Mechanisms of Resistance to Quinupristin-Dalfopristin among Isolates of Enterococcus faecium from Animals, Raw Meat, and Hospital Patients in Western Europe

Abstract

Twenty-eight quinupristin-dalfopristin-resistant isolates of Enterococcus faecium from hospital patients and nonhuman sources in European countries were studied. High-level resistance (MICs, ≥32 μg/ml) was associated with the presence of vat(E) (satG) (14 isolates [50%]) or vat(D) (satA) (6 isolates [21%]). These genes were not detected in eight (29%) isolates with lower levels of quinupristin-dalfopristin resistance (MICs, 4 to 16 μg/ml). This suggests the presence of further mechanisms of resistance to quinupristin-dalfopristin in E. faecium.

Quinupristin-dalfopristin is a semisynthetic, injectable mixture of streptogramin A and B compounds which has recently been licensed for clinical use in the United States and in Europe for the treatment of infections caused by multiresistant gram-positive pathogens, including glycopeptide-resistant Enterococcus faecium (9, 17, 18). Virginiamycin, another streptogramin A and B combination, has been used as a growth promoter in animal feed for many years. It selects for virginiamycin-resistant strains of E. faecium, which are cross-resistant to quinupristin-dalfopristin (11, 12, 19) and which may pose a risk to public health. As a consequence, the use of virginiamycin has been banned in the European Union.

Resistance to mixtures of streptogramin A and B compounds was first reported among staphylococci (10) and requires only resistance to the A component (15), although resistance to both the A and B components may give a higher level of resistance (B. Bozdogan and R. Leclercq, Abstr. 39th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 841, 1999). Several genes conferring resistance to type A streptogramins have been identified in staphylococci (1–5, 7, 8). In this paper we adopt the nomenclature proposed recently by Roberts and colleagues (16). The staphylococcal genes vat(A), vat(B), and vat(C) encode acetyltransferases (2, 4, 7); vga(A) and vga(B) encode ATP-binding proteins involved in active efflux (3, 5). Among enterococci, Enterococcus faecalis is naturally resistant to streptogramin A-B combinations (15), whereas most isolates of E. faecium are susceptible. Two acetyltransferase genes, satA (15) and satG (20), have been characterized for strains of E. faecium resistant to streptogramin A-B combinations. These have been renamed vat(D) and vat(E), respectively (16). However, vat(D) occurs in only a minority of the resistant E. faecium organisms investigated in several countries in the European Union (11, 12), and the prevalence of vat(E) is unknown. In this study we investigated the basis of resistance to quinupristin-dalfopristin in a small, diverse group of E. faecium isolates.

(This work was presented at the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, Calif., September 1999.)

Twenty-eight quinupristin-dalfopristin-resistant (MICs, ≥4 μg/ml) isolates of E. faecium were collected in the United Kingdom (UK) and other European countries prior to both the licensing of quinupristin-dalfopristin and the ban on virginiamycin use: 18 from animals (15 from pigs and 3 from chickens), 4 from raw meat, 2 from sewage, and 4 from UK hospital patients. Isolates from hospital patients, raw meat, and sewage were from a collection of enterococci kept at the Antibiotic Resistance Monitoring and Reference Laboratory. Animal isolates were kindly provided by Pfizer Ltd., Reigate, UK). E. faecium strain BM4145 [vat(D)] (kindly supplied by P. Courvalin, Institut Pasteur, Paris, France) and three strains of staphylococci, BM3093 [vat(A) vga(A) vgb(A)], BM12235 [vat(B) vga(B)], and BM12392 [vgb(B) vat(C)] (kindly supplied by N. El-Solh, Institut Pasteur, Paris, France), were used as positive PCR controls. A vat(E)-positive control strain was not available for this study. The identity of all isolates was confirmed by amplification of the E. faecium-specific gene encoding d-alanyl-d-alanine ligase (21).

The MICs of quinupristin-dalfopristin (Rhone Poulenc Rorer, West Malling, UK) and virginiamycin (Pfizer, Rixensart, Belgium) were determined by agar incorporation in DST agar (Oxoid, Basingstoke, UK) containing 5% lysed horse blood, with an inoculum of 10⁴ CFU per spot. Etest strips (Cambridge Diagnostic Services, Cambridge, UK) were used for some quinupristin-dalfopristin MIC determinations. Isolates were screened by PCR for vat(A), vat(B), vat(C), vat(D), vat(E), vga(A), vga(B), vgb(A), vgb(B), and erm(B). The primer sequences and cycling conditions are listed in Table 1. Degenerate primers M and N (Table 1), derived from motifs conserved within streptogramin A acetyltransferases (14), were used to detect putative genes encoding acetyltransferases in quinupristin-dalfopristin-resistant isolates. Plasmid DNA was isolated from E. faecium isolates by alkaline lysis (22), separated by electrophoresis through 0.8% agarose gels, and transferred to nylon membranes by vacuum blotting. Hybridization was performed at high stringency using digoxigenin-labeled vat(D)- and vat(E)-specific probes, which were prepared by incorporation of digoxigenin-11-dUTP (Roche, Lewes, UK) into PCR products. Escherichia coli strain 39R861 (NCTC 50192), an isolate containing plasmids of known size, was used as a source of molecular size markers.

TABLE 1.

Primer sequences and PCR conditions used for streptogramin resistance genes

Primer	Sequence (5′–3′)	GenBank accession no. and/or gene	Position of amplicon	PCR conditions	Reference(s)
vat-1	TGGAGTGTGACAAGATAGGC	L07778	200–712	Same as for vat(D)	7, 11
vat-2	GTGACAACAGCTTCTGCAGC	vat(A)
vatB-1 vatB-2	GGCCCTGATCCAAATAGCAT GTGCTGACCAATCCCACCAT	U19459L38809, vat(B)	76–634	1 cycle of 3 min at 94°C; 35 cycles of 1 min at 94°C, 1 min at 60°C, and 1 min at 72°C; and 1 cycle of 10 min at 72°C	2, 11
vatC-O vatC-P	ATGAATTCGCAAAATCAGCAAGG TCGTCTCGAGCTCTAGGTCC	AF015628vat(C)	1307–1886	1 cycle of 3 min at 95°C and 2 min at 60°C; 30 cycles of 20 s at 72°C, 20 s at 95°C, and 20 s at 60°C; and 1 cycle of 1 min at 72°C	4
satA-1 satA-2	GCTCAATAGGACCAGGTGTA TCCAGCTAACATGTATGGCG	L12033vat(D)	361–632	1 cycle of 3 min at 94°C; 35 cycles of 1 min at 94°C, 1 min at 55°C, and 1 min at 72°C; and 1 cycle of 10 min at 72°C	11, 15
satG-1	ACTATACCTGACGCAAATGC	AF139725	66–577	1 cycle of 5 min at 94°C;	20
satG-2	GGTTCAAATCTTGGTCCG	vat(E)		30 cycles of 25 s at 94°C, 40 s at 52°C, and 50 s at 72°C; and 1 cycle of 6 min at 72°C
vga-1	AGTGGTGGTGAAGTAACACG	M90056	1149–1808	Same as for vat(D)	5, 11
vga-2	CTTGTCTCCTCCGCGAATAC	vga(A)
vgaB-1 vgaB-2	TGACAATATGAGTGGTGGTG GCGACCATGAAATTGCTCTC	U82085vga(B)	990–1566	1 cycle of 3 min at 94°C; 35 cycles of 1 min at 94°C, 1 min at 52°C, and 1 min at 72°C; and 1 cycle of 10 min at 72°C	3, 11
vgb-1	TACAGAGTACCCACTACCGA	M36022	781–1350	Same as for vga(B)	6, 11
vgb-2	TCAATTCCTGCTCCAGCAGT	vgb(A)
vgbB-Q	CAGCAGTCTAGATCAGAGTGG	AF015628	512–1240	Same as for vat(C)	4
vgbB-R	CATACGGATCCATCTTTTCC	vgb(B)
ermB-1 ermB-2	CATTTAACGACGAAACTGGC GGAACATCTGTGGTATGGCG	M11180 ermB-1	836–1260	1 cycle of 3 min at 94°C; 25 cycles of 1 min at 94°C, 1 min at 52°C, and 1 min at 72°C; and 1 cycle of 10 min at 72°C	13erm(B)
strep-M strep-N	ATHATGAAYGGIGCIAAYCAYMGIATG ICCDATCCAIACRTCRTTICC			1 cycle of 5 min at 95°C; 35 cycles of 2 min at 40°C, 90 s at 72°C, and 30 s at 95°C; and 1 cycle of 4 min at 40°C and 12 min at 72°C	14

The 28 quinupristin-dalfopristin-resistant isolates of E. faecium studied were divided into three categories: 6 (21%) isolates (from live pigs in Belgium, Germany, Spain, and The Netherlands) contained vat(D), 14 (50%) isolates (from UK hospital patients, from live pigs and poultry, and from raw poultry meat in Spain, Germany, and the UK) contained vat(E), and 8 (29%) isolates (from live pigs, raw poultry meat, and sewage in Belgium, Spain, and the UK) contained neither of these genes (Table 2). The MICs of both quinupristin-dalfopristin and virginiamycin were ≥32 μg/ml for all vat(D)- or vat(E)-positive isolates. In contrast, the eight vat(D)-negative, vat(E)-negative isolates had lower levels of resistance to quinupristin-dalfopristin (MICs, 4 to 16 μg/ml) and virginiamycin (MICs, 2 to 4 μg/ml). None of the isolates carried the vat(A), vat(B), vat(C), vga(A), and vga(B) streptogramin A resistance genes or the vgb(A) and vgb(B) streptogramin B resistance genes (4, 6) previously detected in staphylococci. The erm(B) gene, conferring macrolide-lincosamide-streptogramin B resistance (13), was detected in 82% of the isolates tested and in ≥75% of isolates in each quinupristin-dalfopristin resistance category (Table 2). The absence of erm(B) from some vat(D)- or vat(E)-positive isolates does not rule out the presence of other genes conferring resistance to streptogramin B compounds; it has been reported recently that resistance to both the A and B components is necessary to confer high-level resistance to streptogramin combinations on E. faecium (Bozdogan and Leclercq, 39th ICAAC).

TABLE 2.

Categories of quinupristin-dalfopristin-resistant E. faecium based on amplification of acetyltransferase genes

Source or group	No. (%) in PCR group
	vat(D)+	vat(E)+	vat(D) and vat(E) negativea	Total
Human	0	4	0	4
Sewage	0	0	2	2
Animal	6	7	5	18
Raw meat	0	3	1	4
Total	6 (21)	14 (50)	8 (29)	28 (100)
erm(B)+ isolates	6 (100)	11 (79)	6 (75)	23 (82)

^aDid not give a PCR product with M and N, degenerate primers for genes that encode streptogramin A acetyltransferases. 

When degenerate primers M and N were used to detect genes likely to encode streptogramin A acetyltransferases, amplicons of ca. 150 bp were obtained from the 20 vat(D)- or vat(E)-positive isolates but from none of the 8 negative isolates. When the DNA was blotted onto nylon membranes, the M and N amplicons from vat(D)-positive strains hybridized only with a vat(D)-specific probe, and those from vat(E)-positive isolates hybridized only with a vat(E)-specific probe. Plasmid DNA prepared from the eight vat(D)-negative, vat(E)-negative isolates failed to hybridize with either vat(D)- or vat(E)-specific probes. The vat(E)-specific probe hybridized with plasmids in the range of 35 to 100 MDa in all vat(E)-positive isolates (not shown).

The proportion of quinupristin-dalfopristin-resistant isolates shown to contain vat(D) (21%) is similar to that reported by other workers (11, 12). The majority of vat(D)-negative isolates (63% [14 of 22]) carried the recently described vat(E) gene (20). Our data indicate wide dissemination of vat(E), which we confirmed to be plasmid mediated, among enterococci from diverse sources and countries. One of the resistance plasmids now known to carry vat(E) could be transferred in vitro to a sensitive laboratory strain of E. faecium (23). Eight diverse isolates expressed lower levels of streptogramin resistance but contained no detectable acetyltransferase genes. The degenerate primers used yielded products with the five genes, vat(A) to vat(E), known to encode streptogramin A acetyltransferases, but we cannot rule out the possibility that these isolates contain unrelated acetyltransferases. However, their streptogramin resistance may have resulted from a mechanism unrelated to acetyltransferase production. Active efflux associated with ATP-binding cassette transporters (3, 5) is the only other mechanism reported to confer resistance to streptogramin A compounds, and it is possible that these isolates have such a mechanism, although the demonstration of this requires further study.

In conclusion, we have shown that distinct categories of quinupristin-dalfopristin resistance exist in isolates of E. faecium. The Vat(D) and Vat(E) acetyltransferases occurred in enterococci from different European countries and conferred resistance to both quinupristin-dalfopristin and virginiamycin. An undefined mechanism conferred higher levels of resistance to quinupristin-dalfopristin than to virginiamycin. A wide distribution of vat(D) in strains from human and nonhuman sources has been reported previously (11, 12, 19). Similarly, in this study we detected vat(E) on plasmids in E. faecium isolated from farm animals, meat, and UK hospital patients; comparable findings have been reported in Germany (20). None of the UK patients had received quinupristin-dalfopristin. It therefore seems likely that exchange of resistant strains or resistance genes may occur between E. faecium isolates from nonhuman and human sources. Until such time as these fears are proven to be unfounded, the occurrence of virginiamycin-resistant E. faecium in animals must be considered to remain a potential risk to public health, the extent of which has yet to be quantified.