AEM
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hasman, H.
Right arrow Articles by Aarestrup, F. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hasman, H.
Right arrow Articles by Aarestrup, F. M.
Agricola
Right arrow Articles by Hasman, H.
Right arrow Articles by Aarestrup, F. M.

 Previous Article  |  Next Article 

Applied and Environmental Microbiology, September 2006, p. 5784-5789, Vol. 72, No. 9
0099-2240/06/$08.00+0     doi:10.1128/AEM.02979-05
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Copper Resistance in Enterococcus faecium, Mediated by the tcrB Gene, Is Selected by Supplementation of Pig Feed with Copper Sulfate

Henrik Hasman,1* Isabelle Kempf,2 Bérangère Chidaine,2 Roland Cariolet,3 Annette Kjær Ersbøll,4 Hans Houe,4 Hans Christian Bruun Hansen,4 and Frank Møller Aarestrup1

Danish Institute for Food and Veterinary Research, Bülowsvej 27, 1790 Copenhagen V, Denmark,1 Unité de Mycoplasmologie-Bactériologie, Agence Française de Sécurité Sanitaire des Aliments, BP53, 22440 Ploufragan, France,2 Service de Production de Porcs Assainis et d'Expérimentation, Agence Française de Sécurité Sanitaire des Aliments, BP53, 22440 Ploufragan, France,3 The Royal Veterinary and Agricultural University, Grønnegårdsvej 8, 1870 Frederiksberg C, Denmark4

Received 16 December 2005/ Accepted 17 June 2006


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The tcr gene cluster mediates in vitro copper resistance in Enterococcus faecium. Here we describe the selection of tcr-mediated copper resistance in E. faecium in an animal feeding experiment with young pigs fed 175 mg copper/kg feed (ppm), which is the concentration commonly used for piglets in European pig production. tcr-mediated copper resistance was not selected for in a control group fed low levels of copper (6 ppm). We also show coselection of macrolide- and glycopeptide-resistant E. faecium in the animal group fed the high level of copper. Finally, we identify the tcr genes in the enterococcal species E. mundtii, E. casseliflavus, and E. gallinarum for the first time.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Copper is used extensively as a growth-promoting agent in both industrial and organic pig production. In Europe, copper is added to the feed at a concentration of 170 to 175 ppm for piglets (up to 12 weeks of age) and a concentration of 25 to 35 ppm for growing and finishing pigs. In contrast to this, copper supplements are not restricted in the United States. In the United States, typical amounts range from 125 to 250 ppm, independent of the age of the pigs (Gawain Willis, Primary Nutrition LLC, personal communication). These concentrations of copper are far above the minimal nutritional requirement of the animals, which depends in part on the age of the animal as well as the presence of additional nutritional components in the feed (13). At a minimum, piglets require 5 to 6 ppm, while growing pigs and slaughter pigs require 3 to 5 ppm copper for normal growth. Copper recommendations to ensure normal growth of the animals range between 4 and 10 ppm (8).

The growth-promoting effect of copper supplementation has been documented in several studies. The authors of the SCAN report (8) analyzed a large number of dose-response studies involving copper supplements given to pigs of different ages, ranging from the postweaning period (up to 25 kg body weight) to growing pigs (25 to 60 kg body weight) and finishing pigs (above 60 kg body weight). It was concluded that the presence of large amounts of copper (from approximately 65 ppm up to 280 ppm) does improve the growth rate of the youngest age group significantly, but no significant effect was observed for growing and finishing pigs.

Copper is an essential trace element in many biological functions of all living cells. However, copper ions at high levels are toxic to most cells. To survive this toxicity, many bacterial species have developed copper resistance mechanisms. Acquired bacterial copper resistance has been described for both gram-positive and gram-negative organisms (3, 4, 10, 14). Acquired copper resistance mediated by the plasmid-localized tcrB gene has been described for Enterococcus faecium and E. faecalis isolated from both production animals and humans in several different European countries (3, 10). The tcrB gene was recently shown to be one of four genes in an operon structure called the tcrYAZB operon, which is similar to the copYZAB copper homeostasis operon from E. hirae (9). Bacteria carrying the plasmid-borne tcrYAZB operon are able to grow in the presence of up to 28 mM copper sulfate in vitro, whereas bacteria lacking this gene can tolerate copper concentrations of only up to 8 mM (10). However, selection of copper-resistant E. faecium occurs in vitro, even at lower concentrations of copper than the MIC of the susceptible isolates. In fact, a copper concentration of 3 mM is sufficient to create a 24-fold enrichment of copper-resistant isolates within fewer than 35 bacterial generations compared to an isogenic copper-susceptible mutant (9). In contrast to this, 2 mM copper sulfate does not seem to select for copper resistance among E. faecium under the same conditions. Interestingly, 175 ppm corresponds to a concentration of approximately 2.8 mM, leading to a theoretical risk that such high concentrations of copper can lead to selection of copper resistance. However, several factors in the gut, such as the pH, copper speciation ([Cu+] versus [Cu2+]), adsorption, and complex formation with organic material, have an influence on the actual free (reactive) copper concentration. Therefore, it remains elusive whether the concentrations of copper used for growth promotion in pig production are able to select for copper-resistant bacteria.

In Denmark, glycopeptides were banned for animal production in 1995 and macrolides were banned for growth promotion in 1998. After the ban of glycopeptides for pig production in 1995, the level of glycopeptide-resistant E. faecium (GRE) isolated from pigs in Denmark did not change. Only after the ban of macrolides for growth promotion in 1998 was a significant decrease in the level of GRE observed. This enigma was later shown to be caused by coselection of the glycopeptide resistance phenotype through linkage to macrolide resistance, as these resistance determinants turned out to be located on the same plasmid in all Danish GRE isolates (1, 12). Since 1998, the level of glycopeptide- and macrolide-resistant enterococci has been decreasing, but isolates resistant to these antibiotics have not disappeared completely (11). Since macrolides are still used for the treatment of sick animals, this could account for the persistence of macrolide- and glycopeptide-resistant bacteria long after the ban. However, a close genetic link between tcrB and the genetic determinants of macrolide resistance [erm(B)] and glycopeptide resistance (vanA) has been demonstrated for E. faecium isolated from pigs and could be involved in this persistence of resistance towards glycopeptides and macrolides (10, 11). It was further shown that the tcrB, erm(B), and vanA genes are often located together on the same transferable plasmids in E. faecium isolated from pigs (12). In fact, >90% of copper-resistant E. faecium strains isolated from pigs in Denmark in 1998 were simultaneously resistant to macrolides, and >20% were resistant to both macrolides and glycopeptides (10). This could lead to a situation where the use of copper as a growth-promoting agent at high concentrations could result in coselection of these antibiotic determinants and thus counteract the effect of the ban.

The purpose of this study was to examine whether the copper concentration (175 ppm versus 6 ppm) used for the growth promotion of piglets is able to select for copper resistance among E. faecium isolates in vivo and to test whether this could result in coselection of macrolide- and glycopeptide-resistant bacteria.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Strains and bacterial growth conditions.
Ten tcrB-negative (copper-sensitive) and 10 tcrB-positive (copper-resistant) E. faecium strains were used in this study (Table 1). All strains were isolated in 2002 from healthy animals as part of the DANMAP surveillance program (6) and were tested phenotypically and genotypically for resistance towards macrolides and glycopeptides, as described below.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Names and relevant genotypes of strains used to inoculate animals on day 0

 
The 20 strains were streaked on Slanetz agar (BD Difco) plates and grown for 48 h at 42°C. Each strain was inoculated separately into 25 ml of brain heart infusion broth at 37°C for 24 h. The cell densities of all strains were measured, and strains were subsequently mixed in equal numbers (with approximately 109 bacteria/ml of each strain) to be inoculated in the pigs as described below.

Animals.
Twenty-four specific-pathogen-free Large White pigs of 8 weeks of age, obtained from the experimental swine herd of AFSSA (Ploufragan, France), were divided into two groups consisting of four experimental units (pens 1, 2, 3, and 4), with each unit containing six animals. The 24 pigs were obtained from three sows and were randomized (balanced) before the experiment. Pens 1 and 2 were replicates of the same treatment, as were pens 3 and 4. Pigs in pens 1 and 2 were fed a low-copper feed both prior to and during the whole experiment. Pigs in pens 3 and 4 were fed a low-copper feed up to the inoculation day (day 0) and a high-copper feed just after inoculation. The mean animal weights of the animals in the four pens on inoculation day (day 0) were 21.5 to 22.9 kg. Very strict biosecurity measures were implemented in order to avoid contamination of the pigs, including the use of an air filtration system and airlocks for each unit, the use of unit-specific clothes, and compulsory showering after visiting the pigs. During the week, daily clinical examinations consisted of looking for general clinical signs and taking rectal temperatures. On these occasions, the consistency of fecal material was evaluated and scored as follows: 0, solid feces; 1, normal; 2, slightly soft; 3, cow dung; 4, diarrhea; and 5, very liquid. Body weight was also recorded each week during the experiment. Pigs were euthanized on days 35 to 37 postinoculation, and lesions were observed. The animals did not receive any treatment with any antibiotics prior to or during the feeding experiment.

Animal feed.
The feed offered to the animals before inoculation was not supplemented with copper and consisted of a 16% protein, low-copper feed (7.3% [weight/dry weight] minerals, 92.7% [weight/dry weight] organic material, and 6.4 mg of copper/kg of body weight [dry weight, approximately 5.2 mg/kg ± 0.26 mg/kg when corrected for water]). Immediately after bacterial inoculation (day 0), animals in pens 1 and 2 were fed the same feed as before the inoculation, while the animals in pens 3 and 4 were shifted to a high-copper diet containing the same 16% protein feed but with copper sulfate added to a final concentration of 208 mg/kg of copper (dry weight, approximately 179 mg/kg ± 17 mg/kg). The contents of copper in the low- and high-copper feeds were determined after sample grinding followed by digestion of 0.25 g in 6.00 ml 70% HNO3 and 5.00 ml 30% H2O2 for 1 day. Finally, 2.00 ml of 30% HCl was added to the samples and kept at 45°C for 1 day before analysis; four feed samples at each copper level were analyzed. The amounts of copper in diluted digests were determined by graphite furnace atomic absorption spectroscopy using a Perkin-Elmer 5100 Zeeman instrument.

Inoculation.
At the beginning of the experiment (day 0), fecal samples were taken from each of the animals, and pigs were left without food for 4 hours. The animals in the four pens were individually fed 20 ml of the mixture of copper-sensitive and copper-resistant E. faecium strains described above (approximately 109 bacteria of each strain) to ensure that all animals received the same inoculum size.

Sampling.
Fecal samples were collected from each pig 0, 7, 14, 21, and 28 days after inoculation for isolation of E. faecium.

Sample handling and bacterial cultivation and isolation.
Approximately 1 g of feces was diluted 1:10 in tryptone salt broth (a 10–1 dilution). The samples were mixed thoroughly by rotating them for 1 hour at room temperature on an automatic rotator-agitator and then further diluted 10–2 to 10–4. From these tubes, three replicates of 100 µl were spread on Slanetz agar plates (Difco). The plates were incubated at 42°C for 2 days. From each dilution row (10–1 to 10–4 dilutions) belonging to a fecal sample, 15 colonies were randomly chosen and restreaked on blood agar plates as well as agar plates containing tellurite to exclude E. faecalis strains. Of the 15 colonies belonging to a fecal sample, six E. faecium-like isolates were selected, grown in brain heart infusion broth, and frozen after the addition of glycerol (20%) for further characterization.

Species identification.
Species identification of all isolates was performed using motility assessment and sugar fermentation as described previously (5). Selected isolates were further tested using multiplex PCR as described previously (7), and non-E. faecium isolates were identified to the species level by sequencing of PCR-amplified 16S rRNA gene fragments. Primers for amplification and sequencing of 16S rRNA genes were primer 442 (5'-GAC TAC CNG GGT ATC TAA TCC-3') and primer 444 (5'-AGA GTT TGA TCC TGG CTN AG-3').

Detection of the tcrB gene by PCR.
The presence of the tcrB gene was detected by a multiplex PCR consisting of two specific primers for the detection of tcrB and two universal primers for the detection of the 16S rRNA gene, which served as an internal control for the PCRs. Primers designed for the specific detection of tcrB based on the previously submitted tcrB sequence (GenBank accession no. AY048044) originating from the glycopeptide-resistant E. faecium strain A17sv1 were primer 824 (5'-CAT CAC GGT AGC TTT AAG GAG ATT TTC-3') and primer 825 (5'-ATA GAG GAC TCC GCC ACC ATT G-3'), leading to a DNA product of 663 bp. Universal 16S rRNA gene primers were primer 718 (5'-ACG AGC TGA CGA CRR CCA TG-3') and primer 1175 (5'-CAG GAT TAG ATA CCC NGG TAG TC-3'), leading to a DNA product of 292 bp. The following assay conditions were used for all PCRs: 5 min at 94°C and then 30 cycles of 30 s at 94°C, 30 s at 55°C, and 1 min at 72°C, with a final step of 10 min at 72°C and holding at 10°C. E. faecium strain A17sv1 served as a positive control for PCR.

Detection of macrolide and glycopeptide resistance.
All isolates were tested for resistance towards macrolides (erythromycin) and glycopeptides (vancomycin), using disc diffusion susceptibility testing (Oxoid) according to CLSI recommendations (15). Furthermore, the 20 isolates used for inoculation were all tested for the presence of the vanA and erm(B) genes as described previously (1), as were all isolates which were examined by pulsed-field gel electrophoresis (PFGE).

Phenotypic detection of copper resistance.
The phenotypic copper resistance of selected isolates was tested on Mueller-Hinton agar (Difco) plates containing 0, 4, 8, 12, 16, 20, 24, 28, and 32 mM copper sulfate (CuSO4 · 5H2O) adjusted to pH 7.3 with 1 M NaOH. Isolates were picked from a blood agar plate and diluted to a McFarland standard of 0.5 (~108 cells/ml) in 0.9% NaCl. From each suspension, a single drop (approximately 5 x 105 cells) was used for inoculation, as suggested by the CLSI guidelines for susceptibility testing of antimicrobials by the agar dilution method (15). The plates were then incubated at 37°C for 24 to 48 h, and growth was assessed. The copper-resistant E. faecium strain A17sv1 was used as a positive control, and the copper-sensitive E. faecium strain BM4105RF was used as a negative control, as described previously (10). As controls for testing of non-E. faecium isolates, the type strains of E. casseliflavus (ATCC 25788), E. mundtii (ATCC 43186), and E. gallinarum (ATCC 35038) were used.

PFGE.
PFGE using SmaI (Fermentas) was carried out as described previously (12). All 20 E. faecium isolates used for inoculation were subjected to PFGE prior to inoculation. Furthermore, randomly chosen macrolide-resistant isolates from day 7 and day 28 as well as all glycopeptide-resistant isolates isolated during the trials were also subjected to PFGE using SmaI.

Statistical analysis.
On each day from day 0 to day 35, the body weights (W) of the pigs were measured. Differences in average daily weight gain [ADG = (W35 – W0)/35] between pigs fed low- and high-copper diets were tested using the t test.

Initially, we tested if there was a significant difference in copper resistance or erythromycin resistance on day 0 by using logistic analysis with a generalized linear mixed model. Copper concentration and erythromycin resistance on day 0 were the outcome variables. Copper concentration was included as a fixed effect. The following three random effects were included in the model: sow, pen within copper concentration, and pig within pen, copper concentration, and sow. However, the random effect of the sow variable was estimated to be 0 and was excluded from the analyses.

The effects of copper concentration on copper resistance and on erythromycin resistance across time were evaluated in a logistic analysis using a generalized linear mixed model. Copper resistance and erythromycin resistance were the outcome variables. The copper concentration (low or high), day (0, 7, 14, 21, or 28), and interaction between copper concentration and day were included as fixed effects. The following three random effects were included in the model: sow, pen within copper concentration, and pig within pen, copper concentration, and sow. However, the random effect of the sow variable was estimated to be 0 and was excluded from the analyses.

All analyses were performed using the Statistical Analysis System (SAS, version 8.2). A 5% significance level was used.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Clinical signs and macroscopic lesions.
No clinical signs could be observed among any animals throughout the study. The consistency of feces was always normal (fecal score = 1 for all animals). The temperature of the pigs remained stable during the study period (days 0 to 35; range, 38.7 to 40.2°C; mean, 39.4°C for pigs fed a low concentration of copper and 39.5°C for pigs fed a high concentration of copper). Postmortem examinations did not reveal any lesions (except for one pig from pen 2 [low-copper diet] with perihepatitis and another one in pen 1 [low-copper diet] with a very limited lesion of perihepatitis).

Zootechnical performances.
There was no significant difference (P = 0.11) in ADG from day 0 to day 35 between pigs fed a low-copper diet (mean ADG = 866 g; standard deviation = 137 g) and pigs fed a high-copper diet (mean ADG = 947 g; standard deviation = 99 g).

High concentrations of copper sulfate in feed select for tcr-mediated copper resistance among E. faecium isolates.
From each pig on each of the five sampling dates (days 0, 7, 14, 21, and 28), up to six E. faecium-like isolates were collected for further analysis. A minimum of three of these six were identified to the species level to be E. faecium and tested for the presence of the tcr genes. In total, 168 E. faecium isolates were obtained from the animals in pens 1 and 2, and 188 isolates were obtained from the animals in pens 3 and 4. As shown in Fig. 1, the background level of E. faecium carrying tcr genes before the pigs were inoculated and before the feed was changed to a high-copper diet in pens 3 and 4 (day 0) was between 35 and 40% for both animal groups. At all later sampling dates, the fraction of isolates carrying the tcr genes remained between 20% and 40% among bacteria isolated from the animals in pens 1 and 2, which did not receive the high-copper diet. In contrast, the level of isolates containing tcr genes for pens 3 and 4, where animals received 175 ppm of copper, was significantly increased (P < 0.001 for the interaction between day and copper resistance) compared to the level prior to the change of feed. In this case, the level of tcr-positive isolates reached a maximum of 94% 1 week after the diet was changed and remained significantly higher than the level in the control pens for the remainder of the experiment (P < 0.01).


Figure 1
View larger version (9K):
[in this window]
[in a new window]
  		FIG. 1. Percentages of tcrB-positive (copper-resistant) E. faecium strains isolated on the five sampling dates (days 0, 7, 14, 21, and 28), originating either from animals in pens 1 and 2 that were fed 6 ppm copper sulfate (triangles) or from animals in pens 3 and 4 that were fed 175 ppm copper sulfate (squares).

 
Among the 188 bacteria isolated from animals in pens 3 and 4, 28 (15%) were negative in the genotypic tcr PCR test. These isolates were tested for phenotypic copper resistance and compared to 49 tcr-negative isolates from pens 1 and 2. All 67 isolates had a MIC of 8 mM copper sulfate, which was equal to the MIC of the negative control (E. faecium BM4105RF).

PFGE of copper-resistant isolates.
One copper-resistant and tcr-positive isolate (if present) from each pig sampled on day 7 and day 28 (n = 38) was subjected to PFGE, and the PFGE pattern was compared to those of the original 20 isolates. In most cases (n = 33), the isolated bacteria could be identified as belonging to one of the inoculated copper-resistant strains. These 33 isolates were also resistant to macrolides, and all tested positive for erm(B) by PCR. Most of the PFGE profiles belonging to the tcrB-positive isolates which were inoculated initially could be reidentified among the 33 isolates. However, 2 of the 10 inoculated strains could not be reidentified among the 33 PFGE profiles (strains 74-30047-1 and 74-30277-2), while all other PFGE profiles were represented between one and eight times.

Five isolates did not show PFGE patterns comparable to those of any of the inoculated strains. None of these were resistant to macrolides [and they were erm(B) negative by PCR].

High concentrations of copper sulfate in feed coselect for macrolide and glycopeptide resistance.
All isolates were tested further for resistance to macrolides and glycopeptides. As shown in Fig. 2, E. faecium isolates obtained from animals receiving a high-copper diet were significantly (P < 0.001) more resistant to macrolides than those obtained from the low-copper control groups from pens 1 and 2. In fact, only about 20% of the bacteria isolated from the low-copper group were macrolide resistant, as opposed to almost 80% of bacteria from the high-copper group. In addition, all macrolide isolates were copper resistant, regardless of the animal group from which they were isolated (data not shown), while 73% of the copper-resistant isolates were resistant to macrolides.


Figure 2
View larger version (7K):
[in this window]
[in a new window]
  		FIG. 2. Percentages of macrolide-resistant E. faecium strains isolated on the five sampling dates (days 0, 7, 14, 21, and 28), originating either from animals in pens 1 and 2 that were fed 6 ppm copper sulfate (triangles) or from animals in pens 3 and 4 that were fed 175 ppm copper sulfate (squares). Macrolide resistance was defined as having a tablet diffusion zone diameter of <13 mm.

 
Similar to this, more glycopeptide-resistant bacteria were isolated from pigs fed the high-copper diet. In fact, no glycopeptide-resistant bacteria were isolated from pens 1 and 2 at all, while 24 glycopeptide (all vanA positive by PCR)- and copper-resistant isolates were found among animals from pens 3 and 4 after the feed was changed. These 24 isolates were subjected to PFGE analysis and compared to the original 20 inoculated isolates. All 24 isolates had a PFGE profile identical to that of the glycopeptide-, macrolide-, and copper-resistant strain (74-30658-3) initially inoculated into the pigs.

The tcrB gene is present in enterococcal species other than E. faecium.
Among the tcr-positive bacteria isolated from the four pens, a few isolates turned out to belong to enterococcal species other than E. faecium. One tcr-positive E. gallinarum isolate was detected at the beginning of the experiment (day 0; pen 3), while 12 isolates were found at later dates. All 12 isolates originated from pigs in pens 3 and 4. tcr genes were also detected for an E. casseliflavus strain isolated from a pig in pen 2 (day 14) as well as for four E. mundtii isolates (day 28; pens 1 and 2).

The 18 non-E. faecium isolates were tested phenotypically for copper resistance, and their phenotypes were compared to the copper susceptibilities of the type strains of the same species by agar MIC determination. In all 18 cases, the bacteria containing the tcr genes had higher MICs than the same type strains, which did not contain the tcr genes, as determined by PCR. All 18 isolated strains had an MIC of 20 mM, while the type strains had an MIC of 4 mM. The same isolates were tested for resistance towards macrolides and glycopeptides and were all found to be sensitive. This was confirmed by PCR, where all tested negative for the presence of the erm(B) and vanA genes.


   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The present study examined the effect of feeding high doses of copper to piglets on the level of copper-resistant E. faecium in their feces. The feeding experiment clearly showed a significant effect on the level of copper-resistant E. faecium. This supports previous in vitro experiments where sublethal levels of copper sulfate selected for copper-resistant E. faecium in competition with an isogenic copper-sensitive bacterium (9). The fact that we were still able to isolate tcr-negative E. faecium strains from pens 3 and 4 shows that 175 ppm copper sulfate does not result in complete elimination of the copper-sensitive bacterial population from the pigs. To ensure that these bacteria did not have a reduced copper susceptibility compared to that of other copper-sensitive E. faecium isolates, phenotypic testing of the tcr-negative isolates from all four animal pens was conducted. These isolates all had MICs comparable to the previously determined MIC of the copper-sensitive E. faecium population in general (2). There were no significant differences in the distribution of MICs among the tcr-negative isolates, regardless of the amount of copper supplementation that their host animals had received, indicating that no other copper resistance mechanism was present among these isolates.

We did not examine the effect of reducing the copper contents of the animal feed after 12 weeks of animal growth, as required in Danish pig production. Theoretically, this fivefold reduction could result in a copper concentration which would not favor resistant over susceptible bacteria. This is supported by a previous study of the prevalence of the tcr genes among E. faecium strains isolated from pigs in Denmark, Spain, and Sweden (3). In contrast to Denmark and Spain, Sweden has a maximum acceptable level of copper supplementation of 35 ppm copper in the feed. In Sweden, the level of tcr-positive E. faecium was far lower (6%) than the levels in Denmark (75%) and Spain (56%). Therefore, additional studies are required to determine the minimal selective concentration of copper.

Nine of 10 copper-resistant strains fed to the animals at the beginning of the experiment were also resistant to macrolides [erm(B) positive], and one of these was also resistant to the glycopeptide vancomycin (vanA positive). In contrast, only 1 of the 10 copper-sensitive isolates was resistant to macrolides, and none were resistant to glycopeptides. This enabled us to examine the coselective potential of high levels of copper sulfate for these antibiotic markers in vivo. As the data presented above show, the number of macrolide-resistant isolates increased significantly in pens 3 and 4 compared to that in pens 1 and 2. In all cases, macrolide-resistant bacteria were also copper resistant. Similar to this, all glycopeptide-resistant bacteria isolated during the experiment originated from pigs which were fed 175 ppm copper sulfate. Consequently, the high doses of copper sulfate fed to the animals in pens 3 and 4 coselected for both macrolide and glycopeptide resistance under the experimental settings applied here. Most of the copper-resistant E. faecium strains isolated from slaughter pigs in Denmark in 1998 were also resistant towards macrolides. Therefore, our experimental setting resembles to some extent the situation present in the stables of Danish pig producers at the time of the ban of antibiotics as growth promoters. Based on the data presented here, coselection of macrolide- and glycopeptide-resistant E. faecium strains in Danish pig production cannot be excluded and could therefore affect the prevalence of resistance among the porcine isolates. A similar coselection by copper supplementation of macrolide- and glycopeptide-resistant E. faecium could exist in other European countries and in the United States, but this has not been studied in detail. In fact, our previous study of Spanish E. faecium isolates from pigs did show that 73 of 75 copper-resistant E. faecium isolates were also resistant towards macrolides. This indicates that a link between copper and macrolide resistance does exist in countries other than Denmark. In that case, our findings could have international significance, as most European countries use 170 to 175 ppm for piglets and most pig producers in the United States use high doses of copper supplementation in the feed throughout the life span of the pigs.

PFGE examination revealed that most of the isolated copper-resistant E. faecium strains were identical to the copper-resistant isolates fed to the animals at the beginning of the experiment. A few tcr-positive isolates had new PFGE profiles and could have originated either from the original E. faecium flora of the pigs, from the feed given to the animals throughout the experiment, or from horizontal transfer of the resistance phenotype in the gut. The most likely explanation is that these new bacteria originated from the normal pig flora, as our data showed that the pigs did contain a moderate level of copper-resistant E. faecium prior to inoculation. However, this was not investigated further.

The growth-promoting effect of feeding high concentrations of copper sulfate to pigs is well documented in the literature, but only with regard to piglets (<20 kg), while copper supplementation does not seem to have an influence on the growth of older pigs (8). We also compared the ADGs of the two animal groups in pens 1 and 2 and in pens 3 and 4 in our study. We did not observe any statistically significant variation in the ADG between animals receiving 175 ppm copper sulfate and animals receiving a low-copper diet. This is in good agreement with previous reports, as the pigs in our study all weighed >20 kg at the start of the experiment. Since no effect on the ADG of pigs weighing >20 kg has been documented, either here or in the SCAN report, it could be argued that 170- to 175-ppm supplementation to piglets should be used only until the piglets reach 20 kg body weight and not, as is custom in the European Community today, until the pigs reach 12 weeks of age.

We also found tcr genes present among enterococcal species other than E. faecium. In all three species, the presence of the tcr genes led to increased copper resistance compared to that of the type strains of the same species. The presence of the tcr genes in E. gallinarum is especially interesting, as the copper supplementation in our experiment seemed to select for these isolates. However, the statistical material is too sparse to draw any conclusions about these isolates.

In conclusion, our experiment showed that the high levels of copper given to piglets as a feed supplement in pig production do select for copper-resistant E. faecium and that macrolide- and glycopeptide-resistant bacteria are coselected under the experimental settings applied in this study.


   ACKNOWLEDGMENTS
 
We thank André Keranflec'h (Service de Production de Porcs Assainis et d'Expérimentation, AFSSA, Ploufragan, France) for skilled technical assistance with the animal experiments and technicians Hanne Nancke-Krogh, Dorte S. Madsen, Jane N. Larsen, Inge M. Hansen, and Berith Kummerfeldt for their excellent help.

This work was supported by a grant from the Danish Research Agency, Danish Agricultural and Veterinary Research Council (23-01-0090).


   FOOTNOTES
 
* Corresponding author. Mailing address: Danish Institute for Food and Veterinary Research, Bülowsvej 27, 1790 Copenhagen V, Denmark. Phone: (45) 72 34 63 47. Fax: (45) 72 34 63 41. E-mail: hha{at}dfvf.dk. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Aarestrup, F. M. 2000. Characterization of glycopeptide-resistant Enterococcus faecium (GRE) from broilers and pigs in Denmark: genetic evidence that persistence of GRE in pig herds is associated with coselection by resistance to macrolides. J. Clin. Microbiol. 38:2774-2777.[Abstract/Free Full Text]
  2. Aarestrup, F. M., and H. Hasman. 2004. Susceptibility of different bacterial species isolated from food animals to copper sulphate, zinc chloride and antimicrobial substances used for disinfection. Vet. Microbiol. 100:83-89.[CrossRef][Medline]
  3. 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.[Abstract/Free Full Text]
  4. Brown, N. L., S. R. Barrett, J. Camakaris, B. T. Lee, and D. A. Rouch. 1995. Molecular genetics and transport analysis of the copper-resistance determinant (pco) from Escherichia coli plasmid pRJ1004. Mol. Microbiol. 17:1153-1166.[CrossRef][Medline]
  5. DANMAP. 2000. DANMAP 1999—consumption of antimicrobial agents and occurrence of antimicrobial resistance from food animals, food and humans in Denmark. Danish Zoonosis Centre, Copenhagen, Denmark.
  6. DANMAP. 2002. DANMAP 2001—consumption of antimicrobial agents and occurrence of antimicrobial resistance from food animals, food and humans in Denmark. Danish Zoonosis Centre, Copenhagen, Denmark.
  7. 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.[Abstract]
  8. European Commissions Scientific Committee for Animal Nutrition (SCAN). 2003. Opinion of the Scientific Committee for Animal Nutrition on the use of copper in feeding stuffs, p. 1-47. European Commissions Scientific Committee for Animal Nutrition, Brussels, Belgium.
  9. Hasman, H. 2005. The tcrB gene is part of the tcrYAZB operon conferring copper resistance in Enterococcus faecium and Enterococcus faecalis. Microbiology 151:3019-3025.[Abstract/Free Full Text]
  10. Hasman, H., and F. M. Aarestrup. 2002. tcrB, a gene conferring transferable copper resistance in Enterococcus faecium: occurrence, transferability, and linkage to macrolide and glycopeptide resistance. Antimicrob. Agents Chemother. 46:1410-1416.[Abstract/Free Full Text]
  11. Hasman, H., and F. M. Aarestrup. 2005. Relationship between copper, glycopeptide, and macrolide resistance among Enterococcus faecium strains isolated from pigs in Denmark between 1997 and 2003. Antimicrob. Agents Chemother. 49:454-456.[Abstract/Free Full Text]
  12. Hasman, H., A. G. Villadsen, and F. M. Aarestrup. 2005. Diversity and stability of plasmids from glycopeptide-resistant Enterococcus faecium (GRE) isolated from pigs in Denmark. Microb. Drug Resist. 11:178-184.[CrossRef][Medline]
  13. Hill, G. M., P. K. Ku, E. R. Miller, D. E. Ullrey, T. A. Losty, and B. L. O'Dell. 1983. A copper deficiency in neonatal pigs induced by a high zinc maternal diet. J. Nutr. 113:867-872.[Abstract/Free Full Text]
  14. Lee, S. M., G. Grass, C. Rensing, S. R. Barrett, C. J. Yates, J. V. Stoyanov, and N. L. Brown. 2002. The Pco proteins are involved in periplasmic copper handling in Escherichia coli. Biochem. Biophys. Res. Commun. 295:616-620.[CrossRef][Medline]
  15. National Committee for Clinical Laboratory Standards (NCCLS). 2003. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. Approved standard M7-A4. NCCLS, Wayne, Pa.

Applied and Environmental Microbiology, September 2006, p. 5784-5789, Vol. 72, No. 9
0099-2240/06/$08.00+0     doi:10.1128/AEM.02979-05
Copyright © 2006, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:


This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hasman, H.
Right arrow Articles by Aarestrup, F. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hasman, H.
Right arrow Articles by Aarestrup, F. M.
Agricola
Right arrow Articles by Hasman, H.
Right arrow Articles by Aarestrup, F. M.

Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
J. Bacteriol. Microbiol. Mol. Biol. Rev. Eukaryot. Cell All ASM Journals