. faecium strains rapidly replace ampicillin-susceptible, commensal E. faecium strains [18, 19]. Interestingly, after a patient is discharged from the hospital, ampicillin-resistant E. faecium tend to wane [18]. Although the replacement of commensal clade B strains by subclade A1 E. faecium in the hospital environment could be related to the greater fitness, colonization capacity, or virulence potential of subclade A1 strains, our hypothesis is that commensal clade B strains have a better ability to colonize the GIT than clade A isolates, which would explain the vast predominance of clade B in humans in the community and why antibiotic-resistant E. faecium strains are often replaced once patients leave the hospital. In an attempt to better understand the dynamics of E. faecium colonization, we evaluated the ability of 12 E. faecium strains from clades A1, A2, and B to colonize the GIT of mice, individually as well as in competition with a strain of a different clade. Colonization of the gastrointestinal tract (GIT) generally precedes infection with antibiotic-resistant Enterococcus faecium. We used a mouse GIT colonization model to test differences in the colonization levels by strains from different E. faecium lineages: clade B, part of the healthy human microbiota; subclade A1, associated with infections; and subclade A2, primarily associated with animals. After mono-inoculation, there was no significant difference in colonization (measured as the geometric mean number of colony-forming units per gram) by the E. faecium clades at any time point (P > .05). However, in competition assays, with 6 of the 7 pairs, clade B strains outcompeted clade A strains in their ability to persist in the GIT; this difference was significant in some pairs by day 2 and in all pairs by day 14 (P < .0008–.0283). This observation may explain the predominance of clade B in the community and why antibiotic-resistant hospital-associated E. faecium are often replaced by clade B strains once patients leave the hospital. Keywords. Enterococcus faecium; clades; GIT; colonization; fitness. Table 1. Enterococcus faecium Strains Used in This Study and Minimum Inhibitory Concentrations (MICs) of Select Antimicrobial Agents MIC, μg/mL Subclade or Clade, Strain Source (Isolation Site) Country of Isolation/Year MLST AMP ERY GEN VAN Reference [20] A1 C68 Hospitalized patient (feces) USA/1996 16 128 >256 >1024 128 1.230.933 Hospitalized patient (blood) USA/2005 18 128 >256 16 >256 [10] TX82 Endocarditis patient (blood) USA/1999 17 64 256 4 >256 [21] TX0133A Endocarditis patient (blood) USA/2006 17 64 >256 16 >256 TX16 (DO) Endocarditis patient (blood) USA/1992 18 16 >256 16 EnGen12 Hospitalized patient (ascites) Netherlands/1995 27 0.5 >256 16 EnGen35 Hospitalized patient (gut) Netherlands/1979 66 1 4 8 EnGen21 Hospitalized patient (feces) Netherlands/2002 5 8 >256 8 Com15 Healthy volunteer (feces) USA/2007 583 ≤0.25 16 8 1 [10] TX1330 Healthy volunteer (feces) USA/1994 107 1 0.25 8 1 [24] 0.5 [22] [23] A2 >256 0.5 >256 [7] [7] [7] B E980 Healthy volunteer (feces) Netherlands/1998 94 ≤2 32 8 0.5 [9] 1.141.733 Hospitalized patient (wound) USA/2005 327 2 16 16 1 [10] Abbreviations: AMP, ampicillin; ERY, erythromycin; GEN, gentamicin; MLST, multilocus sequence type; VAN, vancomycin. Bacterial Strains, Routine Growth Conditions, and General Techniques E. faecium used from subclades A1 and A2 and clade B an (...truncated)