OUP user menu

Detection of multidrug-resistant Salmonella typhimurium DT104 by multiplex polymerase chain reaction

Ashraf A Khan, Mohammed S Nawaz, Saeed A Khan, Carl E Cerniglia
DOI: http://dx.doi.org/10.1111/j.1574-6968.2000.tb08921.x 355-360 First published online: 1 January 2000


Salmonella typhimurium definitive type 104 (DT104) is a virulent pathogen for humans and animals with many strains having multiple drug resistance characteristics. The organism typically carries resistance to ampicillin, chloramphenicol, florfenicol, streptomycin, sulfonamides, and tetracycline (ACSSuT-resistant). A multiplex PCR method was developed to simultaneously amplify four genes, florfenicol (flost), virulence (spvC), invasion (invA), and integron (int) from S. typhimurium DT104 (ACSSuT-type). Twenty-two ACSSuT-resistant DT104 isolates in our collection gave 100% positive reactions to this PCR assay by amplifying 584-, 392-, 321- and 265-bp PCR products, using primers specific to the respective target genes. One Salmonella strain DT23, ACSSuT-resistant, phage type 711 failed to amplify the 584-bp fragment, indicating that this method is specific for DT104-type ACSSuT-resistant S. typhimurium strains. One clinical and one bovine ASSuT-resistant strains that were sensitive to chloramphenicol and florfenicol did not yield a 584-bp fragment, indicating the absence of the flost gene. This method will be useful for rapid identification of ACSSuT-type DT104 strains from clinical, food and environmental samples.

  • Multiplex PCR
  • Salmonella typhimurium DT104
  • Multi-drug resistant
  • Detection
  • ACSSuT-R type

1 Introduction

Salmonella typhimurium definitive type 104 (S. typhimurium DT104 or DT104) is an increasingly common multiple-antibiotic-resistant strain that has rapidly emerged as a world health problem [1,2]. Most DT104 strains are characterized by chromosomal resistance to ampicillin (A), chloramphenicol (C), streptomycin (S), sulfonamides (Su), and tetracycline (T) and are referred to as ACSSuT-type [3,4].

The Centers for Disease Control and Prevention (CDC, Atlanta, GA) have established a Salmonella and Antimicrobial Monitory Resistance Surveillance system [1,4]. In the 1996 National Salmonella Antimicrobial Resistance Monitoring System Study, the CDC serotyped 3903 Salmonella isolates and determined that 976 (25%) were S. typhimurium. Approximately 28% of the S. typhimurium isolates had the R type ACSSuT compared to just 7% in 1990 [5]. Two other antibiotic resistance monitoring programs, Periodic Surveys of Antimicrobial Drug Resistance in Sentinel Counties and the National Antimicrobial Resistance Monitoring System, have reported similar sharp increases in multidrug-resistant Salmonella spp. [1]. The mechanisms for resistance to sulfonamides, streptomycin, ampicillin and florfenicol in DT104 have been described previously [6,7]. Most of the DT104 ACSSuT-type strains contain at least two integrons, one containing the aminoglycoside resistance gene cassette ant (3″)-la, which encodes resistance to streptomycin, and one containing a β-lactamase resistance gene cassette, pse-1, which encodes resistance to ampicillin [6,7]. A gene encoding sulfonamide resistance (sul-1) was found in the 3′ conserved sequences of both integrons [7]. Florfenicol is a fluorinated analog of chloramphenicol approved by the FDA in 1996 for the treatment of bovine respiratory pathogens. Previous studies have shown that bacterial isolates which were resistant to chloramphenicol were sensitive to inhibition by fluorinated analogs [8,9]. The mechanism of resistance to chloramphenicol and florfenicol in DT104 was recently described [8] and the gene flost confers resistance to both of these antibiotics.

Current diagnostic methods used for the identification and characterization of ACSSuT-type DT104 strains require several days [10]. Since DT104 ACSSuT-type strains are becoming widespread globally, development of a rapid and sensitive method for the diagnosis of DT104 ACSSuT-type strain is desirable. PCR technology combines simplicity with a potential for amplification of a specific fragment of nucleic acid and has been used to identify the presence of specific pathogens directly from clinical specimens, food and water [11,12]. We report a multiplex PCR method with four sets of primers, florfenicol (flost), integron (int), invasion (invA), and virulence (spvC) that permitted specific detection of S. typhimurium DT104 (ACSSuT-type) strains. In addition, we assessed these four target genes in multidrug-resistant DT104 isolates from different sources.

2 Materials and methods

2.1 Bacterial strains

The S. typhimurium strains DT104 and other multidrug-resistant strains used in this study were obtained from the Center for Food Safety and Applied Nutrition, United States Food and Drug Administration, Washington, DC (Table 1). These isolates were from human, cattle, swine, sewage, clinical and cattle feces samples. All isolates were serotyped at FDA laboratories or by their original sources (Table 1). The bacteria were grown and maintained in Nutrient Broth (Difco Laboratories, Detroit, MI) at 37°C.

View this table:
Table 1

Genotypic and phenotypic characteristics of S. typhimurium DT104 isolates

DesignationSourceSerotypeAntibiotic resistance
S. typhimuriumDT1CattleDT104A, C, F, S, Su, T
S. typhimuriumDT2HumanDT104A, C, F, S, Su, T
S. typhimuriumDT3BovineDT104A, C, F, S, Su, T
S. typhimuriumDT4BovineDT104A, C, F, S, Su, T
S. typhimuriumDT5BovineDT104A, C, F, S, Su, T
S. typhimuriumDT6BovineDT104A, C, F, S, Su, T
S. typhimuriumDT7BovineDT104A, C, F, S, Su, T
S. typhimuriumDT8BovineDT104A, S, Su, T
S. typhimuriumDT9ClinicalDT104BA, C, F, S, Su, T
S. typhimuriumDT10BovineDT104BA, C, F, S, Su, T
S. typhimuriumDT11ClinicalDT104A, C, F, S, Su, T
S. typhimuriumDT12ClinicalDT104A, S, Su, T
S. typhimuriumDT13SwineDT104A, C, F, S, Su, T
S. typhimuriumDT14ChickenDT104A, C, F, S, Su, T
S. typhimuriumDT15HumanDT104A, C, F, S, Su, T
S. typhimuriumDT16HumanDT104Su
S. typhimuriumDT17SewageDT104S, Su
S. typhimuriumDT18HumanDT104A, C, F, S, Su, T
S. typhimuriumDT19Cattle fecesDT104Su
S. typhimuriumDT20HumanDT104A, C, F, S, Su, T
S. typhimuriumDT21ClinicalU302A, C, F, S, Su, T
S. typhimuriumDT22Bovine208A, S, Su, T
S. typhimuriumDT23Bovine771A, C, S, Su, T
S. typhimuriumDT24ClinicalU302A, C, F, S, Su, T
S. typhimuriumDT25Bovine771A, S, Su, T
S. typhimuriumDT28CattleDT104A, C, F, S, Su, T
S. typhimuriumDT29CattleDT104A, C, F, S, Su, T
S. typhimuriumDT30CattleDT104A, C, F, S, Su, T
S. typhimuriumDT31CattleDT104A, C, F, S, Su, T
S. typhimuriumDT32CattleDT104A, C, F, S, Su, T
S. typhimuriumDT33PorcineDT104A, C, F, S, Su, T
S. typhimuriumDT34NVSLDT104dA, C, F, S, Su, T
S. typhimurium14028ATCC

2.2 Antibiotic resistance testing

The strains were tested for resistance to ampicillin, chloramphenicol, streptomycin, sulfonamides, tetracycline, and florfenicol, by the disk agar diffusion method performed on Mueller-Hinton (Difco Laboratories, Detroit, MI) agar plates. The antibiotic disks used in this study were purchased from Difco Laboratories unless otherwise specified. Disks contained the following amounts of antibiotic: ampicillin 10 μg (A), chloramphenicol 30 μg (C), tetracycline 30 μg (T), streptomycin 10 μg (S), florfenicol 30 μg (F) (Schering-Plough Animal Health, Kenilworth, NJ), and sulfisoxazole 250 μg (Su) (Becton Dickinson and Company, Cockeysville, MD). S. typhimurium ATCC 14028 was used as a sensitive control. The sensitivity and resistance were determined as per the criteria of the National Committee for Clinical Laboratory Standards [13].

2.3 Preparation of the bacterial samples

Bacteria were lysed with an equal volume of 0.2% (w/v) Triton X-100 and heated at 100°C for 5 min in a boiling water bath. The samples were allowed to cool immediately in ice for 5 min and used immediately in the PCR. Whenever necessary, chromosomal DNA was purified by phenol-chloroform-isoamyl alcohol extraction by the method of Wheatcroft and Watson [14]. Purified DNA was precipitated with ethanol, resuspended in sterile buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) at about 1 μg ml−1, based on the optical density at 260 nm, and stored at −20°C until use.

2.4 Primer design

Oligonucleotide sequences of florfenicol (flost), integron (int), invasion (invA), and virulence (spvC) genes for the multiplex PCR amplification of DT104 are listed in Table 2. The gene primers were designed by using published sequences (GenBank accession numbers AF097407, AF071555, M90846 and M64295, respectively) and DNASTAR primer select software (DNASTAR, Inc., Madison, WI). The flost, int, invA and spvC genes primer sequences were predicted to yield 584, 265, 321 and 392-bp products. The specificity of the primers was confirmed by the GenBank database ‘Blast’ program. The primers used in this study were synthesized by Universal DNA Inc., Tigard, OR.

View this table:
Table 2

Sequences of oligonucleotide primers

PrimerTarget geneSequencePCR product (bp)Accession number

2.5 PCR amplification conditions

The amplification of the individual target genes for all four pairs of primers from one S. typhimurium (ACSSuT-type DT104) strain was performed by using a DNA thermal cycler (PE Applied Biosystems model 480, Foster City, CA) and the GeneAmp kit with Taq DNA polymerase (PE Applied Biosystems) in 0.5-ml microcentrifuge tubes. The reaction mixture (50 μl total volume) consisted of 38.75 μl of sterile water, 5 μl of 10×PCR buffer (100 mM Tris-HCl (pH 8.3), 500 mM KCl, 15 mM MgCl2, 0.1% (w/v) gelatin), 4 μl of deoxyribonucleoside triphosphates (2.5 mm each dATP, dTTP, dGTP and dCTP), 0.5 μl of each primer (stock concentration, 100 μM), 1–10 μl of template, and 0.25 μl (5 U μl−1) of Taq DNA polymerase. After overlaying with sterile mineral oil, the samples were subjected to PCR amplification. Preincubation was at 94°C for 90 s. Thirty PCR cycles were run under the following conditions: denaturation at 94°C for 45 s, primer annealing at 60°C for 45 s, and DNA extension at 72°C for 90 s in each cycle. After the last cycle, the PCR tubes were incubated for 3 min at 72°C and then at 4°C. For multiplex PCR, 100 ng (each) of primers was used. The optimization of multiplex PCR was achieved by individually amplifying all four genes at 1.5, 2.0, 2.5 and 3.0 mM MgCl2. The annealing temperature was also optimized with all four pairs of primers.

2.6 Detection of amplified DNAs

Three microliters of the reaction mixture were analyzed by standard submarine gel electrophoresis (1.5% agarose; 5 V cm−1), and the reaction products were visualized by staining with ethidium bromide (0.5 μg ml−1 in the running buffer). As negative control, Escherichia coli cells or DNA was used. A reagent blank contained all components of the reaction mixture with the exception of template DNA, for which sterile distilled water was used. This blank was included in every PCR procedure. The thermocycler, tips and pipetters used for preparing the PCR reagents and template DNA were kept in a different location from where the gels were loaded, stained and photographed. All reagents used in an experiment were taken from the freezer and discarded at the end of the day.

2.7 Specificity of primers

The specificity of multiplex PCR for DT104 ACSSuT strains was evaluated with 33 S. typhimurium strains (Table 1).

3 Results and discussion

This paper describes a simple and specific multiplex PCR method to detect multidrug-resistant (ACSSuT-type) S. typhimurium DT104 strains in a single step. All Salmonella strains listed in Table 1 were tested for drug resistance using a disk agar diffusion method. We found that ACSSuT-type S. typhimurium DT104 isolates resistant to chloramphenicol were also resistant to florfenicol, except that one strain (DT23, phage type 771) was sensitive to 30 μg of florfenicol. This may be due to the absence of the flost gene. Previous studies have shown that most of the bacterial isolates which were resistant to chloramphenicol were sensitive to inhibition by fluorinated analogs of chloramphenicol [8,9]. The antibiotic resistance profiles of DT104 strains and other multidrug-resistant S. typhimurium strains are listed in Table 1. Among 32 strains studied, 24 were resistant to A, C, S, Su, T and F, four strains (DT 8, 12, 22 and 25) were resistant to A, S, Su and T. One strain DT23 (phage type 771) was sensitive to F but resistant to A, C, S, Su, and T. Strain DT17 was sensitive to A, C, F and T, and DT16 and 19 were sensitive to A, C, S, F and T. Two strains, DT21 and 24, were U302 phage type and had antibiotic profiles similar to other DT104 (ACSSuT-type) strains and were also resistant to florfenicol.

In preliminary experiments to determine the optimum PCR amplification condition for multiplex PCR to simultaneously detect four genes, we utilized purified genomic DNA from multidrug-resistant DT104 strain DT18 as a source of template DNA. Individual primer pairs from four different genes were selected, targeting the invasion gene, virulence plasmid gene, integrase gene and florfenicol resistance gene.

Fig. 1 (lanes 2, 3, 4 and 5) shows the amplified product from strain DT18 with four pairs of individual primer sets at 1.5 mM MgCl2 and an annealing temperature of 60°C. When four pairs of primers were mixed together and subjected to multiplex PCR, using the same annealing temperature and all other conditions, the yields of 548- and 392-bp products were lower than the other two PCR products. Multiplex PCR specificity and sensitivity depends on several parameters like annealing temperature, Mg2+ concentration, and primer concentration. We found that lowering the annealing temperature by 2–5°C yielded several non-specific bands. The optimum temperature for these primers was 60°C. The optimum Mg2+ concentration for the multiplex PCR was determined by adding 1.5, 2.0, 2.5 and 3.0 mM MgCl2 in the PCR reaction. The 2.0 mM MgCl2 gave specific and high yields of all four target genes (Fig. 1, lane 9).

Figure 1

Agarose gel electrophoresis of S. typhimurium DT18 strain PCR amplified products using individual pairs of primers and multiplex PCR. Lanes: 1, 100-bp DNA ladder as a size standard; 2, 392-bp PCR product using VirF and VirR primer pairs; 3, 584-bp PCR product using FloF and FloR primer pairs; 4, 321-bp PCR product using InvF and InvR primer pairs; 5, 265-bp PCR product using IntF and IntR primer pairs; 8, 100-bp DNA ladder as a size standard; 9, multiplex PCR products (584, 392, 321 and 265 bp) using four pairs of primers described above.

This protocol was used to screen all Salmonella strains in this study (Fig. 2A,B). Twenty-two ACSSuT-type DT104 strains and two U302 phage type ACSSuT-type strains gave 100% positive (amplified all four genes) PCR assays. These strains are all resistant to florfenicol. In this study we included Salmonella (strain #DT23, phage type 771) to verify the specificity of the PCR assay. Although this strain was resistant to chloramphenicol, it did not give any 548-bp PCR product, probably since this strain is sensitive to florfenicol. The multiplex PCR of this strain amplified only the invA and int genes, indicating that the flost gene was an important target gene for the detection of ACSSuT-type DT104 strains. A complete nucleotide sequence of the florfenicol resistance gene was determined by Bolton et al. [15] as 1202 bp. They also proposed phenotypic and genotypic methods for the identification of multidrug-resistant S. typhimurium DT104 by the use of a probe to determine chloramphenicol and florfenicol resistance. When 44 multidrug-resistant DT104 strains were tested, all of them were found resistant to florfenicol and chloramphenicol [15].

Figure 2

Agarose gel electrophoresis of multiplex PCR products amplified from S. typhimurium strains described in Table 1, using four pairs of primers. A: Lanes 1 and 20, standard size marker (100-bp DNA ladder as a size standard); lanes 2–19, DT1–DT18. B: Lanes 1 and 17, standard size marker (100-bp DNA ladder); lanes 2–16, DT18–DT34.

The invasion gene operon, invA, is essential for full virulence in Salmonella and it is thought to trigger the internalization required for invasion of deeper tissues [16]. Our PCR results indicate all were positives for the invA gene (Fig. 2A,B). A similar result was reported in a study where 245 Salmonella isolates from poultry products, wastewater, and human sources contained the invA gene [17]. A plasmid, spvC, that is present in Salmonella spp. interacts with the host immune system and is responsible for an increased growth rate in host cells [18]. Bolton et al. [15] found that 98% of all the S. typhimurium tested were positive for invA and 88% percent were positive for spvC gene. Our PCR results also indicated that the spvC gene was present in 97% (31 of 32) of S. typhimurium strains. All multidrug ACSSuT-resistant DT104 strains were positive for spvC.

Integrons capture and express mobile genes known as cassettes, which are, in most cases, antibiotic resistance genes [19,20]. The integrase gene is an essential part of all integrons; it encodes a site-specific recombinase that catalyzes the insertion of the gene cassettes into the integron. The PCR data indicated that 94% (30 of 32) of the strains tested were positive for the int gene. Two DT104 strains (DT16 and 19, sensitive to ACST antibiotics) did not amplify a 265-bp PCR product from the integrase gene. One DT104 strain (DT17, which was resistant to streptomycin and sulfonamide, but sensitive to most of the antibiotics tested) was positive for the int gene. The PCR and antibiotic resistance profiles of ACSSuT-resistant DT104 and U302 phage type were similar; however, the 771 and 208 phage type strains DT22, DT23 and DT25 (Table 1) did not have profiles similar to DT104 strains. These results indicate that these two phage types (DT104 and U302) may be closely related to each other [21]. Similar results were observed by Briggs and Fratamico [21] when they used a long PCR method to amplify the integron sequences between the aadA2 and blaCARB-2 genes of DT104 and U302 type ACSSuT-resistant S. typhimurium[21]. Recently, Sandvang et al. [7] have characterized two different integrons and found that multiresistant DT104 strains possess these integrons. The first integron encodes the aminoglycoside resistance cassette ant (3″)-la, which confers resistance to spectinomycin and streptomycin. The second integron contains the pse-1β-lactamase gene cassette. Integrons have been documented in other Gram-negative organisms, such as Pseudomonas, E. coli and Shigella spp. [22,23]. Therefore, it is important to have combinations of all these target genes for specific detection of multiple-drug-resistant Salmonella DT104 strains.

These data indicate that multiplex PCR analysis utilizing flost, invA, spvC and int primer sets can specifically identify multidrug-resistant ACSSuT-type DT104 S. typhimurium strains. This multiplex PCR method, described in this study, will allow the characterization and identification of environmental and clinical S. typhimurium DT104 ACSSuT-resistant strains.


The authors thank Dr. Ben D. Tall and Dr. Farukh M. Khambaty, Center for Food Safety and Applied Nutrition, U.S. Food and Drug Administration, Washington, DC for providing Salmonella DT104 strains. We thank Dr. John B. Sutherland and Dr. Robert D. Wagner for critical review of the manuscript and Barbara Jacks for illustrations.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
View Abstract