OUP user menu

Detection of the satA gene and transferability of virginiamycin resistance in Enterococcus faecium from food-animals

Anette Marie Hammerum, Lars Bog∅ Jensen, Frank M?ller Aarestrup
DOI: http://dx.doi.org/10.1111/j.1574-6968.1998.tb13267.x 145-151 First published online: 1 November 1998


The satA gene encoding streptogramin A resistance was detected in virginiamycin-resistant Enterococcus faecium isolates from pigs and broilers. The satA gene was present in 22 of 89 (25%) virginiamycin-resistant E. faecium isolates. It was shown that the satA gene and other gene(s) encoding streptogramin resistance could be transferred between isogenic E. faecium strains at frequencies ranging from 2.3×10−4 to 2.2×10−3 transconjugants per donor.

Key words
  • Enterococcus faecium
  • Virginiamycin resistance
  • satA gene
  • Conjugative transfer
  • Food-animal

1 Introduction

Streptogramins are a group of cyclic peptides produced by Streptomyces spp. as a mixture of components classified into two groups: streptogramin A and streptogramin B [1]. Compounds of group A (SgA), including streptogramin A, pristinamycin IIA and virginiamycin M, are polyunsaturated cyclic peptolides. Compounds of group B (SgB), which includes streptogramin B, pristinamycin IB and virginiamycin S, are cyclic hexapeptides. In clinical practice, the A and B streptogramins are administered in combination because they act synergistically [1]. Natural mixtures of streptogramins have been available on the European market since the mid-fifties, and products like pristinamycin and staphylomycin have been used as oral antistaphylococcal agents for human infections [1]. A semisynthetic injectable streptogramin quinupristin/dalfopristin RP59500 (Synercid) has been under development, and Synercid and pristinamycin have recently been used for treatment of vancomycin-resistant Enterococcus faecium[24] and methicillin-resistant Staphylococcus aureus in humans [5].

In 1975 the first staphylococci resistant to a mixture of A and B components were observed in France [6]. Since then, four different genes encoding resistance to the A compounds, vga[7], vgaB[8], vat[9], and vatB[10], and one gene which inactivates the B compounds, vgb[11], have been detected on plasmids isolated from staphylococci.

Resistance to the streptogramin A group was observed in a human E. faecium strain in 1984. In this strain a new gene, satA, located on a plasmid was detected [12]. Recently the vgb gene has also been detected in a human E. faecium strain [13].

In contrast to humans, where streptogramins have only been used rarely, another streptogramin product, virginiamycin, has been widely used as a growth promoter for broilers, pigs and cattle in Europe. In Denmark, streptogramins have never been approved for human therapy, but virginiamycin has been used for animals for decades (58.696 kg active compound from 1989 to 1997), and occurrence of resistance to virginiamycin has frequently been observed among E. faecium isolates from Danish food-producing animals [14]. In this study, the presence of the satA gene among virginiamycin-resistant E. faecium isolates from pigs and broilers is described. Furthermore, the transferability of the satA gene and of unknown gene(s) is described.

2 Materials and methods

2.1 Bacterial isolates

The virginiamycin-resistant E. faecium isolates were collected from October 1995 to December 1996 [14]. A total of 89 isolates from broilers (48) and pigs (41) were investigated. The isolates were recovered from the intestinal contents collected at the slaughterhouse. Only one isolate per animal was included. All isolates had MICs ≥4 μg ml−1 to virginiamycin and were identified as E. faecium according to Devriese et al. [15].

E. faecium BM4145 [12], E. faecalis ATCC 29212 and S. aureus ATCC 29213 were used as quality control strains in the antimicrobial susceptibility testing. E. faecium BM4145 [12], S. aureus BM3093 and S. aureus BM12235 were used as positive controls for the PCR. E. faecium BM4105-RF and E. faecium BM4105-Str [16] were used as recipients for the conjugative transfer.

2.2 Antimicrobial susceptibility

The MICs of pristinamycin IIA RP74502 (Rhône-Poulenc Rorer, Vitry sur Seine Cedex, France) and virginiamycin (Pfizer, Rixensat, Belgium) were determined on freshly made Mueller-Hinton II agar (Mueller-Hinton II; Becton Dickinson Microbiology System, Cockeyville, MD) following NCCLS guidelines [17]. The concentrations of virginiamycin were from 0.06 to 32 μg ml−1 and of pristinamycin IIA from 0.25 to 256 μg ml−1.

2.3 Polymerase chain reaction and DNA sequencing

DNA extractions, PCR amplification and DNA sequencing were performed according to Sandvang et al. [18]. The primers are shown in Table 1. SUPER-TAQ (HT Biotechnology, Cambridge, UK) was used as DNA polymerase. E. faecium BM4145, S. aureus BM3093 and S. aureus BM12235 were used as positive controls. All isolates were investigated for the presence of satA with PCR. Two amplicons were sequenced. Six isolates were PCR-negative for the presence of satA were investigated for the presence of vat, vatB, vga, vgaB and vgb.

View this table:
Table 1

Primers used for identification of all know genes encoding streptogramin resistance

PrimerSequence (5′–3′)PositionAccession number
vatB-2GTGCTGACCAATCCCACCAT568–549U19459 L38809

2.4 Filter mating

Eight isolates of E. faecium containing the satA gene and six isolates containing unknown streptogramin resistance gene(s) were chosen as donors for conjugation experiments.

Conjugation was performed using the filter mating procedure as described by Clewell et al. [19]. One vol. of a log-phase donor culture was mixed with 9 vols. of a similarly prepared recipient culture. The mixed culture (200 μl) was dispensed onto a 47-mm sterile filter (0.2 μm, MFS, California, USA) that had been placed on a bovine blood agar plate. After allowing the mixed culture to ‘dry’ into the filter, the plate was incubated at 37°C. Transferability was determined after 24 h incubation placing the filter in 10 ml 0.9% saline, suspending the culture by vortex mixing, and plating on selective media (1000 μg ml−1 streptomycin and 2 μg ml−1 virginiamycin) or (25 μg ml−1 fusidin, 25 μg ml−1 rifampin and 2 μg ml−1 virginiamycin) according to the recipient.

The resistance genes were first transferred to E. faecium BM4105-RF. Later, the transfer efficiencies were measured in a conjugation experiment from E. faecium BM4105-RF to E. faecium BM4105-Str for two of the satA-positive strains and three of the strains with an unknown gene(s).

2.5 Plasmid profiling and pulsed field gel electrophoresis

The plasmid profiles of transconjugants and donors were prepared according to the method described by Anderson and McKay [20]. A digoxigenin-labeled DNA probe for satA (271 bp) was prepared by PCR amplification using the primers described in Table 1 and was subsequently labeled with the Boehringer Mannheim DNA labeling and detection kit. The obtained PCR product was purified by using Qiagen (Hilden, Germany) spin columns.

The satA-positive strains were hybridized with the satA probe. Pulsed field gel electrophoresis (PFGE) of the donors and transconjugants was performed according to Jensen et al. [21].

3 Results and discussion

The MICs of virginiamycin for the 89 isolates are shown in Table 2. All the virginiamycin (mixture SgA and SgB)-resistant isolates were also resistant to pristinamycin IIA (SgA) (data not shown). In staphylococci, the same observation, that resistance to a synergetic mixture of SgA and SgB is always associated with resistance to SgA, has been made [22].

View this table:
Table 2

MICs of virginiamycin for 89 E. faecium isolates from pigs and broilers in Denmark according to the presence of the satA gene

Detection of satADistribution of isolates according to the MIC of virginiamycin (μg/ml) (no. of isolates)
  • aMIC for S. aureus ATCC 29213.

  • bMIC for E. faecalis ATCC 29212.

  • cMIC for E. faecium BM4145.

A sequence identical to satA was detected in 22 of the 89 (25%) virginiamycin-resistant E. faecium isolates originating from pigs and broilers. Two amplicons were sequenced, and the obtained sequences were identical to the published sequence of satA from E. faecium BM4145 (Fig. 1). There was no relationship between the presence or the absence of satA and the levels of resistance to virginiamycin (Table 2) or pristinamycin IIA (data not shown).

Figure 1

Comparison of the satA gene and two sequences obtained by PCR. The sequences were identical with the satA gene between the two primers. The primers are reported in bold type.

It was possible to transfer plasmids from six of the eight satA-positive strains and five of the six isolates with unknown gene(s) to BM4105-RF (Table 3). All recipients had received plasmids and the satA probe hybridized to all the satA-positive strains (data not shown). The transconjugants and the recipient strains had nearly identical PFGE patterns (two bands difference), indicating that a large plasmid was transferred to the recipient strains (Fig. 2).

View this table:
Table 3

Conjugation of virginiamycin resistance from 14 resistant E. faecium isolates to a sensitive recipient (BM4105-RF)

DonorsatARecipientTransfer possiblea
S9508038-2 (pAH2) E. faecium BM4105-RF+
S9508039-1 (pAH3) E. faecium BM4105-RF
S9630193-3 (pAH4) E. faecium BM4105-RF+
S9630342-3 (pAH5) E. faecium BM4105-RF+
F9630395-1 (pAH7) E. faecium BM4105-RF+
F9631080-1 (pAH9) E. faecium BM4105-RF+
BM4145 (pIP815)+ E. faecium BM4105-RF+
S9631101-1 (pAH12)+ E. faecium BM4105-RF
F9630230-1 (pAH13)+ E. faecium BM4105-RF+
F9508106-1 (pAH14)+ E. faecium BM4105-RF
F9631160-1 (pAH15)+ E. faecium BM4105-RF+
F9631247-1 (pAH16)+ E. faecium BM4105-RF+
S9631095-1 (pAH17)+ E. faecium BM4105-RF+
S9630370-1 (pAH18)+ E. faecium BM4105-RF+
  • aConjugation was done by filter mating as described in the text. Antibiotic selection (μg/ml): rifampin, 25; fusidin, 25; virginiamycin, 2.

Figure 2

CHEF electrophoresis of SmaI digested DNA from recipients, transconjugants and donors; lane 2–3 recipients, BM4105-Str, BM4105-RF; lane 4–8 transconjugants, BM4105-RF (pAH4), BM4105-RF (pAH7), BM4105-RF (pAH9), BM4105-RF (pAH15), BM4105-RF (pAH16); Lane 9–13 donors, S9630193-3 (pAH4), F9630395-1 (pAH7), F9631080-1 (pAH9), F9631160-1 (pAH15), F9631247-1 (pAH16). Lanes 1 and 14, Lambda ladder used as molecular size markers. Size of markers is 339, 291, 242, 194, 145, 97, 48.5 bp.

The transfer frequency ranged from 2.3×10−4 to 2.2×10−3 transconjugants per donor (Table 4). The transfer frequencies are relatively high and further studies may show if it is a sex pheromones mating response as has been observed in many E. faecalis and a few E. faecium strains [23]. In vitro transfer of streptogramin resistance has previously been observed [24]. This suggests that both known and unknown genes encoding streptogramin resistance can spread horizontally to sensitive E. faecium isolates.

View this table:
Table 4

Transfer frequencies of plasmids carrying virginiamycin resistance between E. faecium BM4105 isolates

DonorRecipientTransfer frequencya
E. faecium BM4105-RF (pAH5) E. faecium BM4105-Str3.2×10−4
E. faecium BM4105-RF (pAH7) E. faecium BM4105-Str6.9×10−4
E. faecium BM4105-RF (pAH9) E. faecium BM4105-Str3.3×10−4
E. faecium BM4105-RF (pAH13) E. faecium BM4105-Str2.2×10−3
E. faecium BM4105-RF (pAH15) E. faecium BM4105-Str2.3×10−4
  • aTransfer frequencies, expressed as the number of transconjugants per donor CFU after mating, are means of three independent experiments and were reproducible within one order of magnitude. Antibiotic selection (μg/ml): streptomycin, 1000; virginiamycin, 2.

In the majority of the virginiamycin/pristinamycin IIA-resistant isolates, satA was not detected, but it was possible to transfer resistance to a sensitive E. faecium strain. These isolates were all PCR negative to vat, vatB, vga, vgaB, vgb. This indicates that another streptogramin A resistance gene(s) is present in E. faecium.

The present study showed that virginiamycin-resistant E. faecium isolates from pigs and broilers contained the satA gene and another not yet identified gene(s), conferring resistance to virginiamycin and pristinamycin IIA. Both types of resistance were shown to be transferable.


This study is a part of the Danish Integrated Antimicrobial Resistance Monitoring and Research Program (DANMAP) conducted in collaboration between Statens Serum Institut, the Danish Veterinary and Food Administration and the Danish Veterinary Laboratory, and funded jointly by the Danish Ministry of Health and the Danish Ministry of Food, Agriculture and Fisheries. We would like to acknowledge the following persons for their technical assistance: René Hendriksen, Lissie Kjær Jensen, Karina Absalonsen, Mette Juul, Anne Lykkegaard Lauritsen and Christina Aaby Svendsen. We are grateful to Professor P. Courvalin and Professor R. Leclercq for sending BM4145, to Dr. Claire Poyart and Dr. Patrick Trieu-Cout for sending E. faecium BM4105-RF and E. faecium BM4105-Str and to Dr. Nevine El Sohl for sending S. aureus BM3093 and S. aureus BM12235.


View Abstract