Detection of disinfectant and antibiotic resistance genes in Staphylococcus aureus isolated from the oral cavity of Tunisian children

Abstract

Staphylococcus aureus is an important pathogen associated to dental infection. Many antiseptic agents are used in hygienic handwash to prevent nosocomial infections by methicillin-resistant staphylococci. In this study, 22 Staphylococcus aureus strains isolated from the oral cavity of Tunisian children were investigated for their susceptibility to benzalkonium chloride (0.5–512 μg/ml) and antibiotics. The β-lactams resistance gene blaZ, the erythromycin resistance methylase genes (ermA, ermB and ermC), the macrolide efflux gene (msrA) and the disinfectant resistance genes (qacH, qacA, qacB and qacC) were also investigated. Determination of the minimal inhibitory concentration values revealed that 54.5, 54.5, 68, and 82% of isolates were resistant to benzalkonium chloride, oxacillin, tetracycline and erythromycin respectively. The frequency of strains positive for the antibiotic resistance genes tested was 100 (blaZ), 50 (ermA), 36.4 (ermB), 22.7 (ermC) and 13.6% (mrsA). The qacH and the qacA genes were found in 22.7% of isolates and qacB and qacC in 13.6%. Two strains harboured three qac (qacH, qacA and qacB or qacC) genes. These data highlight the importance of determining the susceptibility to antibiotics and disinfectants of strains isolated in dental medicine in order to monitor the epidemiology and spread of multi-drug resistant staphylococci.

Introduction

The human oral cavity is habited by more than 500 cultivable and non-cultivable bacterial species (Paster et al. 2001). The majority of these species are commensals, while a subset comprises opportunistic pathogens. Their key role in the etiology of periodontitis and dental caries in the world is well established (Meyer and Fives-Taylor 1998), and they has been also implicated in the etiology of a number of systemic diseases, such as infective endocarditis (Barrau et al. 2004), respiratory infections (Mojon and Bourbeau 2003) and cardiovascular diseases (Meyer and Fives-Taylor 1998). Interestingly, in some studies, microorganisms such as staphylococci, coliforms and Candida spp. have been commonly isolated from peri-implant lesions (Leonhardt et al. 1999; Salvi et al. 2008). More recently, Staphylococcus aureus has been demonstrated to have the ability to adhere to titanium surfaces (Gristina 1987), which may be significant in terms of the colonisation of dental implants and subsequent infections (Harris and Richards 2004).

Oral bacteria are exceptionally adept at acquiring resistance to pencillins, tetracycline, macrolides and antiseptic agents (Sweeney et al. 2004). Methicillin-resistant S. aureus (MRSA) is considered to be one of the most difficult bacteria to treat in patients, and successful treatment is compounded by the fact that many strains also possess efflux pumps, which export certain tetracyclines, macrolides and qacA genes that confer resistance to antibiotics and antiseptics (Marshall and Piddock 1997).

Resistance to macrolides in staphylococci may be based on the active efflux mechanism encoded by the msrA gene or on 23S rRNA methylation, which is encoded by erythromycin resistance genes [erm(A), erm(B), erm(C)] (Fluit et al. 2001). It is well-established that in MRSA the tet(M) and tet(K) genes confer tetracycline resistance through ribosomal protection and an active efflux mechanism, respectively (Trzcinski et al. 2000; Fluit et al. 2001).

Benzalkonium chloride (BC), is frequently used in hospitals to disinfect and prevent the spread of pathogens. The widespread use of BC may impose a selective pressure and contribute to the emergence of disinfectant-resistant microorganisms in these environments (Reverdy et al. 1993; Russell 2000). In clinical cases of staphylococci, qacA, qacB and qacC/smr are generally plasmid borne and are widely distributed in the environment (Leelaporn et al. 1995; Noguchi et al. 1999; Mayer et al. 2001).

In the study reported here, we examined the antibiotic and BC susceptibility of 22 S. aureus strains isolated from the oral cavity of Tunisian children. We also investigated the prevalence of blaZ, ermA, ermB, ermC and msrA genes, which are responsible for antibiotic resistance, as well as the qac genes, which are responsible for disinfectant resistance, in S. aureus strains.

Materials and methods

Patients and Bacterial strains

Samples were taken with a sterile swab from the oral cavity of Tunisian children (n = 90). The mean age of the children was approximately 4–12 years. Prior to study commencement, all clinical procedures were approved by the Ethical Committee of the Faculty of Medicine, University of Monastir, Tunisia, and written informed consent was obtained from the parents of all participants. The criteria for inclusion were: no antibiotic treatment during the 4 weeks preceding sampling and no systemic disease.

Species identification was performed using API 20 Staph strips (BioMerieux, Marcy l'Etoile, France) according to the manufacturer’s recommendation, and the results were read using an automated microbiological mini-API (Bio-Merieux).

Molecular identification of S. aureus strains was performed using specific primers (Sa442-1 and Sa442-2), as previously described (Martineau et al. 1998).

Minimal inhibitory concentration determination

The broth microdilution method was used to determine the minimum inhibitory concentration (MIC) of BC (Acros Organics, Geel, Belgium), oxacillin (OX), tetracycline (TET) and erythromycin (ERY) (0–512 μg/ml) according to the Comité de l’antibiogramme de la société française de microbiologie (CASFM 2010). After incubation, bacterial growth was evaluated by the presence of turbidity and a pellet on the well bottom. The MIC was defined as the lowest concentration of compound that resulted in no macroscopically visible growth.

Detection of antibiotic resistance gene by PCR

Genomic DNA was extracted using a Wizard Genomic Purification kit (Promega, Madison, WI). The presence of blaZ DNA was detected by PCR using forward and reverse primers as described previously (Martineau et al. 2000). Multiplex PCR for the detection of macrolide resistance genes ermA, ermB, ermC and msrA was performed as described elsewhere (Martineau et al. 2000; Lim et al. 2002).

Detection of qacH, qacA, qacB and qacC genes by PCR–restriction fragment length polymorphism

The presence of the qacH gene was detected by PCR using forward and reverse primers as described previously (Sidhu et al. 2002).

Multiplex PCR–restriction fragment length polymorphism (RFLP) analysis of the qacA/B and qacC genes was performed as described by Sekiguchi et al. (2004).

Following PCR, the product was digested with 5 U of AluI (Promega) at 37°C for 90 min. The treated product (10 μl) was analysed by gel electrophoresis in 3% agarose gel in 1× Tris-borate-EDTA buffer (TBE, pH 8.3). The amplification products were photographed and their size determined using a 25-bp molecular size marker (Promega). The qacB gene was expected to be digested into two fragments of 176 and 44 bp, respectively. Genes qacA and qacC were not expected to be digested (Sekiguchi et al. 2004).

Results

Molecular identification of S. aureus strains

Molecular identification using specific primers showed the presence of 22 S. aureus among 59 Gram-positive strains isolated from the oral cavity of Tunisian children with dental caries (n = 60) and controls (n = 30) as presented in Table 1 .

Table 1 Relationship between the minimal inhibitory concentration of benzalchonium chloride and the presence of disinfectant resistance genes qacH, qacA, qacB and qacC

Antimicrobial susceptibility

The isolates were screened for their susceptibility to BC (Table 1) according to the MIC values. Among the 22 S. aureus strains, 12 (54.5 %) were considered to be resistant to BC (MIC 4–128 μg/ml) and ten (45.45 %) to be sensitive (MIC ≤2 μg/ml) according to the guideline of the CASFM (2010).

Several strains were resistant toward all three antibiotics tested (TET, OXA, ERY), 12 (54.5%) were resistant to oxacillin (MIC >2 μg/ml), 15 (68%) were resistant to tetracyline (MIC >2 μg/ml) and 18 (82%) were resistant to erthromycin (MIC >2 μg/ml) according to the CASFM 2010. The incidence of strains resistant to tetracycline and erythromycin was higher among MRSA than among methicillin-resistant S. aureus (MSSA) isolates, with 10/12 (83%) MRSA isolates being resistant to tetracycline and 11/12 (91%) of isolates being resistant to erythromycin (Table 2).

Table 2 Relationships between MIC of antibiotics, and the presence of resistance genes blaZ, ermA, ermB, ermC and msrA

Detection of antibiotics resistance genes

All the isolated S. aureus were positive for the blaZ gene. The incidence of the three erythromycin ribosomal methylase genes was as follows: 11 of 22 (50%) S. aureus contained ermA, eight strains harboured ermB (36.4%) and five (22.7 %) strains were positive for ermC. The msrA gene was present in three (13.6%) S. aureus strains. Four strains (18%) were susceptible to erythromycin and did not contain any erythromycin resistance gene. In contrast, no erythromycin resistance gene was detected in one strain (B136), although this strain was resistant to erythromycin (Table 2).

PCR–RFLP detection of qacA, qacB, qacC and qacH genes

Among the 22 S. aureus strains tested, 12 were qac-negative, while qacA and qacH were detected in five (22.7%) strains and qacC and qacB were detected in three (13.6%) strains (Table 1). Notably, two of the ten qac-positive strains (B364 and B291) carried three different genes (qacA, qacB, qacH and qacA, qacC, qacH, respectively).

Discussion

The large consumption of antibiotics worldwide has resulted in the development and spread of a large number of antibiotic resistance determinants among bacterial populations. Concerns have arisen regarding the potential emergence of cross-resistance between widely used disinfectants and antibiotics (Paulsen et al. 1998; Russell 2000).

In this study, we report the presence of S. aureus in carious children. A large number of the S. aureus isolates were resistant toward the three antibiotics tested (OXA, TET and ERY) and harboured the antibiotic resistance genes. Our data demonstrate that all S. aureus (either MSSA and MRSA) strains tested, including the two strains from caries-free children, harboured the blaZ gene. The frequency of MRSA estimated in our study is significantly higher than that reported by Petinaki et al. (2001) who found that only 14.8% of strains were resistant to oxacillin. We also found that among the MRSA isolates, 83% of strains were resistant to tetracycline and 91% of strains were resistant to erythromycin. These results are in contrast with those of Petinaki et al. (2001), who showed that 16.21% of MRSA were resistant to tetracycline and 81% were resistant erythromycin.

Erythromycin resistance in staphylococci is predominantly mediated by erm genes encoded by methylases (Weisblum 1995). Accordingly, our investigation on the prevalence of erythromycin resistance genes showed that of the 17 strains resistant to erythromycin, 15 contained at least one erm gene (among them only two were from caries-free children). Of the remaining two strains, one (B136) did not carry any of the investigated erythromycin resistance genes (Table 2) and showed low-level resistance. Similarly, Sekiguchi et al. (2004) reported discordance between phenotypic susceptibility and the presence of erm genes. This result may be explained by the location of resistance genes in small plasmids that could occasionally be lost (Fluit et al. 2001).

Staphylococcal strains resistant to macrolides and type-B streptogramins frequently harbour msrA, which encodes an ATP-dependent efflux pump (Eady et al. 1993). In contrast, we detected  msrA only in three strains. These findings are in disagreement with the study by Eady et al. (1993) conducted with staphylococci in the UK in which an incidence of 5.9% for ermA, 7.2% for ermB and 33% for msrA was reported. Our results also deviate from those of a recent study of Ardic et al. (2005) in which a higher level of ermA and ermC was found in MRSA isolated from a hospital setting.

Disinfectants using quaternary ammonium compounds (QAC) are frequently used in hospitals to prevent the spread of pathogens. However, the widespread use of these disinfectants in hospitals actually contributes to the emergence of disinfectant-resistant bacteria (Reverdy et al. 1993; Russell 2000). In S. aureus, qacA and qacB, which encode multidrug-transporter proteins (Rouch et al., 1990), have been identified as antiseptic resistance genes that confer resistance to cationic antiseptic agents, including dyes such as ethidium bromide, QACs, such as BC, and biguanides, such as chlorhexidine digluconate (Littlejohn et al. 1992; Putman et al. 2000). Among the bacteria tested in this work, 12 (55%) isolates, 11 from caries-active and one from caries-free children, were considered to be BC resistant (MIC 4–128 μg/ml), and the remaining ten (45%) were considered to be sensitive (MIC ≤2 μg/ml). Similarly, it has been reported that 50% of S. aureus isolated from bloodstream infections were resistant to BC (Sidhu et al. 2002). Different genes, including qacH, qacA and qacB or qacC, confer effux-mediated resistance to lipophilic cationic compounds, including BC and dyes, in staphylococci (Paulsen et al. 1996).

Our molecular detection analysis of the disinfectant’s resistance genes showed that eight strains resistant to BC (7 from caries-active and one from caries-free children) contained at last one of the four qac genes. The QacA and qacH genes were the most frequent (both 22.7%), followed by qacC and qacB (both 13.6%). Previous investigators have reported a similar distribution of these qac resistance genes in clinical S. aureus and CoNS (Leelaporn et al. 1995; Mayer et al. 2001).

On the other hand, our data disagree with those of Sekiguchi et al. (2004), who found that 49% of their MRSA isolates were positive for qacA, 10% were positive for qacC and only one isolate was positive for qacB. It should be noted that, as shown in Table 1, two strains (B364 and B291) harboured three qac resistance genes (qacA, qacB, qacH and qacA, qacC, qacH) but showed different MIC values for BC (128 and 16 μg/ml, respectively). Little is known about the occurrence and possible genetic linkage of qac and antibiotic resistance in staphylococci. Interestingly, we observed that staphylococci resistant to BC were generally more often resistant to antibiotics (Oxa, Ery and Tet) than BC-sensitive isolates (Fig. 1). This result is in agreement with data reported by Sidhu et al. (2002). MRSA isolates resistant to antiseptics and disinfectants have been reported in Australia and in the UK in the last decade (McDonnell and Russell 1999). QacA/B genes are typically located on a transposon of transmissible multidrug-resistant plasmids, such as pSK1 (Littlejohn et al. 1991). Noguchi et al. (1999) reported that when the antiseptic susceptibility and the distribution of antiseptic resistance genes of MRSA isolated in Japan in 1992 were studied, qacA/B genes were detected in 10.2% of the strains tested (10/98). However, 7 years later, in a similar study of MRSA isolates in Japan, qacA/B genes were detected in 47.9% (198/413) of the isolates.

Fig. 1

Percentages of antibiotic-resistant S. aureus isolates among benzalkonium chloride (BC)-resistant and BC-sensitive isolates. Antibiotics used for testing were tetracycline (TET), oxacillin (OXA) and erythromycin (ERY)

In conclusion, monitoring susceptibility to antibiotics and disinfectants in bacteria isolated during dental procedures can greatly contribute to the epidemiological control of multidrug resistant strains.