Antimicrobial Resistance among Enterococci from Pigs in Three European Countries

Abstract

Enterococci from pigs in Denmark, Spain, and Sweden were examined for susceptibility to antimicrobial agents and copper and the presence of selected resistance genes. The greatest levels of resistance were found among isolates from Spain and Denmark compared to those from Sweden, which corresponds to the amounts of antimicrobial agents used in food animal production in those countries. Similar genes were found to encode resistance in the different countries, but the tet(L) and tet(S) genes were more frequently found among isolates from Spain. A recently identified transferable copper resistance gene was found in all copper-resistant isolates from the different countries.

Enterococci have in recent years emerged as important nosocomial pathogens in humans. Most infections are considered of endogenous origin, and infections with enterococci can be difficult to treat because they can be resistant to several antibiotics (13, 18).

Frequent occurrences of antimicrobial agent-resistant enterococci have been observed among food animals (2, 10, 19), and it has been suggested that food animals may be a reservoir of resistant enterococci and resistance genes capable of transferring to humans through the food chain. Several different resistance genes have been identified in enterococci, and similar determinants have been observed among enterococci of widely different ecological origin (2, 9, 11, 16, 19). However, comparable data on the occurrence of resistance and resistance genes from different countries are sparse.

This study compares the levels of occurrence of antimicrobial resistance and presence of selected resistance genes among Enterococcus faecalis and Enterococcus faecium isolates from pigs in Denmark, Spain, and Sweden.

Bacterial isolates.

Bacterial isolates from pigs in Denmark were from a previously described collection (2). A total of 102 E. faecalis and 88 E. faecium isolates, which were collected from pigs in Denmark in 1998, were included. All isolates of the same species were from different herds. Bacterial isolates from pigs in Spain were from the collection of the Network of Veterinary Antimicrobial Resistance Surveillance (VAV) (12). Of the E. faecium strains, 68 were collected in 1998 and 56 were collected in 1999. Sampling, isolation, and strain identification has been published previously (7). The bacterial isolates from Sweden were from intestinal (colon) material (collected from five different slaughterhouses in Sweden during 2000) that had been cultured on Slanetz and Bartley medium and bile esculin plates. Species identification was performed as previously described (1). Only one isolate from each farm was included, making a total of 48 E. faecalis and 18 E. faecium isolates.

Antimicrobial susceptibility testing.

All isolates were tested for susceptibility to the antimicrobial agents avilamycin, bacitracin, chloramphenicol, erythromycin, gentamicin, kanamycin, penicillin, streptomycin, quinupristin-dalfopristin, tetracycline, and vancomycin, using a commercially prepared, dehydrated panel (Sensititre), as previously described (2).

Copper gradient assay and detection of the tcrB gene.

Susceptibility to copper was determined for all isolates on brain heart infusion copper gradient agar plates containing CuSO₄ ·5H₂O in concentrations ranging from 0 to 28 mM, as previously described (6). All isolates were also examined for the presence of the tcrB gene, as previously described (6).

Detection of antimicrobial resistance genes.

Resistant isolates were examined for the presence of genes encoding resistance to chloramphenicol (cat-pIP501), gentamicin (aac6-aph2), kanamycin (aphA3), macrolide-lincosamide-streptogramin B (ermB), quinupristin-dalfopristin [vat(D), vat(E)], and tetracycline [tet(K), tet(L), tet(M), tet(O), tet(S)], using PCR as described previously (2, 8, 10). In addition, chloramphenicol-resistant isolates were examined for the presence of cat-pSBK, using primers P1 (5′-TCCAAGGAATCATCGAAATA-3′) and P2 (5′-TGAACTGTATCCTGCTTTGA-3′), and penicillin-resistant isolates for the presence of pbp5, using primers P3 (5′TGAGCAATTTGTCCAAGC-3′) and P4 (5′TGATCCAGCTTTTCCTCC-3′).

The levels of resistance occurrence are shown in Table 1. Much more frequent occurrence of resistance was observed among the E. faecalis isolates from Denmark compared to those from Sweden. The frequency of resistance to erythromycin was especially high in the isolates from Denmark (85%) compared to those from Sweden (17%). This could be due to the fact that Sweden has not used antimicrobial agents for growth promotion since 1985, whereas there had been major usage of the macrolide tylosin for growth promotion in pigs in Denmark until 1998 (3). Following decreased usage of tylosin in Denmark in 1998, a decrease in macrolide resistance (reaching a frequency of around 30%) was observed in 2001 (3). In contrast, the frequencies of resistance to tetracycline are approximately the same, which could reflect the relatively frequent usage of tetracycline for therapy in both countries. Resistance to copper has not previously been investigated in E. faecalis. In this study, resistance to copper was observed among 17% of E. faecalis isolates from Denmark compared to only one isolate from Sweden. In Denmark, all pigs are fed copper-supplemented feed. Weaners normally receive feed containing 175 ppm of CuSO₄, while pigs raised for slaughter receive approximately 35 ppm. In Sweden, the maximum acceptable level is 35 ppm for all pigs, regardless of age. Thus, it seems likely that the usage of CuSO₄ at higher levels for weaners in Denmark had selected for copper-resistant E. faecalis.

TABLE 1.

Occurrence of antimicrobial resistance and distribution of resistance genes among E. faecalis and E. faecium isolates from pigs in Denmark, Spain, and Sweden^a

Bacterial species	Antimicrobial agent	Occurrence of resistance (%) among isolates from pigs in:			Resistance gene	Prevalence of resistance gene (%) among resistant isolates from:
		Denmark	Spain	Sweden		Denmark	Spain	Sweden
E. faecalis	Avilamycin	0		0
	Chloramphenicol	4		0	cat-pIP501	100
	Erythromycin	85		17	erm(B)	92		100
	Gentamicin	1		0	aac6-aph2	100
	Kanamycin	24		2	aphA3	96		100
	Penicillin	0		0
	Streptomycin	39		15
	Tetracycline	68		65	tet(K)	0		0
					tet(L)	16		0
					tet(M)	100		100
					tet(O)	0		6
					tet(S)	0		0
	Vancomycin	0		0
	Copper	17		2	tcrB	100		100
E. faecium
	Avilamycin	4	4	0
	Chloramphenicol	7	19	0	cat-pIP501	100	78
					cat-SBK		13
	Erythromycin	81	86	6	erm(B)	88	100	100
	Gentamicin	0	5	0	aac6-aph2		0
	Kanamycin	18	45	0	aphA3	62	91
	Penicillin	9	21	6	pbp5	100	100	100
	Quinupristin-dalfopristin	52	71	6	vat(D)	2	6	0
					vat(E)	4	6	0
	Streptomycin	27	64	0
	Tetracycline	63	94	22	tet(K)	0	0	0
					tet(L)	29	83	0
					tet(M)	95	98	100
					tet(O)	0	3	25
					tet(S)	0	26	0
	Vancomycin	17	2	6	vanA	100	100	100
	Copper	75	56	6	terB	100	100	100

^aNo. of isolates per country: Denmark, 102 E. faecalis and 88 E. faecium; Spain, 124 E. faecium; Sweden, 48 E. faecalis and and 18 E. faecium.

Among E. faecium isolates, the highest levels of frequency of resistance were found among the isolates from Spain and Denmark, followed by those from Sweden. The only exceptions were resistance to vancomycin and copper, for which the highest frequencies were found among the isolates from Denmark. Almost all isolates from Sweden were susceptible to all antimicrobial agents tested, except two isolates that were tetracycline resistant. A frequent occurrence of copper resistance was observed among the E. faecium isolates from both Denmark (75%) and Spain (56%), while only a single isolate from Sweden was resistant. In Spain, several chemical forms of copper [Cu(CH₃COO₂), CuCO₃, Cu(OH)₂, CuCl₂, CuO, and Cu(C₅H₁₀NO₂)₂] can be used at a maximum level of 175 ppm in fattening pigs until 16 weeks and from this age until slaughter at a maximum level of 35 ppm. This suggests that, as is the case for E. faecalis, the inclusion of different copper compounds in the feed for pigs has selected for this resistance among the enterococcal populations in Denmark and Spain. The occurrence of resistance was higher among E. faecium than E. faecalis isolates, indicating that E. faecium more easily acquires resistance.

The major differences in the occurrence of resistance between the three different countries are most probably due to differences in the usage of antimicrobial agents. Thus, major restrictions have been imposed in Sweden on the usage of especially powerful antimicrobial agents for growth promotion, including the banning of the use of such products in 1985. Data on the consumption of antimicrobial agents for food animals in the different countries are very limited. In 1999, the European Agency for the Evaluation of Medicinal Products presented estimates of antimicrobial agent consumption levels in the different European Union countries during 1997 (4). It was estimated that in Sweden, an average of 24 mg of therapeutic antimicrobial agents was used for the production of 1 kg of meat. The corresponding figures for Denmark and Spain were 24 and 103 mg/kg, respectively. In addition, 30 and 33 mg of antimicrobial agents per kilogram of meat were used for growth promotion in Denmark and Spain, respectively. Thus, the differences in the occurrence of antimicrobial resistance found in this study seem to fit well with the consumption figures.

The levels of occurrence of antimicrobial resistance genes among resistant isolates are given in Table 1. Tetracycline resistance in enterococci has previously been found to be mediated by tet(L), tet(M), tet(O), and tet(S) (15). As would be expected, tetracycline resistance was mediated in most isolates by the tet(M) gene. However, some differences in the distribution of the other resistance genes were observed. Thus, a very high frequency of tet(L) was found among the tetracycline-resistant E. faecium isolates from Spain compared to those from Denmark and Sweden. In addition, 26% of E. faecium isolates from Spain contained tet(S), a result that was not observed among the isolates from Denmark. Chloramphenicol resistance was mediated in most isolates by cat-pIP501, as has also been reported previously (2, 21). However, three (13%) of the isolates from Spain contained cat-SBK, while no resistance genes could be observed in two isolates. Not surprisingly, resistance to erythromycin, penicillin, and vancomycin was mediated in most isolates by erm(B), pbp5, and vanA, respectively. Kanamycin resistance was mainly mediated by aphA3; however, among 38% of the resistant isolates from Denmark, no gene was found. The bifunctional gene aac6-aph2 has previously been identified as the most common gene mediating gentamicin resistance (2), but this gene was not present in any of six gentamicin-resistant E. faecium isolates from Spain. Two different genes mediating resistance to quinupristin-dalfopristin in E. faecium, vat(D) and vat(E), have been identified (14, 20). These two genes have been found to be widespread among E. faecium isolates from humans and animals in several European countries (5, 10, 17, 21). However, each of these two genes could only be detected in 6 to 12% of the resistant isolates from pigs in Denmark or Spain.

Recently, a transferable copper resistance gene (tcrB) was identified in E. faecium isolates from pigs in Denmark (6). In that study, the gene was found in all copper-resistant E. faecium and E. faecalis isolates from the different countries, whereas it was not detected in any of the copper-susceptible isolates. Thus, the presence of the gene seems to be widespread in different countries and to be responsible for copper resistance in both E. faecalis and E. faecium.

The same genes were in general found to encode resistance in the different countries. This shows that resistance does not emerge independently but is mainly caused by the spread of genes within and among bacterial populations in different countries. This is true not only for genes encoding resistance to antimicrobial agents but also for genes implicated in copper resistance. Nonetheless, some differences were apparent. Thus, the tet(L) and tet(S) genes were frequently found among isolates from Spain. The reason for this is unknown.