OUP user menu

Molecular analysis of methicillin-resistant Staphylococcus aureus reveals an absence of plasmid DNA in multidrug-resistant isolates

Jonathan M. Caddick, Anthony C. Hilton, Jessica Rollason, Peter A. Lambert, Tony Worthington, Tom S. J. Elliott
DOI: http://dx.doi.org/10.1016/j.femsim.2004.12.014 297-302 First published online: 1 June 2005


The number, diversity and restriction enzyme fragmentation patterns of plasmids harboured by 44 multidrug-resistant hospital-acquired methicillin-resistant Staphylococcus aureus (MR-HA-MRSA) isolates, two multidrug-resistant community-acquired MRSA (MR-CA-MRSA), 50 hospital-acquired MRSA (HA-MRSA) isolates (from the University Hospital Birmingham, NHS Trust, UK) and 34 community-acquired MRSA (CA-MRSA) isolates (from general practitioners in Birmingham, UK) were compared. In addition, pulsed-field gel electrophoresis (PFGE) type following SmaI chromosomal digest and SCCmec element type assignment were ascertained for each isolate. All MR-HA-MRSA and MR-CA-MRSA isolates possessed the type II SCCmec, harboured no plasmid DNA and belonged to one of five PFGE types. Forty-three out of 50 HA-MRSA isolates and all 34 CA-MRSA isolates possessed the type IV SCCmec and all but 10 of the type IV HA-MRSA isolates and nine CA-MRSA isolates carried one or two plasmids. The 19 non-multidrug-resistant isolates (NMR) that did not harbour plasmids were only resistant to methicillin whereas all the NMR isolates harbouring at least one plasmid were resistant to at least one additional antibiotic. We conclude that although plasmid carriage plays an important role in antibiotic resistance, especially in NMR-HA-MRSA and CA-MRSA, the multidrug resistance phenotype from HA-MRSA is not associated with increased plasmid carriage and indeed is characterised by an absence of plasmid DNA.

  • MRSA
  • Plasmid DNA
  • Multidrug-resistance

1 Introduction

Methicillin-resistant Staphylococcus aureus (MRSA) is a significant cause of nosocomial and community morbidity and mortality which, over the past two decades, has become a worldwide problem exacerbated by the emergence of multidrug-resistant MRSA (MR-MRSA). Such isolates demonstrate a reduced susceptibility to almost all clinically available antibiotics and are often only susceptible to glycopeptides and investigational drugs [1,2].

The mechanisms by which S. aureus acquires resistance may be classified into two main categories: chromosomal gene mutation and acquisition of resistance genes as a result of conjugation, transduction and transformation. S. aureus can acquire exogenous antibiotic resistance genes by horizontal transfer of genomic islands or phage-mediated transduction, or from mobile genetic elements such as plasmids and transposons.

In the case of acquisition of resistance genes possibly the most important resistance element acquired by S. aureus is the staphylococcal cassette chromosome mec (SCCmec), a highly mobile genomic island that carries the mecA gene that confers methicillin-resistance to S. aureus. Five distinct types of SCCmec designated I–V and a few variants have been described to date; three (I–III) are associated with hospital-acquired MRSA (HA-MRSA) [3] and two (IV and V) are associated with community-acquired MRSA (CA-MRSA) [46].

Arguably the best understood mobile genetic elements are plasmids, which have been widely studied in S. aureus and MRSA since the 1970s [710]. Currently, it is suggested that more than 90% of MRSA strains carry plasmids while numerous studies have supported the important role plasmids play in staphylococcal multidrug-resistance [7,1114]. It might therefore be anticipated that a positive correlation would exist between the number and diversity of plasmid DNA and increasing multidrug-resistance phenotype.

The purpose of this study was to screen HA-MRSA including MR-HA-MRSA and CA-MRSA isolates for the presence of plasmid DNA and to determine the extent to which plasmid carriage is involved in multidrug-resistance.

2 Methods

2.1 Antibiotic sensitivity tests

The resistance profile of all isolates was determined against a panel of 10 antibiotics using the British Society for Antimicrobial Chemotherapy (BSAC) disc diffusion method [15].

2.2 Bacterial isolates

A total of 130 MRSA isolates were investigated in this study. Ninety-four of the isolates were from blood cultures collected within the University Hospital Birmingham, NHS Trust, UK (UHB). Thirty-six of the isolates were samples sent to the UHB by general practitioners in the Birmingham area isolated from soft tissue infections from non-hospitalised patients and were designated CA-MRSA. Forty-four HA-MRSA isolates and two CA-MRSA isolates were designated multidrug-resistant, which for the purposes of this study is defined as resistance to six or more antibiotics. All isolates were stored at −70 °C on cryobeads (Microbank™, Pro-Lab Diagnostics, Canada).

2.3 Pulsed-field gel electrophoresis

Harvesting and preparation of cell blocks prior to lysis was performed as described by Lang et al. [16]. The prepared blocks were incubated for 24 h at 37 °C in 3 ml lysis solution (6 mM Tris pH 7.6, 100 mM EDTA pH 8, 100 mM NaCl, 0.5% lauroyl sarcosine, 1 mg ml−1 lysozyme) with 20 units of lysostaphin (Sigma, UK). The initial lysis solution was removed and the blocks were incubated for 48 h at 50 °C in 3 ml ESP (0.5 M EDTA pH 9, 1.5 mg ml−1 proteinase K (Sigma, UK) and 1% lauroyl sarcosine). The blocks were washed at room temperature twice for 2 h followed by two 1 h washes using TE buffer (10 mM Tris and 1 mM EDTA, pH 8). A portion of each agarose block (1 ×1 ×9 mm) was digested with 20 units of SmaI (Roche, UK) in 0.1 ml buffer for 16 h at 25 °C. The digested DNA samples were subjected to pulsed-field gel electrophoresis (PFGE) (CHEF Mapper system, BioRad, UK) under the conditions outlined by Bannerman et al. [17]. Wells 3 and 8 of all PFGE gels carried a SmaI chromosomal digest from S. aureus strain NCTC 8325 as a control and a molecular weight marker [18,19]. Gels were stained with 1 µg ml−1 of ethidium bromide for 45 min and destained for 45 min in distilled water. Gels were visualised under UV illumination and photographed using the GeneGenius Bio Imaging System (Syngene, UK). Images of PFGE profiles were compared visually and by use of Gel Compar II software (Applied Maths, Kortrijk, Belgium). The similarity between the isolates was calculated by the Dice coefficient, and a dendrogram was constructed using un-weighted pair groups using mathematical averages (UPGMA) clustering. The PFGE profiles were assigned letters according to their position in the dendrogram. Unique PFGE profiles were assigned the letter U followed by a number according to their position in the dendrogram. The interpretation of differences in banding patterns was performed using guidelines by Tenover et al. [19].

2.4 Multiplex PCR for mec element type assignment

Isolates were inoculated onto a BHI agar (Oxoid, UK) plate and incubated overnight at 37 °C. A single loop of colonies taken directly from the plate was suspended in 1 ml of TESS buffer (50 mM Tris, 5 mM EDTA, 50 mM NaCl and 50 mM sucrose, pH 8), centrifuged for 4 min at 5500g and the supernatant discarded. The pellet was washed twice with 1 ml of sterile double distilled water. The cells were subsequently resuspended in 1 ml of sterile double distilled water and transferred to a 1 ml sterile plastic cuvette. The absorbance of the suspension at 600 nm was measured and adjusted to a value of 1.7 with sterile double distilled water. A 0.1 ml aliquot of the adjusted suspension was transferred to a 0.2 ml PCR tube and incubated at 94 °C in a GeneAmp® PCR system 9700 (Applied Biosystems, UK) for 12 min. The heated cell suspension was subsequently centrifuged at 3000g for 10 s. The resulting supernatant containing extracted DNA was cooled to room temperature and used as template in a multiplex PCR. The multiplex PCR was performed in a 25 µl reaction volume using a GeneAmp® PCR system 9700 (Applied Biosystems, UK) with primers and cycle conditions as described by Oliveira and de Lencastre [20].

2.5 Preparation of plasmid DNA

Preparations of plasmid DNA were made using a modified protocol from the Sigma GenElute® plasmid miniprep kit (Sigma, UK). The UHB clinical isolate A02T1, identified as a carrier of plasmid DNA, was used as a positive control during each preparation. Test isolates of MRSA were recovered from frozen stocks by inoculating a single cryobead onto a BHI agar plate and incubating at 37 °C for 16 h. A single loop of 25 colonies was taken directly from the plate and washed in 0.5 ml of TES buffer (50 mM Tris, 5 mM EDTA and 50 mM NaCl, pH 8) and centrifuged at 5500g for 4 min in a bench-top microfuge. The supernatant was removed and the pellet resuspended in 0.1 ml of TESS buffer to which 20 µl of lysostaphin (0.5 units µl−1) (Sigma, UK) was added and briefly vortexed. Samples were incubated at 37 °C for 20 min until clear and then at 80 °C for 10 min. The resulting lysate was treated as described in the manufacturer's instructions, with the following modifications: 80 µl of resuspension solution was added to the sample lysate and in the final stage, plasmid DNA was eluted with 0.1 ml of sterile double distilled water. The quantity and quality of extracted plasmid DNA was determined spectrophotometrically. Only samples with A260/A280 ratios between 1.7 and 1.9 were analysed. Plasmid DNA samples were adjusted to a concentration of 1 µg µl−1 with sterile double distilled water and stored at −20 °C until required.

2.6 Restriction endonuclease digestion of plasmid DNA

Plasmid DNA was cleaved with restriction endonucleases HindIII, XbaI, EcoRI, HhaI, HaeIII, PstI or AluI (Promega, UK) according to the manufacturer's instructions using the appropriate reaction buffer. Digested plasmid and corresponding uncut plasmid were resolved in a 2% agarose gel containing 1 µg ml−1 ethidium bromide. Electrophoresis was performed in 1×TAE buffer (40 mM Tris, 1 mM EDTA and 0.1% (v/v) glacial acetic acid) at 65 volts for 80 min. Gels were visualised under UV illumination and photographed using the GeneGenius Bio Imaging System (Syngene, UK). Images of profiles of the digested plasmid DNA were compared visually and by use of Gel Compar II software (Applied Maths, Kortrijk, Belgium). The similarity between the isolates was calculated by the Dice coefficient, and a dendrogram was constructed using UPGMA clustering. Profiles were numbered according to their position in the dendrogram.

2.7 Statistical analysis

The presence and absence of plasmid DNA in multidrug-resistant and non-multidrug-resistant MRSA isolates was compared by means of the Yates' corrected χ2 statistic (Statistica 6.0, StatSoft, Tulsa, USA).

3 Results

3.1 Antibiotic resistance

The antibiotic resistance profiles (Table 1) indicated that there were 27 MR-HA-MRSA isolates and one MR-CA-MRSA isolate resistant to all antibiotics investigated except rifampicin, tetracycline and vancomycin. Seventeen MR-HA-MRSA isolates and one MR-CA-MRSA isolate were sensitive to mupirocin at high concentration in addition to rifampicin, tetracycline and vancomycin. The non-multidrug-resistant HA-MRSA isolates and CA-MRSA isolates were separated into 11 antibiogram types. All HA-MRSA were resistant to methicillin and sensitive to gentamicin, mupirocin (low and high concentrations) and vancomycin with the exception of two isolates which demonstrated resistance to gentamicin. Two of the HA-MRSA isolates were resistant to rifampicin and a further two were resistant to tetracycline. Ten of the HA-MRSA and nine CA-MRSA isolates were resistant to methicillin only.

View this table:
Table 1

Antibiotic resistance phenotype, pulsed-field gel electrophoresis type following digestion with SmaI, mec element assignment and plasmid type following digestion with AluI

MRAntibiotic resistance phenotypePFGE typemec typeTotal number of isolates (number of isolates)Number of plasmids (number of isolates)Plasmid type (number of isolates)
RSSSSSSSSSCIV10(7, 3)0(7, 3)
RSRSSSSSSSDIV2(1, 1)1(1, 1)p4(1, 1)
RRSSSSSSSSU3New 21(1)1(1)p5(1)
RRSSSSSSSSU4New 21(1)1(1)p5(1)
RRSSSSSSSSAIV6(3, 3)1(3, 3)p5(3, 3)
RRSSSSSSSSBIV5(1, 4)1(1, 4)p5(1, 4)
RRSSSSSSSSCIV18(14, 4)1(14, 4)p5(14, 4)
RRRSSSSSSSU2New 11(1)1(1)p2(1)
RRRSSSSSSSBIV4(2, 2)1(1), 2 (1, 2)p2(1) p4 + p5 (1, 2)
RRRSSSSSSSDIV2(2)2(2)p4 + p5(2)
RRRSSSSSSSU8IV1(1)2(1)p4 + p5(1)
RRRSSSSSSSU14IV1(1)2(1)p4 + p5(1)
RRRSSSSSSSU18IV1(1)2(1)p5 + p8(1)
RRRSSRSSSSU19variant 1a1(1)3(1)p1 + p4 + p7(1)
RRRSRSSSSSCIV2(2)2(2)p2 + p5(2)
RRRSRSRSSSU1variant IIIa1(1)0(1)
RRRRRSSSSSDIV2(2)2(2)p4 + p5(2)
MRRRRSRRSRSSFII16(15, 1)0(15, 1)
MRRRRSRRSRRSGII26(25, 1)0(25, 1)
  • MR designates multidrug-resistance. (Isolates resistant to six or more antibiotics).

  • Antibiotic abbreviations: Mc (methicillin), Em (erythromycin), Tm (trimethoprim), Rf (rifampicin), Fu (Fusidin), Gm (gentamicin), Tc (tetracycline), MuL (mupirocin-low), MuH (mupirocin-high) and Vm (vancomycin).

  • (Hospital acquired and community-acquired isolates).

3.2 SCC mec type

Multiplex PCR for mec element assignment revealed that all of the isolates investigated were mecA positive and characteristic of eight types (Table 1). All 44 MR-HA-MRSA isolates and 2 MR-CA-MRSA isolates were type II, 47 HA-MRSA isolates and 34 CA-MRSA isolates were type IV, four individual HA-MRSA isolates were designated type I, Ia, III, IIIa and three HA-MRSA isolates were characterised as belonging to two new variant types.

3.3 PFGE

Analysis of PFGE data revealed 31 different PFGE profiles. Eight PFGE clusters comprising of multiple isolates were designated types A–H. Two MR-HA-MRSA isolates, twelve HA-MRSA isolates and nine CA-MRSA isolates gave unique profiles and given the designation U followed by a number relating to the position of the isolate in the dendrogram. The MR-HA-MRSA and MR-CA-MRSA isolates were separated into three clusters designated F–H with two unique types. The non-multidrug resistant HA-MRSA (NMR-HA-MRSA) and NMR-CA-MRSA isolates were separated into four main clusters designated A–E with 21 unique types.

3.4 Plasmid DNA

Plasmid DNA was absent from all of the MR-HA-MRSA isolates with significance of p≤ 0.001. In addition, 10 NMR-HA-MRSA isolates and five NMR-CA-MRSA isolates that were only resistant to methicillin were also found not to carry plasmid DNA. One NMR isolate assigned SCCmec type I and another NMR isolate assigned type IIIa were also absent of plasmid DNA but were resistant to at least two additional antibiotics other than methicillin. The remaining NMR isolates all carried at least one plasmid, seven of these carried two plasmids and one carried three plasmids.

3.5 Restriction enzyme fragmentation patterns of plasmid DNA

The restriction endonucleases HindIII, XbaI, EcoRI, HhaI, HaeIII, PstI or AluI were screened to determine their ability to cleave plasmid DNA. Of these enzymes only AluI fully digested plasmid DNA and provided clearly resolved, discriminatory profiles. In total there were eight different AluI plasmid types designated p1–p8, presence of multiple plasmids accounted for nine different plasmid profiles.

4 Discussion

It is widely accepted that plasmid DNA may confer antibiotic resistance [1114,21] and therefore it might be assumed that MR-MRSA isolates would carry additional plasmids compared to MRSA with a less resistant phenotype. However, the results of this study suggest quite the opposite in that MR-MRSA isolates were significantly associated (p≤ 0.001) with an absence of plasmid DNA. The results indicate that isolates with the multidrug-resistance phenotype belong to one of five closely related PFGE profile types, possess the type II SCCmec element and do not carry plasmid DNA. It follows therefore that all resistance determinants in the MR-MRSA isolates are chromosomally encoded.

The type II SCCmec element, in addition to mecA, carries genes that confer resistance to non-β-lactam antibiotics [3]. In the case of type II SCCmec an integrated copy of plasmid pUB110 and transposon TN554 are carried in addition to the mecA gene conferring resistance to a number of antibiotics. The plasmid pUB110 encodes tobramycin and gentamicin resistance while transposon TN554 encodes erythromycin resistance [21,22]. Both erythromycin and gentamicin resistance were observed in the MR-MRSA isolates from UHB. It is possible that the type II SCCmec element in these MR-MRSA isolates is similar to the type II SCCmec elements sequenced from the type strains N315 and Mu50. Both of these SCCmec elements encode resistance to β-lactams, macrolide-lincosamide-streptogramin B, and aminoglycosides [21]. Additional point mutations and possible acquisition of resistance genes that are now chromosomally encoded may further explain the antibiotic resistance phenotype expressed by our MR-HA-MRSA. Two of the CA-MRSA isolates in this study were found to have the multidrug-resistance phenotype. Both possessed the type II SCCmec element and had a PFGE profile the same as the MR-HA-MRSA isolate; one shared the type F PFGE profile and the other shared the type G PFGE profile.

Interestingly, most of the NMR-HA-MRSA were of the type IV SCCmec element assignment; an assignment commonly associated with CA-MRSA [4,6]. Using PFGE, mec type assignment, the antibiotic resistance phenotype and restriction enzyme fragmentation patterns of plasmid DNA it was not possible to distinguish NMR-HA-MRSA from CA-MRSA. Also, each plasmid DNA profile (with one exception) observed in the HA-MRSA isolates was also observed in the CA-MRSA isolates. These observations indicate that MRSA containing the type IV SCCmec element and carrying resistance plasmids may be better able to survive and spread in both the hospital and community environs. The fact that few MR-MRSA isolates were identified in the community may suggest that an MRSA population with a multidrug-resistance phenotype and possessing the type II SCCmec element is only truly sustainable in a hospital environment where antibiotic pressure is relatively high. Currently, it is only possible to speculate what biological cost antibiotic resistance comes at for MR-MRSA and MRSA in both hospitals and in the community [2325].

The present study demonstrates that a multidrug-resistance phenotype in MRSA from the UHB is associated with one of five PFGE profile types, the type II SCCmec assignment and the absence of plasmid DNA and that NMR-HA-MRSA and CA-MRSA are of a similar lineage.


  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].
  24. [24].
  25. [25].
View Abstract