OUP user menu

Characterisation of integrons and antibiotic resistance genes in Danish multiresistant Salmonella enterica Typhimurium DT104

Dorthe Sandvang, Frank Møller Aarestrup, Lars Bogø Jensen
DOI: http://dx.doi.org/10.1111/j.1574-6968.1997.tb12770.x 177-181 First published online: 1 December 1997

This article has a correction. Please see:


The presence and genetic content of integrons was investigated in eight Salmonella enterica Typhimurium DT104 isolates from different pig herds in Denmark. Two different integrons were identified using PCR and sequencing. Each of the integrons carried a single resistance cassette in addition to the sul1 and qacEΔ1 genes characteristic of integrons. The first integron encoded the ant (3″)-Ia gene that specified resistance to spectinomycin and streptomycin. The second contained the pse-1 β-lactamase gene. All the multiresistant strains contained both integrons. The presence of these two integrons did not account for the total phenotypic resistance of all the isolates and does not exclude the presence of other mobile DNA elements.

  • Salmonella enterica Typhimurium DT104
  • Integron
  • Antibiotic resistance

1 Introduction

Multiresistant Salmonella enterica serotype Typhimurium phage type DT104 has been reported with increased frequency in Europe and the United States [13] and since September 1996 multiresistant Typhimurium DT104 has been isolated from five pig herds in Denmark [4]. It has been inferred that the genes encoding resistance are chromosomally located in Typhimurium DT104 [2], but the exact location has not yet been defined nor the resistance genes identified.

A novel group of mobile DNA elements called integrons has been identified in Gram-negative bacteria. Integrons are able to incorporate single or groups of antibiotic resistance genes by site-specific recombination and are found in both chromosomal and extrachromosomal DNA [5, 6]. Integrons comprise two conserved segments, the 5′ conserved segment (5′-CS) and the 3′ conserved segment (3′-CS) and an internal variable region. The latter is the target for integration of gene cassettes that encode antibiotic resistance with a recombination site termed the 59-base element [7]. The 5′-CS contains the integrase gene (intI), integration site (attI) for the integron and promoters for expression of all down stream gene cassettes. The 3′-CS encodes resistance to disinfectant (qacEΔ1) and sulfonamide (sul1) in the class 1 integrons. [8, 9].

This study describes the genetic location of antibiotic resistance genes in different integrons in Danish Typhimurium DT104.

2 Materials and methods

2.1 Bacterial strains

Eight isolates of porcine Typhimurium DT104 from eight different herds were collected as a part of the Danish Salmonella surveillance programme between 1994 and 1997 (Table 1). All isolates were of the same PFGE type [4]. Two strains of Escherichia coli K12 that harboured plasmids R751 and R100,1 were kindly provided by L. Sundström [10] (Pharmaceutical Biosciences, Uppsala, Sweden) and were used throughout as controls of known integron sequences (GenBank accession numbers X72585 and X12870, respectively).

View this table:
Table 1

Genotypic and phenotypic characteristics of Salmonella Typhimurium DT104 isolates

OrganismDesignationDate and SourcePFGE typeaAntibiotic resistanceIntegrons detected
S. Typhimurium DT104941244517.11.1994, porcineXISensitiveNone
S. Typhimurium DT104952167631.07.1994, porcineXISensitiveNone
S. Typhimurium DT104942324513.12.1994, porcineXISp, Sm, SuInC-likeb
S. Typhimurium DT104961636820.11.1996, porcineXIAp, Cm, Sp, Sm, Su, TeInC-like and InDc
S. Typhimurium DT104962192705.07.1996, porcineXIAp, Cm, Sp, Sm, Su, TeInC-like and InD
S. Typhimurium DT104962212106.08.1996, porcineXIAp, Cm, Sp, Sm, Su, TeInC-like and InD
S. Typhimurium DT104962017919.01.1996, porcineXIAp, Cm, Sp, Sm, Su, TeInC-like and InD
S. Typhimurium DT104972092110.04.1997, porcineXIAp, Cm, Sp, Sm, Su, TeInC-like and InD
  • aBaggesen and Aarestrup [4].

  • bInC-like integron: the integron containing the ant (3″)-Ia gene cassette.

  • cInD integron: the integron containing the pse-1 gene cassette.

2.2 Antimicrobial susceptibility

Susceptibility to antimicrobial agents was determined with the two-fold dilution method using the Sensititre system (Accumed International Limited, East Grinstead, West Sussex, UK).

The antibiotics tested were: amoxicillin/clavulanic acid (A/Cl) 1/0.5–64/32 μg ml−1, ampicillin (Ap) 0.5–32 μg ml−1, apramycin (Ay) 0.5–64 μg ml−1, carbadox (Ca) 2–256 μg ml−1, ceftiofur (Ce) 0.25–16 μg ml−1, chloramphenicol (Cm) 1–128 μg ml−1, colistin (Ct) 1–64 μg ml−1, enrofloxacin (En) 0.06–8 μg ml−1, gentamicin (Gm) 0.25–32 μg ml−1, kanamycin (Km) 0.25–32 μg ml−1, nalidixic acid (Nal) 1–128 μg ml−1, neomycin (Nm) 0.5–64 μg ml−1, spectinomycin (Sp) 1–128 μg ml−1, streptomycin (Sm) 2–256 μg ml−1, sulfamethoxazole (Su) 2–256 μg ml−1, tetracycline (Te) 0.5–32 μg ml−1, trimethoprim (Tp) 0.25–32 μg ml−1, trimethoprim/sulfamethoxazole (Tp/Su) 0.12/2.37–8/152 μg ml−1.

2.3 Polymerase chain reaction

One bacterial colony was suspended in 1.0 ml phosphate buffered saline (PBS), centrifuged, the pellet resuspended in 100 μl 10 mM Tris, 1 mM EDTA buffer pH 8.0 and boiled for 10 min. After lysis the suspension was stored at −20°C. For each reaction 2 μl of the lysis suspension was used. The PCR amplification was performed according to Aarestrup et al. [11], using the PCR primers shown in Table 2. The annealing temperature was set 5°C below the calculated melting temperature (Tm) [12]. All strains were tested for the content of the genes intI [9], ant(3″)-Ia[13], pse-1 [14], qacEΔ1 and sul1 [8]. The PCR method was optimised to detect amplicons in the range of 300–1500 bp. The 5′-CS and 3′-CS primers (int1F and int1B) were used in combination with the primers for the antibiotic resistance genes to determine the content of integrons. All amplicons positive with the primers for antibiotic resistance genes were sequenced.

View this table:
Table 2

PCR primers for identification and characterisation of integrons and gene cassettes

Number, name and sequence of PCR primerAccession numberPosition of primerReference
1int1 F (5′-CS)GGC ATC CAA GCA GCA AGCU123381416→1433Collis and Hall [17]
2int1 B (3′-CS)AAG CAG ACT TGA CCT GATU123384831→4814Hall et al. [7]
3sul1 FCTT CGA TGA GAG CCG GCG GCX12869924→943Sundström et al. [18]
4sul1 BGCA AGG CGG AAA CCC GCG CCX128691360→1341
5qacEΔ1 FATC GCA ATA GTT GGC GAA GTX15370211→230Stokes and Hall [5]
6qacEΔ1 BCAA GCT TTT GCC CAT GAA GCX15370436→417
7ant (3″)-Ia FGTG GAT GGC GGC CTG AAG CCM10241514→533Hollingshead and Vapnek [19]
8ant (3″)-Ia BATT GCC CAG TCG GCA GCGM102411040→1023
9pse-1 FCGC TTC CCG TTA ACA AGT ACM69058323→342Huovinen and Jacoby [20]
10pse-1 BCTG GTT CAT TTC AGA TAG CGM69058742→723
  • F=nucleotide sequence forward 5′→3′; B=nucleotide sequence backward 3′←5′. Accession numbers are from the published sequences in the GenBank database.

2.4 DNA sequencing

After PCR amplification the DNA was purified using a QIA quick PCR purification kit (Qiagen, Hilden, Germany) and the nucleotide sequence was determined by cycle sequencer using the AmplitaqFS dye terminator kit and a 373A automatic sequencer (Applied Biosystems/Perkin Elmer, Foster City, CA, USA) [14]. For analysis of data DNAsis software was used (Hitachi Software Engineering Co., Ltd).

3 Results

3.1 Antimicrobial resistance

The resistance patterns of the eight Danish Typhimurium DT104 isolates are shown in Table 1. Two of the eight isolates were sensitive to all the antimicrobial agents tested. One isolate (No. 9423245) was resistant to Sp, Sm and Su and the five remaining isolates were multiresistant and had the resistance pattern: Ap, Cm, Sp, Sm, Su, Te.

3.2 PCR mapping and sequencing of antibiotic resistance gene cassettes and integrons

Two distinct integrons were detected by the use of PCR. From the first integron, a PCR product of 1008 bp was obtained (Fig. 1a) using the int1F and int1B primers. The sequence of the 1008-bp fragment confirmed the presence of the aminoglycoside resistance gene cassette ant (3″)-Ia, which conferred resistance to streptomycin and spectinomycin. The amplicon from the second integron was 1133 bp when the same primers were used and corresponded to the conserved regions of the integron together with the pse-1 gene cassette that encoded a β-lactamase (Fig. 1b).

Figure 1

Relation between integron structure and PCR products obtained using a range of primers for InC and InD detected in Danish Typhimurium DT104. The lines below the integron structure represent amplicons and the black boxes regions that were sequenced. The primers were used in different combinations to reveal the arrangement and content of the integrons. The numbers above the PCR products refer to the primers used (see Table 2). 5′-CS and 3′-CS represents the 5′ and 3′ conserved segments of the integron; qacE and sul1 encode resistance to disinfectant and sulfonamide respectively. #Numbers correspond to sequence position in GenBank accession number D43625 [13].

Both integrons were detected in five of the eight Danish Typhimurium DT104 isolates whereas the first integron with the 1008-bp amplicon was found in the one isolate resistant to Sp, Sm and Su only (Table 1). From this isolate six different PCR products were obtained using primers in different combinations as shown in Fig. 1 a. No amplicon was obtained using primers 9 and 10 (Table 2). The same six amplicons were also obtained from the five multiresistant strains. However, from each of these strains four additional PCR products were also obtained, which corresponded to regions of the pse-1 gene cassette. No PCR product was obtained from the two sensitive isolates with any primer combination indicating that no integrons were present. The amplicons and sequenced regions of the two integrons are illustrated in Fig. 1a,b.

4 Discussion

In the present study two integrons with different resistance gene cassettes were identified in Typhimurium DT104. The 1008-bp amplicon showed complete homology to the integron InC described by Kazama et al. [13] (GenBank accession number D43625) when the sequences were compared. In comparison with aadA2 from plasmid pSa this gene cassette differed by a single nucleotide [9].

The second integron encoded the β-lactamase gene pse-1 (Fig. 1b). Although the sequence published by Zuhldorf and Wiedemann [15] contained the antibiotic resistance gene cassette, 5′-CS and 59 bp, the presence of the integron was not acknowledged. In this study we have demonstrated that pse-1 is located within an integron. The presence of an integron and the pse-1 gene as a single cassette has to our knowledge not been described. We therefore suggest InD as a designation.

Both integrons detected were of class I which is the most prevalent type among clinical isolates [9]. Sallen et al. [16] reported a high incidence of an integron that contained the ant (3″)-Ia gene among clinical isolates of antibiotic-resistant Enterobacteriaceae but they did not find two different integrons in the same isolate. In contrast to these authors we have identified two distinct integrons within the same isolate.

The recognition of the two integrons does not account for the entire phenotypic resistance of the Typhimurium DT104 isolates. It is possible that larger integrons are present but were not detected or that the two observed integrons are positioned in a larger mobile DNA element that contains additional resistance genes.

Since the PFGE of all eight isolates indicated a close relationship between the Typhimurium DT104 isolates it is possible that all isolates came from a common sensitive ancestor and that the resistant isolates have accumulated integrons as a result of selection pressure. Further investigation of both location of the integrons and the additional resistance genes is in progress.


We thank Lars Sundström, Division of Microbiology, Uppsala University, Sweden for the supply of the strains containing the plasmids R751 and R100.1. We are grateful to Dorte Lau Baggesen, Danish Veterinary Laboratory for supplying the Danish Salmonella Typhimurium DT104 isolates, to David J. Platt, Glasgow Royal Infirmary for critically reading the manuscript and to Kirsten Vestergaard for technical assistance. This study was supported by a grant from the Danish Agricultural and Veterinary Research Council (9600012).


View Abstract