Characterization of Vancomycin-Resistant Enterococcus faecium Isolates from Broiler Poultry and Pig Farms in England and Wales

JOURNAL OF CLINICAL MICROBIOLOGY, July 2005, p. 3283 3289 Vol. 43, No. 7 0095-1137/05/$08.00 0 doi:10.1128/jcm.43.7.3283 3289.2005 Characterization of Vancomycin-Resistant Enterococcus faecium Isolates from Broiler Poultry and Pig Farms in England and Wales L. Garcia-Migura, E. Pleydell, S. Barnes, R. H. Davies, and E. Liebana* Department of Food and Environmental Safety, Veterinary Laboratories Agency-Weybridge, Addlestone, Surrey KT15 3NB, United Kingdom Received 29 July 2004/Returned for modification 29 November 2004/Accepted 24 February 2005 This study aimed to investigate the occurrence and molecular epidemiology of vancomycin-resistant Enterococcus faecium (VREF) isolates on poultry and pig farms in England and Wales. A total of 217 VREF isolates were obtained from fresh feces and environmental swabs collected from conventional and organic farms. A predominant pulsed-field gel electrophoresis (PFGE) profile was found for each VREF-positive farm, together with less frequent types. All isolates presented the vana genotype and were esp negative. Seventy-six percent of the VREF isolates were additionally resistant to nine or more antimicrobials, presenting a diverse range of resistance phenotypes. The multiresistance traits did not appear to be specific to individual farms or sample types (i.e., environmental or fecal), nor did they correlate with any specific PFGE type. Ninety-three percent of the isolates were resistant to penicillin, 89% were resistant to tetracycline, 87.5% were resistant to erythromycin, and 50% were resistant to quinupristin-dalfospristin (Synercid). The lack of clonality among these populations may suggest the horizontal transfer of resistance genes and/or a dynamic replacement of clonal lines rather than persistence. Bacteria are capable of developing strategies to enhance their survival in adverse situations such as conditions of high salinity, changes in temperature, the presence of heavy metals, or the presence of antimicrobial activity. Enterococcus faecium, which is part of the normal intestinal flora in humans and animals (12), provides a good example of such adaptive bacterial evolution. This species has emerged as an important opportunistic pathogen causing life-threatening infections in hospitals. The emergence of this pathogen is associated with a remarkable capacity to accumulate resistance to antimicrobials (13). Enterococci are intrinsically resistant to cephalosporins and aminoglycosides and were the first bacteria to acquire resistance to vancomycin (19), a glycopeptide antibiotic commonly used for the clinical therapy of gram-positive nosocomial infections. Multidrug-resistant enterococci, particularly those that are vancomycin resistant, are a major cause of concern for the medical community. The percentage of infections caused by vancomycin-resistant enterococci (VRE) has increased 20-fold over the last decade in the United States (13). Recently, it was shown that the genes responsible for this resistance have the potential to be transferred to other gram-positive pathogens such as Staphylococcus aureus (20), thus intensifying the public health threat, especially now that the first cases of vancomycinand methicillin-resistant S. aureus have been reported in the United States (26, 30). Antimicrobial growth promoters have been used for the past * Corresponding author. Mailing address: Department of Food and Environmental Safety, Veterinary Laboratories Agency-Weybridge, Addlestone, Surrey KT15 3NB, United Kingdom. Phone: 44 1932 357587. Fax: 44 1932 357595. E-mail: e.liebana@vla.defra.gsi.gov.uk. five decades as an effective way of enhancing productivity and animal health during livestock production. Avoparcin is an example of a feed additive that has been used in intensive and integrated agricultural systems across Europe, especially within the pig and poultry industries. This growth promoter has not been used in the United States. Avoparcin is a glycopeptide that induces cross-resistance to vancomycin (32) and has been considered an important factor in the emergence and spread of resistance to vancomycin in enterococcal populations (4). For this reason, the use of avoparcin was banned in Denmark in 1995 and throughout the European Union in 1997. In Europe, VRE carriage occurs mainly among healthy individuals in the community and in farm livestock (8). In contrast, in the United States VRE are normally endemic to hospitals (8, 10, 15). VRE have also recently been found as a result of infections linked to foreign travel and to the consumption of imported food (17). It has been postulated that in Europe, the colonization of humans by VRE may occur predominantly via the food chain (5, 6). Moreover, similar vana elements have been found in strains from animals and humans (9, 25, 29). Although the colonization of humans by animal isolates might be transient (7, 24), the risk of transfer of resistance genes during the colonization period has not yet been assessed and might be crucial. Multiresistant organisms might enter the food chain via farm animals. The aim of the present study was to investigate the occurrence and molecular epidemiology of vancomycin-resistant E. faecium (VREF) on poultry and pig farms in England and Wales. The molecular characterization of isolates collected from farms (including genotyping and identification of specific resistance determinants) will help to develop a better 3283

3284 GARCIA-MIGURA ET AL. J. CLIN. MICROBIOL. understanding of the persistence and dynamics of these organisms in farm environments. MATERIALS AND METHODS Farms and sample collection. Samples were collected from a total of 47 farms. Twenty-five farms (six conventional and seven organic broiler farms and seven conventional and five organic pig farms) were sampled once in 2002 and then again within the same month in 2003. A small number of farms were visited more frequently in order to make preliminary assessments of short-term variability. During each visit, 40 to 60 pooled fecal samples were collected, ensuring that all age groups or production classes present on each farm were represented. Pooled fecal samples were collected from the ground and consisted of approximately 5 g offecal material from each of eight fresh fecal masses. Care was taken during the procedure to avoid environmental contamination of the fecal samples. Representative environmental swabs were also taken from internal building surfaces, feeders, and drinkers. For the remaining 22 farms, the farmers collected eight pooled fecal samples from three of their livestock houses on a single occasion and sent them by post to the laboratory for analysis. Bacteriology. Fecal samples (40 g) were mixed with the same volume of phosphate-buffered saline and vortexed, and then 10 l of each was streaked onto Slanetz and Bartley agar (Oxoid, Basingstoke, United Kingdom) containing vancomycin (6 g/ml) and incubated aerobically for 48 h at 42 C. Environmental samples were enriched in buffered peptone water for 18 h at 37 C before plating onto the same selective medium. Presumptive VREF isolates were selected by their colony morphology and color (purple-pink colonies with a lighter halo), subcultured, and stored at 70 C. The identity of the presumptive E. faecium isolates was confirmed by real-time PCR. Amplification was performed in a LightCycler instrument (Roche Diagnostics UK Ltd., Lewes, United Kingdom) in glass capillaries containing 20- l reaction mixtures including 1 LightCycler FastStart DNA Master SYBR green I (Roche Molecular Biochemicals), 10 pmol each of the ddl E. faecium -specific primers F1 and F2 (11), 4 mm MgCl 2 (final concentration), and 2 l of a crude extract of bacterial DNA template. The PCR run comprised 30 cycles with denaturation at 95 C for 5 seconds, annealing at 50 C for 5 seconds, and extension at 72 C for 22 seconds. The nature of the amplicon was determined by a melting point analysis over a temperature range from 65 to 95 C, with a transition rate of 0.1 C/s and continuous detection of fluorescence in channel 1. The melting temperature (T m ) for the PCR product was 84 C, and the size of the product was 550 bp. The identification of the genes responsible for vancomycin resistance was investigated by a multiplex PCR as described previously (11). Subsequently, a duplex, SYBR green I-based PCR assay was developed for the simultaneous detection of E. faecium and vana genes by the use of the primers described above in a LightCycler assay. The PCR consisted of 30 cycles of 95 C for 5 seconds (denaturation), 54 C for 5 seconds (annealing), and 72 C for 25 seconds (extension). A melting point analysis over a temperature range from 65 to 95 C with a transition rate of 0.1 C/s resulted in the identification of two distinct peaks representing the two targets, i.e., one at 84 C for ddl E. faecium and the other at 87 C for vana. The presence of the esp gene in 137 randomly selected isolates was assessed by PCR. Primers were designed with DnaStar software from a published sequence of the esp gene of E. faecium (GenBank database accession number AX537383). PCRs were prepared as follows: 25- l volumes contained 10 pmol of each primer (esp1, 5 TTAGCGGGAACAGGTCACA; esp2, 5 TGTTGCATCATTTTCCA TAGC), 1.5 mm MgCl 2,1UofAmpliTaq Gold (Applied Biosystems), 1 GeneAmp PCR buffer, and 5 l of template DNA. The PCR cycling parameters were as follows: denaturation at 94 C for 10 min, followed by 30 cycles consisting of 94 C for 30 seconds, 62 C for 1 min, and 72 C for 3 min and a final cycle of 10 min at 72 C. The expected size of the amplicon was 471 bp. PFGE. DNAs were prepared as described previously (9a). A single colony of each isolate was streaked onto yeast extract agar and incubated overnight at 37 C. Using a cotton swab, we transferred a portion of the growth to 3 ml of TE buffer (10 mm Tris, 1 mm EDTA, ph 8.0) and adjusted the cell concentration to 0.59 with a Dade Microscan turbidity meter (Dade Behring). A total volume of 240 l of the suspension was transferred to 1.5-ml microcentrifuge tubes, and 60 l of lysozyme solution (10 mg/ml) was added. The tubes were incubated at 37 C for 10 min. Immediately after incubation, 300 l of 1.2% Seakem Gold agarose (Cambrex, East Rutherford, N.J.) 1% sodium dodecyl sulfate 0.2 mg/ml proteinase K was mixed with the bacterial suspension and pipetted into disposable plug moulds. Three plugs per isolate were transferred to 50-ml polypropylene screw-top tubes with 5 ml of cell lysis buffer (50 mm Tris, 50 mm EDTA, ph 8.0, 1% sarcosyl, 0.15 mg/ml proteinase K) and then incubated at 54 C in a shaking water bath for 2 h. Thereafter, the plugs were washed twice with 15 ml TABLE 1. Antimicrobials tested and breakpoints used for this study Antibiotic/growth promoter of sterile water and four times with TE buffer at 50 C for 15 min. Restriction digestion of chromosomal DNAs was carried out by using 25 units of SmaI (Promega, Southampton, United Kingdom) for 2 h at 25 C. Pulsed-field gel electrophoresis (PFGE) was performed on a CHEF DRIII system (Bio-Rad, Hercules, Calif.) in 0.5 TBE extended-range buffer (Bio-Rad) with recirculation at 14 C. DNA restriction fragments were resolved in 0.8% SeaKem Gold agarose in 0.5 TBE buffer. DNA from Salmonella Braenderup H9812 restricted with XbaI was used as a size marker. Restriction fragments were resolved under the running conditions described by Turabelidze et al. (28). Macrorestriction patterns were compared by the use of BioNumerics software (Applied Maths, Sint-Martens-Latem, Belgium). Antimicrobial susceptibility testing. Resistance patterns and MICs were ascertained for 16 different therapeutic and growth-promoting antimicrobials by a standard Sensititre protocol (Trek Diagnostic Systems Ltd., England). Table 1 gives the concentrations of antimicrobials tested and the breakpoint used for each drug (DANMAP 2002). Briefly, isolates were grown in yeast extract agar. Subsequently, a 0.5 McFarland cell suspension was prepared in demineralized water, and 10 l was inoculated into 10 ml of cation-adjusted Mueller-Hinton broth (Trek Diagnostic Systems, Ohio) for a final inoculum of 10 5 CFU/ml. Aliquots of 50 l of the inoculum were seeded in each well of a microtiter plate, which contained doubling dilutions of the antimicrobials. The plates were sealed and incubated aerobically at 37 C for 24 h. A growth control well was used as a reference for interpreting the growth patterns in each plate. Pseudomonas aeruginosa (ATCC 27853), Enterococcus faecalis (ATCC 29213), S. aureus (ATCC 29213), and Escherichia coli (ATCC 25922 and ATCC 35218) were used as quality control organisms. The MIC was recorded as the lowest concentration of antimicrobial that inhibited visible growth. RESULTS Concn range ( g/ml) Breakpoint ( g/ml) Nitrofurantoin 64 256 128 Penicillin 2 128 16 Tetracycline 1 32 16 Erythromycin 1 32 8 Ciprofloxacin 0.5 16 4 Gentamicin 128 2,048 1,024 Streptomycin 128 2,048 2,048 Kanamycin 128 2,048 2,048 Quinupristin-dalfospristin 0.5 32 4 Vancomycin 1 64 32 Teicoplanin 2 64 16 Chloramphenicol 2 64 32 Florfenicol 2 32 32 Bacitracin 8 256 128 Flavomycin 0.5 32 16 Salinomycin 1 32 16 A total of 217 VREF isolates were obtained from fecal and environmental samples during the study period (January 2002 to February 2003). When comparing the sensitivities of the two different sampling techniques (sampling by farmers versus sampling by the research team) for the detection of VREF by using the unpaired t test, we found no evidence to reject the null hypothesis that there was no difference between the two collection methods (P 0.457). Broiler farms. Twenty-seven of the 33 farms investigated were found to have at least one VREF-positive sample on at least one occasion over the course of the study. All 20 conventional broiler farms that submitted pooled fecal samples by post were found to be positive for VREF. The number of fecal samples from these farms which contained VREF ranged from

VOL. 43, 2005 VANCOMYCIN RESISTANCE IN E. FAECIUM 3285 one of eight to eight of eight. Of the further 13 farms that were visited in person, 35 to 100% of fecal samples from three of the six conventional broiler farms were found to be positive for VREF on every visit. Environmental swabs taken on these farms were also positive. On one conventional farm, VREF isolates were found in environmental swabs collected from the surfaces of two houses that had been cleaned and disinfected, although the organisms had not been found in fecal samples collected from the previous flock. Fecal VREF isolates were also found on three visits to two organic farms, but at a lower prevalence than that for the conventional farms (3 to 10% of fecal samples). On another organic farm, VREF isolates were found in environmental samples, but not in fecal samples, on one of five visits. Pig farms. Fewer VREF isolates were isolated from pig farms in this study, although 4 of the 14 farms investigated did yield at least one positive sample. The two conventional finisher units that sent pooled fecal samples to the laboratory were found to have one of eight and five of eight positive samples. Fecal samples from 2 of the 12 farms that were sampled more intensively were positive for VREF at low levels (2 to 5%) on a single visit. One of these farms was conventional and the other was organic. We failed to find any VREF on a second visit to every one of those 12 farms. Genotyping. All VREF isolates harbored the vana resistance gene and all selected isolates were esp negative. They all exhibited MICs of 64 g/ml of vancomycin and between 8 g/ml and 64 g/ml of teicoplanin. One hundred nineteen PFGE restriction profiles were identified among the 217 isolates by SmaI-PFGE (Fig. 1). Each profile was given a unique identification number. The number of fragments generated ranged from 16 to 24, and their sizes varied from 20 to 350 kb. Our analysis revealed a predominant PFGE profile for each farm together with less represented types. The most prevalent type from one visit was never detected again in isolates from a second visit. Two isolates from farms 3 and 8 were highly related according to PFGE. In order to check the stability of the PFGE types, we repeatedly passaged one colony of each of two presumptively related isolates in the laboratory. PFGE was then repeated with the new subcultures. During the third passage, a new band appeared in the PFGE profile of one of the isolates. This suggests that small changes in PFGE profiles may occur in short periods of time. Antimicrobial resistance. Ninety-three percent of the VREF isolates were resistant to penicillin (MIC, 16 g/ml), 89% were resistant to tetracycline (MIC, 16 g/ml), 88% were resistant to erythromycin (MIC, 8 g/ml), 79% were resistant to streptomycin (MIC, 2,048 g/ml), and 50% were resistant to quinupristin-dalfospristin (Synercid) (MIC, 4 to 32 g/ml). One isolate displayed resistance to chloramphenicol, with an MIC of 32 g/ml, and 8% of the isolates were found to be of intermediate resistance (MIC, 16 g/ml). None of the strains exhibited high-level resistance to gentamicin, and all were sensitive to florfenicol (MIC, 32 g/ml). With the exception of vancomycin and teicoplanin, no correlation among resistances to several antimicrobials was found in this study. An analysis of the resistance profiles indicated that 166 (76%) of the 217 VREF isolates tested were resistant to nine or more antimicrobials, comprising a diverse range of phenotypes (n 70). Eleven percent of the isolates were resistant to the same 12 antimicrobials, as follows: bacitracin (MIC, 256 g/ml), ciprofloxacin (MIC, 4 g/ml), erythromycin (MIC, 32 g/ml), flavomycin (MIC, 32 g/ml), kanamycin (MIC, 2,048 g/ml), nitrofurantoin (MIC, 128 to 256 g/ml), penicillin (MIC, 16 to 128 g/ml), streptomycin (MIC, 2,048 g/ml), quinupristin-dalfospristin (MIC, 4 to 16 g/ml), tetracycline (MIC, 32 g/ml), vancomycin (MIC, 64 g/ml), and teicoplanin (MIC, 64 g/ml). These 24 isolates were all obtained from different conventional poultry farms. Finally, 23% of the isolates were resistant to 11 antimicrobials, 25% were resistant to 10, 17% were resistant to 9, and 11.4% were resistant to 8. Table 2 shows the distribution of genotypes and phenotypes on farms with more than one VREF isolate. DISCUSSION VREF was found in samples from 31 of 47 farms (66%). However, we identified differences in the percentages of positive pooled fecal samples between pig and poultry farms. Furthermore, VREF isolates were repeatedly isolated with a relatively high sample prevalence from conventional broiler farms. Far fewer samples from organic poultry yielded VREF, and VREF was never isolated from any organic broiler farm on more than one occasion. In the United Kingdom, certified organic farms must agree to relatively restrictive policies in terms of antimicrobial usage (http://www.defra.gov.uk/farm/organic/legislation-standards). For instance, organic farms are not permitted to use antimicrobial growth-promoting agents, but they may administer therapeutic treatment if this is judged to be clinically necessary by a veterinary surgeon. None of the participating organic poultry farms used any antimicrobials during this study, and most of them (6/7) had not used such treatment within the 12 months preceding the study. On one conventional farm, VREF was detected in samples originating from floors, walls, wooden roof supports, and dust taken after cleaning and disinfection. This indicates that insufficient cleaning and disinfection may play a role in the persistence of these organisms. Interestingly, all isolates from this farm (which used no therapeutic antimicrobials but did use in-feed avilamycin) were of a single clone. The same pattern was also detected in a positive nondomestic bird sample collected from the concrete concourse outside the houses and in dust samples from an occupied house collected during the same visit. Suitable routine hygiene protocols should be performed thoroughly to prevent the persistence of these organisms on surfaces and to minimize colonization by resistant strains when new flocks enter the premises. Multidrug-resistant VREF has been described previously (1, 2, 14), but the level of multiresistance among isolates found in the present study appears to be unprecedented. To date, such multiresistant profiles have not been reported elsewhere in Europe for animal sources. It is difficult to evaluate the significance of these differences since the use of antimicrobials is a common practice in most European countries. It is also difficult to assess the significance of potential horizontal gene transfer between other bacteria and E. faecium clones in our study, and this would require a detailed characterization of the different genetic elements involved.

3286 GARCIA-MIGURA ET AL. J. CLIN. MICROBIOL.

VOL. 43, 2005 VANCOMYCIN RESISTANCE IN E. FAECIUM 3287 TABLE 2. Distribution of dominant phenotypes and genotypes on farms Farm no. (date of study [day/mo/yr]) No. of VREF isolates No. of phenotypes Dominant phenotype(s) a From the medical point of view, the emergence of multiresistance among VREF isolates is a great cause for concern, since some of the antimicrobials involved are commonly used for the treatment of human VRE infections. This is well illustrated by the examples of gentamicin and streptomycin, which have a synergistic effect when administered with cell wall inhibitors such as vancomycin (18). A high-level resistance to aminoglycosides might pose a serious risk in hospitals, as antimicrobial therapy could be limited. In this study, while no isolates were resistant to high levels of gentamicin, 70% were resistant to high levels of streptomycin. A combination of streptogramins (quinupristin-dalfospristin [Synercid]) has been successfully used for the treatment of VREF infections (27). However, some E. faecium strains have already acquired resistance to these antibiotics. This resistance may be related in % of isolates belonging to dominant type No. of genotypes Dominant genotype(s) % of isolates belonging to dominant type 1 32 10 BcEyFaKnPnSrSnTtVnTi 43.8 13 80 25 2 4 3 CpEyFaPnSrTtVnTi 50 2 66 50 26 50 3 7 5 CpEyFaPnSrTtVnTi 42.9 1 102 100 5 (02/09/02) 18 9 BcCpEyFaKnNtPnSrTtVnTi 33.3 9 50 33.3 5 (24/02/03) 23 12 BcCpEyFaKnNtPnSrTtVnTi 26.1 19 44 13 6 5 3 BcFaKnPnTtVnTi 60 3 10 60 10 (11/11/02) 10 7 BcFaPnSrTtVnTi 30 1 78 100 11 10 6 BcEyFaKnNtPnSrSnTtVnTi 40 7 94 30 12 2 2 BcEyFaVnTi 50 2 24 50 BcEyFaPnVnTi 50 23 50 13 7 4 BcCpEyFaKnNtPnTtVnTi 57.1 2 103 85.7 14 5 1 BcEyFaVnTi 100 3 32 40 34 40 15 5 4 EyFaKnPnSrSnVnTi 40 4 27 40 16 5 4 BcCpEyFaKnNtPnSrSnTtVnTi 40 4 85 40 17 6 2 BcCpEyFaKnNtPnSrSnTtVnTi 66.7 5 115 33.3 18 3 2 BcCpEyFaKnNtPnSrSnTtVnTi 66.7 2 14 66.7 19 5 4 BcCpEyFaKnPnSrSnVnTi 40 4 112 40 20 b (18/02/02) 9 5 BcEyFaKnNtPnTtVnTi 33.3 1 38 100 BcCpEyFaNtPnTtVnTi 33.3 20 b (02/09/02) 3 2 BcCpEyFaKnNtPnSrTtVnTi 66.7 3 65 66.7 21 8 6 BcCpEyFaKnNtPnSrTtVnTi 37.5 6 96 37.5 23 b 6 3 BcEyFaKnNtPnSrTtVnTi 66.7 1 35 100 24 c 5 2 EyFaPnSnTtVnTi 60 1 3 100 25 7 6 BcEyFaNtPnSnTtVnTi 28.6 2 41 71.4 26 4 4 None 2 18 17 50 50 27 4 3 BcEyFaKnNtPnSrSnTtVnTi 50 3 12 50 28 9 4 BcEyFaKnNtPnSrSnTtVnTi 44.4 7 16 22.2 84 22.2 29 6 5 BcCpFaKnNtPnSrSnTtVnTi 33.3 4 53 55 30 b 2 2 EyFaNtVnTi 50 2 71 50 BcEyFaKnPnSrTtVnTi 50 37 50 a Bc, bacitracin; Cp, ciprofloxacin; Ey, erythromycin; Fa, flavomycin; Kn, kanamycin; Nt, nitrofurantoin; Pn, penicillin; Sr, streptomycin; Sn, synercid; Tr, tetracycline; Vn, vancomycin; Ti, teicoplanin. b Organic farm. c Pig farm. 33.3 33.3 part to the use of virginiamycin, a growth-promoting feed additive incorporated in agriculture for poultry and pig production (32). The use of virginiamycin was banned in Denmark in 1998 and in the rest of the European Union in 1999. In the present study, 50% of the isolates were resistant to quinupristin-dalfospristin. Erythromycin resistance was also found at very high levels on every farm. Macrolide resistance encoded by ermb-type genes has been linked to the same conjugative plasmid harboring vana genes (3). Further studies are being carried out to ascertain the possible role of gene linkage and the coselection of VREF on farms in the United Kingdom as a result of the use of macrolides. The multiresistance profiles found in this study do not seem to be specific to a particular farm or sample type, and they do not correlate with any specific PFGE banding pattern. Al- FIG. 1. Dendrogram generated by Gel Compar II software showing the relationships of 119 representative fingerprints (SmaI-PFGE types) for 217 VREF isolates. The analysis of the generated bands was performed by using the Dice coefficient and the unweighted-pair group method with arithmetic averages (optimization of 0.00% and position tolerance of 1.00%).

3288 GARCIA-MIGURA ET AL. J. CLIN. MICROBIOL. though PFGE is still considered the standard typing method for enterococci, there are no standardized criteria for analyzing PFGE patterns (16). Therefore, the interpretation of the results can lead to different conclusions. In addition, the lack of standardization in terms of PFGE running conditions makes interlaboratory comparisons rather limited. The issue of how many band differences account for the description of a new clone has not yet been resolved. This is especially true for long-term studies. In the present study, there was only one case of related PFGE types being isolated on two different premises that were spatially separated and managed by different farmers. Farm 8 was a conventional poultry farm which supplied stock to farm 3 (organic poultry farm). Interestingly, similar PFGE profiles were found on each of these farms, differing in only one band. The link between farms was curtailed during the course of the study, and subsequent visits to farm 3 did not recover further VREF isolates. This event might suggest a transfer of resistant strains from farm to farm. PFGE has provided evidence of the high level of genetic diversity among E. faecium populations in farm environments in the United Kingdom. The PFGE results of our study suggest that the introduction of new clones to the farm by deliveries of new stock or by other reservoirs (e.g., other domestic and wild animals, feed, litter, and water), rather than the persistence of resistant clones, may be the cause for the persistence of vancomycin resistance on these farms. Alternatively, we may have observed the effects of the dynamic interaction of bacterial populations, by which the previously detected clone called the dominant type may still be present on the farm on subsequent visits, but at levels below the limit of detection, whereas new clones that previously survived in small numbers may increase to detectable levels for unknown reasons. Management practices such as the use of disinfectants, sources of replacement stock, or even interactions with other enteric flora might have an important impact on the selection of the new dominant bacterial population. VREF isolates from different habitats are very polyclonal, suggesting the horizontal gene transfer of the vancomycin resistance genes rather than the spread of a single clone. Therefore, if we cannot find the same clones in different environments, but we are able to find the same resistance genes, some clear questions arise. Where is the transfer of van and other genes occurring? Are farm animals a significant long-term reservoir of these resistance genes, and if so, since avoparcin has not been administered to livestock since 1999, are we coselecting for vancomycin resistance by the use of other compounds? There may also be other undetermined factors indirectly selecting for the persistence of the vancomycin resistance genes if these organisms are more suited for environmental survival. The esp gene encodes an enterococcal surface protein (Esp), which contributes to the colonization and infection of the urinary tract by increasing attachment to epithelial surfaces and biofilm production (22). This gene appears to be an enterococcus-specific virulence factor which is highly conserved in E. faecium subpopulations involved in hospital outbreaks (31), independent of the vancomycin susceptibility (23). The isolates tested in our study were negative for this element, which suggests that farm animals may not be a significant source of these genes. The absence of esp genes in our isolates indicates that it is unlikely that they could cause infections in humans, although in vitro conjugative transfer of the esp gene has been demonstrated (21). In our study, we detected antimicrobial resistance even on farms where antimicrobials had not been used for many years, if at all. The factors promoting the persistence of resistant bacteria or resistance genes are not clear. Ideally, farm antimicrobial usage and hospital policies should be implemented to minimize the further development, spread, and persistence of resistant organisms. Statistically representative surveys should be carried out to detect and quantify specific genes in agricultural systems as well as in hospital environments. In addition, epidemiological studies to help to unravel the mechanisms underlying the observed heterogeneity of VREF isolates should be attempted. This would provide useful information to help to prevent and ultimately control the spread of antimicrobial resistance among bacteria. ACKNOWLEDGMENTS We gratefully acknowledge Defra for funding project OD2006. Lourdes Garcia-Migura is a Ph.D. student registered with the University of Liverpool. We also thank Felicity Clifton-Hadley for her corrections and suggestions and the participating farmers for letting us undertake this research on their farms. REFERENCES 1. Aarestrup, F. M., H. Hasman, L. B. Jensen, M. Moreno, I. A. Herrero, L. Dominguez, M. Finn, and A. Franklin. 2002. Antimicrobial resistance among enterococci from pigs in three European countries. Appl. Environ. Microbiol. 68:4127 4129. 2. Aarestrup, F. M., H. Kruse, E. Tast, A. M. Hammerum, and L. B. Jensen. 2000. Associations between the use of antimicrobial agents for growth promotion and the occurrence of resistance among Enterococcus faecium from broilers and pigs in Denmark, Finland, and Norway. Microb. Drug Resist. 6:63 70. 3. Aarestrup, F. M., A. M. Seyfarth, H. D. Emborg, K. Pedersen, R. S. Hendriksen, and F. Bager. 2001. Effect of abolishment of the use of antimicrobial agents for growth promotion on occurrence of antimicrobial resistance in fecal enterococci from food animals in Denmark. Antimicrob. Agents Chemother. 45:2054 2059. 4. Bager, F., M. Madsen, J. Christensen, and F. M. Aarestrup. 1997. Avoparcin used as a growth promoter is associated with the occurrence of vancomycinresistant Enterococcus faecium on Danish poultry and pig farms. Prev. Vet. Med. 31:95 112. 5. Bates, J. 1997. Epidemiology of vancomycin-resistant enterococci in the community and the relevance of farm animals to human infection. J. Hosp. Infect. 37:89 101. 6. Bates, J., J. Z. Jordens, and D. T. Griffiths. 1994. Farm animals as a putative reservoir for vancomycin-resistant enterococcal infection in man. J. Antimicrob. Chemother. 34:507 514. 7. Berchieri, A. 1999. Intestinal colonization of a human subject by vancomycinresistant Enterococcus faecium. Clin. Microbiol. Infect. 5:97 100. 8. Bonten, M. J., R. Willems, and R. A. Weinstein. 2001. Vancomycin-resistant enterococci: why are they here, and where do they come from? Lancet Infect. Dis. 1:314 325. 9. Borgen, K., Y. Wasteson, H. Kruse, and R. J. Willems. 2002. Vancomycinresistant Enterococcus faecium (VREF) from Norwegian poultry cluster with VREF from poultry from the United Kingdom and The Netherlands in an amplified fragment length polymorphism genogroup. Appl. Environ. Microbiol. 68:3133 3137. 9a.Centers for Disease Control and Prevention. Standardized protocol for molecular subtyping of Listeria monocytogenes by PFGE. CDC, Atlanta, Ga. [Online.] http://www.cdc.gov/pulsenet/. 10. Coque, T. M., J. F. Tomayko, S. C. Ricke, P. C. Okhyusen, and B. E. Murray. 1996. Vancomycin-resistant enterococci from nosocomial, community, and animal sources in the United States. Antimicrob. Agents Chemother. 40: 2605 2609. 11. Dutka-Malen, S., S. Evers, and P. Courvalin. 1995. Detection of glycopeptide resistance genotypes and identification to the species level of clinically relevant enterococci by PCR. J. Clin. Microbiol. 33:24 27. 12. Giraffa, G. 2002. Enterococci from foods. FEMS Microbiol. Rev. 26:163 171. 13. Huycke, M. M., D. F. Sahm, and M. S. Gilmore. 1998. Multiple-drug resis-

VOL. 43, 2005 VANCOMYCIN RESISTANCE IN E. FAECIUM 3289 tant enterococci: the nature of the problem and an agenda for the future. Emerg. Infect. Dis. 4:239 249. 14. Iversen, A., I. Kuhn, A. Franklin, and R. Mollby. 2002. High prevalence of vancomycin-resistant enterococci in Swedish sewage. Appl. Environ. Microbiol. 68:2838 2842. 15. Martone, W. J. 1998. Spread of vancomycin-resistant enterococci: why did it happen in the United States? Infect. Control Hosp. Epidemiol. 19:539 545. 16. Morrison, D., N. Woodford, S. P. Barrett, P. Sisson, and B. D. Cookson. 1999. DNA banding pattern polymorphism in vancomycin-resistant Enterococcus faecium and criteria for defining strains. J. Clin. Microbiol. 37:1084 1091. 17. Moubareck, C., N. Bourgeois, P. Courvalin, and F. Doucet-Populaire. 2003. Multiple antibiotic resistance gene transfer from animal to human enterococci in the digestive tracts of gnotobiotic mice. Antimicrob. Agents Chemother. 47:2993 2996. 18. Murray, B. E. 1990. The life and times of the Enterococcus. Clin. Microbiol. Rev. 3:46 65. 19. Murray, B. E. 1998. Diversity among multidrug-resistant enterococci. Emerg. Infect. Dis. 4:37 47. 20. Noble, W. C., Z. Virani, and R. G. Cree. 1992. Co-transfer of vancomycin and other resistance genes from Enterococcus faecalis NCTC 12201 to Staphylococcus aureus. FEMS Microbiol. Lett. 72:195 198. 21. Oancea, C., I. Klare, W. Witte, and G. Werner. 2004. Conjugative transfer of the virulence gene, esp, among isolates of Enterococcus faecium and Enterococcus faecalis. J. Antimicrob. Chemother. 54:232 235. 22. Shankar, N., C. V. Lockatell, A. S. Baghdayan, C. Drachenberg, M. S. Gilmore, and D. E. Johnson. 2001. Role of Enterococcus faecalis surface protein Esp in the pathogenesis of ascending urinary tract infection. Infect. Immun. 69:4366 4372. 23. Shankar, V., A. S. Baghdayan, M. M. Huycke, G. Lindahl, and M. S. Gilmore. 1999. Infection-derived Enterococcus faecalis strains are enriched in esp, a gene encoding a novel surface protein. Infect. Immun. 67:193 200. 24. Sorensen, T. L., M. Blom, D. L. Monnet, N. Frimodt-Moller, R. L. Poulsen, and F. Espersen. 2001. Transient intestinal carriage after ingestion of antibiotic-resistant Enterococcus faecium from chicken and pork. N. Engl. J. Med. 345:1161 1166. 25. Stobberingh, E., A. van den Bogaard, N. London, C. Driessen, J. Top, and R. Willems. 1999. Enterococci with glycopeptide resistance in turkeys, turkey farmers, turkey slaughterers, and (sub)urban residents in the south of The Netherlands: evidence for transmission of vancomycin resistance from animals to humans? Antimicrob. Agents Chemother. 43:2215 2221. 26. Tenover, F. C., L. M. Weigel, P. C. Appelbaum, L. K. McDougal, J. Chaitram, S. McAllister, N. Clark, G. Killgore, C. M. O Hara, L. Jevitt, J. B. Patel, and B. Bozdogan. 2004. Vancomycin-resistant Staphylococcus aureus isolate from a patient in Pennsylvania. Antimicrob. Agents Chemother. 48:275 280. 27. Thal, L. A., and M. J. Zervos. 1999. Occurrence and epidemiology of resistance to virginiamycin and streptogramins. J. Antimicrob. Chemother. 43: 171 176. 28. Turabelidze, D., M. Kotetishvili, A. Kreger, J. G. Morris, Jr., and A. Sulakvelidze. 2000. Improved pulsed-field gel electrophoresis for typing vancomycin-resistant enterococci. J. Clin. Microbiol. 38:4242 4245. 29. van Den Bogaard, A. E., R. Willems, N. London, J. Top, and E. E. Stobberingh. 2002. Antibiotic resistance of faecal enterococci in poultry, poultry farmers and poultry slaughterers. J. Antimicrob. Chemother. 49:497 505. 30. Weigel, L. M., D. B. Clewell, S. R. Gill, N. C. Clark, L. K. McDougal, S. E. Flannagan, J. F. Kolonay, J. Shetty, G. E. Killgore, and F. C. Tenover. 2003. Genetic analysis of a high-level vancomycin-resistant isolate of Staphylococcus aureus. Science 302:1569 1571. 31. Willems, R. J., W. Homan, J. Top, M. van Santen-Verheuvel, D. Tribe, X. Manzioros, C. Gaillard, C. M. Vandenbroucke-Grauls, E. M. Mascini, E. van Kregten, J. D. van Embden, and M. J. Bonten. 2001. Variant esp gene as a marker of a distinct genetic lineage of vancomycin-resistant Enterococcus faecium spreading in hospitals. Lancet 357:853 855. 32. Witte, W. 1997. Impact of antibiotic use in animal feeding on resistance of bacterial pathogens in humans. Ciba Found. Symp. 207:61 71. Downloaded from http://jcm.asm.org/ on January 23, 2019 by guest