OUP user menu

The prevalence of the Staphylococcus aureus tst gene among community- and hospital-acquired strains and isolates from Wegener's Granulomatosis patients

Ruud H. Deurenberg, Rutger F. Nieuwenhuis, Christel Driessen, Nancy London, Frank R. Stassen, Frank H. Tiel Van, Ellen E. Stobberingh, Cornelis Vink
DOI: http://dx.doi.org/10.1016/j.femsle.2005.03.002 185-189 First published online: 1 April 2005

Abstract

To allow rapid identification of toxic shock syndrome toxin-1 (TSST-1)-producing Staphylococcus aureus strains, a real-time PCR assay for the detection of the tst gene, which encodes TSST-1, was developed. The assay was applied to S. aureus isolates from patients with Wegener's Granulomatosis (WG), as well as isolates that were classified as either community- (CA) or hospital-acquired (HA). No significant difference in the percentage of tst-positive strains was observed between isolates from WG patients and CA isolates (24% and 25%, respectively). In contrast, only 14% of the HA isolates were tst-positive (p < 0.05). Investigation of the clonal relationship between tst-positive CA and HA strains could indicated the recent emergence of a virulent S. aureus clone in the community.

Keywords
  • Toxic shock syndrome toxin-1
  • Staphylococcus aureus
  • Real-time PCR
  • PFGE

1 Introduction

The pathogenicity of Staphylococcus aureus can in part be attributed to the production of pyrogenic toxins. Because secretion of these toxins causes excessive stimulation of T-lymphocytes, they are called superantigens (SAgs) [1]. To date, 15 different SAgs have been identified in S. aureus, i.e., staphylococcal enterotoxins (SE) A–M, exfoliative toxins (ET) A and B, and toxic shock syndrome toxin-1 (TSST-1) [24].

TSST-1 is a 29.1-kDa protein that is encoded by the S. aureus tst gene [5,6]. The release of TSST-1 into the bloodstream may give rise to a variety of severe clinical conditions, such as toxic shock syndrome (TSS), sudden infant death syndrome, and Kawasaki syndrome. The tst gene is present in up to 70% of the S. aureus strains isolated from patients with TSS. TSS is characterized by high fever, erythematous rash formation, hypotension and major oxygen involvement, which may lead to multi organ failure. Without appropriate therapy, a lethal shock may develop within 24 h after the onset of symptoms [1,5,7,8]. Although most cases (two thirds) of TSS are associated with tampon use, an increasing number of cases are related to localized infections, surgical complications and insect bites [8]. Musser et al. [9] showed that the majority of female patients affected with TSS had a single clone of S. aureus that was well adapted to colonization in the genital tract. TSS is usually treated with proper drainage of surgical wounds, a high dose of a β-lactam antibiotic and γ-globulin [11]. Nevertheless, TSS still has a lethality rate of about 30% [5].

The potency of TSST-1 lies in its ability to efficiently induce T-cell proliferation and activation (10,000-fold more efficiently than other antigens). This is due to bridging the antigen-presenting cells (APC) and T-lymphocytes through binding the major histocompatibility complex (MHC) class II on APCs and specific variable regions on the β-chain of both CD4 and CD8 antigen receptors. Consequently, a T-helper 1 type response is mounted, which results in the massive release of interleukins and cytokines [7]. It has previously been suggested that TSST-1 as well as other SAgs may play a role in patients suffering from Wegener's Granulomatosis (WG), a disease in which organs are damaged through inflammation of blood vessels. This finding was based on the observation of an increased incidence of nasal carriage of S. aureus in combination with chronic activation of circulating T-cells in WG patients [10].

In order to both control and monitor the spread of S. aureus strains producing TSST-1 in the community as well as in the hospital, it is important to be able to rapidly identify these strains. Therefore, a real-time PCR (TaqMan?) assay for the detection of the S. aureus tst gene was developed. The assay was used to determine the prevalence of tst-positive strains among community-acquired (CA) and hospital-acquired (HA) S. aureus bloodstream isolates, and among isolates from nasal swabs from WG patients. Finally, the clonal relation between the TSST-1-positive CA and HA strains as well as all S. aureus isolates from WG patients was investigated using pulsed-field gel electrophoresis (PFGE).

2 Materials and methods

2.1 Bacterial isolates

One-hundred and six S. aureus strains consisting of 55 TSST-1-negative and 51 TSST-1-positive S. aureus isolates, which were previously characterized using a conventional tst-specific PCR method [3,12], were used to investigate the characteristics of the tst real-time PCR assay. To test for potential cross-reactivity with other staphylococcal species, clinical isolates of Staphylococcus epidermidis (n= 14), S. capitis (n= 3), S. haemolyticus (n= 4), S. chromogenes (n= 1), S. cohnii (n= 1), S. hominis (n= 1), S. sciuri (n= 1) and S. warneri (n= 1) were subjected to the TSST-1 real-time PCR. The strains were identified with the API Staph identification system (Biomérieux, Boxtel, The Netherlands).

A random selection of 86 methicillin-susceptible S. aureus (MSSA) bloodstream isolates from individual patients in our hospital, a tertiary 680-bed university hospital, was investigated for the presence of the tst gene. These strains, isolated between 1999 and 2003, included 51 CA S. aureus isolates (isolated within 48 h after patient admission to the hospital) and 36 HA S. aureus isolates (isolated after 72 h of admission). In addition, 16 S. aureus strains, isolated from nasal swabs in 2004, from WG patients were investigated. Each of these strains was found to be catalase and coagulase positive, confirming the identification of these isolates as S. aureus.

2.2 Primer and probe design

Primers (Sigma-Genosys, Haverhill, United Kingdom) and TaqMan?-probes (Applied Biosystems [ABI], Nieuwerkerk a/d Ijssel, The Netherlands) were designed based on the published sequences of the S. aureus-specific tst and fem A genes (Table 1), using the computer program Primer Express 2.0 (ABI, Nieuwerkerk a/d Ijssel, The Netherlands). The specificity of the primer and probe sequences was confirmed by screening sequence databases using BLAST (http://www.ncbi.nlm.nih.gov/blast).

View this table:
Table 1

Characteristics of primers and probes for the femA and TSST-1 real-time PCR

NameSequence 5′? 3′Probe labelAccession no.LocationReference
femA_FPaAGAGTTTGGTGCCTTTACAGATAGX17688642–665[13]
femA_RPbGTCATAGTGGCCAACAGTTTGC707–687
femA_PRcTGCCATACAGTCATTTCMGB, FAM668–684
TSST-1_FPaTCATCAGCTAACTCAAATACATGGATTJ02615537–563[14]
TSST-1_RPbTGTGGATCCGTCATTCATTGTT624–603
TSST-1_PRcTCCAATAACCACCCGTTTTATCGCTTGAATAMRA, FAM598–570
  • aFP, forward primer.

  • bRP, reverse primer.

  • cPR, probe.

2.3 TaqMan assay for femA and TSST-1

A PCR specific for the S. aureus femA gene was developed to serve as a positive control for the tst-specific PCR. Assay conditions, such as primer and probe concentrations, were optimized according to the guidelines from both the Primer Express 2.0 software program and the manual (Protocol) of the TaqMan? Universal PCR Master Mix (ABI, Nieuwerkerk a/d Ijssel, The Netherlands). The following reaction conditions were used in the TaqMan? assay: 0.3 μM of fem A_FP or TSST-1_FP, 0.3 μM of fem A_RP or TSST-1_RP, 100 nM of fem A-PR or TSST-1_PR, 1× TaqMan? Universal PCR Master Mix (ABI, Nieuwerkerk a/d Ijssel, The Netherlands) and 20 μl of purified genomic DNA (isolated from a 1 McFarland (3 × 108 colony forming units [CFU]/ml) suspension using the Wizard? Genomic DNA Purification Kit (Promega, Leiden, The Netherlands) [15]) or 20 μl of a 1:100 diluted 1 McFarland suspension in a total volume of reaction mixture of 50 μl. Amplification was performed on the ABI PRISM 7000 Sequence Detection System using the following program: 2 min at 50 °C, 10 min at 95 °C, followed by 42 cycles of 15 s at 95 °C and 60 s at 60 °C.

2.4 Pulsed-field gel electrophoresis (PFGE)

PFGE analyses was carried out by digestion of the S. aureus chromosomal DNA with Sma I (Invitrogen, Breda, The Netherlands) essentially as described previously [16]. The PFGE patterns were analyzed with Dice comparison and unweighted pair group matching analysis (UPGMA) settings with GelCompar II 3.5 (Applied Maths, Sint-Martens-Latem, Belgium) according to the scheme of Tenover et al. [17]. The position tolerance was set at 2.0% and isolates with a similarity index of 0.80 or more were classified as a clonal group.

3 Results and discussion

3.1 Determination of the detection limit of the tst real-time PCR assay

Initially, the tst TaqMan assay was tested using both purified bacterial DNA as well as bacterial suspensions as input material. Since the assay performed similarly using both types of material (data not shown), the assay was optimized by adding bacterial suspensions directly to the reaction mixtures.

To assess the detection limit of the assay, five tenfold dilutions were made of a 1 McFarland suspension of a tst-positive S. aureus strain (HT.2004.0349). The number of CFUs in each dilution was determined with an Eddy Jet automatic spiral platter (IUL Instruments, Barcelona, Spain). Then, the bacterial suspensions were tested directly in the real-time PCR assay in three independent runs. As shown in Fig. 1(a), the different bacterial suspensions generated mean Ct values ranging from 19.96 to 37.36. The lower detection limit of the assay was approximately 60 CFU per reaction, which corresponded to approximately 3 CFU/μl, i.e., a 10−5 dilution of a 1 McFarland suspension. The standard curve derived from the amplification plot (Fig. 1 (b)) showed a near optimal slope of −3.4, indicating a near optimal reaction efficiency. Furthermore, the standard curve showed that the assay had a broad dynamic range.

Figure 1

(a) PCR amplification curves of serial dilutions of a tst-positive S. aureus strain (HT.2004.0349). The strengths of the suspensions ranged from 3 × 103 to 3 × 108 CFU/ml (1 McFarland). ΔRn indicates the normalized fluorescent reporter value, subtracted from the background value. 3 × 108 CFU/ml (◻); 3 × 107 CFU/ml (∘); 3 × 106 CFU/ml (♦); 3 × 105 CFU/ml (△); 3 × 104 CFU/ml (––); 3 ×+103 CFU/ml (×); NTC, no template control (●); threshold (- - - -). (b) Standard curve generated with the Ct values from the amplification plots shown in (a).

3.2 Sensitivity and specificity of the assay

To determine the sensitivity and specificity of the assay, known tst-positive (n= 51) and tst–negative (n= 55) S. aureus isolates were tested. The tst-positive strains produced Ct values between 23.34 and 31.83 (mean = 26.51, SD = 1.37), which was in the range of the tst-positive control strain. All tst-negative strains did not generate detectable signals (data not shown). As a control for potential inhibition of PCR reactions, all isolates were also tested for the presence of the S. aureus-specific fem A gene by real-time PCR. As expected, each of these isolates was positive in this assay, with Ct values ranging from 26.01 to 30.05 (mean = 27.98, SD = 0.94) (data not shown).

The tst PCR assay was investigated further by testing a panel (n= 26) of staphylococcal species other than S. aureus. None of these strains generated positive results (data not shown) indicating that there was no cross-reactivity with other staphylococcal species.

3.3 Reproducibility of the assay

The reproducibility of the assay was tested with four tst-positive S. aureus strains in five independent runs. In these PCR runs, mean Ct values were found of 24.59 (SD = 0.45), 26.57 (SD = 0.23), 29.37 (SD = 0.02) and 26.05 (SD = 0.20), respectively. This indicated that the real-time tst PCR was highly reproducible.

3.4 Prevalence of the tst gene in clinical isolates

Twelve of the 51 (24%) CA S. aureus isolates were positive for the tst gene and generated Ct values between 25.13 and 34.40 (mean = 28.73, SD = 3.56). In contrast, only five of 36 (14%) of the HA S. aureus isolates were tst positive and generated Ct values between 27.02 and 32.11 (mean = 29.41, SD = 2.17). Four of the 16 (25%) S. aureus strains isolated from patients with WG were tst positive and generated Ct values between 25.23 and 28.81 (mean = 27.08; SD = 1.65). This percentage corresponded to that reported in a previous study, in which a percentage of 19% among WG patients was described [10].

3.5 Clonal relationship of tst-positive CA and HA S. aureus and WG clinical isolates

The clonal relationship between the tst-positive CA and HA S. aureus isolates was studied by PFGE. From the different PFGE patterns a dendrogram was constructed (Fig. 2). Six clonal groups (A–F) were found, and the majority (9 out of 17) of the TSST-1-positive S. aureus isolates were grouped within major clonal group A. Interestingly, most of these isolates were obtained in 2002 and 2003, and had a CA origin. The other 8 isolates were grouped within five minor clonal groups (B–F). These strains were isolated between 1999 and 2001, and had either a CA or HA origin. The majority of the tst-positive CA and HA S. aureus strains was resistant to amoxicillin and/or penicillin, but a common resistant phenotype was not found among the CA S. aureus strains. In addition, none of the tst-positive S. aureus strains was positive for PVL [18].

Figure 2

Dendrogram of S. aureus isolates (CA, HA and WG) carrying the tst gene. The four columns on the right represent isolate code, year of isolation, origin and clonal group, respectively. Community-acquired (CA); hospital-acquired (HA); Wegener's Granulomatosis (WG).

A link between the patients carrying the CA S. aureus strains was not found. Some of the patients had been hospitalized during a period of 12 months prior to isolation of the CA S. aureus strain. In addition, several patients suffered from different disorders, such as chronic obstructive pulmonary disease (COPD) and diabetes. Taken together, these data might suggest that during the period 1999–2003, tst-positive strains from a single clonal group have become predominant in the community.

The clonal relationship between the S. aureus strains isolated from WG patients was investigated with PFGE. One of the strains could not be typed, due to repeated failure to digest DNA with SmaI. Seven clonal groups (G–M) were found among the remaining 15 strains. The 4 tst-positive S. aureus strains were equally distributed between clonal group H and L. However, a relation between clonal groups of tst-positive and tst-negative S. aureus strains was not found (data not shown). Furthermore, the dendrogram showed that only two of the strains isolated from WG patients had a clonal relationship with the emerging CA strains (clonal group A). It is assumed that these two strains were acquired by the WG patients outside the hospital.

The classification of the tst-positive strains from 1999 to 2003 within six different clonal groups is in contrast to the results from Musser et al. [9] who only identified one single clonal group among TSST-1-producing S. aureus isolates that caused urogenital TSS. Although the latter strains were typed by multilocus enzyme electrophoresis (MLEE) rather than PFGE, these data may indicate that while clonally unrelated tst-positive strains may be disseminated in the community, a specific disease, such as TSS, may only be caused by specific clonal groups of S. aureus.

4 Conclusion

A real-time PCR assay was developed for the detection of the S. aureus tst gene. This assay generated results in less than two hours and was, therefore, significantly faster than conventional PCR methods [3,12]. The assay was very convenient, since it could be applied directly on diluted bacterial suspensions and previous DNA purification is not required. Furthermore, the assay was found to be highly reproducible, specific as well as robust. The percentage of tst-positive strains among S. aureus isolates from WG patients was similar to that among isolates from the community. The tst-positive CA and HA S. aureus strains were grouped in one major (A) and five minor clonal groups (B–F). Since most of the isolates from the major clonal group were isolated during the later years of the sample period, these isolates may have a more virulent phenotype than isolates which belong to other clonal groups. Furthermore, no relation was found between clonal groups of tst-positive and tst-negative S. aureus strains isolated from WG patients. Since virulent S. aureus strains pose an increasing problem worldwide, both in the hospital and in the community, the availability of a rapid, real-time PCR assay for the identification of tst-harboring S. aureus strains may contribute to the control of the spread of virulent clones of S. aureus.

Acknowledgments

J. Etienne, M. Bes (INSERM, Lyon, France) and W. Manson (University of Groningen, The Netherlands) are thanked for the collection and characterization of TSST-1-negative and TSST-1-positive S. aureus strains.

References

  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].
View Abstract